Industrial Waste
Management

-------
This Guide provides state-of-the-art tools and
practices to enable you to tailor hands-on
solutions to the industrial waste management
challenges you face.
           WHAT'S AVAILABLE

           • Quick reference to multimedia methods for handling and disposing of wastes
             from all types of industries
           • Answers to your technical questions about siting, design, monitoring, operation.
             and closure of waste facilities
           • Interactive, educational tools, including air and ground water risk assessment
             models, fact sheets, and a facility siting tool.
           • Best management practices, from risk assessment and public participation to
             waste reduction, pollution prevention, and recycling

-------
                  ^DGEMENTS
The fotowing members of the Industrial Waste Focus Group and the Industrial Waste Steering Commute aregrateftjy
acknowledged far al of their dme and assistance m the development of this guidance document
                        'ICUS
Grou;   .
rout own, lilts lyuw ^iiaiu^a
  Company
Walter Carey. Nestle  USA. Inc and
  New Milford Farms
Rama Chaturvedi Bethlehem Steel
  Corporation
H.C. Clark. Rice University
Barbara Dodds.  League of Women
  voters
Chuck Feerick. Exxon Mobil
  Corporation
Stacey Ford. Exxon Mobil
  Corporation
Robert  Giraud DuPont Company
John Harney. Citizens Round
  Tabte/PURE
Kyle Isakower. American Petroleum
  Institute
Richard Jarman, National Food
  Processors Association
James Meiers, Cinergy Power
  Generation Services
Scott Murto. General Motors and
  American Foundry Society
James Roewer, Edison Electric
  Institute
Edward Repa. Environmental
  Industry Association
Tim Saytor, International Paper
Amy Schalfer, Weyerhaeuser
Ed Skemote, WMX Technologies. Inc
Michael Wach Western
  Environmental Law Center
David Wels, University of South
  *"•*-•"* Medical Center

i-ai owin. Cherokee Nation of
  Oklahoma

                    rocu?.

Wl* wtMMIUd. WMMIU l*IUW
Brian Forrestal. Laidlaw Waste
  Systems
Jonathan Greenberg, Browning-
  Ferns Industries
Michael Gregory, Arizona Toxics
  Information and Sierra Club
Andrew Miles, The Dexter
  Corporation
Gary Robbins, Exxon Company
Kevin San. National Paint & Coatings
  Association
Bruce Steter. American Iron & Steel
Lisa Williams, Aluminum Association




  and Territorial Solid Waste
  Management Officials
Marc Crooks. Washington State
  Department of Ecology
Cyndi Darling. Maine Department of
  Environmental Protection
Jon Dilliard Montana Department of
  Environmental Qualty
Anne Dobbs. Texas Natural
  Resources Conservation
  Commission
Richard Hammond. New York Slate
  Department of Environmental
  Conservation
Elizabeth Haven California State
  Waste Resources Control Board
Jim HuD. Missouri Department of
  Natural Resources
Jim Knudson. Washington State
  Department of Ecology
Chris McGuire,  Florida Department
  of Environmental Protection
Gene Mitchell Wisconsin
  Department of Natural Resources
William Pounds. Pennsylvania
  Department of Envtanmental
  Protection
Bjjan Sharafkhani Louisiana
  Department of Environmental
  Qualty
James Warner,  Minnesota Pollution
  Control Agency



rmitaa omit, ntmio uupwUinfn Of
  Environmental Protection
NormGumenik Arizona Department
  of Environmental Qualty
Steve Jenkins, Alabama Department
  of Environmental Management
Jim North  Arizona Department of
  Environmental Qualty

-------
Industrial waste is generated by the production
of commercial goods, products, or services.
Examples include wastes from the production
of chemicals, iron and steel, and food goods.

-------
                                  METHOD  1311

                  TOXICITY CHARACTERISTIC LEACHING PROCEDURE
1.0  SCOPE AND APPLICATION

       1.1   The TCLP is designed to determine the  mobility  of  both organic and
inorganic analytes present in liquid, solid,  and multiphasic wastes.

       1.2   If  a  total  analysis  of the  waste demonstrates  that individual
analytes  are not  present in the  waste, or that they are present but at such low
concentrations that  the appropriate  regulatory levels could  not  possibly be
exceeded, the TCLP need not be  run.

       1.3   If  an  analysis   of any one of  the liquid fractions  of  the TCLP
extract indicates that  a regulated compound is  present at such high concentra-
tions that, even  after  accounting for dilution  from the other  fractions of the
extract,  the concentration would be above the regulatory level for  that  compound,
then the  waste is  hazardous  and it  is  not  necessary  to  analyze  the  remaining
fractions of the  extract.

       1.4   If an analysis of  extract  obtained using  a bottle extractor shows
that the  concentration  of any  regulated  volatile analyte exceeds  the regulatory
level for that compound, then  the waste is hazardous and extraction using  the ZHE
is not necessary.  However,  extract from a  bottle  extractor cannot be  used to
demonstrate that  the  concentration of volatile compounds is  below  the regulatory
1evel .

2.0    SUMMARY OF METHOD

       2.1   For liquid wastes  (i.e.. those  containing  less  than  0.5%  dry solid
material),  the waste,   after  filtration through a  0.6 to  0.8 urn  glass fiber
filter, is defined as the TCLP  extract.

       2.2   For wastes containing  greater  than or equal  to  0.5%  solids,  the
liquid, if any, is separated from the solid phase and stored  for later  analysis;
the particle size  of  the solid phase  is  reduced, if necessary.  The solid phase
is extracted with  an  amount of extraction fluid  equal to 20  times the  weight of
the solid phase.   The extraction fluid employed  is a function of  the alkalinity
of the  solid phase of the waste.   A special extractor vessel  is  used  when  testing
for volatile analytes  (see Table  1 for a  list of volatile  compounds).   Following
extraction, the liquid  extract  is separated from the solid  phase by filtration
through a 0.6 to  0.8 urn glass fiber filter.

      2.3    If compatible (i.e..  multiple phases will  not form  on combination),
the initial  liquid phase of the  waste  is  added to the liquid extract,  and these
are analyzed together.   If incompatible,  the liquids are analyzed  separately and
the  results  are  mathematically combined to  yield a  volume-weighted  average
concentration.

CD-ROM                              1311-  1                       Revision 0
                                                                 July 1992

-------
3.0   INTERFERENCES

      3.1    Potential  interferences that may be encountered during  analysis are
discussed in the individual analytical methods.

4.0   APPARATUS AND MATERIALS

      4.1    Agitation apparatus:  The  agitation  apparatus must be capable of
rotating the  extraction  vessel  in an  end-over-end fashion  (see  Figure  1) at
30 ± 2 rpm.   Suitable devices known to EPA are identified  in Table 2.

      4.2    Extraction Vessels

             4.2.1    Zero-Headspace Extraction Vessel  (ZHE).   This  device is
      for use only when the waste is being tested for the mobility of volatile
      analytes (i.e..  those listed  in Table 1).  The ZHE (depicted  in Figure 2)
      allows  for  liquid/solid separation  within  the device,  and  effectively
      precludes headspace.   This type  of vessel allows for  initial  liquid/solid
      separation,  extraction, and final extract filtration without opening the
      vessel  (see  Section  4.3.1).   The  vessels  shall have an  internal  volume of
      500-600 ml,  and  be equipped to accommodate a  90-110 mm filter. The devices
      contain VITON®1 0-rings which  should be  replaced frequently.   Suitable ZHE
      devices known to EPA are identified in Table 3.

             For the  ZHE  to be  acceptable for use,  the piston  within  the ZHE
      should be able to be  moved with approximately 15 psi  or less.  If it takes
      more pressure  to  move the piston,  the 0-rings in the device  should be
      replaced.  If this does not solve the problem, the  ZHE is unacceptable for
      TCLP analyses and the manufacturer should be contacted.

             The ZHE should be checked for leaks after every extraction.   If the
      device  contains  a  built-in  pressure  gauge,  pressurize  the  device to
      50 psi,  allow it  to  stand unattended for 1 hour,  and  recheck the pressure.
      If the  device does   not have  a  built-in pressure  gauge,  pressurize the
      device to 50 psi,  submerge it in water, and  check  for the presence of air
      bubbles escaping  from any of the fittings.  If pressure is  lost, check all
      fittings  and  inspect  and  replace  0-rings,  if necessary.    Retest  the
      device.   If  leakage  problems  cannot be  solved, the manufacturer should be
      contacted.

             Some  ZHEs  use gas pressure to actuate the ZHE  piston, while  others
      use mechanical pressure (see Table  3).   Whereas  the volatiles  procedure
      (see  Section 7.3)   refers  to  pounds  per  square inch  (psi), for  the
      mechanically  actuated  piston,   the   pressure  applied is   measured  in
      torque-inch-pounds.    Refer to the  manufacturer's  instructions  as  to the
      proper conversion.
     1  VITON" is a trademark of Du Pont.

CD-ROM                              1311-  2                       Revision 0
                                                                 July 1992

-------
             4.2.2    Bottle  Extraction  Vessel.    When  the  waste   is  being
      evaluated using  the nonvolatile extraction, a jar with  sufficient capacity
      to hold  the sample  and  the extraction  fluid  is needed.   Headspace is
      allowed in this  vessel .

             The extraction bottles may be constructed from various materials,
      depending on the analytes  to  be analyzed  and the nature of the waste  (see
      Section 4.3.3).   It is recommended that borosilicate  glass  bottles be  used
      instead  of  other  types   of  glass,  especially  when  inorganics are of
      concern.   Plastic bottles, other than polytetraf1uoroethylene,  shall  not
      be used if organics are to  be investigated.  Bottles are available from a
      number of laboratory  suppliers.   When  this  type of extraction  vessel is
      used,  the filtration  device discussed in Section  4.3.2  is  used for initial
      liquid/solid separation  and final extract filtration.

      4.3    Filtration Devices:   It  is  recommended  that  all  filtrations be
performed in a  hood.

             4.3.1    Zero-Headspace  Extractor Vessel  (ZHE): When the waste is
      evaluated for volatiles, the zero-headspace extraction  vessel described in
      Section  4.2.1 is  used for filtration.   The device shall  be  capable of
      supporting  and  keeping  in place  the  glass  fiber filter  and be  able to
      withstand the pressure needed to accomplish separation (50 psi).

NOTE:         When  it is  suspected that  the glass fiber  filter has  been ruptured,
             an in-line glass fiber  filter  may be used to filter the material
             within the ZHE.

             4.3.2    Filter Holder:  When the waste is evaluated for other  than
      volatile  analytes, any filter holder capable of supporting a glass  fiber
      filter and able  to withstand  the  pressure needed to accomplish separation
      may be used.  Suitable filter  holders  range from simple  vacuum units to
      relatively complex systems capable of exerting pressures  of up  to 50 psi
      or more.   The type of filter  holder  used  depends on the properties of the
      material  to be  filtered  (see Section 4.3.3).  These devices shall  have a
      minimum internal  volume of 300 ml and be equipped to  accommodate a minimum
      filter size  of 47  mm  (filter  holders having  an internal capacity of 1.5 L
      or greater,  and equipped to accommodate a  142 mm  diameter  filter,  are
      recommended).  Vacuum filtration   can  only  be used  for wastes  with  low
      solids content  «10%) and for highly granular, liquid-containing wastes.
      All other  types  of  wastes  should  be  filtered  using  positive  pressure
      filtration.   Suitable filter holders known to EPA are shown in  Table 4.

             4.3.3    Materials   of   Construction:    Extraction   vessels   and
      filtration devices shall  be made of inert materials which will  not  leach
      or absorb waste  components.    Glass,  polytetraf1uoroethylene  (PTFE), or
      type   316 stainless  steel   equipment  may  be  used  when   evaluating  the
      mobility  of both organic  and inorganic components.  Devices made of  high
      density  polyethylene  (HOPE),  polypropylene (PP),  or  polyvinyl  chloride
      (PVC)  may be used  only when  evaluating the mobility of metals.  Borosili-
CD-ROM                              1311-  3                       Revision 0
                                                                 July 1992

-------
      cate  glass  bottles are  recommended for  use  over other  types  of  glass
      bottles, especially when inorganics are analytes  of  concern.

      4.4    Filters:  Filters shall  be made  of  borosilicate  glass  fiber,  shall
contain  no  binder  materials,  and shall have  an  effective  pore size of 0.6  to
0.8 urn,  or equivalent.   Filters known to EPA which meet these  specifications  are
identified  in  Table 5.   Pre-filters must not  be used.   When evaluating  the
mobility of metals,  filters  shall be  acid-washed prior to use  by rinsing  with  IN
nitric acid followed by three  consecutive  rinses with  deionized  distilled  water
(a minimum of 1  L per rinse is  recommended).   Glass fiber filters are fragile  and
should be handled with care.

      4.5    pH Meters:  The meter  should  be  accurate  to +  0.05  units  at 25  °C.

      4.6    ZHE Extract Collection  Devices:   TEDLAR®2 bags or  glass,  stainless
steel or PTFE gas-tight syringes are used to collect the  initial  liquid  phase  and
the final extract of the waste when  using the  ZHE device.  The  devices listed  are
recommended  for use  under the following conditions:

             4.6.1     If a waste contains  an  aqueous  liquid phase  or if  a  waste
      does not contain a significant amount  of nonaqueous  liquid  (i.e.. <1%  of
      total waste),  the TEDLAR® bag  or a 600 ml syringe should  be used to  collect
      and combine the  initial  liquid and  solid extract.

             4.6.2     If  a  waste contains a  significant  amount of nonaqueous
      liquid in the  initial  liquid phase (i.e.. >1% of total waste), the  syringe
      or the TEDLAR® bag may  be used for  both  the initial  solid/liquid separation
      and the final  extract  filtration.  However, analysts  should use one or  the
      other, not both.

             4.6.3     If  the  waste  contains  no  initial  liquid phase (is 100%
      solid)  or  has no  significant  solid phase  (is  100%  liquid), either  the
      TEDLAR® bag  or the syringe may be used.  If the syringe  is  used,  discard
      the  first  5  mL  of  liquid expressed  from the  device.    The  remaining
      aliquots are  used for analysis.

      4.7    ZHE  Extraction  Fluid Transfer  Devices:   Any  device capable   of
transferring the extraction  fluid into the  ZHE without  changing the nature of  the
extraction  fluid  is acceptable  (e.g. .  a  positive displacement or  peristaltic
pump, a  gas  tight syringe, pressure  filtration unit (see  Section 4.3.2),  or  other
ZHE devi ce).

      4.8    Laboratory  Balance:   Any  laboratory balance accurate to within
± 0.01 grams may be  used  (all  weight  measurements are to be  within  + 0.1  grams).

      4.9    Beaker  or Erlenmeyer flask,  glass, 500 mL.
     2  TEDLAR® is a  registered  trademark  of  Du  Pont.
CD-ROM                              1311-  4                      Revision 0
                                                                 July 1992

-------
      4.10   Watchglass,  appropriate  diameter to  cover beaker  or  Erlenmeyer
flask.

      4.11   Magnetic stirrer.

5.0  REAGENTS

      5.1    Reagent  grade  chemicals  shall   be  used  in  all  tests.    Unless
otherwise  indicated,  it is intended  that all  reagents shall  conform  to  the
specifications  of  the  Committee on Analytical   Reagents of the American Chemical
Society,  where such  specifications  are available.    Other grades  may  be used,
provided  it is  first ascertained that the reagent  is  of  sufficiently high purity
to permit its use  without lessening the accuracy of the determination.

      5.2    Reagent  Water.   Reagent  water  is defined  as  water in  which  an
interferant  is not  observed at or  above  the method's  detection  limit  of  the
analyte(s)  of  interest.   For nonvolatile  extractions,  ASTM Type II  water  or
equivalent meets the  definition of reagent water.   For volatile extractions,  it
is recommended  that  reagent  water be generated by any of the following methods.
Reagent water should be monitored periodically for impurities.

             5.2.1    Reagent  water for  volatile extractions may be generated
      by   passing  tap  water through a  carbon filter bed  containing  about  500
      grams of activated carbon (Calgon Corp., Fi1trasorb-300 or equivalent).

             5.2.2    A  water  purification   system  (Millipore  Super-Q   or
      equivalent)  may  also be  used  to  generate  reagent  water  for  volatile
      extractions.

             5.2.3    Reagent  water  for   volatile   extractions   may  also  be
      prepared  by  boiling water for 15 minutes.  Subsequently, while maintaining
      the water temperature  at 90 ± 5 degrees  C, bubble  a contaminant-free inert
      gas  (e.g.  nitrogen)  through  the water  for  1 hour.   While  still  hot,
      transfer  the water to  a narrow mouth  screw-cap  bottle under  zero-headspace
      and seal  with  a Teflon-lined septum and cap.

      5.3    Hydrochloric acid (IN), HC1,  made from ACS reagent grade.

      5.4    Nitric  acid (IN), HN03, made  from ACS  reagent grade.

      5.5    Sodium  hydroxide  (IN),  NaOH,  made from ACS reagent grade.

      5.6    Glacial  acetic acid,  CH3CH2OOH, ACS reagent grade.

      5.7    Extraction fluid.

             5.7.1    Extraction fluid #  1:    Add  5.7  ml glacial CH3CH2OOH  to
      500 ml of reagent water  (See  Section 5.2),  add 64.3  ml  of IN  NaOH,  and
      dilute to a volume of 1  liter.   When  correctly prepared,  the  pH of this
      fluid will  be  4.93 + 0.05.
CD-ROM                              1311-  5                       Revision 0
                                                                 July 1992

-------
             5.7.2    Extraction fluid  #  2:  Dilute 5.7 mL glacial  CH3CH2OOH with
      reagent water  (See  Section  5.2)  to a  volume of 1 liter.  When correctly
      prepared,  the pH of this fluid will be 2.88 + 0.05.

NOTE:        These  extraction  fluids   should  be  monitored  frequently  for
             impurities.  The pH should  be checked prior to use to ensure that
             these fluids are made  up  accurately.   If impurities are found or
             the pH is not within the above specifications, the fluid shall be
             discarded and fresh extraction fluid prepared.


      5.8    Analytical  standards  shall be prepared according to  the appropriate
analytical method.

6.0   SAMPLE COLLECTION, PRESERVATION, AND HANDLING

      6.1    All samples shall  be  collected  using  an  appropriate sampling plan.

      6.2    The TCLP may place  requirements  on  the  minimal size of the field
sample,  depending  upon the physical state  or  states of the waste and the analytes
of concern.   An  aliquot  is needed for preliminary  evaluation  of which  extraction
fluid is to be used for the nonvolatile  analyte extraction  procedure.  Another
aliquot  may  be needed to actually conduct  the nonvolatile extraction (see Section
1.4  concerning  the use of this  extract  for volatile  organics).   If volatile
organics are of  concern, another aliquot may  be needed.  Quality control measures
may  require  additional  aliquots.   Further,  it is always  wise to collect more
sample just  in case something goes wrong with the  initial  attempt  to conduct the
test.

      6.3    Preservatives shall not be  added to samples before extraction.

      6.4    Samples  may  be  refrigerated   unless   refrigeration results  in
irreversible physical  change  to  the waste.   If precipitation occurs,   the entire
sample (including precipitate) should be extracted.

      6.5    When the waste is to  be evaluated for volatile analytes,  care shall
be taken  to minimize the loss  of volatiles.   Samples  shall  be  collected and
stored  in  a manner intended to  prevent  the loss of  volatile analytes  (e.g. .
samples  should be  collected in Teflon-lined septum capped vials and stored at  4
°C.   Samples should be opened only immediately prior  to extraction).

      6.6    TCLP extracts should  be  prepared for  analysis and analyzed as soon
as possible  following  extraction.  Extracts or portions of extracts for metallic
analyte determinations  must  be  acidified with  nitric acid to a pH < 2,  unless
precipitation occurs  (see  Section 7.2.14 if precipitation  occurs).   Extracts
should be preserved for other  analytes  according  to  the guidance given  in the
individual  analysis  methods.   Extracts  or  portions  of  extracts for  organic
analyte  determinations  shall  not be  allowed  to come  into  contact with  the
atmosphere  (i.e.. no  headspace)  to   prevent  losses.    See  Section 8.0  (QA
requirements) for acceptable sample and  extract holding times.
CD-ROM                              1311-  6                       Revision 0
                                                                 July 1992

-------
7.0   PROCEDURE

      7.1    Preliminary Evaluations

      Perform  preliminary  TCLP  evaluations on  a  minimum 100  gram  aliquot of
waste.   This  aliquot may not actually undergo TCLP extraction.   These preliminary
evaluations include:  (1) determination of the percent solids  (Section 7.1.1);
(2) determination  of  whether  the waste contains  insignificant solids  and is,
therefore,  its  own  extract  after  filtration  (Section 7.1.2);  (3) determination
of whether  the solid  portion  of the  waste  requires particle size  reduction
(Section 7.1.3);  and (4)  determination of which of  the two extraction fluids are
to be used for the nonvolatile TCLP extraction of the waste (Section 7.1.4).

             7.1.1    Preliminary  determination  of  percent   solids:    Percent
      solids is defined as that fraction of a waste sample (as a percentage of
      the total  sample)  from  which  no liquid may be  forced  out  by  an applied
      pressure, as described below.

                      7.1.1.1    If the waste will obviously yield  no liquid when
             subjected  to  pressure  filtration (i.e.. is  100%  solids) proceed to
             Section 7.1.3.

                      7.1.1.2    If  the   sample   is  liquid   or  multiphasic,
             liquid/solid  separation  to  make  a  preliminary  determination  of
             percent solids is  required.   This  involves  the filtration device
             described  in  Section 4.3.2  and  is  outlined  in  Sections  7.1.1.3
             through 7.1.1.9.

                      7.1.1.3    Pre-weigh  the  filter and the container that will
             receive the filtrate.

                      7.1.1.4    Assemble the  filter  holder  and filter following
             the manufacturer's instructions.   Place the filter on the support
             screen and secure.

                      7.1.1.5    Weigh  out  a  subsample of the waste  (100 gram
             minimum) and record the weight.

                      7.1.1.6    Allow  slurries  to  stand  to   permit  the  solid
             phase  to  settle.   Wastes  that  settle  slowly may be centrifuged
             prior to filtration.  Centrifugation is to be used only as an aid
             to filtration.  If  used, the liquid  should be decanted and filtered
             followed by filtration of  the solid  portion of  the waste through
             the same filtration system.

                      7.1.1.7    Quantitatively transfer the waste  sample to the
             filter holder (liquid and solid phases).  Spread  the waste sample
             evenly over the surface of the filter.  If filtration of the waste
             at 4 °C reduces the amount of expressed liquid over what would be
             expressed at room temperature then  allow the sample to warm up to
             room temperature  in the device before filtering.

CD-ROM                              1311-  7                       Revision 0
                                                                 July 1992

-------
NOTE:         If waste material  (>1%  of original sample weight)  has  obviously
             adhered  to  the  container used  to transfer  the  sample  to  the
             filtration  apparatus,  determine  the weight  of this  residue  and
             subtract it  from the sample weight determined  in Section 7.1.1.5 to
             determine the weight of the waste sample that will  be filtered.

                      Gradually apply  vacuum  or gentle pressure  of  1-10  psi,
             until  air or pressurizing  gas moves through  the  filter.   If  this
             point  is not reached under  10 psi,  and if no additional liquid has
             passed through  the filter in any  2 minute interval, slowly increase
             the pressure in 10 psi  increments  to a  maximum of  50 psi.   After
             each incremental  increase  of 10  psi, if the  pressurizing  gas  has
             not moved  through  the  filter,  and if  no additional liquid  has
             passed through  the filter  in any  2  minute interval,  proceed to the
             next 10 psi increment.  When the pressurizing  gas  begins  to  move
             through the  filter, or when liquid flow  has  ceased at  50 psi (i.e..
             filtration  does not  result  in any additional filtrate within  any  2
             minute period), stop the filtration.

NOTE:         Instantaneous  application  of high  pressure can  degrade  the  glass
             fiber  filter and  may cause premature plugging.

                      7.1.1.8    The material  in  the filter holder  is defined as
             the solid phase of the  waste, and  the  filtrate  is  defined  as  the
             liquid phase.

NOTE:         Some wastes,  such as  oily wastes  and  some  paint  wastes,  will
             obviously contain  some material  that appears to be a liquid.   Even
             after   applying vacuum  or pressure  filtration,   as   outlined  in
             Section 7.1.1.7,  this  material  may not  filter.   If  this  is  the
             case,  the material within  the  filtration device is  defined  as  a
             solid.   Do  not replace the original  filter  with  a  fresh  filter
             under  any circumstances.  Use  only one  filter.

                      7.1.1.9    Determine  the  weight of  the  liquid  phase  by
             subtracting the  weight of the   filtrate  container  (see  Section
             7.1.1.3) from  the total weight of  the  filtrate-filled container.
             Determine the  weight  of the solid  phase of  the waste  sample  by
             subtracting the weight of the  liquid phase from the weight of the
             total  waste sample,  as determined in Section  7.1.1.5 or  7.1.1.7.

                      Record  the  weight  of   the  liquid  and  solid  phases.
             Calculate the percent  solids  as  follows:

                        Weight of solid (Section 7.1.1.9)
Percent solids = 	   x 100
                 Total  weight of waste (Section 7.1.1.5 or 7.1.1.7)
             7.1.2    If the  percent  solids  determined in  Section  7.1.1.9  is
      equal  to or  greater  than  0.5%,  then proceed either  to  Section  7.1.3  to

CD-ROM                              1311-  8                       Revision  0
                                                                 July  1992

-------
      determine whether the solid material  requires particle size reduction or
      to Section  7.1.2.1  if it is noticed that a  small  amount of the filtrate is
      entrained in wetting of the filter.   If the percent solids  determined in
      Section 7.1.1.9 is  less than 0.5%, then proceed  to Section 7.2.9  if the
      nonvolatile  TCLP  is to  be performed  and  to Section  7.3  with a  fresh
      portion of the waste if the volatile  TCLP  is to  be performed.

                      7.1.2.1   Remove  the  solid  phase and  filter  from  the
             filtration apparatus.

                      7.1.2.2   Dry the filter and  solid phase  at  100  + 20 °C
             until two successive  weighing  yield the  same value within  + 1%.
             Record the final weight.

NOTE:         Caution should be taken  to ensure that the subject  solid will  not
             flash upon  heating.   It  is recommended  that the drying oven be
             vented to a  hood or other appropriate device.

                      7.1.2.3   Calculate  the percent  dry solids  as  follows:

                        (Wt.  of dry waste + filter) -  tared wt.  of filter
Percent dry  solids  = 	 x 100
                        Initial  wt. of waste (Section  7.1.1.5 or  7.1.1.7)

                      7.1.2.4   If the  percent  dry solids is less than  0.5%,
             then  proceed  to Section  7.2.9  if  the nonvolatile  TCLP  is  to be
             performed,  and  to  Section 7.3  if  the  volatile TCLP  is   to  be
             performed.  If the percent dry solids is  greater than  or equal to
             0.5%, and if  the nonvolatile TCLP is  to be  performed, return to the
             beginning of  this Section  (7.1) and,  with  a  fresh  portion of waste,
             determine whether  particle size reduction  is  necessary (Section
             7.1.3) and  determine  the  appropriate extraction fluid (Section
             7.1.4).   If  only the volatile  TCLP is to be  performed, see the note
             in Section 7.1.4.

             7.1.3    Determination of whether the  waste  requires particle size
      reduction (particle size  is  reduced  during  this  step):   Using  the  solid
      portion of  the waste, evaluate the solid for particle  size.  Particle size
      reduction is required,  unless  the solid has a surface  area per  gram of
      material  equal  to or greater  than  3.1  cm2,  or is  smaller than  1  cm  in its
      narrowest dimension (i.e.. is capable of passing through a  9.5  mm  (0.375
      inch)   standard  sieve).   If the  surface area  is  smaller or the particle
      size larger than described above, prepare  the solid portion of  the  waste
      for extraction  by crushing,  cutting,   or grinding  the waste to  a  surface
      area or particle size as described above.  If the solids are prepared for
      organic  volatiles  extraction,   special  precautions  must  be  taken  (see
      Section 7.3.6).

NOTE:  Surface area criteria are meant for  filamentous  (e.g. .  paper,  cloth, and
      similar) waste  materials.   Actual measurement  of  surface area is  not
      required, nor is it  recommended.  For materials that do not  obviously meet

CD-ROM                             1311- 9                       Revision 0
                                                                 July 1992

-------
      the criteria,  sample  specific  methods  would need  to be  developed  and
      employed to measure the surface  area.  Such  methodology  is  currently  not
      available.

             7.1.4    Determination of  appropriate  extraction  fluid:   If  the
      solid  content of the  waste  is  greater than or equal  to  0.5%  and  if  the
      sample  will  be extracted  for  nonvolatile  constituents  (Section  7.2),
      determine  the   appropriate  fluid  (Section  5.7)  for  the  nonvolatiles
      extraction  as follows:

NOTE:         TCLP  extraction  for  volatile  constituents  uses  only  extraction
             fluid #1  (Section  5.7.1).    Therefore,  if  TCLP  extraction  for
             nonvolatiles is not required,  proceed to  Section 7.3.

                      7.1.4.1    Weigh out a small  subsample of  the  solid phase
             of the waste, reduce the solid (if necessary)  to  a  particle size of
             approximately 1 mm in diameter or less, and transfer 5.0 grams of
             the   solid  phase  of the  waste to a  500  ml beaker  or  Erlenmeyer
             flask.

                      7.1.4.2    Add 96.5  ml of  reagent  water  to the  beaker,
             cover with  a watchglass,  and stir vigorously for 5  minutes  using a
             magnetic stirrer.   Measure and record the pH.   If  the  pH is <5.0,
             use  extraction  fluid  #1.   Proceed to Section  7.2.

                      7.1.4.3    If the  pH  from  Section  7.1.4.2  is  >5.0,  add
             3.5  ml IN HC1,  slurry  briefly,  cover with a watchglass,  heat to 50
             °C,  and  hold at 50  °C  for  10 minutes.

                      7.1.4.4    Let the solution  cool to  room  temperature  and
             record the  pH.  If the pH is <5.0,  use extraction fluid #1.   If the
             pH is >5.0,  use extraction fluid #2.   Proceed to Section 7.2.

             7.1.5    If  the  aliquot of  the waste used  for  the  preliminary
      evaluation   (Sections  7.1.1  -  7.1.4)  was  determined to be  100%  solid at
      Section  7.1.1.1,  then it  can  be  used for  the  Section  7.2  extraction
      (assuming  at  least 100 grams  remain), and  the  Section 7.3  extraction
      (assuming at least  25  grams remain).   If the aliquot  was subjected to  the
      procedure in Section 7.1.1.7, then another  aliquot shall  be used  for  the
      volatile extraction procedure in  Section 7.3.   The  aliquot of the waste
      subjected to the procedure in Section  7.1.1.7 might be  appropriate for use
      for the Section  7.2  extraction  if  an  adequate  amount  of  solid  (as
      determined   by   Section  7.1.1.9)  was  obtained.    The amount  of  solid
      necessary is dependent upon whether a  sufficient  amount of extract will be
      produced to support  the analyses.  If  an adequate amount of  solid remains,
      proceed to  Section 7.2.10  of the nonvolatile TCLP  extraction.

      7.2    Procedure When  Volatiles are not Involved

      A minimum sample size  of 100 grams (solid  and liquid phases) is recommend-
ed.   In  some  cases, a larger sample  size may be  appropriate,  depending  on  the

CD-ROM                             1311- 10                      Revision  0
                                                                 July  1992

-------
solids  content  of  the  waste  sample  (percent solids, See Section 7.1.1), whether
the initial  liquid phase  of  the waste will be miscible with the aqueous extract
of the  solid,  and whether  inorganics,  semivolatile  organics,  pesticides, and
herbicides are  all analytes of concern.  Enough solids should be generated for
extraction such that the volume of TCLP  extract will be sufficient  to  support all
of the  analyses required.   If the amount of extract generated by a single TCLP
extraction will not be sufficient  to perform all of the analyses, more than one
extraction may  be  performed and the extracts from each combined and aliquoted for
analysis.

             7.2.1    If the waste will  obviously yield no liquid  when  subjected
      to pressure  filtration (i.e.. is 100% solid,  see Section 7.1.1),  weigh out
      a subsample  of the  waste  (100 gram minimum) and proceed to  Section 7.2.9.

             7.2.2    If  the sample  is  liquid  or  multiphasic, liquid/solid
      separation  is required.  This involves the filtration device described in
      Section 4.3.2 and is outlined in  Sections 7.2.3 to 7.2.8.

             7.2.3    Pre-weigh the container that will receive  the filtrate.

             7.2.4    Assemble  the filter  holder  and  filter  following  the
      manufacturer's instructions.  Place the filter on the  support screen and
      secure.  Acid wash  the  filter if  evaluating  the mobility of metals (see
      Section 4.4).

NOTE:         Acid  washed  filters  may  be used  for  all  nonvolatile extractions
             even when metals are not of concern.

             7.2.5    Weigh out  a subsample  of  the  waste  (100 gram minimum) and
      record the  weight.   If the  waste contains  <0.5% dry  solids  (Section
      7.1.2), the  liquid portion of the waste,  after filtration, is defined as
      the TCLP extract.  Therefore, enough  of the  sample should be filtered so
      that  the  amount of filtered liquid  will  support  all of  the  analyses
      required  of the TCLP  extract.  For wastes  containing >0.5%  dry  solids
      (Sections 7.1.1  or  7.1.2),  use the percent solids information obtained in
      Section 7.1.1 to determine  the optimum sample size (100 gram minimum) for
      filtration.    Enough solids  should be  generated by filtration to support
      the analyses to be performed on  the TCLP  extract.

             7.2.6    Allow  slurries  to  stand  to permit  the solid  phase  to
      settle.  Wastes  that settle  slowly may be  centrifuged  prior  to  filtration.
      Use  centri f ugati on only  as an   aid  to  filtration.    If  the  waste  is
      centrifuged,  the  liquid  should  be  decanted  and  filtered  followed  by
      filtration  of the solid portion  of the waste through the same filtration
      system.

             7.2.7    Quantitatively transfer the waste sample (liquid  and  solid
      phases) to  the filter  holder  (see  Section  4.3.2).   Spread  the waste sample
      evenly over  the  surface  of  the filter.  If filtration  of the waste at 4 °C
      reduces the  amount  of expressed  liquid over what  would  be expressed at
CD-ROM                             1311- 11                      Revision 0
                                                                 July 1992

-------
      room temperature,  then  allow the sample to warm up to room temperature in
      the device before filtering.

NOTE:         If waste  material  (>1% of the original  sample  weight)  has obviously
             adhered  to  the  container   used  to  transfer the  sample   to  the
             filtration  apparatus,  determine the  weight  of this  residue  and
             subtract  it from the sample  weight determined in Section 7.2.5, to
             determine the weight of the waste sample that will be filtered.

             Gradually apply  vacuum or gentle pressure of 1-10 psi ,  until air or
      pressurizing gas moves  through the filter.   If this point is not  reached
      under 10  psi,  and  if no additional  liquid has  passed  through  the filter in
      any 2 minute interval,  slowly increase the pressure in 10 psi increments
      to a maximum of  50  psi.  After each incremental  increase of  10 psi, if the
      pressurizing gas has not moved  through  the  filter,  and  if no additional
      liquid has passed  through the filter in any 2 minute interval, proceed to
      the  next  10 psi  increment.   When the  pressurizing gas   begins  to  move
      through the filter, or when the liquid  flow  has  ceased  at 50 psi (i.e..
      filtration does not result in  any  additional  filtrate within a  2 minute
      period),  stop the filtration.

NOTE:         Instantaneous application of high pressure can  degrade  the glass
             fiber filter and may cause  premature plugging.

             7.2.8    The material  in the filter holder  is defined as the solid
      phase  of  the  waste, and  the  filtrate  is  defined as the  liquid phase.
      Weigh  the  filtrate.  The liquid  phase may  now be  either  analyzed  (See
      Section 7.2.12)  or stored at  4 °C  until  time of analysis.

NOTE:         Some wastes,  such  as  oily wastes  and some  paint  wastes,  will
             obviously contain  some material that appears to be  a liquid.  Even
             after  applying  vacuum  or   pressure  filtration,   as  outlined  in
             Section 7.2.7, this  material may not filter.  If this is the case,
             the material  within the filtration device is defined  as a solid and
             is carried through the extraction as a solid.  Do  not replace the
             original  filter  with a  fresh filter under  any circumstances.   Use
             only one filter.

             7.2.9    If  the waste  contains <0.5%  dry  solids  (see  Section
      7.1.2), proceed  to  Section 7.2.13.   If the  waste contains  >0.5% dry solids
      (see Section 7.1.1  or 7.1.2), and if particle size reduction of the solid
      was needed in Section  7.1.3,  proceed  to Section 7.2.10.   If the waste as
      received  passes  a  9.5 mm sieve, quantitatively transfer the  solid material
      into  the  extractor bottle along  with the  filter used   to  separate  the
      initial liquid from the solid phase,  and proceed  to Section 7.2.11.

             7.2.10   Prepare the solid  portion  of the  waste for extraction by
      crushing, cutting, or  grinding  the waste  to a surface area or  particle
      size  as  described  in Section  7.1.3.   When the surface area or  particle
      size  has  been  appropriately  altered,  quantitatively transfer  the solid
CD-ROM                             1311- 12                      Revision 0
                                                                 July 1992

-------
      material  into  an  extractor bottle. Include the filter  used  to  separate the
      initial  liquid from the solid phase.

NOTE:         Sieving of  the  waste  is  not  normally  required.    Surface  area
             requirements are  meant  for filamentous  (e.g..  paper,  cloth) and
             similar waste materials.  Actual measurement of surface  area is not
             recommended.   If  sieving  is necessary, a Teflon coated  sieve should
             be used to avoid contamination of the sample.

             7.2.11    Determine the  amount  of extraction  fluid  to  add to the
      extractor vessel  as follows:

                      20 x percent solids (Section 7.1.1)  x weight of waste
                              filtered  (Section 7.2.5 or  7.2.7)
Weight of         =  	
extraction fluid                            100

             Slowly  add this  amount of appropriate extraction fluid  (see Section
      7.1.4)  to the  extractor  vessel.  Close the extractor bottle  tightly (it is
      recommended that Teflon  tape be  used  to ensure a tight seal), secure in
      rotary agitation  device, and  rotate  at 30  ± 2  rpm for   18  +  2 hours.
      Ambient temperature (i.e.. temperature of room in which extraction  takes
      place)  shall be maintained at 23 + 2 °C during the extraction period.

NOTE:         As agitation  continues,  pressure  may  build up  within the extractor
             bottle  for some types of wastes  (e.g. .  limed  or calcium carbonate
             containing waste  may evolve gases  such as carbon  dioxide).   To
             relieve excess pressure, the extractor  bottle may be periodically
             opened  (e.g..  after 15 minutes,  30 minutes,  and 1 hour) and vented
             into a  hood.

             7.2.12    Following  the   18 ±  2  hour  extraction,  separate the
      material   in  the  extractor  vessel  into  its  component liquid  and   solid
      phases  by  filtering  through a  new glass  fiber  filter,   as  outlined  in
      Section  7.2.7.   For  final  filtration of  the TCLP extract,  the glass  fiber
      filter  may be  changed,  if necessary, to  facilitate filtration.  Filter(s)
      shall  be  acid-washed  (see  Section  4.4)  if  evaluating the  mobility  of
      metals.

             7.2.13    Prepare the TCLP extract as follows:

                      7.2.13.1    If   the  waste  contained   no   initial  liquid
             phase,  the filtered liquid material obtained from Section 7.2.12 is
             defined as the TCLP extract.  Proceed to Section 7.2.14.

                      7.2.13.2    If  compatible  (e.g. . multiple phases will not
             result  on  combination),  combine  the filtered liquid  resulting from
             Section 7.2.12  with the  initial  liquid phase of the  waste obtained
             in Section  7.2.7.   This  combined  liquid is  defined  as  the  TCLP
             extract.  Proceed to Section 7.2.14.
CD-ROM                             1311- 13                      Revision 0
                                                                 July 1992

-------
                      7.2.13.3    If  the  initial  liquid phase of the waste, as
             obtained from Section 7.2.7,  is not or may not be compatible with
             the filtered liquid resulting from Section 7.2.12, do not combine
             these liquids.   Analyze  these liquids, collectively defined as the
             TCLP extract,  and combine  the results mathematically, as described
             in Section 7.2.14.

             7.2.14   Following collection of  the TCLP  extract,  the pH of the
      extract should  be  recorded.   Immediately  aliquot and preserve the extract
      for analysis.   Metals  aliquots  must  be  acidified  with nitric  acid to
      pH <2.   If precipitation is observed  upon addition  of  nitric acid to a
      small  aliquot of  the  extract,  then  the  remaining portion of the extract
      for metals analyses  shall  not  be  acidified  and  the  extract  shall  be
      analyzed as  soon  as  possible.   All  other aliquots  must  be stored under
      refrigeration (4 °C)  until  analyzed.   The TCLP extract shall be prepared
      and analyzed according  to appropriate analytical  methods.  TCLP  extracts to
      be analyzed  for metals  shall be  acid  digested  except in those instances
      where  digestion causes loss of  metallic analytes.  If an analysis of the
      undigested  extract shows  that the concentration of any regulated metallic
      analyte exceeds  the  regulatory  level,  then  the waste  is  hazardous  and
      digestion  of the  extract  is  not  necessary.   However, data on undigested
      extracts  alone  cannot  be  used  to  demonstrate  that  the  waste  is  not
      hazardous.    If the  individual  phases  are to  be  analyzed  separately,
      determine  the volume  of the individual  phases  (to  ± 0.5%),  conduct the
      appropriate  analyses,  and  combine the  results  mathematically by using a
      simple volume-weighted average:

                                        (Vj)  (d) + (V2) (C2)
      Final  Analyte Concentration  =   	
                                              Vj  +   V2

      where:

      Vj = The  volume  of  the  first  phase  (L).
      Cj = The concentration of  the  analyte of concern in the  first phase (mg/L).
      V2 = The  volume  of  the  second phase  (L).
      C2 = The  concentration  of  the analyte  of  concern  in  the  second  phase
           (mg/L).

             7.2.15   Compare the  analyte  concentrations  in  the  TCLP extract
      with  the  levels identified  in  the appropriate  regulations.    Refer  to
      Section 8.0 for quality assurance requirements.

      7.3     Procedure When Volatiles are Involved

      Use  the  ZHE  device  to  obtain  TCLP  extract  for  analysis  of  volatile
compounds only.   Extract  resulting  from the use of the ZHE  shall not be used to
evaluate the  mobility  of  nonvolatile analytes (e.g. . metals,  pesticides, etc.).

      The  ZHE device has  approximately a 500 mL internal capacity.   The ZHE can
thus accommodate a  maximum of 25 grams of solid (defined as  that fraction of a

CD-ROM                             1311-  14                       Revision 0
                                                                 July 1992

-------
sample from which  no  additional  liquid may be forced out by an applied pressure
of 50 psi),  due to  the  need to add an amount  of  extraction  fluid  equal  to 20
times the weight of the solid phase.

      Charge the ZHE  with  sample only once and  do not open the device until the
final extract (of  the  solid) has been collected.  Repeated  filling of the ZHE to
obtain 25 grams of solid is not permitted.

      Do not  allow  the  waste,  the initial liquid  phase,  or  the  extract  to be
exposed to the atmosphere for any more time than is absolutely necessary.   Any
manipulation of these  materials  should be done when cold (4 °C) to minimize loss
of volatiles.

             7.3.1    Pre-weigh the  (evacuated)  filtrate  collection container
      (See  Section  4.6)  and set  aside.   If using  a  TEDLAR®  bag,  express all
      liquid from  the  ZHE  device into the  bag,  whether for the initial  or final
      liquid/solid separation,  and  take  an  aliquot from the  liquid in the bag
      for analysis.   The  containers  listed  in  Section 4.6 are recommended for
      use under the conditions stated in Sections 4.6.1 - 4.6.3.

             7.3.2    Place the ZHE  piston within the body of  the  ZHE (it may be
      helpful  first  to moisten  the piston  0-rings slightly with extraction
      fluid).   Adjust the  piston  within  the  ZHE  body  to a  height  that  will
      minimize the distance  the  piston will have to move once  the  ZHE is charged
      with sample  (based upon sample size requirements determined from Section
      7.3,  Section  7.1.1  and/or  7.1.2).   Secure the  gas  inlet/outlet  flange
      (bottom flange) onto  the  ZHE  body  in  accordance with the  manufacturer's
      instructions.   Secure the glass fiber filter between the support screens
      and set aside.   Set liquid  inlet/outlet  flange (top flange) aside.

             7.3.3    If the waste is  100% solid  (see  Section  7.1.1), weigh out
      a subsample  (25  gram maximum)  of the waste, record weight,  and proceed to
      Section 7.3.5.

             7.3.4    If the waste contains <  0.5% dry solids (Section 7.1.2),
      the  liquid  portion  of waste,  after  filtration,  is defined  as  the  TCLP
      extract.   Filter  enough  of the  sample  so  that the amount  of  filtered
      liquid will  support  all  of  the volatile analyses  required.   For  wastes
      containing  1  0.5%  dry  solids  (Sections 7.1.1  and/or  7.1.2),  use  the
      percent  solids  information  obtained  in  Section 7.1.1  to  determine the
      optimum sample size to charge into the ZHE.  The recommended sample size
      is as follows:

                      7.3.4.1     For wastes containing < 5% solids  (see Section
             7.1.1),  weigh  out a  500 gram  subsample of waste and record the
             wei ght.

                      7.3.4.2     For wastes containing > 5% solids  (see Section
             7.1.1),   determine  the  amount of waste to  charge into the  ZHE as
             fol1ows:
CD-ROM                             1311- 15                      Revision 0
                                                                 July 1992

-------
                                            25
Weight of waste to charge ZHE =  	   x 100
                                  percent solids (Section 7.1.1)

             Weigh out  a  subsample of  the  waste  of the  appropriate  size and
             record the weight.

             7.3.5    If particle  size  reduction  of the  solid  portion  of the
      waste  was  required  in  Section  7.1.3,  proceed  to Section  7.3.6.   If
      particle size  reduction  was not  required  in Section  7.1.3,  proceed to
      Section 7.3.7.

             7.3.6    Prepare the waste  for  extraction by  crushing, cutting, or
      grinding the  solid  portion  of the waste to  a  surface area or particle size
      as described  in  Section 7.1.3.  Wastes  and  appropriate  reduction equipment
      should  be  refrigerated,   if  possible,  to  4  °C  prior to  particle size
      reduction.    The means used  to  effect particle size  reduction  must not
      generate heat in and  of  itself.   If  reduction of  the  solid phase of the
      waste  is necessary,  exposure of  the  waste  to the  atmosphere  should be
      avoided to  the  extent possible.

NOTE:         Sieving  of the  waste  is not recommended due  to the possibility that
             volatiles may  be lost.  The use of an  appropriately  graduated  ruler
             is  recommended  as  an   acceptable   alternative.    Surface  area
             requirements are  meant  for filamentous (e.g. .  paper,  cloth) and
             similar  waste  materials.  Actual measurement of  surface area  is not
             recommended.

             When  the  surface   area  or  particle  size   has  been appropriately
      altered, proceed to Section 7.3.7.

             7.3.7    Waste slurries  need  not  be allowed  to  stand to permit the
      solid phase to  settle.  Do not centrifuge  wastes  prior to filtration.

             7.3.8    Quantitatively  transfer  the  entire sample  (liquid and
      solid phases) quickly to  the ZHE.   Secure  the filter and support screens
      onto the top flange  of the device and secure the  top  flange  to  the ZHE
      body in accordance  with the  manufacturer's  instructions.  Tighten all ZHE
      fittings and place the device in the vertical position (gas inlet/outlet
      flange on the bottom).  Do not attach the extract  collection device  to the
      top plate.

NOTE:         If waste  material  (>1%  of  original  sample  weight)  has  obviously
             adhered  to the  container used  to transfer  the  sample  to the ZHE,
             determine the  weight of this  residue and  subtract it  from the
             sample weight determined in Section 7.3.4  to determine the weight
             of the waste sample that will  be filtered.

             Attach a  gas  line to the gas  inlet/outlet  valve  (bottom flange)
      and, with the liquid  inlet/outlet valve  (top flange) open,  begin applying
      gentle pressure  of  1-10 psi  (or  more if necessary)  to force all  headspace

CD-ROM                             1311- 16                      Revision 0
                                                                 July 1992

-------
      slowly out  of  the ZHE device  into  a  hood.   At the  first  appearance  of
      liquid from the  liquid  inlet/outlet valve,  quickly  close  the  valve  and
      discontinue pressure.   If filtration of  the  waste at 4  °C  reduces  the
      amount of expressed liquid over what would be expressed at room tempera-
      ture, then allow the sample to warm up to room temperature in the device
      before filtering.   If  the  waste is 100% solid (see  Section 7.1.1), slowly
      increase  the  pressure  to a  maximum of  50 psi  to  force  most of  the
      headspace out  of the device and proceed to Section 7.3.12.

             7.3.9    Attach  the  evacuated  pre-weighed  filtrate  collection
      container  to the  liquid  inlet/outlet valve  and  open the  valve.   Begin
      applying  gentle  pressure  of  1-10 psi to  force the  liquid phase  of  the
      sample into the filtrate  collection  container.  If no additional  liquid
      has passed through the  filter  in  any  2 minute  interval,  slowly increase
      the pressure in  10 psi  increments  to  a  maximum  of 50 psi.   After  each
      incremental  increase of  10 psi, if no additional liquid  has passed through
      the filter in any  2 minute interval,  proceed  to  the  next 10 psi increment.
      When liquid flow  has  ceased such that continued pressure filtration at  50
      psi does not result in any additional  filtrate within a 2 minute period,
      stop the  filtration.   Close  the  liquid  inlet/outlet  valve,  discontinue
      pressure to the piston,  and disconnect and weigh  the filtrate collection
      contai ner.

NOTE:         Instantaneous application  of high  pressure  can  degrade  the  glass
             fiber filter and  may cause premature  plugging.

             7.3.10    The material  in the ZHE is defined as the solid phase  of
      the waste and  the filtrate is defined as  the liquid phase.

NOTE:         Some wastes,  such  as  oily  wastes  and some  paint wastes,  will
             obviously  contain some  material that appears to be  a liquid.   Even
             after applying  pressure  filtration, this material will  not filter.
             If this  is the case, the material  within the filtration device  is
             defined  as  a solid  and  is  carried through the TCLP  extraction  as a
             solid.

             If  the  original  waste  contained  <0.5% dry solids (see  Section
      7.1.2),  this filtrate is  defined as  the TCLP  extract and  is  analyzed
      directly.  Proceed to Section 7.3.15.

             7.3.11    The liquid phase  may  now  be  either analyzed  immediately
      (See  Sections  7.3.13 through  7.3.15)  or stored  at 4 °C  under  minimal
      headspace  conditions  until  time   of  analysis.   Determine the  weight  of
      extraction fluid #1 to add to the ZHE as  follows:

                               20 x percent solids (Section 7.1.1)  x weight
                                of  waste filtered  (Section 7.3.4 or 7.3.8)
Weight of extraction  fluid =  	
                                                 100
CD-ROM                             1311- 17                      Revision  0
                                                                 July  1992

-------
             7.3.12   The following Sections detail  how to add the appropriate
      amount  of  extraction  fluid  to the  solid material  within  the  ZHE  and
      agitation of the  ZHE  vessel.   Extraction fluid #1  is  used  in  all  cases
      (See Section 5.7).

                      7.3.12.1   With the  ZHE  in the vertical position, attach
             a line  from the  extraction  fluid  reservoir to  the liquid  in-
             let/outlet  valve.   The  line used  shall  contain fresh  extraction
             fluid and  should  be preflushed with  fluid to  eliminate  any  air
             pockets  in  the  line.   Release  gas  pressure on the ZHE piston (from
             the  gas inlet/outlet valve),  open the  liquid inlet/outlet valve,
             and   begin  transferring  extraction  fluid  (by pumping or  similar
             means) into the ZHE.  Continue  pumping  extraction  fluid into  the
             ZHE  until the  appropriate amount of fluid has been introduced into
             the  device.

                      7.3.12.2   After  the  extraction  fluid has been added,
             immediately close  the  liquid inlet/outlet valve  and disconnect  the
             extraction  fluid line.  Check the ZHE to  ensure that all  valves  are
             in their  closed  positions.    Manually  rotate  the  device  in  an
             end-over-end fashion  2  or  3  times.    Reposition  the ZHE  in  the
             vertical  position with  the  liquid  inlet/outlet  valve  on  top.
             Pressurize  the ZHE to  5-10 psi (if necessary) and slowly open  the
             liquid inlet/outlet  valve to bleed  out any headspace (into a hood)
             that  may have  been  introduced due to the  addition  of  extraction
             fluid.  This bleeding shall be done quickly  and shall be stopped at
             the  first appearance of  liquid from the valve.   Re-pressurize  the
             ZHE  with  5-10 psi and check all  ZHE  fittings to ensure  that they
             are  closed.

                      7.3.12.3   Place  the  ZHE  in the rotary agitation appara-
             tus  (if  it  is  not  already there) and rotate at 30 ± 2 rpm for 18 ±
             2 hours.   Ambient temperature  (i.e.. temperature of room in which
             extraction  occurs) shall  be maintained at 23 ±  2 °C during agita-
             tion.

             7.3.13   Following  the  18  ±  2  hour agitation  period,  check  the
      pressure behind the  ZHE piston by quickly opening and closing  the  gas
      inlet/outlet valve  and  noting the  escape  of gas.  If the pressure has  not
      been maintained  (i.e.. no  gas  release  observed),  the  device is leaking.
      Check the ZHE for  leaking as  specified in Section 4.2.1, and perform  the
      extraction  again with a new sample of waste.   If the pressure  within  the
      device has  been  maintained, the material  in the extractor vessel  is once
      again separated  into  its  component  liquid and solid phases.   If  the waste
      contained an initial  liquid  phase, the  liquid may  be  filtered directly
      into the same filtrate collection  container  (i.e.. TEDLAR® bag) holding the
      initial   liquid  phase  of  the  waste.    A  separate filtrate  collection
      container must be  used  if combining would  create multiple phases, or there
      is  not  enough  volume  left  within the   filtrate  collection  container.
      Filter through the  glass  fiber filter, using the ZHE  device as  discussed
      in  Section  7.3.9.   All  extract  shall  be filtered  and  collected  if  the

CD-ROM                             1311- 18                      Revision 0
                                                                 July 1992

-------
      TEDLAR® bag  is  used,  if  the  extract  is multiphasic,  or if  the  waste
      contained an initial liquid phase (see Sections 4.6 and 7.3.1).

NOTE:        An in-line glass fiber  filter  may be  used  to filter the material
             within the ZHE if it is suspected that the glass fiber filter has
             been  ruptured.

             7.3.14   If the original waste contained no initial liquid phase,
      the filtered liquid  material  obtained  from  Section  7.3.13 is defined as
      the TCLP  extract.   If  the waste contained an  initial  liquid phase, the
      filtered  liquid  material  obtained  from  Section 7.3.13  and  the initial
      liquid phase (Section 7.3.9) are collectively  defined as  the TCLP extract.

             7.3.15   Following  collection  of  the  TCLP  extract,  immediately
      prepare the extract  for  analysis and store with  minimal  headspace at 4 °C
      until   analyzed.  Analyze the  TCLP  extract according  to the  appropriate
      analytical  methods.    If  the  individual  phases  are   to   be  analyzed
      separately  (i.e..  are  not   miscible),   determine   the  volume  of  the
      individual phases  (to 0.5%), conduct the appropriate  analyses, and combine
      the results mathematically by using a simple volume-weighted average:
      Final Analyte    =   	
      Concentration

      where:

      Vj  =  The  volume  of  the  first  phases  (L).
      Cj  =  The concentration of the  analyte of concern in the first  phase  (mg/L).
      V2  =  The  volume  of  the  second phase  (L).
      C2  =  The  concentration  of the analyte  of  concern  in  the  second phase
           (mg/L).

             7.3.16   Compare  the  analyte  concentrations   in  the  TCLP  extract
      with  the  levels identified  in  the appropriate  regulations.    Refer to
      Section 8.0 for quality assurance requirements.

8.0  QUALITY ASSURANCE

      8.1    A minimum of one  blank  (using the same extraction  fluid  as used for
the samples)  must be  analyzed  for every 20 extractions that have been conducted
in an extraction vessel .

      8.2    A  matrix spike  shall  be performed  for  each waste  type  (e.g. .
wastewater  treatment  sludge,  contaminated soil,  etc.) unless the result  exceeds
the regulatory  level  and  the  data are being used solely to  demonstrate that the
waste property  exceeds the  regulatory level.   A  minimum  of  one  matrix  spike must
be analyzed  for  each  analytical  batch.   As  a minimum,  follow the matrix spike
addition guidance provided in each analytical  method.
CD-ROM                             1311- 19                      Revision 0
                                                                 July 1992

-------
             8.2.1    Matrix spikes  are to be added after filtration  of the TCLP
      extract and before  preservation.  Matrix spikes should not be added prior
      to TCLP extraction of the sample.

             8.2.2    In  most  cases,  matrix  spikes  should  be   added  at  a
      concentration equivalent to the corresponding  regulatory level.   If the
      analyte concentration  is  less than one  half the  regulatory  level,  the
      spike concentration may be as  low  as  one  half  of the analyte concentra-
      tion, but  may not  be  not  less  than  five times the method detection limit.
      In order to avoid differences  in matrix effects,  the matrix spikes must be
      added  to  the same  nominal  volume of TCLP  extract  as  that which  was
      analyzed for the unspiked sample.

             8.2.3    The  purpose   of  the  matrix  spike  is   to  monitor  the
      performance  of  the  analytical  methods  used,  and to  determine  whether
      matrix  interferences  exist.   Use of  other  internal calibration  methods,
      modification of the  analytical  methods,  or  use  of alternate analytical
      methods may be needed to accurately measure the analyte concentration in
      the  TCLP  extract when  the  recovery  of  the matrix  spike  is below  the
      expected analytical  method performance.

             8.2.4    Matrix spike  recoveries are  calculated  by the following
      formul a :

      %R (%Recovery) = 100  (Xs  - XJ/K
      Xs  =  measured  value  for  the  spiked  sample,
      Xu  =  measured  value  for  the  unspiked  sample,  and
      K = known value of the spike in the sample.
      8.3    All  quality  control measures described in the appropriate  analytical
methods shall  be followed.

      8.4    The  use  of  internal  calibration  quantisation  methods  shall  be
employed for a metallic contaminant if:   (1) Recovery of the contaminant from the
TCLP extract  is  not at  least  50% and  the  concentration does  not  exceed the
regulatory level, and (2) The concentration of the contaminant measured in the
extract is within 20% of the appropriate regulatory level.

             8.4.1.    The method of standard  additions  shall be  employed as the
      internal  calibration quantisation method for each metallic contaminant.

             8.4.2    The  method  of  standard   additions  requires  preparing
      calibration standards in  the sample matrix rather than  reagent water or
      blank solution.    It requires  taking  four  identical  aliquots  of  the
      solution  and adding known  amounts  of standard to three of these aliquots.
      The forth aliquot is the unknown.  Preferably,  the first addition should
      be prepared so that  the  resulting concentration  is approximately 50% of
      the expected concentration  of the  sample.   The second and third additions
      should be prepared so that the concentrations are approximately 100% and

CD-ROM                             1311- 20                      Revision 0
                                                                 July 1992

-------
      150% of the expected concentration of the sample.  All four aliquots are
      maintained at  the  same  final  volume by  adding  reagent  water  or a blank
      solution,  and may need dilution  adjustment to maintain the signals in the
      linear range  of the  instrument technique.  All  four aliquots  are  analyzed.

             8.4.3    Prepare a plot, or subject data to linear regression, of
      instrument signals or external -cal ibration-derived concentrations as the
      dependant  variable  (y-axis)  versus  concentrations of  the  additions of
      standard as the independent  variable (x-axis).   Solve  for  the intercept of
      the abscissa  (the independent  variable, x-axis)  which  is  the  concentration
      in the unknown.

             8.4.4    Alternately, subtract the instrumental  signal or  external-
      calibration-derived concentration  of  the unknown  (unspiked) sample  from
      the instrumental  signals or  external-calibration-derived  concentrations of
      the  standard  additions.    Plot  or subject  to  linear regression  of the
      corrected  instrument signals  or external-calibration-derived  concentra-
      tions as the dependant variable versus the independent variable.  Derive
      concentrations  for unknowns  using  the internal  calibration curve as if it
      were an external calibration  curve.
      8.5
periods:
             Samples  must  undergo TCLP  extraction  within  the  following time
SAMPLE MAXIMUM HOLDING TIMES [DAYS]




V o 1 a t i 1 e s
Semi -vol ati les
Mercury
Metals, except
mercury
From:
Field
col lection
To:
TCLP
extraction
14
14
28
180

From:
TCLP
extraction
To:
Preparative
extraction
NA
7
NA
NA

From:
Preparative
extraction
To:
Determi nati ve
analysis
14
40
28
180



Total
el apsed
time
28
61
56
360

NA = Not appli cable


If sample  holding  times  are exceeded, the  values  obtained  will  be considered
minimal  concentrations.    Exceeding  the  holding  time  is  not  acceptable  in
establishing that a waste does not exceed the regulatory level.  Exceeding the
holding  time  will  not  invalidate characterization  if the waste  exceeds the
regulatory 1evel .
CD-ROM
                                   1311- 21
Revision 0
July 1992

-------
9.0   METHOD PERFORMANCE

      9.1    Ruggedness.   Two  ruggedness studies  have been performed to  determine
the effect of various perturbations on specific elements of the TCLP  protocol.
Ruggedness testing  determines the sensitivity  of  small  procedural variations
which might be expected to occur during routine laboratory application.

             9.1.1    Metals - The following conditions were used  when  leaching
      a waste for metals analysis:
Varying Conditions
Liquid/Solid ratio
Extraction time
Headspace
Buffer n acidity
Acid-washed filters
Filter type
Bottle type
19:1 vs. 21:1
16 hours vs. 18 hours
20% vs. 60%
190 meq vs. 210 meq
yes vs. no
0.7 urn glass fiber vs. 0.45 urn
vs. polycarbonate
borosilicate vs. flint glass
             Of the seven method variations  examined,  acidity of the extraction
      fluid had the greatest impact on the  results.   Four of 13 metals from an
      API  separator  sludge/electroplating  waste  (API/EW)  mixture  and  two of
      three metals from  an  ammonia lime still  bottom waste  were  extracted at
      higher levels by  the more  acidic buffer.  Because of the sensitivity  to pH
      changes, the method  requires that the  extraction  fluids  be prepared so
      that the final  pH is within + 0.05 units as specified.

             9.1.2    Volatile Organic Compounds - The following  conditions  were
      used when leaching a waste for VOC analysis:
Varying Conditions
Liquid/Solid ratio
Headspace
Buffer #1 acidity
Method of storing extract
Al i quotti ng
Pressure behind piston
19:1 vs. 21:1
0% vs. 5%
60 meq vs. 80 meq
Syringe vs. Tedlar®
bag
yes vs. no
Opsi vs. 20psi
CD-ROM
1311-  22
Revision 0
July 1992

-------
             None of the parameters had a significant effect on the results of
      the ruggedness test.

      9.2    Precision.   Many  TCLP  precision (reproducibi1ity)  studies  have  been
performed,  and  have  shown  that,  in  general,  the precision  of  the  TCLP is
comparable to or exceeds  that  of  the EP toxicity test and that method  precision
is adequate.  One of the more significant contributions to poor  precision  appears
to be related to sample homogeneity and inter-laboratory variation  (due to the
nature of waste materials).

             9.2.1    Metals - The  results of a multi-1aboratory study  are shown
      in Table  6,  and indicate  that  a  single analysis of  a  waste  may not be
      adequate for waste characterization and identification requirements.

             9.2.2    Semi -Volati1e  Organic  Compounds  -   The  results  of  two
      studies are  shown  in  Tables 7  and  8.  Single  laboratory  precision was
      excellent with  greater  than  90  percent  of the results exhibiting an RSD
      less than  25 percent.   Over 85 percent of  all  individual  compounds in the
      multi-1aboratory study  fell  in  the  RSD  range of  20  - 120 percent.   Both
      studies concluded  that  the  TCLP  provides adequate  precision.   It was  also
      determined that the high acetate content of  the extraction fluid  did not
      present problems (i.e.. column degradation  of the gas chromatograph) for
      the analytical conditions  used.

             9.2.3    Volatile   Organic   Compounds   -   Eleven   laboratories
      participated in  a  collaborative  study of the use of the ZHE with  two waste
      types  which  were  fortified with a  mixture  of  VOCs.   The results of the
      collaborative study  are  shown in Table 9.  Precision  results  for  VOCs  tend
      to  occur  over a  considerable  range.    However,  the  range  and  mean RSD
      compared very closely to the same  collaborative  study metals results in
      Table 6.   Blackburn and Show concluded that  at the 95% level  of signifi-
      cance:  1) recoveries  among  laboratories were statistically similar,  2)
      recoveries did not  vary  significantly between the  two  sample  types, and 3)
      each laboratory showed  the same  pattern  of  recovery  for each of  the two
      samples.

10.0  REFERENCES

1.    Blackburn,  W.B.  and  Show,  I.    "Collaborative   Study  of  the  Toxicity
Characteristics  Leaching Procedure  (TCLP)."  Draft  Final   Report, Contract  No. 68-
03-1958, S-Cubed, November 1986.

2.    Newcomer,  L.R.,  Blackburn,  W.B.,   Kimmell,  T.A.   "Performance  of  the
Toxicity Characteristic  Leaching  Procedure."  Wilson Laboratories,  S-Cubed,  U.S.
EPA, December 1986.

3.    Williams,  L.R.,  Francis, C.W.; Maskarinec,  M.P., Taylor D.R.,  and Rothman,
N.  "Sing!e-Laboratory Evaluation of Mobility Procedure  for  Solid Waste."  EMSL,
ORNL, S-Cubed,  ENSECO.
CD-ROM                             1311- 23                      Revision 0
                                                                 July 1992

-------
                                   Table 1.
                              Volatile Analytes1'2
Compound                                                      CAS No.
Acetone                                                       67-64-1
Benzene                                                       71-43-2
n-Butyl alcohol                                               71-36-3
Carbon disulfide                                              75-15-0
Carbon tetrachloride                                          56-23-5
Chlorobenzene                                                108-90-7
Chloroform                                                    67-66-3
1,2-Dichloroethane                                           107-06-2
1,1-Dichloroethylene                                          75-35-4
Ethyl  acetate                                                141-78-6
Ethyl  benzene                                                100-41-4
Ethyl  ether                                                   60-29-7
Isobutanol                                                    78-83-1
Methanol                                                       67-56-1
Methylene chloride                                            75-09-2
Methyl ethyl ketone                                           78-93-3
Methyl isobutyl ketone                                       108-10-1
Tetrachloroethylene                                          127-18-4
Toluene                                                      108-88-3
1,1,1,-Trichloroethane                                        71-55-6
Trichloroethylene                                             79-01-6
Trichlorof1uoromethane                                        75-69-4
l,l,2-Trichloro-l,2,2-trifluoroethane                         76-13-1
Vinyl  chloride                                                75-01-4
Xylene                                                      1330-20-7
1  When  testing  for  any  or  all  of  these  analytes,  the zero-headspace
  extractor vessel  shall be used instead of the bottle extractor.

2  Benzene,  carbon tetrachloride,  chlorobenzene,  chloroform,
  1,2-dichloroethane ,  1,1-dichloroethylene ,  methyl ethyl ketone,
  tetrachloroethylene,  and vinyl  chloride are toxicity characteristic
  consti tuents.
CD-ROM                             1311- 24                      Revision 0
                                                                 July 1992

-------
                                   Table 2.
                     Suitable Rotary Agitation Apparatus1
Company
Analytical Testing and
Consulting Services,
Inc.

Associated Design and
Manufacturing Company




Locat
Warn'
(215


Al exa
(703)




i on
ngton, PA
) 343-4490


ndria, VA
549-5999




Model No.
4-
8-
12-
24-
2-
4-
6-
8-
12-
24-
vessel
vessel
vessel
vessel
vessel
vessel
vessel
vessel
vessel
vessel
extractor (DC20S
extractor (DC20)
extractor (DC20B
extractor (DC24C
(3740-2-BRE)
(3740-4-BRE)
(3740-6-BRE)
(3740-8-BRE)
(3740-12-BRE)
(3740-24-BRE)
)

)
)






Environmental  Machine and
  Design, Inc.

IRA Machine Shop and
  Laboratory

Lars Lande Manufacturing
Mi 11i pore Corp.
Lynchburg, VA
(804) 845-6424

Santurce, PR
(809) 752-4004
8-vessel  (08-00-00)
4-vessel  (04-00-00)

8-vessel  (011001)
Whitmore Lake, MI 10-vessel  (10VRE)
(313) 449-4H6     5-vessel  (5VRE)
                   6-vessel  (6VRE)
Bedford, MA
(800) 225-3384
4-ZHE or
4 2-liter bottle
    extractor (YT310RAHW)
1  Any  device  that  rotates  the  extraction  vessel  in  an  end-over-end  fashion  at
30+2 rpm is acceptable.
CD-ROM
   1311- 25
              Revision 0
              July 1992

-------
                                   Table 3.
                  Suitable Zero-Headspace  Extractor Vessels1
Company
 Locati on
Model No.
Analytical  Testing &
  Consulting Services, Inc.

Associated Design and
  Manufacturing Company

Lars Lande Manufacturing2
Mi 11i pore Corporati on
Warrington,  PA
(215) 343-4490

Alexandria,  VA
(703) 549-5999

Whitmore Lake, MI
(313) 449-4116

Bedford, MA
(800) 225-3384
C102, Mechanical
Pressure Device

3745-ZHE, Gas
Pressure Device

ZHE-11,  Gas
Pressure Device

YT30090HW, Gas
Pressure Device
Environmental Machine
and Design, Inc.

Gelman Science
 Lynchburg, VA
 (804) 845-6424

 Ann Arbor, MI
 (800) 521-1520
VOLA-TOX1, Gas
Pressure Device

15400 Gas Pressure
Devi ce
1  Any  device  that  meets  the  specifications  listed  in  Section  4.2.1  of  the
method is suitable.

2  This  device uses  a  110 mm  filter.
CD-ROM
    1311-  26
          Revision 0
          July 1992

-------
                                   Table 4.
                           Suitable Filter Holders1
                                                 Model/
Company                        Location          Catalogue  No.      Size
Nucleopore Corporation
Pleasanton, CA
(800) 882-7711
425910
410400
142 mm
47 mm
Micro Filtration             Dublin,  CA         302400            142 mm
Systems                      (800)  334-7132      311400            47 mm
                             (415)  828-6010

Millipore Corporation        Bedford,  MA        YT30142HW         142 mm
                             (800)  225-3384      XX1004700         47 mm
1  Any  device  capable  of  separating  the  liquid  from  the  solid  phase  of  the
waste is suitable, providing that it is chemically  compatible with  the waste
and the constituents  to be analyzed.  Plastic devices (not listed above)  may
be used when only inorganic analytes are of concern.   The 142 mm size  filter
holder is recommended.
CD-ROM                             1311- 27                      Revision  0
                                                                 July  1992

-------
                                   Table 5.
                            Suitable Filter Media1
Company
Mi 1 1 i pore Corporati on
Nucleopore Corporation
Whatman Laboratory
Products, Inc.
Micro Filtration
Systems
Gelman Science
Location
Bedford, MA
(800) 225-3384
Pleasanton, CA
(415) 463-2530
Clifton, NJ
(201) 773-5800
Dublin, CA
(800) 334-7132
(415) 828-6010
Ann Arbor, MI
(800) 521-1520
Model
AP40
211625
GFF
GF75
66256 (90mm)
66257 (142mm)
Pore
Size
(um)
0.7
0.7
0.7
0.7
0.7
1  Any  filter  that  meets  the  specifications  in  Section  4.4  of  the  Method  is
suitable.
CD-ROM
1311-  28
Revision 0
July 1992

-------
              Table 6.  Multi - Laboratory TCLP Metals,  Precision

Waste
Ammoni a
Lime Sti 1 1
Bottoms



API/EW
Mixture




Fossi 1
Fuel Fly
Ash



Extraction
Fluid
#1
n
#1
n
#1
n
#1
n
#1
n
#1
n
n
n
#1
#2
#1
#2

Metal
Cadmi urn

Chromi urn

Lead

Cadmi urn

Chromi urn

Lead

Cadmi urn

Chromi urn

Lead


X
0.053
0.023
0.015
0.0032
0.0030
0.0032
0.0046
0.0005
0.0561
0.105
0.0031
0.0124
0.080
0.093
0.017
0.070
0.0087
0.0457

S
0.031
0.017
0.0014
0.0037
0.0027
0.0028
0.0028
0.0004
0.0227
0.018
0.0031
0.0136
0.069
0.067
0.014
0.040
0.0074
0.0083

%RSD
60
76
93
118
90
87
61
77
40
17
100
110
86
72
85
57
85
18
%RSD Range = 17 - 118
Mean %RSD = 74
NOTE:  X = Mean results from 6
      Units = mg/L
      Extraction Fluid #1 = pH 4.9
                       #2 = pH 2.9
12 different laboratories
CD-ROM
   1311-  29
Revision 0
July 1992

-------
              Table  7.   Single-Laboratory  Semi -Volati1es,  Precision

Waste
Ammoni a
Lime Stil 1
Bottoms





















API/EW
Mixture







Compound
Phenol

2-Methyl phenol

4-Methyl phenol

2 ,4-Dimethyl phenol

Naphthalene

2-Methyl naphthal ene

Di benzof uran

Acenaphthyl ene

Fl uorene

Phenanthrene

Anthracene

Fl uoranthrene

Phenol

2 ,4-Dimethyl phenol

Naphthalene

2-Methyl naphthal ene

Extraction
Fluid
#1
n
#1
n
#1
n
#1
n
#1
n
#1
n
#1
n
#1
#2
#1
#2
#1
#2
#1
#2
#1
n
#1
n
#1
#2
#1
#2
#1
#2

X
19000
19400
2000
1860
7940
7490
321
307
3920
3827
290
273
187
187
703
663
151
156
241
243
33.2
34.6
25.3
26.0
40.7
19.0
33.0
43.3
185
165
265
200

S
2230
929
297
52.9
1380
200
46.8
45.8
413
176
44.8
19.3
22.7
7.2
89.2
20.1
17.6
2.1
22.7
7.9
6.19
1.55
1.8
1.8
13.5
1.76
9.35
8.61
29.4
24.8
61.2
18.9

%RSD
11.6
4.8
14.9
2.8
17.4
2.7
14.6
14.9
10.5
4.6
15.5
7.1
12.1
3.9
12.7
3.0
11.7
1.3
9.4
3.3
18.6
4.5
7.1
7.1
33.0
9.3
28.3
19.9
15.8
15.0
23.1
9.5
%RSD Range =1-33
Mean %RSD = 12
NOTE:  Units = ug/L
      Extractions were performed in triplicate
      All results were at least 2x the detection limit
      Extraction Fluid #1 = pH 4.9
                       #2 = pH 2.9
CD-ROM
1311-  30
Revision 0
July 1992

-------
              Table 8.  Multi- Laboratory Semi -Volati1es , Precision

Waste
Ammonia Lime
Still Bottoms (A)
API/EW
Mixture (B)
Fossil Fuel
Fly Ash (C)

Compound
BNAs

BNAs

BNAs

Extraction
Fluid
#1
n
#1
n
#1
n

X
10043
10376
1624
2074
750
739

S
7680
6552
675
1463
175
342

%RSD
76.5
63.1
41.6
70.5
23.4
46.3
Mean %RSD = 54
NOTE:  JJm'ts = ug/L
      X = Mean results
      Extraction Fluid
      from
      #1 =
      n =
       3 -
       pH
       PH
 10
4.9
2.9
labs
      %RSD Range
        A, #1
        A, #2
        B, #1
        B, #2
        C, #1
        C, #2
for
Individual
        0
        28
        20
        49
        36
        61
 Compounds
 -  113
 -  108
 -  156
 -  128
 -  143
 -  164
CD-ROM
                  1311-  31
                                            Revision  0
                                            July  1992

-------
              Table 9.  Multi- Laboratory (11 Labs) VOCs, Precision
Waste
Mine
Tai 1 i ngs

















Ammoni a
Lime Sti 1 1
Bottoms
















Compound
Vinyl chloride
Methylene chloride
Carbon disulfide
1 , 1-Di chl oroethene
1 , 1-Di chl oroethane
Chl oroform
1 , 2-Di chl oroethane
2-Butanone
1,1, 1-Tri chl oroethane
Carbon tetrachl ori de
Tri chl oroethene
1,1, 2-Tri chl oroethene
Benzene
1,1,2,2-Tetrachloroethane
Tol uene
Chl orobenzene
Ethyl benzene
Trichlorofl uoromethane
Acryl oni tri 1 e
Vinyl chloride
Methylene chloride
Carbon disulfide
1 , 1-Di chl oroethene
1 , 1-Di chl oroethane
Chl oroform
1 , 2-Di chl oroethane
2-Butanone
1,1, 1-Tri chl oroethane
Carbon tetrachl ori de
Tri chl oroethene
1,1, 2-Tri chl oroethene
Benzene
1,1,2,2-Tetrachloroethane
Tol uene
Chl orobenzene
Ethyl benzene
Trichlorofl uoromethane
Acryl oni tri 1 e
X
6.36
12.1
5.57
21.9
31.4
46.6
47.8
43.5
20.9
12.0
24.7
19.6
37.9
34.9
29.3
35.6
4.27
3.82
76.7
5.00
14.3
3.37
52.1
52.8
64.7
43.1
59.0
53.6
7.10
57.3
6.7
61.3
3.16
69.0
71.8
3.70
4.05
29.4
S
6.36
11.8
2.83
27.7
25.4
29.2
33.6
36.9
20.9
8.2
21.2
10.9
28.7
25.6
11.2
19.3
2.80
4.40
110.8
4.71
13.1
2.07
38.8
25.6
28.4
31.5
39.6
40.9
6.1
34.2
4.7
26.8
2.1
18.5
12.0
2.2
4.8
34.8
%RSD
100
98
51
127
81
63
70
85
100
68
86
56
76
73
38
54
66
115
144
94
92
61
75
49
44
73
67
76
86
60
70
44
66
27
17
58
119
118
%RSD Range = 17 - 144
Mean %RSD = 75
NOTE:  Units = ug/L
CD-ROM
1311-  32
Revision 0
July 1992

-------
              Motor
           (30± 2 rpm)
  Extraction Vessel Holder
                       Figure 1.  Rotary  Agitation Apparatus

                                       Liquid Inlet/Outlet Valve
                    Top Flange

                Support Screen-
                            Filter-
                     Support Screen'
                        Viton o-rings
                Bottom Flange —*£
                  Pressurized Gas •
                  Inlet/Outlet Valve
                                             .Sample
                                              Piston       c
                                               Gas
              Pressure
               Gauge
                     Figure 2.  Zero-Headspace Extractor  (ZHE)
CD-ROM
1311- 33
Revision 0
July 1992

-------
                                     METHOD 1311

                    TOXICITY  CHARACTERISTIC  LEACHATE PROCEDURE
                Sepa rate
               1iqui ds f r om
              solids with 0.6
              - 0.8 um glass
               fi ber filter
                 Discard
                 solids
                                      START
                                      Use a
                                   sub-aample of
                                      was t e
                   Sepa rate
                  11qui da f r om
                 solids with 0.6
                  0.8 um glass
                  fiber filter
                                                             Solid
                                    Extract w/
                                 appropriate fluid
                                 1 )  Bottle extractor
                                 for non-volatiles
                                 2) ZHE device for
                                    volatiles
                    Reduce
                  particle size
                   to <9.5 mm
CD-ROM
1311- 34
Revision  0
July 1992

-------
                             METHOD  1311  (CONTINUED)

                   TOXICITY  CHARACTERISTIC  LEACHATE PROCEDURE

Di s ca r d
solids
Solid



Separate
extract f r om
solids w/ 0.6 -
0 8 urn glass
fiber filter
Li qui d

Store
at
1
4
1 qui d
C

Yes
C omb i ne
extract w/
1 iquid phase
o f was t e


Ana 1 y ze
liquid


                                                              Measure amount of
                                                             liquid and analyze
                                                               [mathematically
                                                              combine result w/
                                                              result of extract
                                                                 analysis)
                                               STOP
CD-ROM
1311-  35
Revision 0
July 1992

-------
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/625/R-95/005
July 1996
Pump-and-Treat
Ground-Water Remediation
A Guide for Decision Makers
and Practitioners

-------
                                                               EPA/625/R-95/OQ5
                                                                      July, 1996
A          for
                           Prepared by
                    Eastern Research Group, Inc.
                        110 Hartwell Avenue
                          Lexington, MA
                U.S. Environmental Protection Agency
                 Office of Research and Development
            National  Risk Management Research Laboratory
            Center for Environmental Research Information
                          Cincinnati, Ohio

-------
                                              Notice
                               The information in this document has been
                           funded wholly, or in part, by the U.S. Environmental
                           Protection Agency (EPA). This document has been
                           subjected to EPA's peer and administrative review
                           and has been approved for publication as an EPA
                           document. Mention of trade names or commercial
                           products docs not constitute endorsement or recom-
                           mendation for use.
   In September, 1995 the Office of Research and Development completed a reorganization of its Laboratories and
Centers. The former Risk Reduction Engineering Laboratory located in Cincinnati, Ohio, the Robert S. Kerr
Research Laboratory located in Ada, Oklahoma, the Air and Energy Research Laboratory, located in Research
Triangle Park, North Carolina, and the Center for Environmental Research Information located in Cincinnati, Ohio,
were merged into the National Risk Management Research Laboratory. No physical relocations were involved. The
documents referenced in this guide were published prior to the reorganization: therefore former laboratory/center
names are shown as they were at the time of publication.

-------
                                           Foreword
     The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land,
air, and water resources. Under a mandate of national environmental laws, the Agency strives to formulate
and implement actions leading to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA's research program is providing data and
technical support for solving environmental problems today and building a science knowledge base neces-
sary to manage our ecological resources wisely, understand how pollutants affect our health, and prevent or
reduce environmental risks in the future.

       The National Risk Management Research Laboratory is  the Agency's center for investigation of
technological and management approaches for reducing risks from threats to human health and the environ-
ment. The focus of the Laboratory's research program is on methods for the prevention and control of
pollution to air, land, water and subsurface resources; protection of water quality in public water systems;
remediation of contaminated sites and ground water; and  prevention and control of indoor air pollution. The
goal  of this research effort is to catalyze development and implementation of innovative, cost-effective
environmental technologies; develop  scientific and engineering information needed by EPA to support
regulatory and policy decisions; and provide technical support and information transfer to ensure effective
implementation of environmental regulations and strategies.

       This publication has been produced as part of the Laboratory's strategic long-term research plan. It is
published and made available by EPA's Office of Research and Development to assist the user community
and to link researchers with their clients.

                                                 E.  Timothy Oppelt, Director
                                                 National  Risk Management Research Laboratory
                                               iii

-------
IV

-------
                                   Contents



List of Figures	v

List of Tables	xi

Acronyms and Abbreviations	xii

Acknowledgments	xiii

1.  Introduction to Pump-and-Treat Remediation	 1

2.  Appropriate Use of Pump-and-Treat Technology	3

3.  Smart Pump-and-Treat Techniques	5
   3.1.   Contaminant Removal/Control	5
   3.2.   Thorough Site Characterization	5
   3.3.   Dynamic Management of the Well Extraction Field	7
   3.4.   Realistic Cleanup Goals 	 10

4.  Anticipating Tailing and Rebound Problems 	 19
   4.1.   Effects of Tailing and Rebound on Remediation Efforts	 19
   4.2.   Contributing Factors	 19
         4.2.1.   Non-Aqueous Phase Liquids (NAPLs)	 19
         4.2.2.   Contaminant Desorption	20
         4.2.3.   Precipitate Dissolution	23
         4.2.4.   Matrix Diffusion	24
         4.2.5.   Ground-Water Velocity Variation 	24
   4.3.   Assessing the Significance of Tailing and Rebound at a Site	26

-------
                                        (continued)
5.  Effective Hydraulic Containment	28
   5.1.  Ground-Water Barriers and Flow Control	28
         5.1.1.  Horizontal and Vertical Capture Zones	28
         5.1.2.  Pressure Ridge Systems	31
         5.1.3.  Physical Barriers	31
   5.2.  Hydraulic Containment: Other Special Considerations	32
         5.2.1.  Effects of Anisotropy	32
         5.2.2.  Drawdown Limitations	32
         5.2.3.  Stagnation Zones	33

6.  Pump-and-Treat System Design and Operation	36
   6.1.  Capture Zone Analysis and Optimization Modeling	36
   6.2.  Efficient Pumping Operations	40
   6.3.  Treating Contaminated Ground Water	42
   6.4.  Monitoring Performance 	43
         6.4.1.  Hydraulic Head Monitoring for Containment	43
         6.4.2.  Ground-Water Quality Monitoring for Containment	44
         6.4.3.  Aquifer Restoration Monitoring	48
   6.5.  Evaluating Restoration Success and Closure	50

7.  Variations and Alternatives to Conventional  Pump-and-Treat Methods	53
   7.1.  Alternative Methods for Fluid Deliver}- and Recovery	53
   7.2.  Vadose Zone Source Control	54
   7.3.  Physical and Chemical Enhancements	54
         7.3.1.  Physical Enhancements	59
         7.3.2.  Chemical Enhancements	59
   7.4.  Biological Enhancements	59
   7.5.  Alternatives to the Pump-and-Treat Approach	61
         7.5.1.  Intrinsic Bioremediation	63
         7.5.2.  In Situ Reactive Barriers	63
                                        VI

-------
                                       (continued)


8.  References	66
9.  EPA Publications Providing Further Information	70
List of Sidebars
       1    Changing Expectations for the Pump-and-Treat Approach	2
       2    Major Types of Hydrogeologic Settings	 11
       3    Computer Graphics as a Site Characterization Tool	 13
       4    The Effect of NAPL Phases on Ground-Water Contamination	 16
       5    Computer Model ing of Well Patterns Versus Hydrogeologic Conditions	38
                                       vn

-------
                                     Figures
1.   Examples of hydraulic containment in a plan view and cross section using
    pump-and-trcat technology: (a) pump well, (b) drain, and (c) well within a
    barrier wall system (after Cohen et al., 1994)	
2.   Contaminant plumes as a function of density and miscibility with ground water:
    (a) light liquids (gasoline and methanol) create contaminant plumes that tend to
    flow in the upper portions of an aquifer; (b) dense liquids (perchloroethylene
    [PCEJ and ethylene glycol) create a plume that contaminates the full thickness
    of an aquifer (adapted from Gorclick ct al.. 1993)	6

3.   This hollow-stem auger is fitted with a 5-foot sampling tube that collects a
    continuous core as the auger advances, allowing detailed and accurate
    observation of subsurface lithology. When drilling is completed, a monitoring
    well also can be installed	8

4.   Hydraulic or vibratory direct-push rigs can be installed on vans, small trucks,
    all-terrain vehicles, or trailers and allow collection of continuous soil cores and
    depth-specific ground-water samples for detailed subsurface mapping if
    contaminants are generally confined to depths of less than 15 meters.
    (Photo courtesy of Gcoprobc Systems.)	9

5.   Conceptual diagram of Dense Non-Aqueous Phase Liquid (DNAPL) Trichloro-
    ethylene (TCE) based on soil and ground-water sampling in a heterogeneous sand
    and gravel  aquifer. The extreme difficulty in cleaning up this site, which includes
    five distinct forms of TCE (vapors and residual product in the vadose zone; pooled.
    residual, and dissolved product in the ground water) led to modification of the
    pump-and-treat system for hydraulic containment rather than restoration
    (adapted from Clausen and Solomon, 1994)	 10

                                        viii

-------
                                       (continued)
6.  GEOS computer screen showing organic contaminant plume in relation
   to subsurface stratigraphy	 12

7.  EPA's SITE3D software, under development at the Ada, Oklahoma,
   laboratory, helps visualize in three dimensions a TCE contaminant plume at a
   Superfund Site. Yellows and reds indicate zone with highest concentrations of
   TCE in ground water. 	 14

8.  Dark NAPL (Soltrol) and water in a homogenous micromodcl after (a) the
   displacement of water by NAPL and then (b) the displacement of NAPL
   by water, with NAPL at residual saturation (Wilson et al, 1990)	 17

9.  Photomicrographs of (a) a single blob occupying one pore body, and
   (b) a doublet blob occupying two pore bodies and a pore throat
   (Wilson etal., 1990)	"	 18

10. Concentration versus pumping duration or volume showing
   tailing and rebound effects (Cohen et al., 1994)	20

11. Contaminants are mobilized when ground water that is undersaturated
   with a contaminant comes in contact with a NAPL (a) or contaminant sorbed
   on an organic carbon or mineral surface (b). High ground-w7ater velocities
   and short contact times will result in low contaminant concentrations, and
   low velocities and long contact times will result in high contaminant
   concentrations (c) (adapted from Gorelick et al., 1993)	21

12. Laboratory model of the transport of DNAPL contaminant through
   an aquifer with varying permeability; note the concentration of
   downward movement in fingers and the DNAPL pools above the low-
   permeability zones (the horizontal discs).
   (Source: U.S. EPA National Risk Management Research Laboratory.)	22

13. Dissolved contaminant concentration in ground water pumped from a
   recovery well versus time in a formation that contains a solid-phase
   contaminant precipitate  (Palme rand Fish, 1992)	23

                                       ix

-------
                                        (continued)
14. Changes in average relative trichloroethene (TCE) concentrations
    in clay lenses of varying thickness as a function of time (NRC, 1994)	24

15. Tailing resulting from ground-water velocity variations: (a) horizontal
    variations in the velocity of ground water moving toward a pumping well
    (Keely. 1989) lead to (b) tailing as higher concentrations of ground water
    in slower pathlines mix with lower concentrations in faster pathlines (Palmer
    and Fish, 1992); (c) in a stratified sand and gravel aquifer,  tailing occurs
    at tl when  clean water from the upper gravel strata mixes with still-
    contaminated ground water in the lower sand strata (Cohen ct al., 1994)	25

16. Zone of residuals created in former cone of depression after cessation of
    LNAPL recovery system (Gorelick et al., 1993)	27

17. Plan view of a mixed containment-restoration strategy. A pump-and-treat
    system is used with barrier walls to contain the ground-water contamination
    source areas (e.g., where NAPL or waste may be present) and then collect
    and treat the dissolved contaminant plume (Cohen etal., 1994)	29

18. In an isotropic aquifer, ground-water flow7 lines (b) are perpendicular
    to hydraulic head contours (a). Pumping causes drawdowns and anew
    steady-state potentiometric surface within the well's zone of influence
    (c). Following the modified hydraulic gradients, ground water within
    the shaded capture zone flows to the pumping well (d).
    (Cohen etal., 1994, adapted from Gorelick etal.,  1993)	30

19. Cross section showing equipotential contours and the vertical  capture zone
    associated  with ground-water withdrawal from a partially penetrating well
    in isotropic media (Cohen etal.,  1994)	31
20. Effect of fracture anisotropy on the orientation of the zone of contribution
    (capture zone) to a pumping well (Bradbury etal.. 1991)	33

-------
                                       (continued)
21.  Capture zone simulation of three pumping wells for an isotropic aquifer
    (a) and anisotropy ratio of 10:1 (b) using the EPA Well Head Protection Area
    (WHPA)code. .."	34

22.  Examples of stagnation zones (shaded where ground-water velocity is less than
    4 L/T): (a) single pumping well and (b) four extraction wells with an injection
    well in the center (Cohen et ah, 1994)	35

23.  Major types of pumping/injection well patterns (Satkin and Bedient, 1988)	39

24.  Ground-water flow line in the vicinity of conceptual pumping centers at
    Lawrence Livermore National Laboratory  superimposed on an
    isoconcentration contour map and showing areas of potential stagnation
    (Cohen et ah, 1994, after Hoffinan, 1993)	41

25.  Effect of adaptive pumping on cleanup time at Lawrence Livermore
    National Laboratory Superfund site (Cohen et ah, 1994, after Hoffman. 1993)	42

26.  The pulsed pumping concept (Cohen et ah, 1994, after Keely, 1989)	43

27.  Nested piezometer hydrograph for 1992 at the Chem-Dyne Superfund site
    (Cohen ct ah, 1994. after Papadopulos & Associates, 1993)	46

28.  Ground-water flow between  and beyond the extraction wells, resulting
    even though hydraulic heads throughout the mapped aquifer are higher
    than the pumping level (Cohen et ah,  1994)	47

29.  Example display of ground-water flow directions and hydraulic
    gradients determined between three observation wells (Cohen et ah, 1994)	48

30.  Influent and effluent VOC concentrations (mg/L) at the Chem-Dyne
    treatment plant from 1987 to 1992
    (Cohen et ah, 1994, after Papadopulos & Associates, 1993)	49

31.  Cumulative mass of VOCs removed from the aquifer at the Chem-Dyne
    site from 1987 to 1992 (Cohen ct ah, 1994, after Papadopulos & Associates, 1993)	50

                                       xi

-------
                                        (continued)
32. Determining the success and/or timeliness of closure of a pump-and-treat
    system (Cohen et al., 1994)	51

33. Stages of remediation in relation to example contaminant concentrations
    in a well at a pump-and-treat site (U.S. EPA, 1992)	52

34. Some applications of horizontal wells: (a) intersecting flat-lying layers,
    (b) intercepting plume elongated by regional gradient, (c) intersecting
    vertical fractures, and (d) access beneath structures (U.S. EPA, 1994)	55

35. Two approaches using trenches or horizontal wells to intercept
    contaminant plumes (U.S. EPA, 1994)	58

36. Process diagram for air sparging with (a) vertical wells, and (b) horizontal wells
    [after National Research Council (NRC), 1994]	60

37. Schematic of chemical enhancement of a pump-and-treat system. Key areas of
    concern are shown in boxes. In some cases, the reactive agent will be recovered
    and reused (Palmer and Fish, 1992)	61

38. Two types of aerobic in situ bioremediation systems: (a) injection well with sparger.
    (b) infiltration gallery (Sims etal., 1992, after Thomas and Ward, 1989)	62

39. Alternative ground-water plume management options: (a) pump-and-treat system,
    (b) intrinsic bioremediation, (c) in situ reaction curtain, (d) funnel-and-gate system
    (adapted from Starr and Cherry, 1994)	64

40. Funncl-and-gatc configurations (Starr and Cherry, 1994)	65
                                        Xll

-------
                                     Tables



1.   Categories of Sites for Technical Infeasibility Determinations (NRC, 1994)	 15

2.   Data Requirements for Pump-and-Treat Systems (Adapted from U.S. EPA, 1991)	37

3.   Applicability' of Treatment Technologies to Contaminated
    Ground Water (U.S. EPA, 1991)	45

4.   Issues Affecting Application of Alternative Methods for Delivery or Recovery
    (U.S. EPA, 1994)	56
                                       Xlll

-------

ACL
CZAEM
DNAPL
DOD/ETTC
EPA
GAEP
LNAPL
LLNL
MCL
MCLG
NAPL
NRC
NTIS
ORD
OSWER
PCB
PCE
RCRA
SVE
TCE
TI
voc
WHPA
Alternate concentration limit
Capture Zone Analytic Element Model
Dense non-aqueous phase liquid
Department of Defense Environmental Technology Transfer Committee
Environmental Protection Agency
Geographic Analytic Element Preprocessor
Light non-aqueous phase liquid
Lawrence Livcrmorc National Laboratory
Maximum contaminant level
Maximum contaminant level goal
Non-aqueous phase liquid
National Research Council
National Technical Information Service
Office of Research and Development
Office of Solid Waste and Emergency Response
Polychlorinatcd biphcnyl
Perchloroethylene
Resource Conservation and Recovery Act
Soil vapor extraction
Trichloroethylene
Technical impracticality
Volatile organic compound
Well Head Protection Area
                                      xiv

-------

   This document was prepared under Contract No.
68-C3-0315, Work Assignment No. 1-33, by Eastern
Research Group. Inc. (ERG), and under the sponsor-
ship of the U.S. Environmental Protection Agency.
                     xv

-------
troduction to Pump-and-Treat Remediatio
         Pump-and-treat is one of the most widely used
       ground-water remediation technologies. Conven-
       tional pump-and-treat methods involve pumping
       contaminated water to the surface for treatment.
       This guide, however, uses the term pump and treat
       in a broad sense to include any system where
       withdrawal from or injection into ground water is
       part of a remediation strategy. Variations and
       enhancements of conventional pump and treat
       include hydraulic fracturing as well as chemical
       and biological enhancements. The pump-and-treat
       remediation approach is used at about three-
       quarters of the Superfund sites where ground
       water is contaminated and at most sites where
       cleanup is required by the Resource Conservation
       and Recovery Act (RCRA) and state laws [Na-
       tional Research Council (NRC),  1994]. Although
       the effectiveness of pump-and-treat systems has
       been called into question (Sidebar 1), after two
       decades of use, this approach remains a necessary
       component of most ground-water remediation
       efforts and is appropriate for both restoration and
       plume containment.
         This guide provides an introduction to pump-
       and-treat ground-water remediation by addressing
       the following questions:
         • When is pump and treat an appropriate
          remediation approach?
  • What is involved in "smart" application of
   the pump-and-treat approach?
  • What are tailing and rebound, and how can
   they be anticipated?
  • What are the recommended methods for
   meeting the challenges of effective hydraulic
   containment?
  • How can the design and operation of a pump-
   and-treat system be optimized and its perfor-
   mance measured?
  • When should variations and alternatives to
   conventional pump-and-treat methods be
   used?
  By presenting the basic concepts of pump-and-
treat technology, this guide provides decision-
makers with a foundation for evaluating the
appropriateness of conventional or innovative
approaches. An in-depth understanding of
hydrogeology and ground-water engineering is
required, however, to design and operate a pump-
and-treat system for ground-water remediation.
Readers seeking more information on specific
topics covered in this booklet should refer to the
U.S. Environmental Protection Agency (EPA)
documents listed at the end of this guide (Section
9).

-------
                                                Sidebar 1
               Changing Expectations for the Pump-and-Treat Approach
  Pump-and-treat systems for remediating ground water
came into wide use in the early to mid-1980s. By the
early 1990s, evaluations by EPA (Keely, 1989; U.S.
EPA, 1989; Haley et al, 1991) and others (Freeze and
Cherry, 1989; Mackay and Cherry, 1989) called into the
question the performance of pump-and-treat systems.
The general "failure" of the pump-and-treat approach
was identified as its inability to achieve "restoration"
(i.e., reduction of contaminants to levels required by
health-based standards) in 5 to 10 years, as anticipated
in the design phase of projects. Although a variety  of
factors contributed to this shortcoming, tailing and
rebound (Section 4) represented the major barrier to
achieving remediation goals. Pump-and-treat systems
were criticized more pointedly by Travis and Doty
(1990), who asserted as a "simple fact" that "contami-
nated aquifers cannot be restored through pumping and
treating."


  ixpectations for the effectiveness of pump-and-treat
technology, however, may have been too high. Ground-
water scientists and engineers generally agree that
complete aquifer restoration is an unrealistic goal for
many, if not most, contaminated sites. Nonetheless,
further experience with pump-and-treat systems
indicates that full restoration at some sites with
relatively simple characteristics is possible; moreover,;
many sites, full restoration of ground-water quality can
be achieved for part of a site (NRC, 1994). For
example, Bartow and Davenport (1995), in a review of
37 applications of pump-and-treat systems in Santa
Clara Valley, California, found that one site had
achieved maximum contaminant levels (MCLs) for all
contaminants and about one-third achieved, or were
near, MCLs for one or more parameters. Bartow and
Davenport's conclusion that pump-and-treat systems had
significantly reduced the mass of volatile organic
contaminants (VOCs) in the region's ground water
indicates how expectations regarding the technology
have changed.


  Combining the pump-and-treat approach with in situ
bioremediation (see Section 7.4) provides further
opportunities for improving the effectiveness of ground-
water cleanup. For example, Marquis (1995) suggested
that in situ bioremediation used with the pump-and-treat
approach should always be considered as an option for
remediation of sand and gravel aquifers contaminated
with biodegradable organic  compounds, especially
volatile aromatic and polyaromatic hydrocarbons.

-------
  riate
  Pump-and-treat systems are used primarily to
accomplish the following:
  • Hydraulic containment. To control the
   movement of contaminated ground water,
   preventing the continued expansion of the
   contaminated zone. Figure 1  illustrates three
   major configurations for accomplishing
   hydraulic containment: (1) a pumping well
   alone, (2) a subsurface drain  combined with a
   pump well, and (3) a well within a barrier
   wall system.
  • Treatment. To reduce the dissolved contami-
   nant concentrations in ground water suffi-
    ciently that the aquifer complies with cleanup
    standards or the treated water withdrawn from
    the aquifer can be put to beneficial use.
  Although hydraulic containment and cleanup
can represent separate goals, more typically,
remediation efforts are undertaken to achieve a
combination of both. For example, if restoration is
not feasible, the primary objective might be
containment. In contrast, where a contaminated
well is used for drinking water but the contami-
nant source has not been identified, treatment at
the wellhead might allow continued use of the
water even though the aquifer remains contami-
nated.

-------
                                                                     (a)
                                                                      (b)
Figure 1.
Examples of hydraulic
containment in a plan
view and cross section
using pump-and-treat
technology: (a) pump well,
(b) drain, and (c) well
within a barrier wall
system (after Cohen et
al., 1994)
                                                                      (c)
                                                                                      Upgradient
                                                                                     Barrier Wall
Plume
                           JDowngradient
                             Barrier Wall
                             Capture
                             Zone Limit

-------
  A fundamental component of any ground-water
remediation effort using the pump-and-treat
approach is contaminant removal or control. Thus,
effective remediation of ground water using
pump-and-treat technology requires knowledge of
contaminants and site characteristics. Addition-
ally, the remediation plan should call for imple-
mentation of dynamic system management based
on a statement of realistic objectives (Hoffman,
1993).
3.1.    Contaminant Removal/Control
  Any ground-water cleanup effort will be
undermined unless inorganic and organic con-
taminant sources are identified, located, and
eliminated, or at least controlled, to prevent
further contamination of the aquifer. Toxic
inorganic substances may serve as a continuing
source of contamination through mechanisms
such as dissolution and desorption. At many
contaminated sites, organic liquids are a major
contributor to ground-water contamination.
Figure 2 illustrates four common types of con-
taminant plumes, each characterized by the
liquid's density relative to water and the degree to
which the  liquid mixes with water. Even when the
organic liquid resides exclusively in the vadose
zone (i.e., the area between the ground surface
and the water table) it can serve as a source of
ground-water contamination. In such situations,
contamination occurs when percolating water
comes in contact with the liquid (sometimes
called product) or its vapors and carries dissolved
material to the ground water. Vapors also might
migrate to the water table and contaminate ground
water without infiltration.
  Source removal is the most effective way to
prevent further contamination. Where inorganic or
organic contaminants are confined to the vadose
zone, removal is usually the preferred option.
When removal is not feasible, as is often the case
with dense non-aqueous phase liquids (DNAPLs)
residing below the water table, containment is an
essential initial step in remediation. In some
situations, containment  can be achieved through
capping, which prevents or reduces infiltration of
rainfall through the contaminated soil. Capping
can be ineffective if water table fluctuations occur
within the zone of contamination or when NAPL
vapors are present.
3.2.   Thorough Site  Characterization
  Comprehensive characterization of the contami-
nated site serves two major functions:
   • Accurately assessing the types, extent, and
    forms of contamination in the subsurface
    increases the likelihood of achieving treat-
    ment goals. This requires an understanding of
    the physical phases  in which contaminants
    exist (mainly sorbed and aqueous phases for
    inorganic contaminants, and sorbed, NAPL,
    aqueous, and gaseous phases for organic
    liquids) and quantification of the distribution
    between the phases. Indeed, inadequate site
    characterization has undermined some pump-

-------
                                                                    Immiscible
                                                                                            Ground Surface
Figure 2.
Contaminant plumes
as a function of
density and
miscibility with
ground water:
(a) light liquids
(gasoline and
methanol) create
contaminant plumes
that tend to flow in
the upper portions of
an aquifer; (b) dense
liquids (perchloro-
ethylene [PCE] and
ethylene glycol)
create a plume that
contaminates the full
thickness of an
aquifer (adapted
from Gorelick et al.,
1993).
(b)
                                   Miscible
                                                          Ground Surface
Soil
                          Methanol
                          Kii
i
                                                           Ethylene Glycol
      Ground Water

-------
    and-treat efforts; for instance, when after a
    few years greater quantities of contaminants
    had been removed than were identified in the
    initial site assessment.
   « A thorough, three-dimensional characteriza-
    tion of subsurface soils and hydrogeology,
    including particle-size distribution, sorption
    characteristics, and hydraulic conductivity',
    provides a firm basis for appropriate place-
    ment of pump-and-treat wells. Such informa-
    tion is also required for evaluating the extent
    to which tailing and rebound may present
    problems at a site (Section 4).
  Three-dimensional characterization techniques
include primarily indirect observations, using
surface and borehole geophysical instalments and
cone-penetration measurements, and direct
sampling of soil and ground water. Important
advances in soil sampling technology have been
made relatively recently, such as continuous
samplers used with a hollow-stem auger (Figure
3) and  smaller continuous-core, direct-push
equipment that also can be used to collect ground-
water samples without installing wells (Figure 4).
Vibrator}- drilling methods are another innovative
technique for collecting soil cores and ground-
water samples. Additionally, sensitive borehole
flow meters that allow measurement of vertical
changes in hydraulic conductivity in a borehole
represent an important recent development. These
techniques allow subsurface mapping to be
generated with a level of detail that generally
would be prohibitively expensive using conven-
tional drilling and sampling methods. Figure 5
presents a conceptual diagram  oftrichloroethene
(TCE)  contamination at a complex site
(Sidebar 2) developed from extensive use of
direct-push sampling techniques.
  Moreover, if sufficient data are obtained, the
interpretation of subsurface data can be enhanced
greatly by performing two- and three-dimensional
computer modeling of the subsurface. Figure 6
shows a conceptual model of a site developed by
combining contour visualization of a contaminant
plume of benzene with subsurface lithologic logs
(Sidebar 3). EPA's SITE3D software, being
developed by the National Risk Management
Laboratory's Subsurface Protection and Remedia-
tion Division, allows three-dimensional visualiza-
tion of contaminant plumes (Figure 7). Statistical
software developed by EPA such as Geo-EAS and
GEOPACK for geostatistical analysis and con-
touring of ground-water contaminant data and
GRITS/STAT for analysis of contaminant concen-
trations are among the many computer-based tools
available for analyzing subsurface data.
3.3.    Dynamic                 of the
       Well Extraction Field
  To be effective, pump-and-treat efforts must go
beyond initial site characterization, using infor-
mation gathered after remediation operations are
under way to manage the well extraction field
dynamically. For instance, information collected
while drilling and installing extraction wells,
operating pumping wells, and tracking changes in
water levels in monitoring wells (Section 6.4) and
contaminant concentrations in observation wells
can refine the portrayal  of the site.
  Dynamic management of the well extraction
field based on more comprehensive information
can provide both economic and environmental
benefits. In general, additional information about
the site and the pump-and-treat effort allows
operators to make informed decisions about the
efficient use  of remediation resources. More
specifically, this flexible site management ap-
proach may facilitate greater success in hydraulic
containment (Section 5). Ultimately, the time
required to achieve cleanup goals might be
minimized.

-------
Figures.
This hollow-stem
auger is fitted
with a 5-foot
sampling tube
that collects a
continuous core
as the auger
advances,
allowing detailed
and accurate
observation of
subsurface
lithology. When
drilling is
completed, a
monitoring well
also can be
installed.

-------
Figure 4.  Hydraulic or vibratory direct-push rigs can be installed on vans, small trucks, all-terrain vehicles, or
          trailers and allow collection of continuous soil cores and depth-specific ground-water samples for
          detailed subsurface mapping if contaminants are generally confined to depths of less than 15 meters.
          (Photo courtesy of Geoprobe Systems.)
  A key component of the dynamic management
approach is the effective design and operation of
the pump-and-treat system. The following tech-
niques can be useful in this regard:
   • Using capture zone analysis, optimization
    modeling, and data obtained from monitoring
    the effects of initial extraction wells to
    identify the best locations for wells (Section
    6.1).
• Phasing the construction of extraction and
 monitoring wells so that information obtained
 from operation of the initial wells informs
 decisions about siting subsequent wells
 (Section 6.2).
1 Phasing pumping rates and the operation of
 individual wells to enhance containment,
 avoid stagnation zones (Section 5.2.3), and

-------
        w= Water Table

         WEST
TCE Vapors
                 Area of Residual DNAPL
                 Above Water Table
                         Area of Residual DNAPL
                         Below Water Table  \
Figure 5.  Conceptual diagram of DNAPL (TCE) based on soil and ground-water sampling in a heterogeneous
          sand and gravel aquifer. The extreme difficulty in cleaning up this site, which includes five distinct forms
          of TCE (vapors and residual product in the vadose zone; pooled, residual, and dissolved product in the
          ground water) led to modification of the pump-and-treat system for hydraulic containment rather than
          restoration (adapted from Clausen and Solomon, 1994).
    ensure removal of the most contaminated
    ground water first (Section 6.3).
3.4.    Realistic Cleanup Goals
  Unrealistic expectations for the pump-and-treat
approach can lead to disappointments in system
performance (Sidebar 1). Indeed, a cleanup goal
that is realistic for one site may not be reasonable
elsewhere. The Committee on Ground Water
Cleanup Alternatives of the National Academy of
     Sciences (NRC, 1994) has identified three major
     classes of sites based on hydrogeology (Sidebar
     2) and contaminant chemistry (see Table 1):
       • Class A. Sites where full cleanup to health-
         based standards should be feasible using
         current technology. Such sites include homo-
         geneous single- and multiple-layer aquifers
         involving mobile, dissolved contaminants.
                                       10

-------

  The Committee on Ground Water Cleanup Alterna-
tives of the National Academy of Sciences has defined
three major hydrogeologic settings for evaluating the
technical feasibility of ground-water cleanup based on
the degree of uniformity of the aquifer material and
layering (Table 1).
 • Homogeneous aquifers consist of materials that do
  not vary significantly in their water-transmitting
  properties. Contaminant movement in homogeneous
  aquifers is largely a function of the hydraulic
  conductivity of the aquifer. For example, a homoge-
  neous aquifer might comprise permeable, well-sorted
  sands or gravels.
  Heterogeneous aquifers consist of materials that vary
  in their water-transmitting properties laterally,
  vertically, or in both directions. Contaminants in
  heterogeneous aquifers move preferentially in the
  high-permeability zones, resulting in more rapid
  transport than would be expected based on the
  average hydraulic conductivity of the aquifer. A sand
  aquifer with lenses of silts and clays is an example of
     leterogeneous aquifer.
   •  Fractured aquifers typically consist of
     low-permeability rock where most ground-water
     flow is in joints and fractures.
    Single-layered aquifers are less complex than
  multiple-layered aquifers, which are separated by
  less-permeable strata, because the possibility of cross
  contamination between aquifers by either upward or
  downward movement becomes a consideration. Note
  that the term aquifer is used in this guide in a broad
  sense to include any area within the saturated zone
  where the presence of ground-water contamination is of
  sufficient concern to require remediation. Figure 2
  illustrates a homogenous, single-layer aquifer.
  Figure 15c (described in Section 4.2.5) illustrates a
  two-layered homogenous aquifer. Figure 5 illustrates a
  multiple-layered heterogeneous aquifer. The challenge
  of ground-water cleanup increases along with aquifer
  complexity  because of difficulties in delineating
  contaminant sources and pathways and the increased
  likelihood of tailing and  rebound effects (Section 4).
1 Class B. Sites where the technical feasibility
 of complete cleanup is likely to be uncertain.
 This class includes a wide range of hydrogeo-
 logic settings and contaminant types that do
 not fall into classes A or C.
1 Class C. Sites where full cleanup of the
 source areas to health-based standards is not
 likely to be technically feasible. Such sites
 include fractured-rock aquifers contaminated
    by free-product light nonaqueous phase
    liquids (LNAPL) or DNAPL and single- or
    multiple-layered heterogeneous aquifers
    contaminated by a free-product DNAPL
    (Sidebar 4).
  Typically, preliminary ground-water cleanup
efforts at contaminated sites are focused on
standards established for drinking water, such as
federal or state maximum contaminant levels
                                         11

-------
                     £ILE   SITE  MAP   CONTOURS   X-S£EimM5
                                                                   CROSS SECTION
Figures.
GEOS computer
screen showing
organic
contaminant
plume in relation
to subsurface
stratigraphy (see
text for
discussion).
                                                      A  550
                                                                                             FILLxTOPSOIL
                                                                                           afVIDY/'SILTY CLflY
parameter
 Benzene
                 (MCLs) or nonzero MCL goals (MCLGs). EPA
                 has established procedures, however, by which
                 efforts can target alternative goals at Superfund
                 and RCRA sites using alternate concentration
                 limits (ACLs) where ground-water discharges into
                 nearby surface water (U.S. EPA,  1988) or demon-
                 strating the technical impracticality (TI) of
                 ground-water cleanup (U.S. EPA, 1993; Feldman
                 and Campbell, 1994). At DNAPL sites where the
                 TI of ground-water cleanup has been demon-
                 strated, the remedial strategy might call for
        removal of as much of the DNAPL as is feasible,
        containment of the remaining DNAPL, and
        treatment of the aqueous contaminant plume
        outside the containment area. Consequently, even
        at Class C and Class B sites where restoration is
        not feasible, application of some form of the
        pump-and-treat approach may be required either
        to help contain the contaminant source and
        aqueous-phase plume or to clean up the contami-
        nated ground water outside the containment area.
                                                      12

-------
  Computer visualization can help focus attention on
the types of additional information needed when
characterizing a site before initiating a remediation
effort. For example, in Figure 6 the contoured benzene
data, collected from the sand/gravel # 1 aquifer (yellow
in the cross section in the upper right) shows two areas
of high concentration (i.e., MW-5 and MW-7). Does this
represent a contaminant plume from a single source, or
does it indicate two contaminant plumes from separate
sources? An examination of the cross section (upper
right of center) indicates that sand/gravel # 1 aquifer at
MW-7 has the lower "high" concentration of benzene,
suggesting that the two plumes might be related.
  The cross section also shows, however, that the
aquifer is quite thin between the two monitoring wells.
Indeed, examination of monitoring well MP-10, which
lies near the cross section (see lower center log) reveals
that the aquifer is missing at this point. This suggests
the absence of a concentration gradient between the two
wells, indicating that two different sources may be
involved.
  Further analysis of the spatial distribution of the sand/
gravel # 1 aquifer, including flow directions as indicated
by potentiometric heads, and possibly additional
sampling for benzene would be required to determine if
one or two sources are contributing to the contamina-
tion.
  This particular example also cautions against relying
exclusively on computer-generated interpolations,
which can suggest  features that are not actually present
(i.e., continuity of the aquifer between MW-5 and
MW-7).
                                             13

-------
Figure 7.  EPA's SITE3D software, under development at the Ada, Oklahoma, laboratory, helps visualize in three-
          dimensions a TCE contaminant plume at a Superfund Site. Yellows and reds indicate zone with highest
          concentrations of TCE in ground water.
                                         14

-------
Table 1. Categories of Sites for Technical Infeasibility Determinations (NRC, 1994)
Contaminant Chemistry

Hydrogeology
Homogeneous,
single layer
Homogeneous,
multiple layers
Heterogeneous,
single layer
Heterogeneous,
multiple layers
Fractured

Mobile
Dissolved
(degrades/
volatilizes)
A
(1)
A
(1)
B
(2)
B
(2)
B
(3)

Mobile,
Dissolved
A
(1-2)
A
(1-2)
B
(2)
B
(2)
B
(3)
Strongly
Sorbed,
Dissolved
(degrades/
volatilizes)
B
(2)
B
(2)
B
(3)
B
(3)
B
(3)

Strongly
Sorbed,
Dissolved
B
(2-3)
B
(2-3)
B
(3)
B
(3)
B
(3)

Separate
Phase
LNAPL
B
(2-3)
B
(2-3)
B
(3)
B
(3)
C
(4)

Separate
Phase
DNAPL
B
(3)
B
(3)
C
(4)
C
(4)
C
(4)
Note:  Shaded boxes at the left end (group A) represent types of sites for which cleanup of the full site to health-based standards should be feasible with
       current technology.  Shaded boxes at the right end (group C) represent types of sites for which full cleanup of the source areas to health-based
       standards will likely be technically infeasible. The unshaded boxes in the middle (group B) represent sites for which the technical feasibility of
       complete cleanup is likely to be uncertain. The numerical ratings indicate the relative ease of cleanup, where 1 is easiest and 4 is most difficult.
                                                                   15

-------
  The light (LNAPLs) and dense (DNAPLs) immiscible
nonaqueous phase liquids shown in Figure 2 pose the
most difficult problems for ground-water cleanup
because of their complex interactions with water and
solids in the subsurface. Figure 5 illustrates these
complexities when trichloroethene (TCE), a DNAPL has
moved through a heterogeneous, multiple-layered
alluvial aquifer (at a site in Tennessee). Four distinct
forms, or phases, of TCE are evident:
 •  The NAPL emits a vapor phase in the unsaturated
   zone that moves by diffusion. DNAPL vapors tend to
   sink until they reach impermeable layers (Figure 5) or
   the water table. Even if the NAPL does not reach the
   ground water, contamination can occur by dissolution
   of the vapors directly into the ground water or by
   water percolating through the unsaturated zone.
 •  Residual NAPL remains after the free product has
   moved through the subsurface by gravity or been
   displaced by water (Figure 8a and b). Residual NAPL
   exists as single- to  complex-shaped blobs that fill
pore spaces (Figure 8b and 9). The amount of
residual NAPL remaining in the subsurface depends
on the subsurface material and the type of NAPL.
Residual saturation in the unsaturated zone typically
ranges from 10 to 20 percent of the subsurface
volume and in the saturated zone generally ranges
from 10 to 50 percent (Cohen and Mercer, 1993).
Figure 5 differentiates residual TCE above and
below the water table.
Free product exists where most of the pore space is
filled by the NAPL. It accumulates wherever a
barrier prevents downward movement. LNAPLs,
such as gasoline, tend to float on top of the water
table, whereas DNAPLS tend to sink until they reach
an impermeable layer (Figure 5).
Dissolved NAPL forms the aqueous contaminant
plume that moves in the direction of ground-water
flow. The residual NAPL and free product can serve
as a source of ground-water contamination as long as
they remain in the subsurface.
                                           16

-------
Figure 8.  Dark NAPL (Soltrol) and water in a homogenous micromodel after (a) the displacement of water by
          NAPL and then (b) the displacement of NAPL by water, with NAPL at residual saturation (Wilson et al.
          1990).
                                       17

-------
Figure 9.
Photomicrographs of (a) a
single blob occupying one
pore body, and (b) a doublet
blob occupying two pore
bodies and a pore throat
(Wilson etal., 1990).
                                                            18

-------
nticipating Tailing and Rebound Prob
         The phenomena of tailing and rebound are
       commonly observed at pump-and-treat sites.
       Tailing refers to the progressively slower rate of
       decline in dissolved contaminant concentration
       with continued operation of a pump-and-treat
       system (Figure 10). Rebound is the fairly rapid
       increase in contaminant concentration that can
       occur after pumping has been discontinued. This
       increase may be followed by stabilization of the
       contaminant concentration at a somewhat lower
       level.
       4.1.   Effects of Tailing and Rebound
              on Remediation Efforts
         Tailing presents two main difficulties for
       ground-water restoration:
          • Longer treatment times. Without tailing,
           contaminants theoretically could be removed
           by pumping a volume of water equivalent to
           the volume of the contaminant plume (Figure
           10). The tailing effect, however, significantly
           increases the time pump-and-treat systems
           must be operated to achieve ground-water
           restoration goals. Indeed, pumping may need
           to be conducted for hundreds of years rather
           than tens of years.
          • Residual concentrations in excess of the
           cleanup standard. When tailing occurs, often
           initially the decline in the rate of contaminant
           concentrations is fairly rapid, followed by a
           more gradual decline that eventually stabilizes
           at an apparent residual concentration level
           above the cleanup standard (Figure 10).
  Rebound is most problematic when a pump-
and-treat system attains the cleanup standard, but
concentrations subsequently increase to a level
that exceeds the standard.
4.2.   Contributing Factors
  The degree to which tailing and rebound
complicate remediation efforts at a site is a
function of the physical and chemical characteris-
tics of the contaminant being treated, the subsur-
face solids, and the ground water. Major factors
and processes that contribute to tailing and
rebound are discussed below.
4.2.1.     Non-Aqueous Phase Liquids
  Although immiscible LNAPLs and DNAPLs
tend to be relatively insoluble in water, unfortu-
nately they often are sufficiently soluble to cause
concentrations in ground water to exceed MCLs.
Consequently, residual and pooled free-product
NAPL will continue to contaminate ground water
that makes sufficient contact to dissolve small
amounts from the NAPL surface (Figure  1 la).
When ground water is moving slowly, contami-
nant concentrations can approach the solubility
limit for the NAPL (Figure lie). Although pump-
and-treat systems increase ground-water velocity,
causing an initial decrease in concentration, the
decline in concentration will later tail off until the
NAPL's rate of dissolution is in equilibrium with
the velocity of the pumped ground water. If
pumping stops, the ground-water velocity slows
and concentrations can rebound, rapidly at first
and then gradually reaching the equilibrium
                                             19

-------
  t
 o
 O
                                     Pumping On
                                           Theoretical Removal
                                           Without Tailing
         Apparent Residual
         Contaminant
         Concentration
             Cleanup
             Standard
                         Pumping Off
Figure 10. Concentration versus pumping duration or volume showing tailing and rebound effects (Cohen et al.
         1994).
concentration (Figure 1 Ic), unless pumping is
resumed.
  As shown in Table 1, DNAPL contamination in
heterogeneous and fractured aquifers is the most
intractable. The reasons for this are
  • DNAPLs create an unstable wetting front in
    the subsurface, with fingers of more rapid
    vertical flow speeding the movement deeper
    into the saturated zone (Figure 12). (This also
    makes accurate delineation of zones of
    residual contamination extremely difficult in
    homogeneous aquifers.)
  • If the volume of DNAPL exceeds the residual
    saturation capacity of the unsaturated and
    saturated zones, the DNAPL will reach lower
    permeability materials and form pools of free
    product (Figure 5).
  • In heterogeneous aquifers, localized lenses of
    low-permeability strata may cause pools of
    free product to develop throughout the
    saturated zone (Figure 12). Low-permeability
    strata also may cause extensive lateral move-
    ment of the DNAPL. DNAPL pools are
    especially problematic because the contami-
    nant will dissolve even more slowly than
    residual DNAPL. It may take tens of years to
    remove 1 cm of contaminant from a DNAPL
    pool (NRC, 1994).
4.2.2.    Contaminant Desorption
  The movement of many organic and inorganic
contaminants in ground water is retarded by
sorption processes that cause some of the dis-
solved contaminant to attach to solid surfaces.
The amount of contaminant sorbed is a function
of concentration, with sorption increasing as
concentrations increase, and the sorption capacity
                                     20

-------
                                                                            Advection
                                                                                             Contaminant Free Product
                                                                                                 Solid Grain
                                                                                           Liquid-Liquid Partitioning
                                                                             (a)
                                                                    Organic Carbon or
                                                                    Mineral Oxide Surface
                                                                              (b)
                                           Desorption of
                                           Adsorbed Contaminants
Figure 11.
Contaminants are mobilized
when ground water that is
undersaturated with a
contaminant comes in
contact with a NAPL (a) or
contaminant sorbed on an
organic carbon or mineral
surface (b).  High ground-
water velocities and short
contact times will result in
low contaminant
concentrations, and low
velocities and long contact
times will result in high
contaminant concentrations
(c) (adapted from Gorelick et
al., 1993).
   Equilibrium Concentration
                    Low Ground-Water Velocities and Long Contact
                    Times Produce High Contaminant Concentrations
                    (approaching equilibrium) in Ground Water
                High Ground-Water Velocities and Short Contact Times
                Produce Low Contaminant Concentrations in Ground Water
       Contact Time Increases to Right

• Ground-Water Velocity Increases to Left


                (C)
                                                                    21

-------
Figure 12. Laboratory model of the transport of DNAPL contaminant through an aquifer with varying permeability;
          note the concentration of downward movement in fingers and the DNAPL pools above the low-
          permeability zones (the horizontal discs). (Source: U.S. EPA National Risk Management Research
          Laboratory.)
                                        22

-------
of the subsurface materials. Sorbed contaminants
tend to concentrate on organic matter and clay-
sized mineral oxide surfaces (Figure 1 Ib). Sorp-
tion is a reversible process, however. Thus, as
dissolved contaminant concentrations  are reduced
by pump-and-treat system operation, contami-
nants sorbed to subsurface media can desorb from
the matrix into ground water. Contaminant
concentrations resulting from sorption and
desorption show a relationship to ground-water
velocity and contact time similar to that of NAPLs
(Figure 1 Ic), causing the tailing  of contaminant
          concentrations during pumping as well as rebound
          after pumping stops.
          4.2.3.    Precipitate Dissolution
            As with sorption-desorption reactions, precipi-
          tation-dissolution reactions are reversible. Thus,
          large quantities of inorganic contaminants, such
          as chromate in BaCrO4, may be found with
          crystalline or amorphous precipitates in the
          subsurface (Palmer and Fish,  1992). Figure 13
          illustrates a tailing curve where the contaminant
          concentration is controlled by solubility. In this
          situation, if pumping stops before the solid phase
          is depleted, rebound can occur.
       o
       "ro
       o
       O
       T3
       O
       CO
       CO
Contaminant Concentration
  Controlled by Solubility
       •B   0.4  -
           0.2  -
                                                                     Solid-Phase
                                                                     Reserve Depleted
                                Pumping Duration or Volume Pumped
Figure 13. Dissolved contaminant concentration in ground water pumped from a recovery well versus time in a
          formation that contains a solid-phase contaminant precipitate (Palmer and Fish, 1992).
                                       23

-------
4.2.4.    Matrix Diffusion
  As contaminants advance through relatively
permeable pathways in heterogeneous media,
concentration gradients cause diffusion of con-
taminant mass into the less permeable media
(Gillham et al., 1984). Matrix diffusion is most
likely to occur with dissolved contaminants that
are not strongly sorbed, such as inorganic anions
and some organic chemicals. During a pump-and-
treat operation, dissolved contaminant concentra-
tions in the relatively permeable zones are re-
duced by advective flushing, causing a reversal in
the initial concentration gradient and slow diffu-
sion of contaminants from the low to high perme-
ability media. Figure 14, based on theoretical
calculations of TCE concentrations in clay lenses
of varying thickness, shows that diffusion is a
slow process. For example, the figure indicates
that the time required to reduce the concentration
of TCE to 10 percent of the initial concentration
would be 6 years for a clay lens 1 foot thick, 25
years for a clay lens 2 feet thick, and 100 years
for a clay lens 4 feet thick. The significance of
matrix diffusion increases as the length of time
between contamination and cleanup increases. In
heterogeneous aquifers, matrix diffusion contribu-
tions to tailing and rebound can be expected, as
long as contaminants have been diffusing into
less-permeable materials.
4.2.5.    Ground-Water Velocity Variation
  Tailing and rebound also result from the vari-
able travel times associated with different flow
paths taken by contaminants to an extraction well
(Figure 15a-c). Ground water at the edge of a
capture zone created by a pumping well travels a
greater distance under a lower hydraulic gradient
                                         Clay Lens Thickness = 1.2 Meters (4 ft)
                            20
                                                                                    100
Figure 14. Changes in average relative trichloroethene (TCE) concentrations in clay lenses of varying thickness
          as a function of time (NRC, 1994).
                                      24

-------
                                          (a)
                                                     Moderate
                                                               Fast
                                                                           Moderate
                                          (b)
Figure 15.
Tailing resulting from
ground-water velocity
variations: (a) horizontal
variations in the velocity of
ground water moving
toward a pumping well
(Keely, 1989) lead to (b)
tailing as higher
concentrations of ground
water in slower pathlines
mix with lower
concentrations in faster
pathlines (Palmer and Fish,
1992); (c) in a stratified
sand and  gravel aquifer,
tailing occurs attl when
clean water from the upper
gravel strata mixes with still-
contaminated ground water
in the lower sand strata
(Cohen etal., 1994).
                                                       1.2
                                                       1.0
                                                    "
                                                       0.8
       0.6
       0.4
       0.2
                                                                                          Tailing Due to
                                                                                          Different Travel
                                                                                          Times Along Flow
                                                                                          Paths to
                                                                                          Recovery Well
                    Pumping Duration or Volume Pumped
Stratified Sand-Gravel Aquifer
                                                                 25

-------
than ground water closer to the center of the
capture zone (Figure 15a). Additionally, contami-
nant-to-well travel time varies as a function of the
hydraulic conductivity in heterogeneous aquifers
(Figure 15c).
4.3.                the                of
       Tailing                 at a
  Determining realistic objectives for apump-
and-treat system requires sufficient site character-
ization to define the complexity of the hydrogeo-
logic setting (Sidebar 2) and the subsurface
distribution of contaminants. Such information
makes it possible for the system operator to assess
whether conditions at the site will result in tailing
and rebound and to evaluate the extent to which
these conditions are likely to increase the time
needed to attain health-based cleanup standards.
The sorption characteristics of contaminants can
be assessed using batch sorption tests with aquifer
materials (Roy et al., 1992), although aquifer
heterogeneity increases the difficulty of interpret-
ing test results. For organics, the potential effects
of sorption can be assessed based on a literature
review of contaminant properties and on site-
specific data on organic carbon in aquifer materi-
als (Piwoni and Keely, 1990). Geochemical
computer codes can be used to assess the potential
for tailing and rebound effects from precipitation-
dissolution reactions.
  Assessing the potential for removal or contain-
ment of free product may be the first priority at
NAPL-contaminated sites, followed by assess-
ment of the extent of residual NAPL contamina-
tion. For DNAPLs, residual saturation may extend
throughout the unsaturated and saturated zones
(Figure 5). Typically, for LNAPLs most residual
contamination is located in the vadose zone, but it
may also extend to the depth of the seasonal low
water table. As Figure 16 shows, pumping to
remove free LNAPL product can cause residual
NAPL to move deeper into the saturated zone.
Consequently, when removing free-product
LNAPL that is floating on the water table, steps
should be taken to avoid or minimize movement
of residual NAPL deeper into the saturated zone.
  Berglund and Cvetkovic (1995) evaluated the
relative importance of the degree of heterogeneity
in hydraulic conductivity and mass transfer
processes and concluded that the rate of mass
transfer and the extent to which contaminants are
sorbed on aquifer solids  are the most important
parameters that affect predicted cleanup time.
                                      26

-------
          Ground Surface
          LNAPL
                   At
                                                          Zone of Residual LNAPL Contamination
                                                          After Initial Recovery Efforts Have Stopped
Figure 16. Zone of residuals created in former cone of depression after cessation of LNAPL recovery system
          (Gorelicket al., 1993).
                                         27

-------
ffective  Hydraulic C
          Hydraulic containment is a design objective of
        nearly all pump-and-treat systems. Where restora-
        tion of an aquifer to health-based standards is the
        overall objective, the primary goal of containment
        must be to prevent farther spread of the contami-
        nant plume during restoration efforts. Where
        NAPLs are present, containment using hydraulic
        and physical barriers might be the primary
        objective for cleanup efforts in the portion of the
        aquifer contaminated by free product and residual
        NAPL (Figures Ic and 17). In such situations a
        conventional pump-and-treat system might be
        used to restore the dissolved contaminant plume
        (Figure 17).
          Effective hydraulic containment using pumping
        wells requires the creation of horizontal and
        vertical capture zones that draw all  contaminated
        ground water to the wells (Section 5.1.1) or other
        hydraulic barriers (Sections 5.1.2 and 5.1.3).
        Failure to take aquifer anisotropy into account
        (Section 5.2.1) or limitations in the ability to
        create sufficient drawdown to establish capture
        zones (Section 5.2.2) may allow contaminants to
        escape from these systems. Additionally, stagna-
        tion zones created by pumping operations or the
        use of injection wells can reduce the effectiveness
        of cleanup efforts (Section 5.2.3). The monitoring
        of both hydraulic heads (Section 6.4.1) and
        ground-water  quality (Section 6.4.2) can provide
        early indications that contaminants  are not being
        contained.
5.1.   Ground-Water Barriers and Flow
       Control
  Hydraulic containment can be accomplished by
controlling the direction of ground-water flow
with capture zones (Section 5.1.1) or pressure
ridges (Section 5.1.2) or by using physical
barriers (Section 5.1.3). Figure 17 illustrates a
pump-and-treat system that uses all three types of
hydraulic controls: (1) the contaminant source
area is surrounded by a barrier wall, (2) extraction
wells around the margins of the dissolved plume
capture the contaminated ground water, and (3)
treated ground water is reinjected to create a
pressure ridge along the axis of the contaminant
plume. Note that the pressure ridge in Figure 17
serves the function of increasing pore-volume
exchange rates rather than functioning as a
barrier. Barrier pressure ridge  systems are created
by placing injection  wells along the perimeter of a
contaminant plume.
5.1.1.     Horizontal and Vertical Capture
          Zones
  Pumping wells provide hydraulic containment
by creating a point of low hydraulic head to which
nearby ground water flows. The portion of an
aquifer where flow directions are toward a
pumping well is called a capture zone. In an
isotropic aquifer, where hydraulic conductivity is
the same in all directions, ground-water flow is
perpendicular to the  hydraulic head contours, also
called equipotential  lines (Figure 18b).
                                             28

-------
                                              Limit of Dissolved Plume
                     Contaminant
                     Source Area
                                 Initial Ground-Water Flow Direction
          Barrier Wall
                                                Q  Injection Well

                                                9  Extraction Well

Figure 17. Plan view of a mixed containment-restoration strategy.  A pump-and-treat system is used with barrier
          walls to contain the ground-water contamination source areas (e.g., where NAPL or waste may be
          present) and then collect and treat the dissolved  contaminant plume (Cohen et al., 1994).
  A pumping well creates a zone of influence
where the potentiometric surface has been modi-
fied (Figure 18c). The capture zone is the portion
of the zone of influence where ground water flows
to the pumping well (Figure 18d). Figure 15a
shows how a capture zone creates flow lines of
varying velocity. The size and shape of a capture
zone depend on the interaction of numerous
factors, such as
  • The hydraulic gradient and hydraulic conduc-
    tivity of the aquifer.
   • The extent to which the aquifer is heteroge-
    neous (Sidebar 2) or anisotropic (Section
    5.2.1).
   • Whether the aquifer is confined or uncon-
    fined.
   • The pumping rate and whether other pumping
    wells are operating.
   • Whether the screened interval of the well
    fully or partially penetrates the aquifer.
  When the screened portion of a pumping well
fully penetrates an aquifer (Figure Ib), a two-
dimensional analysis to delineate the horizontal
                                       29

-------
                                            \
                             (a)
 (b)
                \    \     \   \    \    v
                   Zone of Influence     \       ^
                              (c)
(d)
Figure 18. In an isotropic aquifer, ground-water flow lines (b) are perpendicular to hydraulic head contours (a).
          Pumping causes drawdowns and a new steady-state potentiometric surface within the well's zone of
          influence (c).  Following the modified hydraulic gradients, ground water within the shaded capture zone
          flows to the pumping well (d). (Cohen et al., 1994, adapted from Gorelick et al., 1993).
                                          30

-------
capture zone is usually sufficient. When a pump-
ing well only partially penetrates an aquifer,
however, vertical capture zone analysis also is
required to determine whether the capture zone
will contain a contaminant plume. Figure 19
shows a vertical capture  zone for a partially
penetrating well. If the contaminant plume
extended to the base of the aquifer, some contami-
nants would bypass the well, despite the presence
of apparent upward gradients. In stratified aniso-
tropic media (Section 5.2.1), the vertical hydraulic
control exerted by a partially penetrating well will
be further diminished.
5.1.2.    Pressure Ridge Systems
  Pressure ridge systems are produced by inject-
ing uncontaminated water into the subsurface
through a line of injection wells located
upgradient or downgradient of a contamination
plume. The primary purpose of a pressure ridge is
to increase the hydraulic gradient and hence the
velocity of clean ground water moving into the
plume, thereby increasing flow to the  recovery
wells, which serves to wash the aquifer.
Upgradient pressure ridges also  serve to divert the
flow of uncontaminated ground water around the
plume, and downgradient pressure ridges prevent
further expansion of the  contaminant plume.
Typically, treated ground water from extraction
wells within a contaminant plume supply the
upgradient or downgradient injection wells used
to create a pressure ridge.
5.1.3.     Physical Barriers
  Physical barriers are constructed of low-
permeability material and serve to keep fresh
ground water from entering a contaminated
aquifer zone. They also help prevent existing
areas of contaminant from moving into an area of
clean ground water or releasing additional con-
taminants to a dissolved contaminant plume. Most
systems involving physical barriers also require
ground-water extraction to ensure containment by
maintaining a hydraulic gradient toward the
contained area (see Figure Ic). The advantage of
physical barriers is that the amount of ground
water that must be extracted is greatly reduced
compared to the amount when using hydrody-
namic controls, as described in Sections 5.1.1 and
5.1.2. Major types of barriers include
  • Caps (or covers), which are made of low-
    permeability material at the ground surface,
    can be constructed of native soils, clays,
    synthetic membranes, soil cement, bitumi-
    nous concrete, or asphalt.
  • Slurry trench walls, excavated at the proper
    location and to the desired depth while
Figure 19. Cross section showing equipotential contours and the vertical capture zone associated with ground-
         water withdrawal from a partially penetrating well in isotropic media (Cohen et al., 1994).
                                      31

-------
    keeping the trench filled with a clay slum-,
    keep the trench sidewalls from collapsing and
    backfilling with soil bentonite. cement
    bentonite, or concrete mixtures.
   « Grout curtains are created by injecting
    stabilizing materials under pressure into the
    subsurface to fill voids, cracks, fissures, or
    other openings in the subsurface. Grout also
    can be mixed with soil using larger augers.
   « Sheet piling cutoff walls are constructed by
    driving sheet materials, usually steel, through
    unconsolidated materials with a pile driver or
    more specialized vibrator}? drivers.
  Knox et al. (1984) provide further information
on the design and construction of physical
ground-water barriers.
5.2.    Hydraulic Containment: Other

  Certain site conditions can allow7 contaminants
to escape from a hydraulic containment system if
they are not characterized and anticipated.
5.2.1.            ofAnisotropy
  In anisotropic aquifers, hydraulic conductivity
varies with direction. In flat-lying sedimentary
aquifers, hydraulic conductivity is often higher in
a horizontal than a vertical direction. In fractured
rock and foliated metamorphic rocks, such as
schist, the direction of maximum and minimum
permeability is usually aligned parallel and
perpendicular, respectively, to foliation or bedding
plane  fractures (Cohen et al.. 1994). Where
sedimentary strata and foliated media are inclined
or dipping, significant horizontal anisotropy may
be an  aquifer characteristic. In anisotropic media,
the flow of ground water, as well as contaminants
moving with ground water,  is usually not perpen-
dicular to the hydraulic gradient.
  Figure 20 illustrates how horizontal anisotropy
in fractured rock can change the location of the
capture zone of a pumping well. In an aquifer that
is assumed to be isotropic, the general direction of
ground-water flow should be perpendicular to the
hydraulic gradient (Figure 20a). If fractures cause
hydraulic conductivity to be higher in a north-
south rather than an east-west direction, however,
the direction of ground-water flow will diverge
from the direction of the hydraulic gradient
(Figure 20b). In this example, siting a pumping
well based only on the hydraulic gradient (Figure
20a) would result in its failure to capture any
portion of a contaminant plume, except in the
immediate vicinity of the well.
  A contaminant plume that does not follow the
hydraulic gradient may indicate that anisotropy is
influencing the direction of ground-water flow.
Aquifer heterogeneities, such as buried stream
channels that have a different direction than the
hydraulic gradient, also may allow the direction
of contaminant travel to diverge from the hydrau-
lic gradient Computer programs, such as EPA's
Well Head Protection Area (WHPA) code, can be
useful for evaluating the potential  effects of
anisotropy on w7ell capture zones. Figure 21
shows such a simulation for three pumping wells.
In this case, with a vertical to horizontal anisot-
ropy ratio of 10:1, the orientation of the capture
zones shifts from northwest-southeast (isotropic)
to east-west (anisotropic).
5.2.2.    Drawdown
  Under some conditions creating and maintain-
ing an inward hydraulic gradient for a contami-
nant plume is problematic. In such situations.
injection wells may be required to create pressure
ridges (Section 5.1.2) or physical barriers may
need to be installed (Section  5.1.3). Site condi-
tions that might indicate the need for  such mea-
sures  include (Cohen et al., 1994)
   * Limited saturated thickness of the aquifer
   * Relatively high initial hydraulic gradient
                                      32

-------
(a) Isotropic Aquifer
                                                    (b) Anisotropic Aquifer
Water-
Table
Contours
                                                       Water-Table
                                                        Contours
 Figure 20. Effect of fracture anisotropy on the orientation of the zone of contribution (capture zone) to a pumping
           well (Bradbury etal., 1991).
    • Sloping aquifer base
    • Very high aquifer permeability
    • Low aquifer permeability
   Where these conditions exist and hydraulic
 containment is planned, particular care should be
 taken during site characterization and pilot tests to
 assess drawdown limitations.
 5.2.3.    Stagnation Zones
   Stagnation zones develop in areas where pump-
 and-treat operations create low hydraulic gradi-
 ents and, consequently, low ground-water veloci-
 ties. The stagnation zone associated with a single
 extraction well is likely to be located
 downgradient from the well (Figure 22a). A
                                                   stagnation zone can develop upgradient from an
                                                   injection well, however, and form in low-perme-
                                                   ability zones, regardless of hydraulic gradient.
                                                   When multiple extraction or injection wells are
                                                   involved, a number of stagnation zones may
                                                   develop (Figure 22b). Stagnation zones caused by
                                                   low hydraulic gradients can be identified by
                                                   measuring hydraulic gradients, tracer movement,
                                                   and ground-water flow rates using downhole
                                                   flowmeters and through modeling analysis.
                                                   Stagnation zones within a contaminant plume can
                                                   reduce the efficiency of a pump-and-treat system;
                                                   thus, minimizing stagnation is an important
                                                   objective of capture zone analysis and optimiza-
                                                   tion modeling (Section 6.1).
                                        33

-------
                                        (ft)
                                                                Isotropic
                                    9000
                                    7200
                                    5400
                                    3600
                                    1800
                                                               I
                                                  2100       4200       6300

                                                                    (a)
                                                                                                  (ft)
                                               8400      10500
                                         (ft)
                           Anisotropic
Figure 21.
Capture zone
simulation of
three pumping
wells for an
isotropic aquifer
(a) and
anisotropy ratio of
10:1 (b) using the
EPA WHPA code.
                                    9000
                                    7200
                                    5400
                                    3600
1800
                           I
              (ft)
              2100       4200       6300

                              (b)
8400
10500
                                                           34

-------
                  5 / N-  ,n  K  S  CO  CM
       Hydraulic Head
       Contour Map

      I I  I I I  11 11  1
                                          CO
                                          LO
X      X      \
    11       10     9
                           X
                  Ground-Water Velocity
                  Contour Map
                               Hydraulic Head
                               Contour Map
    O
                                                 (b)
                                                     — 1
                                                            Ground-Water Velocity Contour Map
Figure 22. Examples of stagnation zones (shaded where ground-water velocity is less than 4 L/T): (a) single

          pumping well and (b) four extraction wells with an injection well in the center (Cohen et al., 1994).
                                        35

-------
  The basic operating principle of a pump-and-
treat system calls for locating a well (or wells)
and then pumping at rates that cause all water in a
contaminant plume to enter the well rather than
continue traveling through the subsurface. Table 2
lists types of data required for evaluating the
feasibility of using the pump-and-treat approach
at a contaminated ground-water site and then
designing an appropriate system. This section
describes the key aspects of designing and
operating a pump-and-treat system for optimal
performance.
6.1.    Capture Zone Analysis and
        Optimization Modeling
  In recent years, numerous mathematical models
have been developed or applied to compute
capture zone, ground-water pathlines, and associ-
ated travel times to extraction wells or drains. For
relatively simple hydrogeologic settings (homoge-
neous isotropic aquifers), analytical equations
solved manually, using graphical techniques or
computer codes based on analytical solutions,
may be adequate. For more complex sites, nu-
merical  computer models may be required. These
models provide insight to flow patterns generated
by alternative pump-and-treat approaches and to
the selection of monitoring points and frequency.
The WHPA model (Blandford and Huyakorn,
1991) and Capture Zone Analytic Element Model
(CZAEM) (Haitjema et al., 1994; Stock  et al.,
1994) developed by EPA are examples of rela-
tively simple computer software based on analyti-
cal equations (WHPA) and the innovative analytic
element method (CZAEM) that allows capture
zone and ground-water pathline analysis. The
numerical MODFLOW and MODPATH models
developed by the U.S. Geological Survey are
commonly used to model more complex hydro-
geologic settings. Cohen et al. (1994) identify a
number of computer codes of potential value for
capture zone analysis. More detailed information
about specific models and EPA guidance on the
use of models are available in references on
Ground-Water Modeling at the end of this guide
(Section 9). Sidebar 5 summarizes the results of
computer modeling performed to evaluate the
effect of different hydrogeologic conditions on the
effectiveness of different types of well patterns.
  In addition, optimization programming methods
are being used increasingly to improve pump-and-
treat system design (Gorelick et al., 1993). As
applied to the design of pumping systems, optimi-
zation involves defining an objective function,
such as minimizing the sum of pumping rates
from a number of wells. A set of restrictions, or
constraints, specify various conditions, such as
maximum pumping rates and minimum hydraulic
heads at individual wells, that must be satisfied by
the optimal solution alternative. Hydraulic
containment of a contaminant plume usually
requires only linear optimization methods, but
when contaminant concentrations are specified as
constraints, nonlinear methods are often required
(Rogers et al., 1995). At the Lawrence Livermore
                                     36

-------
Table 2. Data Requirements for Pump-and-Treat Systems (Adapted from U.S. EPA, 1991)

     Data Description                                   Purpose(s)
                                               Source(s)/Method(s)
Hydraulic conductivities and
storativities of subsurface
materials
Contaminant concentrations
and areal extent
Contaminant/soil properties
(density, aqueous solubility,
octanol-water/carbon
partitioning coefficient, soil
organic carbon content,
sorption parameters)

Types, thicknesses, and extent
of saturated and unsaturated
subsurface materials

Depth to aquifer/water table
Ground-water flow direction
and vertical/horizontal gradients
Seasonal changes in ground-
water elevation
NAPL density/viscosity/
solubility; residual saturation of
vadose zone and saturated zone
Ground-water/surface water
connection

Precipitation/recharge
Locations, screen/open interval
depths, and pumping rates of
wells influenced by site
To determine feasibility of
extracting ground water;
applicability of pump-and-treat
approach

To determine seriousness of the
problem; existence of NAPL;
applicability and evaluate
effectiveness

To determine mobility
properties; applicability of
pump-and-treat approach
To develop conceptual design;
applicability/considerations for
implementation

To select appropriate extraction
system type; consideration for
implementation

To determine proper well
locations/spacing considerations
for implementation

To locate wells and screened
intervals; considerations for
implementation

To predict vertical distribution
of contamination; consideration
for implementation and
evaluating effectiveness

To determine impacts of surface
water

To calculate water balance;
consideration for implementing
and evaluating effectiveness

To determine
impacts/interference;
considerations for implementing
and evaluating effectiveness
Pumping test, slug tests,
laboratory permeability tests
Soil and water quality sampling
data
Published literature, laboratory
tests
Hydrogeologic maps, surficial
geology maps/reports, boring
logs, geophysics

Hydrogeologic maps,
observation wells, boring logs,
piezometers

Water level data,
potentiometric maps
Long-term water level
monitoring
Literature, laboratory
measurements
Seepage measurements,
stream gaging

NOAA reports, local weather
bureaus; onsite measurements
Well inventory, pumpage
records
                                                              37

-------
                                               Sidebar 5
      Computer Modeling of Well Patterns Versus Hydrogeologic Conditions
  Satkin and Bedient (1988) used the U.S. Geological
Survey MOC model to evaluate the effectiveness of
seven different well patterns (Figure 23) for restoring
contaminated ground water under eight generic
hydrogeologic conditions. The hydrogeologic settings
were defined as various combinations of three major
factors: maximum drawdown (high > 10 ft; low < 5 ft),
hydraulic gradient (high = 0.008; low = 0.0008), and
longitudinal dispersivity (high = 30 ft;  low =10 ft).
Because the contaminants were assumed to not interact
with aquifer solids, tailing and rebound effects were not
a consideration in the study. Major conclusions of the
computer simulations include the following:
 •  Significant differences in cleanup time were observed
   using various well locations for a given well pattern.
The three-spot, doublet, and double-cell well
patterns are effective under low hydraulic gradient
conditions. These well patterns minimize cleanup
time, volume of water circulated, and volume of
water treated.
The three-spot well pattern performed better than
any of the other well patterns studied under a high
hydraulic gradient, high drawdown, and either a low
or high dispersivity.
None of the well patterns investigated was able to
contain and clean up the contaminated plume in a
setting with high gradient, low drawdown, and high
dispersivity.
The centerline well pattern is effective in achieving
up to 99 percent contaminant reduction under both
low and high gradient conditions, but it may present
a water disposal problem.
The five-spot well pattern was the least effective of
the well patterns studied.
                                          38

-------
                                      Single
    Doublet
Centerline
                                      (..Pumping Well
                                                                   X
                                                                   X
                                                                  .—s
                                                                3-Spot
                        X..Injection Well
Figure 23.
Major types of
pumping/injection
well patterns
(Satkin and
Bedient, 1988).
                                      X    X
                                      X    X,
                                      5-Spot
       X
    X   X
Double Triangle
    X
    X
Double Cell
                                                            39

-------
National Laboratory (LLNL) site, Rogers et al.
(1995) applied an innovative nonlinear optimiza-
tion approach, using artificial neural networks and
a genetic algorithm, to evaluate more than 4
million pumping patterns for the project's 28
extraction and injection wells. The three top-
ranked patterns required 8 to 13 wells, with
projected costs estimated at $41 to $53 million
over the 50-year project life. Using these pumping
patterns was estimated to cost from one-third to
one-quarter the cost of using all 28 wells at an
estimated cost of $1.55 million.
6.2.    Efficient Pumping
  Removal of contaminated ground water should
be a dynamic process that uses information on the
response of the ground-water system to improve
the efficiency of pumping operations (Section
3.3). Elements of efficient pumping operations
can include
   * Combined plume containment and source
    remediation, which can be achieved through
    the design of the initial pumping flow field.
    For example, at the LLNL site a line of
    extraction wells at the downgradient margins
    of the plume were established to prevent the
    movement of contaminants toward municipal
    water-supply wells, while other wells were
    located in the source areas where the contami-
    nant concentrations were highest. This limited
    the area requiring remediation and maximized
    contaminant removal (Hoffman, 1993).
   * Phased construction of extraction wells,
    which allows data on the monitored response
    of the aquifer to pumping operations to be
    used in siting subsequent wells.
   «Adaptive pumping, which involves designing
    the well field such that extraction and injec-
    tion can be varied to reduce zones of stagna-
    tion. Extraction wells can be periodically shut
 off. others turned on, and pumping rates
 varied to ensure that contaminant plumes are
 remediated at the fastest rate possible. Figure
 24 illustrates stagnation zones that would
 develop at the LLNL site if a fixed pumping
 well configuration were used. With this
 approach, remediation at the site would take
 about 100 years (Figure 25). Computer
 modeling of adaptive pumping indicates that
 this technique should make it possible to
 reduce the time required for site cleanup to
 about 50 years (Figure 25). Further refine-
 ments in design might shorten the time even
 further (Hoffman, 1993).
• Pulsed pumping, which has the potential to
 increase the ratio of contaminant mass
 removed to ground-water volume where mass
 transfer limitations restrict dissolved contami-
 nant concentrations. Figure 26 illustrates the
 concept of pulsed pumping. During the
 resting phase of pulse pumping, contaminant
 concentrations increase due to diffusion,
 desorption, and dissolution in slower moving
 ground water (Figure 11). Once pumping is
 resumed, ground water with a higher concen-
 tration of contaminants is removed, thus
 increasing mass removal during pumping.
 Special care must be taken to ensure that the
 hydraulic containment objective is met during
 pump rest periods. Bartow and Davenport
 (1995) have reported that about 19 percent of
 the pump-and-trcat systems in Santa Clara
 Valley, California, use some form of pulsed
 pumping. A recent study by Harvey ct al.
 (1994). however, on the effects of physical
 parameters (e.g., the mass transfer rate
 coefficient) concluded that pulsed pumping
 provides little if any advantage over continu-
 ous pumping at an average rate.
                                     40

-------
                   LEGEND

                   Ground-water
                   flow Line
              O   Extraction
                   location
                |y. Areas of potential
                   ground-water
                   stagnation
 —1—
 —10—
~^QQ"~»
 ^OQ0"**'
Total VOC (ppb)
isoconcentration
contours,
dashed where inferred
                                                        Union Pacific
                                                                   -i-
                                                     i   /    /      Patterson Pass Road
            wells within VOC plumes are
          contoured without regard to 'depth
              Rhonewood Subdivision
Figure 24. Ground-water flow line in the vicinity of conceptual pumping centers at Lawrence Livermore National
          Laboratory superimposed on an isoconcentration contour map and showing areas of potential
          stagnation (Cohen et al., 1994, after Hoffman, 1993).
                                          41

-------
                                      1000
Figure 25.
Effect of adaptive
pumping on
cleanup time at
Lawrence
Livermore
National
Laboratory
Superfund site
(Cohen et al.,
1994, after
Hoffman, 1993).
10  15  20  25   30   35  40  45  50
         Time (yr)
                 6.3.   Treating Contaminated Ground
                        Water
                   Once extraction wells have brought contami-
                 nated water to the surface, treatment is relatively
                 straightforward, provided that appropriate meth-
                 ods have been selected and the capacity of the
                 treatment facility is adequate. Table 3 summarizes
                 the applicability of various treatment technologies
                 to ground water contaminated by any of the major
                 categories of inorganic and organic contaminants.
                 U.S. EPA (1995) describes conventional technolo-
                 gies that have evolved from industrial wastewater
                 treatment and that have been implemented at full
                 scale for treatment of contaminated ground water.
                 These methods fall into two main categories:
                • Biological. Biological treatment methods use
                 microorganisms to degrade organic com-
                 pounds and materials into inorganic products.
                 The methods may be applicable for treatment
                 of ground water contaminated by organic
                 compounds if concentrations are low enough
                 and the biological processes are not inhibited.
                 The best established biological treatment
                 methods include (1) activated sludge  systems,
                 (2) a sequencing batch reactor, (3) powdered
                 activated carbon in activated sludge (bio-
                 physical system), (4) rotating biological
                 contactors, and (5) an aerobic fluidized bed
                 biological reactor.
                                                       42

-------
Figure 26.
The pulsed
pumping concept
(Cohen etal., 1994,
after Keely, 1989).
                                                           Time
                   • Physical/Chemical. Physical, chemical, or a
                    combination of physical and chemical meth-
                    ods can be used to remove contaminants from
                    ground water. The most commonly used
                    methods include (1) air stripping, (2) acti-
                    vated carbon, (3) ion exchange, (4) reverse
                    osmosis, (5) chemical precipitation of metals,
                    (6) chemical oxidation, (7) chemically
                    assisted clarification, (8) filtration, and (9)
                    ultraviolet (UV) radiation oxidation.
                   Various emerging and innovative treatment
                 technologies, such as electrochemical separation
                 and wet air oxidation, are being tested. The EPA
                 reference sources identified for ground-water
                 treatment methods at the end of this guide (Sec-
                 tion 9) provide additional information on estab-
                 lished and innovative treatment technologies.
                 6.4.    Monitoring Performance
                   An appropriately designed monitoring program
                 is essential for measuring the effectiveness of a
                 pump-and-treat system in meeting hydraulic
                 containment and aquifer restoration objectives. In
                 general, containment monitoring involves (1)
measuring hydraulic heads to determine if the
pump-and-treat system creates inward gradients
that prevent ground-water flow and dissolved
contaminant migration across the containment
zone boundary, and (2) ground-water quality
monitoring to detect any contaminant movement
or increase of contaminant mass across the
containment zone boundary. Aquifer restoration
monitoring mainly involves measurement of
contaminant concentrations in pumping and
observation wells to determine the rate and
effectiveness of mass removal. Cohen et al.
(1994) provide more detailed guidance on moni-
toring the performance of pump-and-treat sys-
tems.
6.4.1.    Hydraulic Head Monitoring for
         Containment
  In general, the number of observation wells
needed for monitoring inward hydraulic gradients
in a containment area increases with site complex-
ity and with  decreasing gradients along the
containment perimeter. Strategies for adequately
monitoring inward gradients and hydraulic
containment include (Cohen et al., 1994)
                                                      43

-------
> Measuring hydraulic heads in three dimen-
 sions using nested piezometers for detecting
 vertical gradients. As shown in Figure 19,
 partially penetrating wells may not create an
 adequate vertical capture zone. Where leaky
 confining layers separate aquifers, hydraulic
 gradients should be toward the contaminated
 zone. Figure 27 illustrates observations from
 a nest of piezometers at the diem-Dyne
 Superfnnd site in Ohio. Water levels from the
 deep piezometer are consistently about a foot
 higher than in the intermediate and shallow7
 piezometers, indicating an upward gradient.
• Monitoring water levels in observation wells
 intensively during system startup and equili-
 bration to determine an appropriate measure-
 ment frequency. This may involve using
 pressure transducers and dataloggers to make
 near-continuous head measurements for a few
 days or weeks, then switching sequentially to
 daily, weekly, monthly, and possibly quarterly
 monitoring. Data collected during each phase
 should provide the justification for any
 subsequent decrease in monitoring frequency.
> Making relatively frequent hydraulic head
 measurements when the pumping rates or
 locations are modified, or when the system is
 significantly perturbed in a manner that has
 not been evaluated previously. Significant
 new perturbations can arise from, for ex-
 ample, unusual recharge, flooding, drought,
 and new7 offsite w7ell pumping.
> Measuring hydraulic head as close to the
 same time as possible when monitoring
 inward hydraulic gradients or a potentiomet-
 ric surface so that data are temporally consis-
 tent. This ensures that differences in ground-
 water elevation within a network represent
 spatial rather than temporal variations.
  * Supplementing hydraulic head data with flow-
    path analysis using potentiometric maps or
    particle tracking computer codes. Figure 28
    shows that ground water can flow7 between
    and beyond recover}-- wells even though
    hydraulic heads throughout the mapped
    aquifer are higher than the pumping level.
  * Conducting an analysis to determine if
    containment is threatened or lost when
    hydraulic head data do not indicate a clear
    inward gradient. Rose diagrams can be
    prepared to display the variation over time of
    hydraulic gradient direction and magnitude
    based on data from at least three w7ells (Figure
    29). Even when the time-averaged flow7 is
    toward the pump-and-treat system, contain-
    ment can be compromised if contaminant
    escapes from the larger capture zone during
    transient events or if a net component of
    migration away from the pumping wells
    occurs overtime.
6,4.2.    Ground-Water Quality Monitoring
         for Containment
  Monitor w7ell locations and completion depths
should be selected to provide a high probability of
detecting containment system leaks  in a timely
manner. Consequently, monitor wells with
relatively close spacing are usually located along
or near the potential downgradicnt containment
boundary. Ground-water quality sampling usually
is performed less frequently than the measuring of
hydraulic head because contaminant movement is
a slower process. Because ground-water quality
monitoring is more expensive than hydraulic head
monitoring, designing a cost-effective monitoring
plan requires special care. Strategies that may
help reduce costs without compromising the
integrity of the program include (Cohen et al,
1994) *
                                   44

-------
  Table 3. Applicability of Treatment Technologies to Contaminated Ground Water (U.S. EPA, 1991)





Contaminants
Metals
Heavy metals
Hexavalent chromium
Arsenic
Mercury
Cyanide
Corrosives
Volatile organics
Ketones
Semivolatile organics
Pesticides

Dioxins
Oil and grease/floating
products



(—
eutralizatiol
-z.

X
X
X
X
X
•
X
X
X
X

X





recipitation
Q_

•
•
0
•
X
•
X
X
0
0



(5
3
D)
ro
o
o
c
n
oprecipitatii
O

•
X
•
•
X
X
X
X
0
0







V/Ozone
^>

X
X
0
X
•
X
0
o
•


X

c
o

"(0
-Q

hemical Ox
O

X
X
0
X
•
X
•
•
•


X





eduction
or

O
•
X
•
X
X
X
X
X
X

X





istillation
Q

X
X
X
X
X
O
•
•
•








ir Stripping
<

X
X
X
X
X
X
•
•
X
X

X



D)
c

'o.
Q.
&
E
ro
-2
OT

X
X
X
X
X
X
•
•
•
0

X


c
o
_Q

ctivated Ca
<

O
O
0
•
X
X
•
X
•
•

X





vaporation
LU

•
•
X
X
•
X
X
X
0
0

0


g
"ro

ro
Q.
CD

•
X
0
o
X
X
X
X
0
0

0





c
o
'-*-!
ro
"5
LL

X
X
X
X
X
X
X
X
0
0

0

*
c
o
U-*
ro
L_
CO
Q.

-------
           568
           567
B  566

g

>
.S?
LU   565
           564
           563
           562
                                                                                                565.36
                  Jan  '  Feb  '  Mar  '  Apr  '   May  '  June ' July '  Aug ' Sept  '   Oct


                                                       1992
                                                                                     Nov   Dec
                                                  Explanation

           Water-Level Measurement in Deep Piezometer          —

           Water-Level Measurement in Intermediate Piezometer    — •

           Water-Level Measurement in Shallow Piezometer       --

           Daily Average Water Level From Hourly Data Recorded
           in Shallow Piezometer
                                                          Time-Averaged Water Level in Deep Piezometer

                                                          Time-Averaged Water Level in Intermediate Piezometer

                                                          Time-Averaged Water Level in Shallow Piezometer
Figure 27. Nested piezometer hydrograph for 1992 at the Chem-Dyne Superfund site (Cohen et al., 1994, after Papadopulos &
          Associates, 1993).
                                                         46

-------
Figure 28.
Ground-water
flow between and
beyond the
extraction wells,
resulting even
though hydraulic
heads throughout
the mapped
aquifer are higher
than the  pumping
level (Cohen et
al., 1994).
                                                                                           Constant-Head
                                                                                             Pumping
                                                                                            Elevation in
                                                                                            Each Well Is
                                                                                              360.0ft
                                                                                      	
                                                                                      .*;.A^sXv Vk
                                                                                  363.0 iJSCK
                    1 Sampling more frequently and performing
                     more detailed chemical analyses in the early
                     phase of the monitoring program, and using
                     the information gained to optimize sampling
                     efficiency and reduce the spatial density and
                     temporal frequency of sampling in the later
                     phases.
                    1 Monitoring ground-water quality in perimeter
                     and near-perimeter leak detection wells more
                     frequently than in wells that are at a greater
                     distance from the contaminant plume limit.
                    1 Specifying sampling frequency based on
                     potential containment failure migration rates
                     that factor in hydraulic conductivity and
                     effective porosity of the different media as
 well as the maximum plausible outward
 hydraulic gradients. Consider more frequent
 sampling of more permeable strata in which
 migration might occur relatively quickly as
 compared to the sampling frequency for less
 permeable media.
• Focusing chemical analyses on site contami-
 nants of concern and indicator constituents
 after performing detailed chemical analyses
 during the remedial investigation or the early
 phase of a monitoring program. Conduct
 more detailed chemical analyses less fre-
 quently or when justified based on the results
 of the more limited analyses.
                                                       47

-------
                  6.4.3.    A quifer Restoration Monitoring
                   Aquifer restoration monitoring consists of three
                  main elements:
                    • Ground-water sampling from all extraction
                     wells and selected observation wells within
                     the contaminant plume to interpret cleanup
                     progress. Parameters analyzed should include
                     (1) the chemicals of concern, (2) chemicals
                     that could affect the treatment system, such as
                     iron, which can precipitate and clog treatment
                     units if ground water is aerated, and (3)
 chemicals that may indicate the occurrence of
 other processes of interest, such as dissolved
 oxygen, carbon dioxide, and biodegradation
 products. These sampling data are important
 for making adjustments for efficient well
 operation (Section 6.2).
• Periodic sampling and chemical analysis of
 aquifer materials from representative loca-
 tions in the contamination zone to measure
 removal of nondissolved contaminants.
                                                                      North
                                     West
Figure 29.
Example display of ground-water
flow directions and hydraulic
gradients determined between
three observation wells (Cohen et
al., 1994).
                                                                                                     East
 South
                                                        48

-------
   1 Regular sampling and analysis of treatment
    system influent and effluent to assess (1)
    treatment system performance, (2) change in
    influent chemistry that may influence treat-
    ment effectiveness, and (3) dissolved con-
    taminant concentration trends. Figure 30
    shows influent and effluent VOC concentra-
    tions for the first 6 years of operation at the
    Chem-Dyne Superfund site treatment plant.
    Influent concentrations data showed a large
    drop in the first year, and then a more gradual
    decline over the next 5 years due to tailing
    effects (Section 4).
  The simplest indicator of progress in removing
ground-water contaminants is a plot of the
cumulative mass removed from the aquifer as
measured by influent concentrations to the
treatment system. Figure 31 shows the cumulative
mass of VOC removal at the Chem-Dyne site.
Approximately 27,000 pounds of VOCs have been
removed since the system became operational. As
is apparent from both Figures 30 and 31, however,
the rate of removal slowed significantly in the
sixth year. Consequently, removal of the remain-
ing one-third of the in-place mass will take much
longer than 6 years.
    12
          1987
                       1988
                                     1989
                                                   1990
                                                                 1991
                                                                                 1992
                                            Year
                                     Influent
    O Effluent
Figure 30. Influent and effluent VOC concentrations (mg/L) at the Chem-Dyne treatment plant from 1987 to 1992
         (Cohen etal., 1994, after Papadopulos & Associates, 1993).
                                     49

-------
            30,000

        .Q
        .E   25,000
O
O
-t-«
m
"3
            20,000
        Q-   15,000
        >,
        ^   10,000
        o
            5,000
                            1987
                             1988
1989
  Year
1990
1991
1992
 Figure 31. Cumulative mass ofVOCs removed from the aquifer at the Chem-Dyne site from 1987 to 1992 (Cohen
          etal., 1994, after Papadopulos & Associates, 1993).
6.5.   Evaluating Restoration Success
       and Closure
  Ground-water restoration, as operationally
defined, is achieved when a predefined cleanup
standard is attained and sustained. Figure 32
outlines procedures for determining the success
and/or timeliness of closure of a pump-and-treat
system. U.S. EPA (1992) defines six stages of
remediation using water quality data from a single
well (Figure 33):
  • Stage 1. Site evaluation to determine the need
    for and conditions of a remedial action; define
    cleanup standard.
                                          • Stage 2. Operation of the remediation system,
                                           during which contaminant concentrations
                                           decline.
                                          • Stage 3. Conclusion of treatment after con-
                                           taminant concentrations have remained below
                                           the cleanup standard for a sufficient period of
                                           time based on expert knowledge of the
                                           ground-water system and data collected
                                           during pump-and-treat operations.
                                          • Stage 4. Post-termination monitoring of water
                                           levels and contaminant concentrations to
                                           determine when the ground-water flow
                                           system is reestablished.
                                     50

-------
                                                        Define Attainment Objectives and
                                                              Cleanup Standard
                                                        Develop Sampling and Analysis
                                                             Plan for Performance
                                                                Monitor System
                                                                 Performance
                                                                                     No
                                Continue/
                                  Modify
                                Treatment
    Is the Cleanup
 Standard Reached?
Demonstration of
   Technical
 Impracticability
                                                                                                      Modify
                                                                                                    Remedial
                                                                                                      Action
                                                                                                    Objectives
                                                            Allow System to Reach
                                                                 Steady-State
                            Assess and Revise
                            Treatment Design as
                                Necessary
Verify the Attainment of
  Cleanup Standard
Figure 32.
Determining the
success and/or
timeliness of
closure of a
pump-and-treat
system (Cohen et
al., 1994).
    Is the Cleanup
 Standard Maintained
     Over Time?
               Yes
     Monitor as
     Necessary
                                                            51

-------
          1.2  T
Start
Treatment
       g
       '•4-J
       ro
                                                 End Sampling
                                                 Declare Clean or
                                                 Contaminated
       o
       o
       o
       CD
       a)
       ro
          0.6
          0.4  . .
          0.2  - •
                                              Date

Figure 33. Stages of remediation in relation to example contaminant concentrations in a well at a pump-and-treat
         site (U.S. EPA, 1992).
  • Stage 5. Sampling to assess attainment of the
   cleanup standard. If the treatment standard is
   not met, the treatment design may need to be
   assessed and revised (Figure 32).
  • Stage 6. Declaration that the aquifer is clean
   or still contaminated based on data collected
   during Stage 5.
                          Cohen et al. (1994) and U.S. EPA (1992)
                        address in more detail the types of statistical
                        techniques that are required to analyze short-term
                        and long-term trends in contaminant concentra-
                        tions.
                                      52

-------
                                                                              Methods
  Numerous variations and enhancements of
pump-and-treat systems are possible. Major types
include
  • Using trenches or drains in combination with
    or to replace vertical pumping wells (Section
    7.1). Where site conditions are favorable (i.e.,
    shallow contamination), trenches are a
    commonly used method for intercepting
    contaminated ground water.
  • Using horizontal wells or trenches to replace
    or complement vertical wells (Section 7.1).
    Recent developments in directional drilling
    technology make the use of horizontal or
    inclined wells an attractive alternative ap-
    proach.
  • Inducing fractures in the subsurface to
    improve the yield of wells (Section 7.1).
    Although widely used by the petroleum
    industry, the use of induced fractures is
    considered an emerging technology in
    ground-water remediation with applications
    limited to contaminated ground water in low-
    permeability materials.
  • Implementing vadose zone source control and
    remediation, often as a necessary adjunct to
    ground-water cleanup (Section 7.2).
  • Making chemical enhancements, which can
    have the potential to accelerate aquifer
    remediation (Section 7.3).
  • Making biological enhancements, which can
    present opportunities for eliminating or
    reducing the requirements for surface treat-
    ment of contaminated ground water (Section
    7.4).
  Notably few alternatives to pump-and-treat
systems are without requirements for continuous
energy input for pumping fluids (Section 7.4).
7.1.   Alternative Methods for Fluid
        Delivery and Recovery
  Conventional  pump-and-treat systems usually
involve extraction wells—and possibly injection
wells—placed vertically in an aquifer. Alternative
methods of delivery and recovery of contaminated
ground water might enhance the performance of a
pump-and-treat  system, especially while interim
measures are undertaken, by improving the
effectiveness of containment. These methods also
might augment the performance of a variety of
remedial actions selected as possible long-term
remedies. Major alternatives include
  • Interceptor Trenches. After vertical wells,
    trenches are the most widely used method for
    controlling subsurface fluids and recovering
    contaminants. They function similarly to
    horizontal wells, but also can have a signifi-
    cant vertical component, which cuts across
    and can allow access to the permeable layers
    in interbedded sediments. For shallower
    applications, trenches can be installed at
    relatively low cost using conventional equip-
    ment. Recent innovations combine trench
    excavation and well screen installation into a
                                      53

-------
    single step for depths up to 20 feet (U.S. EPA,
    1994). Where depth is not a constraint,
    interceptor trenches are generally superior to
    vertical wells. In such situations, they are
    especially effective in low-permeability
    materials and heterogeneous aquifers.
  * Horizontal and Inclined Wells. Relatively
    recent advances in directional drilling tech-
    nology, which use specialized bits to curve
    bores in a controlled arc, have revolutionized
    the field of well design. Directional drilling
    methods can create wellbores with almost any
    trajectory. Wells that curve to a horizontal
    orientation are especially suited to environ-
    mental applications (Figure 34).
  «Induced Fractures. EPA research has shown
    that petroleum engineering technology used
    to induce fractures for increased productivity
    of oil wells also can improve the performance
    of environmental wells. Induced fractures are
    used mainly where low-permeability aquifer
    materials create problems  for the recovery of
    contaminants.
  Table 4 rates the potential applications of
alternative methods for delivery or recovery of
subsurface fluids in relation to (1) access, (2)
depth. (3) recovered phases, (4) geology, and (5)
availability. Figure 35 illustrates two ways in
which horizontal wells or trenches can be used to
intercept a contaminant plume. In many applica-
tions, deciding between use of a trench or a
horizontal well hinges on economic rather than
technical issues, with trenches generally being
more cost effective at depths less than 20 feet and
horizontal wells being generally more cost
effective at depths greater than 20 feet. Cost
savings can be substantial compared to vertical
well systems. For example,  initial remediation
plans at a site in North Carolina called for 100
vertical wells to recover a hydrocarbon plume at
an estimated cost of $1 million. Instead, a con-
tinuous excavation and completion system was
installed for less than $350,000 (U.S. EPA, 1994).
EPA's Manual Alterative Methods for Fluid
Delivery and Recovery (U.S. EPA, 1994) provides
more detailed information on design consider-
ations and applications of these methods.
7.2.                            Control
  Removal of contaminants from the vadose
(unsaturated) zone is an essential part of any
remedial action plan to clean up contaminated
ground water. Major methods include
   * Capping to reduce infiltration of precipitation.
   * Excavation to remove contaminated soil for
    ex situ treatment, which is most commonly
    used where contaminants have not penetrated
    deeply into the subsurface.
   • Soil vapor extraction (SVE), which is used to
    extract volatile organic contaminants by
    flushing with air, and bioventing, a SVE
    system in which the addition of nutrients
    further enhances the biodegradation of
    organic contaminants. Both techniques.
    considered innovative technologies a few
    years ago. arc widely used.
   • In situ thermal technologies to enhance the
    mobility' of volatile and semivolatile organic
    contaminants; for example, steam-enhanced
    extraction and radio frequency heating are
    promising innovative technologies.
7.3.   Physical      Chemical  En-
       hancements
  Physical and chemical enhancements to pump-
and-treat systems primarily function by enhancing
the mobility of contaminants, thus increasing their
recover}? in ground water that has been pumped to
the surface for treatment. Some chemical en-
hancements transform contaminants in place in
the subsurface to reduce toxicity.
                                      54

-------
            (b)
 VL  /   /   I   I \  \  /  / j ^ I
!il/M'J^Y\/f^
\
                  \  I' V v' A
           (d)
                          1
Figure 34. Some applications of horizontal wells: (a) intersecting flat-lying layers, (b) intercepting plume elongated
      by regional gradient, (c) intersecting vertical fractures, and (d) access beneath structures (U.S. EPA,
      1994).
                         55

-------
Table 4. Issues Affecting Application of Alternative Methods for Delivery or Recovery (U.S. EPA, 1994)

     Issue                     Horizontal Well                              Induced Fracture
                                                                                                                        Trench
Access

Fragile structures
over target

Poor access over
target

Depth

<6m
6-20 m


>20 m


Recovered Phase

Aqueous

LNAPL
*    Minimal surface disturbance
*    Standoff required
                       *   1 m minimum depth
*    Cost of guidance system
     increases at >6 m

*    No depth limit within
     environmental applications
                       •••   Requires accurate drilling;
                            best if water table fluctuations
                            are minor
•    Evaluate effects of surface
     displacement

*•    Possible with horizontal
     well
                                                1-2 m minimum depth
     No depth limit within
     environmental applications
                                                Best with access to
                                                individual fractures
    Excavation expected to be
    infeasible

    Excavation expected to be
    infeasible
*   Installation with common
    equipment

ifi   Excavation costs increase
    with depth

4-   Specialized excavation
    methods required
                                         *  Widely used to ensure
                                            capture; accommodates
                                            water table fluctuations
DNAPL
                            Requires accurate drilling and
                            site characterization
                                                Caution; steeply dipping
                                                fractures may cause
                                                downward movement
                                         *  Assuming mobile phase
                                            present and accurately
                                            located
Vapor
•    Consider omitting gravel pack
     to save costs
     Best with access to
     individual fractures
    Requires tight seal on top
    of trench
Geology
Normally
consolidated clay
Swelling clay
Silty clay till
Stratified sediment
or rock
® Smearing of bore wall may
reduce performance
® Smearing of bore wall may
reduce performance
® Smearing of bore wall may
reduce performance
•*• Anisotropy may limit vertical
influence of well
*• Induced fractures may be
vertical and limited in size
• Relatively large, gently
dipping fractures expected
• Relatively large, gently
dipping fractures expected
® Stratification may limit
upward propagation and
increase fracture size
«
«
«
«
Large discharge expected
relative to alternatives
Large discharge expected
relative to alternatives
Large discharge expected
relative to alternatives
Good way to access many
thin beds or horizontal
partings
                                                                                                               (Continued)
                                                                    56

-------
Table 4.  (Continued)

     Issue
                                Horizontal Well
                                                                           Induced Fracture
                                                                                                                         Trench
Vertically fractured
sediment or rock
Coarse gravel
Thick sand
*   Orient well normal to
     fractures when possible
     Possible problems with hole
     stability; penetrating cobbles

     May be difficult to access top
     and bottom of formation; hole
     stability problems
     Good where induced
     fractures cross-cut natural
     fractures
     (overconsolidated
     sediment and rock)

     Permeability enhancement
     may be unnecessary

     Permeability enhancement
     may be unnecessary
    Orient trench
    perpendicular to natural
    fractures when possible
*  Stability a concern during
    excavation

*  Stability a concern during
    excavation
Rock
                            Feasible, but drilling costs
                            more in rock than in sediment
                                                 Widely used in oil, gas,
                                                 and water wells drilled in
                                                 rock
                                             Excavation difficult but
                                             blasting possible to make
                                             trench-like feature
Availability
Current Experience
(Approximate)
10 to 20 companies with capabilities;
nationwide coverage but may require
equipment mobilization
                        150 to 250 wells at 50 to 100 sites
Several companies offer service;
nationwide coverage with
equipment mobilization
                                            200 to 400 fractures at 20 to 40
                                            sites
Shallow trench (<6 m)
installation widely available
from local contractors; deep
trench will require mobilization

1,000+ trenches at many
hundreds of sites
Key
 • Good application
 ® Moderately good
 4> Fair, with possible technical difficulties
 • Poor; not recommended using available methods
                                                                     57

-------
                     (a)
                             Horizontal Well
                          Capture Zone
                                                            Source
Figure 35. Two approaches using trenches or horizontal wells to intercept contaminant plumes (U.S. EPA, 1994).
                                        58

-------
7.3.1.    Physical Enhancements
  Air sparing, also known as in situ aeration, is
an approach that is similar to soil vapor extrac-
tion except that air is injected into the saturated
zone rather than the vadose zone (Figure 36). Air
sparging systems can effectively remove a
substantial amount of volatile aromatic and
chlorinated hydrocarbons in a variety of geologic-
settings, but significant questions remain about
the ability of this technology to achieve health-
based standards throughout the  saturated zone
(NRC, 1994). Thermal enhancements, such as
steam and hot-water flooding, increase the
mobility of volatile and semivolatile contami-
nants. Use of induced fractures (Section 7.1) is
another form of physical enhancement to pump-
and-treat systems.
7.3,2.    Chemical Enhancements
  Chemically enhanced pump-and-trcats systems
require use of injection wells to deliver reactive
agents to the contaminant plume and extraction
wells to remove reactive  agents and contaminants
(Figure 37). The major types of chemical en-
hancements are
   • Soil flushing, which enhances recover}' of
   contaminants with low water solubility, free-
   product and residual NAPLs, and sorbed
   contaminants. Two major types of chemical
   agents can be used: (1) cosolvents. which,
   when mixed with water, increase the solubil-
   ity of some organic compounds, and (2)
   surfactants, which may cause contaminants
   to desorb and may increase NAPL mobility
   by lowering the interfacial tension between
   the NAPL and water, increasing the solubil-
   ity. Soil flushing is one of the most promis-
   ing innovative technologies for dealing with
   separate phase  DNAPLs in the subsurface
   (NRC, 1994).
   * In situ chemical treatment, which involves
    reactive agents that oxidize or reduce con-
    taminants, converting them to nontoxic forms
    or immobilizing them to minimize contami-
    nant migration. This innovative technology is
    still in the early stages of development.
  The EPA report Chemical Enhancements to
Pump-and-Treat Remediation (Palmer and Fish,
1992) provides additional information on techni-
cal issues related to this topic.
7.4.
  Biological enhancements to pump-and-treat
systems stimulate subsurface microorganisms,
primarily bacteria, to degrade contaminants to
harmless mineral end products, such as carbon
dioxide and water. In situ bio remediation of
certain types of hydrocarbons (primarily petro-
leum products and derivatives), encouraged by
addition of oxygen and nutrients to the ground
water, is an established technology. Other readily
biodegradable substances, such as phenol, cresols,
acetone, and cellulosic wastes, are also amenable
to aerobic in situ bioremediation. Key elements in
such a system are deliver}7 of oxygen and nutri-
ents by use of an injection well (Figure 38a) or an
infiltration gallery (Figure 38b). A limitation of in
situ bioremediation is that minimum contaminant
concentrations required to maintain microbial
populations may exceed health-based cleanup
standards, particularly where heavier hydrocar-
bons are involved.
  In situ bioremediation of chlorinated solvents is
less well demonstrated because metabolic pro-
cesses for their degradation are more complex
than those for hydrocarbon degradation (NRC.
1994). Nonetheless, methanotrophs are able to
degrade some chlorinated solvents under aerobic
conditions if methane is supplied as an energy
source. Also,  the ability of anaerobic bacteria to
degrade a variety of chlorinated solvents is well
                                      59

-------
                                                                                                 Vent to
                                                                                                 Atmosphere
                                                                                             Surface Soil/Cap
                                                                                                        Vapor
                                                                                                        Treatment
                                                                                             Unsaturated Zone
                                                                                             Saturated Zone
                                  Streamtubes of Air
Direction of Ground-
Water Flow
                                   (b)
Figure 36.
Process diagram
for air sparging
with (a) vertical
wells, and (b)
horizontal wells
(after NRC,
1994).
                                               	Injection Point for Flushing Gas
                                            I   f^- Extraction of Contaminated Gas

                                           Y   I

                                                                                             Surface Soil/Cap
                                                                                             Unsaturated Zone
                Saturated Zone
                                                             60

-------
                                        Injection
                                         of Reactive
                                        Agents

DISPOSAL

	 | Recovery of |
• Reactive Agent |

— |—
Treatment

                             Extraction of
                             Reactive Agents
                             and Contaminants
Figure 37.
Schematic of chemical
enhancement of a pump-and-
treat system.  Key areas of
concern are shown in boxes.
In some cases, the reactive
agent will be recovered and
reused (Palmer and Fish,
1992).
Delivery —>• Reaction —> Removal-
                 documented. Two major obstacles to the use of
                 anaerobic processes for in situ bioremediation are
                 that (1) hazardous intermediate degradation
                 products can accumulate, and (2) undesirable
                 water quality changes, such as dissolution of iron
                 and manganese, can occur.
                   EPA reference sources identified at the end of
                 this guide (Section 9) that are particularly relevant
                 to in situ bioremediation include Norris et al.
                 (1993), Sims et al. (1992), and U.S. EPA (1993,
                 1994).
               7.5.    Alternatives to the Pump-and-
                      Treat Approach
                Nearly all approaches to ground-water cleanup
               involve some degree of ground-water pumping.
               Even when containment is the primary objective,
               low-flow pump-and-treat systems are usually
               required to prevent the escape of contaminated
               water from the confined area. Two remediation
               approaches that eliminate pumping as a compo-
               nent of the system are (1) intrinsic bioremedia-
               tion, and (2) in situ reactive barriers. Although
               both of these methods show promise, they are still
                                                      61

-------
                                      (a)
                                 To Sewer or
                                 Recirculate
                                                                                                    Sparger
                                       (b)
                                                 Air Compressor or
                                                 Hydrogen Peroxide
                                                 Tank
                                                                               Nutrient Addition
                                          Infiltration Gallery
Figure 38.
Two types of
aerobic in situ
bioremediation
systems: (a)
injection well with
sparger, (b)
infiltration gallery
(Sims et al.,
1992, after
Thomas and
Ward, 1989).
      Trapped Hydrocarbons     ,., .
                               Table
   Recirculated Water
   and Nutrients
                   Recovery Well
Monitoring Well
                                                             62

-------
in development and their effectiveness remains to
be demonstrated.
7.5,1.   Intrinsic Bioremediation
  Intrinsic biorcmcdiation relies on indigenous
microbes to biodegrade organic contaminants,
without human intervention in the form of supply-
ing electron acceptors, nutrients, and other
materials. The processes that occur are the same
as those in engineered bioremediation systems.
but they occur more slowly. A decision to refrain
from active site manipulation does not eliminate
the need to conduct ground-water sampling within
the contaminant plume to document that biodeg-
radation is occurring. Moreover, sampling would
still need to be performed outside the contami-
nated area to identify any offsite migration of
contaminants that might require initiation of more
active remedial measures (Figure 39b). There is a
greater risk of failure with intrinsic bioremedia-
tion compared to engineered bioremediation
because no active measures are used to control the
contaminant plume. The possible perception that
intrinsic bioremediation is the equivalent to doing
nothing is also a barrier to its acceptance (NRC,
1994).
7,5.2.   In Situ Reactive Barriers
  The concept of using permeable in situ reactive
barriers to treat a contaminant plume as it moves
through an aquifer under natural hydraulic
gradients (Figure 39c and 39d) was first suggested
by McMurty and Elton (1985), but it has only
recently begun to receive significant attention
from the research community (Starr and Cherry,
1994). The funnel-and-gate concept, which
combines impermeable barriers to contain and
channel the flow of the contaminant plume toward
the reactive barrier has received the most attention
because numerous possible configurations can be
developed to address different types of contami-
nant plumes and geologic settings (Figure 40).
Depending on the contaminants present in the
plume, the reactive zone uses a combination of
physical, chemical, and biological processes.
  The great promise of in situ reactive barriers is
that they will  require little or no energy input once
installed, yet provide more active control and
treatment of the contaminant plume than intrinsic
bioremediation. The main engineering challenges
involve provision of suitable amounts of reactive
materials in a permeable medium and proper
placement to avoid short-circuiting the contact
between the gate and the cutoff wall.
                                      63

-------
                                   Contaminant Source
                                   Zone
                                                 Extraction Well
                     (b)
                                           In Situ Reaction Curtain'
                                                         Remediated
                                                          Plume
                     (d)
In Situ Reactor
'Gate1
Figure 39. Alternative ground-water plume management options: (a) pump-and-treat system, (b) intrinsic
          bioremediation, (c) in situ reaction curtain, (d) funnel-and-gate system (adapted from Starr and
          Cherry, 1994).
                                         64

-------
              (a)
                                                   Single Gate System
              (b)
                            Multiple Gate System              Multiple Reactor Systems
                     Fully Penetrating Gate
Hanging Gate
Figure 40. Funnel-and-gate configurations (Starr and Cherry, 1994).
                                          65

-------
Bartow, G. and C. Davenport. 1995. Pump-and-
    Treat Accomplishments: A Review of the
    Effectiveness of Ground Water Remediation
    in Santa Clara Valley, California. Ground
    Water Monitoring and Remediation
    15(2): 140-146.

Berglund, S. andV. Cvetkovic. 1995. Pump-and-
    Treat Remediation of Heterogeneous Aqui-
    fers: Effects of Rate-Limited Mass Transfer.
    Ground Water 33(4):675-685.

Blandford, T.N. and P.S. Huyakorn. 1991. WHPA:
    Modular Semi-Analytical Model for the
    Delineation of Wellhead Protection Areas,
    Version 2.0. Office of Ground Water Protec-
    tion; Available from EPA Center for Subsur-
    face Modeling Support, Ada, OK. Version 1.0
    was released in 1990 [Four modules:
    MWCAP,  RESSQC, GPTRAC, MONTEC;
    most current disk version is 2.1]

Bradbury, K.R., M.A. Muldoon, A. Zaporozec,
    and J. Levy. 1991. Delineation of Wellhead
    Protection Areas in Fractured Rocks. EPA/
    570/9-91-009. Office of Water, Washington,
    DC. 144 pp.

Cohen, RM. and J.W Mercer. 1993. DNAPL Site
    Evaluation. EPA/600/R-93/002 (NTIS PB93-
    150217). R.S. Kerr Environmental Research
    Laboratory, Ada, OK. [Also published by
    Lewis Publishers as C.K. Smoley edition,
    Boca Raton, FL. 384 pp.]
Cohen, R.M., A.H. Vincent, J.W. Mercer, C.R.
    Faust, and C.P. Spalding. 1994. Methods for
    Monitoring Pump-and-Treat Performance.
    EPA/600/R-94/123. R.S. Kerr Environmental
    Research Laboratory, Ada, OK. 102 pp.

Clausen, J.L. and D.A. Solomon. 1994. Character-
    ization of Ground Water Plumes and DNAPL
    Sources Using a Driven Discreet-Depth
    Sampling System. Ground Water Manage-
    ment 18:435-445 (Proc. of 8th Nat. Outdoor
    Action Conf. on Aquifer Remediation,
    Ground Water Monitoring and Geophysical
    Methods).

Feldman, PR and D.J. Campbell. 1994. Evaluat-
    ing the Technical Impracticality of Ground-
    Water Cleanup. Ground Water Management
    18:595-608 (Proc. of 8th Nat. Outdoor Action
    Conf. on Aquifer Remediation, Ground Water
    Monitoring and Geophysical Methods).

Freeze, RA. and J.A. Cherry. 1989. What Has
    Gone Wrong? Ground Water 27(4):458-464.

Gillham, R.W., E.A. Sudicky, J.A. Cherry, and
    E.O. Frind. 1984. An Advective-Diffusion
    Concept to Solute Transport in Heterogeneous
    Unconsolidated Geologic Deposits. Water
    Resour. Res. 20(3):369-378.

Gore lick, S.M., R.A. Freeze, D. Donohue, and
    J.F. Keely. 1993. Groundwater Contamina-
    tion: Optimal Capture and Containment.
    Lewis Publishers: Boca Raton, FL. 416 pp.
                                    66

-------
Haitjema, H.M., J. Wittman, V. Kelson, and N.
    Bauch. 1994. WhAEM: Program Documenta-
    tion for the Wellhead Analytic Element
    Model. EPA/600/R-94/2K), 120 pp. Available
    from EPA Center for Subsurface Modeling
    Support. Ada, OK. [Includes Geographic
    Analytic Element Preprocessor (GAEP) and
    Capture Zone Analytic Element Model
    (CZAEM)]

Haley, J.L., B. Hanson, C. Enfield, and J. Glass.
    1991. Evaluating the Effectiveness of Ground
    Water Extraction Systems. Ground Water
    Monitoring Rev. 11(1): 119-124. [Summary of
    U.S. EPA (1989)]

Harvey. C.F., R. Haggcrty, and S.M. Gorclick.
    1994. Aquifer Remediation: A Method for
    Estimating Mass Transfer Rate Coefficients
    and an Evaluation of Pulsed Pumping. Water
    Rcsour.  Res. 30(7): 1979-1991.

Hoffman, F. 1993. Ground-Water Remediation
    Using ''Smart Pump  and Treat." Ground
    Water31(l):98-106.

Keely, J.F. 1989. Performance Evaluation of
    Pump-and-Trcat Re-mediations. Superfund
    Issue Paper. EPA/540/8-89/005. R.S. Kerr
    Environmental Research Laboratory, Ada,
    OK. 14pp.

Knox, R.C.. L.W. Canter, D.F. Kincannon, E.L.
    Stover, and C.H. Ward. 1984. State-of-the Art
    of Aquifer Restoration. EPA/600/2-84/
    182a&b (National Technical Information
    Service  [NTIS]  PB85-181071 and PB85-
    181089). R.S. Kerr Environmental Research
    Laboratory, Ada, OK.

Mackay, D.M. and J.A. Cherry.  1989. Groundwa-
    ter  Contamination: Pump-and-Treat Remedia-
    tion. Environ. Sci. Technol. 23(6):630-636.
Marquis, Jr., S. 1995. Don't Give Up on Pump
    and Treat: Enhance It with Bioremediation.
    Soils & Groundwater Cleanup, August-
    September, pp. 46-50.

McMurty, D.C., and R.O. Elton. 1985. New
    Approach to In-Situ Treatment of Contami-
    nated Groundwaters. Environ. Progress
    4(3): 168-170.

National Research Council (NRC). 1994. Alterna-
    tives for Ground Water Cleanup. National
    Academy Press. 336 pp.

Norris, R.D. et al. 1993. In-Situ Bioremediation
    of Ground Water and Geological Material: A
    Review of Technologies. EPA/600/R-93/124
    (NTIS  PB93-215564). R.S. Kerr Environmen-
    tal Research Laboratory, Ada, OK. [13
    authors; see also Norris et al., 1994]

Palmer, C.D. and W. Fish. 1992. Chemical
    Enhancements to Pump-and-Treat Remedia-
    tion. Ground Water Issue Paper. EPA/540/S-
    92/001. R.S. Kerr Environmental Research
    Laboratory, Ada, OK. 20 pp.

Papadopulos & Associates, Inc. and Conestoga-
    Rovers & Associates Ltd. 1993. Chem-Dyne
    Site  Trust Fund; 1992 Annual Report. Chem-
    Dyne Site, Hamilton, OH. April.

Piwom, M.D. and J.W. Kcclcy. 1990. Basic
    Concepts of Contaminant Sorption at Hazard-
    ous Waste Sites. Ground Water Issue. EPA/
    540/4-90/053. R.S.  Kerr Environmental
    Research Laboratory, Ada, OK. 7 pp.

Rogers, L.L., R.U. Dowla, and V.M. Johnson.
    1995. Optimal Field-Scale Groundwater
    Remediation Using Neural Networks and
    Genetic Algorithm. Environ. Sci. Technol.
    29(5): 1145-1155.
                                    67

-------
Roy, W.R., I.G. Krapac, S.F.J. Chou, and R.A.
    Griffin. 1992. Batch-Type Procedures for
    Estimating Soil Adsorption of Chemicals.
    EPA/530/SW-87/006F (NTIS PB92-146190).
    Risk Reduction Engineering Laboratory,
    Cincinnati, OH. 100 pp.

Satkm, R.L. and P.B. Bedient. 1988. Effectiveness
    of Various Aquifer Restoration Schemes
    Under Variable Hydrogeologic Conditions.
    Ground Water 26(4):488-498.

Sims, J.L., J.M. Suflita, andH.H. Russell. 1992.
    Tn-Situ Bioremediation of Contaminated
    Ground Water. Ground Water Issue Paper.
    EPA/540/S-92/003. R.S. Kerr Environmental
    Research Laboratory, Ada, OK. 11 pp.

Starr,  R.C. and J.A. Cherry. 1994. In Situ Reme-
    diation of Contaminated Ground Water: The
    Funnel-and-Gate System. Ground Water
    32(3):465-476.

Strack, O.D.L. et al. 1994. CZAEM User's Guide:
    Modeling Capture Zones of Ground-Water
    Wells Using Analytic Elements. EPA/600/R-
    94/174, 58  pp. Available from EPA Center for
    Subsurface Modeling Support, Ada, OK. [Sec
    also, Haitjema et al., 1994]

Thomas, J.D., and C.H. Ward. 1989. In Situ
    Biorestoration of Organic Contaminants in
    the Subsurface. Environ. Sci. Technol.
    23:760-786.

Travis, C.C. and C.B. Doty. 1990. Can Contami-
    nated Aquifers at Superfund  Site Be
    Remediated? Environ. Sci. Technol.
    24(1): 1464-1466.

U.S. Environmental Protection Agency (EPA).
    1988. Guidance on Remedial Actions for
    Contaminated Ground WTater at Superfund
    Sites. EPA/540/G-88/003. Office of Solid
    Waste and Emergency Response (OSWER)
    Directive 9283.1-2 (NTIS PB89-184618).
    Office of Solid Waste and Emergency Re-
    sponse, Washington, DC.

U.S. Environmental Protection Agency (EPA).
    1989. Evaluation of Ground-Water Extraction
    Remedies: Volume 1, Summary Report (EPA/
    540/2-89/054, NTIS PB90-183583, 66 pp.);
    Volume 2, Case Studies 1-19 (EPA/540/2-89/
    054b); and Volume 3, General Site Data Base
    Reports (EPA/540/2-89/054c). Office of Solid
    Waste and Emergency Response, Washington,
    DC.

U.S. Environmental Protection Agency (EPA).
    1991. Handbook: Stabilization Technologies
    for RCRA Corrective Actions. EPA/625/6-91/
    026. Center for Environmental Research
    Information, Cincinnati, OH. 62 pp.

U.S. Environmental Protection Agency (EPA).
    1992. Methods for Evaluating the Attainment
    of Cleanup Standards, Volume 2: Ground
    Water. EPA/230/R-92/014. Office of Solid
    Waste and Emergency Response, Washington,
    DC.

U.S. Environmental Protection Agency (EPA).
    1993. Guidance for Evaluation the Technical
    Impracticability of Ground-Water Restora-
    tion. EPA/540/R-93/080, OSWER 0234.2-25
    (NTIS PB93-963507). Office of Solid Waste
    and Emergency Response, Washington, DC.

U.S. Environmental Protection Agency (EPA).
    1994. Manual: Alternative Methods for Fluid
    Deliver and Recovery. EPA/625/R-94/003.
    Center for Environmental Research Informa-
    tion, Cincinnati, OH. 87 pp.
                                    68

-------
U.S. Environmental Protection Agency (EPA).      Wilson. J.L., S.H. Conrad, W.R. Mason. W.
    1995. Manual: Ground-Water and Leachate         Peplinski and E. Hagen. 1990. Laboratory
    Treatment Systems. EPA/625/R-94/005.            Investigation of Residual Liquid Organics.
    Center for Environmental Research Informa-       EPA/600/6-90/004. R.S. Kerr Environmental
    tion. Cincinnati, OH.  119 pp.                     Research Laboratory, Ada, OK. 267 pp.
                                     69

-------
,
9.    EPA Publications Providing Further Informatio
               The EPA publications listed below provide more
                  detailed information on the subjects discussed
                  in this document. Publications and additional
                  copies of this brochure can be obtained at no
                  charge (while supplies are available) from the
                  following sources:

               EPA/625-series documents: Office of Research
                  and Development (ORD) Publications, P.O.
                  Box 19968, Cincinnati, OH 45219-0968;
                  phone 513 569-7562, fax 513 569-7562.

               Other EPA documents: National Center for
                  Environmental Publications and Information
                  (NCEPI), 11029 Kenwood Road, Cincinnati,
                  OH 45242; fax 513 891-6685.

               Other documents, for which an NTIS acquisition
                  number is shown can be obtained from the
                  National Technical Information Service
                  (NTIS), Springfield, VA 22161; 800 336-
                  4700, fax 703/321-8547.

               Contaminant Transport and Fate
               Huling, S.G. 1989. Facilitated Transport. Ground
                  Water Issue. EPA/540/4-89/003. R.S. Ken-
                  Environmental Research Laboratory, Ada,
                  OK. 5 pp.

               Huling, S.C. and J.W. Weaver. 1991. Dense
                  Nonaqueous Phase Liquids. Ground Water
                  Issue. EPA/540/4-91/002. R.S. Kerr Environ-
                  mental Research Laboratory, Ada, OK. 21 pp.
McLean, J.E. and B.E. Bledsoe. 1992. Behavior
    of Metals in Soils. Ground Water Issue. EPA/
    540/S-92/018. R.S. Kerr Environmental
    Research Laboratory, Ada, OK. 25 pp.

Palmer, C.D. and R.W. Puls. 1994. Natural
    Attenuation of Hexavalent Chromium in
    Ground Water and Soils. Ground Water Issue.
    EPA/540/S-94/505. R.S. Kerr Environmental
    Research Laboratory, Ada, OK. 13 pp.

Piwoni, M.D. and J.W. Keeley. 1990. Basic
    Concepts of Contaminant Sorption at Hazard-
    ous Waste Sites. Ground Water Issue. EPA/
    540/4-90/053. RS. Kerr Environmental
    Research Laboratory, Ada, OK. 7 pp.

Sims, J.L., J.M. Suflita, andH.H. Russell. 1991.
    Reductive Dehalogenation of Organic Con-
    taminates in Soils and Ground Water. Ground
    Water Issue. EPA/540/4-91/054. RS. Ken-
    Environmental Research Laboratory, Ada,
    OK. 12 pp.

Wilson, J.L., S.H. Conrad, W.R Mason, W.
    Peplinski, and E. Hagen. 1990. Laboratory
    Investigation of Residual Liquid Organics.
    EPA/600/6-90/004. RS. Kerr Environmental
    Research Laboratory, Ada, OK. 267 pp.

Site Characterization
U.S. Environmental Protection Agency (EPA).
    1991. Site Characterization for Subsurface
    Remediation. EPA/625/4-91/026. Center for
                                                   70

-------
    Environmental Research Information, Cincin-
    nati, OH. 259 pp.

U.S. Environmental Protection Agency (EPA).
    1992. Estimating the Potential for the Occur-
    rence of DNAPL at Supcrfund Sites. OSWER
    Publication 9355.4-07/FS. Office of Solid
    Waste and Emergency Response, Washington,
    DC.

U.S. Environmental Protection Agency (EPA).
    1993. Evaluation of the Likelihood of
    DNAPL Presence at NPL Sites. EPA/540/R-
    93/002 (OSWER 0355.4-13). Office of Solid
    Waste and Emergency Response, Washington,
    DC.

U.S. Environmental Protection Agency (EPA).
    1993. Use of Airborne, Surface and Borehole
    Geophysical Techniques at Contaminated
    Sites: A Reference Guide. EPA/625/R-92/007.
    Center for Environmental Research Informa-
    tion, Cincinnati, OH.

U.S. Environmental Protection Agency (EPA).
    1993. Subsurface Characterization and
    Monitoring Techniques: A Desk Reference
    Guide; Vol. I: Solids and Ground Water; Vol.
    II: The Vadose Zone, Field Screening and
    Analytical Methods. EPA/625/R-93/003a&b.
    Center for Environmental Research Informa-
    tion, Cincinnati, OH.

Pump-and-Treat
Cohen, R.M., A.H. Vincent, J.W. Mercer, C.R.
    Faust, and C.P. Spalding. 1994. Methods for
    Monitoring Pump-and-Treat Performance.
    EPA/600/R-94/123. R.S. Kerr Environmental
    Research Laboratory,  Ada, OK. 102 pp.

Keely, J.F. 1989. Performance Evaluation of
    Pump-and-Treat Remediations. Superfund
    Issue Paper. EPA 540/8-89/005. R.S. Kerr
    Environmental Research Laboratory, Ada,
    OK. 14 pp.

Mercer, J.W., D.C. Skipp, and D. Giffin. 1990.
    Basics of Pump-and-Treat Ground-Water
    Remediation Technology. EPA/600/8-90/003.
    R.S. Kerr Environmental Research Labora-
    tory, Ada, OK. 58 pp.

Palmer, C.D. and W. Fish. 1992. Chemical
    Enhancements to Pump-and-Treat Remedia-
    tion. Ground Water Issue  Paper. EPA/540/S-
    92/001. R.S. Kerr Environmental  Research
    Laboratory. Ada, OK. 20 pp.

Repa, E. and D.P. Doerr. 1985. Leachate Plume
    Management. EPA/540/2-85/004  (NTIS
    PB86-122330). Hazardous Waste Engineering
    Research Laboratory, Cincinnati,  OH.

U.S. Environmental Protection Agency (EPA).
    1989. Evaluation of Ground-Water Extraction
    Remedies: Volume  1, Summary' Report (EPA/
    540/2-89/054, NTIS PB90-183583, 66 pp.);
    Volume 2, Case Studies 1-19 (EPA/540/2-89/
    054b); and Volume  3, General Site Data Base
    Reports (EPA/540/2-89/054c). Office of Solid
    Waste and Emergency Response,  Washington.
    DC.

U.S. Environmental Protection Agency (EPA).
    1992. Evaluation of Ground-Water Extraction
    Remedies, Phase II. Oswer Publication
    9355.4-05, Vols. 1-2. Office of Solid Waste
    and Emergency Response, Washington, DC.

U.S. Environmental Protection Agency (EPA).
    1992. General Methods for Remedial Opera-
    tions Performance Evaluation. EPA/600/R-92/
    002. R.S. Kerr Environmental Research
    Laboratory. Ada, OK. 37 pp.

U.S. Environmental Protection Agency (EPA).
    1993. Guidance for Evaluating the Technical
                                    71

-------
    Impracticability of Ground-Water Restora-
    tion. EPA/540/R-93/080, OSWER 0234.2-25
    (NTIS PB93-963507). Office of Solid Waste
    and Emergency Response, Washington, DC.

U.S. Environmental Protection Agency (EPA).
    1993. Bioremediation Resource Guide. EPA/
    542/B-93/004. Office of Solid Waste and
    Emergency Response, Washington, DC.
    [Includes annotated list of more than 80
    significant references]

U.S. Environmental Protection Agency (EPA).
    1994. Bioremediation in the Field. EPA/540/
    N-94/501. Office of Solid Waste and Emer-
    gency Response, Washington. DC. [Periodi-
    cally updated; latest issue No. 11, July, 1994]

U.S. Environmental Protection Agency (EPA).
    1.994. Manual: Alternative Methods for Fluid
    Delivery and Recovery.  EPA/625/R-94/003.
    Center for Environmental Research Informa-
    tion, Cincinnati, OH. 87 pp.

U.S. Environmental Protection Agency (EPA).
    1995. In Situ Remediation Technology Status
    Report: Hydraulic and Pneumatic Fracturing.
    EPA/542/K-94/005. Office of Solid Waste and
    Emergency Response, Washington, DC. 15
    pp.

Ground-Water Treatment
Canter, L.W. and R.C. Knox. 1986. Ground Water
    Pollution Control. Lewis Publishers: Chelsea,
    MI. 526 pp. [Contains mostly same material
    as Knox etal. (1984)1

Knox, R.C., L.W. Canter, D.F.  Kincannon, E.L.
    Stover, and C.H. Ward. 1984. State-of-the Art
    of Aquifer Restoration. EPA 600/2-84/
    182a&b (NTIS PB85-181071 and  PB85-
    1.81089). R.S. Kerr Environmental Research
    Laboratory, Ada, OK. [See also Canter and
    Knox (1985)]
McArdle, J.L., M.M. Arozarena, and W.E.
    Gallagher. 1987. A Handbook on Treatment
    of Hazardous Waste Leachate. EPA/600/8-87/
    006 (NTIS PB87-152328). Hazardous Waste
    Engineering Research Laboratory, Cincinnati,
    OH.

U.S. Department of Defense Environmental
    Technology Transfer Committee (DOD/
    ETTC). 1994. Remediation Technologies
    Screening Matrix and Reference Guide. EPA/
    542/B-94/013 (NTIS PB95-104782). Office
    of Solid Waste and Emergency Response,
    Washington, DC.

U.S. Environmental Protection Agency (EPA).
    1994. Ground-Water Treatment Technology
    Resource Guide. EPA/542/B-94/009. Office
    of Solid Waste and Emergency Response,
    Washington, DC. [Includes annotated list of
    more than 60 significant references]

U.S. Environmental Protection Agency (EPA).
    1994. Innovative Treatment Technologies
    Annual Status Report, 6th ed. EPA/542/R-94/
    005. Office of Solid Waste and Emergency
    Response. Washington, DC.

U.S. Environmental Protection Agency (EPA).
    1994. Superfund Innovative Technology-
    Evaluation Program: Technology Profiles, 7th
    ed. EPA/540/R-94/526. Risk Reduction
    Engineering Laboratory, Cincinnati, OH. 499
    pp.

U.S. Environmental Protection Agency (EPA).
    1995. Manual:  Ground-Water and Leachate
    Treatment Systems. EPA/625/R-94/005.
    Center for Environmental Research Informa-
    tion, Cincinnati, OH.
                                    72

-------
In Situ Ground-Water Treatment
Norris, R.D. et al. 1993. In-Situ Bioremediation
    of Ground Water and Geological Material: A
    Review of Technologies. EPA/600/R-93/124
    (NTIS PB93-215564). R.S. Kerr Environmen-
    tal Research Laboratory. Ada, OK.[13 au-
    thors; see  also Norris et al., 1994]

Norris, R.D. et al. 1.994. Handbook of Bioreme-
    diation. Boca Raton, FL: Lewis Publishers.
    272 pp. [Contains same material as Norris et
    al., 1993J

Sims, J.L., J.M. Suflita, and H.H. Russell. 1992.
    In Situ Bioremediation of Contaminated
    Ground Water. Ground Water Issue Paper.
    EPA/540/S-92/003. R.S. Kerr Environmental
    Research Laboratory, Ada, OK. 11 pp.

U.S. Environmental Protection Agency (EPA).
    1995a. In  Situ Remediation Technology
    Status Report: Thermal Enhancements. EPA/
    542/K-94/009. Office of Solid Waste and
    Emergency Response, Washington, DC. 22
    pp.

U.S. Environmental Protection Agency (EPA).
    1995b. In  Situ Remediation Technology
    Status Report: Surfactant Enhancements.
    EPA/542/K-94/003. Office of Solid Waste and
    Emergency Response, Washington, DC. 22
    pp.

U.S. Environmental Protection Agency (EPA).
    1995c. In  Situ Remediation Technology
    Status Report: Cosolvents. EPA/542/K-94/
    006. Office of Solid Waste and Emergency
    Response, Washington. DC. 6 pp.

U.S. Environmental Protection Agency (EPA).
    1995d. In  Situ Remediation Technology
    Status Report: Electrokinetics. EPA/542/K-
    94/007. Office of Solid Waste and Emergency
    Response, Washington, DC. 20 pp.
U.S. Environmental Protection Agency (EPA).
    1995e. In Situ Remediation Technology
    Status Report: Treatment Walls. EPA/542/K-
    94/004. Office of Solid Waste and Emergency
    Response. Washington, DC. 26 pp.

Ground-Water
Bear, J., M.S. Bcljm, and R.R. Ross. 1992.
    Fundamentals of Ground-Water Modeling.
    Ground Water Issue. EPA/540/S-92/005. R.S.
    Kerr Environmental Research Laboratory,
    Ada, OK.  11 pp.

Schmelling, S.G. and R.R. Ross. 1989. Contami-
    nant Transport in Fractured Media: Models
    for Decisionmakers. Ground Water Issue.
    EPA/540/4-89/004. R.S. Kerr Environmental
    Research Laboratory. Ada, OK. 8 pp.

U.S. Environmental Protection Agency (EPA).
    1988. Selection  Criteria for Mathematical
    Models Used in Exposure Assessments:
    Ground-Water Models. EPA/600/8-88/075
    (NTIS PB88-248752). Office of Health and
    Environmental Assessment. Washington, DC.
    [Contains  summary' tables and descriptions of
    63 analytical solutions and 49 analytical  and
    numerical codes for evaluating ground-water
    contaminant transport]

U.S. Environmental Protection Agency (EPA).
    1994. Assessment Framework for Ground-
    Water Model Applications. EPA/500/B-94/
    003 (OSWER Directive 9029.00). Office of
    Solid Waste and Emergency Response,
    Washington, DC. 41 pp.

van der Heijde, P.K.M. 1994. Identification and
    Compilation of Unsaturated/Vadose Zone
    Models. EPA/600/R-94/028 (NTIS PB94-
    157773). R.S. Kerr Environmental Research
    Laboratory, Ada, OK.
                                    73

-------
van der Heijde, P.K.M. and O.A. Einawawy. 1993.      Ada, OK. [Summary information on models
    Compilation of Ground-Water Models' EPA/       for porous media flow and transport,
    600/R-93/118 (NTIS PB93-209401). R.S.          hydrogeochemical models, stochastic models,
    Kcrr Environmental Research Laboratory,          and fractured rock]
                                    74

-------
                                        PB94-159100
                                       EPA/600/R-93/182
                                       September 1993
         Technical Guidance Document:

QUALITY ASSURANCE AND QUALITY CONTROL
    FOR WASTE CONTAINMENT FACILITIES
                David E. Daniel
          University of Texas at Austin
         Department of Civil Engineering
              Austin, Texas 78712

                     and

              Robert M. Koerner
         Geosynthetic Research Institute
        Wert Wing, Rush Building No. 10
        Philadelphia, P-nnsylvania  19104
    Cooperative Agreement No. CR-S15546-01-0
                Project Oflicer

                David A. Carson
      Risk Reduction Engineering Laboratory
       Office of Research ana i>weiopment
      U.S. Environmental Protection Agency
            Cincinnati, Ohio 45268
RISK REDUCTION ENGINEERING LABORATORY
  OFFICE OF RESEARCH AND DEVELOPMENT
 U.S. ENVIRONMENTAL PROTECTION AGENCY
          CINCINNATI, OHIO 45268
                                      Printod on Racy&ed Paper

-------
                                   DISCLAIMER
      Hie information in the document has been funded wholly or in part by the United States
Environmental Protection Agency under assistance agreement number CR-815546-01-0. It has
been subject to the Agency's peer and administrative review and has been approved for publication
as a U.S. EPA document  Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

      This document contains numerous references to various procedures for performing tests as
part of the process of quality control and quality assurance. Standards published by the American
Society  for Testing and Materials (ASTM) are referenced wherever possible because ASTM
procedures represent consensus standards. Other testing procedures referenced in this document
were generally developed by an individual or a small group of individuals and, therefore, do not
represent consensus standards. The mention of non-consensus standards does not constitute their
endorsement.

      The reader is cautioned against using this document for the direct preparation  of site
specific  quality assurance plans or related documents without giving proper consideration to the
site- and project-specific requirements. To do so would ignore the educational context of the
accompanying text, innovations made since the publication of the document, and the prevailing
unique and site-specific aspects of all waste containment facilities.
                                          il

-------
                                   FOREWORD
      Today's  rapidly developing  and  changing technologies  and  Industrial
products and practices frequently carry with them  the  increased  generation of
materials that, If Improperly dealt with, can threaten both public  health and the
environment.  The United States  Environmental  Protection  Agency  (U.S.  EPA) is
charged by Congress with protecting the Nation's land, air, and water resources.
Under a mandate of national environmental laws, the Agency strives to formulate
and implement actions leading to a compatible balance between human activities
and the  ability of  natural  systems to  support and nurture life.   These laws
direct the U.S.  EPA to perform  research to define  our  environmental problems,
measure the impacts,  and search for solutions.

      The Risk  Reduction  Engineering  Laboratory is responsible  for planning,
Implementing, and managing research, development, and demonstration programs to
provide  an  authoritative,  defensible  engineering  basis  in  support  of  the
policies, programs,  and regulations of the  U.S. EPA with respect  to drinking
water, wastewater, pesticides, toxic substances, solid and hazardous wastes, and
Superfund-related activities.  This publication is  one  of the products of that
research and provides a vital communication link between the researcher and the
user community.

      This document provides information needed  to develop comprehensive quality
assurance plans and to carry out quality control procedures at waste containment
sites.  It discusses  quality  assurance  and quality control issues  for compacted
soil  liners,  soil drainage  systems,  gRosynthetic  drainage  systems,  vertical
cutoff walls, ancillary materials, and appurtenances.
                                                             E. Timothy Oppelt
                                                                      Director
                                         Risk Reduction Engineering Laboratory
                                      ill

-------
                                   ABSTRACT
      This Technical Guidance Document provides comprehensive guidance on
procedures for quality assurance and quality control for waste containment
facilities.  The document Includes a discussion of principles and concepts,
compacted soil liners, soil drainage systems, geosynthetlc drainage systems,
vertical cutoff walls, ancillary materials, appurtenances, and other details.
The guidance document outlines critical quality assurance (QA) and quality
control (QC) Issues for each major segment and recommends specific procedures,
observations, tests, corrective actions, and record keeping requirements.  For
geosynthetlcs, QA and QC practices for both manufacturing and construction ire
suggested.

      The main body of the text details recommended procedures for quality
assurance and control.  Appendices Include a 11st of acronyms, glossary, and
Index.  A companion document was under development by the American Society for
Testing and Materials (ASTH) at the time of this writing that will contain all
of the ASTN standards referenced 1n this guidance document as well as most, 1f
not all, of the other test procedures that are referenced 1n this guidance
document.

      This report was submitted 1n fulfillment of CR-815546 by the University
of Texas, Austin, under the sponsorship of the U.S. Environmental Protection
Agency.  This report covers a period from June 1991 to July 1993, and work was
completed as of August 1993.
                                   1v

-------
                                          \
                                 Table of ^Contents
Disclaimer
Abstract
List of Figures
List of Tables
Acknowledgments

Chapter 1  Manufacturing Quality Assurance (MQA) and Construction Quality
             Assurance (CQA) Concepts and Overview
       1.1  Introduction
              1.1.1
              1.1.2
Scope
Definitions
       1.2  Responsibility and Authority
       1.3  Personnel Qualifications
       1.4  Written MQA/CQA Plan
       1.5  Documentation

              1.5.1   Daily Inspection Reports
              1.5.2   Daily Summary Reports
              1.5.3   Inspection and Testing Reports
              1.5.4   Problem Identification and Corrective Measures Reports
              1.5.5   Drawings of Record
              1.5.6   Final Documentation and Certification
              1.5.7   Document Control
              1.5.8   Storage of Records

       1.6  Meetings

              1.6.1   Pre-Bid Meeting
              1.6.2   Resolution Meeting
              1.6.3   Pre-construction Meeting
              1.6.4   Progress Meetings

       1.7  Sample Custody
       1.8  Weather
       1.9  Work Stoppages
       1.10 References

 Chapter 2  Compacted Soil Linen

       2.1  Introduction and Background

              2.1.1   Types of Compacted Soil Liners

                    2.1.1.1 Natural Mineral Materials
                                                        Page No.

                                                             n
                                                            in
                                                            iv
                                                            xii
                                                           xix
                                                            xx

                                                             1
 1
 2

 3
10
11
11

11
11
12
13
13
14
14
14

15

15
15
16
17

17
17
17
18

19

19

19

19

-------
             2.1.1.2  Bentonite-Soil Blends
             2.1.1.3  Other

      2.1.2   Critical CQC and CQA Issues
      2.1.3   Liner Requirements

             2.1.3.1   Subgrade Preparation
             2.1.3.2   Material Selection
             2.1.3.3   Preprocessing
             2.1.3.4   Placement, Rt.nolding, and Compaction
             2.1.3.5   Protection
             2.1.2.6   Final Surface Preparation

      2.1.4   Compaction Requirements

             2.1.4.1   Compaction Curve
             2.1.4.2   Compaction Tests
             2.1.4.3   Percent Compaction
             2.1.4.4   Estimating Optimum Water Content and Maximum
                      Dry Unit Weight

                   2.1.4.4.1 Subjective Assessment
                   2.1.4.4.2 One-Point Compaction Test
                   2.1.4.4.3 Three-Point Compaction Test (ASTM D-5080)
             2.1.4.5
      mended Procedure for Developing Water
Content-Density Specification
       2.1.5   Test Pads

2.2  Critical Construction Variables that Affect Soil Liners

       2.2. 1   Properties of the Soil Material

             2.2. 1 . 1   Plasticity Characteristics
             2.2. 1 .2   Percentage Fines
             2.2.1.3   Percentage Gravel
             2.2. 1.4   Maximum Particle Size
             2.2. 1 .5   Clay Content and  Activity
             2.2.1.6   Clod Size
             2.2.1.7   Bemonite

       2.2.2   Molding Water Content
       2.2.3   Type of Compaction
       2.2.4   Energy of Compaction
       2.2.5   Bonding of Lifts
       2.2.6   Protection Against Desiccation and Freezing

2.3  Field Measurement of Water Content and Dry Unit Weight

       2.3.1   Water Content Measurement
                                                    19
                                                    19

                                                    21
                                                    21

                                                    22
                                                    23
                                                    23
                                                    24
                                                    24
                                                    24

                                                    24

                                                    24
                                                    26
                                                    27

                                                    27

                                                    28
                                                    28
                                                    29
30

34

35

35

35
37
38
39
39
39
40

42
46
48
52
53

53

53
                                   vi

-------
2.5
2.6
             2.3.1.1  Overnight Oven Drying (ASTMD-2216)
             2.3.1.2  Microwave Oven Drying (ASTM D-4643)
             2.3.1.3  Direct Heating (ASTM EM959)
             2.3.1.4  Calcium Carbide Gas Pressure Tester (ASTM EM944)
             2.3.1.5  Nuclear Method (ASTM D-3017)
       2.3.2   Unit Weight

             2.3.2.1  Sand Cone (ASTM D-1556)
             2.3.2.2  Rubber Balloon (ASTM D-2167)
             2.3.2.3  Drive Cylinder (ASTM D-2937)
             2.3.2.4  Nuclear Method (ASTM D-2922)

2.4  Inspection of Borrow Sources Prior to Excavation

       2.4.1   Samp'- ig for Material Tests
       2.4.2   Mau,iud Tests

             2.4.2.1  Water Content
             2.4.2.2  Atterberg Limits
             2.4.2.3  Particle Size Distribution
             2.4.2.4  Compaction Curve
             2.4.2.5  Hydraulic Conductivity
             2.4.2.6  Testing Frequency
     Inspection during Excavation of Borrow Soil
     Preprocessing of Materials
       2.6.1
       2.6.2
       2.6.3
       2.6.4
       2.6.5
              Water Content Adjustment
              Removal of Oversize Particles
              Pulverization of Clods
              Homogenizing Soils
              Bentonite
             2.6.5.1
                      PugmiU Mixing
             2.6.5.2  In-Place Mixing
             2.6.5.3  Measuring Bentonite Content

       2.6.6  Stockpiling Soils

2.7  Placement of Loose Lift of Soil

       2.7.1   Surface Scarification
       2,7.2  Material Tests and Visual Inspection

             2.7.2.1  Material Tests
             2.7.2.2  Visual Observations
             2.7.2.3  Allowable Variations
             2.7.2.4  Corrective Action

       2.7.3   Placement and Control of Loose Lift Thickness
53
53
54
54
54

56

56
57
57
58

61

61
61

62
63
63
63
63
65

65
67

67
68
68
68
68

70
70
70

72

72

73
73

73
73
73
75

75
                                   vii

-------
2.8  Remolding and Compaction of Soil
       2,8.1   Compaction Equipment
       2.8.2   Number of Passes
       2.8.3   Water Content and Dry Unit Weight
                                                             76
                                                             77
                                                             78
             2.8.3.1   Water Content and Unit Weight Tests                    78
             2.8.3.2   Sampling Patterns                                    78
             2.8.3.3   Tests with Different Devices to Minimize Systematic Errors 80
             2,8.3.4   Allowable Variations ard Outliers                       80
             2.8.3.5   Corrective Action                                     82
       2,8.4   Hydraulic Conductivity Tests on Undisturbed Samples

             2.8.4.1   Sampling for Hydraulic Conductivity Testing
             2.8.4.2   Hydraulic Conductivity Testing
             2.8.4.3   Frequency of Testing
             2.8.4.4   Outliers

       2.8.5   Repair of Holes from Sampling and Testing
       2.8.6   Final Lift Thickness
       2.8.7   Pass/Fail Decision

2.9  Protection of Compacted Soil

       2.9.1   Desiccation

             2.9.1.1   Preventive Measures
             2.9.1.2   Observations
             2.9.1.3   Tests
             2.9.1.4   Corrective Action

       2.9.2   Freezing Temperatures

             2.9.2.1   Compacung Frozen Soil
             2.9.2.2   Protection After Freezing
             2.9.2.3   Investigating Possible Frost Damage
             2.9.2.4   Repair

       2.9.3   Excess Surface Water
2,10 Test Pads

       2.10.1
       2.10,2
       2.10.3
       2.10.4
       2.10.5
       2.10.6
       2.10.7
Purpose of Test Pads
Dimensions
Materials
Construction
Protection
Tests and Observations
In Situ Hydraulic Conductivity
             2.10.7.1  Sealed Double-Ring Fnfiltromettr
82

83
84
84
84

85
85
85

86

86

86
86
86
86

86

86
87
87
88

88

88

88
89
89
89
90
91
91

91
                                                                                 K?^r
                                                                                ft
                                   vni

-------
                   2,10.7.2 Two-Stage Borehole Test
                   2,10.7.3 Other Field Tests
                   2,10.7.4 Laboratory Tests

             2.10.8  Documentation

      2,11 Final Approval
      2.12 References

Chapter 3 Gcomembranes

      3.1  Types of Geomembranes and Their Formulations

             3.1.1   High Density Polyethylene (HOPE)

                   3.1.1.1  Resin
                   3.1.1.2  Carbon Black
                   3.1.1.3  Additives

             3.1.2   Very Low Density Polyethylene (VLDPE)
                   3.1.2.1
                   3.1.2,2
                   3.1.2.3
         Resin
         Carbon Black
         Additives
             3.1.3
             3.1.4
 Other Extruded Geomembranes
 Polyvinyl Chloride (PVC)

3.1.4.1   Resin
3.1.4.2   Plasocizer
3.1,4.3   Filler
3.1.4.4   Additives
             3.1.5   Chlorosulfonated Polyethylene (CSPE-R)

                   3.1.5.1   Resin
                   3.1.5.2   Carbon Black
                   3.1.5.3   Fillers
                   3.1.5.4   Additives
                   3.1.5.5   Reinforcing Scrim

             3.1.6   Other Calendered Geomembranes
      3.2  Manufacturing
             3.2.1   Blending, Compounding, Mixing and/or Masticating
             3.2.2   Regrind, Reworked or Trim Reprocessed Material
             3.2.3   High Density Polyethylene (HDPE)

                   3.2.3.1   Flat Die - Wide Sheet
                   3.2.3.2   Flat Die - Factory Seamed
                   3.2.3.3   Blown Film
 93
 95
 95

 95

 95
 95

 99

 99

100

100
101
103

103

104
104
105

105
106

106
106
107
107

107

107
109
109
109
109

110

110

111
111
113

114
115
115
                                         IX

-------
               3.2.3.4  Textured Sheet

        3.2.4  Very Low Density Polyethylene (VLDPE)

               3.2.4.1  Rat Die - Wide Sheet
               3.2.4.2  Flat Die - Factory Seamed
               3.2.4.3  Blown Film
               3.2.4.4  Textured Sheet

        3.2.5  Cocxtnision Processes
        3.2.6  Polyvinyl Chloride (PVQ

               3.2.6.1  Calendering
               3.2.6.2  Panel Fabrication

        3.2.7  CMorosulfonated Polyethylene-Scrim Reinforced (CSPE-R)

              3.2.7.1   Calendering
              3.2.7.2  Panel Fabrication

        3.2.8   Spread Coated Geomembranes

 3.3  Handling

       3.3.1   Packaging

              3.3.1.1   Roils
              3.3.1.2   Accordion Folded

       3.3.2   Shipment, Handling and Site Storage
       3.3.3   Acceptance and Confonnance Testing
       3.3.4   Placement

              3.3.4.1   Subgrade (Subbase) Conditions
              3.3.4.2  Temperature Effects - Sticking/Cracking
              3.3.4.3   Temperature Effects • Expansion/Contraction
              3.3.4.4   Spotting
              3.3.4.5   Wind Considerations

3.4  Seaming and Joining

       3.4.1   Overview of Field Seaming Methods
       3,4.2   Details of Field Seaming Methods
       3.4.3   Test Strips and Trial Seams

3.5  Destructive Test Methods for Seams

       3.5.1   Overview
       3.5.2   Sampling Strategies

             3.5.2,1   Fixed Increment Sampling
             3.5.2.2   Randomly Selected Sampling
  117

  120

  120
  120
  121
  121

  123
  124

  124
  127

  129

  129
  131

  131

  132

 132

 132
 133

 134
 135
 136

 136
 138
 139
 140
 140

 141

 141
 145
 146

 150

 150
 151

151
152

-------
                   3.5.2.3   Other Sampling Strategies                             153

             3.5.3   Shear Testing of Geomembrane Seams                          153
             3.5.4   Peel Testing of Geomembrane Seams                           157
             3.5.5   General Specification Items                                    160

      3.6  Nondestructive Test Methods for Seams                                  161

             3.6.1   Overview                                                   161
             3.6.2   Currently Available Methods                                   162
             3.6.3   Recommendations for Various Seam Types                      165
             3.6.4   General Specification Items                                    165

      3.7  Protection and Backfilling                                              167

             3.7.1   Soil Backfilling of Geomembranes                             167
             3.7.2   Geosyntheric Covering of Geomembranes                       170
             3.7.3   General Specification Items                                    171

      3.8  References                                                           172

Chapter 4 Geosynthetic Clay Liners                                                174

      4.1  Types and Composition of Geosynthetic Clay Liners                       174
      4.2  Manufacturing                                                       176

             4.2.1   Raw Materials                                               176
             4.2.2   Manufacturing                                               177
             4.2.3   Covering of the Rolls                                        179

      4.3  Handling                                                            180

             4.3.1   Storage at the Manufacturing Facility                           180
             4.3.2   Shipment                                                   181
             4.3.3   Storage at the Site                                           181
             4.3.4   Acceptance and Conformance Testing                           184

      4.4  Installation                                        ^                  185
                                                             -^
             4.4.1   Placement                                                  185
             4.4.2   Joining                                                     187
             4.4.3   Repairs                                                     187

      4.5  Backfilling or Covering                                                188
      4.6  References                                                          189

Chapters Soil Drainage Systems                                                  191

      5.1  Introduction and Background                                          191
      5.2  Materials                                                            191
      5.3  Control of Materials                                                   195
      5.4  Location of Borrow Sources                                           196

                                          xi

-------
       S.S  Processing of Materials
       5.6  Placement

             5.6,1   Drainage Layers
             5.6.2   Drainage Trenches

       5.7  Compaction
       3*8  Protection
       5.9  References

Chapter 6  Geosynmetic Drainage Systems
       6.1
       6.2
Overview
Geotextiles

  6.2.1   Manufacturing of Geotextiles

         6.2.1.1  Resins and Their Additives
         6.2.1.2  Fiber Types
         6.2.1.3  Geotextile Types
         6.2.1.4  General Specification Items

  6.2.2   Handling of Geotextiles
                   6.2.2.1
                   6.2.2.2
                   6.2.2.3
                   6.2.2.4
                   6.2.2.5
                   6.2.2.6

             6.2.3   Seaming

                   6.2.3.1
                   6.2.3.2
                   6,2.3.3
                 Protective Wrapping
                 Storage at Manufacturing Facility
                 Shipment
                 Storage at Field Site
                 Acceptance and Conformance Testing
                 Placement
                 Seam Types and Procedun
                 Seam Tests
                 Repairs
            6.2.4   Backfilling or Coveting

      6.3  Geonets and Geonet/Geotextile Gcocomposites

            6.3.1   Manufacturing of Geonets
            6.3.2   Handling of Geonets

                   6.3.2.1   Packaging
                   6.3.2.2   Storage at Manufacturing Facility
                   6.3.2.3   Shipment
                   6.3.2.4   Storage at the Site
                   6.3.2.5   Acceptance and Conformance Testing
                   6.3.2.6   Placement

            6.3.3   Joining of Geonets
 1%
 197

 197
 197

 198
 199
 201

 202

 202
 202

 202

 204
 206
 207
 207

 210

 210
 210
 212
 212
 212
 213

 214

 214
 217
 217

 217

 218

 218
 221

 222
 222
 222
 222
 222
223

223
                                         xii

-------
             6.3.4   Geonet/Geotextile Geocomposites

      6. 4  Other Types of Geocomposites

             6.4. 1   Manufacturing of Drainage Composites
             6.4.2   Handling of Drainage Geocomposites

                   6.4.2.1   Packaging
                   6.4.2.2   Storage at Manufacturing Facility
                   6.4.2.3   Shipment
                   6.4.2.4   Storage at Field Site
                   6.4.2.5   Acceptance and Conformance Testing
                   6.4.2.6   Placement

             6.4.3   Joining of Drainage Geocomposites
             6.4.4   Covering

      6.5  Reference.,

Chapter? Vertical Cutoff Walls

      7.1  Introduction
      7.2  Types of Vertical Cutoff Walls

             7.2.1   Sheet Pile Walls
             7.2.2   Geomcmbrane Walls
             7.2.3   Walls Constructed with Slurry Techniques

      7.3  Construction of Slurry Trench Cutoff Walls

             7.3.1   Mobilization
             7.3.2   Site Preparation
             7.3.3   Slurry Preparation and Properties
             7.3.4   Excavation of Slurry Trench
             7.3.5   Soil-Bentonite (SB) Backfill
             7.3.6   Cement-Bemonite (CB) Cutoff Walls
             7.3.7   Geomembrane in Slurry trench cutoff walls
             7.3.8   Other Backfills
             7.3.9   Caps
7.4  Other Types of Cutoff Walls
7.5  Specific CQA Requirer
7.6  Post Construction Tests tor v-
7.7  References
                                        .inuity
Chapter 8 Ancillary Materials, Appurtenances, and Other Details

       8. 1   Plastic Pipe (aka "Geopipe")

             8.1.1   Polyvinyl Chloride (PVC) Pipe
             8. 1 .2   High Density Polyethylene (HDPE) Smooth Wall Pipe
             8.1.3   High Density Polyethylene (HDPE) Corrugated Pipe
                                                                         225

                                                                         226

                                                                         22*
                                                                         231

                                                                         231
                                                                         231
                                                                         231
                                                                         231
                                                                         231
                                                                         232

                                                                         232
                                                                         233

                                                                         233

                                                                         235

                                                                         235
                                                                         237

                                                                         237
                                                                         238
241

241
241
241
243
244
248
250
250
251

251
251
252
252

253

253

254
256
257
                                         xiii

-------
       8.1.4  Handling of Plastic Pipe
              8.1.4.1  Packaging
              8.1.4.2  Storage at Manufacturing Facility
              8.1.4.3  Shipment
              8.1.4.4  Storage ai Held Site
       8.1.5  Conformance Testing and Acceptance
       8.1.6  Placement
8,2  Sumps, Manholes and Risen
8.3  Liner System Penetrations
8.4  Anchor Trenches
       8.4.1   Geomembranes
       8.4.2   Other Geosynthetics
8.5  Access Ramps
8.6  Geosynthetic Reinfi
                                  it Materials
             8.6.1   Geotexriles for Reinforcement
             8.6.2   Geogrids
       8.7  Geosynthetic Erosion Control Materials
       8.8  Floating Geomembrane Covers for Surface Impoundments
       8,9  References
Appendix A Listof Acroynms
Appendix B Glossary
Appendix C Index
 258
 259
 259
 259
 259
 259
 261
 261
 264
 266
 266
 268
 268
 270
 272
 273
 275
 278
 280
282
285
301
                                 xiv

-------
                                   List  of Figures
Figure    Title                                                              Page No.

1.1       Organizational Structure of MQA/CQA Inspection Activities                    4

2.1       Examples of Compacted Soil Liners in Liner and Cover Systems               20
2.2       Tie-In of New S-il Liner to Existing Soil Liner                              22
2.3       Compaction Curve                                                      25
2.4       One-Point Compaction Test                                               29
2.5       Form of Water Content-Dry  Unit Weight Specification Often Used in the Past    30
2.6       Recommended Procedure to  Determine Acceptable Zone of Water Content/
          Dry Unit Weight Values Based Upon Hydrai.ae Conductivity Considerations    32
2.7       Acceptable Zone of Water Content/Dry Unit Weights Determined by
          Superposing Hydraulic Conductivity and Shear Strength Data                  33
2.8       Relationship between Hydraulic Conductivity and Plasticity Index              36
2.9       Relationship between Hydraulic Conductivity and Percent Fines                37
2.10      Relationship between Hydraulic Conductivity and Percentage Gravel Added
          to Two Clayey Soils                                                     38
2.11      Relationship between Hydraulic Conductivity and day Content                40
2,12      Relationship between Clay Content and Plasticity Index                       41
2.13      Effect of Addition of Bentonitc to Hydraulic Conductivity of Compacted
          S" iy Sand                                                             42
2.14      Effect of Molding Water Content on Hydraulic Conductivity                   43
2.15      Photograph of Hi ghly Plastic Clay Compacted with Standard Proctor Effort
          at a Water Content of 16% (1 % Dry of Optimum)                            44
2,16      Photograph of Highly Plastic Clay Compacted with Standard Proctor Effort
          at a Water Content of 20% (3% Wet of Optimum)                            45
2.17      Four Types of Laboratory Compaction Tests                                46
2,18      Effect of Type of Compaction on Hydraulic Conductivity                      47
2.19      Footed Rollers with Partly and Fully Penetrating Feet                         48
2.20      Effect of Compactive Energy on Hydraulic Conductivity                      49
2.21      Illustration of Why Dry Unit Weight Is a Poor Indicator of Hydraulic
          Conductivity for Soil Compacted Wet of Optimum                           50
2.22      Line of Optimums                                                       51
2.23      Flow Pathways Created by Poorly Bonded Lifts                             52
2.24      Schematic Diagram of Nuclear Water Content • Density Device                 55
2.25      Sand Cone Device                                                      55
2.26      Schematic Diagram of Rubber Balloon  Device                               58
2.27      Schematic Diagram of Drive Ring                       %.                59
2.28      Measurement of Density with Nuclear Device by (a) Direct Transrr Ission and
          (B) Backscattenng                                                      60
2.29      Recommended Procedure for Preparation of a Test Specimen Using Variable
          (But Documented) Compactive Energy for Each Trial                         64
2.30      Schematic Diagram of Pugmill                                            71
2.31      Schematic Diagram of Soil Liner Test Pad                                  90
2.32      Schematic Diagram of Sealed Double Ring Infiltrometer (SDRI)                91
2.33      Three Procedures for Computing Hydraulic Gradient from Infiltration Test      92
2.34      Schematic Diagram of Two-Stage Borehole Test                             94

3.1       HDPE Resin Pellets                                                    100
3.2       Carbon Black in Paniculate Form and as a Concenuate                      102
                                          XV

-------
3.3      CSPE Resin Pieces                                                   108
3.4      Photographs of Materials to be Reprocessed                               112
3.5      Cross-Section Diagram of a Horizontal Single-Screw Extruder for
         Polyethylene                                                         113
3.6      Photograph of a Polyethylene Geomembrane Exiting from a Relatively
         Narrow Flat Horizontal Die                                            114
3.7      Photograph of Blown Film Manufacturing of Polyethylene Geomembranes
         and Sketch of Blown Film Manufacturing of Polyethylene Geomembranes      116
3.8      Various Methods Currently Used to Create Textured Surfaces on HDPE
         Geomembranes                                                      118
3.9      Geotncmbrane Surface Temperature Differences Between Black and
         White Colors                                                         123
3.10     Sketches of Various Process Mixers                                      125
3.11     Various Types of Four-Roll Calenders                                   126
3.12     Photographs of Calendered Rolls of Geomembranes After Manufacturing
         and Factory Fabrication of Rolls into Large Panels for Field Deployment       128
3.13     Multiple-Ply Scrim Reinforced Geomembrane                              130
3.14     Rolls of Polyethylene Awaiting Shipment to a Job Site                      133
3.15     Photograph of Truck Shipment of Geomembranes                          135
3.16     Photographs Showing the Unrolling (Upper) and Unfolding (Lower) of
         Geomembranes                                                      137
3.17     HDPE Geomembrane Showing Sun Induced Wrinkles                      139
3.18     Wind Damage to Deployed Geomembrane                                141
3.19     Various Methods Available to Fabricate Geomembrane Seams                143
3.20     fabrication of a Geomembrane Test Strip                                 147
3.21     Photograph of a Field Tensiometer Performing a Geomembrane Seam Test     148
3.22     Test Strip Process  Flow Chan                                          149
3.23     Completed Patch on a Geomembrane Seam Which had Previously Been
         Sampled for Destructive Tests                                           152
3.24     Shear Test of a Geomembrane Seam Evaluated in a CQC/CQA Laboratory
         Environment                                                         154
3.25     Peel Test of a Geomembrane Seam Evaluated in a CQC/CQA Laboratory
         Environment                                                         158
3.26     Advancing Primary Leachate Collection Gravel in "Fingers" Over the
         Deployed Geomembrane                                               169

4.1      Cross Section Sketches of Currently Available Geosymhetic Clay Linen
         (GCLs)                                                             175
4.2      Schematic Diagrams of the Manufacture of Different Types of Gcosynthetic
         Clay Liners (GCLs)                                          x        178
4.3      Indoor Factory Storage of Geosynthetic Clay Liners (GCLs) Waiting for
         Shipment to a Job Site                                                 180
4.4      Fork Lift Equipped with a "Stinger" and GCL Rolls on a Flat-Bed Trailer      182
4.5      Photograph of Temporary Storage of GCLs in their Shipping Trailers and
         at Project Site                                                        183
4.6      Field Deployment of a GCL on a Soil Subgrade and an Underlying
         Geosynthetic                                                         186
4.7      Premature Hydration of a Geosynthetic Gay Liner Being Gathered and
         Discarded  due to its Exposure to Rainfall  Before Covering                   189

5.1      Grain Size Distribution Curve                                           192
5.2      Filter Layer Used to Protect Drainage Layer from Plugging                  194
                                        xvi

-------
               5.3      Typical Design of a Drainage Trench                                      198
               5.4      CQC and CQA Personnel Observing Placement of Select Waste on
                        Drainage Layer                                                        200

               6.1      Cross Section of a Landfill Illustrating the Use of Different Geosyntherics
                        Involved in Waste Containment Drainage Systems                          203
               6.2      Polyester Resin Chips and Carbon Black Concentrate Pellets Used for
                        Geotextile Fiber Manufacturing                                          205
               6.3      Types of Polymeric Fibers Used in the Construction of Different Types of
                        Geotextiles                                                           206
               6.4      Three Major Types of Geotex tiles                                        208
               6.5      Photographs of Temporary Storage of Geotextiles                          211
               6.6      Various Types of Sewn Seams for Joining Geotextiles                       215
               6.7      Fabrication of a Geotextile Field Seam in a "Flat" or "Prayer" Seam Type       216
               6.8      Typical Geonets Used in Waste Containment Facilities                       219
               6.9      Counter Rotating Die Technique for Manufacturing Drainage Geonets md
                        Example of Laboratory Prototype                                        220
               6.10     Geonets Being Temporarily Stored at the Job Site                           223
               6.11     Photograph of Geonet Joining by Using Plastic Fasteners                    224
               6.12     Various Types of Drainage Geocornposites                                 22"?
               6.13     Vacuum Forming System for Fabrication of a Drainage Geocomposiie         229
               6.14     Photograph of Drainage Core Joining via Male-to-Female Interlock            232

               7.1      Example of Vertical Cutoff Wall to Limit Flow of Ground Waier into
                        Excavation                                                           235
               7.2      Example of Vertical Cutoff Wall to Limit Flow of Ground Water through
                        Buried Waste                                                         236
               7.3      Example of Vertical Cutoff Wall to Restrict Inward Migration of Ground
                        Water                                                                236
               7.4      Example of Vertical Cutoff Wall to Limit Long-Term Contaminant Transport    236
               7.5      Interlocking Steel Sheet Piles                                            237
               7.6      Examples of Interlocks for Geomembrane Walls                            238
               7.7      Hydrostatic Pressure from Slurry Maintains Stable Walls of Trench            239
               7.8      Diagram of Construction Process for Soil-Bentoniie-Backfilled Slurry Trench
                        Cutoff Wall                                                          240
               7.9      Construction of Dike to Raise Ground Surface for Construction of Slurry
                        Trench                                                              242
               7,10     Backhoe for Excavating Slurry Trench                                    244
               7.11     Clamshell for Excavating Slurry Trench                                   245
               7.12     Mixing Backfill with Bentonite Slurry                                    246
               7.13     Pushing Soil-Bentonite Backfill Into Slurry Trench with  Dozet               247
               7.14     Examples of Problems Produced by  Improper Backfilling of Slurry Trench     248
               7.15     Diaphragm-Wall Construction                                           249

               8.1      Cross Section of a Possible Removal Pipe Scheme in a Primary Lcachate
                        Collection and Removal System                                         253
               8.2      Plan View of a Possible Removal Pipe Scheme in a Primary Lcachate
                        Collection and Removal System                                         254
               8.3      Photograph of PVC Pipe to be Used in a Landfill Leach ate Collection System   255
               8.4      Photograph of HOPE Smooth Wall Pipe Risers Used as Primary and
                        Secondary Removal Systems from Sump Area to Pump and Monitoring
                        Station                                                              256
                                                       xvii
L_

-------
8.5       Photograph of HDPE Corrugated Pipe Being Coupled and After Installed      258
8.6       A Possible Buried Pipe Trench Cross Section Scheme Showing Soil Backfill
          Terminology and Approximate Dimensions                                262
8.7       Various Possible Schemes for Leachate Removal                           263
8.8       Pipe Penetrations through Various Types of Barrier Materials                 265
8.9       Various Types of Geomembrane Anchors Trenches                         267
8.10      Typical Access Ramp Geometry and Cross Section                         269
8.11      Geogrid or Geotextile Reinforcement of (a) Cover Soil above Waste.
          (b) Ltachate Collection Layer beneath Waste, and (c) Liner System Placed
          above Existing Waste ("Piggybacking")                                   271
8.12      Photographs of Geogrids Used as Soil (or Waste) Reinforcement Materials     274
8.13      Examples of Geosynthetic Erosion Control Systems                         277
8.14      Surface Impoundments with Geomembrane Floating Covers along with
          Typical Details of the Support System and/or Anchor Trench and Batten Strips  279
                                        xviii

-------
Table
                                   List or Tables
Title
PaeeNo.
i. 1       Recommended Implementation Program for Construction Quality Control
          (CQQ for Geosynthetics (Beginning January 1,1993)                         8
1.2       Recommended Implementation Program for Construction Quality Assurance
          (CQA) for Geosynthetics (Beginning 'anuaiy 1,1993)                         9
1.3       Recommended Personnel Qualifications                                    10

2.1       Compaction Test Details                                                 26
2.2       Materials Tests                                                         62
2.3       Recommended Minimum Testing Frequencies for Investigation of Borrow
          Source                                                                65
2.4       Criteria for Describing Dry Strength (ASTM D-2488)                         66
2.5       Criteria for Describing Plasticity (ASTM D-2488)                            67
2.6       Recommended Tests on Bentonite to Determine Bentonite Quality and Gradation  70
2.7       Recommended Tests to Verify Bentonite Content                            72
2.8       Recommended Materials Tests for Soil Liner Materials Sampled after
          Placement in a Loose Lift (Just Before Compaction)                          74
2.9       Recommended Maximum Percentage of Failing Material Tests                  75
2,10      Recommended Tests and Observations on Compacted Soil                     79
2.11      Recommended Maximum Percentage of Failing Compaction Tests              82

3.1       Types of Commonly Used Geomembranes and Their Approximate
          Weight Percentage Formulations                                          99
3.2       Fundamental Methods of Joining Polymeric Genniembranes                  142
3.3       Possible Fseld Seaming Methods for Various Geomembranes Listed in
          this Manual                                                          145
3.4       Recommended Test Method Details for Geomembrane Seams in Shear and
          in Peel and for Unseamed Sheet                                          155
3.5       Nondestructive Geomembrane Seam Testing Methods                        163
3.6       Applicability Of Various Nondestructive Test Methods To Different Seam
          Types And Geomembrane Types                                         166
3.7       Critical Cone Heights For Selected  Geomembranes In Simulated Laboratory
          Puncture St udies                                                      167
3.8       Coefficients Of Thermal Expansion/Contraction Of Various Nonreinforced
          Geomembrane Polymers                                               170

5.1       Effect of Fines on Hydraulic Conductivity of a Washed Filter Aggregate        193
5.2       Recommended Tests and Testing Frequencies for Drainage Material            196

6.1       Compounds Used in The Manufacture of Geotextiles (Values Are
          Percentages Based on Weight)                                           204
                                         XIX

-------
                                 Acknowledgments
      The authors gratefully acknowledge the following individuals and organizations who
provided many of the photographs included in this report: Stephen T. Butchko of Tensar
Environmental Systems, Inc.. Richard W. Carnker of James Gem Corporation, Steven R. Day of
Geo-Con, Inc., Anthony £. Eith of RUST Environmental and Infrastructure, Inc., VitoGalante of
Waste Management of l.'onh America, Inc., Gary Kolbasuk of National Seal  Co., David C.
Lauwers of Occidental Chemical Corp., David L. Snyder of Webtec, Edward Staff, Jr., of Staff
Industries,  Inc., Fred Strove of Gundle  Lining  Systems, Inc., Michael T. Taylor of JPS
Elastomerics Corp., and Dennis B. Wedding of Hocchst Celanese Corp.

      The authors wish  to express their sincere appreciadon to the following individuals, who
reviewed and critiqued an earlier version of this manuscript:  Craig H. Benson, Gordon P.
Boutwell, Mark CadwaJlader, Craig R. Calabria, James G. Collin, Jose D. Constantino, Steven R.
Day, Robert Denis, Richard H. Dickinson, Thomas N. Dobras, Lee Embrey, Jeffrey C. Evans,
Richard A. Goodrum, Dave Guram, Janice HaJl, Bill Hawkins, Georg  Heerten, William A.
Hoffman III, Ted Koemer, David C. Lauwers, Larry D. Lydick, Lance Mabry, Stephen F. Maher,
William C.  Neal, Leo Ovcrmann, Ian Peggs, Gregory N. Richardson, Charles  Rivette, Mark D.
Sieracke, Edward C. Staff, James Stcnborg, Frank Taylor, Stephen J. Trautwein, James F.  Urek,
Luke van't Hoog, Mark H.  Wayne, Felon R. Wilson, and John P. Workman.  The authors also
thank Joseph A. Dieitz who, through the Industrial Fabrics Association International, provided
assistance in obtaining reviews of an earlier draft of this document.

      The authors also gratefully acknowledge the many individuals, too numerous to name here,
who over the years have shared their  experiences and recommendations concerning quality
assurance and quality control with the authors.  The member organizations of ihe Geosynthetic
Research Institute are thanked for support of this effort.  Finally, the assistance of and input by
Mr. Robert E.  Landreth with the U.S. EPA is gratefully acknowledged.

-------
                                      Chapter 1

                   Manufacturing Quality Assurance  (MQA) and
          Construction Quality Assurance (CQA) Concepts and Overview
1.1    Introduction

       As a prelude to description of the detailed components of a waste containment facility,
some introductory comments are felt to be necessary. These comments are meant to clearly define
the role of the various parties associated with the manufacture, installation and inspection of all
components of a total liner and/or closure system for landfills, surface impoundments and waste
piles.

1.1.1  Scope

       Construction quality assurance (CQA) and construction quality control (CQC) are widely
recognized is critically important factors in overall quality management for waste containment
facilities. The best of designs and regulatory requirements will not necessarily translate to waste
containment facilities that are protective of human health and the environment unless  the waste
containment and closure facilities are properly constructed.   Additionally, for geosynthetic
materials, manufacturing quality assurance (MQA) and manufacturing quality control (MQC) of the
manufactured product is equally important. Geosynthetics refer to factory fabricated polymeric
materials like geomembranes, geotextiles, geonets, geogrids, geosynthetic clay liners, etc.

       The purpose of this document is to provide detailed guidance for proper MQA  and CQA
procedures for waste containment facilities. (The document also is applicable to MQC and CQC
programs on the pan of the manufacturer and contractor).  Although facility designs are different,
MQA and CQA procedures are the same. In this document, no distinction is made concerning the
type of waste to be contained (e.g., hazardous or nonhazardous waste) because the MQA and CQA
procedures needed to inspect quality lining systems, fluid collection and removal systems, and
final cover systems are the same  regardless of the waste type. This technical guidance document
has been written to apply to all types of waste disposal facilities, including new hazardous waste
landfills and impoundments, new municipal solid  waste landfills, nonhazardous waste liquid
impoundments, and final coven for new facilities and site remediation projects.

       This document is intended to aid those who are preparing MQA/CQA plans, reviewing
MQA/CQA plans, performing MQA/CQA observations and tests, and reviewing field MQC/CQC
and MQA/CQA procedur  •. Permitting agencies may use this doi urnent as a technical resource to
aid in the review of site-specific MQA/CQA plans and to help in identification of any deficiencies in
the MQA/CQA plan. Owner/operators and their MQA/CQA consultants may~consu!t this document
for guidance on the plan, the process, and the final certification report.  Field inspectors may use
this document and the references herein as a guide to field MQA/CQA procedure;. Geosynthetic
manufacturers may use the document to help in establishing appropriate MQC procedures and as a
technical resource to explain the reasoning behind MQA procedures. Construction personnel may
use this document to help in establishing appropriate CQC procedures and as a technical resource
to explain the reasoning behind CQA procedures.

       This technical  guidance  document is intended to update and expand EPA's Technical
Guidance Document,  "Construction Quality  Assurance for Hazardous Waste Land  Disposal

-------
                                                                                                V;
Facilities," (EPA, 1986).    The scope of this document includes all natural and geosynthetic
components that might normally be used in waste containment facilities, e.g., in liner systems,
fluid collection and removal systems, and cover systems.

       This document draws heavily upon information presented in three EPA Technical Guidance
Documents: "Design, Construction, and Evaluation of Clay Liners for Waste  Management
Facilities" (EPA,  1988a), "Lining of Waste Containment and Other Impoundment Facilities"
(1988b), and "Inspection Techniques for the Fabrication of Geomembrane Field Seams" (EPA,
1991a). In addition, general technical backup information concerning many of the principles
involved in construction of liner and cover systems for waste containment facilities is provided in
two  additional EPA documents:  "Requirements for Hazardous Waste  Landfill Design.
Construction, and Closure" (EPA. 1989) and "Design and Construction of RCRA/CERCLA Final
Covers" (EPA, 1991b).  Additionally, there are numerous books and technical papers in the open
literature which form a large data base from which information and reference will be drawn in the
appropriate sections.

1.1.2  Definitions.

       It is critical to define and understand the differences between MQC and MQA and between
CQC and CQA and to counterpoint where the different activities contrast and/or complement one
another. The following definitions are made,

       *  Manufacturing Quality Control (MQC): A planned system of inspections that is used to
         directly monitor and control the manufacture of a material which is factory originated.
         MQC is normally performed  by  the manufacturer of geosynthetic materials and is
         necessary to ensure minimum for maximum) specified values in  the manufactured
         product.  MQC refers to measures taken by the manufacturer to determine compliance
         with the requirements for materials and workmanship as stated in certification documents
         and contract plans.

       •  Manufacturing Quality Assurance (MQA): A planned system of activities that provides
         assurance that the materials were constructed as specified in the certification documents
         and contract plans. MQA includes manufacturing facility inspections, verifications,
         audits and evaluation of the raw materials and geosynthetic products to assess the quality
         of the manufactured materials.  MQA refers to measures taken by the MQA organization
         to determine if the manufacturer is in compliance with the product certification and
         contract plans  for a project.

       «  Construction Quality Control (CQC):  A planned system of inspections that is used to
         directly monitor  and control the quality of a construction  project (EPA,  1986).
         Construction quality control is normally performed by the geosynthetics Installer, or for
         natural soil materials by the earthwork contractor, and is necessary to"achievc quality in
         the  constructed or installed system.  Construction quality control (CQC)  refers to
         measures  taken by the  installer or contractor to determine compliance  with  the
         requirements for materials and workmanship as stated in the plans and specifications for
         the project.

       •  Construction Quality Assurance (CQA):  A planned system of activities thai provides the
         owner and permitting agency assurance that the facility was constructed as Specified in
         the  design {EPA, 1986). Construction  quality assurance  includes inspections,
         verifications,  audits, and evaluations of materials and  workmanship  necessary to
         determine and document the quality of the constructed facility. Construction quality


-------
         assurance (CQA) refers to measures taken by the CQA organization to assess if the
         installer or contractor is in compliance with the plans and specifications for a project
       MQA and CQA are performed independently from MQC and CQC. Although MQA/CQA
and MQC/CQC are separate activities, they have similar objectives and, in a smoothly running
con'Jtructioi project, the processes will complement one another.  Conversely, an effective
MQA/CQA program can lead to identification of deficiencies in the MQC/CQC process, but a
MQA/CQA program by itself (in complete absence of a MQC/CQC program) is unlikely to lead to
acceptable quality management Quality is best ensured with effective MQC/CQC and MQA/CQA
programs.  See Fig. 1.1  for the usual interaction of the various elements in a total inspection
program,

1.2    Responsibility and Authority

       Many  individual are involved directly or  indirectly in MQC/CQC and  MQA/CQA
activities. The individuals, their affiliation, and their responsibilities and authority are discussed
below.

       The principal organizations and individuals involved in designing, permitting, constructing,
and inspecting a waste containment facility are:

       »  Permitting Agency.  The permitting agency is often a state regulatory agency but may
         include local or regional agencies and/or the federal U. S. Environmental Protection
         Agency (EPA). Other federal agencies, such as the U.S. Army Corps of Engineers, the
         U.S. Bureau of Reclamation, the U.S. Bureau of Mines, etc., or their regional or state
         affiliates are  sometimes also involved.  It is the responsibility of the permitting agency to
         review the owner/operator's permit application, including the site-specific MQA/CQA
         plan, for compliance with the agency's regulations and to make a decision to issue or
         deny a permit based on this review. The permitting agency also has the responsibility to
         review all MQA/CQA documentation during or after construction of a facility, possibly
         including visits to the manufacturing facility and construction site  to  observe  the
         MQC/CQC and MQA/CQA practices, to confirm that the approved MQA/CQA plan was
         followed and that the facility was constructed as specified in the design.

       •  Owner!Operator. This is the organization that will own and operate the disposal unu
         The  owner/operator is responsible for the design, construction, and operation of the
         waste disposal  unit. This responsibility includes complying^with the requirements of the
         permitting agency,  the submission of MQA/CQA documentation, and assuring  the
         permitting agency that the facility was constructed as specified in the construction plans
         and specifications and as approved by the permitting agency. The owner/operator has
         the authority to select and dismiss organizations charged with design, construction,  and
         MQA/CQA. If the owner and operator of a facility are different organizations, the
         owner is ultimately responsible for these activities.  Often the owner/operator, or owner,
         will  be a municipality rather than a private corporation.  The interaction of a state office
         regulating another state or local organization should have absolutely no impact on
         procedures,  intensity of effort and ultimate decisions of the MQA/CQA or MQC/CQC
         process as described herein.

-------
                                         Natural Soil
                                        I Components
                                        I  Earthwork
                                        I Subcontractor
Figure I.I - Organizational Structure of MQA/CQA Inspection Activities

-------
Owner's Representative, The owner/operator usually has an official representative who
is responsible for coordinating  schedules, meetings, and field activities.   This
responsibility includes communications to dnsr members in the owner/operator's
organization, owner's representative, permitting agency, material suppliers, general
contractor, specialty subcontractors or installers, and MQA/CQA engineer.

Design Engineer,  The design engineer's primary responsibility is *o design a waste
containment facility that fulfills the operational requirements of the  owner/operator,
complies with accepted design practices for waste containment facilities, and meets or
exceeds the minimum requirements of the permitting agency.  The design engineer may
be  an  employee  of the owner/operator or a design  consultant hired by the
owner/operator. The design engineer may be requested to change some aspects of the
design if unexpected conditions are encountered during construction (e.g., a change in
site conditions, unanticipated  logistical  problems during construction, or lack of
availability of certain materials). Because design changes during construction are not
uncommon, the design engineer is often involved in the MQA/CQA process. The plans
and specifications referred to in this manual will generally be the product of the Design
Engineer. They are a major and essential part of the permit application process and the
subsequently constructed facility.

Manufacturer.  Many  components, including ail geosynthetics, of a waste containment
facility are manufactured materials. The manufacturer is responsible for the manufacture
of its materials and for quality control during manufacture, i.e., MQC. The minimum or
maximum (when appropriate) characteristics of acceptable materials should be specified
in the permit application. The manufacturer is responsible for certifying that its materials
conform to those specifications and any more stringent requirements or specifications
included in the contract of sale  to the owner/operator or its agent.  The quality control
steps taken by a manufacturer are critical to overall quality management in construction
of waste containment facilities. Such activities often take the form of process quality
control, computer-aided quality control and the like. All efforts at producing better
quality materials are highly encouraged. If requested, the manufacturer should provide
information to the owner/operator, permitting agency, design engineer, fabricator,
installer, or MQA engineer that describes the quality control (MQC) steps that are taken
during the manufacturing .of the product.  In  addition, the manufacturer should be
willing to allow the owner/operator, permitting agency, design  engineer, fabricator,
installer, and MQA engineer to observe the manufacturing process and quality control
procedures if they so desire.  Such visits should be able to be made on an announced or
unannounced basis. However, such visits might be coordinated with the manufacturer
to assure that the appropriate people are present to conduct The tour and that the proper
geosynthetic is scheduled for that date so as to obtain the most information from the
visit.  The manufacturer should have a designated individual who is in charge of the
MQC  program and to whom questions can be directed and/or through whom visits can
be  arranged. Random samples of materials should be able to be  taken for subsequent
analysis and/or archiving. However,  the manufacturer should retain the right to insist
that any proprietary information concerning the manufacturing of a product be held
confidential. Signed agreements of confidentiality are at the option of the manufacturer.
The owner/operator, permitting agency, design engineer, fabricator, installer, or MQA
engineer may request that they be allowed to observe the manufacture and quality control
of some or all of the raw materials and final product to be utilized on a particular job; the
manufacturer should be willing to accommodate such requests.  Note that these same
comments apply to marketing organizations which represent a manufactured product
made by others, as well as the manufacturing organization itself.
                                  5

-------
Fabricator.  Some materials are fabricated from manufactured components.  For
example, certain  geomembrancs  are  fabricated  by seaming  together  smaller,
manufactured geomembrsne sheets  at  the fabricator's  facility.  The minimum
characteristics of acceptable fabricated materials are specified in the permit application.
The  fabricator is  responsible for  certifying that  its materials  conform  to those
specifications and  any more stringent requirements  or specifications included in the
fabrication contract with the owner/operator ur its agent. The quality control steps taken
by a fabricator are critical to overall quality  in construction of waste containment
facilities. If requested, the fabricator should provide information to the owner/operator,
permitting agency, design engineer, installer, or MQA  engineer that describes the quality
control steps that are taken during the fabrication of the  product.  In addition, the
fabricator should be willing to allow the owner/operator, permitting agency, design
engineer, installer, or MQA engineer to observe the fabrication process and quality
control procedures if they so desire. Such visits may be made on an announced or
unannounced basis. However, such visits might be coordinated  with the fabricator to
assure that the appropriate people are present to conduct the tour and that the proper
geosymhetic is scheduled for that date so as to obtain the most  information  from the
visit  Random samples of materials should be able to be taken  for subsequent analysis
and/or archiving.  However, the fabricator should retain the  right to insist that any
proprietary information concerning  the fabrication of a product  be held confidential.
Signed  agreements  of confidentiality  are  at the  option  of the fabricator.   The
owner/operator, permitting agency, design engineer, or MQA engineer may request that
they be allowed  to observe the fabrication process and quality control of some or all
fabricated materials to be utilized on a particular job; the fabricator should be willing to
accommodate such a requests.

General Contractor, The general contractor has overall responsibility for construction of
a waste containment facility and for CQC during construction.  The general contractor
arranges for purchase of materials that meet specifications, enters into a contract with
one or more fabricators (if fabricated materials are needed) to  supply those materials,
contracts with an installer (if separate from the general contractor's organization), and
has overall control over the construction operations, including scheduling and CQC.
The general contractor has the primary responsibility for ensuring that a facility is
constructed in accord with the plans and specifications that have been developed by the
design engineer and approved by the permitting agency. The general contractor is also
responsible  for informing  the owner/operator and  the MQA/CQA engineer of the
scheduling and  occurrence of all  construction activities.  Occasionally, a waste
containment facility may be constructed without a general contractor. For example, an
owner/operator may arrange for all the necessary material, fabrication, and installation
contracts.  In such cases,  the owner/operator's representative  will serve the same
function as the general contractor.

Installation Contractor  Manufactured products (such as geosynthetics) are placed and
installed in the field by an installation contractor who is the  general contractor, a
subcontractor to the general contractor, or is a specialty contractor hired directly by the
Owner/dperator.  The installer's personnel may be employees of the owner/operator,
manufacturer, or fabricator, or they may work for an independent installation company
hired by the  general contractor or by the owner/operator directly. The installer is
responsible for handling, storage, placement, and installation of manufactured and/or
fabricated materials.  The installer should have a CQC plan to detail the  proper manner
that materials are handled, stored, placed, and installed.  The installer is also responsible
for informing the owner/operator and the MQA/CQA engineer of the scheduling and

-------
         occurrence of all gcosynthetic construction activities.

         Earthwork Contractor. The earthwork contractor is responsible for grading the site to
         elevations and grades shown on the plans and for constructing earthen components of
         the waste containment facility, e.g., compacted clay liners and granular drainage layers
         according to the specifications. The earthwork contractor may be hired by the general
         contractor  or if the owner/operator serves  as the  general  contractor, by  the
         owner/operator directly. In some cases, the general contractor's personnel may serv; as
         the earthwork contractor.  The earthwork contractor is responsible not only for grading
         the site to p-->per elevations but also for obtaining suitable earthen materials, transport
         and storage of those materials, preprocessing of materials (if necessary), placement and
         compaction of materials, and protection of materials during and (in some cases) after
         placement.  If a test pad is required, the earthwork contractor is usually responsible for
         construction of the test pad.  It is highly suggested that the same earthwork contractor
         that constructs the test fill also construct the waste containment facility compacted clay
         liner so that the experience gained from the test fill process will not be lost. Earthwoiic
         functions must be carried out in accord with plans and specifications approved by the
         permitting agency. The earthwork contractor should have a CQC plan (or agree to one
         written by others) and is responsible for CQC operations aimed at controlling materials
         and placement of those materials to conform with project specifications.  The earthwork
         contractor is also responsible for informing the owner/operator and the CQA engineer of
         the scheduling and occurrence of ail earthwork construction activities.

         CQC Personnel. Construction quality control personnel  are individuals who work for
         the general contractor, installation contractor, or earthwork contractor and whose job it is
         to ensure that construction is taking place in accord with the plans and  specifications
         approved by the permitting agency.  In some cases, CQC personnel, perhaps even a
         separate company, may also oe pan of the installation or construction crews. In other
         cases, supervisory personnel  provide CQC or, for  large projects, separate CQC
         personnel, perhaps even a separate company, may be utilized.  It is recommended that a
         certain portion of the  CQC staff should be certified* as per the implementation schedule
         of Table 1.1.  The examinations have been available as of October,  1992.

         MQA/CQA   Engineer.   The  MQA/CQA engineer  has  overall responsibility for
         manufacturing quality assurance and  construction quality assurance. The engineer is
         usually an individual experienced in a  variety of activities although particular specialists
         in soil placement, polymeric materials and geosynthetic placement will invariably be
         involved in a project. The MQA/CQA engineer is responsible  for reviewing the
         MQA/CQA  plan as well as general plans and specifications forjhe project so that the
         MQA/CQA plan can be implemented with no contradictions or unresolved discrepancies.
         Other responsibilities of the MQA/CQA engineer include education  of inspection
         personnel on MQA/CQA requirements and procedures and special  steps that are needed
         on a particular project, scheduling and coordinating of MQA/CQA inspection activities,
         ensuring that proper procedures are  followed,  ensuring that testing laboratories are
         conforming to MQA/CQA requirements and procedures, ensuring that sample custody
         procedures arc followed, confirming that test data are accurately reported and that test
         data are mai; _ined for later reporting, and preparation of periodic reports. The most
         important duty of the MQA/CQA engineer is overall responsibility for confirming that
         the facility was constructed in accord with plans and specifications approved by the
* A certification program is available from the National Insiiiulc for Certification of Engineering Technologies
(N1CET); 1420 King Slreeu Alexandria. Virginia 22314 (phone: 703-684-2835)

-------
 permitting agency.  In the event of nonconformance with the project specifications or
 CQA Plan, the MQA/CQA engineer should notify the owner/operator as to the details
 and, if appropriate, recommend work stoppage and possibly remedial actions.  The
 MQA/CQA engineer is normally hired by the owner/operator and functions separately of
 the contractors and owner/operator.  The MQA/CQA engineer must be a registered
 professional engineer who has shown competency and experience in similar projects and
 is considered qualified by the permitting agency.  It is recommended that the person's
 resume and  record on like facilities must be submitted in writing and accordingly
 accepted by the permitting agency before activities commence. The permitting agency
 may request additional information from the prospective MQA/CQA engineer and his/her
 associated organization including experience record, education, registry and ownership
 details.   The  permitting agency may accept or deny  the MQA/CQA  engineer's
 qualifications based on  such data and revelations.  If the permitting agency requests
 additional information or denies the MQA/CQA engineer's qualifications it should be
 done prior to construction, so that alternatives caa  be made which do not negatively
 impact on the progress of the work. The MQA/CQA engineer is usually required to be at
 the construction site during all major construction operations to oversee MQA/CQA
 personnel. The MQA/CQA engineer is usually the MQA/CQA certification engineer who
 certifies the completed project
 Table 1.1-  Recommended Impentation Program for Construction Quality Control
            (CQQ forGeosynthetics* (Beginning January 1,1993)
No. of
Field C«ws»*
At Each Site
1-4
25

End of
IS Months
(i.e., June 30, 1994|
1 -Level n
i-uvcin
2 - Level 1
End of
36 Monil"
O.e,, January 1, 1996)
1- Level III"**
1 - Level HI***
1 - Level I
•Certification for natural materials is under development as of tins writing
••Perfoiraing a Crilkal Operation; Typically 4 to 6 People/Crew
***Or PE wiiJi applicable experience
 MQA/CQA Personnel.  Manuta, luring quality assurance and construction quality
 assurance personnel are responsible for making observations and performing field tests
 to ensure that a facility is constructed in accord with the plans and specifications
 approved by the permitting agency. MQA/CQA personnel normally are employed by the
 same firm as the MQA/CQA engineer, or by a firm hired by the firm employing the
 MQA/CQA engineer.  Construction MQA/CQA personnel report to the MQA/CQA
 engineer, A relatively large proportion (if not the entire group) of the MQA/CQA staff
 should be certified.  Table  1.2  gives the currently recommended implementation
 schedule. As mentioned previously, certification examinations have been available as of
 October,  1992, from the National Institute for Certification of Engineering Technologies
 in Alexandria, Virginia.

-------
                       Testing  Laboratory.  Many MQC/CQC  and MQA/CQA tests  arc performed by
                       commercial laboratories. The testing I* .boratory should have its own internal QC plan to
                       ensure that laboratory procedures conform to the appropriate American Society for
                       Testing and Materials (ASTM)  aundards or other applicable testing standards.  The
                       testing laboratory is responsible for ensuring that tests are performed in accordance with
                       applicable methods and  standards, for  following  internal  QC procedures,  for
                       maintaining sample chain-of-custody records, and for reporting data.  The testing
                       !  boratory must be  willing to allow the owner/operator, permitting agency, design
                       engineer, installer, or MQA/CQA engineer to observe the sample preparation and testing
                       procedures, or record-keep ing  procedures, if they  so desire.  The owner/operator,
                       permitting agency, design engineer, or MQA/CQA engineer may request that they be
                       allowed to observe some or all tests on a particular job at any time, either announced or
                       unannounced.  The testing laboratory personnel must be willing to accommodate such a
                       request,  but the observer should not interfere with the testing or slow the testing
                       process.


                       Table 1.2- Recommended Implementation Program for Construction Quality Assurance
                                  (CQA) for Geosynthetics* (Beginning January 1, 1993)
                             No. of                    Emlof                            End of
                          Field O.ws»*                 ISMwnhs                         36Monihs
                          At Each Site             (i.e.. June 30.1994)                 .'i.e.. January 1.

                              1-2                    J-Level II                       1-Level II!*
3-4
25
1 - Level 11
1 - Level I
1- Level II
2 • Level I
I - Level II!***
1 - Level I
1 - Level III***
1 - Level 11
1 • Level I
                      •Certification for natural materials is under development as of this writing
                      "Performing a Critical Operation; Typically 4 to 6 People/Crew
                      ***Or PE with applicable experience
                       MQA/CQA Certifying Engineer. The MQA/CQA certifying engineer is responsible for
                       certifying tc the owner/operator and permitting agency that, isT his or her opinion, the
                       facility has  been constructed in accord with plans and specifications and MQA/CQA
                       document approved by the permitting agency.  The certification statement is normally
                       accompanied by  a final  MQA/CQA  report that  contains  all the appropriate
                       documentation, including daily observation reports, sampling locations, test results,
                       drawings of record or sketches, and other relevant data. The MQA/CQA certifying
                       engineer may be the MQA/CQA engineer or someone else in the MQA/CQA engineer's
                       organization who is a registered professional engineer with experience and competency
                       in certifying like installations.
L

-------
1.3     Personnel Qualifications

        The key individuals involved in MQA/CQA and their minimum recommended qualifications
are listed in Table 13.
                        Table 1.3 - Recommended Personnel Qualifications
Individual
Minimum Recommended Qualifications
IVsIgH Engineer

Owner's Representative


Manufacturer/Fabricator
MQC Officer
Geosyntheuc Installer's
Representative

CQCPtnomel
CQ A Personnel
MQA/CQA Engineer
MQA/CQA Certifying Engineer
Registered Professional Engineer

The specific individual designated by the owner wiih knowledge
of the project, its plans, specifications and QC/QA documents.

Experience in manufacturing, or fabricating, at least
1,000,000 m2 (10,000,000 ft2) of similar geosyntheuc
materials.

Manufacturer, or fabricator, trained personnel in charge of
quality control of the geosynihcoc materials to be used in the
specific waste containment facility.

tie individual specifically designated by a manufacturer or
fabricator, in charge of geosyniheiic material quality control

Experience installing at least 1.000,000 m2 (10,000,000 ft2)
of similar geosyniheuc materials.

Employed by ITS general contractor, installation contnctcr or
earthwork contractor involved in waste containment facilities;
certified to the extent shown in Table 1,1.

Employed by an organization that operates separately from the
contractor and the owner/operator, certified to the extent shown
in Table  1.2.

Employed by an organization that operate* separately from the
contractor and owner/operator; registered Professional Engineer
and approved by permitting agency.

Employed by an organization that operates separately from the
contractor and owner/operator, registered Professional Engineer
in the state in which the waste containment facility is
constructed and approved by the appropriate permitting agency.
                                                  10

-------
1.4   Written MOA/COA Plan

      Quality assurance begins with a quality assurance plan. This includes both MQA and
CQA.  These activities are never ad hoc processes that are developed while they are being
implemented.  A written MQA/CQA plan must precede any field construction activities.

      The MQA/CQA plan is the owner/operator's written  plan for MQA/CQA activities.  The
MQA/CQA plan should include a detailed description of all MQA/CQA activities that will be used
during materials manufacturing and construction to manage the installed quality of the facility.  The
MQA/CQA plan should be tailored to the specific facility to be constructed and be completely
integrated into the project plans and specifications.  Differences should be settled before any
construction work commences.

      Most state and federal regulatory agencies require that a MQA/CQA plan be submitted by
the owner/operator and be approved by that agency prior to construction.  The MQA/CQA plan is
usually pan of the permit application.

      A copy of the site-specific plans and specifications, MQA/CQA plan, and MQA/CQA
documentation reports should be retained at the facility by the owner/operator or the MQA/CQA
engineer. The plans, specifications, and MQA/CQA documents may be reviewed during a site
inspection by  the permitting agency and will be the chief means for the facility owner/operator to
demonstrate to the permitting agency that MQA/CQA objectives for a project are being met

      Written MQA/CQA plans vary greatly from project to project. No general outline or
suggested list of topics is applicable to all projects or all regulatory agencies. The elements covered
in this document  provides guidance on topics that should be addressed in the written MQA/CQA
plan.

1.5   Docurnentan'on

      A major  purpose of the MQA/CQA process is to provide documentation  for those
individuals who were unable to observe the entire construction process (e.g., representatives of the
permitting agency) so that those individuals can  make informed judgments about the quality of
construction for a project MQA/CQA procedures and results must be thoroughly documented.

1.5.J Daily Inspection Reports

      Routine daily reporting  and documentation procedures should be required.  Inspectors
should prepare daily written inspection reports that may ultimately be included in  the final
MQA/CQA document Copies of these reports should be available from the MQA/CQA engineer.
The daily reports should include information about work that wag%accomplished, tests and
observations that  were made, and descriptions of the adequacy of the work that was performed.

1.5.2 Daily Sumnr  y Reports

      A daily written summary report should be prepared by ihe MQA/CQA engineer.  This
report provides a chronological framework for identifying and recording all other reports and aids
in tracking what was done and by whom. As a minimum, the daily summary reports should
contain the following (modified from Spigolon and Kelly, 1984. and EPA, 1986):
                                          11

-------
      •  Date, project name, location, waste containment unit under construction, personnel
         involved in major activities and other relevant identification information;

      •  Description of weather conditions, including temperature, cloud cover, and precipitation;

      •  Summaries of any meetings held and actions recommended or taken;

      •  Specific work un;'s and locations of construction underway during that particular day;

      •  Equipment and personnel being utilized in each work task, including subcontractors;

      •  Identification of areas or units of work being inspected;

      •  Unique identifying sheet number of geomembranes for cross referencing and document
         control;

      •  Description of off-site materials received, including any quality control data provided by
         the supplier;

      •  Calibrations or recalibrations of test equipment, incluv,. _ actions taken as a result of
         recalibration;

      •  Decisions made regarding approval of units of material or of work, and/or corrective
         actions to be taken in instances of substandard or suspect quality;

      •  Unique identifying sheet numbers of inspection data sheets and/or problem reporting and
         corrective measures used to substantiate any MQA/CQA decisions described in the
         previous item;

      •  Signature of the MQA/CQA engineer.

1.5.3  Inspection and Testing Reports

      All observations, results of field tests, and results of laboratory tests performed on site or
off site should be recorded on a suitable data: heet.  Recorded observations may take the form of
notes, charts, sketches, photographs, or any combination of these. Where possible, a checklist
may be useful to ensure that pertinent factors are not overlooked.

      As a minimum, the inspection data sheets  should  include the following information
(modified from Spigolon and Kelly, 1984, and EPA, 1986):

      •  Description or title of the inspection activity,

      •  Location of the inspection activity or location from which the sample was obtained;

      •  Type of inspection activity and procedure used (reference to standard method when
         appropriate or specific method described in MQA/CQA plan);

      •  Unique identifying geomembrane sheet number for cross referencing and document
         control;
                                           12

-------
      *  Recorded observation or test data;
      *  Results of the inspection activity (e.g., pass/fail); comparison with specification
         requirements;
      *  Personnel involved in the inspection besides the individual preparing the data sheet;
      *  Signature of the MQA/CQA inspector and review signature by the MQA/CQA engineer.
1.5.4 Problem Identification an4 Corrective Measures Report^
      A problem is defined as material or workmanship thai does not meet the requirements of the
plans, specifications or MQA/CQA plan for a project cr any obvious defect in material or
workmanship, even if there is conformance with plans, specifications and the MQA/CQA plan. As
a minimum, problem identification and corrective measures reports should contain the following
information (modified from EPA, 1986):
      *  Location of the problem;
      *  Description of the problem (in sufficient  detail and  with supporting  sketches or
         photographic information where appropriate) to adequately describe the problem;
      *  Unique identifying geomembrane sheet number for cross referencing an 3 document
         control;
      *  Probable cause;
      *  How and when the problem was located  (reference to  inspection data sheet or daily
         summary report by inspector);
      •  Where relevant, estimation of how long the problem has existed;
      *  Any disagreement noted by the inspector between the inspector and contractor about
         whether or not a problem exists or the cause of the problem;
      *  Suggested corrective measure(s);
      *  Documentation of correction if corrective action was  taken and completed prior  to
         finalizatton of the problem and corrective measures report (reference to inspection data
         sheet, where applicable);
      •  Where applicable, suggested methods to prevent similar problems;
      *  Signature of the MQA/CQA inspector and review signature of MQA/CQA engineer.
1.3.3 Drawings of Record
      Drawings of record (also called "as-built" drawings) should be prepared to document the
actual lines and grades and conditions of each  component of the disposal unit.   For soil
components, the record drawings shall include survey data that show bottom and top elevations of
a particular component, the plan dimensions of the component, and locations of ail destructive test
samples. For geosyntnetic component , the record drawings often show the dimensions  of ail
                                          13

-------
geomembrane field panels, the location of each panel, identification of all seams and panels with
appropriate identification numbering or lettering, location of all patches and repairs, and location of
all destructive test samples. Separate drawings are often needed to show record cross sections and
special features such as sump areas.

1.5.6 Final pocjrggpMi0" and.Certifiqation

       At the completion of a project, or a component of a large project, the owner/operator should
submit a final report to the permitting agency. This report may include all of the daily inspection
reports, the  daily MQA/CQA enginver's summary reports, inspection data  sheets, problem
identification and corrective measures reports, and other documentation such as quality control
data provided by manufacturers or  fabricators, laboratory test results, photographs, as-built
drawings, internal MQA/CQA memoranda or reports with data interpretation  or analyses, and
design changes made by  the design engineer during construction.  The document should be
certified correct by the MQA/CQA certifying engineer.

       The final documentation should emphasize that areas of responsibility and tines of authority
were clearly defined, understood, and accepted by all panics involved in we project (assuming that
this was the case). Signatures of the owner/operator's representative, design engineer, MQA/CQA
engineer, general contractor's representative, specialty subcontractor's representative, and
MQA/CQA certifying engineer may be included as confirmation that each party understood and
accepted 'he areas of responsibility and lines of authority outlined in the MQA/CQA plan.

1.5.7
       The MQA/CQA documents which have been agreed upon should be maintained under a
documem r.ontrol procedure.  Any portion of the documcnt(s) which are modified must be
communicated to and agreed upon by all parties involved.  An indexing procedure should be
developed for convenient replacement of pages in the MQA/CQA plan, should modifications
become necessary, with revision status indicated on appropriate pages.

       A control scheme should be implemented to organize and index all MQA/CQA documents.
This scheme should be designed to allow easy access to ail MQA/CQA documents and should
enable a reviewer to identify and retrieve original inspection  reports or data sheets for any
completed work element.

1.5.8 Storage of Jtecorjjs,

       During construction, the MQA/CQA engineer should be responsible for all MQA/CQA
documents. This includes a copy of the design criteria, plans, specifications, MQA/CQA plan, and
originals of all data sheets and reports.  Duplicate records should be kept at another location to
avoid loss of this valuable information if the originals are destroyed

       Once  construction is complete,  the document originals  should  be stored by  the
owner/operator in a manner that will allow for easy access while still protecting them from damage.
An additional copy should be kept at  the facility if this is in a different location  from  the
owner/operator's main files.  A final  copy should  be kept by the permitting agency.   All
documentation should be maintained through the operating and post-closure monitoring periods of
the facility by the owner/operator and the permitting agency in an agreed upon format (paper hard
copy, microfiche, electronic medium, etc.).
                                          14

-------
1.6    Meetings

       Communication is extremely important to quality management Quality construction is
easiest to achieve when all parties involved understand clearly their responsibility and authority.
Meetings czn be very helpful to make sure that responsibility and authority of each organization is
clearly  understood.   During  construction,  meetings  can help  to  resolve problems or
misunderstandings and to find solutions to unanticipated problems that have developed.

1.6.1  Pre-Bid Meeting

       The first meeting is held to discuss the MQA/CQA plan and to resolve differences of
opinion before  the project is let for bidding.  The pre-bid meet.ng is held after the permitting
agency has issued a permit for a waste containment facility and betore a construction contract has
been awarded.  The  pre-bid meeting is held before construction bids are prepared so that the
companies bidding on the construction will better understand the level  of MQA/CQA to be
employed on the project. Also, if the  bidders identify problems with the MQA/CQA plan, this
affords the owner/operator an opportunity to rectify those problems early in the process.

1.62  Resolution Meeting

       The objectives of the resolution meeting are to establish lines of communication, review
construction  plans and specifications, emphasize the critical aspects of a project necessary to ensure
proper quality, begin planning and coordination of tasks, and anticipate any problems that might
cause  difficulties or delays  in  construction.   The  meeting should   be  attended by the
owner/operator's representative, design engineer, representatives of the general contractor and/or
major subcontractors, the MQA/CQA engineer, and the MQA/CQA certifying engineer.

       The resolution meeting normally involves the following activities:

       •  An individual is assigned to take minutes (usually a representative of the owner/operator
         or of the MQA/CQA engineer's organization);

       •  Individuals are introduced  to one another and their responsibilities  (or  potential
         responsibilities) are identified;

       •  Copies of the project plans and specifications are made available for discussion;

       •  The MQA/CQA plan is distributed;

       •  Copies of any special permit restrictions that are relevant to construction or MQA/CQA
         are distributed;
                                                                   ^»
       •  The plans and specifications are described, any unique design features are discussed (so
         the contractors will understand the rationale behind the general design), any potential
         construction problems are identified and discussed, and questions from any of the
         panics concerning the construction are discussed;

       •  The  MQA/CQA plan is reviewed and discussed, with the  MQA/CQA engineer and
         MQA/CQA certifying engineer identifying their expectations and identifying the most
         critical components;
                                           15

-------
      »  Procedures for MQC/CQC proposed by installers and contractors are reviewed and
         discussed;

      •  Corrective actions to resolve potential construction problems are discussed;

      *  Procedures for documentation and distribution of documents are discussed;

      *  Each organization's responsibility, authority, and lines of communication are discussed;

      *  Suggested modifications to the MQA/CQA plan that would improve quality management
         on the project are solicited; and

      »  Construction  variables (e.g., precipitation, wind, temperature) and schedule are
         discussed.

      It is very important that the procedures for inspection and testing be known to ail, that the
criteria for pass/fail decisions be clearly defined (including the resolution of test data outliers), that
all parties understand the key problems that the MQA/CQA personnel will be particularly careful to
identify,  that each individual's responsibilities and authority be understood, and that procedures
regarding resolution of problems be understood.  The resolution  meeting  may  be held in
conjunction with either the pit-bid meeting (rarely) or the pre-construction meeting (often).

1.6.3  Pni-construction Meeting

      The pre-construction meeting is held after a general construction contract his been awarded
and the major subcontractors and material suppliers are established. It is usually held concurrent
with the  initiation of construction.  The purpose: of this meeting is to review the details of the
MQA/CQA plan, to make sure that the responsibility and authority of each  individual is clearly
understood, to agree on procedures to resolve construction problems, and to establish a foundation
of cooperation in quality management.  The pre-construction meeting should be attended by the
owner/operator's representative,  design engineer, representatives of the general contractor and
major subcontractors,  the  MQA/CQA engineer, the MQA/CQA certifying  engineer, and a
representative from the permitting agency, if that agency expects to visit  the  site  during
construction or independently observe MQA/CQA procedures.

      The prc-construction meeting should include the following activities:

      •  Assign an individual (usually representative of MQA/CQA engineer) to take minutes;

      *  Introduce parties and identify their responsibility and authority;

      •  Distribute the MQA/CQA plan, identify any revisions made after the resolution meeting,
         and answer any qur :aons about the MQA/CQA plan, procedures, or documentation;

      *  Discuss responsibilities and lines of communication;

      *  Discuss reporting procedures, distribution of documents, schedule  for any  regular
         meetings, and resolution of construction problems;

      •  Review site requirements and logistics, including safety procedures;
                                           16

-------
      *  Review the design, discuss the most critical aspects of the construction, and discuss
         scheduling and sequencing issues;

      *  Discuss MQC procedures that the geosynthetics manufacturers) will employ;

      •  Discuss CQC procedures that the installer or contractor will employ,  for example,
         establish and agree on geomembrane repair procedures;

      *  Make a list of action items that require resolution and assign responsibilities for these
         items.
1.6.4  Proress
       Weekly progress meetings should be held.  Weekly meetings can be helpful in maintaining
lines of communication, resolving problems, identifying action items, and improving overall
quality management. When numerous critical work elements are being performed, the frequency
of these meetings can be increased to biweekly, or even daily.  Persons who should attend this
meeting are those involved in the specific issues being discussed.  At all times the MQA/CQA
engineer, or designated representative, should be present.

1.7    Sample Custody

       All samples shall be identified as described in the  MQA/CQA plan. Whenever a sample is
taken, a chain of custody record should be made for that sample.  If the sample is transferred to
another individual or laboratory, records shall be kept of the transfer so that chain of custody can
be traced. The purpose  of keeping a record of sample custody is to assist in tracing the cause of
anomalous test results or other testing problem, and to help prevent accidental loss of test samples.

       Soil  samples  are usually discarded after testing.  Destructive testing  samples  of
geosynthetic materials are often taken in triplicate, with one sample tested by CQC personnel, one
tested by CQA personnel, and the third retained in storage as prescribed in the CQA plan.

1.8    Weather

       Weather can play a critical role  in  the  construction of waste  containment facilities.
Installation of all geosynthetic materials (including geosynthetic clay liners) and natural clay liners
is particularly sensitive to weather conditions,  including temperature, wind,  humidity,  and
precipitation.  The contractor or installer is responsible for complying with ths contract plans and
specifications (along with the MQC/CQC plans for the various'rcmponents of the system).
Included in this information should be details which restrict the weather conditions  in which certain
activities can take place.  It is the tesponsibility of the contractor or installer to make sure that these
weather restrictions are observed during construction.

1.9    Work Stoppages

       Unexpected work stoppages can occur due to a variety of causes, including labor strikes,
contractual  disputes, weather,  QC/QA problems, etc.  The MQA/CQA engineer should be
particularly careful during such stoppages  to determine (1) whether in-place materials are covered
and protected from damage (e.g., lifting of a geomembrane by  wind or premature hydration of
geosynthetic clay liners); (2) whether partially covered materials are protected from damage (e.g.,
desiccation of a compacted clay liners); and (3) whether manufactured materials are properly
stored and properly or adequately protected (e.g., whether geotextiles are protected from ultraviolet
                                           17

-------
exposure).  The cessation of construction should not mean the cessation of MQA/CQA inspection
and documentation.

1.10  References

Spigolon, S J., and M.F. Kelly (1984), "Geotechnical Assurance of Construction of Disposal
   Facilities," U. S. Environmental Protection Agency, EPA 600/2-84-040, Cincinnati, Ohio.

U.S. Environmental Protection Agency (1986), 'Technical Guidance Document, Construction
   Quality Assurance for Hazardous Waste Land Disposal Facilities," EPA/53Q-SW-86-031,
   Cincinnati, Ohio, 88 p.

U.S. Environmental Protection Agency (1988a), "Design, Construction, and Evaluation of Clay
   Liners for Waste Management Facilities," EPA/530-SW-86-007F, Cincinnati, Ohio.

U.S. Environmental Protection Agency  (1988b). "Lining of Waste Containment and Other
   Impoundment Facilities," EPA/600/2-88/052, Cincinnati, Ohio.

U. S. Environmental Protection Agency (1989), "Requirements for Hazardous Waste Landfill
   Design, Construction, and Closure," EPA/625/4-89/022, Cincinnati, Ohio.

ILS. Environmental Protection Agency (1991a), "Inspection Techniques for the Fabrication of
   Geomembrane Field Seams," EPA/S3Q/SW-91/051, Cincinnati, Ohio.

U. S. Environmental Protection Agency (1991b), "Design and Construction of RCRA/CERCLA
   Final Covers," EPA/625/4-91/025, Cincinnati. Ohio.
                                         18

-------
                                      Chapter 2

                               Compacted  Soil  Liners
2.1
Introduction and Backwound
2.1.1  Types of Compacted Soil Liners

      Compacted soil liners have been used for many years as engineered hydraulic barriers for
waste containment facilities.  Some liner and cover systems contain a single compacted soil liner,
but others may contain two or more compacted soil liners. Compacted soil liners are frequently
used in conjunction  with gcomembnnes to form a composite liner, which usually consists of a
geomembrane placed directly on the suifcce of a compacted soil liner. Examples of soil liners used
in liner and cover systems are shown in Fig. 2.1.

      Compacted soil liners are composed of clayey materials that are placed and compacted in
layers called lifts. The materials used to construct soil liners include natural mineral materials
(natural soils), bentonite-soil blends, and other material

2.1,1.1  Natural Mineral Materials.

        The most common type of compacted soil liner is one that is constructed from naturally
occurring soils that contain a significant quantity of clay. Soils are usually classified as CL, CH,
or SC soils in the Unified Soil Classification System (USCS) and ASTM D-2487. Soil liner
materials are excavated from locations called borrow pits. These borrow areas are located either on
the site or offsitc. The soil in the  borrow pit may be used directly without processing or may be
processed to alter the water content, break down  large pieces of material, or remove oversized
panicles. Sources of natural soil  liner materials include lacustrine deposits, glacial tills, acolian
materials, deltaic deposits, residual soils, and other types of soil deposits. Weakly cemented or
highly weathered rocks, e.g., mudstoncs and shales, can also be  used for soil liner materials,
provided they are processed property.

2.1.1.2  Benjonjfe-Soil Blends

      If the soils found in the vicinity of a waste disposal facility are not sufficiently clayey to be
suitable for direct use as a soil liner material, a common practice is to blend natural soils available
on or near a site with bentonite.  The term bentonite is used in different ways by different people.
For purposes of this discussion, bentonite is any commercially processed material that is composed
primarily of the mineral smectite.  Bentonite may be supplied in granular or pulverized form. The
dominant adsorbed cation of commercial bentonite is usually sodium or calcium,  although the
sodium form is much more commonly used for soil sealing applications. Bentonite is mixed with
native soils either in thin layers or in a pugrrull.

2.1.1.3  Other

      Other materials have occasionally been used for compacted soil liners. For example,
bentonite may be blended with fly ash to form a liner  under certain circumstances.  Modified soil
minerals and commercial additives, e.g., polymers, have sometimes been used.
                                           19

-------
                       TYPICAL LINES SYSTEMS
                  Single Comoosita LJnort
           Concoct* _

           \JtXK
                  Doubla Composite Liner-
           CompMrn-
           Umr
           ComeoM*
           LMw
                        TYPICAL COVER SYSTEM
                                      Top Soil
                                      Soli cr Geotertl* Fih»r

                                      Drainage Layer

                                      Geomembran*
                                      Low Permeatality
                                      Compacted Soil
                                      Uner
                                      Wasta
Rgure 2.1 * Examples of Compacted Soil Liners in Liner and Cover Systems



                                 20

-------
2.1.2   Critical CQC and COA Issues

      The CQC and CQA processes for soil liners are intended to accomplish three objectives:

      1.     Ensure that soil liner materials are suitable.

      2.     Ensure that soil liner materials are properly placed a. ,. compacted.

      3.     Ensure that the completed liner is properly protected

Some of these issues, such as protection of the liner from desiccation after completion, simply
require application of common-sense procedures. Other issues, such preprocessing of materials,
are potentially much more complicated because, depending on the material, many construction
steps may be involved.  Furthermore, tests alone will not adequately address many of the critical
CQC and CQA issues - visual observations by qualified personnel, supplemented by intelligently
selected tests, provide the best approach to ensure quality in the constructed soil liner.

      As discussed in Chapter 1, the objective of CQA is to ensure that the final product meets
specifications.  A  detailed program of tests  and observations is necessary  to accomplish this
objective. The objective of CQC is to control the manufacturing or construction process to meet
project specifications. With geosynthetics, the distinction between CQC and CQA is obvious:  the
geosynthetics  installer performs CQC  r-'hile an independent organization conducts CQA.
However, CQC and CQA activities for  soils are more  closely linked  than in geosynthetics
installation. For example, on many earthwork projects the CQA inspector will typically determine
the water content of the soil and report the value to the contractor, in effect, the CQA inspector is
also providing CQC input to the contractor.  On some projects, the contractor is required to
perform extensive tests as part of the CQC process, and the CQA inspector performs tests t~ check
or confirm the results of CQC tests.

      The lack of clearly separate r  s for CQC  and CQA inspectors in the earthwork industry is
a result of historic practices and procw TS. This chapter is focused on CQA procedures for soil
liners, but the reader should understand that CQA and CQC practices are often closely linked in
earthwork. In any event, the QA plan should clearly establish QA procedures and should consider
whether there will be QC tests and observations to  complement the QA process.

2.1.3 yner ^ecjuirements

      The construction of soil liners is a challenging task that requires many careful steps.  A
blunder concerning any one detail of construction can have disastrous impacts upon the hydraulic
conductivity of a soil liner.  For example, if a liner is allowed to desiccate, cracks might develop
that could increase the hydraulic conductivity of the liner to  above the specified  requirement

      As stated in Section 2.1.2, the CQC and CQA processes for soil liners essentially consist
of using suitable materials, placing and compacting the materials properly, and protecting the
completed liner. The steps required to fulfill these  requirements may be summarized as follows:

       1,     The subgradc on which the soil liner will be placed should be properly prepared.

      2.     The materials employed in constructing the  soil liner should be suitable and should
             conform to the plans and specifications for the project.
                                           21

-------
       3.     The soil liner material should be preproccssed, if necessary, to adjust the water
             content, to remove oversized particles, to break down clods of soil, or to add
             amendments such as bentonitc.

       4.     *P>: soil should be placed in lifts of appropriate thickness and then be properly
             remolded and compacted.

       5.     The completed soil liner should be protected from damage caused by desiccation or
             freezing temperatures.

       6.     The final surface of the soil liner should be properly prepared to support the next
             layer that will be placed on top of the soil liner.

       The six steps mentioned above are described in more detail in the succeeding subsections to
provide the reader with a general introduction to the nature of CQC and CQA for soil liners.
Detailed requirements are discussed later.

2.1.3.1 Subgrade Preparation

       The subgrade on which a soil liner is placed should be properly prepared, i.e., provide
adequate support for compaction and be free from mass movements. The compacted  soil liner may
be placed on a natural  or geosynthetic material, depending on the particular design and  the
individual component in the liner or cover system.  If the soil liner is the lowest component of the
liner system, native soil or rock forms the subgrade. In  such cases the subgrade  should be
compacted to eliminate soft spots.  Water should be added or removed as necessary to produce a
suitably firm subgrade per specification requirements. In other instances the soil liner may be
placed on top of geosynthetic components of the liner system, e.g., a geotextilc.  In such cases, the
main concern is the smoothness of the geosynthetic on which soil is placer1, and conformity of the
geosynthetic to the underlying material (c.g., no bridgin" over ruts left by vehicle traffic).

       Sometimes it is necessary to "tie in" a new  section of soil liner to an old one, e.g., when a
landfill is being expanded laterally. It is recommended that a  lateral excavation be made about 3 to
6 m (10 to 20 ft) into the existing soil liner, and that the existing liner be stair-stepped as shown in
Rg. 2.2 to tie the new liner into the old one.  The surface  of each of the steps in the old liner
should be scarified to m  
-------
L
              2.1.3.2 Material Selection

                     Soil liner materials are selected so that a low hydraulic conductivity will be produced after
              the soil is remolded and compacted. Although the performance specification is usually hydraulic
              conductivity, CQA considerations dictate that restrictions be placed on certain properties of the soil
              used to build a liner.  For example, limitations may be placed on the liquid limit, plastic limit,
              plasticity index, percent fines, and percent gravel allowed in the soil liner material

                     The process of selecting construction materials and verifying the suitability of the materials
              varies from project to project In general, the process is as follows:

                     1.      A potential borrow source is located and explored to determine the vertical and
                            lateral extent of the source and to obtain representative samples, which are tested for
                            properties such as liquid limit, plastic limit, percent fines, etc.

                     2.      Once construction begins, additional CQC and CQA observations and tests may be
                            performed in the borrow pit to confirm the suitability of materials bring removed.

                     3.      After a lift of soil has been placed, additional CQA tests should be performed for
                            final verification of the suitability of the soil liner materials.

              On some projects, the process may be somewhat different. For example, a materials company may
              offer to sell soil liner materials from a commercial pit, in which  case thr first step listed above
              (location of borrow source) is not relevant.

                     A variety of tests is performed at various stages of the construction process to ensure that
              the soil liner material conforms with specifications. However, tests alone will not necessarily
              ensure an adequate material -- observations by  qualified CQA inspectors are essential to confirm
              that deleterious materials (such as stones or large pieces of organic or other deleterious matter) are
              not present in the soil liner material.

              2.1.3.3 Preprocessing

                     Some soil liner materials must be processed prior to use. The principal preprocessing steps
              that may be required include the following:

                     1.      Drying of soil that is too wet.

                     2.      Wetting of soil that is too dry.
                                                                                -^
                     3.      Removal of oversized panicles.

                     4.      Pulverization of clods of soil.

                     5,      Homogenization of nonuniform soil.

                     6.      Addition of bentonite.

              Tests are performed by CQA personnel to confirm proper preprocessing, but visual observations
              by CQC and CQA personnel a.e needed to confirm that proper procedures have been followed and
              that the soil liner material has been properly prcprocessed.
23

-------
2.1.3.4  Placement. Remolding, and Compactfpq

      Soil liners are placed and compacted in lifts. The soil liner material must first be placed in a
loose lift of appropriate thickness. If a loose lift is too thick, adequate compactive energy may not
be delivered to the bottom of a lift.

      The type and weight of compaction equipment can have an important influence upon the
hydraulic conductivity of the constructed liner.  The CQC/CQA program should be designed to
ensure that the soil liner material will be properly placed, remolded, and compacted as described in
the plans and specifications for the project.

2.1.3.5  Protection

      The completed soil liner must be protected from damage caused by desiccation or freezing
temperatures.  Each completed  lift of the  soil liner, as well as  the completed liner, must be
protected.

2.1.3.6  Final Surface Preparation

      The surface of the liner must be properly compacted and smoothed to serve as a foundation
for an overlying geomembrane liner or other component of a liner or cover system. Verification of
final surface preparation is an important pan of the CQA process.

2.1.4  Compaction Requirements

      One of the most important aspects of constructing soil liners that have low hydraulic
conductivity is the proper remolding and compaction of the soil. Background information on soil
compaction is ntsented in this subsection.

2.1.4.1  Compaction Curve

      A compaction curve is developed by preparing several samples of soil at different water
contents and then sequentially compacting ear '.j of the samples into a mold of known volume with a
specified compaction procedure.  The total unit weight (y), which is also called the wet density, of
each specimen is determined by weighing the compacted specimen and dividing the total weight by
the total volume.  The water content (w) of each compacted specimen is determined by oven drying
the specimen. The dry unit weight (Yd),  which is sometimes called the dry density, is calculated as
follows:
                                 Yd -  Y/U
(2.1)
The (w, Yd) points are plotted and a smooth curve is drawn between the points to define the
compaction curve (Fig. 2.3).  Judgment rather than an analytic algorithm is usually employed to
draw the compaction curve through the measured points.
       The maximum dry unit weight (Yd^nax) occurs at a water content that is called the optimum
water content, w^ (Fig. 2.3). The main reason for developing a compaction curve is to determine
the optimum water content  and maximum dry unit weight  for a given soil and compaction
procedure.
                                          24

-------
                           WEIGHT-VOLUME TERMINOLOGY
                        Weights      Volumes
                                Air
                               oooe
                                Soikte,
                                                         w.
                                                    Td "   V
f
$
            o
                   d.nax
                                 COMPACTION CURVE
                           Maximum Dry
                           Unit Weight
                                                   Zero Air Voids Curve
                                                 Optimum
                                                 Water
                                                 Content
                                         w
                                           Of*
                                     Molding Water Content (w)
                            Rgure 2.3 - Compaction Curve


      The zero cur voids curve (Fig. 7.3), also known as the 700% saturation curve, is a curve
that relates diy unit weight to water content for a saturated soil that contains no air.  The equation
for the zero air voids curve is:
                                         25

-------
                                    - Wl*
(2.2)
where Gs is the specific gravity of solids (typically 2.6 to 2.8) and YW is the unit weight of water.
If the soil's specific gravity of solids changes, the zero air voids curve will also change.
Theoretically, no points on a plot of dry unit weight versus water content should lie above the zero
air voids curve, but in practice some points usually lie slightly above the zero air voids curve as a
result of soil variability and inherent limitations in the accuracy of water content and unit weight
measurements (Schmertmajin, 1989).

      Benson and Boutwell (1992) summarize the maximum dry unit weights and optimum water
content measured on soil liner materials from 26 soil liner projects and found that the degree of
saturation at the point of (w^, y donax) ranged from 71% to 98%,  based on an assumed Gs value
of 2.75.  The average degree of saturation at the optimum point was 85%.

2.1.4.2  Compaction Tests

      Several methods of laboratory compaction are commonly employed.  The two procedures
that are  most commonly used are standard and modified compaction.  Both techniques usually
involve compacting the soil into a mold having a volume of 0.00094 m3 (1/30 ft3). The number of
lifts, weight of hammer, and height of fall are listed in Table 2.1. The compaction tests are
sometimes  called Proctor tests after Proctor, who developed th; tests and wrote  about the
procedures in several 1933 issues of Engineering News Recffl-j. Thus, the compaction curves are
sometimes called Proctor curves, and the maximum dry unit weight may be termed the Proctor
density.
                           Table 2.1 • Compaction Test Details
Compaction
PtoCCQfffC
Standni
Modified
Number
of Lifts
3
5
Weight of
Hammer
24.JN
(SJ Ibs)
44.5N
(10 Ibs)
Height of
Fall
305 mm
(12 in.)
457 mm
(18 in.)
Compacuvc
Encrw
594 kN-m/m3
'.12J7S fl-Ib/ft3)
2,693 kN-m/m3
(56,250 fl-lh/ft3)
       Proctor's original test, now frequently called the standard Proctor compaction test, was
developed to control compaction of soil bases for highways and airfields. The maximum dry unit
weights attained from the standard Proctor compaction test were approximately equal to unit
weights observed in the field on well-built fills using compaction equipment available in the 1920s
and 1930s. During World War II, much heavier compaction equipment was developed and the
unit weights attained from field compaction sometimes exceeded the laboratory values.  Proctor's
original procedi're was modified by increasing compaetive energy. By today's standards:

-------
                  »  Standard  Compaction  (ASTM D-698) produces  maximum dry  unit  weights
                     approximately equal to field dry unit weights for soils that are well compacted using
                     modest-sized compaction equipment

                  *  Modified  Compaction  (ASTM D-1557) produces maximum  dry  unit  weights
                     approximately equal to field dry unit weights for soils that are well compacted using the
                     heaviest compaction equipment available.


             2.1.4.3 Percent Compaction

                    The compaction test is used to help CQA  personnel  to determine: 1) whether the soil is at
             the proper water content for compaction, and 2) whether the  soil has received adequate compactive
             effort  Field CQA personnel will typically measure the water content of the field-eomoacted soil
             (w) and compare that value with the optimum water content (WQBI) from a laboratory compaction
             test The construction specifications may limit the value of w relative to w0m  e.g., specifications,
             may require  w to be between 0 and 44 percentage points of w0pt-  Field CQC personnel should
             measure the water content of the soil prior to remolding and compaction to ensure that the material
             is at the proper water content before the soil is compacted. However, experienced earthwork
             personnel can often tell if the soil is at the proper water content from the look and feel of the soil.
             Field CQA personnel should measure the water content and  unit weight after compaction to verify
             that the water content and dry unit weight meet specifications. Field CQA personnel often compute
             the percent compaction, P, which is  defined as follows;

                                                                                              (2.3)

             where  Yd 's the dry  unit weight of the field-compacted soil. Construction specifications often
             stipulate a minimum  acceptable value of P.

                    In summary, the purpose of the laboratory compaction test as applied to CQC and CQA is
             to provide water content (WOM) and dry unit weight (Yd,max) reference points. The actual water
             content of the field-compacted soil liner may be compared to the optimum value determined from a
             specified laboratory compaction  test  If the water content is not  in the proper  range,  the
             engineering properties of the soil are not likely to be in the range desired. For example, if the soil
             is too wet, the shear strength of the soil may be too low.  Similarly, the dry unit weight of the
             field-compacted soil may be compared to the  maximum dry unit weight determined from a
             specified laboratory compaction test If the percent compaction is too low, the soil has probably
             not been adequately compacted in the field. Compaction criteria may also be established in ways
             that do not involve percent compaction, as discussed later, but one way or another, the laboratory
             compaction test provides a reference point

             2.1.4.4 Estimating Optimum Water Content and Maximum Dry Unit Weight

                    Many CQA plans require that the water content and dry unit weight of the field-compacted
             soil be compared to  values determined from laboratory compaction tests. Compaction tests are a
             routine part of nearly all CQA programs. However, from a practical standpoint, performing
             compaction tests introduces two problems:

                    1.    A compaction test often takes 2 to 4 days to complete -- field personnel cannot wait
                          for the completion of a laboratory compaction test to make "pass-fail" decisions.
                                                       27
i
I

-------
      2.     The soil will inevitably be somewhat variable -- the optimum water content and
             maximum dry unit weight will vary. The values of w0pt and Yd,nm appropriate for
             one location may not be appropriate for another location. This has been termed a
             "mismatch" problem (Noorany, 1990).

      Because dozens  {sometimes hundreds) of field  water content and density tests are
performed, it is impractical to perform a laboratory compaction test each and every time a field
measurement of water content and density is obtained.  Alternatively, simpler techniques for
estimating the maximum dry unit  weight are almost always employed for rapid field CQA
assessments. These ^chniques are subjective assessment, one-point compaction test, and three-
point compaction lest.

2.1.4.4.1  Subjective Assessment

      Relatively  homogeneous fill mat?rials produce similar results when repeated compaction
tests are performed on the soil.  A common approach is to estimate optimum water content and
maximum dry unit weight based on the results of previous compaction tests. The results of at least
2 to 3 laboratory compaction tests should be available from tests on borrow soils prior to actual
compaction of any soil liner material for a project. With subjective assessment. CC A personnel
estimate the optimum water content and maximum dry unit weight based upon the results of the
previously-completed compaction tests and their evaluation of the soil at a particular location in the
field.  Slight variations in the composition of fill materials will cause only slight variations in wopt
and Ydjnax- As an  approximate guide, a relatively homogeneous borrow soil would be  considered
a material in which w0pi does not vary by more than ± 3 percentage points and Yd .max does not
vary by more than ± 0,8 kN/ft-* (5 pcf)-  The optimum water content and maximum dry unit
weight should not  be estimated in this manner if the soil is heterogeneous - too much guess work
and opportunity for error would exist.

2.1.4.4.2  Qne-Point Compaction Test

      The results of several complete compaction tests should always be available for a particular
borrow source prior to construction, and the data base should expand as a project progresses and
additional compaction tests are performed.  The idea behind a one-point compaction  test is shown
in Fig. 2.4. A sample of soil is taken from the field and dried to a water content that appears to be
just dry of optimum.  An experienced field technician can u^ally tell without much difficulty when
the  water content is just dry of optimum. The sample of soil is compacted into a mold of known
volume according  to the compaction procedure relevant to a particular project, e.g., ASTM D-698
or D-I557.  The weight of the compacted specimen is measured and the  total unit weight is
computed. Ths sample is dried using one of the rapid methods of measurement discussed later to
determine water content. Dry unit weight is computed from Eq. 2,2. The water content-dry unit
weight poini from the one-pt int compaction test is plotted  as shown in Fig.  2.4 and used in
conjunction with available  compaction curves to estimate w0pt and Yd,max- O*z assumes that the
shape of  the compaction is similar to the  previously-developed compaction curves  and passes
through the one point that has been determined.

      The dashed curve in Fig. 2,4 is the estimated compaction curve. The one-point compaction
test is commonly used for variable soils.  In extreme cases, a one-point compaction test may be
required for nearly all  field water content and density measurements for purposes of computing
percent compaction.  However, if the material is so variable to require a one-point compaction test
for nearly all field density measurements, the material is probably »oo variable 10 be suitable for use
in a soil liner. The best use of the one-point compaction test is to assist with estimation of the
optimum  water content and maximum dry unit weight for questionable materials and to fill in data


                                          28

-------
gaps when results of complete compaction tests are not availaole quickly enough.
              t
                                    Previously-Developed
                                    Compaction Curve
                                                    Result of One-Point
                                                    Compaction Test
                                                           Assumed Compaction
                                                           Curv*
                                                                 Previously-Developed
                                                                 Compaction Curv*
                                                       Estimated wopt
                                    Water Content
                          figure 2.4 - One-Point Compaction Test
2.1.4.4.3 Tlree-poin? Compactign.Tgt fASTM D-5Q80V

       A more reliable technique than the one-point compact ion test for estimating the optimum
water content and maximum dry unit weight is to use a minimum of three compaction points to
define a curve rather than relying on a single compaction point. A representative sample of soil is
obtained from the field at the same location where the in-place water content and dry unit weight
have been measured. The first sample of soil is compacted at the field water content.  A second
sample is prepared at a water content two percentage points wetter than the first sample and is
compacted.  However, for extremely wet soils that are more than 2%  wet of optimum (which is
often the case for soil liner materials), the second sample should be dried 2% below natural water
content Depending on the outcome of this compaction test, a third sample is prepared at a water
content either two percentage points dr> of the first sample or two percentage poims wet of the
second sample (or, for wet soil liners, 2 percentage points dry of the second sample). A parabola

                                           29

-------
is fitted to die three compaction data points and the optimum water content and maximum dry unit
weight are determined from the equation of the best-fit parabola.  This technique is significantly
more time consuming than the one-point compaction test but offers 1) a standard ASTM procedure
and 2) greater reliability and repeatability in estimated w^ and Ydjnax-

2.1.4.5 Recommetided Procedure jar Developing Water Content-Density Specification

       One of the most important aspects of CQC and CQA for soil liners is documentation of the
water content and dry unit weight of the soil immediately after compaction.  Historically, the
method used to specify water i jntent and dry unit weight  has been based upon experience with
structural fill Design engineers often require that soil liners be compacted within a specified range
of water content and to a minimum dry unit weight  The "Acceptable Zone" shown in Fig. 2.5
represents the  zone of acceptable water content/dry unit weight combinations that is often
prescribed. The shape of the Acceptable Zone shown in Fig.  2.5 evolved empirically from
construction practices applied to roadway bases, structural fills, embankments, and earthen dams.
The specification is based primarily upon the need to achieve a minimum dry unit weight for
adequate strength and limited compressibility. As discussed by Mundell and Bailey (1985),
Boutweil and Hedges (1989), and Daniel and Benson (1990), this method of specifying water
content and dry unit weight is not necessarily the best method for compacted soil liners.
         Sf
         I
                f»T
                                                Zero Air Vmdt Curv*
                                                   SpvcifM
                                                    Rung*
                                               opt
                                      Molding Water Content (w)
    Figure 2J5 - Form of Water Content-Dry Unit Weight Specification Often Used in the Past
                                          30

-------
      The recommended approach is intended to ensue that the soil liner will be compacted to a
water content and dry unit weight that will lead to low hydraulic conductivity and adequate
engineering performance with respect to other considerations, e.g., shear strength.  Rational
specification of water content/dry unit weight criteria should be based upon test data developed for
each particular soil. Field test data would be better than laboratory data, but the cost of determining
compaction criteria in the field through a series of test sections would almost always be prohibitive.
Because the compactive effort will vary in the field, a logical approach is to select several
compact!\e efforts in the laboratory that  span the range of compactive effort that might be
anticipated in the field.  If this is done, the water content/dry unit weight criterion that evolves
would be expected to apply to any reasonable compactive effort

      For most earthwork projects, modified Proctor effort represents a reasonable upper limit on
the compactive effort likely to be delivered to the soil in the field  Standard compaction effort
(ASTM D-698) likely represents a medium compactive effort It is conceivable that soil in some
locations will be compacted with an effort less than that of standard Proctor compaction.  A
reasonable lower limit of compactive energy is the "reduced compaction" procedure in which
standard compaction procedures (ASTM D-698) are followed except that only IS drops of the
hammer per lift are used instead of the usual 25 drops. The reduced compaction procedure is the
same as tlie 15 blow compaction test described by the U.S. Army Corps of Engineers (1970). The
reduced compactive effort is expected to correspond to a reasonable minimum level of compactive
energy for a typical soil liner or cover.  Other compaction methods, e.g., kneading compaction,
could be used. The key is to span the range  of compactive effort expected in the field  with
laboratory compaction procedures.

      One satisfactory approach is as follows:

      1.     Prepare and compact soil in tht  .boratory with  modified, standard, and reduced
             compaction procedures to develop compaction curves as shown in Fig. 2.6a. Make
             sure that the soil preparation procedures are appropriate; factors  such as clod size
             reduction may influence the results (Benson and Daniel, 1990). Other compaction
             procedures can be used if they better simulate field compaction and span the range
             of compactive effort expected in the field.  Also,  as  few as  two compaction
             procedures can be used if field construction procedures make either the lowest or
             highest compactive energy irrelevant

      2.     The compacted specimens should be permeated, e.g., per ASTM D-5084.  Care
             should be taken to ensure that permeation procedures are correct, with important
             details such as degree of saturation and effective confining stress carefully selected.
             The measured hydraulic conductivity should be  plotted as a function of molding
             watrr content as shown in Rg. 2.6b.

      3.     As shown in Fig. 2.6c, the dry unit weight/water content points should be replotted
             with different symbols used  to represent compacted specimens that had hydraulic
             conductivities greater than the  maximum acceptable value and specimens  with
             hydraulic conductivities less than or equal to the maximum acceptable value. An
             "Acceptable Zone" should be drawn to encompass the data points representing test
             results meeting or exceeding the design criteria.  Some  judgment is usually
             necessary in constructing the Acceptable  Zone from the daw points. Statistical
             criteria (e.g., Boutweil and Hedges, 1989) may be introduced at this stage.
                                           31

-------
             The Acceptable Zone should be modified (Fig. 2,6d) based on other considerations
             such as shear strength. Additional tests are usually necessary in order to define the
             acceptable range of water content and dry unit weight that satisfies both hydraulic
             conductivity and shear strength  criteria.  Figure 2.7 illustrates how one might
             overlap Acceptable Zones defined from hydraulic conductivity and shear strength
             considerations to define a single Acceptable Zone. The same procedure can be
             applied to take into consideration other  factors such as shrink/swell  potential
             relevant to any particular project
   o
a-
i
I
a
                                                                             Murtmum
                                                                            Vtfl»
                                                                            a,
                                                      (B)
                ttoMing Wntr Com*nt
    \
    $
f

i
e>
o
                  Melding WMM Camwn
             Moldinfl Wi»f CanMU
Figure 2.6 -  Recommended Procedure to Determine Acceptable Zone of Water Content/Dry Unit
            Weight Values Based Upon Hydraulic Conductivity Considerations (after Daniel and
            Benson, 1990).
                                           32

-------
                I
Acceptable Zone
Based on Shear
Sfrength Criterion
                                              Overall Acceptable Zone
                                              Based on AD Criteria
                                                              Acceptable Zone
                                                              Based on Hydraulic
                                                              Conductivity Criterion
                                      Mowing WM»r Content
Figure 2.7 - Acceptable Zone of Water Content/Dry Unit Weights Determined by Superposing
            Hydraulic Conductivity and Shear Strength Data (after Daniel and Benson, 1990).
       The same general procedure just outlined may also be used for soil-bentonite mixtures.
However, to keep the scope of testing reasonable, the required amount of bentonite should be
determined before the main part of the testing program is initiated. The recommended procedure
for soil-bentonite mixes may be summarized as follows:

       1.     The type, grade, and gradation of bentonite that will be used should be determined.
              This process usually involves estimating costs from several potential suppliers. A
              sufficient quantity of the bentonite likely to be used for the project should be
              obtained and tested to characterize the bentonite (characterization tests are discussed
              later).

       2.     A representative sample of the soil to which the bentonite will be added should be
              obtained
                                           33

-------
      3.     Batches of soil-bentonite mixtures should be prepared by blending in bentonite at
             several percentages, e.g., 2%, 4%, 6%, 8%, and 10% bentonite. Bentonite content
             is defined as the weight or mass of bentonite divided by the weight or mass of soil
             mixed with bentonite. For instance, if 5 kg of bentonite are mixed with 100 kg of
             soil, the bentonite content is 5%. Some people use the gross weight of bentonite
             rather than oven dry weight Since --ir-dry bentonite usually contains 10% to 15%
             hygroscopic water by weight, the use of oven-dry, air-dry, or damp  weight can
             make a difference in the percentage. Similarly, the weight of soil may be defined as
             either moist or dry (air- or oven-dry) weight. The contractor would rather work
             with total (moist) weights since the materials used in forming a soil-bentonite blend
             do contain some water. However, the engineering characteristics are controlled by
             the  relative amounts of dry materials.   A dry-weight  basis is generally
             recommended for definition of bentonite content, but CQC and CQA personnel
             must recognize that the project specificat'cni may or may not be on a dry-weight
             basis.

      4.     Develop compaction curves for each soil-bentonite mixture prepared from Step 3
             using the method of compaction appropriate to the project, e.g., ASTM D-698 or
             ASTMD-1557.

      5.     Compact samples at 2% wet of optimum for each percentage of bentonite using the
             same compaction procedure employed in Step 4.

      6.     Permeate the soils prepared from Step 5 using ASTM D-5084 or  some other
             appropriate test method.  Graph hydraulic conductivity versus percentage of
             bentonite.

      7.     Decide how much bentonite to use  based on the minimum required amount
             determined from Step 6. The minimum amount of  bentonite used  in the field
             should always be greater than the minimum amount suggested by laboratory tests
             because mixing in the field is usually  not as thorough as in the  laboratory.
             Typically, the amount of bentonite used in the field is one to four percentage points
             greater than the minimum percent bentonite indicated by laboratory tests.

      8.     A master batch of material should  be  prepared by mixing bentonite  with a
             representative sample of soil at the average bentonite content expected in the field.
             The procedures described earlier for determining the Acceptable Zone of water
             content and dry unit weight are then applied to the master batch.

2.1.5 Testj»adS

      Test pads are sometimes constructed and tested prior to construction of the full-scale
compacted soil liner. The test pad simulates conditions at the time of construction of the soil liner.
If conditions change, e.g., as a result  of emplacement of waste materials over the liner, the
properties of the liner will change in ways that are not normally simulated in a test pad.  The
objectives of a test pad should be as follows:

      1.     To verify that the materials and methods of construction will produce a compacted
             soil liner that meets the hydraulic conductivity objectives defined for a project,
             hydraulic conductivity should be measured with techniques that  will characterize the
             large-scale hydraulic conductivity and identify any construction defects that cannot
             be observed with small-scale laboratory hydraulic conductivity tests.

                                          34

-------
       2.     To verify that the proposed CQC and CQA procedures will result in a high-quality
             soil liner that will meet performance objectives.

       3.     To provide a basis of comparison for full-scale CQA: if the test pad meets the
             performance objectives  for  the  liner (as verified by appropriate hydraulic
             conductivity tests) and the full-scale liner is constructed to standards that equal or
             exceed those used in building the test pad, then assurance is provided that the full-
             scale liner will also meet performance objectives.

       4.     If appropriate,  a test pad provides an opportunity for the facility owner to
             demonstrate that unconventional materials or construction techniques will lead to a
             soil liner that meets performance objectives.

       In terms of CQA, the test pad can provide an extremely powerful tool to  ensure  that
performance objectives are met  The authors recommend a test pad for any project in which failure
of the soil liner to meet performance objecv ves would have a potentially important, negative
environmental impact

       A test pad need not be constructed if results are already available for a particular soil and
construction methodology. By the same token, if the materials or methods of construction change,
an additional test pad  is recommended to test  the  new materials or construction procedures.
Specific CQA tests and observations that are recommended for the test pad are  described later in
Section 2.10.

2.2    Critical Construcrion^Variables that Affect Soil Liners

       Proper  construction of compacted soil liners requires careful attention to construction
variables.  In this section, basic principles are reviewed to set the stage for discussion of detailed
CQC and CQA procedures.

2.2.1  Properties of the So^ Material

       The construction specifications place certain restrictions on the materials that can be used in
constructing a soil liner.  Some of the restrictions are more important than others, and it is
important for CQC and CQA  personnel to understand how material properties  can influence the
performance of a soil liner.

2.2.1.1 P]asriciry Characteristics
                                                                 -^
       The plasticity of a soil refers to the capability of a material to behave as a plastic, moldable
material.  Soils are said to be either plastic or non-plastic. Soils that contain clay  are usually plastic
whereas those that do not contain clay are usually non-plastic.  If the soil is non-plastic, the soil is
almost always considered unsuitable for a soil  liner unless additives such  as bentonite are
introduced.

       The plasticity characteristics of a soil are quantified by three parameters: liquid limit, plastic
limit, and plasticity index.  These terms are defined as follows:

     •   Liquid Limit (LL): The water content corresponding to the arbitrary  limit between tiie
        liquid and plastic states of consistency of a soil.

     •   Plastic Limit (PL): The water content corresponding to the arbitrary  limit between the


                                           35

-------
        plastic and solid states of consistency of a soil.

    «   Plasticity Index (PI):  The numerical difference between liquid and plastic limits, ie.. LL
        -PL.

The liquid limit and plastic limit are measured using ASTM 0-4318.

       Experience  has shown that if the soil has extremely low plasticity, the soil will possess
insufficient clay to develop low hydraulic conductivity when the soil is compacted. Also, soils that
have very low Pi's tend to grade into non-plastic soils in some locations. The question of how
low the PI can be before the soil  is not sufficiently plastic is impossible to answer universally.
Daniel (1990) recommends that the soil have a PI j> 10% but notes that some soils with Pi's as low
as 7% have  been used successfully to build soil liners with extremely low in situ hydraulic
conductivity (Albrecht and Cartwright, 1989).  Benson et al.  (1992) compiled a data base from
CQA documents and related the  hydraulic conductivity measured in the laboratory on small,
"undisturbed" samples of field-compacted soil to various soil characteristics. The observed
relationship between hydraulic conductivity and plasticity index is shown in Fig. 2.8.  The data
base reflects a broad range of construction conditions, soil materials, and CQA procedures. It is
clear from the data base that many soils with Pi's as lew as approximately 10% can be compacted
to achieve a hydraulic conductivity six ID*7 cm/s.
                1.000E-6C
           *2   1.0006-7
               t.oooe-s
               1.000E-9
                        0     10    20    30     40    50     60    70

                                       Piasuctty Index
Figure 2.8 - Relationship between Hydraulic Conductivity and Plasticity Index (Benson et al.,
           1992)
                                          36

-------
      Soils with high plasticity index (>30% to 40%) tend to form hard clods when dried and
sticky clods when wet.  Highly plastic soils also tend to shrink and swell when wetted or dried.
With highly  plastic soils, CQC and CQA personnel should be particularly watchful for proper
processing of clods, effective remolding of clods during compaction, and protection from
desiccation.

2.2.1.2 Percentage Fines

      Some earthwork specifications place a minimum requirement on the percentage of fines in
the soil liner material. Fines are defined as the fraction of soil that passes through the openings of
the No. 200 sieve (opening size * 0.075 mm). Soils with inadequate fines typically have too little
silt- and clay-sized  material to produce suitably  low hydraulic conductivity.  Daniel (1990)
recommends that the soil liner materials contain at least 30% fines.  Data from Benson et aJ.
(1992), shown in Fig.  2.9, suggest that a minimum of 50%  fines might be an  appropriate
requirement for many soils. Field inspectors should check the soil to make sure the percentage of
fines meets or exceeds the minimum stated in the construction specifications and should be
particularly watchful for soils with less than 50% fines.
                1.0006-6
          •=•    1.000E-7
                1.000E-8
               1.000E-9
                       40     50     SO     70      80     90     luu

                                               Fines
Figure 2.9 - Relationship between Hydraulic Conductivity and Percent Fines (Benson et aL, 1992)
                                          37

-------
2.2.1.3 Percentage Gravel

        Gravel is herein defined as particles that will not p-.as through the openings of a No. 4
sieve (opening size » 4.76 mm). Gravel itself has a high hydraulic conductivity. However, a
relatively large percentage (up to about 50%) of gravel can be uniformly mixed with a soil liner
material without significantly increasing the hydraulic conductivity of the material (Fig. 2.10).  The
hydraulic conductivity of mixtures of gravel and clayey  soil is iow because the clayey soil fills the
voids between the gravel particles.  The critical observation for CQA inspectors to make is for
possible segregation of gravel into  pockets that do not contain sufficient soil to plug the voids
between the gravel particles. The uniformity with which the gravel is mixed with the soil is more
important than the gravel content itself for soils with no more than 50% gravel by weight Gravel
also may possess the capability of puncturing geosynthetic materials — the maximum size and the
angularity of the gravel are very important for the layer of soil that will serve as a foundation layer
fora geomembrane.
       10'
               Note: Hydraulic Conductivity of
                    Gravel Atone »170 cm/s
       10-
                              Percent Gravel (by Weight)
•  Kadinite
•  Mine Spoil
Figure 2.10 - Relationship between Hydraulic Conductivity and Percentage Gravel Added to Two
            Oayey Soils (after Shelley and Daniel, 1993).
                                           38

-------
2.2.1.4 Maximum Particle Size

       The maximum particle size is important because; (1) cobbles or large stones can interfere
with compaction, and (2) if a geomembrane is placed on top of the compacted soil liner, oversized
particles can damage the geomembranc. Construction specifications may stipulate the maximum
allowable particle size, which is usually between 25  and 50 mm (1  to 2 in.) for compaction
considerations  Hut which may be much less for protection against  puncture of an adjacent
geomcmbrane.  If a geomembrane is to be placed on the soil liner, only the upper lift of the soil
liner is relevant in terms of protection against puncture. Construction specifications may place one
set of restrictions on all lifts of soil and place more stringent requirements on the upper lift to
protect the geomcmbrane from puncture. Sieve analyses on small samples will not usually lead to
detection of an  occasional piece of oversized material. Observations by attentive CQC and CQA
personnel are the  most effective way to ensure  that oversized materials have been removed.
Oversized materials are particularly critical for the top lift of a soil liner  if a geomembrane is to be
placed on the soil liner to form a composite geomernbrane/soil liner.

2.2.1.5 Clay Content and Activity

       The clay content of the soil  may be defined in several ways but  it is usually considered to
be the percentage of soil that has an equivalent panicle diameter  smaller than 0.005 or 0.002 mm,
with 0.002 mm being the much more common  definition.  The clay content is  measured by
sedimentation analysis (ASTM D-422).  Some construction specifications specify a minimum clay
content but many do not

       A parameter that is sometimes useful is the activity, A, of the soil, which is defined as the
plasticity index (expressed as a percentage) divided by the percentage of clay (< 0.002 mm) in the
soil. A high activity (> 1) indicates that expandable clay minerals such as  mommorillonite are
present.  Lam be  and Whitman  (1969)  report that  the activities  of kaolinite, illitc, and
montmorillonite (three common clay minerals) are 0.38, 0.9, and 7.2, respectively.  Activities for
naturally occurring clay liner materials, which contain a  mix of minerals, is frequent!y in the range
       Benson et al. (1992) related hydraulic conductivity to clay content (defined as particles <
0.002 mm) and reported the correlation shown in Fig. 2.1 1. The data suggest that soils must have
at least 10% to 20% clay in order to be capable of being compacted to a hydraulic conductivity £ 1
K  10*7 cm/s.  However, Benson et al, (1992) also found that clay content correlated closely with
plasticity index (Fig. 2.12). Soils with PI >10% will generally contain at least 10% to 20% clay.

         It is recommended that construction specification writers and regulation drafters indirectly
account for clay content by requiring the soil to have an adequate percentage of fines and a suitably
large plasticity index - by necessity the soil will have an adequate amount of clay.

2.2.1.6  Clod Size

       The term clod refers to chunks of cohesive soil. The maximum size of clods may be
specified in the construction specifications. God  size is very important for dry. hard, clay-rich
soils (Benson and Daniel, 1990). These materials generally mu*r be broken down into small clods
in order to be properly hydrated, remolded, and compacted. God size is less important for wet
soils — soft, wet clods can usually be remolded into a homogeneous, low-hydraulic-conductivity
mass with a reasonable compactive effort.
                                           39

-------
                 1.000E-6c
            *   f.OOOE-7
                1.000E-8
                1.000E-9
                                                                  a
                         10   20    30    40     §0    iO    70    80

                                      Clay Content (2 micron)
Figure 2.11 - Relationship between Hydraulic Conductivity and Gay Content (Benson et al.,
            1992)
      No standard method is available to determine clod size. Inspectors should observe the soil
liner material and occasionally determine the dimensions of clods by direct measurement with §
niter to verily conformance with construction specifications.

2,2.1.7  Bqafppfle

      Bentonite may be added to clay-deficient soils in order to fill the voids between the soil
particles with bentonite and to produce a material that, when compacted, has a very low hydraulic
conductivity. The effect of the addition of bentoniie upon hydraulic conductivity is shown in Fig.
2.13 for one silty sand, for thij particular soil addition of 4% sodium bentonite was sufficient to
lower the hydraulic conductivity to less than 1 % IO7 cm/s.
                                         40

-------
                                                                                                II
fV
60

50


40
—.
of 30
20

10


8 .
0 0
o

o
<$>Q
0 0 8
O
o o° ° o o
o * 
-------
to different degrees.  A fine, powdered bemonitc will behave differently from a coarse, granular
bentonite ~ if the bentonitc was supposed to be finely ground bat loo coarse a grade was delivered,
the bentonite may be unsuitable in the mixture amounts specified. Because bentonite is available in
variable degrees of pulverization, a sieve analysis (ASTM D422) of the processed dry bentonite is
       ended to determine the grain size distribution of the material.
       The most difficult parameters to control are sometimes the amount of bentonite added to the
soil and the thoroughness of mixing.  Field CQC and CQA personnel should observe operational
practices carefully.
                       10-*
                       10-H
                                      !         10         15

                                      Percent Sodium BentonUe
20
 Figure 113- Effect of Addition of Bentonite to Hydraulic Conductivity of Compacted Silty Sand
2.2.2  foldingWater Content

       For natural soils, the degree of saturation of the soil liner material at the time of compaction
is perhaps the single most important variable  that controls the engineering properties of the
compacted material.  The typical relationship between  hydraulic conductivity and molding water
content is shown in Fig. 114.  Soils compacted at water contents less thin optimum (dry of
optimum} tend to have a relatively high hydraulic conductivity; soils compacted at water contents
greater than optimum (wet of optimum)  tend  to have  a low  hydraulic conductivity and low
strength. For some soils, the water content relative to the plastic limit (which is the water content
of the soil when the soil is at the boundary between being a solid and plastic material) may indicate
the degree to which the soil can be compacted to yield low hydraulic conductivity.  In general, if
the water content is greater than the plastl. limit, the soil is in a plastic state and should be capable
of being remolded into a low-hydraulic-conductivity material.  Soils *ith water contents dry of the
plastic limit will exhibit very little "plasticity" and may be difficult to compact into a low-hydraulic*
conductivity mass without delivering enormous compactive energy to the soil. With soil-bentonite
mixes, molding water content is usually not is critical as it is for natural soils.
                                           42

-------
                         t
                                     Molding Water Content
                                     Molding Water Content
          Figure 2.14 - Effect of Molding Water Content on Hydraulic Conductivity
      The water content of highly plastic soils is particularly critical.  A photograph of a highly
plastic soil (PI» 41%) compacted 1% dry of the optimum water content of 17% is shown in Fig.
115.  Large inter-clod voids are visible; the clods c* clay were too dry and hard to be effectively
remolded with the compactive effort used.  A photograph of a compacted specimen of the same soil
moistened to 3% wet of optimum and then compacted is shown in  Fig. 2.16.  At this water
content, the soft soil could be remolded into a homogenous, low-hydraulic-conductivity mass.

-------
                                   16
                     STANDARD

                       PROCTOR
Figure 2.15 - Photograph of Highly Plastic Clay Compacted with Standard Proctor Effort at a
            Water Content of 16% (1% Dry of Optimum).
                                                                                               : •  I
g*;*

 'I
r,i
                                                                                                fc
                                         44

-------
                    STANDARD
                      PROCTOR
Figure 2,16 - Photograph of Highly Plastic Clay Compacted with Standard Proctor Effort at a
            Water Content of 20% (3% Wet of Optimum).


       It is usually preferable to compact the  soil  wet of optimum to minimize hydraulic
conductivity. However, the soil must not be placed at too high a water content.  Otherwise, the
shear strength may be too low, there may be gr^at risk of desiccation cracks forming if the soil
dries, and ruts may form when construction vehicles pass over the liner. It is critically important
that CQC and CQA inspectors verify that the water content of the soil is within the range specified
in the construction documents.
                                          45

-------
2.2.3 Type of Compaction
      In the laboratory, soil can be compacted in four ways:
       1.      Impact Compaction: A ram is repeatedly raised and dropped to compact a lift soil
              into a mold (Fig. 2.17a), e.g., standard and modified Proctor.
      2.
                    gtion:  A piston compacts a lift of soil with a constant stress (Fig.
       2.17b).
3.     Kneadfog Compaction.: A "foot" kneads the soil (Fig. 2.1 7c).
4.     Vibratory Compaction; The soil is vibrated to densify the material (Fig. 2. 17d).
            A. Impact Compaction
                                                 S.  Static Completion
                                             Controtod Fore*
            C. Knaading Compaction
              Conirottsd Force
                                                0, Vibratory Compaction
              mm.
           Figure 2.17 - Four Types of Laboratory Compaction Tests
                                     46

-------
       Experience from the laboratory has shown that the type of compaction can affect hydraulic
conductivity, e.g., as shown in Fig. 2.18.  Kneading the soil  helps to break down clods and
remold the soil into a homogenous mass that is free of voids or large pores.  Kneading of the soil
is particularly beneficial for highly plastic soils. For certain bentonite-soil blends that do not furtn
clods, kneading is not necessary. Most soil liners are constructed with "footed" rollers. The "feet"
on the roller penetrate into a loose lift of soil and knead the soil with repeated passages of the
roller.  The dimensions of the feet on rollers vary considerably.  Footed rollers with short feet (»
75 mm or 3 in.) are called "pad foot" rollers; the feet are said to be "partly penetrating" because the
foot is too short to penetrate fully  a typical loose lift of soil.  Footed rollers with long feet (<• 200
mm or 8 in.) are often called "sheepsfoot" rollers; the feet fully penetrate a typical loose lift Figure
2,19 contrasts rollers with partly and fully penetrating feet
               10 •€
               10  -7
               10 -8
                            S
                            *
                            i
                            E
                                                       A   Static Compaction
                                                       •   Kneading Compaction
                   18         18        20        22        24X

                                        Mokfng Water Content (%)
26
28
Figure 2.18 - Effect of Type of Compaction on Hydraulic Conductivity (from Mitchell et al, 1965)
                                             47

-------
                  Hollar with
               Fully Penetrating
                     Feet
                                     Roller with
                                        Partly
                                     Penetrating
                                        Feet
                                    Loose Utt of Soil
-y-xVV/Vx'J
 .vvvV,V
                                 Compacted Lin of Soil
                                                             --'
   Fully Penetrating F««t on Roilw
   Compact BOM ot Ntw, IJOOM of Soil
   Into Surface ot Old, Previously
   Compacted IHt
                                      Partly Penetrating Feet on Roller Do
                                      Not Extend to Ban oJ New, loose
                                      Uft of Soil and Do Not Compact New
                                      Uft kite Surface of Old Lift
              figure 119 - Footed Rollers with Partly and Fully Penetrating Feet
       Some construction specifications place limitations on the type of roller that can be used to
compact a soil liner. Personnel performing CQC and CQA should be watchful of the type of roller
to nuke sure it conforms to construction specifications. It is particularly important to use a roller
with fully penetrating feet if such a roller is required; use of a non-footed roller or pad foot toller
would result in less kneading of the soil.

2.2.4  Energy of Conrogc, tlgtl
       The energy used to compact soil can have an important influence on hydraulic conductivity.
The data shown in Fig. 2.20 show that increasing the compactive effort produces soil that has a
greater dry unit weight and lower hydraulic conductivity.  It is important that the soil be compacted
with adequate energy if low hydraulic conductivity is to be achieved

       In the field, compactive energy is controlled by:

       1 .     The weight of the roller and the way  the weight is distributed (greater weight
              produces more compactive energy).

       2.     The thickness of a loose lift (thicker lifts produce less compactive energy per unit
              volume of soil).

       3.     The number of passes of the compactor (more passes produces more compactive
              energy).
                                            48

-------
                2
                «
                E
                g
                i
                I
                      14
                        10
               15          20

            Molding Water Content (%)
                     10*
                     10*
10
                     10'
                        10
               15           20

            Molding Water Content (%)
Figure 2.20 - Effect of Compactive Energy on Hydraulic Conductivity (after Mitchell et aL, 1965)
      Man/ engineers and technicians assume that percent compaction is a good measure of
compactive energy.  Indeed, for soils near optimum water content or dry of optimum, percent
compaction is a good indicator of compactive energy:  if the percent compaction is low, then the
compactive energy was almost certainly low.  However, for soil compacted wet of optimum,
                                        49

-------
percent compaction is not a particularly good indicator of compactive energy. This is illustrated by
the curves in Fig. 221. The same soil is compacted with Compactive Energy A and Energy B
(Energy B > Energy A)  to develop the compaction curves shown in Fig. 2.21.  Next, two
specimens are compacted to the same  water content (w^ * WB).  The dry unit weights are
practically identical (Yd^A  " Yd3) despite the fact that the energies of compaction were different.
Further, the hydraulic conductivity (k) of the specimen compacted with the larger energy (Energy
B) has a. lower hydraulic conductivity than the specimen compacted with Energy A despite the fact
that Y4A " Y4B- The percent compaction for the two compacted specimens is computed as follows:
             o
             u
             X
"EmrgyB
                                     Molding Water Content
              I
              e
              a
                a>
                                    Molding Water Content
Rgure 2.21 - Illustration of Why Dry Unit Weight Is a Poor Indicator of Hydraulic Conductivity
            for Soil Compacted Wet of Optimum
                                          50

-------
                          PA - Y4A/lt [YdjnaxlA. then PA > PB- Thus, based on percent compaction,
since PA > PB» °r*e might assume Soil A was compacted with greater compactiv- energy than Soil
B. In fact, just the opposite is true. CQC and CQA personnel are strongly encouraged to monitor
equipment weight, lift thickness, and number of passes (in addition to dry unit weight) to ensure
that appropriate compactive energy is delivered to the soil Some CQC and CQA inspectors have
failed to realize that footed rollers towed by a dozer must be filled with liquid to have the intended
large weight.

      Experience has shown that effective CQC and CQA for soil liners can be accomplished
using the line of optimums as a reference.  The "line of optimums" is the locus of Oopt( Yd .ma*)
points for compaction curves developed on the same soil with different compactive energies (Fig.
2.22). The greater the percentage of actual (w,y,j) points that lie above the  line of optimums the
better the overall quality of construction (Benson and Bourwell,  1992). Inspectors are encouraged
to monitor the percentage of field-measured (w.yd) points that lie on or above the line of optimums.
If the percentage is less than 80%  to 90%, inspectors should carefully consider whether adequate
compactive energy is being delivered to the soil (Benson and Boutwell, 1992).
                                                    Un* of Optimums
                                           'opt
                                                  Molding Water Content (w)
                             Figure 2.22 - Line of Optimums
                                          51

-------
2.2.5  Bonding of Lifts

       If lifts of soil are poorly bonded, a zone of high hydraulic conductivity will develop at
interfaces between lifts.  Poorly bonded lift interfaces provide hydraulic connection between more
permeable zones in adjacent lifts (Rg. 2J23).  It is important to bond lifts together to the greatest
extent possible, and to maximize hydraulic tortuosity along lift interfaces, in order to minimize the
overall hydraulic conductivity.

       Bonding of lifts is enhanced by:

       1.     Making sure the surface of a previously-compacted lift is rough before placing the
              new lift of soil (the previously-compacted lift is often scarified with a disc prior to
              placement of a  new  lift), which promotes bonding and increased hydraulic
              tortuosity along the lift interface,.

       2.     Using a fully-penetrating footed roller (the feet pack the base of the new lift into the
              surface of the previously-compacted lift).

Inspectors should pay particular attention to requirements for scarification and the length of feet on
rollers.
               Good Bonding of Lifts
     Poor Banding o( Lite
        Good Bordirxj of lifts Causn
        Hydraulic Defects in Adjacent
        Lift* To 8* Hydraulcalty
        Unconnected
Poor Bonding of Lifts Causes
Hydraulic Detects in Adjacent
Lifts To Ba Hydrauikalty
Connected To Each Otfcir
                 Figure 2.23 - Flow Pathways Created by Poorly Bonded Lifts


                                             52

-------
2,2.6 Protection Against Desiccation and f .feezing

      Clay soils shrink when they are dried and, depending on the amount of shrinkage, may
crack. Cracks that extend deeper than one lift can be disastrous. Inspectors must be very careful
to make sure that no significant desiccation occurs during or after construction.  Water content
should be treasured if there are doubts.

      Freezing of a soil liner will cause the hydraulic conductivity to increase. Damage caused by
superficial freezing to a shallow depth is easily repaired by rcrolling the surface. Deeper freezing is
not so easily repaired and requires detailed investigation discussed in Section 2.9.2.3.  CQC &
CQA personnel should be watchful during periods when freezing temperatures are possible.
                                            i
2.3   Held Measurement of Water Conjent ar»4 Djy Unit Weight

2.3.1 Water Content Measurement

2.3.1.1  Overnight Oven Drying fASTM D-2216)

      The standard method for determining the water content of a soil is  to oven dry the soil
overnight in a forced-convention oven at 11CK1 This is the most fundememal and most accurate
method  for determining the water content of a soil.  All other methods of measurement ire
referenced to the value of water content determined with this method.

      Were it not for the fact that one has to wait overnight to determine water content  with this
method, undoubtedly ASTM D-2216 would be the only method of water content measurement
used in  the CQC and CQA processes for  soil liners.  However, field personnel cannot wait
overnight to make decisions about continuation with the construction process.

2.3.1.2  Microwave Oven Drying (ASTM D-4643)

      Soil samples  can be  dried in a microwave oven to obtain water contents much more quickly
than can be obtained with conventional overnight oven drying.  The main problem with microwave
oven drying is that if the soil dries for too long in the microwave oven, the temperature of the soil
will rise significantly above 110°C.  If the soil is heated to a temperature greater than 110*C, one
will measure a water content that is greater than the water content of the soil determined by drying
at 110*G  Overheating the soil drives water out of the crystal structure of some minerals and
thereby leads to too much loss of water upon oven drying.

      To guard against overdrying the soil, ASTM method D-4643 requires that the soil be diied
for three minutes and then weighed.  The soil  is then dried for-an  additional minute and
reweighed.  The process of drying for one minute and weighing the soil prevents overheating of
the soil  and forces  the operator to cease the  drying process once the weight of the soil has
stabilized.

      Under ideal conditions, microwave oven drying can yield water contents that are almost
indistinguishable from values measured with conventional overnight oven drying.  Problems that
are sometimes encountered  with microwave oven drying include problems in operating the oven if
the soil contains significant  metal and occasional problems with samples exploding from expansion
of gas in the  interior  of  the sample during microwave oven drying.  Because  errors can
occasionally arise with microwave oven drying, the water content d'etermined with microwave
oven drying should be periodically checked with the value determined by conventional over-night
oven drying (ASTM D-2216).
                                          53

-------
2.3,1.3  Direct Hearing (ASTM EM959)

      Direct heating of the soil was common practice up until about two decades ago. To dry a
soil with direct heating, one typically places a mass of soil into a metallic container (such as a
cooking utensil) and then heats the soil over a flame, e.g., a portable cooking stove, until the soil
first appears dry.  The mass of the soil plus container is then measured. Next, the soil is heated
some more and then re-weighed. This  process is repeated until the mass ceases to decrease
signTicantly (Le., to change by < 0.1% or less).

      The main problem with direct heating is that if the soil is overheated during drying, the
water content that is measured will be too large.  Although ASTM D-4959 does not eliminate this
problem, the ASTM method does warn the user not to overheat the soil. Because errors can do
arise with  direct heating, the water content determined with direct heating should be regularly
checked with the value determined by conventional over-night oven drying (ASTM D-2216).

2,3.1.4  Calcium, Carbide Qag Pressure Tesyr (ASTM Q-4944)

      A known mass of moist soil is placed in a testing device and calcium carbide is introduced.
Mixing is accomplished by shaking and agitating the soil with the aid of steel balls and a shaking
apparatus. A measurement is made of the gas pressure produced. Water content is determined
from a calibration curve. Because errors can occasionally arise with gas pressure testing, the water
content  determined with gas pressure testing should be periodically checked with the value
determined by conventional over-night oven drying (ASTM D-2216).

2.3.1.5  Nuclear Method fASTM D-3017^

      The most widely  used method of measuring the water content of compacted soil is the
nuclear method.  Measurement of water content with a nuclear device involves the moderation or
thermalization of neutrons provided by a  source of fast neutrons. Fast neutrons are neutrons with
an energy  of approximately 5 MeV.  The radioactive source of fast neutrons is embedded in the
interior pan of a nuclear water content/density device (Fig. 2.24). As the fast neutrons move into
the soil, they undergo a reduction in energy every time a hydrogen atom is encountered.  A series
of energy reductions takes place when a neutron sequentially encounters hydrogen atoms.  Finally,
after an average of nineteen collisions with hydrogen atoms, a neutron ceases  to lose further energy
and is said to be a "thermal" neutron with an energy of approximately 0.025 MeV. A detector in
the nuclear device senses the number of  thermal neutrons that are encountered.  The number of
thermal  neutrons that are encountered over a given period of time is a function of the number of
fast neutrons that are emitted from the source and the density of hydrogen atoms in the soil located
immediately below the nuclear device.  Through appropriate calibration, and with the assumption
tha ae only source of hydrogen in the soil is water, the nuclear device provides a measure of the
wtesr content of the soil over an average depth of about ?.00 mm (8 in.).

      There are a number of potential sources of error with the nuclear water content measuring
device.  The most important potential source of error is extraneous hydrogen atoms not associated
with water. Possible sources of hydrogen other than water include hydrocarbons, methane gas,
hydrous minerals (e.g., gypsum), hydrogen-bearing minerals  (e.g., kaolinite, illite, and
monimorillonitc), and organic matter in  the soil.  Under extremely ;  favorable conditions the
nuclear device can yield water content measurements that are as much as ten percentage points in
error (almost always on the high side). Under favorable conditions, measurement error is less than
one percent  The nuclear device should be calibrated for site specific soils and changing conditions
within a given site.

-------
                               Source Rod
                                Gamma
                                Ray Source
                                (C8137)
                                                       Guide Rod
                                                        Fast Neutron Source
                                                        (Am 241 + Be)
                                                                        Neutron
                                                                        Detector
                                                                             Gamma
                                                                               lectors
                        Figure 2.24 - Schematic Diagram of Nuclear Water Content * Density Device
                     Another potential source cf error is the presence of individuals, equipment, or trenches
               located within one meter of the device (all of which can cause an error).  The device must be
               wanned up for an adequate period of time or the readings may be incorrect. If the surface of the
               soil is improperly prepared and the device is not sealed properly  against a smooth surface,
               erroneous measurements can result If the standard count, which is a measure of the intensity of
               radiation from the source, has not been taken recently an erroneous reading may result. Finally,
               many nuclear devices allow the user to input a moisture adjustment factor to correct the water
               content reading by a Fixed amount.   If the  wrong moisture  adjustment factor is stored in the
               device's computer, the reported water content will be in error.

                     It .is very important that the CQC and CQA personnel  be well versed in the proper use of
               nuclear water content measurement devices.  There are many  opportunities for error if personnel
               are not properly trained or do net  correctly  use the equipment. As  indicated later, the nuclear
               device should be checked with other types of equipment to ensure that site-specific variables are
               not influencing test results. Nuclear equipment may be checked against other nuclear devices
               (particularly new devices or recently calibrated devices) to minimize potential for errors.
L
                                                         55


-------
2.3.2 UnitWeiyht

2.3.2.1  Sand Cone fAgTM
      The sand cone is a device for determining the volume of a hole that has been excavated into
soii The idea is to determine the weight of sand required to fill a hole of unknown volume.
Through calibration, the volume of sand that fills the hole can be determined from the weight of
sand needed to fill the hole. A schematic diagram of the sand cone is shown in Fig. Z25.
                                     Sand
                                                    Plastic or
                                                   'Glass Jar
                                     Xn^-
                                     XL"
                                   /  V^


                            Figure 2.25 - Sand Cone Device
      The sand cone is used as follows. First, a template is placed on the ground surface.  A
circle is scribed along the inside of the hole in the template. The template is removed and soil is
excavated from within the area marked by the scribed circle.  The soil that is excavated is weighed
to determine the total weight  (W) of the soil excavated. The excavated soil is oven dried (e.g.,
with a microwave oven) to determine the water content of the soiL The bottle in a sand cone device
is filled with sand and the full bottle is weighed. The template is placed over the hole and the sand
cone device is placed on top of the template. A valve on the sand cone device is opened, which
allows sand to rain down through the inverted funnel of the device and inside the excavated hole.
                                         56

-------
When the hole and funnel are filled with sand, the valve is closed and the bottle containing sand is
weighed.  The difference in  weight before and after the hole is dug is calculated.  Through
calibration, the weight of sand needed tn fill the funnel is subtracted, and the volume of the hole is
computed from the weight of sand that filled the hole.  The total unit weight is calculated by
dividing the weight of soil excavated by the computed volume of the excavated hole. The dry unit
weight is then calculated from Eq. 2.1.

       The sand cone device provides a reliable  technique for determining the dry unit weight of
the soil.  The primary sources of error arc improper calibration of the device, excavation of an
uneven hole that has sharp edges or overhangs that ;an produce voids in the sand-Filled hole,
variations in  the sand, excessively  infrequent calibrations, contamination of the sand by  soil
panicles if the sand is reused, and vibration as from equipment operating close to the sand cone.

2.3.2.2  Rubber Balloon fASTM D-21671

       The rubber balloon is similar to the sand cone except that water is used to fill the excavated
hole rather than sand. A rubber balloon device  is sketched in Fig. 2.26. As with the sand cone
test, the test is performed with the device located on the template over the leveled soil.  Then a hole
is excavated into the soil and the density measuring  device is again placed on top of a template at
the ground surface.  Water inside the rubber balloon device  is pressurized with air  to force the
water into the excavated hole. A thin membrane (balloon) prevents the water from entering the
soil. The pressure in the water forces the balloon  to conform to the shape of the excavated hole. A
graduated scale on the rubber balloon device enables  one to determine the volume of water required
to fill the hole.  The total unit weight is calculated by dividing the known weight of soil excavated
from the hole by the volume of water required to fill the hole with the rubber balloon device.  The
dry unit weight is computed from Eq, 2.1.

       The primary sources of error with the rubber  balloon device are improper excavation of the
hole (leaving small zones that cannot be filled by the pressuri; ed balloon), excessive pressure that
causes local deformation of the adjacent soil, rupture of the balloon, and carelessness  in operating
the device (e.g., not applying enough pressure to force the balloon to fill the hole completely).

2.3.2.3  Drive Cylinder (ASTM p.29371

       A drive cylinder is sketched in Fig. 2.27,  A drop weight is used to drive a thin-walled tube
sampler into the soil. The sampler is removed from the soil and the soil sample is trimmed flush to
the bottom and top of the sampling tube.  The soil-filled tube is weighed and the known weigh' of
the sampling tube itself is subtracted fo determine the gross  we'«ht of the  soil sample.   "Tie
dimensions of the sample are measured to  enable calculation of volume. The unit  weight is
calculated by dividing the known weight by the known volume of the sample. The sample is oven
dried (e.g., in a microwave oven) to determine water content. The dry unit weight is computed
from Eq. 2.1.

       The primary problems with the drive cylinder are sampling disturbance caused by rocks or
stones in the soil, densification of the soil caused by compression resulting from driving of the
tube into the soil, and nonuniform driving of the tube into the soil.  The drive cylinder method is
not recommended for stony or gravely soils.  The drive cylinder method works best for relatively
soft, wet clays that do not tend to densify significantly when the tube is driven into the soil and for
soils that are free of gravel or stones. However, even under favorable circumstances, densification
of die soii caused by driving the ring into the soil cart cause an increase in total unit weight of 2 to 5
pcf (0.3 to 0.8 kN/m3).
                                           57

-------
                  Air Pressure
                  Fitting
                   Rubber Saloon
                 Figure 2.26 - Schematic Diagram of Rubber Balloon Device
2-3.2.4 Nuclear Method f ASTM P-2922'i

       Unit weight can be measured with a nuclear device operated in two ways as shown in Fig.
L28. The most common usage is called direct transmission in which a source of gamma radiation
is lowered down a hole made into the soil to be tested (Fig. 2.28a). Detectors located in the
nuclear density device sense the intensity of gamma radiation at the ground-surface. The intensity
of gamma radiation detected at the surface is a function of the intensity of gamma radiation at the
source  and the total unit weight of the soil material. The second mode of operation of the nuclear
density device is called backscattering.  With this technique the source of gamma radiation is
loaned at the ground surface (Fig. 2.28b).  Hie intensity of gamma radiation detected at the surface
is 2 function  of the: density of  the soil as well as the radioactivity of the source.  With the
backscattcring technique, the measurement is heavily dependent upon the density of the soil within
the upper 25 to 50 mm of soil. The direct transmission method is the recommended technique for
soil liners because direct transmission provides a measurement averaged over a greater depth than
backscattcring.
                                           58

-------
                                                   GuctnRod
                                                   Drop Hammer
                                                        QrtvaHwd
                                                      Sampling Tuba
                       Rgure 2.2? • Schematic Diagram of Drive Ring
       The operation of a nuclear density device in the direct transmission mode is as follows.
First, the area to be tested is smoothed, and a hole is made into the soil liner material by driving a
rod (called the drive rod) into the soil. The diameter of the hole is approximately 25 mm (1 in.)
and the depth of the hole is typically SO mm (2 in.) greater than the depth to which the gamma
radiation source will be lowered below the surface. The nuclear device is then positioned with the
source rod directly over the hole in the soil liner material.  The source rod is then lowered to a
depth of approximately 50 mm (2 in.) above the base of the hole.  The source is then  pressed
against the surface of the hole closest to the detector by pulling on 'he nuclear device and forcing
the source to bear against the side of the hole closest to the detector. The intent is to have good
contact between the source and soil along a direct line from source to detector. The intensity of
radiation at the detector is measured for a fixed period of time, e.g., 30 or 60 s.  The operator can
select the period of counting. The longer the counting period, the more accurate the measurement,
However, the counting period cannot be extended too much because productivity will suffer.
                                           59

-------
                                (A) Direct Transmission
                                 (B) Backscattering
Figure 2.28 - Measurement of Density with Nuclear Device by (a) Direct Transmission and (B)
             Backscattering
       After total unit weight has been determined, the measured water content is used lo compute
dry unit weight (Eq. 2.1).  The potential sources of error with the nuclear device are fewer and less
significant in the density-measuring mode compared to the water content measuring mode.  The
most serious potential source of en or is improper use of the nuclear density device by the operator.
One gross error that is sometimes made is to drive  the source rod into the soil rather than insening
the source rod into a hole that had been made earlier with the drive rod.  Improper separation of
the source from  the base of the hole, an inadequate period of counting, inadequate  waim-up,
spurious sources of gamma radiation, and inadequate calibration are other potential sources of
error.
                                           60

-------
2.4   Inspection of Borrow Sources Prior to Excavation

2.4.1  Sampling for Material Tests

      In order to determine the properties of the borrow soil, samples are often obtained from the
potential borrow area  for laboratory  analysis prior to actual excavation but as pan of the
construction contract. Samples may be obtained in several ways. One method of sampling is to
drill soil borings and recover samples of soil from the borings. This procedure can be very
effective in identifying major strata and substrata within the borrow area. Small samples obtained
from the borings are excellent for index property testing but often do not provide a very good
indication of subtle stratigraphic changes in the borrow area.  Test pits excavated into the borrow
soil with a backhoe, frontend loader,  or other excavation equipment can expose a large cross-
section of the borrow soil.  One  can obtain a much better idea of the variability of soil in the
potential borrow area by examining exposed cuts rather than viewing small soil samples obtained
from borings.

      Large bulk samples of soil are required for compaction testing in the laboratory.  Small
samples of soil taken with soil  sampling devices do not provide a sufficient volume of soil for
laboratory compaction testing. Some engineers combine samples of soil taken at different depths
or from different borings to produce a composite sample of adequate volume.  This technique is
not recommended because a degree of mixing takes place in forming the composite laboratory test
sample that would not take place in the field. Other engineers prefer to collect material from auger
borings  for use  in performing laboratory compaction tests. This technique is  likewise not
recommended without careful borrow pit control because vertical mixing of material takes place
during auguring in a way that would not be expected to occur in the field unless controlled vertical
cuts are made.  The best method for obtaining large bulk  samples of material for laboratory
compaction testing is to take a large sample of material from one location in the borrow source. A
large, bulk sample can be taken from the wall or floor of a test pit that has been excavated into the
borrow area. Alternatively, a large piece of drilling equipment such as a bucket auger can be used
to obtain a large volume of soil from a discreet point in the ground.

2.4.2. Material Tests

      Samples  of  soil must  be taken for laboratory testing to ensure conformance with
specifications for parameters such as percentage fines and plasticity index.  The samples are
sometimes taken in the borrow nit, are sometimes taken from the loose lift just prior to compaction,
and are sometimes taken from both. If samples are taken from the borrow area, CQA inspectors
track the approximate volumes of soil excavated  and sample at the frequency prescribed in the (X<'»
plan.  Sometimes borrow-source testing is performed prior to issuing  of a contract to purchase the
borrow material. A CQA program cannot be implemented for work already completed.  The CQA
personnel will have ample opportunity to check the  properties of soil materials later during
excavation and placement of the soils. If the CQA personnel  for a project did not observe borrow
soil testing, the CQA personnel should review the results of borrow soil  testing to ensure that the
required tests have been performed. Additional testing of the borrow material may be required
during excavation of the material.

      The material tests that are normally performed on borrow soil are water content, Atterberg
limits, panicle size distribution, compaction curve, and hydraulic conductivity (Table 2.2).  Euch
of these  tests is discussed below.
                                          61

-------
                                   Table 12 - Materials Tests
Parameter
ASTMTest
Method
Tide of ASTMTest
Water Con tent
Liquid Limit,
Plastic Limit, &,
Plasticity Index

Panicle Size
Distribution

Compaction
Curve
Hydraulic
Conductivity
D-2216


EM643


D-4944



D-4959


EM318



D-422


D-698



D-1557



D-5084
Laboratory Determination of Water (Moisture)
Content of Soil and Rock

Determination of Water (Moisture) Content of Soil
by the Microwave Oven Method

Field determination of Water (Moisture) Content of
Soil by the Calcium Carbide Gas Pressure Tester
Method

Determination of Water (Moisture) Content by Direct
Heating Method

Liquid Limit, Plastic Limit, and Plasticity Index of
Soils
Panicle Size Analysis of Soil
Moisture-Density Relations for Soils and Soil-
Aggregate Mixtures Using 5.5-lb. (2.48-kg)
Rammer and 12-in. (305-mm) Drop

Moist is-Density Relations for Soils and Soil-
Aggregate Mixtures Using 10-lb. (454-kg)
Rammer and 18-in. (457-mm) Drop

Measurement of Hydraulic Conductivity of
Saturated Porous Materials Using A Flexible Wall
2.4.2.1  Water Content

       It is important to know the water content of the borrow soils so that the need for wetting or
drying the soil prior to compaction can be identified.  The water content of the borrow soil is
normally measured following the procedures outlined in ASTM D-2216 if one can wait overnight
for results.  If not, other test methods described in Section 2.3.1 and listed in Table 2.2 can be
used to produce results faster.
                                               62

-------
2.4.2.2  Atterberg Limits

      Construction specifications for compacted soil liners often require a minimum value for the
liquid limit and/or plasticity index of the soil.  These parameters are measured in the laboratory
with the procedures outlined in ASTM D-4318.

2.4.2.3  Particle Size Distribution

      Construction specifications for soil liners often place limits on the minimum percentage of
fines, the maximum percentage of gravel, and in  some cases the minimum percentage of clay.
Panicle size analysis is performed following  the procedures in ASTM D-422.  Normally the
requirements for the soil material are explicitly stated in the construction specifications. An
experienced inspector can often judge the percentage of fine material and the percentage of sand or
gravel in  the soil. However, compliance with specifications is best documented by laboratory
testing.

2.4.2.4  Compaction Curve

      Compaction curves are developed utilizing the method of laboratory compaction testing
required in the construction specifications. Standard compaction (ASTM D-698) and modified
compaction (ASTM D-1557) are two common methods of laboratory compaction specified for soil
liners. However, other compaction methods (particularly  those unique  to state highway or
transportation departments) are sometimes specified.

      Great care should be taken to follow the procedures  for soil preparation outlined in the
relevant test method.  In particular, the drying of a cohesive material can change the Atterberg
limits as well as the compaction characteristics of the soil. If  the test procedure recommends that
the soil not be dried, the soil should not be dried. Also, care must be taken when sieving the soil
not to remove clods of cohesive material.  Rather, clods of soil retained on a sieve should be
broken apart by hand if necessary to cause them to pass through the openings of the sieve. Sieves
should only be  used to remove stones or other large  pieces of material  following ASTM
procedures.

2.4.2.5  Hydraulic Conductivity

      The hydraulic conductivity of compacted samples of borrow material may be measured
periodically to verify that the soil liner material can  be compacted to  achieve the required low
hydraulic conductivity.  Several methods of laboratory permeation are available, and others are
under development. ASTM D-5084 is the only ASTM procedure currently available. Care should
be taken not to apply excessive effective confining stress to test specimens. If no value is specified
in the CQA plan, a maximum effective stress of 35 kPa (5  psi) is recommended for both liner and
cover systems.

       Care should be taken to prepare specimens for hydraulic conductivity testing properly. In
addition to water content and dry unit weight, the method of compaction and the compactive energy
can have a significant influence on the hydraulic conductivity  of laboratory-compacted soils. It is
particularly important not to deliver too much compactive energy to attain a desired dry unit weight.
The purpose of the hydraulic conductivity  test is to verify that borrow soils can be compacted to the
desired hydraulic conductivity using a reasonable compactive energy.

       No  ASTM compaction method  exists for preparation of hydraulic conductivity test
specimens.  The following procedure is recommended:
                                           63

-------
      1.     Obtain a large, bulk sample of representative material with a mass of approximately
             20kg,

      2.     Develop  a laboratory compaction curve using the procedure specified in the
             construction specifications for compaction control, e.g., ASTM D-698 or D-I557.

      3,     Determine the target water content (wtare«3 and dry unit weight (Yd,targei) for the
             hydraulic conductivity test specimen.  The value of wtarga is normally the lowest
             acceptable water content and Yd,targ« is normally the minimum acceptable dry unit
             weight (Kg. 2.29).

      4.     Enough soil to make several test specimens is mixed to wtarget- The compaction
             procedure used in Step 2 is used to prepare a compacted specimen, except that the
             energy of compaction is reduced, e.g., by reducing the number of drops of die ram
             per lift.  The dry unit weight (Yd) is determined. If Yd - Yd, target, toe compacted
             specimen may be  used for hydraulic conductivity testing.  If Yd  * Yd.targei. then
             another test specimen is prepared with a larger or smaller (as appropriate)
             compactive energy. Trial and error preparation of test specimens  is repeated until Yd
             "" Yd, tared- The procedure is illustrated in Fig. 2.29. The actual compactive  effort
             should be documented along with hydraulic conductivity.

      5.     Atterbrrg limits and percentage fines should be determined for  each bulk sample.
             Water content and dry density should be reported for each compacted specimen.
                   I
                   2*     d.target
                                                      Second Trial
                                                4     secona i

                                                     First TriafS

                                           Water Content
Figure 2.29 -  Recommrnded Procedure for Preparation of a Test Specimen Using Variable (But
             Documented) Compacnve Energy for Each Trial
                                          64

-------
2.4.2.6 Testing Frequency

       The CQA plan should stipulate the frequency of testing. Recommended minimum values
are shown in Table 2.3. The tests listed in Table 2.3 arc normally performed prior to construction
as pan of the characterization of the borrow source.  However, if time or circumstances do not
permit characterization of the borrow source prior to construction, the samples for testing are
obtained during excavation or delivery of the soil materials.
  Table 2.3 - Recommended Minimum Testing Frequencies for Investigation of Borrow Source
Parameter
Frequency
Wata Content

Auerberg Limits

Percentage Fines

Percent Gravel

Compaction Curve

Hydraulic Conductivity
1 Test per 2000 m3 or Each Change in Material Type

1 Tea per 5000 m^ or Each Change in Material Type

1 Test per 5000 m3 or Each Change in Material Type

1 Tea per 5000 m3 or Each Change in Material Type

1 Tea per 5000 m3 or Each Change in Material Type

1 Test per 10,000 m3 or Each Change in Material Type
Note:  Iyd3-0.76m3
2.5    Inspection during Excavation of Borrow Soil

       It is strongly recommended that a qualified inspector whrxreports directly to the CQA
engineer observe all excavation of borrow soil in the borrow pit.  Often the best way to determine
whether deleterious material is present in the borrow soil is to observe the excavation of the soil
directly.

       A key factor for inspectors to observe is the plasticity of the soil.  Experienced technicians
can often determine whether or not a soil has adequate plasticity by carefully examining the soil in
the Held.  A useful practice for field identification of soils is ASTM D-2488, "Description and
Identification of Soils (Visual-Manual Procedure)."  The  following procedure is  used for
identifying clayey soils.
                                            65

-------
         Dry strength: The technician selects enough soil to mold into a ball about 25 mm (1 in.)
         in diameter. Water is added if necessary to form three balls that each have a diameter of
         about 12 mm (1/2 in.). The balls are allowed to dry in the sun. The strength of the dry
         balls is evaluated by crushing them between the fingers. The dry strength is described
         with the criteria  shown in Table 2.4. If the dry strength is none  or low, inspectors
         should be alerted to the possibility that the soil lacks adequate plasticity.

         Plasticity:  The soil is moistened or dried so that a test specimen can be shaped into an
         elongated pat and rolled by hand on a smooth surface or between the  palms into a thread
         about 3 mm (1/8 in.) in diameter.  If the sample is  too wet  to roll  easily it should be
         spread into a thin layer and allowed  to lose some water by  evaporation.  The sample
         threads are re-rolled repeatedly until the thread crumbles at a diameter of about 3 mm (1/8
         in.). The thread will crumble at a diameter of 3 mm when the soil is near the plastic limit.
         The plasticity is described from the criteria shown in Table 2.5, based upon observations
         made during the toughness test.  Non-plastic soils are usually unsuitable for use as soil
         liner materials without use of amendments such as bentonite.
               Table 2.4 - Criteria for Describing Dry Strength (ASTM D-2488)
Description
Criteria
None


Low


Medium


High



Very High
The dry specimen crumbles into powder with mere
pressure of handling

The dry specimen crumbles into powder with some
finger pressure

The dry specimen breaks into pieces or crumbles
with considerable finger pressure

The dry specimen cannot be broken with finger
pressure. Specimen will break into pieces between
thumb and a hard surface

The dry specimen cannot be broken between ihe
thumb and a hard surface
                                            66

-------
                 Table 2.5 - Criteria fa- Describing Plasticity (ASTM D-2488)
Description
Criteria
Nonplasuc


Low


Medium




High
A 3 mm (1/8-in.) thread cannot be rolled at any
water content

The thread can barely be rolled and ihe lump cannot
be formed when drier thai the plastic limit

A thread is easy to roll and not much time is  .
required to reach the plastic limit. The thread   I
cannot be rerolled after reaching the plastic limit.
The lump crumbles when drier than the plastic limit

It takes considerable time rotlirg and kneading to
reach the plastic limit. The thread can be rerolled
several times after reaching the plastic limit. The
lamp can be forned without crumbling when drier
than the plastic limit
2.6    Preprocessing of Materials

       Some soil liner materials are ready to be used for final construction immediately after they
are excavated from the borrow pit. However, most materials require some degree of processing
prior to placement and compaction of the soil.

2.6.1  Water Content Adjustment

       Soils that are too wet must first be dried. If the water content needs to be reduced by no
more than about three percentage points, the soil can be dried after it has been spread in a loose lift
just prior to compaction.  If the water content must be reduced by more than about 3 percentage
points, it is recommendec. that drying take place in a separate processing area. The reason  for
drying in a separate processing area is to allow adequate time for the soil to dry uniformly and to
facilitate mixing of the material during drying.  The soil to be dried is spread in a lift about 225 to
300 mm (9 to 12 in.) thick and allowed to dry.  Water content is periodically measured using one
or more of the methods listed in Table 2.2. The contractor's CQC personnel should check the soil
periodically to determine when the soil has reached the proper water content.

       The  CQA inspectors should check to be sure that the soil is periodically mixed with a disc
or rototiller to ensure uniform drying. The soil cannot be considered to be ready for placement and
compaction unless the water is uniformly Distributed; water content measurements alone do  not
ensure that water is uniformly distributed within the soil.
                                             67

-------
       If the soil must be moistened prior to compaction, the saute principles discussed above for
drying apply; water content adjustment in a separate preprocessing area is recommended if the
water content must be increased by more than about 3 percentage points.  Inspectors should be
careful to verify that water is distributed uniformly to the soil (a spreader baron the back of a water
truck is the recommended device for moistening soil unifoimly), that the soil is periodically mixed
with a disc or rototiller, and that adequate time has been allowed for uniform hydration of the soil.
If the water content is increased by more than three percentage points, at least 24 to 48  hours
would normally be required for uniform absorption of water and hydration of - Ml panicles. The
construction  specifications may limit the  type of water that  can be used; in some cases,
contaminated water, brackish water, or sea water is not allowed.

2.6.2  Rerpoyaj of Oversize
       Oversized stones and rocks should be removed from the soil liner material.  Stones and
rocks interfere with compaction of the soil and may create undesirable pathways for fluid to flow
through the soil liner. The construction specifications should stipulate the maximum allowable size
of particles in the soil liner material.

       Oversized particles can be removed with mechanical equipment (e.g., large screens) or by
hand. Inspectors should examine the loose lift of soil after the contractor has removed oversized
panicles to verify that oversized panicles are not present. Sieve analyses alone do not provide
adequate assurance that oversized materials have been removed -- careful visual inspection for
oversized material should be mandatory.

2.6.3  Pulverisation of Clods,

       Some specifications for soil liners place limitations on the maximum  size of chunks or
clods of clay present in the soil liner material. Discs, rototillers, and road recyclers are examples cf
mechanical devices that will pulverize clods in a  loose lift. Visual inspection of the loose lift of
material is normally performed tc ensure that clods of soil have been pulverized to the extent
required in the construction specifications. Inspectors should be able to visually examine the entire
surface of a loose lift to determine whether clods have been adequately processed.  No standard
method exists for determining clod size.  Inspectors normally measure the dimensions of an
individual clod with a ruler.

2.6.4  Homogenizing Soils

       CQC  and  CQA are very difficult to perform for heterogeneous materials.  It may be
necessary to blend and homogenize soils prior to their use in constructing soil liners in order to
maintain proper CQC and CQA. Soils can  be blended and homogenized"^ a pugmill.  The best
way to ensure adequate mixing of materials is through visual inspection of the mixing process
itself.

2.6.5  l|en|pnite

       Bentonite is a common additive to soil liner materials that do not contain enough clay to
achieve the desired  low hydraulic conductivity. Inspectors must ensure that the bentonite being
used for a project is  in conformance with specifications (i.e., is of the proper quality and gradation)
and that the bentonite is uniformly mixed with soil in the required amounts.

       The parameters that are specified for the bentonite quality vary considerably from project to
project The construction specifications should stipulate the criteria to be met by the bentonite and

                                           68

-------
ihc relevant test methods.  The quality of bemonite is usually measured with some type of
measurement of water adsorption ability of the clay.  Direct measurement of water adsorption can
be accomplished using the plate water adsorption test (ASTM E-946). This test is used primarily
in ;he taconite iron ore industry to determine the effectiveness of bentonite, which is used as a
binder during the pelletizing process to soak up excess water in the ore. Brown (1992) reports that
thousands of plate water adsorption tests have been performed on bentonite, but experience has
been that the test is time consuming, cumbersome, and extremely sensitive to variations in the test
equipment and test conditions. The plate water adsorption test is not recommended for CQO'CQA
of soil liners.

       Simple, alternative tests that provide an induct indication of water adsorption are available.
One indirect test for water adsorption is measurement of Aaerberg (liquid and plastic) limits via
ASTM D-4318.  The higher the quality of the bentonite, the higher the liquid limit and plasticity
index. Although liquid and plastic limits tests are very common for natural soils, they have not
been frequently used as indicators of bentonite quality in the bentonite industry. A comrnonly-used
test in the bentonite industry is the free swell test.  The free swell test  is used to determine the
amount of swelling  of bentonite wh?n bentonite is  exposed  to water in a glass beaker.
Unfortunately, there is currently no ASTM  test for determining free swell of bentonite, although
one is under development.  Until such time as an ASTM standard is developed, ine bentonite
supplier may be consulted for a suggested testing procedure.

       The liquid limit test and free swell test are recommended as the  principal quality control
tests for the quality of bentonite being used on a project.  There are no widely accepted cutoff
values for the liquid limit and free swell. However, the following is offered for the information of
CQC and CQA inspectors. The liquid limit of calcium bentonite is frequently in the range of 100 10
150%. Sodium bentonite of medium quality is expected to have a liquid limit of approximately 300
to 500%.  High-quality sodium bentonite typically has a liquid limit in the range  of about 500 to
700%. According to Brown (1992), e*lci;im bentonnes usually have a free swell of less than 6 cc.
Low-grade sodium bentonites typically have a free swell of 8 - 15 cc.  High-grade bemonites often
have free swell values in the range of 18 to 28 cc.  If high-grade sodium bentonite  is to be used on
a project, inspectors should expect that the liquid limit will be S 500% and the free swell will be 2
18 cc.

       The bentonite must usually also meet gradational requirements.  The gradation of the dry
bentomie may be determined by carefully sieving the beatonite following procedures outHned in
ASTM D-422. The CQA inspector should be particularly careful to ensuse that the oentonite has
been  pulverized to  the extent required  in the  construction specifications.   The  degree of
pulverization is frequently overlooked. Finely-ground, powdered bentonite will behave differently
when blended into soil than more coarsely ground, granular bentonite.  CQC/CQA  personnel
should be particularly careful to make sure that the bentonite is sufficiently finely ground and is not
delivered in too coarse a form (per project specifications); sieve tests'»n the raw bentonite received
at a job site are recommended to verify gradation of the bentonite.

       The benionitc supplier is expected to certify that the bentonite rr.rets :he specification
requirements.  However, CQA inspectors should perform their own tests to ensure compliance
with the specifications. The recommended CQA tests and testing frequencies for bentonite quality
and gradation are summarized in Table 2.6.
                                           69

-------
   Table 2.6 - Recommended Tests on Bcntonite to Dctcrrrune Bcntonite Quality and Gradation
                                  Frequency
                     Test Method
Liquid Limit
Free Swell
Grain Size of Dry Bentomte
IperTrackload
or 2 per Rail Car
I perTrucktoad
or 2 per Rail Car

IfwTrueUaSEt
or 2 per Rail Car
ASTM D-4318, "Liquid Limit,
Plastic Limit, and Plasticity Index
of Soils"

No Standard Procedure Is Available
ASTM CM22, "Particle Size
Analysis of Soil**
2.6.5.1 Pupnill Mixing

       A pygmill is a device for mixing dry materials,  A schematic diagram of a typicaJ pugmill is
shown in Fig. 2.30.  A conveyor belt feeds soil into a mixing unit, and bentonite drops downward
into the mixing unit  The materials are mixed in a large box that contains rotating rods with mixing
paddles. Water may be added to the mixture in the pugmiil, as well.

       The degree  of automation of pugmills varies considerably.  The most  sophisticated
pugmills have computer-controlled devices to monitor the amounts of the ingredients being mixed.
CQA personnel should monitor the controls on  the mixing equipment.

2.6.5.2 In-P|ace
       An alternative mixing technique is to spread the soil in a loose lift, distribute bentonite on
rhe surface, and mix the bentonite and soil using a rototiller or other mixing equipment. There are
several potential problems with in*place mixing.  The mixing equipment may not extend to an
adequate depth and may not fully mix the loose lift of soil with bentonite. Alternatively, the mixing
device may dig too deeply into the ground and actually mix the loose lift in with underlying
materials. Bentonite (particularly powdered bentonite) may be blown away by wind when  it is
placed on the surface of a  loose lift, thui reducing the amount of bentqnite that is  actually
incorporated into the soil. The mixing equipment may fail to pass over all areas of the loose lift
and may inadequately  mix certain portions of the loose lift.  Because of these problems many
engineers believe that pugmill mixing provides a more reliable means for mixing bentonite with
soil. CQA personnel should carefully examine the mixing process to ensure  'hat the problems
outlined above, or other problems, do not compromise the quality of the mixing process. Visual
examination of the mixture to verify plasticity (see Section 2.5 and Table 2.5) is recommended.

2,6.5.3  Measuring Pentopite Content

       The best way to control the amount of bentonite mixed with soil is to measure the relative
weights of soil  and bentonite blended together at the lime of mixing. After bentonite has been
                                            70

-------
                                                                                                     S4
                                                                                                     1-9
mixed with soil there arc several techniques available to estimate the amount of bentonite in the
soil. None of the techniques are particularly easy to use in all situations.

       The recommended technique for measuring the amount of bentonite in soil is the methylene
blue test (Alther. 1983). The methylene blue test is a type of ntration test. Methylene blue is
slowly titrated into a material and the amount of meihylcne blue required to saturate the material is
determined.  The more  bentonite in the soil the greater the amount of methyiene blue that must be
added to achieve saturation. A ca'-bration curve is developed between the amount of methylene
blue needed to saturate the material and the bentonite content of the soil. The methylene blue test
works very  well when bemoi.ite is added into a non-clayey  soil. However, the amount of
methylene blue that must be added to the soil is a function of the amount of clay present in the soil.
If clay minerals other than bentonite are present, the clay minerals interfere with the determination
of the bentonite content. There is no standard methylene blue test; the procedure outlined in Alther
(1983) is suggested until such time as a standard test method is developed.
                         Figure 2.30 - Schematic Diagram of Pugmill
       Another type of test that has been used to estimate bentonite content is the filter press test.
This test is essentially a water absorbency test: the greater the amount of clay in a soil, the greater
the water holding capacity.  Like the methylene blue  test, the filter press test works well if
bentonite is the only source of clay in the soil.  No specific test procedure was available at the time
of this writing.


                                            71

-------
       Measurement of hydraulic conductivity provides a means for verifying that enough
bentcnite has been added to the soil to achieve the desired  low hydraulic conductivity.   If
insufficient bentonite has been added, the hydraulic conductivity should be unacceptably large,
However, just because the hydraulic conductivity is acceptably low for a given sample does not
necessarily mean that the required amount of bentonite has been added to the scii at all locations.
indeed, extra bentonite beyond the minimum amount required is added to soil so that there will be
sufficient bentonite present even at those locations that are "lean" in bentonite.

       The recommended tests and testing frequencies to verify proper addition of bentonite are
summarized in Table 2.7. However, the CQA personnel must realize that the amount of testing
depends on the degree of control in the mixing process: the more control during mixing, the less is
the need for testing to verify the proper bentonite content.
                 Table 2.7 - Recommended Tests to Verify Bentonite Content
Parameter
Frequency
Test Method
Mc'hylcnc Blue Test

Compaction Curve for
Soil-Bcmunitc Mixture
(Needed To Prepare Hydraulic
Conductivity Tcsi Specimen)

Hydraulic Cc xJuctiviiy
of SoiJ-Bcntomte Mixture
Compacted to Appropriate
Waier Content jixJ Dry
Unit Weight
1 per 1,000 m3

I per 5,000 m*
3/ha/Lift
(1/Actc/Uft)
A!ther(1983)


Per Project Specifications, e.g.,
ASTMD-W«orD-I537
ASTM D-5084, "Hydraulic
Conductivity of Saturated Porous
Materials Using a Rouble Wall
Pertneametet"*
Note:  1 yd3 - 0.76
2.6.6
       After the soil has been preprocesscd it is usually necessary to ensunT^hat the water content
does not change prior to use. The stockpiles can be of any size or shape. Small stockpiles should
be covered so that the soil cannot dry or wet.  For large stockpiles, it may not be necessary to
cover the stockpile, particularly if the stockpile is sloped to promote drainage, moisture  is added
occasionally to offset drying at  the surface, or other steps are taken to minimize wetting or drying
of the stockpiled soil.

2.7    gtacemeni of Lqose Lift of Soii

       After a soil has been fully processed, the soil is hauled to the final placement area. Soil
should not be  p,ac*d iu adverse weather corditions, e.g., heavy rain.  Inspectors  are usually
responsible for docuf lenting weather conditions during all earthwork operations.  The surface on
                                             72
                                                                                                          *J

-------
which the soil will be placed must be properly prepared and the materiel must be inspected after
placement to make sure that the material is suitable. Then the CQA inspectors must also verify that
the lift is not too thick.  For side slopes, construction specifications should clearly state whether
lifts are parallel to the slope or horizontal. For slopes inclined at 3(H):i(V) or flatter, lifts are
usually parallel to the slope.  For slopes inclined  at 2(H):1(V) or steeper, lifts are usually
horizontal. However, horizontal lifts may present problems because the hydraulic conductivity for
flow parallel to lifts is expected to be somewhat greater than for flow perpendicular to lifts. Details
of testing are described in the following subsections.

       Transport vehicles cart pick up contaminants while hauling material from the borrow source
or preprocessing area.  If this occurs, measures  should be taken to prevent contaminants from
falling off transport vehicles into the soil liner material. These measures may include restricting
vehicles to contaminant free haul roads or removing contaminants before the vehicle enters the
placement area,

2.7.1   Surface Scarification

       Prior to placement of a new lift of soil, the surface of the previously compacted lift of soil
liner should be roughened to promote good contact  between the new and old lifts.  Inspectors
should observe the condition of the surface of the previously compacted lift to make sure that the
surface has been scarified as required in the construction specifications. When soil is scarified it is
usually roughened to a depth of about 25 mm (1 in.). In some cases the surface may not require
scarification if the surface is already rough after the end of compaction of a lift  It is very important
that CQA inspectors ensure that the soil has been properly scarified if construction specifications
require scarification. If the soil is scarified, the scarified zone becomes part of the loose lift of soil
and should be counted in measuring the loose lift thickness.

2.7.2  Material Tests and Visual Inspection

2.7.2.1  Material Tests

         After a loose lift of soil has been placed, samples are periodically taken to confirm the
properties of the soil liner material. These samples are in addition to samples taken  from the
borrow area (Table 2.3). The types of tests and frequency of testing arc  normally specified in the
CQA documents.  Table 2.8 summarizes recommended minimum tests and testing frequencies.
Samples of soils can be taken either on a grid pattern or on a random sampling pattern (see Section
2.8.3.2). Statistical tests and criteria can be applied but are not usually applied to soil liners in part
because enough data have  to be gathered to apply statistics, and yet decisions have  to be made
immediately, before very much data are collected.

2.7.2.2  Visual Observations

       Inspectors should position themselves near the working face of soil liner material as it is
being placed.  Inspectors should look for deleterious materials such as stones, debris, and organic
matter.  Continuous inspection of the placement of soil Hner material is recommended to ensure that
the soil liner material is of the proper consistency.

2.7.2.3  Allowable Variations

       Tests on soil liner materials may occasionally fail to conform  with required specifications.
It is unrealistic to think that 100% of a soil liner material will be in  complete conformance with
specifications.  For example, if the construction documents require a minimum plasticity index it
                                            73

-------
may be anticipated that a small fraction of the soil (such as pockets of sandy material) will fail to
conform with sp *ifications.  It is neither unusual nor unexpected that occasional failing material
will be encountered in .toil liners.  Occasional imperfections in soil liner materials are expected.
Indeed, one of ths reasons why multiple lifts are used in soil liners is to account for the inevitable
variations in the materials of construction employed in building soil liners. Occasional deviations
from construction specifications are not harmful. Recommended maximum allowable variations
(failing tests) are listed in Table 2.9.
Table 2.8 -    Recommended Materials Tests for Soil Liner Materials Sampled after Placement in a
              Loose Lift (lust Before Compaction)
                             Test Method
                                                   Minimum Testing Frequency
Percent Pines
(No* I)
Percent Gravel
(Nwe3)
Liquid & Plastic limits
Percent Ben tonne
(N«e4)
Compaction Curve
Construction Oversight
ASTMCM140
ASTMEW22
ASTMD-4318
A1ther(1983)
As Specified
Observation
1 per 800m3 (Notes 2 & 5)
1 per 8QOm3 (Notes 2 &S)
1 per 800m3 (Notes 2 & 5)
1 per 800 m3 (Noes 2 45)
1 per 4,000m3 (Note 5)
Continuous
Notes:

I.

2.


3.

4.


5.
Percent Tines is defined as percent passing the No. 200 sieve.

In addition, at least one test should be performed each day thai soil is placed, and additional tests should be
performed on any suspect material observed by CQA personnel,

flucem gravel is defined as percent retained on the No. 4 sieve.

This test is only applicable to soil-bcnionite  liners.

1 yd3 » 0.76m3.
                                              74

-------
            Table 2.9 - Recommended Maximum Percentage of Failing Material Tests
Parameter
Maximum Allowable Percentage of Outlier.
Auerberg L'mia

Percent Fines

Pcrctnt Gravel

God Size

Percent Bemonite

Hydraulic Conductivity of
Laboratory  Compacted Soil
3% and Outliers Not Concentrated in One Lid or One Area

5% and Outliers Not Concentrated in One Lift or One Area

10% and Outliers Not Concentrated in One Lift or One Area

10% and Outliers Not Concentrated in One L ft or One Area

5% and Outliers Not Concentrated in One Lift or One Area

5% and Outliers Not Concentrated in One Lift or One Area
2.7.2.4  Corrective Action

       If it is determined that the materials in an area dp not conform with specifications, the first
step is to define the extent of the area requiring repair.  A sound procedure is to require the
contractor to repair the  lift of soil out to the limits defined by passing CQC/CQA tests.  The
contractor should r.ot be allowed to guess at the extent of the area that requires repair.  To define
the limits of the area that requires repair, additional tests are often needed. Alternatively, if the
contractor chooses not to request additional tests, the contractor should repair the area that extends
from the failing test out to the boundaries defined by passing tests.

       The usual corrective action is to wet or dry the loose lift of soil in place if the water content
is incorrect.  The water must be added uniformly, which requires mixing the soil wiih a disc or
rototiller (see  Section 2.6.1).   If the soil contains oversized material, oversized panicles are
removed from  the material (see Section 2.6.2). If clods are too large,  clods can be pulverized in
the loose lift (see Section 2.6.3).  If the soil lacks adequate plasticity, contains too  few fines,
contains too much gravel, or lacks adequate bentonite, the material is normally excavated and
replaced.

2.7.3  Placement and Control of Loose Lift Thickness

       Construction specifications normally place limits on the maximum thickness of a loose lift
of soil, e.g., 225 mm (9 in.).  The thickness of a loose lift  should not exceed this value with
normal equipment.  The thickness of a loose lift  may be determined in several ways.  One
technique is  for an inspector standing near the working face of soil being placed to observe the
thickness of the lift  This is probably the most reliable technique for controlling loose lift thickness
for CQA inspectors. If there is a question about loose lift thickness one should dig a pit through
the loose lift  of soil and into the underlying layer.  A cross-beam is used to measure the depth from
the surface of a loose lift to the top of the previously compacted lift If the previously compacted
lift was scarified, the zone of scarification should be counted in the loose lift thickness for the new
layer of soil. Continuous observation of loose lift thickness is recommended during placement of

                                             75

-------
soil liners.

       Some earthwork contractors control lift thickness by driving grade stakes into the subsoil
and marking the grade stake to indicate the proper thickness of the next layer. This practice is very
convenient fot equipment operators because they can tell at a glance whether the loose lift thickness
is correct. However, this practice is strongly discouraged for the second and subsequent lifts of a
soil liner because the penetrations into the previously-compacted lift made by the grade stakes must
be repaired. Also, any grade stakes or fragments from grade stake- left in a soil liner could
puncture overlying geosymhetics. Repair of holes left by grade stakes is very difficult because one
must dig through  the loose lift of soil to expose the grade stake, remove the grade stake without
breaking the stake and leaving some of the stake in the soil, backfill the hole left by the grade stake,
and then replace the loose soil in the freshly-placed lift. For the first lift of soil liner, repair of
grade stake holes may not be relevant (depending on the subgrade and what its function is), but
grade stakes are discouraged even for the first lift of soil because the stakes may be uften broken
off and incorporated into the soil. Grade stakes resting on a small platform or base do not need to
be driven into the  underlying material and are, therefore, much more desirable than ordinary grade
stakes. If grade stakes are used, it is recommended that they be numbered and accounted for at the
end of each shift; this will provide verification that grade stakes are not being abandoned in the fill
material.

       The recommended survey procedure for control of lift thickness involves laser sources and
receivers. A laser beam source is set at a known elevation, and reception devices held by hand on
rods or mounted to grading equipment are used to monitor lift thickness. However, lasers cannot
be used at all sites. For instance, the liner may need to be a minimum distance above rock, and the
grade lines may  follow the contours of underlying rock.  Further, every site has  ajeas such as
comers, sumps, and boundaries of cells, which preclude the use of lasers,

       For those areas where lasers cannot be used, it is recommended that either flexible plastic
grade stakes or metallic grade stakes (numbered and inventoried as  pan of the QA/QC process) be
used.  It is preferable if the stakes are mounded on a base so that the stakes do not have to be
driven into the underlying lift.  Repair of grade stake holes should  be required; the repairs should
be periodically inspected and the repairs documented.  Alternatively (and preferably for small
areas), spot elevations can be obtained on the surface of a loose lift with conventional level and rod
equipment, and adjustments made by the equipment operator based on the levels.

       When soil is placed, it is usually dumped into a heap  at the working face arH spread with
dozers.  QA/QC  personnel should stand in front of the working face to observe the soil for
oversized materials or other deleterious material, to visually observe loose lift  thickness, and  to
make  sure that the dozer does not damage an underlying layer.

2.8    Reropldjpg and Compaction of Soil

2.8.1   Com.pactlQn, Equipment

       The important parameters concerning compaction equipment are the type and weight of the
compactor, the characteristics of any feet on the drum, and the weight of the roller per unit length
of drummed surface. Sometimes construction specifications will stipulate a required type of
compactor or minimum weight of compactor. If this is the case inspectors should confirm that the
compaction equipment is in conformance with specifications.  Inspectors should be particularly
cognizant of the weight of compactor and length of feet on drummed rollers. Heavy compactors
with long feet that fully penetrate a loose lift of soil are generally thought to be the best type of
compactor to use for soil liners. Footed rollers may not be necessary or appropriate for some

-------
bentonite-soil mixes; smooth-drum rollers or rubber tired rollers may produce best results for soil-
bentonitc mixtures that do not require kneading or remolding to achieve low hydraulic conductivity
bnl only require densification.

       Some compactors are self-propelled while other compactors are towed. Towed, footed
rollers are normally ballasted by filling the drum with water to provide weight that will enable
significant compactive effort to be delivered to the soil.  Inspectors should be very careful to
determine whether or not all drums on towed rollers have been filled with liquid.

       Compacting soil liners on side slopes can present special challenges, particularly for slopes
inclined at 3(H):1(V) or steeper. Inspectors should observe  side-slope compaction carefully and
watch for any tendency for the compactor to slip down slope or for slippage or cracking to take
place in the soil. Inspectors should also be watchful to make sure that adequate compactive effort
is delivered to the soil. For soils compacted in lifts parallel to the slope, the first lift of soil should
be "knitted" into existing subgrade to minimize a preferential flow path along the interface and to
minimize development of a  potential slip plane.

       Footed rollers can become clogged with soil between the feet. Inspectors should examine
the condition of the roller  to make sure that the space between feet is not plugged with soil. In
addition, compaction equipment is intended to be operated at a reasonable speed.  The maximum
speed of the compactor should be specified in the construction specifications.  CQC and  CQA
personnel should make sure the speed of the equipment is not toe great.

       When soils are placed directly on a fragile layer, such as a geosynthetic material, or a
drainage material,  great care must be taken in placing and compacting the first lift so as not to
damage the fragile material  or mix clay in with the underlying drainage material. Often, the first lift
of soil is considered  a sacrificial lift that is placed, spread  with dozers,  and only nominally
compacted with the dozers or a smooth-drum or rubber-tire roller. QA/QC personnel should be
particularly careful to observe all placement and compaction operations of the first lift of soil for
compacted soil liners placed directly on a grosynthctic material or drainage layer.

       It is not  uncommon for a contractor to use more than one type of compaction equipment on
a project. For example, initial compaction may be with a heavy roller having long feet that fully
penetrate a loose lift of soil. Later, the upper part of a lift may be compacted with a heavy nibber-
tired roller or other equipment that is particularly effective in compacting near-surface materials.

2.8.2  Number of Passes

       The compactive effort delivered by a roller is a function of the number of passes of the
roller over a giver  area of  soil. A pass may defined as one pass of thcxonstruction equipment or
one pass of a drum over a given point in the soil liner.  It does not matter whether a pass is defined
as a pass of the  equipment or a pass of a drum, but the construction specifications and/or CQA plan
should define what is meant by a pass. Normally, one pass of the vehicle constitutes a pass for
self-propelled rollers and one pass of a drum constitutes a pass for towed rollers.
 follows:
       Some construction documents require a minimum coverage.  Coverage (C) is defined zs
                           C = (Af/Ad x N x 100%
(2.4)
 where N is the number of passes of the roller, Af is the sum of the area of the feet on the drums of
 the roller, and Ad is the area the drum itself. Construction specifications sometimes require 150% -
                                           77

-------
200% coverage of the roller.  For a given roller and minimum percent coverage, the minimum
number of passes (N) may be computed.

       The number of passes of a compactor over the soil can have an important influence on the
overall hydraulic conductivity of the soil lirer.  It is recommended that periodic observations be
made of the number of passes of the roller over a given point. Approximately 3 observations per
hectare per lift (one observation per acre per lift) is the recommended frequency of measurement.
The minimum number of passes that is reasonable depends upon many factors a,nd cannot be stated
in general terms. However, experience has been that at lecst 5 to 15 passes of a compactor over a
given point is usually necessary to remold and compact clay liner materials thoroughly.

2.8.3  Water Content and Dry Unit Weight

2.8.3.1  Water Content and Unit Weight Tests

        One of the most important CQA tests is measurement of water content and dry unit
weight.  Methods of measurement were discussed in Section  2.3.  Recommended  testing
frequencies are listed in Table 2.10. It is stressed that the recommended testing frequencies are die
minimum values.  Some judgment should be applied to these numbers, and the testing frequencies
should be increased or kept at  the minimum depending on the specific project and other QA/QC
tests and observations.  For example, if hydraulic conductivity tests are  not performed on
undisturbed samples (see Section 2.8.4.2), more water content/density tests may be required than
the usual minimum.

2.8.3.2 Sampling Patterns

       There are several ways in which sample locations may be selected for water content and
unit weight tests. The simplest and least desirable method is for someone in the field to select
locations at the time samples must be taken. This is undesirable because the selector may introduce
a bias into the sampling pattern.  For example, perhaps on the previous project soils of one
particular color were troublesome. If the individual were to focus most of the  tests on the current
project on soils of that same color a bias might be introduced.

       A common method of selecting sample locations is to establish a grid pattern. The grid
pattern is simple and ensures a  high probability of locating defective areas so long as the defective
areas are of a size greater than or equal to the spacing between the sampling points.  It is important
to stagger the grid patterns in successive lifts so that sampling points are not  at the same location in
each lift One would not want to sample at the same location in successive lifts because repaired
sample penetrations would be stacked on top of one another.  The grid pattern sampling procedure
is the simplest one to use that avoids the potential for bias described in the previous paragraph.

       A third alternative for  selecting sampling points is to locate  sampling points randomly.
Tables and examples are given in Richardson (1992). It is recommended that no sampling point be
located within 2 meters of another sampling point.  If a major portion of the area to be sampled has
been omitted as a result of  the random sampling proems, CQA inspectors may add additional
points to make sure the area receives some testing. Random sampling is sometimes preferred on
large projects where statistical procedures  will be used to evaluate data. However, it can be
demonstrated that for a given  number of sampling points, a grid pattern will be more likely  to
detect a problem area provided ihat the dimensions of the problem area are greater than or equal to
the spacing between samplir g points. If the problem area is smaller thar. the spacing between
sampling points, the probability of locating the problem area is approximately the same with both a
grid pattern and a random pattern of sampling.
                                          78

-------
              Table 2.10 - Recommended Tests and Observations on Compacted Soil
Parameter
                                  Test Method
Minimum Testing Frequency
Water Content (Rapid)
    (Motel)
Water Content
    (Nose 3)
Total Density (Rapid)
    (Note 4)

Total Density
    (Note 5)
Number of Passes
Construction Oversight
                                  ASTM D-3017
                                  ASTM D-4643
                                  ASTMD-4944
                                  ASTM D-4959

                                  ASTMD-2216
                                  ASTM D-2922
                                  ASTMD-2937

                                  ASTM D-1556
                                  ASTM D-1587
                                  ASTM D-2167

                                  Observation
                                  Observation
I3/ha/!ift(5/acre/lift)
(Notes 2 & 7)
One in every 10 rapid water
content tests
(Notes 3 & 7)

13/ha/lifl(5/acre/lift)
(Notes 2,4 & 7)

One in every 20 rapid density teas
(Noes 5, 6, & 7)
3/ha/lifl (I/acre/lift)
(Notes 2 & 7)

Continuous
Notes:
2.
3.
4.
5.
6.
    ASTM D-3017 is a nuclear method, ASTM D-4643 is microwave oven drying, ASTM D-4944 is a calcium
    carbide gas pressure tester method, and ASTM D-4959 is a direct heating method. Direct water content
    determination (ASTM D-22I6) is the  standard against which nuclear, microwave, or other methods of
    measurements are calibrated for on-sue soils.

    In addition, at ieasl one test should be performed each day soil is compacted and additional tests should be
    performed T aras for which CQA personnel have reason to suspect inadequate compaction.

    Evert  1/6- th the diameter of the sample,
7.  1 acre » 0.4 ha.
                                                 79

-------
X
      No matter which method of determining sampling points is selected, it is imperative that
CQA inspectors  have the responsibility to perform additional tests on any suspect area.  The
number of additional testing locations that ate appropriate varies considerably from project to
project

2.8.3.3  Tests with Different Device? to^Mjnfoiize Systematic Iforors.

      Some methods of measurement may introduce a systematic error. For example, the nuclt-r
device for measuring water conteut may consistently produce d water content measurement that is
too high  if there is an extraneous source of hydrogen atoms besides water in the soil.   It is
important that devices that may introduce a significant systematic error be periodically correlated
with measurements that do not have such error. Water content measurement tests have the greatest
potential  for systematic error.  Both the nuclear method as well as microwave oven drying can
produce significant systematic error under certain conditions. Therefore, it is recommended that if
the nuclear method or any of the rapid methods of water content measurement (Table 2.2) are used
to measure water content, periodic correlation tests should be made with conventional overnight
oven drying (ASTM D-2216).

      It  is suggested that at the beginning of a project, at least 10 measurements of water content
be determined on representative samples of the site-specific soil using any rapid measurement
method to be employed on the project as well as ASTM D-2216. After this initial correlation, it is
suggested (see Tables 2.10) that one in ten rapid water content tests be crossed check with
conventional overnight oven drying.  At the completion of a project a graph should be presented
that correlates the measured water content with a rapid technique against the water content from
conventional overnight oven drying.

      Some methods of unit  weight measurement may also introduce bias.  For example, the
nuclear device may not be properly calibrated and could lead to measurement of a unit weight that
is either too high or too low.  It is recommended that unit weight be measured independently on
occasion  to provide a check against systematic errors. For example, if the nuclear device is the
primary method of density measurement being employed on a project, periodic measurements of
density with the  sand cone or rubber balloon device can  be used to check the nuclear device.
Again, a  good practice is to perform about 10 comparative tests on representative  soil prior to
construction.  During construction, one in every  20 density tests (see Table 2.10) should be
checked with the sand  cone or rubber balloon.  A  graph should be made of the unit weight
measured with the nuclear device versus the unit weight measured with the sand cone or rubber
balloon device to show me correlation.  One could either plot dry unit weight or total unit weight
for the correlation.  Total unit weight in some ways is more sensible because the methods of
measurement are actually total unit weight measurements; dry unit weight is calculated from the
total unu  weight and water content (Eq. 2.1.).                            -._

2.8.3.4  AJlowabjg Variations and Outliers

      There are several reasons why a field water content or density test may produce a failing
result, i.e., value outside of the specified range. Possible causes for a variation include a human
error in measurement of water content or dry unit weight, natural variability of the soil or the
compaction process leading to an anomaly at an isolated location, limitations in the sensitivity and
repeatability of the test methods, or inadequate construction procedures that reflect broader-scale
deficiencies.

      Measurement errors are made on every project. From time to time it can be expected that
CQC and CQA personnel will incorrectly measure either the water content or the dry unit weight
                                                        80

-------
Periodic human errors are to be expected and should be addressed in the CQA plan,

       If it is suspected that a test result is in error, the proper procedure for rectifying the error
should be as follows.  CQC or CQA personnel should return to the point where the questionable
measurement was obtained. Several additional tests should be performed in close proximity to the
location of the questionable  test.  If all of the repeat tests provide satisfactory results the
questionable test result may be disregarded as an error. Construction quality assurance documents
should specify the number of tests required to negate a blunder.  It is recommended that
approximately 3 passing tests be required to negate the results of a questionable test.

       One of the main reasons why soil liners are  built of multiple lifts is a realization that the
construction process and the materials themselves vary. With multiple lifts no one particular point
in any one lift is  especially significant even if that point consists of unsatisfactory material or
improperly compacted material. It should be expected that occasional deviations from construction
specifications will be encountered for any soil liner.  In fact, if one were to take enough soil
samples, one can rest assured that a failing point on some scale would be located.

       Measurement techniques for compacted soils are imperfect and produce variable results.
Turnbull et al. (1966) discuss statistical quality control  for compacted soils. Noorany (1990)
describes 3 sites in the San Diego area for which 9 testing laboratories measured water content and
percent compaction on the same fill materials. The ranges in percent compaction were very large:
81-97% for Site 1,77-99% for Site 2, and 89-103%  for Site 3.

       Hilf (1991) summarizes statistical data from 72 earth dams; the data show that the standard
deviation in water content is typically 1 to 2%, and the standard deviation in dry density is typically
0.3 to 0.6 kN/m^  (2 to 4 pcf)-  Because the standard deviations are themselves or the same order
as the  allowable range of these parameters in many earthwork specifications, it is statistically
inevitable that there will be some failing tests no matter now well built the soil liner is.

       It is unrealistic ta expect that  100%  of all CQA tests  will t* in  compliance with
specifications. Occasional deviations should be anticipated. If there are only a few randomly-
located failures, the deviations in no way compromise the quality or integrity of a multiple-lift liner.

       The CQA documents  may provide an allowance for an occasional failing test.  The
documents may stipulate that failing tests not be permitted to be concentrated in any one lift or in
any one area.  It is recommended that a small percentage of failing tests be allowed rather than
insisting upon the unrealistic requirement that  100% of all  tests meet project objectives.
Statistically based requirements provide a convenient yet safe and reliable technique for handling
occasional failing test results.. However, statistically based methods require that enough data be
generated to  apply statistics reliably.  Sufficient data to apply statistical methods may not be
available, particularly in the early stages of a project

       Another approach is to allow a small percentage of outliers but to require repair of any area
where the water content is far too low or high or the dry unit weight is far too low. This approach
is probably the simplest to implement — recommendations are summarized in Table 2.11.

-------
         Table 2.11 - Recommended Maximum Percentage of Failing Compaction Tests
Parameter
Maximum Allowable Percentage of Outliers
Water Content
Dry Density
Number of Passes
3% and Oinlien Not Concentrated in One Lift or One Aita,
and No Wuer Content Less tnan 2% or More than 3% of
the Allowable Value

3% and Outlien Not Concentrated in One Lift or One Area,
and No Dry Density Less than 0.8 kN/m3 (5 pcf) Below the
Required Value

5% and Outlieis Not Concentrated in One Lift or One Area
2.8.3.5 Corrective Action

       If it is determined that an area does not conform with specifications and that the area needs
to be repaired, the first step is to define the extent of the area requiring repair. The recommended
procedure is to require the contractor to repair the lift of soil out to the limits defined by passing
CQC and CQA tests.  The contractor should not be allowed to guess at the extent of the area that
requires repair. To define the limits of the area that requires  repair, additional tests are often
needed. Alternatively, if the contractor chooses not to request additional tests, the contractor
should repair the area that extends from the failing test out to the boundaries defined by passing
tests.

       The usual problem requiring corrective action at this stage is inadequate compaction of the
soil. The contractor is usually able to rectify the problem with additional passes of die compactor
over the problem area.

2.8.4  Hydraulic Conductivity Tests op Undisturbed Samples

       Hydraulic conductivity tests are often performed on "undisturbed" samples of soil obtained
from a single lift of compacted soil liner.  Test specimens are trimmed from the samples and are
permeated in the laboratory.  Compliance with the  stated hydraulic conductivity criterion is
checked.                                                               "-

       This type of test  is given far too much weight in most QA programs.  Low hydraulic
conductivity of samples taken from the liner is necessary for a well-constructed liner but is not
sufficient to demonstrate that the large-scale, field hydraulic conductivity is adequately low. For
example, Elsbury et ai.  (1990) measured hydraulic conductivities on undisturbed samples of a
poorly constructed liner that averaged 1 x IQ~9 crn/s, and yet the actual in-field value was 1 x lQrs
cm/s. The cause for the discrepancy was the existence of macro-scale flow paths in the field that
were not simulated in the small-sized (75 mm or 3 in. diameter) laboratory test specimens.

       Not only does the flow pattern through a 75-rum-diameter test specimen not necessarily
reflect flow patterns on a larger field scale, but the  process of obtaining a sample for testing
inevitably disturbs the soil. Layers are distorted, and gross alterations occur if significant gravel is
                                           82


-------
present in the soil. The process of pushing a sampling tube into the soil densifies the soil, which
lowers its hydraulic conductivity. The harder and drier the soil, the greater the disturb?nce. As a
result of these various factors, the large-scale, field hydraulic conductivity is almost always greater
than or equal to the  small-scale, laboratory-measured hydraulic conductivity.  The difference
between values from a small laboratory scale and a large field scale depends on the quality of
construction — the better the quality of construction, the less the difference,

      Laboratory hydraulic conductivity tests on undisturbed samples of compacted liner can be
valuable in some situations.  For instance, for soil bentom'te mixes, the laboratory test provides a
check on whether enough bemonite has been added to the mix to achieve the desired hydraulic
conductivity. For soil liners in which a test pad is not constructed, the  laboratory tests provide
some verification that appropriate materials have been used and compaction was reasonable (but
hydraulic conductivity tests by themselves do not prove this fact).

      Laboratory hydraulic conductivity tests constitute a major inconvenience because the tests
usually take at least several days, ar.' sometimes a week or two, to complete. Their value as QA
tools is greatly diminished by the lor ^ testing time — field construction personnel simply cannot
wait for  the results  of the  tests to proceed with construction,  nor would the QA personnel
necessarily want them to wait  because opportunities exist for damage of the liner as a result of
desiccation.  Thus, one should  give very  careful consideration as to  whether the laboratory
hydraulic conductivity tests are truly needed for a given project and will serve a sufficiently useful
purpose to make up for the inconvenience of this type of test.

      Research is currently underway to determine if larger-sized samples from field-compacted
soils can give more reliable results than the usual 75-mm (3 in.) diameter samples.  Until further
data are developed, the following recommendations are made concerning  the approach to utilizing
laboratory hydraulic conductivity tests for QA on field-compacted soils:

       1.     For gravely soils or other soils that cannot be consi*»;ntly sampled without causing
              significant disturbance, laboratory hydraulic conductivity tests should not be a part
              of the QA program because representative samples cannot  realistically be obtained.
              A test pad (Section 2.10) is recommended to verify hydraulic conductivity.

      2.     If a test pad is constructed and it is demonstrated that the field-scale hydraulic
              conductivity is satisfactory on the test pad, the QA program for the actual soil liner
              should focus on establishing that the actual liner is built of similar materials and to
              equal or better standards compared to the test pad  —  laboratory  hydraulic
              conductivity testing is not necessary to establish this.

       3.     If no test pad is constructed and it is believed that representative samples can be
              obtained for hydraulic conductivity testing, then laboratory hydraulic conductivity
              tests on undisturbed samples from the field are recommenddrt.

2.8.4.1   Sampling for Hydraulic Conductivity Testing

      A thin-walled tube is pushed into the soil to obtain a sample. Samples of soil should be
taken in the manner that minimizes disturbance such as described in ASTM D-1587.  Samples
should be sealed and carefully stored  to prevent  drying and transported  to the laboiatory in a
manner that minimizes soil disturbance as described in ASTM D-4220.

       It is particularly important that the thin-walled sampling tube be pushed into the soil in the
direction perpendicular to the plane of compaction. Many CQA inspectors will push the sampling


                                            83

-------
tube into the soil using the blade of a dozer or compactor.  This practice is not recommenced
because the sampling tube tends to rotate when it is pushed into the soil. The recommended way of
sampling the soil is to push the sampling tube straight into the soil using a jack to effect a smooth,
straight push.

       Sampling of gravely soils for hydraulic conductivity testing is often a futile exercise. The
gravel particles that are encountered by the sampling tube tend to tumble and shear during the push,
which caused major disturbance of the soil sample.  Experience has been that QA/QC personnel
may take several samples of gravely soil before a sample that is sufficiently free of gravel to e- able
proper sampling is finally  obtained;  in  these cases, the badly disturbed, gravely sample are
discarded.  Clearly, the process of discarding samples because they contain too much gravel to
enable proper sampling introduces a bias into the process.  Gravely soils are not amenable to
undisturbed sampling.

2,8.4.2 Hydraulic Conductivity Testing

       Hydraulic conductivity tests are performed utilizing a flexible wall permeameter and the
procedures described in ASTM D-5084.  Inspectors should H careful to make sure that the
effective confining stress utilized in the hydraulic conductivity test is not excessive. Application of
excessive confining stress can produce an artificially low hydraulic conductivity.  The CQA plan
should prescribe the maximum effective  confining stress that will be used; if none is specified a
value of 35 kPa (5 psi) is recommended for both liner and cover systems.

2.8.4.3 Frequency of Testing

       Hydraulic conductivity tests are typically performed at a frequency of 3 tests/ha/lift (1
test/acre/lift) or, for very  thick liners  (S 1.2 m or 4 ft) per  every  other lift.   This is the
recommended frequency of testing, if hydraulic conductivity testing is required.  The CQA plan
should stipulate the frequency of testing.

2.8.4.4  Outliers

       The results of the above-described hydraulic conductivity tests are often given far too much
weight A passing rate of 100% does not necessarily prove that the liner was well built, yet some
inexperienced individuals falsely believe this to be the case.  Hydraulic conductivity tests are
performed on small samples; even though small samples may have low hydraulic conductivity,
inadequate construction or CQA can  leave remnant macro-scale defects such as fissures and
pockets of poorly compacted soil.  The fundamental problem is that laboratory hydraulic
conductivity tests are usually performed on 75-mm (3 in.) diameter samples, and these samples are
too small to contain a representative distribution of macro-scale defects (if any such defects are
present). By the same token, an occasional failing test does not necessarily prove that a problem
exists.  An occasional failing test only shows that either (1) there are occasional zones that fail to
meet performance criteria,  or (2) sampling disturbance (e.g., from the sampling tube shearing
stones in the soil) makes confirmation of low hydraulic conductivity difficult or impossible. Soil
liners built of multiple lifts are expected to have occasional, isolated imperfections — this is why the
liners are constructed from multiple lifts.  Thus, occasional failing hydraulic conductivity tests by
themselves do not mean very much. Even on the best built liners, occasional failing test results
should be anticipated.

       It is recommended that a multiple-lift soil liner be considered acceptable even if a small
percentage (approximately  5%) of the hydraulic conductivity tests fail.  However, one should
allow a small percentage of hydraulic conductivity  failures only if the overall CQA program is


                                           84
 i
 :•!
m
 vf

-------
thorough. Further, it is recommended that failing samples have a hydraulic conductivity that is no
greater than one-half to one order of magnitude above the target maximum value.  If the hydraulic
conductivity at a particular point is more than one-half to one order of magnitude too high, the zone
should be retestcd or repaired regardless of how isolated it is.

2.8.5  Repair of Holes from Sampling and Testing

       A number of tests,  e.g., from nuclear density tests  and  sampling  for  hydraulic
conductivity, require that a penetration be made into a lift of compacted soil.  It is  extremely
important that all penetrations he repaired. The recommended procedure for repair is as follows.
The backfill material should first be selected.  Backfill may consist of the soil liner material itself.
granular or pelletized bcntonite. or a mixture of bemonite and soil liner material. The backfill
material should be placed in the hole requiring repair with a loose lift thickness not exceeding about
SO mm (2  in.).   The loose lift of  soil should be tamped several times with a steel rod or other
suitable device that compacts the backfill and ensures no bridging of material that would leave large
aii pockets. Next, a new lift of backfill should be placed and compacted.  The process is repeated
until the hole has been filled.

       Because  it is critical  that  holes be properly repaired, it is recommended that periodic
inspections and  written  records made of the repair of holes.  It is suggested that approximately
20% of ail the repairs be inspected and thu the backfill procedures be documented for these
inspections. It is recommended that the inspector of repair of holes not be the same person who
backfilled the hole.

2.8.6  FmaJ Lift Thickness

       Construction documents may place restrictions on the maximum allowable final (after-
compaction) lift  thickness. Typically, the maximum thickness is 150 mm (6 in.).  Final elevation
surveys should be used to establish thicknesses of completed earthwork segments. The specified
maximum lift thickness is a nominal value. The actual value may be determined by surveys on the
surface of each completed lift, but an acceptable practice (provided there is good CQA on loose lift
thickness) is to survey the liner after construction and calculate the average thickness of each lift by
dividing the total thickness by  the number of lifts.

       Tolerances should be specified on final lift thickness. Occasional outliers from these
tolerances are not detrimental to  the  performance of a multi-lift liner.  It is recommended by
analogy to Table 2.9 that no  more than 5% of the final lift thickness determinations be out of
specification and that no out-of-specification thickness be more than 25 mm (1 in.) more than the
maximum allowable lift thickness.

2.8.7  Pass/Fail Decision

       After all  CQA tests have been  performed, a pass/fail decision must be made. Procedures
for dealing with  materials problems were discussed in Section 2.7.2.4. Procedures for correcting
deficiencies in compaction of the soil were addressed in Section 2.8.3.5. A final pass/Tail decision
is made by the CQA engineer based upon all the data and test results. The hydraulic conductivity
test results may  not be available for several days after construction of a lift has been completed.
Sometimes the contractor proceeds at risk with placement of additional lifts before all test results
are available. On occasion, construction of a liner proceeds without final results from a test pad on
the assumption that results will be acceptable.  If a "fail" decision is made at this late stage, the
defective soil plus any overlying materials that have been placed should be removed and replaced.
                                            85

-------
2.9    |»roteciion of Compacted Soil

2.9.1  Desiccation

2.9,1.1  Preventive Measures

       There are several ways to prevent compacted soil liner materials from desiccating. The soil
nay be smooth roiled with a steel drummed roller to produce a thin, dense skin of soil on the
surface.  This thin skin of very dense soil helps to minimize transfer of water into cr out of the
underlying material. However, the smooth-rolled surface should be scarified prior to placement of
• new lift of soil.

       A far better preventive measure is to water the soil periodically.  Care must be taken to
deliver water uniformly to the soil and not to create zones of excessively wet soil.  Adding water
by hand is not recommended because water is not delivered uniformly to the soil.

       An alternative preventive measure is to cover the soil temporarily with a gcomcmbrane,
moist geotextile, or moist soil. The geomembrane or gcotextile should be weighted down with
sand bags or other materials to prevent transfer of air between the geosyntheric cover and soil. If a
geomembrane is used, care should be taken to ensure that the underlying soil does  not become
netted and desiccate; a light-colored geomembrane may be needed to prevent overheating. If moist
soil is placed over the soil liner, the moist soil is removed using grading equipment.

2.9.1.2  Observations

       Visual observation is the best way to ensure that appropriate preventive measures have been
taken to minimize desiccation. Inspectors should realize that soil liner materials can dry put very
quickly (sometimes in a matter of just a few hours).  Inspectors should be aware that drying may
occur over weekends and provisions should be made to provide appropriate observations.

2.9,1.3  Tests

       If there are questions about degree of desiccation, tests should be performed to determine
the water content of the soil.  A decrease in water content of one to two percentage points is not
considered particularly serious and is within the general accuracy of testing. However, larger
reductions in water content provide clear evidence that desiccation has taken place.

2.9.1,4  Corrective Action

       If soil has been desiccated to a depth less than or equal to the thickness of a single lift, the
desiccated lift may be disked, moistened, and recompacted.  However, disking may produce large,
hard clods of clay that will require pulverization. Also, it should be recognized that  if the soil is
wetted, time must be allowed for water to be absorbed into the clods of clay and hydration to take
place uniformly.  For this reason it may be necessary to  remove the  desiccated soil from the
construction  area, to process the lift in a  separate  processing area, and to replace the soil
accordingly.

2.9.2  Freezing Temperatuiej

2.9.2.1  Compacting Frozen Soil

       Frozen soil should never be used to construct  soil liners.  Frozen soils form hard pieces

-------
that cannot be properly remolded and compacted. Inspectors should be on the lookout for frozen
chunks of soil when construction takes place in freezing temperatures.

2.9.2.2  Protection After Freezing

       Freezing of soil liner materials can produce significant increases in hydraulic conductivity.
Soil liners must be protected from freezing before and after construction.  If superficial freezing
t&kes piace on the surface of a lift of soil, the surface may be scarified and recompacted. If an
entire lift has been frozen, the entire lift should be disked, pulverized, and recompacted. If the soil
is frozen to a depth greater than one lift, it may be necessary to strip away and replace the frozen
material.

2.9.2.3  Investigating Possible Frost Damage

       Inspectors usually cannot determine from an examination of the surface the depth to which
freezing took place in a completed or partially completed soil liner that has  been exposed to
freezing.  In such cases it may be necessary to investigate the soil liner material for possible frost
damage.  The extent of damage is difficult  to determine.   Freezing temperatures cause the
development of tiny microcracks in the soil.  Soils that have been damaged due to frost action
develop fine cracks that lead to the formation of chunks of soil  when the soil is excavated. The
pushing of a sampling tube into the soil will probably close these cracks and mask the damaging
effects of frost upon hydraulic conductivity. The recommended procedure for evaluating possible
frost damage to soil liners involves three steps:

       1.     Measure the water content of the soil within and beneath the zone of suspected frost
              damage. Density may also be measured, but freeze/thaw has littic effect on density
              and may actually cause an increase in dry unit weight.   Freeze/thaw  is often
              accompanied by desiccation; water content measurements will help  to determine
              whether drying has taken place.

       2.     Investigate the morphology of the soil by digging into the  soil and examining its
              condition. Soil damaged by freezing usually contains hairline cracks, and the  soil
              breaks apart in chunks along larger cracks caused by freeze/thaw. Soil that has not
              been froi.cn should not have tiny cracks nor should it break apart in small chunks.
              The morphology of the soil should be examined by excavating a small pit into the
              soil liner and peeling off sections from the wall of the pit. One should not attempt
              to cut pieces from the side wall; smeared scil will mask cracks.  A distinct depth
              may be obvious; above this depth the soil breaks into chunks along frost-induced
              cracks, and below this depth there is no evidence of cracks produced by freezing.

       3.     One or  more samples of  soil  should be carefully  hand trimmed for hydraulic
              conductivity testing. The soil is usually trimmed with the aid of a sharpened section
              of tube of the appropriate inside diameter. The tube is set on the soii surface with
              the sharpened end facing downward, soil is trimmed away near the sharpened edge
              of the trimming ring, the tube is pushed a  few millimeters into the  soil, and the
              aimming is repeated. Samples may be taken a: several depths to delineate the depth
              to which freeze/thaw damage occurred. The minimum diameter of a cylindrical test
              specimen should be 300 mm (12 in.). Small test specimens, e.g., 75 mm (3 in.)
              diameter  specimens, should not be used because freeze/thaw can create
              morphological structure in the soil on a scale too large to permit representative
              testing with small samples. Hydraulic conductivity tests should be performed as
              described in ASTM D-5084. The effective confining stress should not exceed the
                                            87

-------
              smallest vertical effective stress to which the soil will be subjected in the field,
              which is usually the stress at the beginning of service for liners. If no compressive
              stress is specified, a  value of 35 kPa (5 psi) is recommended for both liner and
              cover system.

       The test pit and all other penetrations should be carefully backfilled by placing soil in lifts
and compacting the lifts. The sides of the test pit should  be sloped so that the compactor can
penetrate through to newly placed material without interference from the vails of the pit

2.9.2.4  Repair

       If it is determined that soil has been damaged by freezing, the damaged material is usually
repaired as follows. If damage is restricted to a single lift, the lift may be disked, processed to
adjust water content or to reduce clod size if necessary, and recompacted. If the damage extends
deeper, damaged materials should be excavated and replaced.

2.9.3  Excess Surface Water

       In some cases exposed lifts of liner material, or the completed liner, are subjected to heavy
rains that soften the soil.  Surface water creates a problem if the surface is uneven (e.g., if a footed
roller has been used and the  surface has not been smooth-rolled with a smooth, steel wheeled
roller) —  numerous small puddles of water will develop in the depressions low areas.  Puddles of
water should be removed before further lifts of material, or other components of the liner or cover
system, are constructed.  The material should be disked repeatedly to allow  the soil to dry, and
when the soil is at the proper water content, the soil should be compacted. Alternatively, the wet
soil may be removed and replaced.

       Even if puddles  have net formed, the soils may be too soft  to permit  construction
equipment to operate on the soil without  creating ruts. To deal with this problem, the soil may  be
allowed to dry slightly by natural processes (but care must be taken to ensure that it does not dry
too much and does not crack excessively during the drying process). Alternatively, the soil may  be
disked, allowed to dry while it is periodically disked, and then compacted.

       If soil  is reworked and recompacted,  QA/QC tests should  be  performed at the same
frequency as for the rest  of the project.  However, if the area requiring reworking is very small,
e.g., in a sump, tests should be performed in the confined area to confirm proper compaction even
if this requires sampling at a greater frequency.

2.10   Test Pads

2.10.1   Purpose of Test Pads

       The purpose of a test pad is to  verify that the materials and methods of construction
proposed for a project will lead to  a soil liner with the required large-scale, in-situ, hydraulic
conductivity. Unfortunately, it is impractical to perform large-scale hydraulic conductivity tests on
the actual soil liner for two reasons:  (1) the testing would produce significant physical damage to
the liner, and the repair of the damage would be questionable; and (2) the time required to complete
the testing would be too long - the liner could become damaged due to desiccation while one
waited for the test results.

       A test  pad may also be used to demonstrate that unusual materials  or  construction
procedures will work. The process of constructing and testing a test pad is usually a good learning
                                           88

-------
experience for the contractor and CQC/CQA personnel;  overall quality of a project is usually
elevated as a result of building and testing the test pad.

       A test pad is constructed with the soil liner materials proposed for a project utilizing
preprocessing procedures, construction equipment, and construction practices that are proposed for
the actual liner.  If the required hydraulic conductivity is demonstrated for the test pad,  it is
assumed that the actual liner will have a similar hydraulic conductivity, provic'-d the actual liner is
built of similar materials and to standards that equal or exceed those used in building the test pad.
If a test pad is constructed and hydraulic conductivity is verified on the test pad, a key goal of
CQA/CQC for the actual liner is to verify that the actual liner is built of similar materials and to
standards that equal or exceed those used in building the test pad.

2.10.2  Dimensions

      Test pads (Fig. 2.31) normally measure about 10 to 15 m in width by 15 to 30 m in length.
The width of the test pad is typically at least four times the width of the compaction equipment, and
the length  must be adequate for the compactor to reach normal operating speed in the test area.  The
thickness of a  test pad is usually no less than the thickness of the soil liner proposed for a facility
but may be as  little as 0.6 to 0.9 m (2 to 3 feet) if thicker liners are to be employed at full scale. A
freely draining material such as sand is often placed  beneath the test pad to provide a known
boundary condition in case infiltrating water from a surface hydraulic conductivity test (e.g., sealed
double ring infiltrometcr) reaches the base of the liner.  The drainage layer may be drained with a
pipe or other means.  However, infiltrating water will not reach the drainage layer if the hydraulic
conductivity is very low; the drainage pipe would only convey water if the hydraulic conductivity
turns out  to be very large. The sand drainage material may not provide adequate foundation
support for the first lift of soil liner unless the sand is compacted sufficiently.  Also, the first lift of
soil liner material on the drainage layer is often viewed as a sacrificial lift and is only compared
nominally to avoid mixing clayey soil in with the drainage material.

2.10.3  Materials

      The test pad is constructed of the same materials that are proposed for the actual project.
Processing equipment and procedures should be identical, too. The same types of CQC/CQA tests
that will be used for the soil liner are performed on the test pad materials. If more than one type of
material will be used, one test pad should be  constructed for each type of material.

2.10.4  Construction

      It is recommended that test strips be built before constructing the test pad. Test strips allow
for the detection of obvious problems and provide an opportunity uvfine-tune soil specifications,
equipment selection, and procedures so that problems are minimized and the probability of the
required hydraulic conductivity being achieved in the test pad  is maximized.  Test strips are
typically two lifts thick, one and a half to two equipment widths wide, and about 10m (30 ft) long.

      The test  pad is built using the same loose  lift thickness, type of compactor, weight of
compactor, operating speed, and minimum number  of passes that are proposed for the actual soil
liner. It is important that the test pad not be built to standards that will exceed those used in
building the actual liner. For example, if the test pad is subjected to 15  passes of the compactor,
one would want the actual soil liner to be subjected to at least 15 passes  as well. It is critical that
CQA personnel document the construction practices that are employed in building the test pad.  It is
best if the  same contractor builds the test p?d and actual liner so that experience gained from the test
pad process is  not lost.  The same a*    -    CQC and CQA personnel.
                                            89

-------
                                                      Drainagt l&tiwtai
                         W « 3 Compaction V«Meta W» tht. Minimum
                         L • A Vahi* NO Smaller than W and Suffieimt tor Equipment
                            Hi Raacft Prop«r Operating Spted In T«$l Ar«
                   Figure 2.31 - Schematic Diagram of Soil Liner Test Pad
       The test pad must be protected from desiccation, freezing, and erosion in the area where hi
situ hydraulic conductivity testing is planned. The recommended procedure is to cover the ttst pad
with a sheet of white or clear plastic and then either spread a thin layer of soil on the plastic if no
rain is anticipated or, if rain may create an undesirably muddy surface, cover the plastic with hay or
straw.

-------
2,10.6  Tests and
       The same types of CQA tests that are planned for the actual liner are usually performed on
the test pad.  However, the frequency of testing is usually somewhat greater for the test pad.
Material tests such as liquid limit, plastic limit, and percent fines are often performed at the rate of
one per lift  Several water content-density tests are usually performed per lift on the compacted
soil A typical rate of testing would be one water content-density test for each 40 m2 (4^0 ft2 ).
The CQA plan should describe the testing frequency for the test pad.

       There is a danger in over testing the test pad -- excessive testing could lead to a greater
degree of construction control in the test pad than in the actual liner. The purpose of the test pad is
10 verify that the materials and methods of construction proposed for  a project can result in
compliance with performance objectives concerning hydraulic conductivity.  Too much control
over the construction of the test pad runs counter to this objective.

2.10.7   In Situ Hydraulic Conductivity

2.10.7.1  Sealed Ootibj^-Ring Infiltrometer

       The most common method of measuring in situ hydraulic conductivity on test pads is the
sealed double-ring infiltrometer (SDRI). A schematic diagram of the SDR1 is shown Fig. 2.32.
The test procedure is described in ASTM D-5093.
           Tensiometer
                           Inner Ring
                                                                 Rexible Bag
              Gro
                                      rainage Layer
         Figure 2.32 - Schematic Diagram of Sealed Doable Ring Infiltrometer (SDRI)
       With this method, the quar, tity of water that flows into the test pad over a known penod of
time is measured. This flow rate, which is called the infiltration rate (I), is computed as follows:

                                                                                   (2.5)
                                           91

-------
where Q is the quantity of water entering the surface of the soil through A cross-sectional area A
and over a period of time t.

       Hydraulic conductivity (K) is computed from the infiltration rate and hydraulic gradient (i)
as follows:

                                 K»W                                         (2.6)

       Three procedures have been used to compute the hydraulic gradient. The procedures are
called (1) apparent gradient method; (2) wetting front method; and (3) suction head method. The
equation for computing hydraulic gradient from each method Is shown in Fig. 2.33.
                       Apparent Hydradlc Conductivity Method
                                                              H+D
                              Suction H*ad Method
                                                            ,, H + D
                              Wetting Front Method
                                ,„  H t.O.
                                      o
     Figure 2.33 - Three Procedures for Computing Hydraulic Gradient from Infiltration Test


                                          92

-------
       The apparent gradient method is the most conservative of the three methods because this
method yields the lowest estimate of i and, therefore, the highest estimate of hydraulic
conductivity. The apparent gradient method assumes that the test pad is fully soaked with water
over the entire depth of the test pad. For relatively permeable test pads, the assumption of full
soaking is reasonable, but for soil liners with K < 1 x 10*7 cm/s, the assumption of full soaking is
excessively conservative and should not be used unless verified.

       The second and most widely used method is the wetting front method.  The wetting front is
assumed to partly penetrate the test pad (Fig. 2.33) and the water pressure at the wetting front is
conservatively assumed to equal atmospheric pressure. Tensiometers are used to monitor the depth
of wetting of the soil over time, and the variation of water content with depth  is determined at the
end of the test. The wetting front method is conservative but in most cases not excessively so.
The wetting front method is the method that is usually recommended.

       The third method, called the suction head method, is the same as the wetting front method
except that the water pressure at the wetting front is not assumed to be atmospheric pressure.  The
suction head (which is defined as the negative of the pressure head) at the wetting front is Hs and is
added to the static head of water in the infiltration ring to calculate hydraulic gradient (Fig. 2.37).
The suction head Hs is identical to the wetting front suction  head employed in analyzing water
infiltration with the Green-Ampt theory. The suction head Hs is rjoj the ambient suction head in the
unsaturated soil and is generally very difficult to determine (Brakensiek, 1977). Two techniques
available for determining Hs are:
                                                                                    (2.7)
1.      Integration of the hydraulic conductivity function (Neuman,  1976):


          Hs=|    Kr*s

               Jhsc

       where hx is the suction head at the initial (presoaked) water content of the soil, Kr
       is the relative hydraulic conductivity (K at particular suction divided by the value of
       K at full saturation), and hs is suction.

       Direct measurement  with air  entry permeameter (Daniel, 1989, and references
       therein).
       2.
              therein).
Reimbold (1988) found that Hs was close to zero for two compacted soil liner materials. Because
proper determination of Hs is very difficult, the suction head method cannot be recommended,
unless the testing personnel take the time and make the effort to determine Hs properly and reliably.
                                                                >•
       Corrections may be made to account for various factors.  For example, if the soil swells,
some of the water that infiltrated into the soil was absorbed into the expanded soil. No consensus
exists on various corrections and these should be evaluated case by case.

2.10.7.2  Two-Stage Borehole Test

       The two-stage borehole hydraulic conductivity was developed by Boutwell (the test is
sometimes called the Boutwell Test) and was under development as an ASTM standard at the time
of this writing.  The device is installed by drilling a hole (which is typically 100 to 150 mm in
diameter), placing a casing in the hole, and sealing the annular space between the casing and
borehole w-th grout a« ahown in Fig. 2.34. A scries of falling head tests is performed  and the
                                           93

-------
hydraulic conductivity from this first stage (ki) is computed. Stage one is complete when ki
ceases to change significantly. The maximum vertical hydraulic conductivity may be computed by
assuming that the vertical hydraulic conductivity is equal to lq.  However, the test may be
continued for a second stage by removing the top of the casing and extending the hole below the
casing as shown  in Fig. 2.34. The casing is reassembled, the device is again filled with water, and
failing head tests are performed to determine the hydraulic conductivity from stage two (k2).  Both
horizontal and vertical hydraulic conductivity may be computed from the values of ki and k2-
Further details on methods of calculation are provided by Boutwell and Tsai (1992), although the
reader is advisee to refer to the ASTM standard when it becomes available.
                A. Stage
S. Stage II
                                                                 lili
                                                                             ^

                Figure 2.34 - Schematic Diagram of Two-Stage Borehole Test
       The two-stage borehole test permeates a smaller volume of soil than the sealed double-ring
infiltrometer. The required number of two-stage borehole tests for a test pad is a subject of current
research.  At the present time, it is recommended that at least 5 two-stage borehole tests be
performed on a test pad if the two-stage test is used. If 5 two-stage borehole tests are performed,
then one would expect that all five of the measured vertical hydraulic conductivities would be less
than or equal to the required maximum hydraulic conductivity for the soil liner.
                                         94

-------
2.10.7.3   Other Field Tests

       Several other methods of in situ hydraulic conductivity testing are available for soil liners.
These methods include open infiltrometers, borehole tests with a constant water level in the
borehole, porous probes, and air-entry permeameters.  The methods are described by Daniel
(1989) but are much less commonly used than the SDRI and two-stage borehole test

2.10.7.4   Laboratory Tests

       Laboratory hydraulic conductivity tests may be performed for two reasons:

    1.  If a very large sample of soil is taken from the field and permeated in the laboratory, the
        result may be representative of field-scale hydraulic conductivity. The question of how
        large the  laboratory test specimen needs to be is currently a matter of research, but
        preliminary results indicate that a specimen with a diameter of approximately 300 mm (12
        in.) may be sufficiently large (Benson et al., 1993).

    2.  If laboratory hydraulic conductivity tests are a required component of QA/QC for the
        actual liner, the same sampling  and  testing procedures  are used  for the test pad.
        Normally, undisturbed soil samples are obtained following the procedures outlined in
        ASTM D-1587, and soil test specimens with  diameters of approximately 75 mm (3 in.)
        are permeated in flexible-wail permeameters in accordance with ASTM D-5084.

2.10.8  Documentation

       A report should be prepared that describes ail of the test results from the test pad. The test
pad documentation provides a basis for comparison between test pad results and the CQA data
developed en an actual construction project.

2.11   Final Approval

       Upon completion of the soil liner, the soil liner should be accepted and approved by the
CQA engineer prior to deployment or construction of the next overlying layer.
2.12   Referer.ces

Aibrecht, K .A., and K. Cartwright (1989), "Infiltration and Hydraulic Conductivity of a
    Compacted Earthen Liner," Ground Water, Vol. 27,  No. 1, pp. 14-19L_

Alther, G. R. (1983), The Methylene Blue Test for Bentonite Liner Quality Control,"
    Gcotechnical Testing Journal, Vol. 6, No. 3, pp. 133-143.

ASTM D-422, "Particle-Size Analysis of Soils"

ASTM D-698, "Laboratory Compaction Characteristics of Soils Using Modified Effort (12,400 ft-
    lbf/ft3(600kN-m/m3))"

ASTM D-l 140, "Amount of Material in Soils Finer than the No. 200 (75-|om)Sieve"
                                          95

-------
ASTM D-1556, "Density and Unit Weight of Soil In Place by Sand-Cone Method"
ASTM D-1557, "Laboratory Compaction Characteristics of Soils Using Standard Effort (56,000
   ft-lbf/ft3 (2,700 kN-nVm3))"
ASTM D-1587, Thin-Walled Tube Sampling of Soils"
ASTM D-2167, "Density ai i Unit Weight of Soil Li Place by Rubber Balloon Method*
ASTM D-2216, 'Laboratory Determination of Water (Moisture) Content of Soil and Rock"
ASTM D-2487. "Classification of Soils for Engineering Purposes (Unified Soil Classification
   System)"
ASTM D-24SS, "Description and Identification of Soils (Visual-Manual Procedure)"
ASTM D-2922. "Density of Soil and Soil-Aggregate In Place by Nuclear Methods (Shallow
   Depth)
ASTM D-2937, "Density and Unit Weight of Soil In Place by Drive-Cylinder Method"
ASTM D-3017, "Water Content of Soil and Rock In Place by Nuclear Methods (Shallow Depth)"
ASTM D-4220, "Preserving and Transporting Soil Samples"
ASTM D-4318, "Liquid Limit, Plastic Limit, and Plasticity Index of Soils"
ASTM D-4643. "Determination of Water (Msikure) Content of Soil by Microwave Oven Method"
ASTM D-4944, "Held Determination of Water (Moisture) Content of Soil by Calcium Carbide Gas
   Pressure Tester Method"
ASTM D-4959, "Determination of Water (Moisture) Content of Soil by Direct Heating Method"
ASTM D-5080, "Rapid Determination of Percent Compaction"
ASTM D-5084, "Measurement of Hydraulic Conductivity of Saturated Porous Materials Using a
   Flexible Wall Permeameter"
ASTM D-5093, "Reid Measurement of Infiltration Rate Using a Double-Ring"kifiltrometer with a
   Sealed Irmer Ring"
ASTM E-946, "Water Adsorption of Bentonite by the Porous Hate Method*
Benson, C.  H., and  D,E. Daniel (1990), "Influence of Gods on Hydraulic Conductivity of
   Compacted Gay," Journal of Geotechnical Engineering, Vol. 116, No. 8, pp. 1231-1248.
Benson, C H., and G. P. Boutweil (1992), "Compaction Control and Scale-Dependent Hydraulic
   Conductivity of Clay Liners," Proceedings, Fifteenth Annual Madison Waste Conference,
   Univ. of Wisconsin, Madison, Wisconsin, pp. 62-83..
Benson, C.  H., Zhai. H.,  and Rashad, S. M.  (1992), "Assessment of Construction Quality
                                        96

-------
   Control Measurements and Sampling Frequencies for Compacted Soil Liners," Environmental
   Geotechnics  Report No.  92-6, University of Wisconsin,  Department  of Civil  and
   Environmental Engineering, Madison, Wisconsin, lOOp.

Benson, C.  H., Hardianto, F. S., and Motan, E. S. (1993), "Representative Sample Size for
   Hydraulic Conductivity Assessment of Compacted Soil Liners," Hydraulic Conductivity and
   Waste Contaminant Transport in Soils, ASTM STP 1142, D. E. Daniel and S. J. Trautwein
   (Eds.), American Society for Testing and Materials, PhiladeJohia, (in review).

Bourwell, G., and Hedges, C. (1989), "Evaluation of Waste-Retention Liners by Multivariate
   Statistics," Proceedings of the Twelfth International Conference on Soil Mechanics and
   Foundation Engineering, A. A. Balkema, Rotterdam, Vol. 2, pp. 815-818.1

Boutwell, G .P., and Tsai, C. N. (1992), "The Two-Stage Field  Permeability Test for Clay
   Liners," Geotechnical News, Vol. 10, No. 2, pp. 32-34.

Brakensiek, D.L. (1977), "Estimating the Effective  Capillary Pressure in the Green and Ampt
   Infiltration Equation," Water Resources Research, Vol. 12, No. 3, pp. 680-681.

Brown, R. K. (1992), Personal Communication.

Daniel, L.E. (1989), "In Situ Hydraulic Conductivity Tests for Compacted Clay," Journal of
   Geotechnical Engineering, Vol. 115, No. 9, pp. 1205-1226.

Daniel, D.E. (1990), "Summary  Review of Construction Quality Control for Compacted  Soil
   Liners," in Waste Containment Systems:  Construction, Regulation,  and Performance, R.
   Bonaparte (Ed.), American Society of Civil Engineers, New York, pp. 175-189.

Daniel, D.E., and C. H. Benson (1990), "Water Content - Density  Criteria for Compacted Soil
   Liners," Journal of Geotechnical Engineering, Vol. 116, No.  12, pp. 1811-1830.

Elsbury, B. R., Daniel, D.  E., Sraders, G. A., and Anderson, D. C. (1990), "Lessons Learned
   from Compacted Clay Liner," Journal of Geotechnical Engineering,  Vol. 116, No. 11, pp.
   1641-1660.

Hilf, J. W. (1991), "Compacted Fill," in Foundation Engineering Handbook^ H. Y. Fang (Ed.),
   Van Nostrand Reinhold, New York, pp. 249-316.

Lambe, T. W., and Whitman, R. V. (1969), Soil Mechanics, John Wiley & Sons, New  York, 533
   P-

Mitchell, J. K.,  Hooper, D. R., and Campanella, R. G. (1965), "Permeability of Compacted
   Clay," iournal of the Soil Mechanics and Foundations Division, ASCE, Vol. 91,  No. SM4,
   pp. 41-65.

\fundell, J. A., and Bailey *\ (1985), "The Design and Testing of a Compacted Clay Barrier
   Layer to Limit Percolation through Landfill Covers," Hydraulic Barriers in Soil and Rock,
   ASTM STP 874, A. I. Johnson et al. (Eds.), American Society for Testing and  Materials,
   Philadelphia, pp. 246-262.

Neuman, S.P. (1976), "Wetting Front Pressure Head in the Infiltration  Model of Green and
   Ampt," Water Resources Research, Vol. 11, No.  3, pp. 564-565.

                                          97

-------
Noorany, I. (1990), "Variability in Compaction Control," Journal of Geotechnical Engineering,
   Vol.  116, No. 7, pp. 1132-1136.

Rcimbold, M.W. (1988), "An Evaluation of Models for Predicting Infiltration Rates in
   Unsatunued Compacted day Soils," MS. Thesis, The University of Texas at Austin, Austin,
   Texas, 128 p.

Richardson. G. N. (1992), "Constractk - Quality Management for Remedial Action and Remedial
   Design Waste Containment Systems," U.S. Environmental Protection Agency, EPA/S40/R-
   92/073, Washington, DC

Schmcrtmann, I.R (1989), "Density Tests Above Zero Air Voids Line," Journal of Geotechnical
   Engineering, Vol. 115, No. 7, pp. 1003-1018.

Shelley, T. L., and Daniel, D. E. (1993), "Effect of Gravel on Hydraulic Conductivity of
   Compacted Soil Liners," Journal of Geotechnical Engineering, Vol. 119, No. 1, pp. 54-68.

Turnbull, W. J., Compton, J, P., and R. G.  Ahlvin (1966),  "Quality Control of Compacted
   Earthwork," Journal of the Soil Mechanics and Foundations Division* ASCE, Vol. 92, No.
   SMI, pp. 93-103.

U.S. Army Corps of Engineers  (1970), "Laboratory Soils Testing," Office of the Chief of
   Engineers, Washington, DC, EMI 110-2-1906.
                                        98

-------
                                      Chapter 3

                                   Geomembranes
      This  chapter focuses upon the manufacturing quality assurance (MQA) aspects of
geomembrane formulation, manufacture and fabrication, and on the construction quality assurance
(CQA) of the complete installation of the gcomembranes in the field.  Note that in previous
literature these liner materials were called flexible membrane liners (FML's), but the more generic
name of gcomembranes will be used throughout this document.

      The geomembrane materials discussed in this document are those used most often at the
time of writing. However, there are other polymer types that are also used. Aspects of quality
assurance of these materials can be inferred from information contained in this document. In the
future, new materials  will be developed and the reader is  advised to seek the appropriate
information for evaluation of such new or modified materials.

3.1 Types of Geomembranes and Their Formulations

      It must be recognized that all geomembranes are actually formulations of a parent resin
(from which they derive their generic  name) and several other ingredients. The most commonly
used geomembranes for solid and  liquid waste containment are listed below.  They are listed
according to their commonly referenced acronyms which will be explained in the text to follow.
Other geomembranes  in limited use or under initial field trials will also be mentioned  where
appropriate but will be covered in less detail than the types listed below.


Table 3.1 - Types of Commonly Used Geomembranes and Their Approximate Weight Percentage
          Formulations*
Geomembram
TVpe
HOPE
VLDPE
Other Extruded Types •*
PVC
CSPE***
Other Calendered Tvpes*«
Resin
95-98
94-%
95-98
50-70
4(WO
40-97
Plastic izer
0
0
0
25.35
0
0-30
Filler Carbon Black
or Piemem
0
0
0
0-10
40-50
0-50
2-3
2-3
2-3
2-5
S40
2-30
Additives
025-1.0
1-4
1-2
2-5
5-15
0-7
*   Note that this Table should not be directly used for MQA or CQA Documents, since neither the Agency nor
    the Authors of the Report intend to provide prescriptive formulations for manufacturers and their respective
    geomembranes.
**  Other geomembranes than those listed in this Table will be described in the appropriate Section.
*** CSPE geomcjnb.ancs are generally fabric (scnrr.) reinforced.
                                           99

-------
      It must be recognized thai Table 3.1 and the references to it in the text to follow are meant to
reflect on the current state-of-the-art. The values mentioned are not meant to be prescriptive and
future research and development may result in substantial changes.

3.1.1 High Densir^Pjriyejhyleng fHDPE.)

      As noted in Table 3.1, high density polyethylene (HDPE) geomembranes arc made from
polyethylene resin, carbon black and aadin'ves.

3.1.1.1  %es|n

      The polyethylene resin used for HDPE geomembranes  is prepared by low  pressure
polymerization of ethylene as the principal monomer and having the characteristics listed in ASTM
D-1248. As seen in Fig. 3.1, the resin is usually supplied to the manufacturer or formulator in an
opaque pellet form.
                                Poly»thylttn»  Pallets
                            Figure 3.1 - HDPE Resin Pellets

      Regarding the preparation of a specification or MQA document for the resin component of
an HDPE geomembranc, the following items should be considered:

       1.  The polyethylene resin, which is covered in ASTM D-1248, is to be made from virgin,
          uncontaminated ingredients.

       2.  The quality control tests performed on the incoming resin will typically be density
          (either ASTM D-792 or D1505) and melt flow index which is ASTM D-1238.
                                         100

-------
      3 .  Typical natural densities of the various resins used are between 0.934 and 0,940 g/cc.
          Note that according to ASTM D-1248 this is Type II polyethylene and is classified as
          medium density polyethylene.

      4.  Typical melt flow index values are between 0.1 and 1.0 g/10 min as per ASTM D-
           1238, Cond. 190/2.16.

      S .  Other tests which can be considered for quality control of the resin ere melt flow ratio
          (comparing high-to-low weight melt flow values), notched constant tensile load test as
          per ASTM D-5397, and a single point notched constant load test, see Hsuan and
          Koerner (1992) for details. TV Utter tests would require a plaque to be made from the
          resin from which test specimens a>c taken. The single point notched constant load test
          is then performed at 30%  yie'J strength  and the  test specimens are currently
          recommended not to fail within 200 hours.

      6.  Additional quality control certification procedures by the manufacturer (if any) should
          be implemented and followed.

      7.  The frequency of performing each of the preceding tests should be covered in the
          MQC plan and it should be implemented and followed.

      8.  An HOPE geoinembrane formulation should consist of at least 97% of polyethylene
          resin.  As seen in Table 3.1  the balance is carbon black  and additives. No fillers,
          extenders, or other materials should be mixed into the formulation.

      9 .  It should be noted that by adding carbon black and additives to the resin* the density of
          the final formulation is generally 0.941 to 0.9S4 g/cc.  Sines this numeric value is now
          in the high density polyethylene category according to ASTM D-1248, geomcmbrancs
          of this type are commonly referred to as high density polyethylene (HOPE).

      10. Regrind or rework chips (which  have been previously processed by the same
          manufacturer but never used as a  geomembrane, or other) are often added to the
          extruder during processing. This topic will be discussed in section 3.2.2.

      1 1 .  Reclaimed material (which is  polymer material that has seen previous service life and is
           recycled) should never be allowed in the formulation in any quantity. This topic will
           be discussed in section 3.2.2.
3.1.1.2
      Carbon black is added into an HOPE geomembrane formulation for general stabilization
purposes, particularly for ultraviolet light stabilization. It is sometimes added in a powder form at
the geomembrane manufacturing facility during processing, or (generally) it is added as a
preformulated concentrate  in pellet  form.  The latter is the usual case. Figure 3.2 shows
photographs of carbon black powder and of concentrate pellets consisting of approximately 25%
carbon black in a polyethylene resin carrier.

      Regarding  the preparation of a specification or MQA document for the carbon black
component of HDPE geomembranes, the following items should be considered.

      1 . The carbon black used in HDPE geomembranes  should be a Group 3 category, or
         lower, as defined in ASTM D-1765.
                                          101

-------
                                                 *  *
Figure 3.2 - Carbon Black in Paniculate Form (Upper Photograph) and as a Concentrate (Lower
           Photograph)
                                         102

-------
      2. Typical amounts of carbon black are from 2.0% to 3.0% by weight per ASTM D-1603.
         Values less than 2.0% do not appear to give adequate long-term ultraviolet protection;
         values greater than 3.0% begin to adversely effect physical and mechanical properties,

      3. Current carbon black dispersion requirements in the final HOPE geomembrane are
         usually required to be A-1, A-2 or B-1 according to ASTM D-2663. Sample preparation
         is via ASTM D-3015. It should be  noted, however, that this test method is directed at
         polymeric materials containing relatively large amounts of carbon black, e.g., thermjset
         elastomers with carbon black contents of approximately 18% by volume. ASTM D-35
         Committee on Geosynthetics has a Task Group formulating a new standard focused at
         carbon black dispersion for formulations containing less than 5% carbon black. Thus
         this standard will be applicable for the 2 to 3% carbon black currently used in
         polyethylene formulations.

      4. In the event that the carbon black is mixed into the formulation in the form of a
         concentrate rather than a powder, the carrier resin of the concentrate should bt the same
         generic type as the base polyethylene resin.

3.1.1.3 Additives

      Additives are introduced into an HOPE geomembrane formulation for the purposes of
oxidation prevention,  long-term durability and as a lubricant and/or processing aid during
manufacturing.  It is quite difficult to write a specification for HDPE geomembranes around a
particular additive, or group of additives, because they arc generally proprietary.  Furthermore,
there is research and development ongoing in this area and thus additives are subject to change over
time.

      If additives are included in a specification or MQA document, the description must be very
general as to the type and amount However, the amount can probably be bracketed as to an upper
value,

      1. The nature of the additive package used in the HDPE compound may be requested of the
         manufacturer.

      2. The maximum amount of additives  in a particular formulation should not exceed 1.0%
         by weight.

3.1-2 Very Low Denjjry Polyethylene fVLDPEl
                                                             T*.
      As seen in Table 3.1, very low density polyethylene (VLDPE) geomembranes are made
from polyethylene resin, carbon black and additives.  It should be noted that there are similarities
between VLDPE and certain types of linear low density polyethylene (LLDPE).  The linear
structure and lack of long-chain branching in both LLDPE and  VLDPE arise from their similar
polymerization mechanisms although the catalyst technology is different.  In the low-pressure
polymerization of LLDPE, the random incorporation of alpha olefin comonomers produces
sufficient short-chain branching to yield densities in the range of 0.915 to 0.930 g/cc. The even
lower densities of VLDPE resins (from 0.890 to 0.912 g/cc) are achieved by  adding more
comonomer (which produces more short-chain branching than occurs  in LLDPE, and thus a lower
level of crystallinity) and using proprietary catalysts and reactor technology. Since VLDPE is more
commonly used than LLDPE for geomembranes in waste containment applications, this section is
written around VLDPE. It can be used for LLDPE if the density is  at the low end of the above
mentioned range. The situation is  under discussion by many groups as of the writing of this
                                         103

-------
document

3.1.2.1 Resin

      The polyethylene resin used for VLDPE geomembranes is a linear polymer of ethylene with
other alpha-olefins. As with HOPE, the resin is generally supplied to the manufacturer iii the form
of pellets, recall Fig. 3.1.

      Some specification or MQA document items for VLDPE resins follow:

      1. The very low density polyethylene resin is to be made from completely virgin materials.
         The natural density of the resin is less than 0,912 g/cc, however, a unique category is
         not yet designated by ASTM.

      2. A VLDPE geomembrane formulation should consist of approximately 94-96% polymer
         resin.  As seen in Table 3.1, the balance is carbon black and additives.

      3. Typical quality control tests for VLDPE resin will  be density, via ASTM D-792 or
         D1505, and melt flow index via ASTM D-1238.

      4. Additional quality control certification procedures of the manufacturer (if any) should be
         implemented and followed.

      5. The frequency of performing each of the preceding tests should be covered in  the MQC
         plan and it should be implemented and followed.

      6. Regrind or rework  chips (which have  been previously processed ty the same
         manufacturer but  never used  as a geomembrane, or other) are  often added to the
         formulation during processing.  This topic will be discussed in section 3.2.2.

      7. Reclaimed material (which is polymer that has seen previous service life and is recycled)
         should never be allowed in any quantity. This topic will be discussed in section 3.2.2.

3.1.2,2 Carbon  Black

      Carbon black is added  to VLDPE geomembrane formulations for general stabilization
purposes, particularly for ultraviolet  light stabilization.  It is added either in a powder form at the
geomembrane manufacturing facility, or it is added as a preformulated concentrate in pellet form,
recall Fig. 3.2.

      Some items to be included in a specification or MQA document follow:

      1. The carbon black  used  in VLDPE geomembranes should be a Group 3 category, or
         lower, as defined in ASTM D-1765.

      2. Typical amounts of carbon blai •• ?..« from 2.0% to 3.0% by weight as per ASTM D-
         1603.  Values less than 2.0% do not appear to give adequate long-term ultraviolet
         protection, while values greater than 3.0%  begin to negatively effect physical and
         mechanical properties.

      3. Current carbon black dispersion requirements in the final HDPE geomembrane are
         usually  required to  be A-l,  A-2 or  B-l according to ASTM D-2663W.  Sample

                                         104

-------
         preparation is via ASTM D-3015.  It should be noted, however, that this test method
         was directed at polymeric materials containing relatively large amounts of carbon black,
         e.g., thermoset elastomers with carbon black contents of approximately 18% by volume.
         ASTM D-35 Committee on Geosynthetics has a Task Group formulating a new standard
         focused at carbon black dispersion  for formulations containing less than 3%  carbon
         black which is the amount used in formulation of VLDPE geomembranes.

      4. in the  event that the carbon black is mixed into the formulation  in the form of a
         concentrate rather than ? powder, the carrier resin of the concentrate should be identified,

3.1.2.3 Additives

      Additives are introduced into a VLDPE formulation for the purposes of anti-oxidation,
long-term durability and as a lubricant and/or processing aid during manufacturing.  It is quite
difficult to write a specification for VLDPE geomembranes around a particular additive, or group
of additives, because they are generally  proprietary.  Furthermore,  there is research  and
development ongoing in this area and thus additives are subject to change over time.

      If additives were included  in a specification or MQA document, the description must be
very general as to the type and amount. However, the amount can probably be bracketed as to an
upper value.

      1. The nature of the additive package used in the VLDPE compound may  be requested of
         the manufacturer.

      2. The maximum amount of additives in a particular formulation should not exceed 2.0%
         for smooth sheet or 4.0% for textured sheet by weight.
3.1.3 Other Extruded Geomembranes

      Recently, there have been developed other variations of extruded geomembranes.
have seen commercialization and will be briefly mentioned.
Four
      One variation is a coextruded light colored surface layer onto a black base layer for the
purpose of reduced surface temperatures when the geomembrane is exposed for a long period of
time. The usual application for this material is as a liner for surface impoundments which have no
soil covering or sacrificial sheet covering. In the formulation of the light colored surface layer the
carbon black is replaced by a pigment (often metal oxides, such as titanium dioxide) which acts as
an ultraviolet screening agent.  This results in a white, or  other Hght colored surface.   The
coextruded surface layer is usually relatively thin, e.g., 5 to 10 percenrof the total geomembrane's
thickness.

      A second coextrusion variation is HDPE/VLDPE/HDPE sheet where the two surface layers
of HDPE are relatively thin with respect to the VLDPE core. Thickness percentages of 20/60/20
are sometimes used. The interface of these coextruded layers cannot be visually distinguished
since  the polymers merge into one another while they are  in  the molten  state, i.e., such
geomembranes are not laminated together after processing, but are coextruded during processing.

      A third variation of coextrusion is to add a foaming agent, such as nitrogen gas, into the
surface layer extruders). This foaming agent expands and bursts at the surface of the sheet as it
cools. The resulting surface is very rough and is generally referred to as textured. This variation
will be described in  Sections 3.2.3.4 and 3.2.4.4 for HDPE and VLDPE, respectively.


                                          105

-------
      A  fourth variation of extruded gepmembranes is a generic polymer group under the
classification of fully crosslinked elastomeric alloys (FCEA). This group of polymers is described
in ASTM D-5046.  The particular geomembrane type that has been used in waste containment
applications is a thermoplastic elastomeric alloy of polypropylene (PP) and ethylene-propylene
dime monomer (EPDM). The EPDM is fully crosslinked and suspended in a PP matrix in a
process called dynamic vulcanization. The mixed polymer is extruded in a manner similar to the
geomembrane types discussed in this section.

3.1.4  Polwinvl Ciloride (PVO

      As seen  in Table 3.1, polyvinyl chloride (PVC) geomembranes are made from poly vinyl
chloride resin, piasticizer(s), fillers and additives.

3.1.4.1 Resin

      The polyvinyl chloride resin used for PVC geomembranes is made by cracking ethylene
dichloride into a vinyl chloride monomer. It is then polymerized to make PVC resin.  The PVC
resin (in the form of a white powder) is then compounded with other components to form a PVC
compound.

      In the preparation of a specification or MQA document, the following  items concerning the
PVC resin should be considered.

      1. The polyvinyl chloride resin should be made from completely  virgin materials.

      2. A PVC compound will generally consist of 50-70% PVC resin, by weight.

      3. Typical quality control tests on the resin powder  will be contamination, relative
         viscosity, resin gels, color and dry time. The specific test procedures will be specified
         by the manufacturer.  Often they are other than ASTM tests.

      4. The frequency of performing each of the preceding tests should be  covered in the MQC
         plan and it should be implemented and followed.

      5. Quality control certification procedures used by the manufacturer should be implemented
         and followed.

3.1.4.2 Plasricizer

      Plasticizers are added to PVC formulations  to impart flexibility, improve handling and
modify physical and mechanical properties. When blended with the PVC resin the plasticizer(s)
must be completely mixed into the resin. Since  the resin is a powder, and the plasticizers are
liquid, mixing of the two components continues until the liquid is completely absorbed by the
powder.  The result is usually a  powder which can be readily conveyed.  However, it is  also
possible to  wet blend  with acceptable results. There are two general categories of possible
plasticizers; monomeric plasticizers and polymeric  plasticizers. There are  many specific types
within each category. For example, monomeric plasticizers are sometimes  phthalates, epoxides
and phosphates, while polymeric plasticizers are sometimes polyesters, ethylene copolymers and
nitrile rubber.

      For a specification or MQA document written around PVC plasticizer(s), the following
items should be considered.
                                         106

-------
                    1. If more than one type of plasticizer is used in a PVC formulation they must  be
                       compatible with one another.

                    2. The plastitizer{s) in a PVC compound are generally from 25-35% of the total compound
                       by weight.

                    3. The exact type of plasticizer(s) used by the manufacturers are rarely identified.  This is
                       industry-wid- practice and due to the long history of PVC is generally considered to be
                       acceptable.

                    4. The plasticizer(s) should be certified by the manufacturer as having a successful past
                       performance or as having been used on a specific number of projects.

              3.1.4.3 Filler

                    The filler used in a PVC formulation is a relatively small component (recall Table 3.1), and
              (if used at all) is generally not identified. Calcium carbonate, in powder form, has been used but
              other options also exist Certification as to successful past performance could be requested.

              3.1.4,4 Additives

                    Other additives for the purpose of ease of manufacturing, coloring and stabilization are also
              added to the formulation. They are generally not identified.  Certification as to successful past
              performance may be requested.

              3.1-5  Chlorosulfonated Polyethylene (CSPE-IO

                    As seen  in Table 3.1, chlorosdfonated polyethylene  (CSPE) geomembranes consist of
              Chlorosulfonated polyethylene resin, fillers, carbon black (or colorants) and additives.  The
              finished geomembrane is usually fabricated with a fabric reinforcement, called a "scrim", between
              the individual  plys of the material. It is then designated as CSPE-R.

              3.1.5.1 Resin

                    There are two different types of Chlorosulfonated polyethylene resin used to make CSPE
              geomembranes. One is a completely amorphous polymer while the other is a thermoplastic
              material containing a controlled amount of crystallinity to provide useful physical properties in the
              uncured state while maintaining flexibility without the need of any plasticizers. The second type is
              generally used to manufacture geomembranes. CSPE is made directly from branched polyethylene
              by adding chlorine and sulfur dioxide.  The chlorosulfonic groups acres preferred cross-linking
              sites during the polymer aging process.  In the typical commercial  polymer there is one
              chlorosulfonyl group for each 200 backbone carbon atoms.

                    CSPE resin pieces usually arrive at the sheet manufacturing facility in large canons. They
              are somewhat pillow shaped (about 1 cm diameter) and 2 cm in length. The resin pieces (see Fig.
              3.3) are relatively spongy in their resistance  to finger pressure.  Alternatively, CSPE can be
              premixed with carbon black in slab form which is then referred to as a master batch.  The master
              batch is usually  made by a formulator and shipped to the manufacturing facility in a prepared form.
L
107

-------
                              Fig. 3.3 - CSPE Resin Pieces
      In preparation of a specification or MQA document, the following items concerning the
CSPE resin should considered.

      1. The CSPE resin should be made from completely virgin materials.

      2. The formulation will usually be based on 40 to 60% of resin, by weight.

      3. Typical MQC tests on the CSPE resin will be Mooney viscosity, chlorine content, sulfur
         content and a series of vulcanization properties (e.g., rheometry and high temperature
         behavior).

      4. The CSPE resin can be premixed with carbon black in slab form (referred to as a "master
         batch") and shipped to the manufacturers facility.

      5. Additional quality control certification procedures used by the manufacturer should be
         implemented and followed.

      6. The frequency of performing each of the preceding tests should be covered in the MQC
         plan and it should be implemented and followed.
                                          108

-------
3.1.5.2 Carbon Black

      The amount of carbon black in CSPE geomembranes varies from 5 to 36%. The carbon
black functions as an ultraviolet light blocking agent, as a filler and aids in processing. The usual
types of carbon black used in CSPE formulations are N 630, N 774, N 762 and N 990 as per
ASTM D-1765. When low percentages of carbon black are used N 1 10 to N 220 should be used.
When the carbon black is premixed with the resin and produced in the form of a master batch of
pellets, it is fed directly into the mixer with the other components, such as fillers, stabilizers and
processing aids.

      A specification on carbon black in CSPE geomembranes, could be framed around the type
and amount of carbon black as just described, but this is rarely the case. Typical MQC certification
procedures should be available and implemented.

3.1.5.3 Fillers

      The purposes of blending fillers  into the CSPE compound are to provide workability and
processability.  The common types of fillers are  clay and calcium carbonate. Both are added in
powder form and in quantities ranging from 40 to 50%.

      Specifications are rarely written around this aspect of the material, however MQC
certification procedures should be available and implemented

3.1.5.4 Additives

      Additives are used in CSPE compounds for the purpose of stabilization which is used to
distinguish the various grades.  The industrial grade of CSPE geomembranes uses lead oxide as a
stabilizer, whereas the potable water grade uses magnesium oxide or magnesium hydroxide.
These stabilizers function as acid acceptors during the polymer aging process. During aging,
hydrogen chloride or sulfur dioxide releases from 'he polymer and the metal oxides react with these
substances inducing cross linking over time.

      Specifications are rarely written  around the type and quantity  of additives used in CSPE,
however MQC certification procedures should be  written around each additive, be available and be
implemented.

3.1.5.5 Retnfore|ng S crin|

      CSPE geomembranes are usually fabricated with a reinforcing "sc/im" between two plys of
polymer sheets. This results in a three-ply laminated geomembrane consisting of geomembrane,
scrim, geomembrane which is sealed together, under pressure, to form a unitized system.  The
geomembrane is said to be reinforced and then carries the designation CSPE-R.  Other options of
multiple plys are also available. The scrim imparts dimensional stability to the material which is
important during storage, placement and seaming. It also imparts a major increase in mechanical
properties over the umtinforced type, particularly in the tensile strength, modulus of elasticity and
tear resistance of the final geomembrane.

      The  reinforcing scrim for CSPE geomembranes is a woven fabric made from polyester
yarns in a standard "basket" weave.  Note that there are usually many fine fibers (of very fine
diameter) per individual yam, e.g., 100 to 200 fibers per yain depending on the desired strength.
The yarns,  or "strands" as they are referenced in the industry, are spaced close enough to one
another to achieve the desired properties, but far  apart enough to allow open space between them
                                          109

-------
so that the opposing geomembrane sheet surfaces can adhere together. This is sometimes referred
to as "strike-through" and is measured by a ply-adhesion test.  The designation of reinforcing
scrim is based on the number of yams, or strands, per inch of woven fabric.  The general range is
from 6 x 6 to 20 x 20, with 10 x 10 being the most common. A 10 x 10 scrim refers to 10 strands
per inch in the machine (or warp) direction and an equal number of 10 strands per inch in the cross
machine (or weft) direction.

      It must also be mentioned that ths polyester scrim yams must be coated for them to have
good bonding to the upper and lower CSPE sheets.  Various coatings, including latex, polyvinyl
chloride and others, have been used. The exact formulation of the coating material  (or "ply
enhancer") is usually proprietary.

      Regarding a specification or MQA document for the fabric scrim in CSPE-R geomembranes
the following applies.

      1. The type of polymer used for the scrim is usually specified as polyester, although nylon
         has been used in the past It should be identified accordingly.

      2. The strength of the fabric scrim can be specified and, when done, is best accomplished
         in tensile strength units of pounds per individual yarn rather than  individual fiber
         strength.

      3. The strike-through is indirectly quantified  in specifications on the basis of ply adhesion
         requirements.  This will be discussed later.

3.1.6 Qth.er Calendered Geomembranes

      Within thw  category of calendered geomembranes there are other types that have not been
described thus far. They will be briefly noted here  along with similarities  and/or differences to
those just described.

      Chlorinated polyethylene (CPE) has been used as a polymer resm in the past for either non-
reinforced or scrim reinforced geomembranes. Its production and ingredients are similar to CSPE,
or CSPE-R, with the obvious exception of the nature of the resin itself. In contrast to CSPE, CPE
contains no sulfur  in its formulation.

      Ethylene interpolymer alloy (EIA) is always used as a reinforced geomembrane, thus EIA-R
is its proper designation. The resin is a blend of ethylene vinyl acetate and polyvinyl chloride
resulting in a thermoplastic elastomer. The fabric reinforcement is a tightly woven polyester which
requires the polymer to  be individually spread coated on both sides of the fabric. Note, however,
that there are other related products being developed under different trademarks in this general
category.

      Among the newer geomembranes is polypropylene (PP) which is a very flexible olefmic
polymer based on new  polypropylene resin technology.  This polymer has been converted into
sheet by calendering, with  and without scrim reinforcement, and by flat die and  blown film
extrusion processes. Factory fabrication of large panels is possible. The initial field trials of this
type of geomembrane are currently ongoing.

3.2 Manufacturing

      Once the specific type of geomembrane formulation that is specified has been  thoroughly
                                          110

-------
mixed it is then manufactured into a continuous sheet.  The two major processes used for
manufacturing of the various types of sheets of geomembranes are variations of either extrusion
(e.g., for HDPE, VLDPE, and LLDPE) or calendering (e.g., for PVC, CSPE and PP).  Spread
coating (the least used process) will be briefly mentioned in section 3.2.8.

3.2.1 pending. Compounding. Mixing and/or Masticating

      Blending, compounding, mixing and'"** /nasticating of the various components described in
Section 3.1 is conventionally done on a weight percentage basis. However, each geomembrane's
processing is somewhat unique in its equipment and procedures. Even for a particular  type of
geomembrane, manufacturers will use different procedures, e.g., batch methods versus continuous
feed systems, for blending or mixing.

      Nevertheless, a few general considerations are important to follow in the preparation of a
specification or MQA document

      1. The blending, compounding, mixing and/or masticating equipment must be clean and
        completely purged from prsviously  mixed materials of a different formulation.  This
        might require sending a complete cycle  of purging material  through the system,
        sometimes referred to as a "blank".

      2. The various components of the formulation are added on a weight percentage basis to an
        accuracy set by industry standards. Different components are often added to the mixture
        at different locations in the processing, i.e., the entire batch is not necessarily added at
        the outset.

      3. By the rime the complete formulation is ready for extrusion or calendering it trust be
        completely homogenized.  No traces of segreta.tion, agglomeration, streakir>j or
        discoloration should be visually apparent in the finished product.

3JL2 Regrind. Reworked or Trim Reprocessed Material

      "Reg-rind", "reworked" or "trim"  are all terms  which can be  defined as finished
geomembrane sheet material which has been cut from edges or ends of rolls, or is off-specification
from a surface  blemish, thickness or other property point of view.   Figure 3.4(a) shows a
photograph of HDPE regrind chips, VLDPE chips appear similar to HDPE. Figure 3.4(b) shows
a photograph of PVC edge strips i.e., edge of sheet material cut off to meet specific roll width
requirements. Excess edge trimmings  of PVC sheet is fed back into  the production system.
CSPE-R trim can be added similarly, however without any reinforcing scrim.
                                                            "*»
      These materials are reintroduced during the blending, compounding and/or mixing stage in
controlled  amounts as a matter of cost efficiency on the part of the manufacturer.  Note that
regrind, rework and trim material must be clearly distinguished from "recycled", or "reclaimed",
material which is finished sheet material that has actually seen some type of service performance
and has subsequently been returned to the manufacturing facility for reuse into new sheet material.

      In preparing a specification or MQA document on the use of reprocessed material, the
following items should bs considered:

      1. Regrind, reworked or trim materials in the form of chips or edge strips may be added if
        the material is from the same manufacturer and is exactly the same formulation as the
        geomembrane being produced.
                                         in

-------
           Figure 3.4(a) - HOPE Regrind Chips
\
             Figure 3.4(b) - PVC Edge Strips
   Figure 3.4 - Photographs of Materials to he Reprocessed
                          112

-------
      2. Generally HDPE and VLDPE will be added in chip form as "regrind" in controlled
         amounts into the hopper of the extruder.

      3. Generally PVC, CSPE and PP will be added in the form of a continuous strip of edge
         trimmings into thi roll mill  which precedes calendering.   For scrim reinforced
         geomembranes it is important that the edge trim does not contain any portion of the
         fabric scrim.

      4. The maximum amount of regrind, reworked or trim material to be added is a topic of
         considerable debate. Its occurrence in the completed sheet is extremely difficult, if not
         impossible,  to identify much less to quantify  by current chemical fingerprinting
         methods.  Thus its maximum  amount is not suggested in this manual.  It should be
         mentioned that if regrind is not permitted to be used, the manufacturer may charge a
         premium over current practice.

      5. It is generrlly accepted that no amount of "recycled", or "reclaimed" sheet material (in
         any form whatsoever) should be added to the formulation.

3.2.3 High Density Polyethylene
      High density polyethylene (HDPE) geomembranes are manufactured by taking the mixed
components described earlier and feeding them into a hopper which leads to a horizontal extruder,
see Fig. 3.5.  In the manufacturing of HDPE geomembranes many extruders are 200 mm (8.0
inch) diameter systems which are quite large, e.g., up to 9 m (30 ft. long).  In an extruder, the
components enter a feed hopper and are transported via a continuous screw through a feed section,
compression stage, metering stage, filtering screen and are then pressure fed '-mo a die. The die
options currently  used for HDPE  geomembrane production are either fl%; :.cri/;.ital dies  or
circular vertical dies,  the latter production technique often bei"g refened to as "blown film"
extrusion. The length of flat dies and the circumference of circular dies determine the width of the
finished sheet and vary greatly from manufacturer to manufacturer. Some detail is given below.
                                                  Continuous
                                  Breaker Plate and
                                  Filter Screen
Drive
Mechanism

Feod
Section
                                          Compression     Metering
                                          Section       I
   Figure 3.5 - Cross-Section Diagram of a Horizontal Single-Screw Extruder for Polyethylu.e
                                         113

-------
3.2.3.1 PalPie-Wide Sheet

      A conventional HDPE geomembrane sheet extruder can feed enough polymer to produce
sheet up to approximately 4.5 m (15 ft.) wide in typical HOPE thicknesses of 0.75 to 3.0 mm (30
to 120 mils), see Fig. 3.6. Recently, one manufacturer has used two such extruders in parallel to
produce sheet approximately 9.0 m (30 ft) wide.
Figure 3.6 - Photograph of a Polyethylene Geomembrane Exiting from a Relatively Narrow Hal
            Horizontal Die
      Insofar as a specification or MQA document for finished HDPE geomembranes made by
flat die extrusion, the following items should be considered.

      I. The finished geomembrane sheet must  be free from pinholes, surface blemishes,
         scratches or o'.her defects (e.g., nonuniform color, streaking, roughness, carbon black
         agglomerates, visually discernible regnnd, etc.).

      2. The nominal and minimum thicknesses of the sheet should be specified. The minimum
         value is usually related to the nominal thickness as a percentage. Values range from 5%
         to 10% less than nominal.
                                          114

-------
      3. The maximum thickness of the sheet is rarely, if ever, specified. This is for the obvious
         reason that if a manufacturer wishes to supply sheet thicker than specified it is generally
         acceptable.  It is also done, however, to allow for those manufacturers with unique
         variations of flat die extrusion (such as horizontal ribs or factory fabricated seams) to not
         be excluded from the market.

      4, The finished sheet width should be controlled to be within a set tolerance.  This is
         usually done by ci eating a sheet larger than called for, and trimming the edges
         immediately before final rolling onto the wind-up core. (The edge trim is subsequently
         ground into chips and used as regrind as previously described). Flat  die extrusion of
         HOPE sheet should meet a ± 2.0% width specificauon.

      5. Other MQC tests such as strength, puncture, tear, etc. should be part of a certification
         program which should be available and implemented.

      6. The frequency of performing each of the preceding tests should be covered in the MQC
         plan and it should be implemented and followed.

      7 . The trimmed and finished sheet is wound onto a hollow wind-up core  which is usually
         heavy cardboard or (sometimes) plastic pipe. The outside diameter of the core should be
         at least 150 mm (6.0 in).  It obviously must be stable enough to support the roll without
         buckling or otherwise failing during handling, storage and transportation.

      8. Partial rolls for site specific project details may be cut and prepared for  shipment per the
         contract drawings.

3.2.3.2 Flat Die - Factory Seamecj

      Since there arc commercial extruders which produce sheets less than 6 m  (20 ft) wide, the
resulting sheet widths can be factory seamed into wider panels before shipment to the field. All of
the specification details just described apply to narrow sheets as well as to wide sheets.

      The method of factory seaming should be left to the discretion of the manufacturer.  The
factory seams,  however, must meet the same specifications as the field seams  (to be described
later).
3.2.3.3
               Flm
       By using a vertically oriented circular die the extruder\can feed molten polymer in an
upward orientation creating a large cylinder of polyethylene sheet, see Fig, 3.7. Since the cylinder
of polymer is closed at the top where it passes over a set of nip rollers which advances the
cylinder, air is generally blown within it to maintain its dimensional stability. Note that upward
moving air is also outside of the cylinder to further aid in stability.  After passing through the nip
rollers, the collapsed cylinder is cut longitudinally, opened to its full width, brought down to floor
level and rolled onto a wind-up core.  Note that collapsing the cylinder and passing it through the
nip rollers results in two creases. After slitting the collapsed cylinder and opening it to full width,
remnants of the two creases remain.
                                           115

-------
Figure 3.7 (a) - Photograph of Blown Film Manufacturing of Polyethylene Geomembianes

                                  Nip Ratter*       ^_
(P*
y
/
A
bub** J
sp va
\
V


f;
up
0|
l**H.
                                                             2-StBtlBn WlrxJ
                                                             up lor ConUnuou*
                                                             OpanUon
                                           Cooling Ring
                                           •nd Blowers
   Rg. 3.7(b) - Sketch of Blown Film Manufacturing of Polyethylene Geomembranes
                                       116

-------
       Regarding a specification  or MQA  document for blown film  produced  HDPE
geomembranes, the following applies:

      1.  The finished geomembrane sheet shall be free from pin holes, surface blemishes,
         scratches or other defects (e.g., nonuniform color, streaking, roughness, carbon black
         agglomerates, visually discernible regrind, etc.).  Note that two machine direction:
         creases from nip rollers are automatically induced into the finished sheet at  the 1/4
         distances from each edge.

      2. The nominal and minimum thickness of the sheet should be specified. The minimum
         value is usually  related to the nominal thickness as a percentage. Values referenced
         range from 5% to 10% less than nominal.                                        \

      3. The maximum thickness of the sheet is rarely, if ever, specified. This is for the obvious
         reason that if a manufacturer wishes to supply sheet thicker than specified it is generally
         acceptable.

      4. The finished sheet width should be controlled to be within a set tolerance.  HDPE
         geomembrane made from the blown film extrusion method should meet a ± 2.0% width
         specification.

      5. Other MQC tests  such as tensile strength, puncture, tear, etc., should be part of a
         certification program which should be available and implemented.

      6. The finished sheet is wound  onto a  hollow wind-up core which  is usually heavy
         cardboard or sometimes plastic pipe.  The outside diameter of the core should be at least
         150 mm (6.0 in.).  It must be stable enough to support  the roll without buckling or
         otherwise fa'ling during handling, storage and transportation.

      7. It is important that the two creases located at the  1/4-points from the edges of the sheet
         are wound on the core such that they will face upward when deployed in the field. The
         reason for this is so that scratches will not occur on the creases if the sheets are shifted
         on the soil subgrade when in an open and flat position.

      8. Partial rolls for site specific project details may be cut and prepared for shipment as per
         the contract drawings.
      By creating a roughened surface on a smooth HDPE sheet, a process called "texturing" in
this document, a high friction surface can be created. There are currently three methods used to
texturize smooth HDPE geomcmhn  *s: coexmision, impingement and lamination, see Fig. 3.8.

      The coexirusion  method utilizes a blowing agent in the molten extrudate and delivers it
from a small extruder immediately adjacent to the main extruder. When both sides of the sheet are
to be textured, two small extruders (one internal and one external to the main extruder) are
necessary.  As the extrudate from these smaller extruders meets  the cool air the blowing agent
expands, opens to the atmosphere and creates the textured surfacc(s).
                                          117

-------
                      Mtbur
                                 (•) Coextnukm with Nitrogen Cut
                                                                   Finished
                                                                   Tc.:mrad
                                                                   Sheet
                                                                   Die



                                                                   Main Core Extruder
                                                                   Internal Extruder (N2 Gas)
                                                                   External Extruder (N2 Gat)
             I Smooth
             'Roll
                                                                       . Finished
                                                                    Of) Textured
                                                                      J Roll
                                    (b) Impingement of Hot Polyethylene Particle
                                  Hot PE Foam
SprecderBar
                                Smooth Roll            Finished Texmred Roll
                                jmwwiwu    (Repeal Oppotin Side lerOovble Sided Tenm)

                                   'c) Lamination with Polyethylene Foam
Figure  3.8  -  Various  Methods  Currently  Used  to  Create  Textured Surfaces on HDPE
                 Geomembranes
                                                 118

-------
      Impingement of hot HDPE panicles against the finished HDPE sheet is a second method of          ;•
texturing. In this case, hot particles are actually projected onto the previously prepared sheet on          •
one or both of its surfaces in a secondary operation. The adhesion of the hot particles to the cold
surface(s) should be as great, or greater, than the shear  strength of the adjacent soil or other          :
abutting material. The lengthwise edges of the sheets can be left non-textured for up to 300 mm          i.
(12 in.) so that thickness measurements and field seaming can be readily accomplished.                     I
                                                                                                 I
      The third method for texturizing HDPE sheet is by lamination of an HDPE ioam on  the          j
previously manufactured smooth sheet in a secondary operation.  In this method a foaming ar   it          f-
contained within molten HDPE provides a froth which produces a rough textured laminate adh  red          •
to the previously prepared smooth sheet. The degree of adhesion is important with respect i   the          \
shear strength of the adjacent soil or other abutting material. If texturing on both sides  .  the          i
geomembrane is necessary, the roll must go through another cycle but now on its opposi'  side.          £
The lengthwise edges of the sheets can be left non-textured for up to 300 mm (12 in.)  j that
thickness measurements and field seaming can be readily accomplished.                                  *

       Regarding  the  writing  of a specification or MQA  document on  textured  HDPE
geomembranes the following points should be considered.

       1. The surface texturing material should be of the same type of polymer and formulation as
         the base sheet polymer and its formulation. If other chemicals are added to the texturing
         material they must be identified in case of subsequent seaming difficulties.

       2. The degree of texturing should be sufficient to develop the amount of friction as needed
         per the manufacturers specification and/or the project specifications.

       3. The quality control of the texturing process can be assessed for uniformity using an
         inclined plane test method, e.g., GRI GS-7*.

       4. The actual friction angle for design purposes should come from a large scale direct shear
         test simulating site specific conditions as closely as possible,  e.g., ASTM D-5321.

       5. The thickness of the base geomembrane should be micrometer measured (according to
         ASTM D-751) along the smooth edge strips  of  textured geomembranes  made by
         impingement or lamination.  For those textured geomembranes with no smooth edge
         strips, i.e., for blown  film coextruded materials, an overall average thickness can be
         estimated on the basis  of the roll weight divided by total area  with suitable incorporation
         of the density of the material. Alternatively, a tapered point micrometer for measuring
         screw threads has also  been used for point-to-point measurements.

       6. Other MQC tests such  as tensile strength,  puncture, tear, etc., should  be  part of a
         certification program which should be available and implemented.

       7. The frequency of performing each of the preceding tests should be covered in the MQC
         plan and it should be implemented and followed.
* The Geosyniheuc Research Institute (CRT) provides interim test methods for a variety of geosynihetic related
topics until such time as consensus organizations (like ASTM) adopt a standard on the same topic. At that time the
GRI standard is abandoned.


                                          119
                                                                                                 •i

-------
3.2.4 Very Low Density Polyethylene (VLDPIft

      Very low density polyethylene (VLDPE) geomembranes are manufactured by taking the
mixed components described earlier and feeding them into a hopper which leads to a horizontal
extruder, recall Fig. 3.5. In the extruder, the blended components enter via a feed hopper and are
transported via a continuous screw, through a feed section, compression stage, metering stage,
filtering screen and are then pressure fed into a die. The die options currently used for VLDPE
i,iomembrane production are either flat horizontal dies or circular vertical dies, the latter often
being referred to as "blown film" extrusion.  The width of flat dies and the circumference of
circular dies vary greatly from manufacturer to manufacturer. The techniques are the same as were
described in the manufacture of HOPE geomembranes.

3.2.4.1  Flat Die - Wide Sheet

      A conventional VLDPE sheet extruder can feed enough polymer to produce sheet up to
approximately 4.5 m (15 ft.) wide in typical VLDPE thicknesses of 0.75 to 3.0 mm (30 to 120
mils), recall Fig. 3.6. In developing a specification or MQA  document for  the manufacture of
VLDPE geomembranes the following should be considered:

      1. The finished geomembrane  sheet must be free from pinholes,  surface blemishes,
        scratches or other defects (e.g. carbon black agglomerates, visually  discernible regrind,
        etc.).

      2. The minimum thickness of the sheet should be specified. It is usually related to the
        nominal thickness as a percentage. Values range from 5% to 10% less than nominal.

      3. The maximum thickness of the sheet is rarely, if ever, specified. This is for the obvious
        reason Jiat if a manufacturer wishes to supply sheet thicker than specified it is generally
        acceptable. It  is also done, however, to allow for  those manufacturers with  unique
        variations of flat die extrusion  (such as horizontal ribs or factory fabricated seams) to not
        be excluded from the market.

      4. The finished sheet width  should  be controlled to be within a set tolerance. This is
        usually  done  by creating a  sheet larger than called  for, and trimming  the edges
        immediately before final rolling onto the wind-up core. (The edge trim is subsequently
        ground into chips and used as regrind as previously described). Flat die extrusion of
        VLDPE sheet can readily meet a ± 0.25% width specification.

      5. Other MQC tests such as tensile strength, puncture, tear, etc. should be  part of a
        certification program which should be available and implementcd>

      6. The trimmed and finished sheet is wound onto a hollow wind-up core which is usually
        heavy cardboard or sometimes plastic pipe.  The outside diameter of the core should be
        at least 150 mm (6.0 in). It obviously must be stable enough to support the roll without
        buckling or otherwise failing.

      7. Partial rolls for site specific project details may be cut and prepared for shipment as per
        contract drawings.

3.2.4.2  Flat Die - Factory  Seamed

      Since there are commercial extruders which pnxluce significantly narrower sheet than just
                                         120

-------
discussed, the resulting narrow sheet widths can be factory seamed into wider panels before
shipment to the field. All of the specification details just described apply to narrow sheets as well
as to wide sheets.

      The method of factory seaming should be left to the discretion of the manufacturer. The
factory seams, however, must be held  to the same destructive and  nondestructive testing
procedures as with field seams (to be described later).

3.2.4.3 Blown Film

      By using a circular die oriented vertically the extruder can feed molten polymer in an
upward orientation  creating a large cylinder of polymer, recall Fig.  3.7. Since the cylinder is
closed at the top where it passes over a  set of nip rollers which advances the cylinder, air is
generally contained within it maintaining its dimensional stability.  Note that upward moving air is
also outside of the cylinder to further aid in stability.  After passing beyond the nip rollers the
cylinder is cut longitudinally, opened to its full width, brought down to floor level and rolled onto
a stable core.

      The following items should be considered in preparing a specification or MQA document
for blown film VLDPE geomembranes.

      1, The finished geomembrane sheet shall be free from pinholes, surface blemishes,
         scratches or other defects (carbon black agglomerates, visually discernible regrind, etc.).
         Note that two machine direction creases from nip rollers are automatically induced into
         the finished sheet at the 1/4 distances from each edge.

      2. The minimum thickness of the  sheet should be specified.  It is usually related to the
         nominal thickness as a percentage. Values referenced range from 5% to 10% less than
         nominal.

      3. The maximum thickness of the sheet is rarely, if ever, specified. This is for the obvious
         reason that if a manufacturer wishes to supply sheet thicker than specified it is generally
         acceptable.                                              ,

      4. The finished sheet width should be controlled to be within a  set tolerance.  VLDPE
         geomembrane made from the blown film extrusion method should meet a ± 2.0% width
         specification.

      5. Other  MQC tests such as  tensile strength, puncture,  tear.^etc. should be part of a
         certification program which should be available and implemented.

      6. The finished sheet is wound onto a hollow wind-up core which is usually heavy
         cardboard or sometimes plastic pipe. The outside diameter of the core should be at least
         150 mm (6.0 in.).  It obviously must be stable enough to support the roll without
         buckling or otherwise failing.

      7. Partial rolls for site specific project details may be cut and prepared for shipment as per
         contract drawings.

3.2.4.4 Textured Sheet

      By creating a roughened surface on a smooth VLDPE sheet, a process called "texturing" in


                                           12!

-------
this document, a high friction surface can be created.  There are currently three methods used to
texturize smooth VLDPE geomembranes:  eoextrusion, impingement and lamination, recall Fig.
3.8.

      The eoextrusion method utilizes a blowing agent in the molten extrudate and delivers it
from a small extruder immediately adjacent to the main extruder. When both sides of the sheet are
to be textured, two small  extruders, one internal and one external to the main extruder,  are
necessary.  As the extrudate from these smaller extruders meets the cool air the blowing agent
expands, opens to the atmosphere and creates the textured surface(s).

      Impingement of hot polyethylene particles against the finished VLDPE sheet is a second
method of texturing. In this case, hot particles are actually projected onto the previously prepared
sheet on one or both of its surfaces in a secondary operation. The adhesion of the hot particles to
the cold surface(s) should be as great, or greater, than the shear strength of the adjacent soil or
other abutting material. The lengthwise edges of the sheets can be left non-textured for up to 30
cm (12 in.) so that thickness measurements and field seaming can be readily accomplished.

      The third method for texturizing VLDPE sheet is by lamination  of a hot polyethylene foam
on the previously manufactured smooth sheet in a secondary operation. In this method a foaming
agent contained in molten polyethylene provides a froth which  produces a rough textured laminate
adhered  to the previously prepared smooth sheet. The degree of adhesion is important with respect
to the shear strength of the adjacent soil or other abutting material. If texturing of both sides of the
geomembrane  is necessary the roll must go through another cycle but now on its opposite side.
The lengthwise edges of the sheets can be left non-textured  for up to 300 mm (12 in.) so that
thickness measurements and field seaming can be readily accomplished.

      Regarding  the writing  of a specification  or MQA document on textured VLDPE
geomembranes the following points should be considered.

      I. The surface texturing material should be polyethylene of density equal to the VLDPE, or
         greater.  The latter is often the case.  If other chemicals are  added to  the texturing
         material they must be identified in case of subsequent seaming difficulties.

      2. The degree of texturing should be sufficient to develop the amount of friction as needed
         per the manufacturers specification and/or the project specifications.

      3. The quality  control of the texturing process can be assessed for uniformity using an
         inclined plane test method, e.g., GRI GS-7.

      4. The actual friction angle for design purposes should come from a large scale direct shear
         test simulating site specific conditions as closely as possible, e.g., ASTM D-532I.

      5, The thickness of the base  geomembrane should be micrometer measured  (according to
         ASTM D-751) along the smooth edge strips of textured geomembranes made by
         impingement or lamination.  For those textured VLDPE geomembranes with no smooth
         edge strips, i.e., for blown film coextruded materials, an overall average thickness can
         be estimated on the basis of the roll weight divided by total area with suitable
         incorporation of the density of the material.  Alternatively, a tapered point micrometer for
         measuring screw threads  has  also been used for point-to-point measurements.  Care
         must be exercised, however,  because VLDPE thickness measurements with a point
         micrometer are very sensitive to pressure.
                                          122
1
fs
i

?«
i
13
ff
F
 i

-------
      6. Other MQC tests such as tensile strength, puncture, tear, etc., should be part of =
         certification program which should be available and implemented.

      7. The frequency of performing each of the preceding tests should be covered in the MQC
         plan and it should be implemented and followed.

3,2,5  Coextrusion Processes

      As mentioned previously in Section 3.1.3, there are other variations of manufacturing
polyethylene geomembrancs. The basic manufacturing principle of adding the desired components
to an extruder and having the molten polymer exit a flat horizontal die or a circular vertical die is
always the same. What is different between these variations and the single component HDPE or
VLDPE just described is the coextrusion process along  with the idiosyncrasies of the particular
materials utilized.

      In coextrusion, two or three extruders simultaneously introduce molten polymer into the
same die. As the different materials exit the die and are cooled they commingle with one another
such that local blending and molecular entanglement occur and no discrete separation layer exists.
Thus coextrusion is fundamentally different from the lamination of different surfaces together or of
preformed sheets together under heat  and pressure.  Different  variations of coextrusion of
polyethylene geomembranes are described as follows.

      Since polyethylene resin is supplied as a opaque pellet, the addition of colorants (rather than
carbon black) can product white, blue, green, etc., colored geomembranes. The benefit for
geomembranes having these light colors  is to reduce ihe surface temperature of the geomembrane
when  it is required to be exposed, e.g., as liners for surface impoundments or floating covers for
reservoir5.  Figure 3.9 shows how the temperature differences between white and black can be
very significant The white (or light) colors  generally  utilize titanium dioxide (or other metal
oxides) in  amounts not  exceeding  1.0% by weight.   Note that only a thin  surface layer
(approximately  10-20% of the total thickness) is treated in  this  manner.  The  balance of the
geomembrane contains carbon black and is treated in the same manner as described previously.
Tomperaturo ('C)
60-
50-
40-
30.
-— a «
f



g 6lac* Geomembrana
* White GsomemOrane

r^
X
60     120     180

      Tim* (mins.)
                                                       240
300
  Figure 3.9 - Geomembrane Surface Temperature Differences Between Black and White Colors

      A second variation of polyethylene is to coextrudc a "sandwich" of HDPE on each side of
VLDPE in the center. The purpose of such a combination is to provide high chemical resistance on
the top and bottom of the sheet (via the HDPE) and to have high  flexibility and out-of-plane

                                                                                               f Mi

                                                                                               I'J1
                                          123

-------
elongation properties  within the core (via the VLDPE). The thickness percentages of these
components are approximately 20%, 60% and 20% of the total thickness of the sheet, respectively.

      Third, it is possible to coextrude a surface layer to conventional HDPE or VLDPE which
contains a gas that expands when cooled.  Thus the molten polymer moves  through the die in a
regular manner only to have the expanding gas rapidly exit on its surface(s).  This forms a
roughened, or textured, surface which depends on  the  amount of gas and thickness  of the
coextruded surface lu/er. Similar extruders can be used on both sides of the parent sheet. The
purpose of such texturing is to increase the interface friction between the textured geomembrane
and the material above and/or below it, refer to Sections 3.2.3.4 and 3.2.4,4.

      Lastly, it is possible to coextrude other polymers than polyethylene.  As noted in Section
3.1.3, fully crosslinked elastomeric alloys (FCEA) can be extruded or could be coextruded with
other polymers.

3.2.6 Polwinvl Chloride fPVQ

      Polyvinyl chloride (PVC) geomembranes are manufactured by taking proportional weight
amounts of PVC resin (a dry powder) and plasticizer (a liquid) and premixing them until the
plastictzer is absorbed into the resin. Filler (in the form of a dry powder) and other additives (also
usually dry powders) are then added to the plasticized resin and the total formulation is mixed in a
blender. Various types of high intensity or low intensity blenders can be used.  Note that PVC
rework in the form of chips, rather than edge trim, can be introduced at this point.

      The resulting free-flowing powder compound is fed into a mixer which has heat introduced
thereby initiating a reaction between the various components. These mixers can be either batch
type (e.g., Banbury) or continuous types (e.g., Parrel), see Figs.  3.10(a) and (b), respectively. In
these mixers, the temperature is appiuximately 180°C  (350°F) which melts the mixture into a
viscous mass.   The mixed material is then removed from the discharge door or port onto a
conveyor belt.  From  the  conveyor belt the viscous material  is  further worked  (called
"masticating") in a rolling mill (or mills) into a smooth, consistent, uniform color, continuous mass
of 100-150 mm (4-6 in.) in diameter. Finished product edge trim can also be introduced into the
rolling mill at this point.  The fully mixed formulation is then fed by conveyor directly into the
sizing calender.

3-2.6,1 Calendering

       PVC formulations, irrespective of the pre-processing procedures, are manufactured into
continuous geomembrane sheets by a calendering process. The viscous feed of polymer coming
from the rolling mill(s) is  worked and  flattened  between  counter-rotating rollers  into a
geomembrane sheet.  Most calenders are "inverted-L" configurations, see  Fig.  3.11, but other
options also exist. The rollers are. usually smooth  surfaced (they can be slightly textured) stainless
steel cylinders and are up to 200 cm (80 in.) in  width. The opening distance between adjacent
cylinders is set for the desired thickness of the final sheet. A rolling bank of molten material is
formed between adjacent rolls.   In an inverted four roll "L" calender, 3 such banks are formed,
They act as reservoirs for the molten material, and help to fill the  sheet to full thickness as it passes
between the rolls. As the geomembrane exits from the calender, it enters an additional series of
rollers for the purposes of pickoff, embossing, stripping, cooling and cutting. At least one, and
perhaps two, rollers in PVC manufacturing are embossed so as to impart a surface texture on the
geomembrane. The purpose of this embossing is to prevent the rolled geomembrane from sticking
together, i.e., "blocking", during wind-up, storage and transportation.
                                          124

-------
                 Ram
 Sliding
Discharge
  Door
                           Feed Hopper
                                Rotors
                                Cooling/
                                Healing
                                Channels
                    (a) Batch Process Mixer
                                                       Feed
      Discharp
        Oifice
        Gate
y
Discharge
                        (b) Continuous Type Mijrer

        Figure 3.10 - Sketches of Various Process Mixers
                            125

-------
                                                          Feed
         Feed
      Roll ing Bank
                                                            (b) INVERTED!.
                (a) VERTICAL
                                                     Rolling Bank
                                            (e) INCLINED Z
                       Figure 3.11 - Various Types of Four-Roll Calenders
      In developing  a specification or MQA document for the manufacturing of PVC
georoembranes the following considerations are important:

       1.  The finished genmcmbrane sheet should be free from pinholes, surface blemishes,
          scratches or other defects (agglomerates of various additives or fillers, visually
          discernible rework, etc.)

       2.  The finished geomembrane sheet surfaces should be of a uniform color.

       3.  The addition of a dusting powder, such as talc, to eliminate blocking  is not an
          acceptable practice. The powder will invariably attach to the sheet or be trapped within
                                         126

-------
      4.
the embossed irregularities and eventually be contained in the seamed area as a
potential contaminant which could effect the adequacy of the seem.

The nominal and minimum thickness of the sheet should be specified. The minimum
thickness of  the finished geomembrane sheet  is usually limited to  the nominal
thickness minus 5%.
      5 .  The maximum thickness of the finished geomembnne sheet is generally not specified.

      6,  The widt'ii of the finished PVC geomembrane is dependent on the type of calender
          used by the manufacturer.

      7.  The geomembrane sheet should be edge trimmed to result in a specified width.  This
          should be controlled to within ± 0.25%.

      8.  Various MQC tests such as tensile strength, puncture, tear, etc. should be part of a
          certification program which should be available and implemented.

      9.  The frequency of performing each of the preceding tests should be covered in the
          MQC plan and it should be implemented and followed.

       10. The finished geomembrane sheet should be roUed onto stable wind-up cores of at least
          75 mm (3.0 in.) in diameter.

3.2.6.2 Panel Fabrication

      PVC geomembranes as just described are typically 100 to 200 cm (40 to 80 in.) wide and
are transported in rolls weighing up to 6.7 kN (1500 pounds) to a panel fabrication facility, see
Fig. 3.12 (upper photo). When a specific job order is placed, the rolls are unwound and placed
directly on top of one another for factory seaming into a panel, sec Fig. 3.12 (lower photo). A
panel will typically consist of 5 to 10 rolls which are accordion seamed to one another, i.e., the left
side of a particular roll is seamed to the underlying roll while the right side is seamed to the
overlying roll. After seaming, the completed panel is again accordion folded (now in a lengthwise
direction) and placed on a wooden pallet It is then covered with a protective wrapper and shipped
to the job site for deployment. To be noted is that some fabricators use other procedures for panel
preparation.

      Regarding a specification or MQA document  for factory fabrication of PVC geomembrane
panels, the following items should be considered.

      1 . The factory seaming of PVC rolls into panels should be performed by thermal or
         chemical seaming methods,  see ASTM D-4545. It should be noted that dielectric
         seaming is a factory seaming method for joining PVC rolls.  This is a thermal (or heat
         fusion)  method that is acceptable  and  is unique  to  factory seaming  of  flexible
         thermoplastic geomembranes. It is currently not a field seaming method.

      2. Factory seams should be subjected to the  same type of destructive and nondestructive
         tests as field scams (to be described later).

      3. When factory seams are made by chemical methods they aie generally protected against
         blocking by covering them with  a 100 mm (4 in.) wide strip of thin polyethylene film.
         When the panels are unfolded in the field these strips are discarded.
                                          127

-------
        ."qH* ••

        *'V»-'*!5-K?^!*
          • .if •* Tfc^F™
         ^. ^>r*
-------
      4. The finished and folded panels must be protected against accidental damage and
         excessive exposure during handling, transportation and storage.  Usually they are
         protected by covering them in a heavy cardboard enclosure and placed on  a wooden
         pallet for shipping.

      5. The cardboard enclosures should be labeled and coded according to the specific job
         specifications.

3.2.7 Chlorosulfonated Polyethylene-Scrim Reinforced (CSPE-R)

      Chlorosulfonated polyethylene geomembranes are made by mixing CSPE resin with carbon
black (or their colorants) thereby making a "master batch" of these two components. Added to this
master batch are fillers, additives and lubricants in a batch type mixer, e.g., a Ban bury mixer,
recall Fig. 3.10(a). Within the mixer the shearing action of the rotors against the ingredients
generates enough heat to cause melting and subsequent chemical reactions to occur.  After the
mixing cycle is complete, the batch is dropped from the Banbury onto a two-roll mill, then to a
conveyor leading to a second two-roll mill. In moving through the roll mill it is further mixed into
a completely homogenized material having a uniform color and texture. It should be  noted that
edge trim is often taken from finished sheet and routed back to the roll mill for mixing and reuse.

      A conveyor now transports the material directly to the calender, as shown in Fig, 3.11, and
feeds it between the appropriate calender rolls.

3.2.7.1 Calendering

      All CSPE formulations are manufactured into  geomembrane sheets by a calendering
process. Here the viscous ribbon of polymer is worked and flattened into  a geomembrane sheet.
Most calenders are "inverted-L" configurations, recall Fig. 3.11, but other options also exist.  As
the geomembrane exits the calender, it enters a series of rollers  for the purposes of pickoff,
stripping, cooling and cutting.

      The inverted-L type calender provides  an opportunity to  introduce two simultaneous
ribbons of the mixed and masticated polymeric compound thereby making two individual sheets of
geomembranes.  While this section of the manual is written around CSPE, it should be recognized
that many other geomembrane types which are calendered can be made in multiple ply form as
well.  Since they  are  separately formed  geomembrane sheets, they  are brought together
immediately upon exiting the calender to provide a laminated geomembrane consisting of two plys.
Additional plys can also be added as desired, but this is not usually done in  the manufacture of
CSPE geomembranes.

      While producing the two separate plys in an inverted-L calender as mentioned above, a
woven fabric, called a reinforcing scrim, can be introduced between the two plys, see Fig. 3.13.
The CSPE geomembrane is then said to be reinforced and is designed CSPE-R. It is common
practice, however, to just use the acronym CSPE when referring to either the nonreinforeed or
reinforced variety of CSPE. The scrim is usually a woven polyester yarn with 6x6,10 x 10 or 20
* 20 tuuni. These numbers refer TO ine number of yarns per incn in tne macnine and cross machine
directions, respectively.  Other scrim counts are also possible.

-------
                 Figure 3.13 - Multiple-Ply Scrim Reinforced Geomfmbrane
      Regarding the preparation of a specification or MQA document for multiple-ply scrim
reinforced CSPE-R geomembranes the following should be considered,

      1.  The finished geomembrane should be free from surface blemishes, scratches and other
         defects (additive agglomerates, visually discernible rework, etc.).

      2,  The finished geomembrane sheet should be of a uniform color (which may be black, or
         by the addition of colorants, be white, tan, gray, blue, etc.), gloss and surface texture.

      3.  A uniform reinforcing scrim pattern  should be reflected on both sides  of the
         geomembrane and should be free from such anomalies as knots, gathering of yarns,
         delaminarions or nonuniforrn and deformed scrim.          "

      4,  The sheet should not be embossed since the surface irregularities caused by the scrim
         are adequate to prohibit blocking.

      5.  The thickness of the sheet should be measured over the scrim and at a minimum should
         be the nominal thickness minus 10%.

      6.  The geomembrane sheet should  have a salvage, i.e.,  geomembrane ply directly on
         geomembrane ply with  no  fabric scrim,  on both edges.  This salvage shall be
         approximately 6 mm (0.25 in.).

      7.  Various MQC tests such as strength, puncture, tear, ply adhesion, etc., should  be part
         of a certification program which should be available and implemented.
                                          130

-------
      8.  The frequency of performing each of the preceding tests should be covered in the MQC
         plan and it should be implemented and followed.

      9.  The finished geomembrane sheet should be rolled onto stable wind-up cores of at least
         75 mm (3.0 in.) in diameter.
3.2.7.2 Panel Fabrication

      CSPE-R geomcmbranes as just described are typically 100 to 200 cm (40 to 80 in.) wide
and are transported in rolls weighing up to 6.7 kN (1500 pounds) to a panel fabrication  facility.
When a specific job order is placed, the rolls are unwound and placed on top of one another for
factory seaming into a panel, recall Fig. 3.12. A  panel will typically consist of 5 to  10 rolls
accordion seamed to one another.  After seaming, the panel is accordion folded in its length
direction and placed onto a wooden pallet.  It is then appropriately covered and shipped to the job
site for deployment.  To be noted is that some  fabricators use other procedures for panel
preparation.

      In preparing  a specification or MQA document for CSPE-R geomembrane panels, the
following items should be considered.

      1. Factory seaming of CSPE-R rolls should use thermal, chemical or bodied chemical
         fusion methods, see ASTM D-4545. It  should be noted that dielectric seaming is a
         factory seaming method for joining CSPE-R rolls. This is a thermal, or heat fusion,
         method that is acceptable  and is currently unique  to factory seaming of  flexible
         thermoplastic geomembranes. It is not a field seaming method.

      2. Factory seams should be subjected to the same type of nondestructive tests  as field
         seams (to be described later). A start-up seam is made prior to making panel production
         seams from which destructive tests are taken (to be described later).

      3. When factory seams are made by chemical fusion methods they are generally protected
         against sticking to the adjacent sheet (i.e., blocking) by covering them with 100 mm (4
         in.) wide thin strip of polyethylene film. When the panels arc unfolded in the field these
         strips are discarded.  Other systems may not require this film.

      4. The folded panels must be protected against accidental damage and excessive exposure
         during handling, transportation and storage.  Usually they-are protected by containing
         them in a heavy cardboard enclosure and placed on  a wooden pallet for shipping.

      5. The cardboard enclosures are labeled  and  coded according to the specific job
         specifications.

3.2.8  Spread Coated Geomembranes

      As mentioned previously, an exception to the calendering method of producing  flexible
geomembranes, is the spread coating process. This process is currently unique to a geomembrane
type called ethy'.ene interpolvmer alloy (EIA-R), but has been used to produce other specialty
geomembranes in the past. The process utilizes a dense fabric substrate, commonly either a woven
or nonwoven textile, and spreads the molten polymer on its surface. Due to the dense structure of
the fabric, penetration of the viscous polymer to the opposite side is usually not complete. When


                                          131

-------
cooled, the sheet must be turned aver and the process repeated on the opposite side. Adherence of
die polymer to the fabric is essential.

      Geomembranes produced by the spread coating method are indeed multiple-ply reinforced
materials, but produced  by a method other  than calendering.  MQC and  MQA  plans and
specifications should be framed in a  similar manner as described previously for CSPE-R
geomembranes.

3.3
      While there should be great concern and care focused on the manufacturers and installers of
geomembranes, it is also incumbent that they are packaged, handled, stored, transported, re-
stored, re-handled and deployed in a manner so as not to cause any damage. This section is
written with these many ancillary considerations in mind.

3.3.1  Packaging

      Different types of geomembranes require  different types of packaging after the> are
manufactured. Generally HOPE and VLDPE are packaged around a core in roll form, while PVC
and CSPE-R are accordion folded in two directions and packaged onto pallets.

3.3.1.1 Balls

      Both HDPE and VLDPE geomembranes are manufactured and fed directly to a wind-up
core in Mi-width rolls. No external wrapping or covering is generally needed, nor provided.
These rolls, which weigh up to 22 kN (5000 pounds), are either moved by fork-lifts using a long
rod inserted into the core (called a "stinger") or they are picked up by fabric slin&d with a crane or
hoist. Note that the slings are often dedicated to each particular roll and follow along with it until
its actual deployment. The rolls are usually stored in an outdoor area. They are stacked such that
one roll is nested into the valley of the two underlying rolls, see Fig. 3.14.

      Regarding a specification or MQA document for finished rolls of HDPE geomembranes the
following applies.

       1 . The cores on which the rolls of geomembranes are wound should be at least 150 mm
         (6.0 in.) outside diameter.

      2. The cores  should have a sufficient inside diameter such that fork lift stingers can be used
         for lifting and movement.

      3. The cores should be sufficiently strong that the roll can be lifted by a stinger or with
         slings without excessively deflecting, nor structurally buckling the roll.

      4. The Stacking of rolls at the manufacturing facility should not cause buckling of the cores
         nor flattening of the rolls. In general, the maximum stacking limit is 5 roils high.

      5. If storage  at the manufacturer's facility is for longer than 6 months, the rolls should be
         covered by a sacrificial covering, or placed within a temporary or permanent enclosure.

      6. The manufacturer should identify all rolls with the manufacturer's name, product
         identification, thickness, nailer number, roll dimensions and date manufactured.

-------
             Figure 3.14 - Rolls of Polyethylene Awaiting Shipment to a Job Sl:e
3,3.1.2 Accordion Folded

      PVC and CSPE-R geomembranes are initially manufactured in rolls and are then sent to a
fabricator for factory seaming into panels. At ihe fabrication facility they are unrolled directly on
top of one another, factory seamed along alternate edges ot' the rolls and are then accordion folded
both width-wise and length-wise and placed onto wooden  pallets for packaging and shipment.
PVC and CSPE-R  geomembranes are generally not stored longer than a few weeks it the
fabrication facility.

      Regarding items for a specification or MQA document, the following applies.

      1. The wooden pallets on which the accordion folded geomembranes are placed should be
         structurally sound and of good workmanship so that fork lifts or cranes can transport
         and maneuver them without structurally failing or causing damage to the geomembrane.

      2. The wooden pallets should extend at least 75 mm (3 in.) beyond the edge of the folded
         geomembrane panel on all four sides.

      3. The folded geomembrane panel should be packaged in treated cardboard or plastic
         wrapping for protection from precipitation and direct ultraviolet exposure.

      4, Banding straps around the geomembrane and pallet should be properly cushioned so as
         not to cause damage to any pan of the go •"•membrane panel.
                                          133
                                                                                                a

-------
      5. Palleted geomembranes should be stored only on level surfaces since the folded material
         is susceptible to shifting and possible damage.

      6. The stacking of palleted geomembrane panels on top of one another should not be
         permitted.

      7. If storage at the fabricator's facility is for longer than 6 monUis, the palleted panels
         should be covered with a sacrificial covering, temporary shelter or placed within a
         permanent enclosure.

      8. The fabricator should identify all panels with the manufacturers name, product
         information, thickness, panel number, panel dimensions and date manufactured.

3.3.2 Shipment. Handling and Site Storage

      The geomembrane rolls or pallets are shipped to the job site, offloaded, and temporarily
stored at a remote location on the job site, see Fig. 3.15.

      Regarding items for a specification or CQA document*, the following applies:

      1. Unloading of rolls or pallets at the job site's  temporary -storage location should be such
         that no damage to the geomembrane occurs.

      2. Pushing, sliding or  dragging of rolls or pallets of geomembranes should not be
         permitted.

      3. Offloading at the job site should be performed with crones or fork lifts in a workmanlike
         manner such that damage does not occur to any part of the geomembrane.

      4. Temporary storage at the job site should be in an area where standing  water cannot
         accumulate at any time.

      5. The ground surface should be suitably prepared such that no stones or other rough
         objects which could damage the geomembranes are present

      6. Temporary storage of rolls of HDPE or VLDPE geomembranes in the field should not
         be so high that crushing of the core or flattening of the rolls occur. This limit is typically
         5 rolls high.

      7. Temporary storage of pallets of PVC or CSPE-R geomembranes by stacking should not
         be permitted.

      8. Suitable means of securing the rolls or pallets should be used such that shifting, abrasion
         or other adverse movement does not occur.

      9. If storage of rolls or pallets of geomembranes at the job site is longer than 6 months, a
         sacrificial covering or temporary shelter should be provided for  protection against
         precipitation, ultraviolet exposure and accidental damage.
* Note that the designations of MQC and MQA will now shift to CQC and CQA since field construction personnel
are involved. These designations will cany forward throughout the remainder of this Chapter.
                                          134

-------
                                                                                                 p..)
                                                                                                  -1
                                X
               Figure 3.15 - Phcr^graph of Truck Shipment of Geomcmbianes

3.3.3 Acceptance and Conformance Testing

      It is the primary duty of the installation contractor, via the CQC personnel, to see that the
geomembrane supplied to the job site is the proper material that was called for in the contract, as
specified by the Plans and Specifications.  It is also the duty of the CQA Engineer to verify this
material to be appropriate. Clear marking should identify all rolls or pallets with the information
described in Section 3.3.1, A complete list of roll numbers should be prepared for each material
type.

      Upon delivery of the rolls or pallets of geomembrane, the CQA Engineer should ensure that
conformance test samples are obtained and sent to the proper laboratory for testing.  This will
generally be the laboratory of the CQA Firm, but may be that of the CQC firm if so designated in
the CQA documents.  Alternatively, conformance testing could be performed at the manufacturers
facility and  when completed the particular lot should be marked for the particular site under
investigation.

      The following items should be considered for a specification or CQA document with regard
to acceptance and conformance testing.

      1. The particular tests selected for acceptance and conformance testing can be all of those
         listed previously, but this is rarely the case since MQC and MQA testing should have
         preceded the field  operations.   However,  at a minimum,  the following tests  are
         recommended  for field  acceptance and conformance testing for  the particular
                                          135

-------
        geomembrane type.
         (a) HDPE: thickness (ASTM D-5199), tensile strength and elongation (ASTM D-638)
            and possibly puncture (FTM Std 101C) and tear resistance (ASTM D-1004, Die C)

         (b) VLDPE:  thickness (ASTM D-5199), te..sile strength and elongation (ASTM D-
            638), and possibly puncture (FTM Std 101Q and tear resistance (AS'.M D-1004,
            DieC)

         (c) PVC:  thickness (ASTM D-5199), tensile strength and elongation (ASTM D-882),
            tear resistance (ASTM D-1004, Die Q

         (d) CSPE-R: thickness (ASTM D-5199), tensile strength and elongation (ASTM D-
            751), ply adhesion (ASTM D-413, Machine Method, Type A)

      2. The method of geomembrane  sampling should be prescribed.  For geomembranes on
        rolls,  1 m (3 ft.) from the entire width of the roll on the outermost wrap is usually cut
        and removed. For geomernbranes folded on  pallets, the protective covering must be
        removed, the uppermost accordion folded section opened and an appropriate size sample
        taken. Alternatively, factory seam retains can be shipped on top of fabricated panels for
        easy access and use in conformance testing.

      3. The machine direction must be  indicated with an arrow on all samples using a permanent
        marker.

      4, Samples are usually taken on the basis of a stipulated area of geomembrane, e.g., one
        sample per 10,000 m2 (100,000 ft2).  Alternatively, one could take samples at the rate of
        one per lot, however, a lot must be clearly defined. One possible definition could be that
        a lot is a group of consecutively numbered rolls or panels from the same manufacturing
        line.

      5. All conformance test results should be reviewed, accepted and reported by the CQA
        Engineer before deployment of the geomembrane.

      6. Any nonconformance of test results should be reported to the Owner/Operator.  The
        method of a resolution of such differences should be clearly stated in  the CQA
        document. One  possible  guidance document for failing conformance tests could be
        ASTM D-4759 titled "Determining the Specification Conformance of Geosynthetics".

3.3.4 Placenyqt

      When the subgrade or subbase (either soil or some other geosymhetie) is approved as being
acceptable, the rolls or pallets of the temporarily stored geomembranes are brought to their intended
location, unrolled or unfolded, and accurately spotted for field seaming, see Fig, 3.16.

3.3.4.1  Siibgrade££uj?base| Conditions

      Before beginning to move the geomembrane roils or pallets from their temporary storage
location at the job site, the soil subgrade (or other subbase material) should be checked for its
preparedness.
f .* 3
f-as
 * •*»
 I
 > *'1
 r
                                         136

-------
                                                  "-AJ&W*-/
      Wte  w-£M

      t£-:^  -"

      •?~  -.T—
      --**
       ••»••%>
                    -—sSr-/"?i53/'

        ^l^^^i^^^--: :fcm^^-^
       ,-jii-^*  u'-"t~«— "^i-r:'  ' "  • ^^"rSL  't*<^S£^J
       .*-              ^-  ^«  .v^a     ""'"'" •^**J*-
                       ^_ "    *"'': '      •  •••»«•.
           ^:-- ^,.s,.^--^->.-.,.  _-:.
      •i;-'-"*#&-*.;.'    •"•-i^*'4'*:  t-^1?**^.  '••
      ^     ^<"'	—.^"-^^i^.       ,tj -..-..-^
               ;•.---,* •*»,* v**..i^..-^^.   -*&*a*^.^r^&*£'Z
                                                                               M
Figure 3.16 - Photographs Showing the Unrolling  (Upper)  and  Unfolding (Lower) of

            Geomembranes
                                   137

-------
      Some items recommended for a specification or CQA document include the following;

      1,  The soil subgrade shall be of the specified grading, moisture content and density as
         required by the installer and as approved by the CQA engineer for placement of the
         geomembrane. See Chapter 2 for these details for compacted clay liner subgrades.

      2,  Construction equipment deploying the rolls or pallets shall not deform or rut the soil
         subgrade excessively. Tirr or track deformations beneath the geomembrane should not
         be greater than 25 mm (1.0 in.) in depth.

      3.  The geomembrane shall not be deployed on frozen subgrade where ruts are greater than
         12 mm (0.5 in.) in depth.

      4.  When placing the geomembrane on another geosynthelic material (geotextilc, geonet,
         etc.). construction equipment should not be permitted to ride directly on the lower
         geosynthetic material.  In cases where rolls must be moved over previously placed
         geosynthetics it is necessary to move materials by hand or by using small pneumatic
         tired lifting units. Tire inflation pressures should be limited to a maximum value of 40
         kPa (6 Ib/in2),

      5,  Underlying geosynthetic materials (such as geotextiles or geonets) should have all folds,
         wrinkles and other undulations removed before placement of the geomembrane.

      6.  Care, and planning, should be taken to unroll or unfold the geomembrane close to its
         intended, and final, position.

3.3.4.2 Temperature Effects - Sticking/Cracking

      High temperatures can cause geomembrane surfaces on rolls, or accordion folded on
pallets, to stick together, a process commonly called "blocking". At the other extreme, low
temperatures can cause geomembrane sheets to crack when unrolled or unfolded.  Comments on
unrolling, or unfolding of geomembranes at each of these temperature extremes follow.
items.
For example, a specification or CQA document should have included in it the following


1. Geomembranes when unrolled or unfolded should not stick together to the extent where
   tearing, or visually observed straining of the geomembrane, occurs.  The upper
   temperature limit is very  specific to the particular type of geomembrane.  A sheet
   temperature of 50°C (122°F) is the upper limit that a geomembrane should be unrolled or
   unfolded unless it is shown otherwise to the satisfaction of the CQA engineer.

2. Geomembranes which have torn or have been excessively deformed should be rejected,
   or shall be repaired per the CQA Document

3. Geomembranes when unrolled or unfolded in cold weather should not Crack, craze, or
   distort in texture.   A sheet temperature of 0°C (32°F) is the lower limit that a
   geomembrane should be  unrolled or unfolded unless it is shown otherwise to  the
   sad sfaction of the CQA engineer.
                                         138

-------

3.3.4.3 Temperature Effects - Expansion/Contraction

      Polyethylene geomembranes expard when they are heated and contract when they are
cooled.  Other types of geomembranes may slightly contract when heated.  This expansion and
contraction must be considered when placing, seaming and backfilling geomerrtbranes in the field.
Fig. 3.17 shows a wrinkled polyethylene liner which has expanded due to thermal warming from
the sun.
             Figure 3.17 - HDPE Geomembrane Showing Sun Induced Wrinkles
      Either the contract plans and specifications, or the CQA documents should cover the
expansion/contraction situation on the basis of site specific and geomembrane specific conditions.
Some items to consider include the following:

      1. Sufficient slack shall be placed in the geomembrane to compensate for the coldesl
         temperatures envisioned so that no tensile stresses are generated in the geomembrane or
         in its seams either during installation or subsequently after the geomembrane is covered

      2. The geomembrane shall have adequate sLck such that it  do«s not lift up off of the
         subgrade or substrate material at any location within the facility, i.e., no "trampolining"
         of the geomembrane shall be allowed to occur at any time.
                                         139

-------
      3. The geomembrane shall not have excessive slack to the point where creases fold over
         upon themselves either during placement and seaming, or when the proactive soil or
         drainage materials are placed on the geomembranc.

      4. Permanent (fold-over type) creases in the covered geomembrane should not be permitted
         at any rime.

      5. The amount of slack to be added »**ie deployed and seamed geomembrane should be
         carefully considered and calculated, taking into account the type of geomembrane and the
         geomembrane's temperature during installation versus its final temperature in  the
         completed facility.

3.3.4,4 Spotting

      When a geomembrane roll or panel is deployed it is generally required that some shitting
will be necessary before field seaming begins.  This is called "spotting" by many installers.

      Some items for a specification or CQA document should include the following:

      I. Spotting of deployed geomembranes should be done with no disturbance to the soil
         subgradc or geosynthetie materials upon which they are placed.

      2. Spotting should be done with a minimum amount of dragging of the geomembrane on
         soil subgrades.

      3. Temporary tack  welding (usually with a hand  held hot air gun) of  all  types of
         thermoplastic geomembranes should be allowed at the installers discretion.

      4. When temporary tack welds of geomembranes are utilized,  the  welds should  not
         interfere with the primary seaming method, or with the ability to perform subsequent
         destructive seam tests.

3.3,4.5 Wind Considerations

      Wind damage to geomembranes, unfortunately, is not an uncommon occurrence, see Fig.
3.18.  Many deployed  geomembranes have been uplifted  by wind and have been damaged. In
some cases the geomembranes have even been torn out of anchor trenches. This is sometimes
referred to as "blow-out" by field personnel.  Generally,  but  not always,  the unseamed
geomembrane rolls or panels acting individually are most vulnerable to wind uojift and damage.

      The contract plans and specification, or at least the CQA documents, must be  very specific
as to resolutions regarding geomembranes that have been damaged due to shifting by wind.  Some
suggestions follow.

      1. Geomembrane rolls or panels which have been displaced by wind should be inspected
         and approved by the CQA engineer before any further field operations commence.

      2. Geomembrane rolls or panels which have been damaged (torn, punctured, or deformed
         excessively and permanently) shall be rejected and/or repaired as directed in the contract
         plans, specifications or CQA documents.

      3. Permanent crease marks, or severely folded (crimped) locations,  in geomembranes
                                         140

-------
         should not be permuted unless it can be shown that such distortions have no adverse
         effect on the properties of the geomembrane.  If this cannot be done, these areas should
         be cut out and properly patched as per the contract documents and approved by the CQA
         Engineer.

      4. If patching of wind damaged geomembranes becomes excessive (to the limit set forth in
         the specifications or CQA plan), Je entire roll or panel should be rejected.
                  Rgure 3.18 - Wind Damage to Deployed Geomembrane
3.4 Seaming and Joining

      The field seaming of the deployed geomembrane rolls or panels is a critical aspect of their
successful functioning as a barrier to liquid (and sometimes vapor) flow.  This section describes
the various seaming methods in current use, references a recently publbhe•: EPA Technical
Guidance Document on  seam fabrication techniques (EPA, 1991), and describe ; the concept and
importance of test strips (or trial seams).

3.4.1  Overview of Field Seaming Methods

      The fundamental mechanism of seaming  polymeric geomembrane sheets together is to
temporarily reorganize, i.e., melt, the polymer structure of the two surfaces to be joined in a
                                         141

-------
controlled manner that, after the application of pressure and after the passage of a certain amount of
time, results in the two sheets being bonded together. This reorganization results from an input of
energy that originates ftom either thermal or chemjgal processes. These processes may involve the
addition of extra polymer in the bonded area.

      Ideally, seaming two geomcmbrane sheets would result in no mt loss of tensile strength
across the two sheets and the joined sheets would perform as one single  gsomembrane sheet,
However, due to stress concentrations resulting from the seam geometry, current seaming
techniques may result in minor tensile strength loss relative to the parent geomembrane sheet Tht
characteristics of the seamed area are a function of the rype of geomembrane and the seeming
technique used. These characteristics, such as residual strength, geomembrane type, and seaming
type, should be recognized by the designer when applying the appropriate design factors-of-safety
for the overall geomembrane function and facility performance.

      It  should  be noted that the seam can be the  location of the lowest tensile strength in a
geomembrane liner. Designers and inspectors should be aware of the importance of seeking only
the highest quality geomembrane seams. The minimum seam tensile strengths (as determined by
design) for various geomembranes must be predetermined by laboratory testing, knowledge of past
field performance, manufacturers literature, various trade journals or other standards setting
organizations that maintain current information on seaming techniques and technologies.

      The methods of seaming at the time of the printing of this document and discussed herein
are given in Table 3.2 and shown schematically in Fig. 3.19.
            Table 3.2.  Fundamental Methods Of Joining Polymeric Geomembranes
             Thermal Processes
Chemical Processes
                • Fillet

                • Flat

              Fusion:

                • Hot Wedge

                • Hot Air
Chemkai:

  • Chemical Fusion

  • Bodied Chemical Fusion

Adhesive:

  • Chemical Adhesive

  • Contact Adhesive
      Within the entire group of thermoplastic geomembranes that will be discussed in this
manual, there are four general categories of seaming methods extrusion welding, thermal fusion or
melt bondjng. chemical fusion and adhesive; seaming. Each will be explained along with their
specific variations so as to give an overview of field seaming technology.
                                          142

-------
f
                                  Dual Hot wedge
                          (Single Track is Also Possible}
                                                                                 Rat-Type
                                                       (a) Extrusion Seams
      Single Hot Air
(Dual Track is Also Possible)
                                                       (b) Fusion Seams
                                                                                  Bodied Chemical
                                                       (c) Chemical Seams
                                                                               Contact Adhesive
                                                     (d) Adhesive Seams
                               Figure 3.19 - Various Methods Available to Fabricate Geomembrane Seams
                                                              143

-------
      Extrusion  welding  is  presently  used exclusively on  geomembranes made  from
polyethylene. A ribbon of molten polymer is extruded over the edge of, or in between, the two
surfaces to be joined.  The molten extrudate causes the surfaces of the sheets to become hot and
melt, after which the entire mass cools and bonds together.  The technique is called extrusion fillet
seaming when the extrudate is placed over the leading edge of the seam, and is called extrusion flat
seaming when the extrudate is placed between  the two sheets to be joined. It should be noted that
extrusion fillet seaming  is  essentially the only  practical  method for seaming polyethylene
geomembranc patches, for seaming in poorly accessible areas such as sump bottoms and around
pipes and for seaming of extremely short seam lengths. Temperature and seaming rate both play
important roles in obtaining an acceptable bond; excessive melting weakens the geomembrane and
inadequate melting results  in poor extrudate flow across the seam interface and low seam strength.
The polymer used for the extrudate is also very  important and should generally be the  same
polyethylene compound used to make the geomembrane. The designer should specify acceptable
extrusion compounds and how to evaluate them in the specifications and CQA documents.

      There are two thermal fusion or melt-bonding methods that can be used on all thermoplastic
geomembranes. In both of them, portions of the opposing surfaces are truly melted.  This  being
the case, temperature, pressure, and seaming rate all play important roles in that excessive melting
weakens the geomembrane and  inadequate melting results in low seam strength. The  hot wedge.
or hot shoe, method consists of  an electrically heated resistance element in  the shape of a wedge
that travels between the two  sheets to be seamed. As it melts the surface of the two sheets  being
seamed, a shear flow occurs across the upper and lower surfaces of the wedge. Roller pressure is
applied as the two sheets converge at the tip of the wedge to form the final seam. Hot wedge units
are controllable as far as temperature, amount of pressure applied and travel rate.  A standard hot
wedge creates a single uniform width seam, while a dual hot wedge (or "split" wedge) forns two
parallel seams with a uniform unbonded space between them. This space can be used  tc evaluate
seam  quality and continuity of the seam by pressurizing the unbonded space  with air and
monitoring any drop in pressure  that may signify a leak in the seam.

      The hot air method makes use of a device consisting of a resistance  heater, a blower, and
temperature controls  to  force hot air between  two sheets to  melt the opposing surfaces.
Immediately following the melting of the surfaces, pressure is applied to the seamed area to bond
the two sheets.  As with the hot wedge method, both single and dual seams can be produced. In
selected situations, this technique may also be used to temporarily "tack" weld two sheets together
until the final seam or weld is made and accepted.

      Regarding the chemical  fusion seam types; chemical fusion seams make use of a liquid
chemical applied between the two geomembrane sheets to be joined. After a  few seconds, required
to soften the surface, pressure is applied to make complete contact and bond the sheets together.
As with any of the chemical seaming processes to be described,  the two adjacent materials to be
bonded are transformed into a viscous phase. Care must be used to see that the proper amount of
chemical is applied in  order to  achieve the desired results.  Bodied chemical fusion seams are
similar to chemical fusion  seams except that 1% to 20% of the parent lining resin or compound is
dissolved in the chemical and then is used to make the seam. The purpose of adding the resin or
compound is to increase the viscosity of the liquid for slope work and/or adjust the evaporation rate
of the chemical. This viscous liquid is applied between the two opposing surfaces to  be bonded.
After a few seconds, pressure is applied to make complete contact. Chemical adhesive seams make
use of a dissolved bonding agent (an adherent) in the chemical or bodied chemical which is left
after the seam has been completed and cured. The adherent thus become1, an additional element in
the system.  Contact adhesives are applied to both mating surfaces.  After reaching the proper
degree of tackiness, the two  sheets are placed on top of one another, followed by application of
roller pressure. The adhesive forms the bond and is an additional element in the system.
                                          144

-------
      Other emerging seaming methods use ultrasonic, electrical conduction and magnetic
induction energy sources. Since these methods are in the developmental stage, they will not be
described further in this document See EPA (1991) for further details.

      In order to gain an overview as to which seaming methods are used for the various
thermoplastic geomembrancs described in this document, Table 3.3 is offered. It is generalized,
but it is used to introduce the primary seaming methods versus the type of geomembrane that is
customai Jy seamed by that method.
  Table 3.3  Possible Field Seaming Methods for Various Geomembranes Listed in this Manual
   Type of Seaming
      Method
Type of Geomembrane
                        HDPH     VLDPE   Other PE     PVC     CSPE-R  Other Flexible
extrusion
(fillet and flat)
thermal fusion
(hot wedge and
hot air)
chemical
{chemical and
bodied chemical)
adhesive
(chemical and
contact)
A

A


n/a


n/a


A

A


n/a


n/a


An/a n/a

A A A


n/a A A


n/a A A


A

A


A


A


Note:  A  » method is applicable
      n/a  « method is "not applicable"
3.42 Details of Field Seaming Methods.

      Full details of field seaming methods for the edges and ends of geomembrane rolls or
panels has recently been described in EPA Technical Guidance Document, EPA/530/SW-91/051,
entitled: "Inspection Techniques for the Fabrication of Geomembrane Seams".  In this document
(EPA, 1991) are separate chapters devoted to the following field seaming methods.

      * extrusion fillet seams
                                         145

-------
      * extrusion flat seams

      * hot wedge seams

      • hot air seams

      * chemical and bodied chemical fused seams

      * chemical adhesive seams

There is also a section on emerging technologies for geomemfarane seaming. The interested reader
should consult this document for details regarding all of these seaming methods.

      Whenever the plans and specifications are not written around a particular seaming method
the actual method which is used becomes a matter of choice for the installation contractor.  As seen
in Table 3.3, there are a number of available choices for each geomembrane type.  Furthermore,
even when the installation contractor selects the particular seaming method to be used, its specific
details are rarely stipulated even in the specification or CQA documents. This is to give the
installation contractor complete latitude in selecting seaming temperatures, travel rates, mechanical
roller pressures, chemical type, tack time, hand rolling pressure, e;c.  The role of  the plans,
specifications and CQA documents is to adequately provide for destructive tests (on test strips and
on production seams) and nondestructive tests (on production seams) to assure that the seams are
fabricated to the highest quality and uniformity and are in compliance with the project's documents.

      This is not to say that the specification never influences the type of seaming method. For
example, if the specifications call for a nondestructive constant air pressure test to be conducted,
the installation contractor must use a thermal fusion technique like the dual hot wedge or dual hot
.Jr methods since they are the only methods that can produce such a seam.

3.4.3  Tes| Strips, and Trial Seams

      Test strips and trial seams, also called qualifying seams, are considered to be an important
aspect of CQC/CQA procedures.  They are meant  to serve  as a prequaJifying experience for,
personnel, equipment and procedures for making seams on the identical geomembrane material
under die same climatic conditions as the actual field production seams  will be made. The test
strips are usually made on two narrow pieces of excess geomembrane varying in length between
1.0 to 3.0 m (3 to  10 ft.), see Fig. 3.20, The test strips  should be made in sufficient lengths,
preferably as a single continuous seam, for ail required testing purposes.

      The goal of these test strips is to reproduce all aspects of the actual-production field seaming
activities intended to be performed in the immediately upcoming work session so as to determine
equipment and operator proficiency. Ideally, test strips can be used to estimate the quality of the
production  seams while  minimizing damage to the installed  geomembrane through destructive
mechanical testing. Test strips are typically made every 4 hours (for example, at the beginning of
the work shift and after the lunch break). They are also made whenever personnel or equipment
are changed and  when climatic conditions reflect wide changes in geomembrane temperature or
when other conditions occur that could affect seam quality. These details should be stipulated in
the contract specifications or CQA documents.

      The destructive testing of the test strips should be done as soon as the installation contractor
feels that the strength requirements of the contract specification or CQA documents can be met.
Thus it behooves the contractor to have all aspects ot the test strip seam fabrication in complete
                                          146

-------

working order just as would be done in the case of fabricating production field seams. For
extrusion and thermal fusion seams, destructive testing can be done as soon as the seam cools. For
chemical fusion and adhesive seams this could  take several days and the use of a field oven to
accelerate the curing of the seam is advisable.

                    Figure 3.20 - Fabrication of a Geomembrane Test Strip
       From two to six test specimens are cut from the test strip using a 25 mm (1.0 in. wide die).
They are selected at random by the CQA inspector. The specimens are then tested in both peel and
shear using a field tensiometer, see Fig. 3.21.  (Generally peel tests are more informative in
assessing the quality of the seam). If any of the test specimens fail, a new test strip is fabricated.
If additional specimens fail, the seaming apparatus and seamer should not be accepted and should
not be used for seaming until the deficiencies are corrected and successful trial welds are achieved
The CQA inspector should observe all trial seam procedures and tests.  If the specimens pass,
seaming operations can move directly to production seams in the field. Pass/fail criteria for
destructive seam tests wil' be described in Section 3.5.
                                           147

-------
    Figure 3 21 - Photograph of a Field Tensiometer Performing a Geomembrane Seam Test
      The flow chart illustrated in Fig. 3.22 gives an idea of the various decisions that can be
reached depending upon the outcome of destructive tests on test strip specimens. Here it is seen
that failed test strips are linked to an  increased frequency  of destructive tests to be taken on
production field seams made during the time interval between making the test strip and  its testing.
Furthermore, it is seen that there  are only two chances at  making adequate test strips before
production field seaming is stopped and repairs are initiated. These details should  be covered in
either the project specification or the CQA documents,

      Some specification or CQA document nems regarding the fabrication ol" geomembrane seam
test snips include  the following:

      1. The frequency of making test strips should be clearly stated.  Typically this is at the
        beginning of the day, after  the  noon break and whenever changed conditions are
        encountered, e.g., changes in  weather, equipment, personnel.

      2. The CQA Engineer should have the option of requesting test strips of any field seaming
        crew or device at any time.
                                          14K

-------
       Uake TMI Strip 1
                            Uake Production
                             Field Seam*
      Make Teat Strip 2
             I
Take Destructive Samplee
From Production FleM
                                      Seama
    Increase Frequency of
    Destructive Sampling
                No
    Hatt Production fWd
    Seaming and Repair par
    CQA/COC Documents to
    Point of Previous
    Acceptance wMi
    Approved Seaming
    Crew and/or Equtpmenr
                                                           Continue Production
                                                             FMd Seeming
                                           1
                • Ncti: Seaming Crew Falling to
                  Prepare Acceptable Teat Stripe
                  Hay Require Retraining m
                  Accordance witti COC/CQA
                  Document*

                   Figure 3.22 - Test Strip Process Flow Chan
3. The procedure for sampling and evaluating the field test strip samples chouicl be ckariy
   outlined, i.e., the number of peei and sliezr test specimens to be cut and tested frotr the
   test strip sample, the rate of testing and what the required strength values are in these
   two different modes of testing.

4. The fabrication of the field test strip and testing of test specimens should be observed by
   the CQA personnel.
                                       149

-------
      5, The time for testing after the test snip is fabricated varies between seam types. For
         extrusion and fusion fabricated seams, the testing can commence immediately after the
         polymer cools to ambient temperature.  For chemical fusion and adhesive fabricated
         seams, the testing mu; t wait until adequate curing of the seam occurs. This can take as
         long as  1 to 7 days. During this time all production seaming must be tracked and
         documented.

      o. Accelerated oven curing of chemical and adhesive fabricated seams is acceptable so as to
         hasten the curing process and obtain test results as soon as possible.  GR1 Test Method
         GM-7 can be used for this purpose.

      7. The required inspection protocol and implications of failed test specimens from the test
         strips must be clearly stated. The protocol outlined in Fig. 3.22 is suggested

      8. Field test strips are usually discarded after the destructive test specimens are removed
         and tested.  If this is not the case, it should be clearly indicated who receives the test
         strip samples and what should be the utilization (if any) of these samples.

3.5 Destructive Test Methods for Seams

      The major reason that plans and specifications do not have to be specific about the type of
seaming methods and their particular details is that gee-membrane seams can be readily evaluated
for their quality by taking samples and destructively testing them either at the job site or in a timely
manner at a testing laboratory thereafter.

3.5.1  Overview

      By destructivel v testing geomembrane seams it is mrant to actually cut out (i.e., to sample)
and remove a portion of the completed production seam, and then to further cut the sample into
appropriately sized test specimens.  These specimens are then tested according to a specified
procedure to failure or to yield depending upon the type of geomembrane.

      A possible procedure is to select the sampling location and cut two closely spaced 25 mm
(1.0 in.) wide test specimens from the seam. The distance between  these two test specimens is
defined later.  The individual specimens arc then tested in a peel mode using a field lensiormter
(recall Fig. 3.21).  If the results are acceptable, the complete seam between  the two field test
specimens is removed and properly identified and distributed. If either test specimen fails, two
new locations  on either side of the failed specimen(s) are selected until acceptable seams are
located.  The seam distance between acceptable seams is usually repaired by cap-stripping but other
techniques are also possible. The exact procedure must be stipulated in the-specifications or CQA
document

      The length dimension of  the field scan? sample between the two  test specimens just
described varies according to whatever is  stipulated in the plans and specifications, or in
accordance with the CQA documents. Some common options are to sample the seam for a distance
of either 36 cm (14 in.), 71 cm (28 in.)  or 106 cm (42 in.) along its length.  Since the usual
destructive seam tests  are either shear or peel tests and both types are 25 mm  (1.0 in.) wide test
specimens, this allows for approximately 10, 20 or 30 tests (half shear and half peel) to  be
conducted on the respective lengths cited  above. The sample width perpendicular to the seam is
usually 30 cm (12 in.) with the seam being centrally located within this dimension.
                                          ISO

-------
      The options of seam sample length between the two peel test specimens mentioned above
that are seen in various plans, specifications, and CQA documents, are as follows:

      •  A 36 cm (14 in.) sample is taken from the seam  and cut  into 5 shear and 5  peel
         specimens.  The tests are conducted in the field or at a remote laboratory by, or under the
         direction of, the responsible CQA organization.

      •  A 71 cm (28 in.) long sample is .aken from the seam and cut in half. One half is further
         cut into 5 shear and 5 peel test specimens which are tested in the field or at a remote
         laboratory by the CQC organization (usually the installation contractor). The other half is
         sent to a remote laboratory for testing by the CQA organization who also does 5 shear
         and 5 peel tests. Alternatively, sometimes only the CQA organization does the testing
         and the second half of the sample is left intact and archived by the owner/operator.

      •  A 106 cm (42 in.) long sample is taken from the seam and cut into thre-, individual 36
         cm  (14 in.) samples.  Individual samples go to the CQC organization,  the CQA
         organization and the owner/operator.  The CQC and  CQA organizations each cut their
         respective samples into 5 shear and 5 peel test specimens and conduct the appropriate
         tests immediately. The remaining sample is archived by the owner/operator.

      Whatever is the  strategy for taking samples from the production seams for destructive
testing it must be clearly outlined in the contract plans and specifications and further defined and/or
corroborated in the CQA documents.

      Obviously, the hole created in the production seam from which the test sample  was
originally taken must be patched  in an appropriate manner.  See Fig. 3.23 for such  a  patched
sampling location. Recognize that the seams of such patches are themselves candidates for field
sampling and testing.  If this is done, one would have the end result of patch on a patch,  which is a
rather unsightly and undesirable condition.

3.5.2  Sampling Strategies

      The sampling  of production seams of installed geomembranes  represents a dilemma of
major proportions. Too few samples results in a poor statistical representation of the strength of
the seam, and too many samples requires an additional cost and a risk of having the  necessary
repair patches being problems in themselves.  Unfortunately, there is no clear strategy for all cases,
but the following are some of the choices that one has in formulating a specification or CQA plan.

      Note also that in  selecting a sampling strategy the sampling frequency is tied directly into
the performance of the test strips described in Section 3.4.3.  If the test strips fail during the time
that production seaming is ongoing, the frequency of destructive sampling and testing must be
increased. The following strategies, however, are for situations where geomembrane seam test
strips are being made in an acceptable manner.

3.5.2.1 Fixed Jncrement Sampling

      By far the most commonly used sampling  strategy is the "fixed  increment sampling"
method.  In this method, a seam sample is taken at fixed increments along the total length of the
seams. Increments usually range  from 75 to 225 m (250 to 750 ft) with a commonly specified
value  being one destructive test sample every 150 m (500 ft). Note that this value can be applied
either directly to the record drawings during layout of the seams, to each seaming crew as  they
progress during the work period, or to each individual seaming device.  Once the increment is
5V«-
«'»».
                                          151

-------
decided upon, it should be held regardless of the location upon which it falls, e.g., along side
slopes, in sumps, etc.  Of course, if the CQA documents allow otherwise, exceptions such as
avoiding sumps, connections, protrusions, etc, can be made.
Figure 3,23 - Completed Patch on a Geomembrane Seam Which had Previously Been Sampled
             for Destructive Tests
3.5.2.2 Randomly Selected Sampling

      In random selection of destructive seam sample locations it is first necessary to preselect a
preliminary estimate of the total number of samples to be taken.  This is done by taking the total
seam length of the facility and dividing it by an arbitrary interval, e.g., ISO m (500 ft), to obtain
the total number of samples that are required. Two choices to define the actual sampling locations
                                           152

-------

are now available: "stratified" random sampling, or "strict" random sampling.  The stratified
method takes each pre-selscted interval (e.g., a ISO m (500 ft) length) and randomly selects a
single sample location within this interval. Thus with stratified random sampling one has location
variability within a fixed increment (unlike fixed frequency sampling which is always* at the exact
end of the increment). The strict method uses the total seam length of the facility (or cell) and
randomly selects sample locations throughout the facility up to the desired number of samples.
Thus with strict random sampling a group of samples may be taken in close proximity to one
another, which necessarily leaves other areas with sparse Campling.

      There are various ways of randomly selecting the specific location within an interval, e.g.,
in a specific region of great concern, or within the total project seam length.  These are as follows:

      *  Use a random number generator from statistical  tables  to predetermine the sampling
         locations within each interval or for the entire project,

      *  Use a programmable pocket calculator with a  random number generator program to
         select the sampling location in the field for each interval or for the entire project.

      •  Use a random number obtained by simply multiplying two large numbers together to
         form an 8-digit result. A pocket calculator with an adequate register will be necessary.
         The center two digits in such a procedure are quite randomly distributed and can be used
         to obtain the sampling  location.  For example,  multiplication  of the following two
         numbers "4567" by 4567" gives 2085J489 where the central two digits, i.e., the "57",
         are used to select the location within the designated sampling interval.  If this  interval
         were 500 ft, the sampling location within it would be at 0.57 * 500 = 285 ft. from the
         beginning of the interval.  The next location of the sample  would require a new
         calculation resulting in a different central two-digit number  somewhere within the next
         500 ft sampling interval and would be located in a • imilar fashion.

3.3.2.3 Other Sampling Strategies

      There are two other sampling strategies wnich might be selected in determining how many
destructive seam samples should be taken. Both are variable strategies in that repeated acceptable
seam  tests are rewarded  by requiring fewer samples and repeated failures are penalized by
requiring more frequent samples. These two strategies are called the "merhod of attributes" and the
use of "control charts".  Both set  upper and lower bounds which require either fewer or more
frequent testing than the initially  prescribed sampling frequency.  Each  of these methods are
described fully in Richardson (1992).

      Whatever the sampling strategy used, it should never limit or prohibit the ability to select a
destructive seam sample from a suspect area. This should ultimately be an option left to the CQA
engineer.

3.5.3  Shear Testing of Geomembrane Seams

      Shear testing of specimens  taken from field fabricated geomembrane seams represents a
reasonably  simulated performance test.  The possible exception is that a normal  stress  is  not
applied to the surfaces of the test specimen thus it is an "uneonfined" tension test. A slight rotation
may be induced during tensioning of the specimen, making the actual test results tend toward
conservative values. The configuration of a shear test in a tension testing machine is shown in Fig.
3.24.

I
                                          153

-------
 Figure 3.24 - Shear Test of a Geomembrane Seam Evaluated in a CQC/CQA Laborawy
              Environment
      Commonly recommended shear tests for HOPE, PVC, CSPE-R and EIA-R seams, along
with the methods of testing the unseamed sheet material in tension, are given in Table 3.4.  The
VLDPE data presented was included in a way so as to parallel the HOPE testing protocol except for
the strain rate values which are faster since breaking values, rather than yield values are required.
There is no pronounced yield value when tensile testing VLDPE geomembranes.
                                                                                                ,
                                                                                               6?
                                                                                               fcr
                                                                                               r
                                         154

-------
       Table 3.4  Recommended Test Method Details for Georocmbrane Seams in Shear and in Peel and for Unseamed Sheet
Ut
ut
Type of Test
Shear Test on Seams
ASTM Test Method
Specimen Shape
Specimen Width (in.)
Specimen Length (in.)
Gage Length (in.)
Strain Rate (ipm)
Strength (psi) or (ppi)
Peel Test on Scams
ASTM Test Method
Specimen Shape
Specimen Width (in.)
Specimen Length (in.)
Gage Length (in.)
Strain Rate (ipra)
Strength (psi) or (ppi)
Tensile Test on Sheet
AST^t Test Method
Specimen Shape
Specimen Width (in.)
Specimen Length (in.)
Gage Length (in.)
Strain Rate (ipm)
Strength (psi) or (Ib)
Strain (in./m.)
Modulus (psi)
where n/a «• not applicable
t *= geontembrane
psi = pounds/square
HOPE

D4437
Strip
1.00
6,00 + team
4.00 + seam
2.0
Fbrce/(l.tt)xi)

D4437
Slrip
1.00
400
n/a
20
Foree/(1.00xi)

D638
Dumbbell
t' 0.25
4.50
1.30
2.0
Force/(0.25xt)
EkmgVlJO
Prom Graph

thickness
inch of specimen cross section
VLDPE

D4437
Strip
1.00
6.00 + seam
4.00+ seam
20
ForccAlXJOxi)

D4437
Strip
LOO
4.00
n/a
20
Fofc«AJ.OO>a)

D638
DumbbeU
0.25
4.50
1.30
20
ForceA0.25xt)
Elons71.30
From Graph



PVC

D3083
Strip
1.00
6 .00 + seam
4,00+ seam
20
FonxAl.OOxi)

D413
Strip
1.00
4.00
nA
2A
Force/! .00

D882
Strip
1.00
6.00
2.00
20
ForccAl.OOxt)
Elong/2.00
From Graph



CSPER

D751
Grab
4.00 (1.00 grab)
9.00 + seam
6.00+ seam
12
Force

D413
Strip
1.00
4.00
•tfa
2.0
Force/1.00

D^Sl
Grab
4.00 (1.00 Grab)
6.00
3.00
12
Force
Ekmg/3.00
n/a



ppi = pounds/linear inch width of specimen
ipm » inchssAninute




Force » maximum force attained at specimen failure (yield or break)

-------
      Insofar as the shear testing of nonreinforced geomembrane seams (HDPE, VLDPE and
PVC), all use a 25 mm (1.0 in.) wide test specimen with the seam being centrally located within
the testing grips. For the reinforced geomembranes (CSPE-R and EIA-R) a "grab" test specimen
is used. In a grab tension test the specimen is 200 mm (4.0 in.) wide but is only gripped in the
central 25 mm (1.0 in.). The test specimen is tcnsioned, at its appropriate strain rate, until failurr
occurs.  If the seam delaminat-s (i.e., pulls apart in a seam separation mode), the seam fails -i
what is called  a "non-film tear bond", or non-FTB. In this case, it is rejected as a failed sear.
Details on various types of seam failures and on the interpretation of Fl'B are found in Hax-.
(1988).  Conversely, if the seam does not delaminate, but fails in the adjacent sheet material o
either side of the seam, it is an acceptable failure mode, i.e,, called a "film tear bond", or FTB, anrf
the seam strength is then calculated.

      The seam strength (for HDPE, VLDPE and PVQ is the maximum force attained divided by
either the original specimen width (resulting in units of force per  unit  width), or the original
specimen cross sectional area (resulting in units of stress). It is general procedure to use force per
unit width as it is an absolute strength value which can be readily compared to other test results.  If
stress units are desired, one can use the nominal thickness of the geomembrane, or continuously
measure the actual thickness of each test specimen. This latter alternative requires considerable
time and effort and is generally not recommended. The procedure is slightly different for the
reinforced geomembranes  (CSPE-R and EIA-R) which use a grab test method. Here the strength
is based on the maximum tensile force that can be mobilized and a stress value is not calculated.

      The resulting value of seam shear strength is then compared to the required seam strength
(which is the usual case) or to the strength of the unseamed geomembrane sheet. If the latter, the
procedures for obtaining this value are listed in Table 3.4. In each case the test protocol for seam
and sheet are  the same, except for HDPE anH  VLDPE. The sheet strength value for these
polyethylene geomembranes are based on a ASTM p-638 "dumbbell-shaped" specimens, although
the strength  is calculated on  the reduced section width. With all of these sheet tension tests, the
nominal thickness of the unseamed geomembrane sheet is used for the comparison value.  If actual
thickness of the sheet is considered, the results will be reflected accordingly.  Note, however, that
this will require a large amount of additional testing (to get average strength values) and is not a
recommended  approach.

      Knowing the seam shear strength and the unseamed sheet strength (ether by a specified
value or by testing), allows for a seam shear efficiency calculation to be made as follows;
where
              "shear
 seam in shear
r
 unseamed sheet
                                    (100)
                                                                                   (3.1)
               = seam efficiency in shear (%)

               - seam shear strength (force or stress units)

               = sheet tensile strength (force or stress units)

      The contract plans, specifications or CQA documents should give the minimum allowable
seam shear strength efficiency.  As a minimum, the guidance listed below can be used whereby
                                           156

-------
percentages of seam shear efficiencies (or values) are listed:

      HDPE = 95% of specified minimum yield strength
      VLDPE -  typically 1200 lb/in2
      PVC = 80%
      CSPE-R =  80% (for 3-ply reinforced)
      EIA-R - 80%

      Generally an addirional requirement of a film tear bond, or FTB, will also be required in
addition to a minimum strength value.  This means that the failure must be located in the sheet
material on either side of the sram and not  within the  seam itself.  Thus  the seam cannot
delaminate.

      Lastly, the number of failures allowed per number of tests conducted should be addressed.
If sets of 5 test specimens are performed for each field sample, many specifications allow for one
failure out of the five tested. If the failure number is larger, then the plans, specifications or CQA
documents must be clear on the implications.

      When a destructive seam test sample fails, many specifications and CQA documents require
two additional samples to be taken, one on each  side of the original sample each spaced 3 m (10 ft)
from it  If either one of these samples fail, the  iterative process of sampling every 3 m (10 ft) is
repeated until passing test results are observed.  In this case the entire seam  between the two
successful test samples must be questioned.  For  example,  remedies for polyethylene
geomembranes are to cap strip the entire seam or if the seam is made with a thermal fusion method
(hot air or hot wedge) to extrude a fillet weld over the outer seam edge.  When such  repairs are
concluded the seams on the cap strip or extrusion fillet weld should be sampled and tested as just
described

      Note that elongation of the specimens during shear testing is usually not  monitored
(although current testing trends are in this direction), the only value under consideration is the
maximum force that the seam can sustain.  It should also be mentioned that the test is difficult to
perform on the inside of the tracks facing the air channel of a dual channel thermal fusion seam.
For small air channels the tab available for gripping will be considerably less than that required in
test methods as given in Table 3.4.  Regarding the testing of the inside or outside tracks (away
from the air channel) of a dual channel thermal  fusion seam, or even both tracks, the specification
or CQA document should be very specific.

3.5.4  Peel Testing of Georrumbrane Seams

      Peel testing of specimens taken  from field fabricated geomembrane seams represent a
quality control type of index test. Such tests are not meant to simulate in-situ performance but are
very important indicators of the overall quality  of the seam. The configuration  of a peel test in a
tension testing machine is shown in Fig. 3.25.

      The recommended peel tests for HDPE, PVC, CSPE-R and EIA-R seams, along with the
unseamed sheet material in tension are given in Table 3.4. The VLDPE data was included in a way
so as to parallel the HDPE testing protocol.

      Insofar as the peel testing of geomembrane seams is concerned, it is seen that all of the
geomembranes listed have a 25 mm (1.0 in.) width test specimen.  Furthermore, the specimen
lengths and strain rate are also equal for all geomembrane types.  The only difference is that HDPE
and VLDPE use the thickness of the geomembrane to calculate a tensile strength value in stress
                                          157

-------
units, whereas PVC, CSPE-R and EIA-R calculate the tensile strength value in units of force per
unit width, i.e., in units of pounds per linear inch of seam.
Fig. 3.25 - Peel Test of a Geomembrane Seam Evaluated in a CQC/CQA Laboratory Environment
      In a peel test the test specimen is tensioned, at its appropriate strain rate, until failure occurs.
If the seam delaminates (i.e., pulls apart in a seam separation mode), it is called a "non-film tear
bond or non-FTB", and is recorded accordingly.  Conversely, if the seam does not delaminate, but
fails in the adjacent sheet material on either side of the seam it is called a "film tear bond or FTB"
and the seam strength  is calculated.  Details  on various  types of seam failures and on  the
interpretation of FTB are found in Haxo (1988). The seam strength is the maximum force attained
divided by the specimen width (resulting in units  of force per unit width), or by the specimen cross
sectional area (resulting in units of stress). The  former procedure is the most common, i.e., peel
strengths are measured in force per unit width units. If stress units are desired the thickness of the
                                           158

-------
geomembrane sheet must be included. The nominal sheet thickness is usually used. If the actual
sheet thickness is used, a large amount of thickness measurements will be required to obtain a
statistically reliable value. It is not a recommended procedure.

      The resulting value of seam peel  strength is then compared to a specified value (the usual
case) or to the strength of the unseamed geomembrane sheet. The testing procedures for obtaining
these values are listed in Table 3.4. It can be seen, however, that only with PVC is tVe same width
test specimen used for peel and sheet testing.  For HDPE and VLDPE one is comparing a 1.0 in.
uniform width peel test with a dumbbell shaped specimen, while for CSPE-R and EIA-R one is
comparing a  uniform  width peel test with the strength from a grab shaped test specimen.  If,
however, one does have a specified sheet strength value or a measured value, a seam peel strength
efficiency calculation can be made as follows:
              -peel
-    seam in peel
   r
    unseamed sheet
                                   (100)
                                                                                   (3.2)
where
             = seam efficiency in peel (%)
             = seam peel strength (force or stress units)
             = sheet tensile strength (force or stress units)

      The contract plans, specifications or CQA documents should give the minimum allowable
seam peel strength efficiency.  As a minimum, the guidance listed below can be used whereby
percentage peel efficiencies (or values) are listed as follows:

      HDPE    = 62% of specified minimum yield strength and FTB
      VLDPE   = typically 1000 lbAn2
      PVC      = 10 Ib/in.
      CSPE-R  = 10 Ib/in. or FTB
      EIA-R    = 10 Ib/in.

Lastly, the number of failures allowed per number of tests conducted should be addressed. If sets
of 5 test specimens are performed for each field sample, many specifications allow for one failure
out of the five tested. If the failure number is larger, then the plans, specifications  or CQA
documents must be clear on the implications.

      When a destructive seam test sample fails, many specifications require an additional two
samples to be taken, one on each side of the original spaced 3 m (10 ft) from it.  If either one of
these samples fail the  iterative process of sampling every 3 m (10 ft) is repeated  until successful
samples result.  In this case, the entire seam between the last successful test samples must be
questioned.  Remedies are to cap strip the entire seam or if the seam  is HDPE or VLDPE made
with a thermal fusion  method (hot air or hot wedge) to extrude a fillet weld over the outer seam
edge. When this is done the seams on the cap strip or extrusion fillet weld may  be sampled and
tested as just described.

       Note that neither elongation of the specimen nor peel separation, during the test is usually
monitored (although current testing trends are in this direction), the only value under consideration
is the maximum tensile force that the seam can sustain.  It should also be mentioned that both
frontward and backward peel tests can be performed thereby challenging both sides of a seam.  For
                                          159

-------
dual channel seams, both insides of the tracks facing the air channel can be tested, but due to the
narrow width of most air channels the tab available for gripping will be considerably less than that
given in Table 3.4.  Regarding the testing of the inside or outside cracks (away from  the air
channel) of a dual channel seam, or even both tracks, the specification or CQA document should be
very specific.

3-5-5  General Specification fterns

      Regarding field sampling of geomembrane seams and their subsequent destructive testing, a
specification or CQA document should consider the following items.

      1.  CQA personnel should observe all production seam sample cutting.

      2.  All samples should be adequately numbered and marked with permanent identification.

      3.  All sample locations should be indicated on the geomembrane layout (and record)
          drawings.

      4.  The reason for  taking the sample  should be indicated, e.g., statistical routine,
          suspicious feature, change in sheet temperature, etc.

      5.  The sample dimensions should be given insofar as the length of sample and its width.
          The seam will generally be located along the center of the length of the sample.

      6.  The distribution of various portions of the  sample (if more than one) should be
          specified.

      7.  The number of shear and peel tests to  be conducted on each sample (field  tests and
          laboratory tests) should be specified

      8.  The specifics of conducting the shear and peel tests should be specified, e,g.» use of
          actual sheet thickness, or of nominal sheet thickness. The following are suggested
          ASTM test methods for each geomembrane type:
Geofncmbranc
HDPE
VLDPE
PVC
CSPE-R
EIA-R
Seam Shear Test
D-4437
D-4437
D-3083
D-751
D-751
Searn Peel Test
D-4437
D-4437 x
D-413
D-413
D-751
Sheet Test
D-638
D-638
D-882
D-751
D-751
s
5"
it.
I
                                                                                                    Frf,
      9.  The CQA personnel should witness all field tests and see that proper identification and
          details accompany the test results. Details should be provided in the CQA documents.
          Such details as follows are often required.

                                                                                                    F3K,
                                          160

-------
                                                                                                it
                                                                                                !"

                                                                                                I'
          *   date and time

          *   ambient temperature

          •   identification of seaming unit, group or machine

          *   name of master seamer

          •   welding apparatus temperature and pressure, or chemical type and mixture

          •   pass or fail description

          *   a copy of the report should be attached to the remaining portion of die sample

       10. The CQA personnel should verify that samples sent to the testing laboratory are
          properly marked, packaged and shipped so as not to cause damage.

       11. Results of the laboratory tests should come to the CQA Engineer in a stipulated time.
          For extrusion and thermally bonded seams, verbal test results are sometimes required
          with 24 to 72 hours after the laboratory receives the samples. For chemically bonded
          seams, the rime frame is longer and depends on whether or not accelerated heat curing
          of the seams is required.  In all  cases, the CQA Engineer must inform the Owner's
          representative of the results and make appropriate recommendations.

       12. The procedures for seam remediation in the event of failed destructive tests should be
          clear and unequivocal.  Options usually are (a) to repair the entire seam between
          acceptable sampling locations, or (b) to retest the seam on both sides in the vicinity of
          the failed sample. If they are acceptable only this section of the seam is repaired. If
          they are not, a wider spaced set of samples are taken and tested.

       13. Repairs to locations where destructive samples were removed should be stipulated.
          These repairs are specific to the type of geomembrane and to the seaming method.
          Guidance in this regard is available in EPA (1991).

       14. Each repair of a patched seam  where a test sample had been removed should be
          verified. This is usually done by an appropriate nondestructive test.  If, however, the
          sampling strategy selected calls for a destructive test to be made at the exact location of
          a patch it should be accommodated. Thus the final situation will require a patch to be
          placed on an earlier patch.  If this (unsightly) detail is to*fee avoided, it should be stated
          outright in the specifications or CQA document.

       15. The time required to retain and store destructive test samples on the part of the CQC
          and CQA organizations should be stipulated.

3.6 Nondestructive Test Methods for Seams

3.6.1 Qvervjew

      Although it is obviously important to conduct destructive tests on the fabricated scams, such
tests do not  give adequate information on the continuity and completeness of the entire seam
between sampling locations. It does little good if one section of a seam meets the specification
requirements, only to have the section  next to it missed completely  by the field-seaming crew.
 !R
 $•
 I
I

I
                                         161
                                                                                              ft

-------
Thus cominuoi s methods of & nondestructive testing (NDT) nature will be discussed here. In each
of these methods the gosl is to validate 100% of the seams or, at minimum, a major percentage of
them,

3.6.2  Currently Available Methods
           currently available NDT methods for evaluating the adequacy of geomembrane field
seams are listed in Table 3.5 in the order that they will be discussed.

      The air lance  method uses a jet of air at approximately 350 kPa  (50 lb/in.2) pressure
coming through an orifice of 5 mm (3/16 in.) diameter. It is directed beneath the upper edge of the
overlapped seam and is held within 1 00 mm (4.0 in.) from the edge of the seamed area in order to
detect unbonded areas. When such an area is located, the air passes through the opening in the
seam causing an inflation and fluttering in the localized area. A distinct change in sound emitted
can generally be heard. The method works best on relatively thin, less than 1.1 mm (45 mils),
flexible geomembranes, but works only if the defect is open at the front edge of the seam, where
the air jet  is directed.  It  is essentially a geomembrane installer's  method to  be used in  a
construction quality control (CQC) manner.

      The mechanical point stress or "pick" test uses a dull tool, such as a blunt screw-driver,
under the top edge of a seam. With care, an individual can detect an unbonded area, which would
be easier to separate  than a properly bonded  area.  It is a rapid test that obviously depends
completely on the care and sensitivity of the person doing it. Detectability is similar to that of using
the air  lance, but both are very operator-dependent.  This test is to be performed only by the
geomembrane installer as a CQC method.  Design or inspection engineers should not use the pick
test but rather one or more of the techniques to be discussed later.

      The pressurized dual seam method was mentioned earlier in connection with the dual hot
wedge or dual hot air thermal seaming methods.  The air channel that results between the dual
bonded tracks is inflated using a  hypodermic needle and pressurize! to approximately 200 kPa (30
lb/in.2 ).  There is no limit as to- the length of the seam that is tested. If the pressure drop is within
an allowable amount in the designated time period (usually 5 minutes), the scam is acceptable; if a
unacceptable drop occurs, a number of actions can be taken:

      *  The distance can be systematically halved until the leak is located.

      «  The section can be tested by some other leak detection method.

      •  An extrusion  fillet weld can be placed over the entire edge.

      •  A cap strip can be seamed over the entire edge.

Details of the test can  be found in GRI Test Method GM6.  The test is an excellent one for long,
straight-seam lengths.  It is generally performed by the installation contractor, but usually with
CQA personnel viewing the procedure and documenting the results.


                                           162

-------
Table 3.5 - Nondestructive Gcomembrane Seam Testing Methods, Modified fiom Richardson and Koenier (1988)
Nondestructive
Test Method
1. air lance
2. mechanical
point (pick)
stress
3. dual seam
(positive
pressure)
4, vacuum
chamber
(negative
pressure)
5, elccincwire
6. electric field

7. ultrasonic
(Hike echo
8. ultrasonic
impedance
9. ultrasonic
shadow
Primary Tscr
CQC CQA
yes —
yes -—


yes


yea yes



yes yes
yes yes

— yes
tr
yes

yes

General Comments
Cost of
Equipment
$200
nil


$200


$1000



$500
$20.000

$5000

$7000

$5000

Speed of
Tests
fast
fast


fast


slow



fast
slow

moderate

moderate

moderate

Cost of Tests
low
nil


moderate


very high



nil
high

high

high

high

Type of
Result
yes-no
yes-no


yes-no


yes-no



yes-no
yes-no

yes-no

qualitative

qualitative

Recording
Method
manual
manual


manual


manual



manual
manual and
automatic
automatic

automatic

automatic

Operator
Dependency
high
very high


low


moderate



fejgh
low

moderate

unknown

moderate


-------
      The vacuum chamber {box) method uses a box up to 1.0 m (3 ft) long with a transparent
top that is placed over the seam; a vacuum of approximately 20 kPa (3 Ib/in.1) is applied. "Wften a
leak is encountered the soapy solution originally placed over the seam shows bubbles thereby
reducing the vacuum.  This is due to air entering from beneath the geomembrane and passing   .          '
through  the unbonded zone.   The test is slow to perform (a  10  sec dwell time is currently
recommended) and is often difficult to make a vacuum-tight joint at the bottom of tLe box where it             ;
passes over the sear1 edges. Due to apwarf deformations of the liner into the vacuum box, only             •
geomembrane thickness greater than 1.0 mm (40 mils) should be tested in this manner. For             ,
thinner, more  flexible geomembranes  an open grid wire mesh can be used along the bottom of the,             -
box to prevent uplift. It should also be noted that vacuum boxes are the most common form of             I
nondestructive test currently  used by design engineers and CQA inspectors for polyethylene             ;,
geomembranes.  It should be recognized that 100% of the field seams cannot be inspected by this             i
method.  The test cannot cover portions of sumps, anchor trenches, and pipe peneoatic is with any             |;
degree of assurance.  The  method is also very awkward to use on side slopes.  The  adequate             K,
downward pressure required to make a good seal is difficult to mobilize since ii is usually done by             '
standing on top of the box.                                                                           |

      Electric sparking (not mentioned in Table 3.5) is a technique used to detect pinholes in             i_,;
thermoplastic liners. The method uses a high-voltage (15 to 30  kV) current, and  any leakage to             (
ground (through an opening or hole)  results in sparking. The method is being investigated for             [.
possible field use.  The electric wire  method places a copper or  stainless steel wire between the             |
overlapped geomembrane region and actually embeds it into the completed seam. After seaming, a             {*
charged  probe of about 20,000 volts is connected to one end of the  wire and slowly moved over             |;
the lengin of the seam. A  seam defect between the probe and the embedded wire  results in an
audible alarm from the unit

      The electric field test utilizes a potential which is applied  across the geomembrane  by
placing a positive  electrode  in water within the geomembrane and a ground electrode  in the
subgrade or in the sump of the leak detection system. A current will only flow  between the             §?
electrodes through a hole (leak) in the geomembrane.  The potential gradients in the ponded water             ||5
are measured by "walking" the area with a previously calibrated probe. The operator walks along a             ||
calibration grid layout and identifies where anomalies exist  Holes less than 1 mm diameter can be
identified. These locations can be recheeked after the  survey is completed by other methods, such
as the vacuum box. In deep water, or for hazardous liquids, a remote probe can be dragged from
one side of the impoundment  to the other across the surface of the geomembrane.  On side slopes
that are not covered by water, a positively charged stream of water can be directed onto the surface
of the geomembrane.  When the water stream encounters and  penetrates a hole, contact with the
subgrade is made.  At this point current flow is indicated, thus locating the hole.  Pipe j/enetrations              p^
through  the geomembrane  and soil cover that goes up  the side slope ant' contacts the  subgrade              ""
reduce the sensidviry of the method.
      The last group of nondestructive test methods noted in Table 3.5 can collectively be called              fS
ultrasonic inethods. A number of ultrasonic methods arc available for seam testing and evaluation.              1
The ultrasonic pulse echo technique is basically a thickness measurement technique and is only for              p
use with nonreinforced geomembranes.  Here a high-frequency pulse is sent into the upper              •"--'
geomembrane and (in the case of good acoustic coupling and good contact between the upper and              =••
lower sheets) reflects off of the bottom of the  lower one. If, however, an unbonded area is              i
present, the reflection will occur at the unbonded interface.  The use of two transducers, a pulse              |
generator, ?jid a CRT monitor are required.  It cannot be used for extrusion fillet seams, because of              i
their nonuniform thickness.  The ultrasonic impedance plane method works on the principle of              \
acoustic impedance.  A continuous wave of  160 to 185 kHz  is sent through the seamed              »
geomembrane, and a characteristic dot pattern is displayed on c CRT screen. Calibration of the dot              ;


                                          164

-------
pattern is required to signify a good seam; otherwise, it is not. The method has potential for all
types of geomembranes but still needs additional developmental work. The ulirasonic shadow          '•
method uses two roller transducers:  one sends a signal into the upper geomembrane and the other
receives the signal from the lower geomembrane on the other side of the seam (Richardson and          ;
Koerner, 1988). The technique can be used for all types of seams, even those in difficult
locations,  such as around  manholes, sumps, appurtenances,  etc   It is best  suited for
semicrystalline geomembunes, including HDPE, and will not work for scrim-reinforced liners.             [

3.6,3 Recommendations for Various Seam Types                                                    i

      The various NDT methods listed in Table 3.5 have certain uniqueness and applicability to          ;
specific seam and geomembrane types.  Thus a specification should only be framed around the          (•
particular seam type and geomembrane type for which it has been developed. Table 3.6 gives          K-
guidance in this regard. Even within Table 3.6, there are certain historical developments. For          I
example, the air lance method is used routinely on the flexible geomembranes seamed by chemical          ;
methods, whereas the vacuum chamber method is used routinely on the relatively stiff HDPE          !
geomembranes.  Also to be noted is that the dual seam can technically be used  on all          i;
geomembranes, but only when they are seamed by a dual track thermal fusion method, i.e., by hot          ?
wedge or hot air seaming methods. Thus by requiring such a dual seam pressure test method one          h
mandates the type of seam  which is to be used by the installation contractor.                              ft;

       Lastly, it should be mentioned that only three of the nine methods listed in Table 3.5 are          i.
used routinely at this point in time. They are the  air lance, dual seam  and vacuum chamber          '
methods. The others arc either uniquely used by the installation contractor (pick test and electric          k
wire), or are in the research and development stage (electric current and the various ultrasonic test          |v
methods).                                                                                      |i>'
                                                                                               fc ,'
3.6,4 General Specification Items                                                                 f

                                                                                               &
       Regarding field evaluation of geomembrane seams and their nondestructive testing, a          (•
specification or CQA document should consider the following items:                                    f~
                                                                                               K"
       1.  The  purpose of nondestructive testing should  be clearly  stated.  For example,          p
          nondestructive testing is meant to verify the continuity of field seams and not  to          1
          quantify seam strength.                                                                 !v'
                                                                                               O
       2.  Generally nondestructive testing is conducted as the seaming work progresses or as          ;';
          soon as a suitable length of seam is available.                                              lf{-
                                                                                               * ~.
       3.  Generally nondestructive testing of some type is required for 100% of the field seams.          II'
          For geomembranes supplied in factory fabricated panels, the factory  seams may, or          \ii
          may not, be specified to be nondestructively tested in the field.  This decision depends          $'-
          on the degree of MQC (and MQA) required on factory fabricated seams.                       !

       4.  The  specification should recognize that the same type of nondestructive test cannot be          ; *
          used in every location. For example, in sumps and at  pipe penetrations the dual air          [*'
          channel and vacuum box methods may not be usable.                                        ;

       5.   It must be recognized that there are no current ASTM Standards on any of the NDT
           methods presented in Table 3.5 although many are in progress. Thus referencing to
           such consensus documents is not  possible.  For  temporary guidance, there is a GR1          !
           Standard available for dual seam air pressure test method, GR1 GM-6.                        ;


                                         165

-------
                                                                                                    *
                                                                                                    )r<
      6.  CQA personnel should observe all nondestructive testing procedures.

      7.  The location, data, test number, name of test person and outcome of tests must be
         recorded.

      8.  The Owner's representative should be informed of any deficiencies.

      9.  The method of repair of deficiencies found by nondestructive testing should be clearly
         outlined in the specifications or CQA documents, as should the retesting procedure.
Table 3.6    Applicability Of Various Nondestructive Test Methods To Different Seam Types
            And Geornembrane Types
NOT Method
I. air lance
2. mechanical point stress
3. dual seam
4. vacuum chamber
5, electric wire
6. electric current
7. ultrasonic pulse echo
Seam Types*
C. BC, Chem A, Cont. A
aD
HW.HA
all
all
all
HW.HA
Geornembrane Types
all except HOPE
ail
all
all
all
all
HOPE, VLDPE, PVC
8.  ultrasonic impedance
9.  ultrasonic shadow
CBC.
Chem. A. Cent A

HW.HA
C. BC,
Chem. A, Cont. A

E Fil.. E Fit. HW. HA
*E Fil.
 EFlt.
 HW
 HA
 C
 BC
 Chem. A
 Cont A
           » extrusion fillet
           - extrusion Hat
           »hotwedge
           «hotair
           » chemical
           » bodied chemical
           = chemical adhesive
           • contact adhesive
                                          166
                                                             HOPE, VLDPE, PVC
                                                             HOPE, VLDPE

-------
                                                                                                 p?-
                                                                                                 gt
3.7 Protection and Backfilling

      Hie field deployed and seamed geomembrane must be backfilled with soil or covered with a
subsequent layer of geosynthetics in a timely manner after is acceptance by the CQA personnel. If
the covering layer is soil, it will generally be a drainage material like sand or gravel depending
upon the required permeability of the overlying layer. Depending upon the particle size, hardness
and angularity of this soil, a geotexti.j or other type of protection layer may be necessary. If the
covering layer is a geosynthetic, it will generally  be a geonet or geocomposite drain, which is
usually placed directly upon the  geomembrane.   This is obviously a  critical step since
geomembranes are relatively thin materials with puncture and tear strengths of finite proportions.
Specifications should be very clear and unequivocal regarding this final step in  the installation
survivability of geomembranes.

3.7.1 Soil Backfilling of Geornembranes

      There  are at  least three important considerations concerning  soil  backfilling of
geomembranes: type of soil backfill material, type  of placement equipment and considerations of
slack in the geomembrane.

      Concerning the type of soil backfilling material; its particle size characteristics, hardness and
angularity are important with regard to the puncture and tear resistance of the geomembrane. In
general, the maximum soil particle size is very important, with additional concerns over poorly
graded soils, increased angularity and increased hardness being of significance. Past research on
puncture resistance of geomembranes has shown tha HDPE and CSPE-R geomembranes are more
sensitive to puncture than are VLDPE and PVC geomembranes for conventional thicknesses of the
respective types of geomembranes.  Using truncated cones in laboratory tests K* simulate the
puncturing phenomenon (Mailings and Koerner, 1991), the critical cone height valuv^ which were
obtained are listed in Table 3.7.  It should be cautioned,  however, that these values are not based
on actual soil subgrades, nor on geostatic type stresses. The values are meant to give relative
performance between the different geomembrane types.
                                                                                                 5*;
                                                                    m
                                                                   ***.
                                                                   fiT:
Table 3.7.   Critical Cone Heights For  Selected Geomembranes In Simulated Laboratory
            Puncture Studies (Richardson and Koerner, 1988)
Geomembrane Type
Geomembrane "Thickness
mm            mil
Critical Cone Height
mm          inch
HDPE
VLDPE
PVC
CSPE-R
1.5
1.0
0,5
0.9
60
40
20
36
12
89
70
15
0.50
3.50
2.75
0.60
Although the truncated cone hydrostatic test is an extremely challenging index-type test, the data of
Table 3.7 does not reflect creep and/or stress relaxation of the geomembrane.  In reviewing
numerous CQA documents it appears that the maximum backfill particle size for use with HDPE
and CSPE-R geomembranes should  not exceed  12-25 mm (0.5-1.0 in.).  VLDPE and  PVC
geomembranes appear to be able to accommodate larger soil backfill panicle sizes. If the soil
                                          167
                                                                   Ste-:
                                                                  *^«
                                                                   ^j'**>
                                                                  ifi*«
                                                                   Ik
                                                                  ^r^
                                                                   «•-

-------
panicle size must exceed the approximate limits given (e.g., for reasons of providing high
permeability in a drainage layer), then  a protection  material must be placed on top of the
geomembrane and beneath the soil.  Geotextiles, as well as other protection materials, have been
used in this regard. New materials, e.g., recycled fiber  geotextiles and rubber matting, are being
evaluated.

      Concerning the type of placement -quipment, ftt  initial lift height of the backfill soil is very
important  (Note that construction equipment shoulu never be allowed to move directly on any
deployed geomembrane.  This includes rubber tired vehicles such as automobiles and pickup
trucks but does not include light weight equipment like all-terrain vehicles (ATV's). The minimum
initial lift height should be determined for the type of placement equipment and soil under
consideration, however, 150 mm (6 in.) is usually considered to be a minimum.  Between this
value and approximately 300 mm (12.0 in.), low ground pressure placement equipment should be
specified. Ground contact pressure equipment of less than 35 kPa (S.O lb/in2) is recot   ended.
For lift heights of greater than 300 mm (12.0 in.),  proportionately heavier placement equipment
can be used.

      Placement of soil b: :kf;51ing should proceed from a stable working area adjacent to the
deployed geomembrane ar,  gradually progress outward. Soil is never to be dropped from dump
trucks or front end loaders c;,-rctly onto the geomembrane. The soil should be pushed forward in
an uoward tumbling action so as  not to impact directly on the geomembrane. It should be placed
by a bulldozer or front end loader, never by a motor grader which would necessarily have its front
wheels riding directly on the geomembrane. Sometimes "fingers" of backfill are pushed out over
the geomembrane with controlled amounts of slack between them.  Figure 3.26 shows a sketch
and photograph of this type of soil covering placement.  Backfill is then widened so as to connect
the "fingers", with the controlled slack being induced into the geomembrane. This procedure is at
the discretion of the design engineer and depends on site  specific materials and conditions.

      If a predetermined amount of slack is to be placed in the geomembrane; the temperature of
the geomembrane itself during  backfilling is  important and should be contrasted against the
minimum service temperature that the geomembrane will eventually experience. This difference in
temperature, assuming the geomembrane temperature at the time of backfilling is higher than the
minimum service temperature, is multiplied by the distance between backfilling "fingers" and by
the coefficient of thermal expansion/ contraction of the  particular geomembrane. Coefficients of
thermal expansion/contraction found in the literature are given in Table 3.8, Note, however, that
the coefficient of expansion/contraction of the site  specific geomembrane should be available for
such calculations

      While many geomembrane polymers fall in the same general range of coefficient of thermal
expansion/contraction (as seen in Table 3.8), it is the  stiff and relatively  thick geomembranes,
which are Troublesome during backfilling.  Here the slack accumulates in a wa« which should not
be allowed to crest over on itself, lest a fold is trapped beneath the backfill.  In such cases, the
"fingers" of backfilling must be relatively close together. If the situation becomes unwieldy due to
very high geomembrane temperature, the backfilling should temporarily cease until the  ambient
temperature decreases.  This will have  the effect of  requiring less slack to be placed in the
geomembrane.
                                          168

-------
                   Not*: Arrows Indicate Advancement of
                       Cover Soil Over Oeomembran*
Figure 3.26 - Advancing Primary Leachate Collection Gravel in "Fingers" Over the Deployed
Gcomembrane
                                         !69

-------
Table 3.8 -    Coefficients Of Thermal Expansion/Contraction Of Various  Nonrcinforced
             Geomembrane Polymers (Various References)*
Polymer Type
   Thermal linear expansivity x
perrF
Polyethylene
 high density
 medium density
 tow density
 very low density
Polyvinyl chloride
 unpiasticized
 plasticued
  7-12
   64
   5-7
 11-16

   3-S
  3-10
  4-14
12.22
11-15
 9-13
20-30
 5-18
 7-25
* Values are approximate and change somewhat with the particular formulation and with the actual temperature range
over which the values are measured.
3.7.2 Geosynthejie {^vering of Geomemjjpnes

      Various geosynthetic materials may be called upon to cover the deployed and seamed
geomembrane.  Often a geotexule or a geonet will be the covering material Sometimes, however,
it will be a geogrid (for cover soil reinforcement on slopes) or even a drainage geocomposite (again
on slopes to avoid instability of natural drainage soils).  As with the previous discussion on soil
covering,  no construction vehicles of any type should be allowed to move directly on  the
geomembrane (or any other geosynthetic for that matter). Generators, low tire inflation ATV's,
and other seaming related equipment are allowed as long as they do not damage the geomembrane.
As a result, the movement of large rolls of geotextile or geonet becomes very labor intensive.
Proper planning and sequencing of the operations is important  for logistical control.  The
geosynthetic materials are laid directly on the geomembrane with no bonding of any type to the
geomembrane being allowed. For example, thermally fusing of a geonet to a gcxxncmbrane should
not be permitted.  Temperature compensation (as described earlier) should  be added based on
material characteristics.

      The geosynthetics placed above the geomembrane will either be overlapped (as with some
geotextiles), sewn (as with other geotextiles), connected  with plastic ties  (as with geonets),
mechanically joined with rods or bars (as with geogrids), or male/female joined (as with drainage
composites). These details will be described in Chapter 6 on geosynthetic materials other than
geomembranes.
                                           170

-------
3.7.3
             Secification. Items
      The specification or CQA document for backfilling should be written around the concept
that the geomembrane must be protected against damage by the overlying material. Since soil,
usually sand or gravel, is the most common backfilling material, the items that follow should be
considered.

       1 .   The temperature during soil backfilling should be considered. Expansion, contraction,
           puncture, tear and  other properties vary in accordance with the geomembrane
           temperature.

       2.   In general, backfilling in warm climates or during summer months should  be
           performed at the coolest part of the day.

       3 .   In extreme cases of excessively high temperatures, backfilling may be required during
           non-typical work hours, e.g.t sunrise to 10:00 AM or 5:00  PM to sunset

       4.   If soil backfilling is to be done between sunset and sunrise, i.e., at night, the work
           area should be suitably lit for safety, constructabitity and inspection considerations.

       5,   If soil  backfilling  is to be done at night, excessive  equipment noise may not  be
           tolerated by people in the local neighborhood. This is an important and obviously site
           specific condition which should be properly addressed.

       6.   When a geotextile or other protection layer is to be placed  above the geomembrane it
           should be done so according to the plans and specifications.

       7.   Soil placer, Hint equipment should never move, or drive, directly on the geomembrane.

       8.   Personnel or materials vehicles (automobiles, pickup trucks, etc.) should never drive
           directly on the geomembrane.

       9.   The  soil particle  size characteristics should be stipulated as part of the design
           requirements.

       10. The  minimum soil  lift thickness should be stipulated in the design requirements.
           Furthermore, the thickness should be clear as to whether it is loose or compacted
           thickness.

       1 1 . The  maximum ground contact pressure of the placement equipment  should  be
           stipulated in the design requirements.                   ~~

       12. For areas regularly traversed by heavy equipment, e.g., the access route for loaded
           dump trucks, a larger than usual fill height should be required.

       13. The  CQA  personnel  should be available at  all times during backfilling of the
           geomembrane.  It  is the last time when anyone will see the completely installed
           material.

       1 4, Documentation should include the soil type, lift thickness, total thickness, density and
           moisture conditions (as appropriate).
                                          171

-------
3.8 References
ASTMIM13, "Rubber Propcny-Adhesion to Flexible Substrate"
ASTM D-638, Tensile Properties of Plastics"
ASTM D-751, 'Test Methods for Coated Fabrics'*
ASTM D-7W, "Specific Gravity ami Density of Plastics by Displacement"
ASTM D-882, Test Methods for Tensile Properties of Thin Plastic Sheeting"
ASTM D-1004, "Initial Tear Resistance of Plastic Him and Sheeting"
ASTM D-1238, "How Rates of Thermoplastics by Extrusion Plastometer"
ASTM D-1248, "Polyethylene Plastics and Extrusion Materials"
ASTM D-1505, "Density of Plastics by the Density-Gradient Technique"
ASTM D-1603, "Carbon Black in Olefin Plastics"
ASTM D-1765, "Classification System for Carbon Black Used in Rubber Products"
ASTM D-2663, "Rubber Compounds - Dispersion of Carbon Black"
ASTM D-3Q15, "Recommended Practice for Microscopical Examination of Pigment Dispersion in
    Plastic Compounds"
ASTM D-3083, "Specification for Flexible Poly (Vinyl  Chloride) Plastic Sheeting for Pond,
    Canal, and Reservoir Lining"
ASTM D-4437, "Practice for Determining the Integrity of Held Seams Used in Joining Flexible
    Polymeric Sheet Geomembranes"
ASTM D-4545,  "Practice for Determining the Integrity of Factory Seams Used in Joining
    Manufactured Flexible Sheet Geomembranes"
ASTM D-4759, "Determining the Specification Conformance of Geosynthetics"
ASTM D-5046, "Specification for Fully Crosslinked Elastomeric Alloys"
ASTM D-5199, "Measuring Nominal Thickness of Geotextiles and Geomembranes"
ASTM D-5321 "Determining  the Coefficient  of Soil and Geosynthetic  or Geosynthetic and
    Geosynthetic Friction by the Direct Shear Method"
ASTM D-5397, "Notched Constant Tensile Load Test for Polydefin Geotnembranes"
                                       172

-------
FTM Std. 101C "Puncture Resistance and Elongation Test," Federal Test Method 2065," March
    13, 1980.

GRIGM-6, "Pressurized Air Channel Test for Dual Seamed Geomembranes"

GRI GM-7, "Accelerated Curing of Geomembrane Test Strips Made by Chemical Fusion
    Methods"

GRI GS-7, "Determining the Index Friction Properties of Geosynthetics"

Haxo, H. E.,  (1988), "Lining of Waste Containment and Other Impoundment Facilities,"
    EPA/600/2-88/052, Washington, DC

Hsuan, Y. and Koerner,  R. M.  (1992), "Stress Cracking Potential and Behavior of HDPE
    Geomembranes," Final Report to U.S. EPA, Contract No. CR-815692.

Hullings, D. E, and Koerner, R.  M. (1991), "Puncture Resistance of Geomembranes Using a
    Truncated Cone Test," Proceedings, Geosynthetics '91, IFAI, pp. 273-286.

Richardson, G. N. and Koerner, R. M. (1988), "Geosynthetic Design Guidance for Hazardous
    Waste Landfill Cells and Surface Impoundments," EPA/600/S2-87A)97.

Richardson, G. N. (1992), "Construction Quality Management for Remedial Action and Remedial
    Design Waste Containment Systems," U.S. EPA, EPA/540/R-92/073, Washington, DC

U. S. Environmental Protection Agency (1991) "Inspection Techniques for the Fabrication of
    Geomembrane Field Seams," EPA Technical Guidance  Document, EPA/530/SW-91/051.
                                       173

-------
                                      Chapter 4

                              Geosynthetic Clay Liners
4.1  Types and CgmposiHon of Geosynthe.tic Clav Liners.

      As with most types of manufactured products within a given category, there are sufficient
differences such that no two products are truly equal to one another. Geosynthetic clay liners
(GCLs) are no exception. Yet, there are a sufficient number of common characteristics such that
the current commercially available products deserve a separate category and a separate treatment in
this manual. GCLs can be defined as follows:

      "Geosynthetic ciay liners (GCLs) are factory manufactured, hydraulic barriers
      typically consisting of bentonite clay or other very low permeability clay
      materials, supported by geotextiles and/or geomembranes which arc held
      together by needling, stitching and/or chemical adhesivcs"

Other names that GCLs have been listed under, are "clay blankets", "clay mats", "bentonite
blankets", "bentonite mats", "prefabricated bentonite clay blankets", etc. GCLs are hydraulic
barriers to water, leachate or  other liquids.  As such, they  are used to augment or replace
compacted clay liners or geomembranes, or they are used in a composite manner to augment the
more traditional clay liner or geomembrane materials.

      Cross section sketches of the currently available GCLs at the time of writing are shown in
Rg. 4.1.  General comments regarding each type follow:

      *  Figure 4. l(a) illustrates a bentonite clay mixed with a water soluble adhesive which is
         supported by individual geotextiles on both its upper and lower surfaces.

      *  Figure 4.1 (b) illustrates a stitchbonded variation of the above type of product whereby
         the upper and lower geotextiles are joined by continuous sewing in discrete rows
         throughout  the machine direction  of the product as well as a recent product which
         consists of bentonite powder alone with no admixed adhesive.

      *  Figure 4,1 (c) illustrates a  bentonite clay powder or granules, containing no adhesive,
         which is supported by individual geotextiles on  its upper and lower surface1! and is
         needle punched throughout to provide for its stability. Several variations of this type of
         GCL are available including styles with ciay infilled in the voids of the upper geotextile.

      *  Figure  4.1(d) illustrates a bentonite  clay which is admixed with an adhesive and is
         supported by a geomembrane on its lower surface, as shown, or it can be used in an
         inverted manner with  the geomembrane side facing upward.  Variations  of this product
         are also available with textured or raised geomembrane surfaces.

All of the GCL products available in North Ame..ca use sodium bentonite clay (predominately
smectite) powder or granules at as-manufactured mass per unit areas in the range of 3.2 to 6.0
kg/m2 (0.66 to 1.2 lb/ft2). The clay thickness in the various products vary between the range of
4.0 to 6.0 mm (160 to 320 mils).  GCLs  are delivered to the job site at moisture contents which
                                          174

-------
                                                              Upper Geotextile
                                                              Lower Geotextile
        (a) Adhesive Bound Qay to Upper and Lower Geotextiles
     -5mm
                                                               Upper Geotextile
 Stitch Bonded
 in Rows
                                                              Lower Geotextile


        (b; Stitch Bonded day Between Upper and Lower Geotextiles
    -4-6 mm
                                                               Upper Geotextile
Needle Punched
Fibers Throughout

 Lower Geotextile
          (c) Needle Punched Clay Through Upper and Lower Geotextiles
   - 4.5 mm
               (d) Adhesive Bound Qay to a Geomembrane
                                                              Lower or Upper
                                                             ' Geomembrane
Figure 4.1 - Cross Section Sketches of Currently Available Geosynthetic Ciay Liners (GCLs)
                                       175

-------
vary from 5 to 23%, depending upon the local humidity. Note that this is sometimes referred to in
the technical literature as the dry" state. The types or geotexales used with the different products
vary widely in their manufacturing style (e.g., woven slit film, needle punched nonwoven,
spunlaced, heat bonded nonwovens, etc.) and in their mass per unit area [e.g., varying from 85
g/m2 (15 oz/yd2) to  1000 g/m2 (30 o?/yd2). The particular product with a geomembrane backing
can also vary in its type, thickness and surface texture.

      GCLs are  factory made in widths of 2.2 to 5.2 m (7 to 17 ft) and lengths of 30 to 61 m
(100 to 200 ft). Upon manufacturing GCLs are roiled onto a core and are covered with a plastic
film to prevent additional moisture gain during storage, transportation .and placement prior to their
final covering with an overlying layer.
4.2 Manufacjurifig

      This section on  manufacturing of GCLs
manufacturing of the rolls, and covering of the rolls.

4.2.1 Raw Materials
will  discuss the various raw  materials,
      The bentonite clay materials currently used in the manufacture of GCLs are all of the
sodium mantmotillomte variety which is a naturally occurring mineral in the Wyoming and North
Dakota regions of the USA. After the clay is mined, it is dried, pulverized, sieved and stored in
silos until it is transported to a GCL manufacturing facility.

      The other raw material ingredient used in the manufacture of certain GCLs (recall Section
4,1) is an adhesive which is a proprietary product among the two manufacturers that produce this
type of GCL. Additionally, gcotcxtiles and/or geomembranes are used as substrate (below the
clay) or superstrata (above the clay) layers which are product specific as was mentioned in the
previous section.

      Regarding a specification or MQA  document for the various raw materials used in the
manufacture of GCLs, the following items should be considered.

      1. The clay should meet the GCL manufacturer's specification for quality control
         purposes.  This is  often 70% to 90% sodium montmorillonite clay from the
         Wyoming/North Dakota "Hack Hills" region of bentonite deposits.  A certificate of
         analysis should be submitted by the vendor for each lot of clay supplied.  While the
         situation is far from established, the certificate may include the various compounds of
         the clay, per X-Ray diffraction or methylene-blue absorption, particle size per ASTM
         D-422 or C-136, moisture content per ASTM D-2216 or D-4643, bulk density per
         ASTM B-417, and free swell.

      2. The GCL marufacturer should have a MQC plan which describes the procedures for
         accomplishing quality in the  final product, various tests to be conducted and  their
         frequency. This MQC document should be fully implemented and followed.

      3. The MQC test methods that the GCL manufacturer performs  on the clay component
         may include the following; free swell  per USP-NF-XVHI or ASTM draft standard,
         "Determination of Volumetric Free Swell of Powdered Bentonite Clay," plate water
         absorption per ASTM E-946, moisture content  per ASTM D-2216or D-4643 and
         (sometimes) panicle size per ASTM D-422, fluid  loss per API 1?B. pH per ASTM D-
         4972, and liquid/plastic limit per ASTM D-4318.
                                         176

-------
      4.  For those products which use adhesives, the composition of ihe proprietary adhesive is
          rarely specified.  If a statement is required, it should signify that the adhesive selected
          has been successfully used in th; past and to what extent.

      5.  The geotextiles used as the substrate or the superstrate, or the geomembranc vary
          according to the particular style of product.  Manufacturers current literature should be
          used in this regard.  If a statement is required it should signify that the products selected
          have been successfully used in the past and to what extent.

      6.  If further detail  is needed  as 10 a specification for the geotextiles, see Chapter 6.
          Similarly, specifications for geomembranes are found in Chapter 3.

      7.  The type of sewing thread (or yam) which is used in joining the products is rarely
          specified. If a statement is required it should signify that the materials selected have
          been successfully used in the past and to what extent

4.2.2 Manufacturing

      The raw materials just described are used to make the final GCL product. The production
facilities are all relatively large operations where the products are made in a continuous manner.
Process quality control is obviously necessary and is practiced by all GCL manufacturers. Figure
4.2 illustrates, in schematic form, the various processing methods used for those GCLs which
have adhesives mixed with the clay and those which are stitch bonded and needle punched. Figure
4.2(a) illustrates an adhesively bonded clay product which has an adhesive sprayed in a number of
layers with intermittent additions of bentonite. The clay is placed either between geotextiles or on a
geomernbrane. Figure 4.2(b) illustrates the needle punching or stitch bonding of a bentonite clay
powder after it is placed between the covering geotextiles. Windup around a core and placement of
the protective covering is common among all GCLs.

      There are numerous items which should be includwu in a specification or MQA document
focused on the manufactured GCL product.

       1.  There should be verification that the actual geotextiles or geomembrane used meet the
          manufacturer's specification for that particular type and style.

      2.  A statement should be included that the geotextile property values are based on the
          minimum average  roll value (MARV) concept  The geomembranc's properties are
          generally based on average values.

      3.  Verification that  needle  punched nonwoven geotextiles have been inspected
          continuously for the presence of broken needles using an in-line metal detector. There
          should also be a magnet or other device, for removal of broken needles.

      4.  Verification that the proper mass per unit area of bentonite clay has been added to the
          product should  be provided.   At a minimum, this should consist of providing a
          calculated value based on  the net weight of the final roll divided by its area (with
          deduction for the mass per unit area of the geosynthetics and the adhesive, if any).

      5.  Thickness measurements  are  product dependent, i.e.,  some GCLs  can  be quality
          controlled via thickness while others cannot.
                                          177

-------
   Lower Geotettile
   or Geommtgne
                                         Bcmoiite
   Upper
  Gaxextile
  IWl(epl)
                                                                        Cil aider
                                                                         Sliion
                                                                         or Oven
                                  (a) Adhesive Mixed with Clay

    LoMr
   GeotexUJe

   O.
                                 Bentonte
                              Hopper "B* (opt.)
                                              Upper
Nsxilingor Sith
Bonding SuBoa
                   Windup
                        (b) Needle Punched or Stitch Bonded Through Ch y
Figure 4.2 - Schematic Diagrams of ihe Manufacture of Different Types of Geosynthetic day
            Liners (GCLs)
                                         178

-------
      6.  It is recommended that the overlap distance on both sides of the GCL be marked with
          two continuous waterproof lines guiding the minimum overlap distances.

      7,  The product should be wrapped around a core which  is structurally sound such that it
          can support the weight of the roll without excessive bending or buckling under normal
          handling conditions as recommended by the manufacturer,

      8.  The GCL manufacturer sLjuld have a MQC plan for the finished product, which
          includes sampling frequency, and it should he implemented and followed.

      9.  The manufacturer's quality control tests on the finished product should be stipulated
          and followed. Typical tests include thickness per ASTM D-1777 or ASTM D-5199,
          total product mass per unit area per ASTM D-5261, clay content mass per unit area per
          ASTM D-5261, hydraulic conductivity (permeability) per ASTM D-5084 or GRIGCL2
          and sometimes shear strength at various locations such as top, mid-pl-uie and bottom
          per ASTM D-5321. Other  tests as recommended by the manufacturer are also
          acceptable.

4.2.3 Covering of the Rolls

      The final step in the manufacturing of GCLs is their covering with a waterproof, tightly-fit,
plastic covering. This covering is sometimes a spirally wound polyethylene film approximately
0.05 ro 0.08 mm (2 to 5 mils) thick and is the final step in production. The covering can also be a
plastic bag, or sheet, pulled over the product as a secondary operation. Figure 4.3 shows the
factory storage of GCLs, with their protective covering, before shipment to the field.

      Some items for a specification or MQA document with regard to the covering of GCLs are
the following:

      1.  The manufacturer  should clearly stipulate the type of protective covering and the
          manner of cover placement. The covering should be verified as to its capability for safe
          storage and proper transportation of the product.

      2.  The covering should be placed around the GCL in a workmanlike manner so as to
          effectively protect the product on all of its exposed surfaces and edges.

      3.  The central core should be accessible for handling by fork lift vehicles fitted with a long
          pole (i.e., a "stinger") attached. For wide GCLs, e.g., wider than approximately 3,5 m
          (11.5 ft), handling should be by overhead cranes utilizing two dedicated slings
          provided on each roll at approximately the one-third points.

      4.  Clearly visible  labels should identify the name and address of  the manufacturer,
          trademark, date of manufacture, location of manufacture, style, roll number, lot
          number,  serial number, dimensions, weight and other important items for proper
          identification.  Refer to ASTM D-4873 for proper labeling in this regard.  In some
          cases, the roll  number itself is adequate to trace the entire MQC record and
          documentation.
                                          179

-------
Figure 4,3 - Indoor Factory Siorage of Geosynthetic Clay Liners (GCLs) Waiting for Shipment to
           a Job Site
4.3
      A number of activities occur between the manufacture of a GCL, its final positioning in the
field and subsequent backfilling.  Topics such as storage at the factory, transportation, storage at
the site and acceptance/conformance testing wilt be described in this section.
4.3.1
                                Facilit
      Storage of GCLs at the manufacturers facility is common. Storage times typically range
from days to six months. Figure 4.3 illustrated typical GCL storage at a fabrication facility.

      Some specifications or MQA items to consider for storage and handling of GCLs are the
following:

      1 .  GCLs should always be stored indoors until they are ready to be transported to (he field
          site,

      2.  Handling of the GCLs should be such that the protective wrapping is not damaged.  If
          it is, it must be immediately rewrapped by machine or by hand.  In  the case of minor
          tears it may be taped.
                                          180

-------
      3.  Placement and stacking of rolls should be done in a manner so as to prevent thinning of
          the product at the points of contact with the storage frame or with one another. Storage
          in individually supported racks is common so as tu more efficiently use floor space.

4.3.2 Shipment

      Rolls of GCLs  are shipped from the manufacturers storage facility to the job site via
common carrier. Ships, railroads and trucks have ail been used depending upon the  locations of
the origin and final destination. The usual carrier within the USA is truck, which should be with
the GCLs contained in an enclosed trailer as shown in Fig.  4.4(a), or on an open flat-bed trailer
which is tarpaulin covered as shown in Fig. 4.4 the field. When storage is required for a short
period of time i.e., days or a few weeks, and the product is delivered in trailers, the trailers can be
unhitched from their tractors and used as temporary storage.  See the photograph of Fig. 4.5(a).
Alternatively, storage at the job site can also be acceptable  if the GCLs are properly positioned,
protected and maintained, see Fig. 4.5(b).

      If storage of GCLs is permitted on die job site, offloading of the rolls should be done in an
acceptable manner. Some specification or CQA* document items to consider are the following.

       1.   Handling of rolls of GCLs should be done in a competent manner such that damage
           does not occur to the product nor to its protective wrapping. In this regard ASTM D-
           4873, "Identification, Storage and Handling of Geotextilcs", should be referenced and
           followed.
* Note that the designations of MQC and MQA will now shift to CQC and CQA since field construction personnel
ire involved.
                                          181

-------
F
               Figure 4.4(a) • Fork Lift Equipped with a "Stinger"
                 Figure 4.4(b) - GCL Rolls on a Flat-Bed Trailer
                                     182

-------
Figure 4.5(a) - Photograph of Temporary Storage of GCLs in their Shipping Trailers
     Rgure 4.5(b) - Photograph of Temporary Storage of GCLs at Project Site
                                   183

-------
      2. The location of temporary field storage should not be in areas where water can
         accumulate.  The rolls should be stored on high flat ground or elevated off of the
         ground so as not to form a dam creating the ponding of water.  It is recommended to
         construct a platform so that GCL rolls are continuously supported along their length.

      3. The rolls should not be stacked so high as to cause thinning of the product at points of
         contact  Furthermore, they should be stacked in such  a  way that access  for
         conformance testing is possible.

      4. If outdoor storage of rolls is to be longer than a few weeks particular care. e.g., using
         tarpaulins, should be taken to minimize moisture pickup or accidental damage. For
         storage periods longer than one season a temporary enclosure should be placed over the
         rolls, or they should be moved within an enclosed facility.

4.3.4 Acceptance and Conformanee Testing

      Upon delivery of the GCLs to the field site, the CQA officer should see that conformance
test samples are obtained  These samples are then sent to the CQA Laboratory for testing to ensure
that the GO, confcvms to the projea plans and specificaticns. The samples are taken from selected
rolls  by removing the protective wrapping and cutting full-width, 1 m (3 ft.) long samples from the
outer wrap of the selected roll(s). Sometimes one complete outer revolution of GCL is discarded
before the test sample is taken. The rolls are immediately re-wrapped and replaced in the shipping
trailers or in the temporary field storage area.  Alternatively, conformance testing could be
   formed at the manufacturer's facility and when completed the particular lot should be identified
   the particular project under investigation..

      Items to consider for a specification or CQA document in this regard are the following:

      1. The samples should be identified by type, style, lot and roll numbers.  The machine
         direction should be noted on the samplers) win a waterproof marker.

      2. A lot is usually defined as a group of consecutively numbered roils from the same
         manufacturing line. Other definitions are also possible and should be clearly stated in
         the CQA documents.

      3. Sampling should be done according to the project specification and/or CQA documents.
         Unless otherwise  stated, sampling should be based on a lot basis.  Different
         interpretations of sampling frequency within a lot are based on total area or on number
         of rolls.  For example, sampling could be based on 10,000 mHl 00,000 ft2) of area or
         on use of ASTM D-4354 which is based on rolls.

      4. Testing at the CQA laboratory may include mass per unit area per ASTM D-5261, and
         free swell of the clay component per GRI-GCL1. The sampling frequency for these
         index tests should be based on ASTM D-4354. Other conformance tests, which are
         more performance oriented, could be required  by the project specifications but at a
         reduced frequency compared to the above mentioned index  tests.  Examples  are
         hydraulic conductivity (permeability) ASTM D-5084 (mod.) or GRIGCL2 and direct
         shear testing per ASTM D-5321. The sampling frequency for these performance tests
         might be based on area, e.g., one test per 10,000 m~ (100,000 ft2).
                                         184

-------
      5 .  If testing of the geotextiles, or geomembrane. covering the GCLs is desired it should be
          done on the original mils of the geotextiles, or geomembrane, before they are fabricated
          into the GCL product Once fabricated their properties wBI change considerably due »
          the needling, stitching and/or gluing during manufacturing. ,

      6.  Peel testing of needle punched or stitch bonded GCLs should be done in accordance
          with ASTMD-413(mod.). The sampling frequency is recornmended to he one test per
          2000m2 OQ.OOO ft*).

      7.  Conformance test results should be sent to the CQA engineer prior to installation of any
          GCL from the lot under review.

      8.  The CQA engineer should review the results and should report any nonconformance to
          the Owner/Operator's Project Manager.

      9.  The resolution of failing conformance tests must be clearly stipulated  in  the
          specifications or CQA documents.  Statements should be based upon ASTM D-4759
          entitled "Determining the Specification Conformance of GeosyntheQcs."

4.4 Installation

      This section will cover the placement, joining, repairing and covering of GCLs.

4.4.1 Placemen^

      The installation contractor should remove the protective wrapping from the rolls to be
deployed only after die substrate layer (soil or other geosynthetic) in the field has been approved by
CQA personnel. The specification and CQA documents should be written in such a manner as to
ensure that the GCLs are not damaged in any way. A CQA inspector should be present at all times
during the handling, placement and covering of GCLs. Figure 4.6(a) shows the typical placement
of a GCL in the field on soil subgrade and  Fig. 4.6(b) shows placement (without heavy
equipment) on an underlying geosynthetic.

      The following items should be considered  for inclusion in a specification or CQA
      1.  The installer should take the necessary precautions to protect materials underlying the
          GCL. If the substrate is soil, construction equipment can be used to dsploy the GCL
          providing excessive rutting is not created. Excessive rutting should be clearly defined
          and quantified.  In some cases 25 mm (1 .0 in.) is the maximum rut depth allowed. If
          the ground freezes, the depth of ruts should be further reduced to a specified value. If
          the substrate is a geosynthetic material, GCL deployment should be by hand, or by use
          of small jack lifts or light weight equipment on pneumatic tires having low ground
          contact pressure.

      2.  The minimum overlap distance which is specified should be verified. This is typically
          150 to 300 mm (6 to 12 in.) depending upon the particular product and site conditions.
                                          185

-------
      Figure 4.6(a> - Field Deployment of a GCI. «Tt .1 S->i! Si.
I'i^urc 4 <>(h) • Tu-iJ IK'pio\ incut <>! a (i('l. i>ri .in I  n,iii!\ ;;:•:' ii'i

-------
      3.  Additional oentonite clay should be introduced into the overlap region with certain types
          of GCLs. There are typically those with needle punched nonwoven gcotextiles on their
          surfaces. The cky is usually added by using a line spreader or line chalker with the
          bentonite clay in a dry state. Alternatively. a bentonite clay paste, in the mixture range
          of 4 to 6 parts water to 1 part of clay, can be extruded in the overlap region.
          Manufacturer's recommendations on type and quantity  f clay to be added should be
          followed.

      4.  During placement, care must be taken not to entrap in or beneath the GCL, fugitive
          clay, stones, or sand that could damage a geomembrane, cause clogging of drains or
          filters, or hamper subsequent seaming of materials either beneath or above the GCL.

      5.  On side slopes, the GCL should be anchored at the top and then unrolled so as to keep
          the material free of wrinkles and folds.

      0.  Trimming of die GCL should be done with great care so that fugitive clay particles do
          not come in contact with drainage materials such as geonets, geocomposites or natural
          drainage materials.

      7.  The deployed GCL should be visually inspected to ensure that no potentially harmful
          objects are present, e.g., stones, cutting  blades, small tools, sandbags, etc.

4.4.2 Joining

      Joining of GCLs is generally accomplished by overlapping without sewing  or other
mechanical connections. The overlap distance requirements should be clearly stated. For all GCLs
the required overlap distance should be marked on the underlying layer by a pair of continuous
guidelines. The overlap distance is typically 150 to 300 mm (6 to 12 in.).  For those GCLs, with
needle punched nonwoven geotextiles on their surfaces, dry bentonite is generally placed in the
overlapped region.  If this is the case, utmost  care should be  given to avoid fugitive bentonite
particles from coming into contact with leachate collection systems. Another variation, however,
has been to extrude a moistened tube of bentonite into die overlapped region.
      Items to
corrader for a specification or CQA document follow:
       1.  The amount of overlap for adjacent GCLs should be stated and adhered to in field
          placement of die materials.

       2.  The overlap distance is sometimes different for the rol' ends versus the roll edges.  The
          values should be stated and followed.
                                                               -Hf

       3.  If dry or moistened bentonite clay (or other material) is to be placed in the overlapped
          region, the type and amount should be stated in accordance with the manufacturer's
          recommendations and/or design considerations.  Index testing requirements for proper
          verification of the clay should be specified accordingly. Furthermore, the placement
          procedure  should be clearly outlined so as to have enough material  to make an
          adequately tight joint and yet not an excessive amount which could result in fugitive
          clay particles.

4.4.3  Repairs

       For the geotextile*related GCLs, holes, tears or rips in the covering geotexdles  made during
                                          187

-------
transportation, handling, placement or anytime before backfilling should be repaired by patching
using a geotextile.  If the bentonite component of the GCL is disturbed either by loss of material or
by shifting, it should be covered using a full GCL patch of die same type of product

      Some relevant specification or CQA document items follow.

      1.  Any patch, used for repair of a tear or rip in the geotextile, should be done using the
          same type as the damaged geotextile or other approved geotextile by the CQA engineer.

      2.  The size of the geotextile patch must extend at least 30 cm (12 in.) beyond any portion
          of the damaged geotextile and be adhesive or heat bonded to the product to avoid
          shifting during backfilling with soil or covering with another geosymhetic.

      3.  If bentonite particles are lost from within the GCL or if the clay has shifted, the patch
          should consist of the full GCL product  It should extend at least 30 cm (12 in.) beyond
          the extent of the damage at all locations. For those GCLs requiring additional bentonite
          clay in overlap seaming, the similar procedure should be use for patching.

      4.  Particular care should be exercised in using a GCL patch since fugitive clay can be lost
          which can find its way into drainage materials or onto geomentbranes in areas which
          eventually are to be seamed together.

4,5 Backfilling or Covering

      The layer of material placed above the deployed GCL will be either soil or another
geosvnthetic. Soils will vary from compacted clay layers to coarse aggregate drainage layers.
Geosynthetics will generally be geomembranes although other geosynthetics may also be used
depending on the site specific design. The GCL should generally be covered before a rainfall or
snow event occurs. The reason for covering with the adhesive bonded GCLs is that hydration
before covering can cause changes in thickness as a result of uneven swelling or whenever
compressivc or shear loads are encountered.  Hydration before covering may be less of a concern
for the needled and stitch bonded types of GCLs, but migration of the rally hydrated clay in these
products might also be possible under sustained compressive or shear loading.  Figure 4.7 shows
the premature hydration of a GCL being gathered up by hand to be discarded in the adjacent
landfill.

      Some recommended specifications or CQA document items are as follows:

      1.  The GCL should be covered with its subsequent layer before a  rainfall  or snowfall
          occurs.                                                x

      2.  The GCL should not be covered before observation and approval by the CQA
          personnel. This requires close coordination between the installation crew and die CQA
          personnel.

      3.  If soil is to cover the GCL it should be done such that the GCL or underlying materials
          are not damaged. Unless otherwise specified, the direction of backfilling should
          proceed in the direction cf downgradient shingling of the GCL overlaps.  Continuous
          observation of the soil placemen* is recommended.

      4.  If a geosymhetic is tc cover a GCL. both underlying and the newly deployed material
          should not be damaged.
                                         188

-------
      5.  The overlying material should not be deployed such that excess tensile stress is
          mobilized in the GCL. On side slopes, this requires soil backfill to proceed from the
          bottom of the slope upward. Other conditions are site specific and material specific.
Figure 4.7 - Premature Hydration of a Geosymhetic Clay Liner Being Gathered and Discarded due
           to its Exposure to Rainfall Before Covering
4.6 References

API 13B, "Fluid Loss of Bentonite Clays"

ASTM B-417, "Apparent Density of Non Free-Flowing Metal Powders"

ASTM C-136, "Sieve Analysis of Fine and Coarse Aggregates"
                                         189

-------
ASTM D-413, "Rubber Property - Adhesion to Flexible Substrate"
ASTM EM22, "Panicle Size Analysis of Soils"
ASTM D-1777, "Measuring Thickness of Textile Materials"
ASTM D-2216, "Laboratory Determination of Water (Moisture) Content of Soil and Rock"
ASTM D-4318, * Liquid Limit, Plastic Limit, and Plasticity Index of Soils"
ASTM D-4354, "Sampling of Geosynthetics for Testing"
ASTM D-4643, "Determination of Water (Moisture Content) of Soil by Microwave Oven Method"
ASTM D-4759, "Determining the Specification Conformance of Geosynthetics1*
ASTM D-4873, "Identification, Storage and Handling of Geotextiles"
ASTM D-4972, "Method for pH of Soils"
ASTM D-5084, "Hydraulic Conductivity of Saturated Porous Material Using A Flexible Wall
ASTM D-5199, "Nominal Thickness of Geotextiles and Geomembranes"
ASTM D-5261, "Measuring Mass per Unit Area of Geotextiles"
ASTM D-5321, "Determining the Coefficient of Soil and Geosynthetic or Geosynthetic and
    Geosynthetic Friction by the Direct Shear Method"
ASTM E-946, "Water Absorption of Bemonite of Porous Plate Method"
GRIGCL1, "Free Swell Conformance Test of day Component of a GCL"
GRIGCL2, "Permeability of Geosynthetic Clay Liners (GCLs)"
USP-NF-XVll, "Swell Index Test"
                                       190

-------
                                      Chapter 5

                               Soil Drainage  Systems
5.1   Introduction and Background

      Nature] sot! drainage materials are used extensively in waste containment units. The most
common uses arc

       1.     Drainage layer in final cover system » reduce the hydraulic head on the underlying
             barrier layer and to enhance slope stability by reducing seepage forces in the cover
             system.

       2.     Gas collection layer in final cover systems to channel gas to vents for controlled
             removal of potentially dangerous gases.

       3.     Leachate collection layer in liner systems to remove leachate for treatment and to
             remove precipitation from the disposal unit in areas where waste has not yet been
             placed.

       4.     Leak detection layer in double liner systems to monitor performance of the primary
             liner and, if necessary,» serve as a secondary leachate collection layer.

       5.     Drainage trenches to collect horizontally-flowing fluids, e.g., ground water and
             pa-

Drainage layers are also used in miscellaneous ways, such as to drain liquids from backfill behind
retaining walls or to relieve excess water pressure in critical areas such as the toe of slopes.

3.2    Material^

       Soil drainage systems are constructed of materials that have high hydraulic conductivity.
High hydraulic conductivity is not only required initially, but the drainage material must also
maintain a high hydraulic conductivity over time and resist plugging or clogging. The hydraulic
conductivity of drainage materials depends primarily on the pain size of the finest particles present
in the soil.  An equation that is occasionally used to estimate hydraulic conductivity of granular
    rials is Hazen's formula:

                                                                                  (5.1)

where k is the hydraulic conductivity (cm/s) and DIQ is the equivalent grain diameter (mm) at
which 10% of the soil is finer by weight. To determine the value of DJQ, a plot is made of the
pain-size distribution of the soil (measured following ASTM D-422) as shown in Fig. 5.1.  The
equivalent grain  diameter (Dio) te determined from the grain size distribution curve as shown in
Rg. 5.1.

       Experimental data verify that the percentage of fine material in the soil dominates hydraulic
conductivity. For example, the data in Table 5.1 illustrate the influence of a small amount of fines
                                          191

-------
upon the hydraulic conductivity of a filter sand. The addition of just a few percent of fine material
to adraiMge material can reduce the hydraulic cxjnductivity of the drainage material by 100 fold or
tnoxe.
                100
                 80
                 eo
           S
           I
                 20
                           SIM
                                Utan Grain Dlanwttr 0jjg Scato)
                        Figure 5.1 - Gnin Size Distribution Curve
       Construction specifications usually stipulate a minimum hydraulic conductivity for the
drainage layer. The value specified varies considerably from project to project but is typically in
the range of 0.01 to 1 cm/*. The method used to determine hydraulic conductivity in the laboratory
UASTMD-2434.
                                          192

-------
Table 5.1
     Effect of Fines on Hydraulic Conductivity of a Washed Filter Aggregate (from
     Cedergren, 1989)
Percent Passing
No. 100* Sieve
0
2
4
6
7
Hydraulic Conductivity (an/a)
0.03to0.11
0.004 lo 0.04
0.0007 to 0.02
0.0002 to 0.007
0.00007 to 0.001
•Opening size is 0.15 nun.
       Drainage materials may also be required to serve as filters. For instance, as shown in Fig.
5.2, a filter layer may be needed to protect a drainage layer ftom plugging.  The filter layer must
serve three functions:
              The filter must prevent passage of significant amounts of soil through the filter,
              Lc., the filter must retain soil.

              The filter must have a relatively high hydraulic conductivity, e.g., the filter should
              be more permeable than the adjacent soil layer.

              The soil particles within the filter must not migrate significantly into the adjacent
              drainage!
1.
2.
3.
       Filter  specifications vary somewhat, but the  design procedures  are  similar.  The
determination of requirements for a filter material proceeds as follows:
1 .
2.
              The grain size distribution curve of the soil to be retained^ (protected) is determined
              following procedures outlined in ASTM D-422. The sift of the protected soil at
              which 15% is finer (Dts.Mi) and 85% is finer (Dgj.Mi) is determined.
              Experience shows that the particles of the protected soil will not significantly
              penetrate into the filter if the size of the filter at which 15% is finer (Dis. fuu») is
              less than 4 to 5 times DSJ of the protected soil:
                        , fillers (4 to 5) DM,
                                                                             (5.2)
                                           193

-------
       4.
       5.
             Experience shows that the hydraulic conductivity of the filter will be significantly
             greater than that of the protected soil if the following criterion is satisfied:
                    Di5, filler^ 4 DIS.
                                                                      (5.3)
To ensure that the particles within the filter do not tend to migrate excessively into
the drainage layer, the following criterion may be applied:
                        , dramS (4 to 5) Oi5,
                                                                      (5.4)
Experience shows that the hydraulic conductivity of the drain will be significantly
greater than that of the filter if. the following criterion is satisfied:
                    Dl5, drain£ 4 DIS, filter
                                                                      (5.5)
       Filter design is complicated significantly by the presence of biodegradable waste materials,
e.g., municipal solid waste, directly on top of the filter.  In such circumstances, the usual filter
criteria may be modified  to satisfy site-specific requirements.  Some degree of reduction in
hydraulic conductivity of the filter layer may be acceptable, so long as the reduction does not
impair the ability of the drainage system to serve its intended function.  A laboratory test method to
quantify the hydraulic properties of both soil and geotextile filters that are exposed to leachate is
ASTM D-1987. However, regardless of specific design criteria, the gradational characteristics of
the filter material control  the behavior of the  filter. CQC/CQA personnel should focus their
attention on ensuring that the drainage material and filter material meet the grain-size-distribution
requirements set forth in the construction specifications, as well as other specified requirements
such as mineralogy of the materials.
                                Layer Whose Particles Must Not
                           Migrate into Underlying Drainage Layer
                               Filer Layer to Prevent Migration
                             of Soil Particles into Drainage Layer
            Figure 5.2 - Filter Layer Used to Protect Drainage Layer from Plugging
                                           194

-------
5.3    Control of Materials

       The recommended procedure for verifying the hydraulic conductivity for a proposed
drainage material is as follows. Samples of the proposed material should be obtained and shipped
to a laboratory for testing. Samples should be compacted in the laboratory to a density that will be
representative of the density to be used in the field. Hydraulic conductivity should be measured
following procedures in ASTM D-2434 and compared with the required minimum values stated in
the construction specifications. If the hydraulic conductivity exceeds the minimum value, the
material is tentatively considered to be acceptable. However, it should be realized that the process
of excavating and placing the drainage material will cause some degree of crushing of the drainage
material and will produce additional fines.  Thus, the construction process itself tends to increase
the amount of fines in  the drainage material and to decrease the hydraulic conductivity of the
material If the drainage material just barely meets the hydraulic conductivity requirements stated in
the construction specifications from initial tests, there is a good possibility that the material will fail
to meet the required hydraulic conductivity standard after die material has been placed. As a rule of
thumb, approximately one-half to one percent of additional fines by weight will be generated every
time a drainage material is handled, e.g., one-half to one percent additional fines  would be
generated when the drainage layer material is excavated and an additional one-half to one percent of
fines would be generated when the material is placed.  Also, the reproducibility of hydraulic
conductivity tests is not well established; a material may just barely meet the hydraulic conductivity
standard in one test but fail to meet minimum requirements in another test Finally, if the drainage
materials are found to be suitable prior to placement but unsuitable after placement, an extremely
difficult situation arises — it is virtually impossible to remove and replace the drainage material
without risking damage to underlying geosynthetic components, e.g., a geomembrane. Therefore,
some margin of safety should be factored into the selection of drainage material.

       Because it is extremely  difficult to remove  and replace a drainage material without
damaging an underlying geosynthetic component, testing of the drainage matcria' should occur
prior to placement of the material. The CQC personnel should have a high degree of confidence
that the drainage material is suitable prior to placement of the material. Because the construction
process may alter the characteristics of the drainage material, it is important that CQA tests also be
performed on the material after it has been placed and compacted (if it is compacted).

       The usual tests involve determination of the grain size distribution of the soil (ASTM D-
422) and hydraulic conductivity of the soil (ASTM D-2434). Hydraulic conductivity tests tend to
be time consuming and relatively difficult to reproduce  precisely; the test apparatus  that is
employed, the compaction conditions for the drainage material, and other details of testing may
significantly influence test results.  Grain-size distribution analyses are simpler. Therefore, it is
recommended that the CQA  testing program emphasize grain-size distribution analyses,  with
particular attention paid to the amount of fines present in the "drainage material, rather than
hydraulic conductivity  testing. The percent of fines is normally defined as the percent on a dry
weight basis passing through a No. 200 sieve (openings of 0.075 mm). Again, it is emphasized
that close testing and inspection  of the borrow source or the  supplier prior to placement of the
material is critical, particularly if the drainage material is underlain by a geosynthetic material.

       The  recommended tests  and frequency of testing  are shown in  Table 5.2.  The same
principles for sampling strategies discussed in Chapter 2 may be applied to location  of tests or
location of samples for drainage layer materials.  Also, occasional failing tests may be allowed, but
it is recommended that no  more than 5% of the  CQA tests  be allowed to deviate from
specifications, and the  deviations should be relatively minor, i.e., no more than about 2% fines
beyond the  maximum  value allowed and  no less than about one-fifth the  minimum allowable
hydraulic conductivity.
                                          195

-------
         Table 52 - Recommended Tests and Testing Frequencies for Drainage Material
Location of Sample
Type of Tea
MuwmBn Fre<]uency
Potential Borrow Source
OB Site* After Placement
and Compaction
Grain Size
(ASTMEM22)


Hydraulic Conductivity
(ASTMD-2434)

Carbonate Content*
(ASTMCM373)

Grain Size       • •
(ASTMD422)
                                   Hydraulic Conductivity
                                   (ASTMD-2434)
                                   Carbonate Content*
                                   (ASTMD-4373)
Ipw 2,000 m^



1 per 2,000 m3
1 per Ifectare for Drainage
Layers; 1 per 500m3 for
Other Utes

1 per 3 Hectare* for Drainage
Layers: 1 per 1.500m3 for
Other Use*

1 per 2,000m3
*"rte frequency at carbonate content testing should be greatly reduced to 1 per 20 JXX)m3 for thote drainage materials
that obviously do not and cannot contain sipriflcam carbonates (94, crushed basalt).
5.4    Location of Borrow Sources

       The construction specificationr usually establish criteria that must be met by the drainage
material. Earthwork contractors are normally given latitude in locating a suitable source of material
that ineets constraction specifications.  On occasion the materials may be available on site or from a
nearby piece of property, but most frequently the materials are supplied by a commercial materials
company. If the materials are supplied by an existing materials processor, stockpiles of materials
are Usually readily available for testing and no geotecnnical investigations are required, other than
to test the proposed borrowed material.

5,5    Processing of Materials

       Materials may be processed in several v   s.  Oversized stones or rocks are typically
removed by sieving.  Fine material may also be removed by sieving. Washing the fines out of a
sand or gravel can be particularly effective in removing silt and clay sized particles from granular
                                            196

-------
material. For drainage layer materials that are supplied from a commercial processing facility, die
facility owner is usually experienced in processing die material to remove fines,

       For the CQA inspector the main processing issues  are removal of oversized material,
removal of angular material (if required to minimize potential to puncture a geomembrane), and
assurance that excessive fines will not be present in the material.

       On occasion the amount of limestone, dolostone. dolomite, calcite, or other carbonates in
the drainage material may be an issue. Carbonate materials are slightly soluble in water.  If the
drainage material contain.* excessive carbonate, the carbonate may dissolve at one location and
precipitate at another, plugging the material. CQA inspectors should also be cognizant of the need
to make sure that carbonate components are not present in excessive amounts. If the specifications
place a limit on carbonate content, tests should be performed to confirm compliance (Table 5.2).

5.6

       Drainage materials may be placed in layers (e.g., as leachate collection layers) or they may
be placed in  drainage trenches (e.g., to provide drainage near the toe of a slope).  Placement
considerations differ depending on the application.

5.6.1  Drainage; l^yfflj

       Granular drainage materials are usually hauled to the placement  area in dump trucks,
loosely dumped from the truck, and spread with bulldozers. The contractor should  dump and
spread the drainage material in a manner that minimizes generation of fine material. For instance,
light-contact-pressure dozers can be used to spread the drainage material and minimize die stress on
the gniiular material. Granular materials placed on top of geosynthetic components on side slopes
should be placed from the bottom of the slope up.

       When granular drainage material is placed on a previously-placed geomembrane or
geotextUe and spread with a dozer, the sand or gravel should be lifted and tumbled forward so as to
minimize shear forces on the underlying geosynthetic.  The dozer should not be allowed to
"crowd" the  blade into the  granular material and drag it over the surface of the underlying
geosynthetic material.

       Granular materials are often placed with a backhoe in small, isolated areas such as sumps.
Some drainage materials may even be placed by hand, e.g., in sumps and around drainage pipes.

       CQA personnel should position themselves in front of the working  face of the  placement
operation to be able to observe the materials as they are spread and to ensure that there  is no
puncture of underlying materials. CQA personnel should observe placement of drainage layers to
ensure that fine-grained soil is not accidentally mixed with drainage material.

5.6.2  Drainage Trenches

       Drainage materials are often placed in trenches to provide for subsurface drainage of water.
A typical trench configuration is shown in Fig. 5.3. Often, a perforated pipe will be placed in the
bottom of the trench.  Geotextile filters are often required along die side walls to prevent migration
of fine particles into the drainage material. CQA personnel should carefully review the plans and
specifications to ensure that die drainage and fitter components have been properly located in die
trench prior to backfill.
                                          197

-------
      Geotextile Filter
                V-V-V.V.:•,•..••.-,•.••••"•.•".•.•",'••',•••",'••',•
                                  '-
                      Figure 5.3 - Typical Design of a Drainage Trench
       CQC/CQA personnel should be aware of all applicable safety requirements for inspection
of trenches.  Unsupported trenches can pose a  hazard to personnel working in the trench or
inspecting the trench. For trenches that are supported by shoring, CQA personnel should review
with the contractor the plan for pulling the shoring in terms of the timing for placement of materials
and ensure that the procedures are in accord with the specifications for the project.

       Granular backfill is usually placed in a  trench by a back hoe.  For narrow trenches, a
"tremie" is commonly used to direct the  material  into the trench without allowing the material to
come into contact with soil on the sidewalls of the trench. Sometimes drainage materials are placed
by hand for very small trenches.

       A special  type of trench involves support of the trench wall with a biodegradable
("biopolymcr") slurry.  The trench is excavated into soil using a biodegradable, viscous fluid to
maintain the stability of the trench. The backfill is placed into the fluid-fitted trench. An agent is
introduced to promote degradation of the viscous drilling fluid, which quickly loses much of its
viscosity and allows the granular backfill to attain a  high hydraulic conductivity without any
plugging effect from the slurry. This technology allows construction of deep, continuous drainage
trenches but is used much more often for remediation of contaminated sites than in new waste
containment facilities. Further details are given by Day (1990).

5.7    Compaction

       Many construction specifications  stipulate a minimum percentage compaction for granular
drainage layers. There is rarely a need  to compact drainage materials.  However, on occasion,
there may be a need to compact a drainage material for one of the following reasons:
                                           198

-------
      3.
If a sett.,  .ait-sensitive structure h to be placed on top of the drainage layer, the
drainage la^or may need to be compacted to minimize settlement

If dynamic  loads  might cause  loose drainage material to liquefy or settle
excessively, the material may need to be compacted.

If the drainage material must have exceptionally high strength, the material may
need to be compacted.
      Only in rare instances will the problems listed above be significant Settlement-sensitive
^ructures are rarely built on top of liner or cover systems. u_qurf action is rarely an issue because
die hydraulic conductivity of the drainage maerial is normally sufficiently large to preclude the
possibility of liquefaction. Strength is  rarely a problem with granular materials.  Reasons not to
compact the drainage layer are as follows:

      1.     Compacting the drainage material increases th« amount of fines  in the drainage
             material, which decreases hydraulic conductivity,

      2.     Compacting the drainage layer reduces the porosity of the material, which decreases
             hydraulic conductivity,

      3.     Dynamic compaction stresses may damage underlying geosynthetics,

      Unless there is a sound reason why the drainage material  should be compacted, it is
recommended mat the drainage material not be compacted. Hie main goal of the drainage layer is
to remove liquids, and this can only be accomplished if the drainage layer has high hydraulic
conductivity. The -jncompacted drainage layer may be slightly compressible, but the amount of
compression is expected to be small.

      There is a potential problem with drainage layer materials placed on side slopes. In some
situations the friction between the drainage layer and underlying geosymhetic component may not
be adequate to maintain stability of the side slope. CQA personnel should assume that the designer
has analyzed slope stability and designed stable slide slopes for assumed materials and conditions.
However, CQA personnel should be vigilant for evidence of slippage at the interface between the
drainage layer and an underlying geosynthetic component.  If problems are noted, the design
engineer should be notified immediately.

5.8
       The main protection required for the drainage layer is to ensure that large pieces of waste
material do not penetrate excessively into the layer and that fines do not contaminate the layer.
Many designs call for placement of protective soil or select waste on top of the leachate collection
layer. As shown in Fig. 5.4. CQA personnel should stand near the working face of the first lift of
solid waste placed on top of a leachate collection layer in a solid waste landfill to observe placement
of select material.

       Wind-borne fines may contaminate drainage materials. Soil erosion from adjacent slopes
may also lead to accumulation of fines in the drainage material.  The  CQA personnel cannot
complete their job until the drainage material is fully covered and protected.

       Residual fines may be  washed by rain from other soils, or the drainage material itself,
during rain storms and accumulate in low areas.  The accumulation of fines in sumps or other low


                                          199

-------
points can reduce the effectiveness of the drainage system, CQOCQA personnel should be aware
of this potential problem and watch for (1) areas where fines may be washed into the drainage
material; and (2) evidence of lack of free drainage in low-lying areas (e.g., development or ponds
of water in the drainage material in low-lying areas).  If excessive fines are washed into a portion
of the drainage material, the design engineer should  be contacted for further evaluation prior to
covering die drainage material by fie next successive layer in the system.
 Rgure 5.4 » CQC and CQA Personnel Observing Placement of Select Waste on Drainage Layer.

-------
5.9
ASTM D-422, "Panicle Size Analysis of Soils"
ASTM D-1987, "Biological Clogging of Geotextile of Sofl/GoMewile Fillers"
ASTM D-2C4, "Permeability of Granular Soils"
ASTM D-4373, "Calcium Carbonate Content of Sofls"
Cedergitn, H.R. (1989), Seepage, Drainage, and Flow Nets, Third Edition, John Wiley & Sons,
    New York, 465 p.
Day, S. R. (1990), "Ewavation/Inierccption Trenches by the Bio-Polymer Slurry Drainage Trench
    Technique." Superfund 90, Hazardous Materials Control Research Institute, Silver Spring,
    Maryland, pp. 382*385.
                                         201

-------
                                      Chapter 6

                          Geosynthetic Drainage  Systems
6.1  Overview
      The collection of liquids in waste containment systems, their drainage and c 'entual removal
represents an important element in the successful functioning of these facilities.  Focus in this
chapter is on the primary and secondary leachate collection systems beneath solid waste and on
surface water and gas removal systems in the cover above the waste. This chapter parallels
Chapter 5 on natural soil drainage materials but now using geosynthetics. Combined systems such
as geocomposites and geospacers are often used; however we will generally  focus on the
individual geosynthetic components. The individual materials to be described are the following:

      *  geotextiles used as filters over various drainage systems (geonets, geocomposites, sands
         and gravels)

      *  geotexriles used for gas collection

      *  geonets used as orimary and/or secondary leachate ccJlection systems, and gas collection

      *  other geosynthetic drainage systems used as surface water collection systems and
         possibly as primary and/or secondary leachate collection systems

The locations of the various geosynthetic materials listed above are illustrated in the sketch of fig.
6.1.
      Geotextiles. which some refer to as filter fabrics or construction fabrics, consist of
polymeric yams (fibers) made into woven or mmwoven textile sheets and supplied to the job site in
large rolls. When ready for placement, the rolls are removed from their protective covering,
property  positioned and unrolled over the substrate material.  The substrate upon which the
geotextile is placed is usually a geonet, geocomporite. drainage soil or other soil material  The roll
edges and ends are either overlapped for a specified distance* or are sewn together. Afterapproval
by the CQA personnel, the geotextile is covered with the overlying material. Depending on site
specific conditions, this overlying material can be a geomembrane, geosynthetic clay liner,
compacted clay liner, geonet, or drainage soil.

      This section presents the MQA aspects of geotextiles insofar as their manufacturing is
concerned and ihe CQA aspects as far as handling, seaming and backfilling is concerned.

6.2.1 Manufacturing of Geotextile*

      The manufacturing of geotextiles made from polymeric fibers follows traditional textile
manufacturing methods and uses similar equipment. It should be recognized at the outset that most
manufacturing facilities have developed their respective geotextile products to the point where
product qualify control procedures and programs are routine and fully developed.

      Three discrete stages in the manufacture of geotextiles should  be recognized from an MQA
perspective: (1) the polymeric materials; (2) yarn or fiber type; and (3) fabric type (IFAI, 1990).
                                          202

-------
GT   -
ON   -Oeooet
OM  — OffomfniNiM
CO. -OeoiymhtticCUyUnef
OC   "Oeooimpottle
CCL » Compacted Clay IJ»f
                                                                              ocGC
                                                                         CCL
Figure 6.1 - Qoss Section of • Undfill Illustrating the Use of Different Geosynthetici Involved
          in Waste Containment Drainage Systems

-------
6.2.1.1 Resins and Theiy Additives

      Approximately 75% of geotextiles used today are based on polypropylene resin.  An
additional 20% are polyester and the remaining 5% is a range of polymers including polyethylene,
nylon and others used for specialty purposes.  As with all geosymhetics, however, the base resin
has various additives formulated with it resulting in the final compound Additives for ultraviolet
light protection and as processing aids are common, see Table 6.1.


Table 6.1 •   Compounds Used in The Manufacture of Geotextiles (Values Are Percentages Based
            on Weight)
Generic None
 Resin
Carton Black
Other Additives
Polypropylene

Polyester

Ottos
93-91

97-98

95-98
   0-3

   0-1

   1-3
    0-2

    0*2

    1-2
      The resin is usually supplied in the form of pellets which is then blended with carbon black,
either in the form of concentrate pellets or chips, or as a powder, and the additive package. The
additive package is usually a powder and is proprietary with each particular manufacturer. For
some manufacturers, the pellets are precompounded with carbon black and/or the entire additive
package,  figure 6.2 shows polyester chips and carbon  black concentrate pellets used in the
manufacturer of polyester geotextiles. Polypropylene pellets and carbon Mack are similar to those
shown in the manufacture of polyethylene geomembranes. Refer to Chapter 3 for details and in
particular to Section 3.2.2 for use of recycled and/or reclaimed material.

      The following items should be considered for a specification or MQA document for resins
and additives used in the manufacture of geotextiles for waste containment applications.

      1.  The resin should meet MQC requirements. This usually requires a certificate of analysis
          to be submitted by the resin vendor for each lot supplied. Included will be various
          properties, their specification limits and th*  appropriate  test  methods.  For
          polypropylene resin, the usual requirement* -   ..teh flow index^and other properties
          felt to be relevant by the manufacturer. For polyester resin, the usual requirements are
          intrinsic viscosity, solution viscosity, color, moisture content and other properties felt
          10 be relevant by the manufacturer.

      2,  The internal quality control of the manufacturer should be reported to verify that the
          gcotextile manufactured for the project meets the proper specifications.

      3.  The frequency of performing each of the preceding tests should be covered in the MQC
          plan and should be implemented and followed.
                                          204

-------
Polyester Chips
      •  »
Figure 6.2 - Polyester Resin Chips (t'pper) and Carbon Black Concentrate Pellets
           far Geotex tile Fihrr Munufacturirg
                                            Used
     205

-------
      4.  The percentage, according to ASTM D-1603, and rype of carbon black should be
          specified for the particular formulation being used, although it is low in comparison to
          Heumei iiPianes.

      5.  The type and amount of stabilizers are rarely specified.  If a statement is required it
          shouldsignify that the stabilizer package has been successfully used in the past and to
          what extent

6.2.1.2 Fiber Types

      The resin, carbon black and stabilizers are introduced to an exruder which supplies heat,
mixuig action and filtering. Utlien forces the tnoltentraterial to exit tta^
small orifices called a "spinnerette". Here the fibers, called "yams", are usually drawn (work
hardened) by mechanical tension, or impinged by air, as they are stretched and cooled.  The
resulting yams, called "filaments", can be wound onto a bobbin, or can be used directly to form
the finished product. Other yarn manufacturing variations include those made from staple fibers
and flat, tape-like, yarns called "slit-film". Each type (filament, staple or slit-film) can be twisted
together with omen as shown in Fig. 6.3.  Note that "yam" is a generic term for any continuous
strand (fiber, filament or tape) used to form a textile fabric.  Thus all of the examples in Fig. 6.3
are yams, except for staple, and can be used to manufacture geotextiles.
                      n      n        ii        n
                             i
                                  ^
                                  ^
                                  s
                                  ^
            ¥••
               Multif,
lirwnt
Supto
Fiben
                                     If      1         IT
                                  /      /
                                  to        /
         Supie
         Y«n
  Slit-film
Monofilament
   Yam
             Slic-film
            FiMtltted
Figure 6.3 - Types of Polymeric Fibers Used in the Construction of Different Types of Geotextiles
                                          206

-------
&2.1.3 Geotextile, Types

      The yams just described are joined together to make a fabric, or geotextile.  Generic
classifications are woven, nonwoven and knit Knit geotextiles. however, are rarely used in waste
containment systems and will not be described further in this document

      The manufacturer of a woven geotextile uses the desired type of yarn from a bobbin and
constructs the fabric on a weaving loom.  Fabric weaving technology is well established over
literally centuries of development  Most  woven fabrics used for geotextiles are "simple", or
"basket-type" weaves consisting of each yam going over and under an intersecting yarn on an
alternate basis. Figure 6.4(a) shows a micrograph of a typical woven geotextile pattern.

      In contrast to this type of uniformly woven panne are nonwoven fabrics as shown in Figs.
6.4(b) and (c). Here the yams are utilized directly from the extruding spinncrette and laid down on
a moving belt in a random fashion. The speed of the moving belt dictates the mass per unit area of
the final product  While positioned on the belt the material is "lofty",  and  the yarns are  not
structurally bound in 
-------
   it III lit I II I LI I ill ill Ml Llf II I tilt
   | Illlill aJlMIftlllllMlftill.ll*llll
   1 i 1 1 tl 1 1» • i I • 1 • 1 1 1 • II I II J It I * • I M •
   iiimiLiiiitiiiiLii.iittlfcltllltJlt
   ft iii s LI i tit ii i ill Lie iii ftiiina.il 1.1
   lilliilLlllltltlilltlllllMffcllllf
   in M i *i • ii in i n i ti i tiibiiiii&i tt
   Ilillllltllllf illillLlliMLIIfrCltl
  • lillftllftlllllllltllLllfttlMtfctfftf
  I LI 1 1.1 1 t,B I »l I it tftl f i-1 1 EMUl tfrtf til
  ItlMllllMIMIlllllirfltllitlMtft
          (a) Woven Geotextile at 4X Magnification
   (b) Nonwoven Needlepunched Geotextile at 24X Magnification

Figure 6,4 - Three Major Types of Gcoiextiles (Continued on Next Page).


                     20K

-------
           (c) Nonwovcn Heatbonded Geotextile at 24X Magnification
  Figure 6.4 - Three Major Types of Geotextilcs (Continued from Previous Page)
5.  The ultraviolet light resistance should be specified which is usually a certain percentage
    of strength or elongation retained after exposure in a laboratory weathering device,
    Usually ASTM D-4355 is specified and retention after 500 hours is typically 5(Wc to
    90%.

6.  The frequency of performing each of the preceding tests should  be covered  in the
    manufacturer's MQC plan and it should be implementedand followed.

7,  Verification  thai needle-punched, nonwoven  geoiextiles have  been  inspected
    continuously for the presence of broken needles using an in-line mewl detector with an
    adequate sweep rate should be provided. Furthermore, a needle removal system, e.g.,
    magnets, should be implemented.

8.  A statement indicating if. and to what extent, reworked polymer, or fibers, was added
    during manufacturing. If used, the statement should note that the rework polymer, or
    fibers, was of die same composition as the intended product.

9.  Reclaimed or recycled, i.e., fibers or polymer that has been previously used, should
    not be  added to the  formulation  unless specifically allowed for in ihe project
                                    209

-------
          specifications.  Note, however, that reclaimed fibers may be used in geotextiles in
          certain waste containment applications. The gas collection layer above the waste and
          the geotextile protection layer between drainage store and a geomembrane are likely
          locations. These should be design decisions and should be made accordingly.

6.2.2 Handling of Geote^tiles

      A number of activities occur between the manufacture of geotextiles and their final
positioning at the waste facility.  These activities involve protective wrapping, storage at  the
manufacturing facility, shipment, storage at the site, product acceptance, conformance testing and
final placement at the facility. Each of these topics will be described in this section.

       Protective Wranning
      All rolls of geotextiles, irrespective of their type, must be enclosed in a protective wrapping
that is opaque and waterproof.  The object is  to prevent any degradation from atmospheric
exposure (ultraviolet light, ozone, etc.), moisture uptake (rain, snow) and to a limited extent,
accidental damage.  It must be recognized that geotextiles are the most sensitive of all geosynthetics
to degradation induced by ultraviolet light exposure. Geotextile manufacturers use tightly wound
plastic wraps or loosely fit plastic bags for this purpose. Quite often the plastic is polyethylene in
the thickness  range of 0.05 to 0.13  mm  (2 to 5 mil).  Several  important issues should be
considered in a specification or MQA document

       1.  The protective wrapping should be wrapped around (or placed around) the geotextile in
          the manufacturing facility and should be included as the final step in the manufacturing
          process.

      2.  The packaging should not interfere with the handling of the rolls either by slings or by
          die utilization of the central core upon which the geotextile is wound.

      3.  The protective wrapping should prevent exposure of the geotextile to ultraviolet light,
          prevent it from moisture uptake and limit minor damage to the roll.

      4.  Every roll must be labeled with the manufacturers name, geotextile style and type, lot
          and roll numbers, and roll dimensions (length, width  and gross weight). Details
          should conform to ASTM D-4873.

6.2.2.2 Storage at Manufacturing Facility

      The manufacturing  of geotextiles is  such  that temporary storage  of rolls at the
manufacturing facility is necessary. Storage times range from a few days to a year, or longer.
Figure 6.5(a.) shows geotextile storage at a manufacturer's facility.

      Regarding specification and MQA document items, the following should be considered.

      1.  Handling of rolls of geotextiles should be done in a competent manner such that
          damage does not occur to the geotextile nor to its protective wrapping. In this regard
          ASTM D-4873 should be referenced and followed.

      2.  Rolls of geotextiles should not  be  stacked  upon one another to the extent that
          deformation of the core occurs or to the point where accessibility can cause damage in
          handling.


                                          210

-------
              (a) Storage at Manufacturing Facility
                    (b) Storage at Field Silt-
Figure 6.5 - Photographs of Temporary Storage ot"Cicoicxiiics
                           211

-------
      3.  Outdoor storage of rolls at the manufacturer's facility should not be longer than six
          months. For storage periods longer than six months a temporary enclosure should be
          put over the rolls, or they should be moved to within a enclosed facility.

6.2.2.3 Shipment

      Geotextile rolls are shipped from the manufacturer's (or their representatives) storage
facility to the job site via common carrier.  Ships, railroads and trucks have all been  used
depending upon the locations of the origin and final destination. The usual carrier from within the
USA, is truck. When using flat-bed trucks the rolls are usually loaded by means of a crane with
slings wrapped around the individual rolls. When the truck bed is closed, i.e., an enclosed trailer,
the rolls are usually loaded by fork lift with a "stinger" attached. The "stinger" is a long tapered
rod which fits inside the core upon which the geotextile is wrapped.

      Insofar as specification and MQA/CQA documents are concerned the following items
should be considered.

      1.  The method of loading  the geotextile rolls, transporting them and off-loading them at
          the job site should not cause any damage to die geotextile, its core, nor its protective
          wrapping.

      2.  Any protective wrapping that is accidentally damaged or stripped off of the rolls should
          be repaired immediately or the roll should be moved to a enclosed facility until its repair
          can be made to the approval of the CQA personnel.

6.23.4 Storage at Field Site

      Off-loading of geotextile rolls at  the site and temporary storage which must be done  in an
acceptable manner. Figure 6.5(b) shows typical storage at the field site. Some specification and
CQA document items to consider are the  following.

      1.  Handling of rolls of geotextiles should  be done in a competent  manner  such that
          damage does not occur  to the geotextile nor to its protective  wrapping. In this regard
          ASTM D-4873 should be referenced and followed.

      2.  The location of field storage  should not be in areas where water can accumulate. The
          rolls  should be elevated off of the ground so as not to form a dam creating the ponding
          of water.

      3.  The roils should be  stacked in such a way  that cores are-jiot crushed nor is the
          geotextile damaged. Furthermore, they should be  stacked in such a way that access for
          conformance testing is possible.

      4.  Outdoor storage of rolls should not exceed manufacturers recommendations or longer
          than  six months, whichever is less. For storage periods longer than six  months a
          temporary enclosure should be placed over the rolls, or they should be moved within an
          enclosed facility.

6.2J2.5 Acceptance and Conformance Testing

      Upon delivery of the rolls of geotextiles to the project site, and temporary stoiuge thereof,
the CQA engineer should see tha: conformance test samples are obtained. These samples are then


                                          212
                                                                                             _.ffca

-------
sent to the CQA laboratory for testing to ensure that die supplied geotextile conforms to the project
plans and specifications. The samples are taken from selected rolls by removing the protective
wrapping and cutting full-width, 1 m (3 ft) long samples off of the outer wrap of the selected
roU(s). Sometimes the outer revolution of geotextile is discarded before die test sample is taken.
The rolls are immediately re-wrapped and replaced in temporary field storage. The samples rolls
not be relabeled for future identification. Alternatively, conformance testing could be performed
at the manufacturer's facility and when  completed the particular lot should be marked for the
particular site under investigation. Items to be considered in a specification an-" CQA documents in
this regard are tne following:

      1. The samples should be identified by type, style or, lot and roll numbers. The machine
         direction should be noted on the samplers) with a waterproof marker.

      2. A lot is defined as a unit of production, or a group of other units or packages having
         one or more common properties and being readily separable from other similar units.
         Other definitions are also possible and should be clearly stated in the CQA documents,
         seeASTMD-4354.

      3. Sampling should  be done according to the job specification and/or CQA documents.
         Unless otherwise stated, sampling should be based on one per let Note that a lot is
         sometimes defined as 10,000 m2 (100,000 ft2) of geotextile. Utilization of ASTM D-
         4354 may be referenced and followed in  this regard but it might result in a different
         value for sampling than stated above.

      4. Testing at the CQA laboratory may include mass per unit area per ASTM D-5261, grab
         tensile strength per ASTM D-4632, trapezoidal tear strength per ASTM D-4533, burst
         strength per ASTM D-3786, puncture strength per ASTM D-4833, and possibly
         apparent opening  size per ASTM D-4751, and permittivity per ASTM D-4491. Other
         conformance tests may be required by die project specifications.

      5. Conformance test results should be sent to the CQA engineer prior to deployment of
         any geotextile from the lot under review.

      6. The CQA engineer should review the results and should report any nonconformancc to
         the Owner/Operator's Project Manager.

      7. The resolution of failing conformance tests must be clearly stipulated in the
         specifications or CQA documents.  Statements should be based upon ASTM D-4759
         entitled "Determining the Specification Conformance of Geosynthetics".

      8. The  geotextile rolls which are sampled  should be immediately rewrapped in their
         protective covering to the satisfaction of the CQA personnel.

6JL2.6 Placement

      The geosynthetic installation contractor should remove the protective wrappings from the
geotextile rolls to be deployed only after the substrate layer, soil  or other geosynthetic, has been
documented and approved by the CQA personnel. The specification and CQA documents should
be written in such a manner as to  ensure that the geotextiles are not damaged nor excessively
exposed to ultraviolet degradation. The following items should be considered for inclusion in a
specification or CQA document.
                                         213

-------
      1. The ia»aller shouki take the necessary pttcaudow to i»o^ the undcriying layers upon
         which the geotextile will be placed. If the substrate Is soil, construction equipment can
         be used provided that excess rutting is not created. Excess rutting should be clearly
         defined and quantified by the design engineer.  In some cases 25 ram (1.0 in.) is the
         maximum rut depth allowed. If the ground freezes, the depth of ruts should be further
         reduced to a specified value. If the substrate is a geosynthetic material, deployment
         must be by hand, by use of small jack lifts on pneumatic tires having low ground
         contact pressure, or by use of all-terrain vehicles, ATV's, having low ground contact
         pressure.

      2. During placement, care must be  taken not to entrap (either within or beneath the
         geotextile) stones, excessive dust or moisture that  could damage a georaembrane,
         cause clogging of drains or filters, or hamper subsequent seaming.

      3. On side stopes, the geotexriles should be ancnc>rcd at the top and then unroUed so
         keep tte geotextile oree of wrinkles and folds.

      4. Trimming of the geotextiles should be performed using only an upward cutting hook
         blade.

      5. Nonwoven geotextiles placed on textured geomembranes can be troublesome  due to
         sticking and are difficult to align or even separate after they are placed on one another.
         A thin sheet of plastic on the geomembrane during deployment of the geotextile  can be
         very helpful in this regard. Of course, it is removed after correct positioning of the
         geotextile.

      6. The geotextile should be weighted with sandbags, or the  equivalent,  to provide
         resistance against wind uplift This is a site-specific procedure and completely the
         installer's decision. Uplifted and moved geotextiles cvi generally be reused but only
         after approval by the owner and observation by the CQA personnel

      7. A visual exarrariation of the deployed geotextile should be earned out to ensure that no
         potentially  harmful  objects are present, e.g., stones, sharp objects, small  tools,
         sandbags, etc.

6-2.3  Seaming

      Seaming of geotextiles, by sewing, is sometimes required (versus overlapping with no
sewn seams) of all geotextiles placed in waste facilities.  This generally should be the case for
geotextiles used in filtration, put may be  waived for geotextiles used in separation (e.g.,  as gas
collection layers above the waste or as protective layers for geomembranes) as per the plans and
specifications.  In such cases, heat bonding is also an acceptable alternate method of joining
separation geotextiles.  In cases where overlapping is  permitted, the overlapped distance
requirements should be clearly stated in the specification and CQA documents. Geotextile seam
types and procedures, seam tests and geotextile repairs are covered in mis section.

6.2.3.1 ScfllTl Tvpes and Procedures

      The three types of sewn geotextile seams are shown in Fig. 6.6. They are *he "flat" or
"prayer" seam, the "J** seam and the "butterfly" seam.  While each can be made by a single thread,
or by a two-thread chain stitch, as illustrated, the latter stitch is recommended. Furthermore, a
single, double, or even triple, row of stitches can be made as illustrated by the dashed lines in the


                                          214

-------
figures. Figure 6.7 show:, a photograph of the fabrication of a flat seam and see Diaz (1990) for
further details regarding geotextile seaming.
                    SS*-1
                   SSn-l
                   SSd-1
  SSa-2             SSa-3
   TUt" or "Prayer" Soon
SSn-2               SSn-3
       TSean
          SSd-2

   "Butterfly* Scam
                 "101* Single Thread Chainstitch
                   "401* Two-Thread Chainnhch
      Figure 6.6 - Various Types of Sewn Seams for Joining Geotextiles (after Diaz, 1990)
                                           215

-------

Figure 6,7 - lubrication of a Geotextile Field Seam in a "Flat" or "Prayer" Seam Type
 The project specification or CQA documents should address the following considerations.

 1.  The type of seam, type of stitch, stitch count or number of stitches per inch and number
    of rows should be specified based on the tendency of the fabric to fray, strength need
    and toughness of the fabric.  For Filtration and separation geotextiles a flat seam using a
    two-thread chain stitch and one row is usually specified  For reinforcement geotextiles,
    stronger and more complex  seams arc utilized.  Alternatively, a minimum seam
    strength, per ASTM D-4HH4, could he specified.

 2.  The seams should be continuous,  i.e.. spot sewing is generally not allowed.

 3.  On slopes greater than  approximately 5  (horiz.) to 1  (vert.), seams should be
    constructed parallel to the slope gradient. Exceptions are permitted for small patches
    and repairs.

 4.  The thread typ-:  must he polymeric with  chemical and  ultraviolet light  resistant
    properties equal or greater than that of the gcntcxiiic itself.

-------
      5.  The color of the sewing thread should contrast that of the color of the geotextile for
          ease in visual inspection.  This may not be possible due to polymer composition in
          some cases.

      6.  Heat seaming of geotextiles may be permitted for certain seams. A number of methods
          are available such as hot plate, hot knife and ultrasonic devices.

      7.  Overlapped seams of geotextiles may be permitted for certain seams.  The overlap
          distance should be stated depending on the site specific conditions.

6.2.3,2 SeamTests

      For geotextiles used in filtration and separation, seam samples and subsequent strength
testing are not generally required.  If they are, however, they should be stipulated in the
specifications or CQA documents. Also, the sampling and testing frequency should be noted
accordingly. The test method to evaluate sewn seam test specimens is ASTM D-4884.

42.3.3 Repairs

      Holes, or tears, in geotextiles made during placement or anytime before backfilling should
be repaired by patching. Some relevant specifications and CQA document items follow.

      1.  The patch material used for repair of a hole or tear should be the same type of polymeric
          material as the damaged geotextile, or as approved by the CQA engineer.

      2.  The patch should extend at least 30 cm (12 in.) beyond any portion of the damaged
          geotextile.

      3.  The patch should be sewn in place by hand or machine so as not to accidentally shift
          out of position or be moved during backfilling or covering operations.

      4.  The machine direction of the patch should be aligned with die machine direction of the
          geotextile being repaired.

      5.  The thread should be of  contrasting color to the geotextile and of chemical and
          ultraviolet light resistance properties equal or greater than that of the geotextile itself.

      6.  The repair should be made to the satisfaction of die specification and CQA documents.

&Z.4 Backfilling or Covering
      MI^^BMMnMMM^^M^M^BMBftBMMBMI^P                                  fc_
                                                             1*
      The layer of material placed above the deployed geotextile will be either soil, waste or
another geosymhetic.   Soils will vary from compacted clay layers to coarse aggregate drainage
layers. Waste should be what is referred to as "select" waste, i.e., carefully separated and placed
m as not to cause damage. Geosynthetics will vary from geomembranes to geosynthetic clay
liners. Some considerations for a specification and CQA document to follow:

      1.  If soil is to cover the geotextile it should be done such that the geotextile is not shifted
          from its intended position and underlying materials are not exposed or damaged.

      2.  If a geosynthetic is to cover the geotextile, both the underlying geotextile and the newly
          deployed material should not be damaged during the process.
                                          217

-------
      3.  If solid waste is to cover the geotextile, (he type of waste should be specified and visual
          observation by CQA personnel should be required.
      4.  The overlying material should not be deployed such that excess tensile stress is
          mobilized in the geotextile.  On side slopes, this requires soil backfill to proceed from
          the bonom of the slope upward.

      5.  Soil backfilling or covering by another geosynthetic, should be done within the time
          .rame stipulated for the particular type of geotextile. Typical time frames for geotextiles
          are within 14 days for polypropylene and 28 days for polyester geotextiles.

6.3 Geojefean
-------
                 (a) Rolls of Drainage Ceonets
                (b) Closeup of Rib Intersection
Figure 6.8 - Typical Geonets Used in Waste Containment Facilities
                             219

-------
Polymer Feed
              Spreadini Mandrel
                    «nd
                Quench Tanh
    Figure 6.9 - Counter Rotating Die Technique (Left Sketch) for Manufacturing Drainage Gconets
               and Example of Laboratory Prototype (Right Photograph)

-------
      Regarding the preparation of a specification or MQA document for the resin component of
HOPE geonets, the following items should be considered:

      1. Specifications may call  for the  polyethylene resin to be made from virgin,
         uncontaminated ingredients.  Alternatively, geonets can be made with  off-spec
         gepmembrane material as a large, or even major part, of their total composition provided
         this material is of the same formulation as the intended geonet and does not consist of
         recycled and/or reclaimed material Recycled and/o- reclaimed material is generally not
         allowed. It is acceptable, and is almost always the case, that the density of the resin is in
         the medium density range for polyethylene, i.e., that its density is equal to or less than
         0.940 g/cc.

      2. Typical quality control tests on die resin are density, via ASTM D-1505or D-792 and
         melt flow index"via ASTM EM 238.

      3. An HOPE geonet formulation should consist of at least 97% of polyethylene resin, with
         the balance being carbon black and additives. No fillers, extenders, or other materials
         should be mixed into the formulation.       /
                                                /
      4. It should be noted that by adding carbon bl/tck and additives to the resin, the density of
         the final formulation is generally over 0.9','i  g/cc. Since this value is in the high density
         polyethylene category, according to AST?, i D-1248, geonets of this type are customarily
         referred to as  high density polyethylene C.-1DPE).

      5. Regrind or reworked polymer which/is previously processed HDPE geonet in chip
         form, is often added to the extruder d/iring processing. It is acceptable if it is the same
         formulation as the geonet being produced.

      6. No amount of "recycled" or "reclaimed" material, which has seen prior use in another
         product should be added to the formulation.

      7, An acceptable variation of the process just described is to add a foaming agent into the
         extruder which then is processed in the standard manner.  As the geonet is formed and is
         subsequently  quenched, the founing agent expands within the ribs creating innumerable
         small spherical voids.  The voids are approximately 0.01 mm (0,5 mil) in diameter.
         This type of  geonet is calico a "foamed rib" geonet, in contrast to the standard type
         which is a "solid rib" geonet. Foamed rib geonets are currently seen less frequently in
         drainage systems than previously.

      8. Quality control certificate-; from the manufacturer should  include proper identification of
         the product and style and results of quality control tests.
                                                              >*
      9. The frequency of performing each of the preceding tests should be covered in the MQC
         plan and it should be implemented and followed.

6.3.2  Handling of Geonets

      A number of activities occur between the manufacture of geonets and their final positioning
where intended at the waste  facility.  These activities involve packaging, storage at the
manufacturing facility, shipment, storage at the site, acceptance and conformance testing and final
placement at the facility, Each of these topics will be described  in this section.
                                          221

-------
6.3.2.1 Packaging

      As geonets come from the quenching tank they are wound on a core until the desired length
is reached. The geonet is then cut along its width and the entire roll contained by polymer straps so
as not to unwind during subsequent handling. There is generally no protective wrapping placed
around geonets, however, a plastic wrapping can be provided if necessary.

      Specifications or a MQA document should be formed around a few important points.

      1 . The core must be stable enough to  support the geonet foil while it is handled by either
         slings around it, or from a fork lift "stinger** inserted in it.

      2. The core should have a minimum 100 mm  (4.0 in.) inside diameter.

      3. The banding straps around the outside of  the roll should be made from materials with
         adequate strength yet should not damage the outer wrap(s) of the roll

6.3.2.2 $toragt; at Manufacturing Faculty

      The storage of geonet rolls at the manufacturer's facility is similar to that described for
HDPE geomembranes. Refer to Section 3.3.1 for a complete description.

6.3.2.3
      The shipment of geonet rolls from the manufacturer's facility to the project site is similar to
that described for HDPE geomembranes. Refer to Section 3.3.2 for a complete description.

6.3.2,4 Sfomge at the Site

      The storage of geonet rolls at the project site is similar to that described with HDPE
geomembranes. Refer to section 3.3.2 for a complete description, see Fig. 6.10. An important
exception is that a ground cloth should be placed under the geonets if they are stored on soil for
any tune longer than one month. This is to prevent weeds from growing into die lower rolls of the
geonet.  If weeds do grow in the geonet during storage, the broken pieces must be removed by
hand on the Job when the geonet is deployed.
6.3.2*5
                     Confermaiiee. Tsrin
      The acceptance and conformance testing of geonets is similar to that described for HDPE
geomembranes.  Refer to Section 3.3.3 for a complete description.  For geonets, the usual
conformance tests are the following:

      • density, per ASTM D-1505 or D-792

      * mass per unit area, per ASTM D-5261

      » thickness, per ASTM D-5199

Additional conformance tests such as compression per ASTM D-1621 and transmissivity per
ASTM D-47 1 6 may also be stipulated,
                                          222

-------
                Figure 6.10 - Geonets Being Temporarily Stored at the Job Site
6.3.2.6 Placement

      The placement of geonets in the field is similar to that described for geotextiles.  Refer to
Section 6.2.2.6 for a complete description.
6.3.3 Joining of Geonets
     MMMMHM^MMMBBIMMMMMMaM
              are generally joined together by providing a stipulated overlap and using plastic
fasteners or polymer braid to tie adjacent ribs together at minimum intervals, see Ing. 6.1 1 .

      Recommended items for a specification or CQA document on die joining of geonets include
the following:

      1 .  Adjacent roll edges of geonets should be overlapped a. minimum distance. This is
          typically 75-100 mm (3-4 in.).

      2.  The roll ejdl of geonets should be overlapped 150-200 mm (6*8 in.) since flow is
          usually in the machine direction.
                                          223

-------
                        |>S^^'i^^^fe:^S^

      Figure 6. II - Photograph ot'Oeonet Joining by Using Plastic Fasteners
3.  Ail overlaps should be joined by tying with plastic fasteners or polymeric braid,
            lies or fasteners are not allowed.
4 .  The tying devices should be white or yellow, as contrasted to the black geonet, for ease
    of visual inspection.

5.  The tying interval should be specified. Typically tie intervals are every 1.5 in (5.0 ft)
    along the edges and every 0. 15 m (6.0 in.) along the ends dud in anchor trenches.

6.  Horizontal seams should not be allowed on side slopes. This requires that the length of
    the  geonet should be at least as long as the side slope, anchor trench and a minimum run
    out  at the bottom of the facility.  If horizontal seams are allowed, they should be
    staggered from one roll to the adjacent roll.

7.  In difficult areas, such as  corners  of  side slopes, double layers  of geonets are
    sometimes used. This should be stipulated in the plans and specifications.

 8.   If  double geonets are used, they should be  layered on top of one another such that
     interlocking does not occur.
                                     224

-------
      9.   If double geonets are used, roll edges and ends should be staggered so that the joints
           do not lie above one another.

      10.  Holes or tears in the  gepnet should be repaired by placing a geonet patch extending a
           minimum of 0.3 m (12 in.) beyond the edges of the hole or tear.  The patch should be
           tied to the underlying geonet at 0.15 m (6.0 in.) spacings.

      11.  Holes or tears along  more than 50% of the width of the geonet on side slopes should
           require the entire length of geonet to be removed and replaced.

6.3.4 Geonet/Geotextile Geocomposites

      Geonets are always covered with either a geomembrane or a geotextile, i.e., they are never
directly  soil covered since the soil particles would fill the apertures of the geonet rendering it
useless.  Many geonets have a  geotextile bonded to one, or both, surfaces. These are then referred
to as  geocpmposites in the  geonet  manufacturer's  literature.  In this document, however,
geocomposites will refer to many different types of drainage core structures.  Clearly, covered
geonets  are included in this group. However, geocomposites also consist of fluted, nubbed and
cuspated cores, covered with geotextiles and/or geomembranes and will be described separately in
section 6.4.  Still further, some manufacturers refer to the entire group of geosynthetic drainage
mfltPPflly as "geospacers".

      Regarding a specification  or CQA document for geonet/geotextile drainage geocomposites,
a few comments are offered:

      1.  The geotextile(s) covering a geonet should be bonded together in such  a way that
          neither component  is  compromised to the point where proper functioning is impeded.
          Thus  adequate, but  not excessive, bonding of the geotextile(s)  to  the geonet is
          necessary.

      2.  If bonding is by heating, the geotextile(s) strength cannot be compromised to the point
          where failure could occur.  The transmissivity under load test, ASTM D-4716, should
          be performed on the intended geocomposite product.

      3.  If bonding is by adhesives, the type of adhesive must be identified, including its  water
          solubility and organic content. Excessive adhesive cannot be used since it could fill up
          some of the geonet's  void space.  The transmissivity under load  test, ASTM D-4716,
          should  be performed  on  the  intended geocomposite product.  The geotextile's
          permittivity could be evaluated using ASTM D-4491. ^

      4.  If the shear strength of the geotextile(s) to the geonet is of concern an adapted form of
          an interface shear test, e.g., ASTM D-5321, can be performed with the geotextile firmly
          attached to a wooden  substrate, or other satisfactory arrangement Alternatively, a ply
          adhesion test may be adequate, see ASTM D-413 which might be suitably modified for
          geotextile-to-gconet adhesion.

      5.  For factory fabricated geocomposites with geotextiles placed on both sides of a geonet,
          the geonet must be  free from all din, dust and accumulated debris before covering.
                                          225

-------
      6.  For field placed geotextiles, the geonet should be free of all soil, dust and accumulated
          debris before covering with a geomembrane or geotextile. In extreme cases this may
          require washing of the geonet to accumulate the paniculate material at the low end
          (sump) area where it is subsequently removed by hand.

      7.  When placing geosynthetk clay liners (p*i$) above geocomposites, cleanliness is
          particularly important in assuring that fugitive bentonitc clay panicles do not find their
          way into the geonet

      8.  Placement of a covering geomembrane should not shift the geotextile or geocomposite
          out of position nor damage the underlying geonet

      9.  An overlying geomembrane or geotextile should not be deployed such that excess
          tensile stress is mobilized in die geocompo*:;-.

6.4 Other Typey of Geocompo^ps

      Geocomposite drainage systems consist of a polymer drainage core protected by a geotextile
acting as both a filter and a separator to the adjacent material.  Thus a geonet, with a geotextile
attached to one surface or to both surfaces as described in section 6.3.4, is indeed a drainage
geocomposite. However, for the drainage geocomposites discussed in this section the geotextile
filter is always attached to the drainage core and the core can take a wide variety of non-geonet
shapes and configurations. In some cases, the geotextile is only on one side of the core (the side
oriented toward the inflowing liquid), in other cases it is wrapped completely around the drainage
core.

      There are three different types of drainage geocomposites referred to in this document; sheet
drains, edge drains and strip (or wick) drains.  Typical variations are shown in Fig.  6.12. For
drainage systems associated with waste containment facilities, sheet drains, Fig.  6.12a, are
sometimes used as surface water collectors and drains in cover systems of closed landfills and
waste piles, refer to Fig. 6.1. Infiltration water that moves within the cover soil enters the sheet
drain and flows gravitationally to the edge of the site (or cell) where it is generally collected by a
perforated pipe, or edge drain.   Pipes will be discussed separately  in Chapter 8.  The other
possible use for sheet drains is for primary leachate collection systems in landfills. The required
flow rate in some landfills is too great tor a geonet, hence the  greater drainage capacity of a
geocomposite is sometimes required.  Of course, when used in this application the drainage
geocomposite must resist the compressive and shear stresses imposed byjhc waste and it must be
chemically resistant to the leachate, but these are design consideratipns/The use of strip (wick)
drains, Fig. 6.lib, in waste containment has been as vertical drains within a solid waste landfill to
promote leachate communication between individual lifts. The edge drains, shown in Fig. 6.12(c),
nave potential applicability around the perimeter of a closed landfill facility to accumulate the
surface water coming from a cap/closure system. A variety of perimeter drains could utilize such
geocomposite edge drains.

      Of the different types of drainage geocomposites shown in Fig. 6.12, only sheet drains will
be described since they have the greatest applicability in waste containment systems.
                                          226

-------
                     (a) Gcocomposite Sheet Drains
                  (b) Gcocomposite Snip (Wick) Drains
Figure 6,12 - Various Types of Drainage Geocomposites (Continued on Next Page)
                                 227

-------
                             (c) Geocomposite Edge Drains
    Rgure 6.12 - Various Types of Drainage Geocomposites (Continued from Previous Page)
6.4.1 Manufacturing of Drainage Composites
      The  manufac"'ire of the drainage core of a geocomposite sheet  drain is generally
accomplished by taking the desired type of polymer sheet and then vacuum forming dimples,
protrusions or cuspations which give rise to the protrusions.  The polymer sheets of drainage
geocomposites have been made from a wide variety of polymers, Cornjnercial products that are
currently available consist of the following polymer formulations:      "*
      * polystyrene
      • nylon
      • polypropylene
      • polyvinyl chloride
      * polyethylene
      • polyethylene/poIystyrenVrx>lyethylene(cc*xtnision)
                                         228

-------
With coextrasion there exists a variety of possibilities in addition to those listed above. Recognize,
however, that coarse fibers, entangled webs, filament mattings, and many other variations are also
possible.

      Upon deciding on the proper type and thickness of polymer sheet, a geocomposite core
usually goes through a vacuum forming step. In this step a vacuum draws portions of the polymer
sheet i"to cusps at prescribed locations. Depending on the particular product, the protrusions are at
12 to 25 mm (0.5 to 1.0 in.) centers and are of a controlled depth and shape. Figure 6.13 shows a
sketch of a vacuum forming system. In many of the systems the protrusions are tapered for ease in
manufacturing during release of the vacuum and for a convenient maJe-to-fcmale coupling of the
edges and/or ends of the product in the field.  The different types of drainage geocomposites are
made in either continuous roils or in discrete panels.
                          Infrarad HMt«r»
                                                                 Extruded
                                                                 Sh**t
                             Vacuum
                                                                  D*form*d
                                                                  Stwat
                              Vacuum
       Figure 6.13* Vacuum Forming System for Fabrication of a Drainage Geocomposite
       The geotextile, which acts as both a filter to allow liquid into the drainage core and as a
separator to keep soil out of the core by spanning from cusp to cusp is put onto the core as a
secondary operation. Quite often an adhesive is placed on the  tops of the cusps to adhere the
geotextile to the core. Alternatively, heat bonding can be utilized. A variety of geotextiles can be
                                           229

-------
used and the site specific design will dictate the actual selection. As far as the MQA/CQA of the
geotextile it is the sanx as was described in Section 6.2.

      There are several items which should be included in a specification or MQA document for
drainage geocoraposite cores.

      1.  There should be verification and certification that the actual geocomposite core
          properties meet the manufacturers specification for that particular type and style.

      2.  Quality control certificates should include at a minimum, polymer composition,
          thickness of sheet per ASTM D-5199, height of raised cusps, spacing of cusps,
          cornpressive strength behavior (both strength and deformation values at core failure) per
          ASTM D-1621, and transmissivity using site specific conditions per ASTM D-4716.

      3.  For drainage  systems consisting of coarse fibers, entangled webs and/or filament
          mattings the thickness under load per ASTM D-5199 and transmissivity under load per
          ASTM D-4716 are the main tests for QC purposes.

      4.  Values for each property should meet, or exceed, the manufacturers listed values or the
          project specification values, whichever are higher.

      5.  A statement indicating  if, and to what extent, regrind  polymer was added  during
          manufacturing. No amount of reclaimed polymer should be allowed.

      6.  The frequency of performing each of die preceding tests should be covered in the MQC
          plans and it should be implemented and followed.
      Additionally, there are several items which should be included in a specification or MQA
document for the geotextik(s)/dnunage core geocomposite.

      1.  The type of geotextile(s) should be identified and properly evaluated. See section 6.2
          for these details,

      2.  For strip (wick) drains and edge drains, see Figs. 6,12(b) and (c) respectively, the
          geotextile complete surrounds the drainage core and generally no fixity is required.  For
          sheet drains, Fig. 6.12{a), this is not the case.

          The geotextile(s) covering of a drainage core should be bonded in such a way that
          neither component is compromised to the point where proper functioning is impeded.
          Thus adequate, but not excessive, bonding of the geotextile(s) to the drainage core is
          necessary.

          If bonding is by heating, the geotextile(s) strength cannot be compromised to the point
          where failure could occur. The transmissivity under load test, ASTM D-4716, should
          be perfoiined on the intended geocomposite product.

          If bonding is by adhesive*, the type of adhesive must be identified, including its water
          solubility and organic content Excessive adhesive cannot be used since it could fill up
          some of the drainage core's void space. The transmissivity under load test, ASTM D-
          4716, should be performed on the intended geocomposite product The geotextile's
          permittivity could be evaluated using ASTM D-4491.
                                         230

-------
      6.  If the shear strength of the geotextile(s) to the core is of concern an adapted form of an
          interface shear test, e.g., ASTM D-5321, can be performed with a wooden substrate, or
          other satisfactory arrangement Alternatively, a ply adhesion test may be adequate, see
          ASTM D-413 which might be suitably modified for geotex'ile-to-core adhesion.

      7.  For factory fabricated geocomposites with geotextiles pitted  on both sides of the
          drainage core, the core must be free from all din, dust ind accumulated debris before
          covering.

6.4.2 Handling of Drainage Geocomposites

      A number of activities occur between the manufacture of drainage geocomposites and their
final positioning where intended at the waste facility. These activities involve packaging, storage at
the manufacturing facility, shipment, storage at the site, acceptance and conformance testing, and
final placement at the facility. Each of these topics will be described although most will be by
reference to the appropriate geotextiie section.

6.4.2.1  Packaging

      Usually a manufacturer will not attach the geotextile to the core until an order is received
and shipment is imminent. Thus warehousing is not a major issue. The cores are cither rolled
onto themselves or are laid flat if they are in panel form.  When an order is received, die geotextiie
is bonded  to the core, the rolls are banded together with polymer straps and, if panels, they are
banded in a similar manner.

6.4.2.2  Storage at Manufacturing Facility

      Storage of the drainage core: 
-------
      *   thickness of raised cusps per ASTMD-1621

      *   spacing of raised cusps per ASTMD-1621

Optional conformance tests such as compression per ASTM D-1621 and transmissivity per ASTM
D^4716 may also be stipulated. The frequency of conformance tests of the drainage core must be
stipulated. In general, one test per 5,000 m2 (50,000 ft2) should be the minimum test frequency.

6.4.2,6 Placement

      The placement of drainage geocomposites in the field is similar to that described for
geotextilcs. Refer to Section 6.2.2.6 for details.

6.4.3 Joining i^fPrainagf! fieocomposite^

      Drainage geocomposites are usually joined together by folding back the geotextile from the
lower core and inserting it into the bottom void space of the upper core, see Fig. 6.14,  Where this
is not possible a tab should be available at the edges of the core material for the purpose of
overlapping.  The geotextile must be refolded over the connection area assuring a complete
covering of the core surface.
        Figure 6.14 - Photograph of Drainage Core Joining via Male-to-Female Interlock
                                         232

-------
      Recommended '.terns for a specification or CQA document on the joining of drainage
geocompositea include the following:

      1.  Adjacent edges of drainage cores should be overlapped for at least two tows of cusps.

      2.  The ends of drainage cores (in the direction of flow) should be overlapped for at least
          four rows of cusps.

      3.  The geotextiles co. ering the joined cores must provide a complete seal against backfill
          soil entering into the core.

      4.  Horizontal scams should not be allowed on sideslopes. This requires that the drainage
          geocomposiie be provided in roils which are at least as long as the side slope.

      5.  Holes or tears in drainage cores are repaired by placing a patch of the same type of
          material over the damaged area. The patch should extend at least four cusps beyond the
          edges of the hoi; or tear.

      6.  Holes or tears of more than 50% of the width of the drainage core on side slopes should
          require the entire length of the drainage core to be removed and replaced.

      7.  Holes or tears in the geotextile covering the drainage core should be repaired as
          described in Section 6.2.3.3.

6,4.4  Covering

      Drainage geocomposites, with an attached geotextile, are covered with either soil, waste or
in some cases a geo membrane. Regarding a specification or CQA document some comments
should be included.

      1.  The core of the drainage geocomposite should be free of soil, dust and accumulated
          debris before backfilling or covering with a geomembrane.  In extreme cases this may
          require washing of the core to accumulate die paniculate material to the low end (sump)
          area for removal.

      2.  Placement of the backfilling soil, waste or geomembrane should not shift the position of
          the drainage  geocomposite nor damage the underlying  drainage geocomposite,
          geotextile or core.

      3.  When using soil or waste as backfill on side slopes, the work progress should begin at
          the toe of the slope and work upward.

&S Reference*
ASTM EMI 3, "Rubber Property-Adhesion to Flexible Substrate"

ASTM D-792, "Specific Gravity and Density of Plastics by Displacement"

ASTM D-1238, "Flow Rates of Thermoplastics by Extrusion Plastometcr"

ASTM D-1248, "Polyethylene Plastics and Extrusion Materials'*
                                         233

-------
ASTM D-1505, "Density of Plastics by the Density-Gradient Technique"
ASTM D-1603. "Carbon Black in Olefin Plastics"
ASTM D-1621, "Compressive Properties of Rapid Cellular Plastics"
ASTM D-3786, "Hydraulic Bursting Strength of Knitted Goods and Nonwoven Fabrics:
    Diaphragm Burning Strength Tester Method"
ASTM D-4354, "Sampling of Geosynthetics for Testing"  I
ASTM D-435S, "Deterioration of Geotextiles from Exposure to Ultraviolet Light and Water
    (Xenon-Arc Type Apparatus)"
ASTM D-4491, "Water Permeabilby of Geotextiles by Perminsviry*'
ASTM D-4533, "Trapezoidal Tearing Strength of Geotextiles"
ASTM D-4632, "Breaking Load and Elongation of Geotextiles (Grab Method)"
ASTM D-4716, "Constant Head Hydraulic Transmissivity On-Plane Flow) of Geotextiles and
    Geotextik Related Products"
ASTM D-4751, "Determining the Apparent Opening Size of a Geotextik"
ASTM D-4759, "Determining the Specification Conformance of Geosynthetics"
ASTM D-4833, "Index Puncture Resistance of Geotextiles, Geomembranes and Related Products"
ASTM D-4873, "Identification, Storage and Handling of Geosynthetics"
ASTM D-4884, "Seam Strength of Sewn Geotexoles"
ASTM D-5199, "Measuring Nominal Thickness of Geotextiles and Geomembranes"
ASTM D-5261, "Measuring Mass Per Unit Area of Geotextiles'1
ASTM D-5321, "Determining the Coefficient of Soil and Geosynthetic or Geosynthetic and
    Geofynthetic Fric^on by the Direct Shear Method"
Diaz, V, A. (1990), The Seaming of Geosynthetics," IFAI Publ., St. Paul, MN, 1990.
IF AI (1990), "A Design Primer, Geotextiles and Related Materials," Industrial Fabrics Association
    International, St. Paul, MN.
                                        234

-------
                                      Chapter 7

                                Vertical Cutoff Walls
7.1    TnpnductJQn

       Situations occasionally arise in w1 ich it is necessary or desirable to restrict horizontal
movement of liquids with vertical cutoff walls. Examples of the use of vertioJ cutoff walls include
the following:

    1.  Control of ground water seepage into an excavated disposal csll to maintain stable side
        slopes or to limit the amount of water that must be pumped from the excavation during
        construction (Fig. 7.1).

    2.  Control of horizontal ground water flow into buried wastes at older waste disposal sites
        that do not contain a liner (Fig. 7.2).

    3.  Provide  a "seal" into an aquitard flow-permeability stratum), thus  "encapsulating" the
        waste to limit inward movement of clean ground water in areas where ground water is
        being pumped out and treated (Fig. 7.3).

    4.  Long-term barrier to impede contaminant transport (Fig. 7.4).


       Vertical walls are also sometimes used to provide drainage. Drainage applications are
discussed in Chapters 5 and 6.
               Pumpt Low«r Ground
               WiMrUwlSwMMtti
               Excavated C*l
Shiny Wii RwtitettW
Flow into th*C*l
   Figure 7.1 - Example of Vertical Cutoff Wall to Limit Flow of Ground Water into Excavation.
                                          235

-------

Figure 7.2 - Example of Vertical Cutoff Wall to Limit Flow of Ground Water through Buried
          Waste.
                       mtm*
                       •T™~
                                             .

                       ffjt&fefe^A?^
 W&tsf lowered from
and Treat Refnedia«on-_-j
   Figurc 7.3 • Example of Vertical Cutoff Wai! to Restrict Inward Migration of Ground Water.
    Figure 7.4  Example of Vertical Cutoff WaH to Limit 'xjng-Tcrm Contaminant Transport.
                                      236

-------
7.2    Types of Vertical Cutoff Wa|ls

        The principal types of vertical cutoff walls are sheet pile walls, geomembrane walls, and
slurry trench cutoff walls.  Other techniques, such as grouting and deep soil mixing, are also
        , but have rarely been used for waste containment applications.
7.2.1  Sheet Pile Walls

       Sheet pile walls are interlocking sections of steel or plastic materials (Fig. 7,5). Steel sheet
piles are used for a variety of excavation shoring applications; the same type of steel sheet piles are
used for vertical cutoff walls. Plastic sheet piles are a relatively recent development and are used
on a limited basis for vertical cutoff walls. Sheet piles measure approximately 0.5 m (18 in.) in
width, and interlocks join individual sheets together (Rg. 7.5). Lengths are essentially unlimited,
but sheet piles are rarely longer than about 10 to 15 m (30 to 45 ft).
                                                     Interlock
                         Figure 7.5 • Interlocking Steel Sheet Piles.
              Plastic sheet piles are different from geomembrane panels, which are discussed
later. Plastic sheet piles tend to be relatively thick-walled (wall thickness > 3 mm or 1/8 in.) and
rigid; geomembrane panels tend to have a smaller thickness (< 2.5 mm or 0.1 in.), greater width,
and lower rigidity.

       Sheet pile walls are installed by driving or vibrating interlocking steel sheet piles into the
ground.  Alternatively, plastic sheet piles can be used, but special installation devices may bf.
needed, e.g^ a steel driving plate to which the plastic sheet piles*re attached. To promote a seal, a
cord of material that expands when hydrated and attains a very low permeability may be inserted in
the interlock.  Other schemes have been devised and will continue to be de* Joped for attaining a
water-tight seai in the interlock.

       Sheet pile walls have a long history of use for dewatering applications, particularly where
the sheet pile wall is also used as a structural wall. Sheet pile walls also have been used on several
occasions to cutoff horizontal seepage through permeable strata that underlie darns (Sherard et al.,
1943).   %

       Sheet pile walls have historically suffered from problems with leakage through interlocks,
although much of the older experience may not be applicable to modern sheet piles with expanding
material located in the interlock (the expandable material is a relatively recent development).
                                           237

-------
Leakage through sheet pile interlocks depends primarily on the average width of openings in the
interlocking connections, the percentage of the interlocks that leak, and the quality and integrity of
any sealant placed in the interlock. The sheet piles may be damaged during installation, which can
create niptures in the sheet pUe material or separation of sheet piles at interlocks. Because of these
problems, sheet pile cutoffs have not been used for waste containment facilities as extensively as
some other types of vertical cutoff walls. Sheet pile walls are not discussed further in this report

7.2.2  neomembrane Walls

       Geomembrane walls represent a relatively new type of vertical barrier that is rapidly gaining
in popularity.  The georoembrane wall consists of a series of geomembrane panels joined with
—ial interlocks (examples of interlocks are sketched in Fig. 7.6) or installed as a single unit If
   i icomembrane panels contain interlocks, a water-expanding cord is used to seal the interlock.
Figure 7.6 * Examples of Interlocks for Geomembrane Walls (Modified from Manassero and
            Pasqualini, 1992)
       The technology has its roots in Europe, where slurry trench cutoff walls that are backfilled
with cement-bentonite have been commonly used for several decades. One of the problems with
cement-bentooite backfill, at discussed later, is that it is difficult to make the hydraulic conductivity
of die cetnem-bentonite backfill less than or equal to 1 x 10"7 cm/s, which is often required of
regulatory agencies in die U.S. To overcome this limitation in hydraulic conductivity and to
improve the overall containment provided by the vertical cutoff wall, a geomembrane may be
inserted into the cement-bentonite backfill. The geomembrane may actually be installed either in a
slurry-filled trench or it may be installed directly into the ground using a special insertion plate.
                                          238

-------
7.2.3  Walls Qmftnipted with Slurry Techniques

       Walls constructed by slurry techniques (sometimes called "slurry trench cutoff walls") are
described by Xanthakos (1979), D'Appolonia (1980), EPA (1984), Ryan (1987), and Evans
(1993). With this technique, an excavation is made to the desired depth using a backhoe or
clamshelL The trench is filled with a clay-water suspension ("mud" or "slurry"), which maintains
stability of sidewalls via hydrostatic pressure.  As the trench is advanced, the slurry tends to flow
into the surrounding soil.  Clay panicles are  filtered out, forming a thin skin of relatively
impermeable material along the wall of the trench called a "filter cake." The filter cake has a very
low hydraulic conductivity and allows the pressure from the slurry to maintain stable walls on the
trench (Fig. 7.7). However,  the level of slurry must generally be higher than the surrounding
ground water tabk in order to maintain stability. If the water table is at or above die surface, a dike
may be constructed to raise the surface elevation along the alignment of the slurry trench cutoff
wall.
            Wright of Starry
            CrMtc* Prwsum """""
            Ading on FttwCafc*

                                        r V* ','»* *W
                                    swwyw.
                                        II


                                              »s»«S££«£^
                                             ^^^^^^^>
                                              iiiiiilil^i *'
        Figure 7.7 - Hydrostatic Pressure from Slurry Maintains Stable Walls of Trench.
       In most cases, sodium bcntonitc is the clay used in the slurry. A problem with bentonite is
that it does not gel properly in highly saline water or in some heavily contaminated ground waters.
la such cases, an alternative clay mineral such as attapulgite may be used, or other special materials
may be used to maintain a viscous slurry.

       The slurry trench must either be backfilled or the slurry itself must harden into a stable
material - otherwise clay will settle out of suspension, the slurry will cease to support the walls of
the trench, and the walls may eventually collapse.  If the slurry is allowed to harden in place, the
slurry is usually a cemem-bentonite (CB) mixture.  If the slurry trench is backfilled, the backfill is
usually a soil-bentonite (SB) mixture, although plastic concrete may also be used (Evans, 1993).
                                           239

-------
      In the U.S., slurry trenches backfilled with SB have been the most commonly used vertical
cutoff trenches for waste containment applications.  In Europe, the CB method of construction has
been used more commonly. "Die reason for the different practices in the U.S. and Europe stems at
least in part upon die fact that abundant supplies  of high-quality sodium bentonite are readily
available in the  U.S. but not in Europe.  Also,  in most situations, SB backfill will have a
somewhat lower hydraulic conductivity than cured CB slurry, and in  the U.S. regulations have
tended to drive the requirements for hydraulic conductivity to tower values than in Europe.

      The construction sequence for a soil-bentonite backfilled trench is shown schematically in
Fig.  7.8.
                    Backfill
                  Mixing Area
Trench Spoils
Rgure 7.8 -   Diagram of Construction Process for Soil-Bentonite-Backfilled Slurry Trench
             CutoffWall.


      The main reasons why slurry trench cutoff walls are so commonly used for vertical cutoff
walk are:

      1.     The depth of the trench may be checked to confirm penetration to the desired depth,
             and excavated materials may be examined to confirm penetration into a particular
             stratum;

      2.     The backfill can be checked prior to placement to make sure that its properties are as
             desired and specified;


                                          240

-------
      3.     The wall is relatively thick, (compared to a sheet pile waM or a geomembrane wall);

      4.     There are no joints between panels or construction segments with the most common
             type of sluny trench cutoff wall construction.


      In scneral, in comparison to sheet-pile walls, deep-soil-mixed  -alls, and grouted walls,
there is more opportunity with a sluny trench cutoff wall to check the condition of the wall and
confirm mat the wall has been constructed as designed.  In contrast, it is much more difficult to
confirm that a sheet pile wall has been installed without damage, mat grout has fully penetrated aU
erf the desired pore spaces in the soil, or that deep mixing 35 taken place as desired.

7,3   Construction of Slurry Trench Cutoff Walk

      The major  construction  activities  involved  in building a slurry cutoff wall are
preconstruction planning and mobilization, preparation of the site* slurry mixing and hydration,
excavation of  soil, backfill  preparation, placement  of backfill, clean-up of the site, and
demobilization.  These activities are described briefly in the paragraphs that follow.
7.3.1
Mobilization
       The first major construction activity is to make an assessment of die site and to mobilize for
construction. The contractor locates the slurry trench cutoff wall in the field with appropriate
surveys. The contractor determines the equipment that will be needed, amounts of materials, and
facilities that may be required. Plans are made for mobilizing personnel and moving equipment to
the site.

       A preconstruction meeting  between the designer, contractor, and CQA engineer is
recommended.  In this meeting, materials, construction procedures, procedures for MQA of the
bentonite and CQA of all aspects of the project, and corrective actions are discussed (see Chapter
1).
7.3.2
Site Preparation
       Construction begins with preparation of the site.  Obstacles are removed, necessary
relocations of utilities are made, and the surface is prepared.  One of the requirements of slurry
trench construction is that the level of slurry in the trench be greater than the level of ground water.
If die ground water table is high, it may be necessary to construct a dike to ensure that the level of
slurry in the trench is above the ground water level (Fig. 7.9).  There may be grade restrictions in
the construction specifications which will require some regrading of the surface or construction of
dikes in low-lying areas.  The site preparation work will typically  also include preparation of
working surfaces for mixing materials.  Special techniques may be required for exacavation around
utility lines.
7.3.3
Slurrv Preparation and Properties
       Before excavation begins, as well as during excavation, the slurry must be prepared. The
slurry usually consists of a mixture of bentonitic clay with water, but sometimes omer clays such
as aaapulgjte are used.  If the clay is bentonite. (he specifications should stipulate the criteria to be
met, e.g., filtrate loss, and the testing technique by which the parameter is to be determined. The
criteria can vary considerably from project to project.
                                           241

-------
              High Water
              Table
Dike
  Figure 7.9 - Construction of Dike to Raise Ground Surface for Construction of Slurry Trench,
       The clay may be mixed with water in either a batch or flash mixing operation. In the batch
system specified quantities of water and bentonite are added in a tank and mixed at high speeds
with a pump, paddle mixer, or other device thai provides adequate high-speed colloidal shear
mixing. Water and clay are mixed until hydration is complete and the desired properties of the
slurry have been achieved. Complete mixing is usually achieved in a few minutes. The size of
batch mixers varies, but typically a batch mixer will produce several cubic meters of mixed slurry
at a time.

       Flash mixing is achieved with a venturi mixer.  With this system, bentonite is fed at a
predetermined rate into a metered water stream that is forced through a nozzle at a constant rate.
The slurry is subjected to high shear mixing for only a fraction of a second. The problem with this
technique is that complete hydration does not take place in the short period of mixing.  After the
claj is mixed with water, the resulting slurry is tested to make sure the density and viscosity are
within the requirements set forth in the CQA plan.

       The mixed slurry may be pumped directly to the trench or to a holding pond or tank. If the
slurry is stored in a tank or pond, CQA personnel should check thfl. properties of the  slurry
periodically to make sure that the properties have not changed due to thixptroptc processes or
sedimentation  of material from the slurry. The specifications for the project should stipulate
mixing or circulation requirements for slurry that is stored after mixing,

       The properties of the slurry used to maintain the stability of the trench are important.  The
following pertains to a bentonite slurry that will ultimately be displaced by soil-bentonitc or other
backfill; requirements for cement-b^ntonite slurry are discussed later in section 7.3.6. The slurry
must be sufficiently dense and viscous to maintain  stability of the trench. However, the slurry
must not be too dense or viscous: otherwise, it will be difficult to displace the slurry when backfill
is placed. Construction specifications normally set limits on the properties of the slurry. Typically
about 4-8% bentonite by weight is added to fresh water to form a slurry that has a specific gravity
of about 1 .OS to 1.15. During excavation of the trench additional fines may become suspended in
                                           242

-------
the slurry, and the specif'c gravity is likely to be greater than the value of the freshly mixed slurry.
The specific gravity of the slurry during excavation is typically on die order of 1.10 -1.25.

       The density of the slurry Is measured with the procedures outlined in ASTM D-4380.  A
known volume of slurry is poured into a special "mud balance," which contains a cup on one end
of a balance. The weight is determined and density calculated from the known volume of the cup.

       The  viscosity of the slurry is usually measured with a Marsh funnel. To  determine the
Marsh viscosity, fluid is poured into the funnel to a prescribed level. The numoer of seconds
required to  discharge 946 mL (1 quart) of slurry into a cup is  measured.  Water has a Marsh
viscosity of about 26 seconds at 23*G Freshly hydrated bemonite slurry should have a Marsh
viscosity in the range of about 40-50 seconds. During excavation, the viscosity typically
increases to  as high as about 65 Marsh seconds. If the viscosity becornes too large the thick slurry
must be replaced, treated (e.g., to remove sand), or diluted with additional fresh slurry.

       The  sand content of a slurry may also be specified. Although sand is not added to fresh
slurry, the slurry may pick up sand in the trench during the construction process.  The sand content
by volume is measured with ASTM D-4381. A special glass measuring tube is used for the test.
The slurry is poured onto a No. 200 sieve (0.075 mm openings), which is repeatedly washed until
the water running thrmigh the sieve is clear. The sand is washed into the special glass measuring
tube, and the sand con^ot (volumetric) is read directly from graduation marks.

       Other criteria may be established for the slurry. However, filtrate loss and density, coupled
with viscosity, are the primary control variables.  The specifications should set limits on these
parameters as well as specify the test method.  Standards of the American Petroleum Institute
(1990) are often cited for slurry test methods. Limits may also be set on pfi, gel strength, and
other parameters, depending on the specific application.

       The  primarily responsibility for monitoring the properties of t!ie slurry  rests with the
construction quality control (CQC) team.  The properties of the slurry directly affect construction
operations but may also impact the final quality of the slurry trench cutoff wall. For example, if
the slurry is too dense or viscous, the slurry may  not be properly displaced by backfill. On the
other hand, if the slurry is too thin and lacks adequate bemonite, the soil-bentonite backfill (formed
by mixing soil with the bentonite slurry) may also lack adequate bemonite. The CQA inspectors
may periodically perform tests on the slurry, but  these  tests are usually conducted primarily  to
verify test results from the CQC team. CQA persunrrl should be especially watchful to make sure
that: (1) the slurry has a sufficiently high viscosity and density (if not, the trench walls may
collapse); (2) the level of the slurry is maintained near the top of the trench and above the water
table (usually the  level must be at least 1 m above the  ground water table to maintain a stable
trench); and (3) the slurry does not become too  viscous or dense (otherwise backfill will not
properly displace the slurry).

7.3.4     Excavation of Blurry Trencrj

       The  slurry trench is excavated with a backhoe (Ing. 7.10) or a clam  shell (Fig. 7.11).
Long-stick  backhoes can dig to depths of approximately 20 to 25 m (60 to 80 ft). For slurry
trenches that can be excavated with a backhoe, the backhoe is almost always the most economical
means of excavation.  For trenches that ue too deep to be excavated with a backhoe, a clam shell is
normally  used.  The  trench may be excavated first with a backhoe to the  maximum depth  of
excavation that is  achievable with the backhoe and to further depths with a clam  shell. Special
chopping, chiseling, or other equipment may be used as necessary. The width of  the excavation
tool is usually equal to the width of the oench and is typically 0.6 to 1.2 m (2 to 4 ft).
                                           243

-------
                                i-*i   ,w. &  \K
                                ^x,*f   T.MI' ii  ^
                                Sifrvv   l-S»?  >&  i

                                          '       -
- - ;:/*„,;•*'". — .";
    ^3 - " 21 r" p" -

-------
                    Figure 7.11. Clamshell for Excavating Slurry Trench.
       Two parameters concerning the backfill are very important:w( 1) the presence of extremely
coarse material (i.e., coarse gravel and cobbles),  and (2) the presence of fine material.  Coarse
gravel is defined as material with particle sizes between 19 and 75 mm (ASTM D-2487).  Cobbles
are materials with panicle sizes greater than 75 mm.  Fine material is material passing the No. 200
sieve, which has openings of 0.075 mm. Cobbles  will tend to settle and segregate in the backfill;
coarse gravel may also segregate, but the degree of segregation depends on site-specific
conditions. In some cases, the backfill may have to be screened to remove pieces that exceed the
maximum size allowed in the specifications.  The hydraulic conductivity of the backfill is affected
by the percentage of fines present {D'Appolonia, 1980;  Ryan. 19M7; and Evans. 1993).  Often, a
minimum percentage of fines is specified.  Ideally, the backfill material should contain at least 10 to
30% fines to achieve low hydraulic conductivity (< 107 cm/s).
                                          245

-------
       The bentonite may be added in two ways: (1) soil is mixed with the bentonite slmry
(usually with a denser, as shown in Fig. 7.12) to form a viscous SB material; and (2) additional dry
powdered bentonite may be added to die soil-bentonite slurry mixture.  Dry, powdered bentonite
may or may not be needed. D'Appolcnia (1980) and Ryan (1987) discuss many of the details of
SB backfill design.
                     Figure 7.12 - Mixing Backfill with Bentonite Slurry.
       When SB backfill is used, a more-ur-less continuous process of excavation, preparation of
Backfill, and backfilling is used. To initiate the process, backfill is placed by lowering it to the
bottom of die trench, e.g.,  with a clamshell bucket, or placing it below the slurry surface with a
marie pipe (similar to a very long funnel) until the backfill rises above the surface of the slurry
trench at the starting point of the trench. Additional SB backfill is then typically pushed into the
trench with a dozer (Fig. 7.13). The viscous backfill sloughs downward and displaces the slurry
in the trench. As an alternative method to initiate backfilling, a separate trench that is not pan of the
final slurry trench cutoff wall, called a lead-in trench, may be excavated outside at a point outside
of the limits of the final slurry trench and backfilled with the procev just described, to achieve full
backfill at the point of initiarion of the desired slurry trench.
                                           246

-------
          Figure 7,13 - Pushing Soil-Bcntonite Backfill Into Slurry Trench with Dozer.
       After the trench has been backfilled, low hjdraulic conductivity is achieved via two
mechanisms: (1) the SB backfill itself has low hydraulic conductivity (typical design value is < 10
7 cm/s), and (2) the filter cake enhances the overall function of ihe wall as a barrier. Designers do
not normally count on the filter cake as a component of ihe harrier, it is viewed as a possible source
of added impermeanility that enhances the reliability of the wall.

       The compatibility of the backfill material with the ground water at a site should be assessed
prior to construction.  However. CQA personnel should be watchful  for ground water conditions
that may differ from those assumed in the compatibility testing projram.  CQA  personnel should
familiarize themselves with the compatibility testing program.  Sunstances that are  particularly
aggressive today backfills include-non-water-soluble organic chemicals, high and low pH liquids,
and highly saline water.  If there is any question  about ground water conditions  in relationship to
the conditions covered in the compatibility testing  program, the CQA  engineer and/or design
engineer should be consulted.

       Improper backfilling of slurry trench cutoff walls can produce defects (Fig. 7,14).  More
details are given by Evans < 1993). CQA  personnel should watch out for accumulation of sandy
materials during pauses in construction, e.g., during shutdowns or overnight: an airlift  can be used
to remove or re suspend the  sand, if necessary.
                                           247

-------
                   Soil-Bentonite
                   Backfill
            Colapse of Trench
            During Construction
                                                          Sand
                                                          Deposited
                                                          after Pause in
                                                          Backfilling
                 Inadequate Key
                 into Bedrock
                                                       Sand on Bottom
                                                       of Slurry Trench
    figure 7.14 - Examples of Problems Produced by Improper Backfilling of Slurry Trench.
       Some slurry trench cutoff walls fully encircle an area. As the slurry trench reaches the
point of initiation of the slurry trench cutoff wall, closure is accomplished by excavating into the
previously-backfilled wail.
                                                                  -»
       Hydraulic conductivity of SB backfill is normally measured by testing of small cylinders of
material formed from field samples.  Ideally, a sample of backfill material is scooped up from the
backfill, placed in a cylinder of a specified type, consolidated to a prescribed effective stress, and
permeated.  It is rare for borings to be drilled into the backfill to obtain samples for testing.

7.3,6 Cemcnt-gentonite (C&) Cutoff Walls

       A cement-bentonite (CB) cutoff wall is constructed with a  cement-bentonite-water mixture
that hardens and attains low  hydraulic conductivity. The slurry trench is excavated, and excavated
soils are hauled away. Then the trench is backfilled in one of two ways. In the usual method, the
slurry used  to maintain a stable trench during construction is CB rather than just bentonite-water.
                                          248

-------
and the slurry is left in place to harden. A much-less-common technique is to construct the slurry
trench with a bentonite-water slurry in discrete diaphragm cells (Fig. 7.15), and to displace the
bentonhe-water slurry with CB in each ceU.

      The CB mixture cures with time and hardens to the consistency of a medium to stiff clay
(CB backfill is not nearly as strong as structural concrete).  A typical CB slurry consists on a
weight basis  of  75 to  80% water, IS to 20% cement, 5% bcmonite, and a small amount of
viscosity reducing material.  Unfortunately, CB backfill is usually more permeable than SB
backfill. Hydraulic conductivity of CB backfill is often i.i the range of 10'6 to 10'5 cm/s, which is
about an order of magnitude or more greater than typical SB cutoff walls.
           (A) Excavate Panels
                                Excavated Panels
                                           Panel Being
                                             Excavated
           (B)
Excavate Between Panels
Excavation Between
Previously-Excavated
Panels

                         Figure 7.15 • Diaphragm-Wall Construction.


                                           249

-------
      The CB cutoff wall is constructed using procedures almost identical to those employed in
building structural diaphragm walls. In Europe, CB backfilled slurry trench cutoff walls are much
more common than in the U.S., at least partly because the diaphragm-wall construction capability
is more broadly available in Europe and because high-grade sodium bcntonite (which is critical for
soil-bentonite backfilled walls) is not readily available in Europe. In Europe, die CB often contains
other ingredients besides cement, bentonite, and water, e.g., slag and fly ash.
7.3.7 G
forane in Slurrv Trench Cutoff Walls
       Geomembranes may be used to form a vertical cutoff wall.  The geomembrane may be
installed in one of at least two ways:

       1.     The geomembrane may be inserted in a trench filled with CB slurry to provide a
             composite CB-geoniembrane barrier (Manassero and Pasqualini, 1992).  The
             geomembrane is typically mourned to a frame, and the frame is lowered into the
             slurry.  The base of the geomembrane  contains a weight such mat when the
             geomembrane is released from the frame, the frame can be removed without the
             geomembrane floating to the top. CQA personnel should be particularly watchful to
             ensure that the geomembrane is properly weighted and does not float out of
             position.  Interlocks between geomembrane panels (Fig. 7.6) provide a seal
             between panels.  The panels are typically relatively wide (of the order of 3 to 7  m)
             to minimize the number of interlocks and to speed installation. The width of a panel
             may be controlled by the width of excavated sections of CB-filled panels  (Fig.
             7.15).

       2.     The geomembrane may be driven directly into the CB backfill or inn the native
             ground. Panels of geomembrane with widths of the order of 0.5 to 1 m (18 to 36
             in.) are attached to a guide or insertion plate, which is driven or vibrated into the
             subsurface. If the panels are driven into a CB backfill material, the panels should
             be driven before the backfill sets up. Interlocks between geomembrane panels (Fig.
             7.6) provide a seal between panels. This methodology is essentially the same as
             that of a sheet pile wall.

       Although use of geotnembranes in slurry trench cutoff walls  is relatively new, the
technology is gaining popularity.  The promise of a practically impermeable vertical barrier, plus
excellent chemical resistance of HOPE geomembranes, are compelling advantages. Development
of more efficient construction procedures will make this type of cutoff wal! increasingly attractive.

7.3.8  Other Backfills

       Structural concrete could be used as a backfill, but if concrete is usedVthe material normally
contains bentonite and is termed plastic concrete (Evans, 1993). Plastic concrete is a mixture of
cement, bentonite, water, and aggregate.  Plastic concrete is different from structural concrete
because it contains bentonite and is different from SB backfill because plastic concrete contains
aggregate.   Other ingredients, e.g., fly ash, may be  incorporated into the plastic concrete.
Construction is typically  with the panel method (Fig. 7.15). Hydraulic conductivity of the backfill
can be < 1
-------
7.3.9    Cans

       A cutoff wall cap represents the final surface cap on top of the slurry trench cutoff wall.
The cap may be designed to minimize infiltration, withstand traffic loadings, or serve other
purposes.  CQA personnel should also inspect the cap as well as the wall itself to ensure that the
cap conforms with specification.

7.4    Other Types of Cutoff Walls

       Evans (1993) discusses other types of cutoff walls.  These include vibrating beam cutoff
walls, deep soil mixed walls, and other types of cutoff walls.  These are not discussed in detail
here because these types of walls have been used much less frequently than the other types.

7.5    Specific CQA Requirements

       No standard types of tests or frequencies of testing have evolved in the industry for
construction of vertical cutoff walls. Among the reasons for this is the fact that construction
materials and technology are continually improving.  Recommendations from this section were
taken largely from recommendations provided by Evans (personal communication).
       For slurry trench cutoff walls, the following comments are applicable. The raw bentonite
(or other clay) that is used to make the slurry may have specific requirements that must be met If
so, tests should be performed to verify those properties.  There are no standard tests or frequency
of tests for the bentonite. The reader may wish to consult Section 2.6.5 for a general discussion of
tests and testing frequencies for bentonite-soil liners.  For the slurry itself, common tests include
viscosity,  unit weight, and filtrate loss, and other tests often include pH and sand content The
properties of the slurry are normally measured on a  regular basis by the contractor's CQC
personnel; CQA personnel may perform occasional independent checks.

       The soil that is excavated from the trench should be continuously logged by CQA personnel
to verify that subsurface conditions are siro'lai to those that were anticipated. The CQA personnel
should look for evidence of instability in  the walls of the trench (e.g., sloughing at the surface next
to the trench or development of tension cracks). If the trench i* to extend into a particular stratum
(e.g., an aquitard), CQA personnel should verify that adequate penetration has occurred. The
recommended procedure is to measure the depth of the trench once the excavator has encountered
the aquitard and to measure the depth again,  after adequate penetration is thought to have been
made into  the aquitard.

       After the slurry has been prepared, and CQC tests indicate that the properties are adequate,
additional  samples are often taken of the slurry from the trench. The samples are often taken from
near the base of the trench using a special sampler that is capable of trapping slurry from the
bottom of  the trench. The unit weight is particularly important because sediment may collect near
the bottom of the trench.  For SB backfill, the slurry must not be heavier than the backfill. The
depth of the trench should also be confirmed by CQA personnel just prior to backfilling. Often,
sediments  can accumulate near the base of the trench - the best time to check for accumulation is
ju«t prior to backfilling. CQA personnel chould be particularly careful to check for sedimentation
after periods when the slurry has not been agitated, e.g., after an overnight work stoppage.

       Testing of SB backfill usually includes unit weight, slump, gradation, and  hydraulic
conductivity.  Bentonite content may also be measured, e.g., using the methylene blue test (Alther,
1983). Slump testing is u,c same as for concrete (ASTM C-143).  Hydraulic conductivity testing
is often performed using the API (1990) fixed-ring device for the filter press test. Occasional
                                          231

-------
comparative tests with ASTM D- 5084 should be conducted. There is no widely-applied frequency
of testing backfill materials.
7.6    Posj Construction Tests for Continuity
       At the present time, no testing procedures are available to determine the continuity of a
completed vertical cutoff wall.
7.7    References
Alther, G.  R. (1983), "The Methylene  Blue Test for Bentonite Liner Quality Control,**
   Geotechnical Testing Journal, Vol. 6, No. 3. pp. 133-143.
American Petroleum Institute (1990), Recommended Practice f^r Standard Procedure for Field
   Testing Drilling Fluids, API Recommended Practice 13-B-l, Dallas, Texas.
ASTM C-143, "Slump of Hydraulic Cement Concrete."
ASTM D-2487, "Classification of Soils for Engineering Purposes (Unified Soil Classification
   System)."
ASTM D-4380, "Density of Bentonitic Slurries."
ASTM D-4381, "fand Content by Volume of Bentonite Slurries.''
ASTM D-5084, "Measurement of Hydraulic Conductivity of Saturated Porous Materials Using a
   Flexible Wall Permeameter."
D'Appolonia, D. J. (1980), "Soil-Bentonite Slurry Trench Cutoffs," Journal of Geotechnical
   Engineering, Vol. 106, No. 4, pp. 399-417.
Evans, J.  C (1993), "Vertical Cutoff Walls," in Geotechnical Practice for Waste Disposal, D. E.
   Daniel (Ed.), Chapman and Hall, London, pp. 430-454.
Manassero,  M., and E.  Pasqualini (1992), "Ground Pollutant Containment Barriers," in
   Environmental  Geotechnology, M.A. Usmen and Y.B. Acar (Eds.),  A.A. Balkema,
   Rotterdam, pp. 195-204.
Ryan, C.  R. (1987), "Soil-Bentonite Cutoff Walls," in Geotechnical Practice for Waste Disposal
   '87, R. D. Woods (Ed.), American Society of Civil Engineers, New York, pp. 182-204.
Sherard,  J.  L., Woodward, R. J., Gizienski, S. E, and W. A, Clevenger (1963), Earth  and
   Earth-Rock Dams, John Wiley and Sons, Inc., New York, 725 p.
U.S. Environmental Protection  Agency (1984),  "Slurry Trench Construction for Pollution
   Migration Control," Office of Emergency and  Remedial  Response, Washington, DC, EPA-
   540/2-84-001.
Xanthakos, P. P. (1979), Slurry Walls, McGraw-Hill Book Company, New York, 822 p.
                                        252

-------
                                      Chapter 8

                Ancillary Materials, Appurtenances  and Other Details
      This chapter is devoted toward ancillary materials used within a waste containment facility,;
various appurtenances which  are necessary for proper functioning of the system and other
important details. Ancillary materials such as plastic pipe for leach ate transmission, sumps for
collection of leachate, manholes and pipe risers for removal of leachate will be covered in this
chapter.  Appurtenances, such as penetrations made through various barrier materials, will be
covered.  Lastly, other important details requiring careful inspection, such as anchor trenches,
internal dikes and berms, and access ramps, will also be addressed.

8.1 Plastic Pipe (aka "Geopipe^

      Whenever the primary or secondary leachate collection system at the bottom of a waste
containment facility is a natural soil material, such as sand or gravel, a perforated piping system
should be located within it to rapidly transmit the leachate to a sump and removal system. Figure
8.1 illustrates the cross section of such a pipe system which is generally located directly on top of
the geomembrane or geotextile to 225 mm (9.0 in.) above the primary liner material This is a
design issue and the plans and specifications must be clear and detailed regarding these
dimensions.
                                                                          Geotextile
                                                                          Filters
                                                                          Drainage
                                                                          Stone

                                                                          GeotextiJe
                                                                          Protect*?
                                                                          Layer
                                                                          Geomembrane
Figure 8.1 - Cross Section of a Possible Removal Pipe Scheme in a Primary Leachate Collection
            and Removal System (for illustration purposes only).
      The pipes are sometimes placed in a manifold configuration with feeder lines framing into a
larger main trunk line thus covering the entire footprint of the landfill unit or cell, see Fig. 8.2.
The entire pipe network flows gravitationally to a low point where the sump and removal system
                                          253

-------

                                                                          Vertical
                                                                          Removal
                                                                          Sump and
                                                                          Manhoteor
                                                                          SldMlop*
                                                                          RfMT
Figure 8.2 -  Plan View of a Possible Removal Pipe Scheme in a Primary Leachate Collection and
            Removal System (for illustration purposes only).
consisting of either a manhole or pipe riser is located.  The diag?r.
-------
Figure 8.3 - Photograph of PVC Pipe to be Used in a Landfill Leachate Collection System.
   1.  The basic resin should be made from PVC as defined in ASTM D-1755. Details are
      contained therein.

   2.  Other materials in the formulation, such as fillers, carbon black/pigment and additives
      should be stipulated and certified as to the extent of their prior use in plastic pipe.

   3.  dean rework material, generated from the manufacturer's own pipe or fitting production
      may be used by the same manufacturer providing that the rework material meets the
      above requirements.  See section  3.2.2 for a description of possible use of reworked
      and/or recycled material.

   4.  Pipe tolerances and properties must meet the applicable standards for the particular grade
      required by the plans and specifications.  For PVC pipe specified as Schedule 40, 80
      and 120, the appropriate specification is ASTM D-1785.  For PVC pipe in the standard
      dimension ratio (SDR) series, the applicable specification is ASTM D-224],
                                      255

-------
      5. Both of the above referenced ASTM Standards have sections on product marking and
         identification which should be followed as well as requiring the manufacturer to provide
         a certification statement stating that the applicable standard has been followed.

      6. PVC pipe fitu.igs should be in accordance with ASTM D-3034.  This standard includes
         comments on solvent cement  and elastomeric gasket joints as well as a section on
         product marking and certification.
8.1-2 Hif h Densify Polyethylene
                                     Smooth Wall Pie
      High density polyethylene (HOPE) smooth wall pipe has been used in waste containment
systems for leachate collection and removal in a number ot different locations and configurations,
The pipe can be perforated or not depending on the site specific design.  The pipes are often
supplied in 6. 1 m (20 ft) lengths which are generally joined together using butt-end fusion using a
hot plate as per the gas pipe construction industry. Other joining variations such as bell and spigot,
male-to-female and threading are also available.  The HDPE material itself consists of 97-98%
resin, approximately 2% carbon black and  up to 1% additives. Figure 8.4 illustrates the use of
HDPE smooth pipe.
Figure 8.4 - Photograph of HDPE Smooth Wall Pipe Risers Used as Primary and Secondary
            Removal Systems from Sump Area to Pump and Monitoring Station.
                                         256

-------
      The following items should be considered regarding the contract specification or MQA
document on HOPE solid wall pipe and fittings:

      1.  The basic material should be  made of HDPE resin and should conform to the
         requirements of ASTM D-1248. Details are contained therein.

      2. Quality control tests on the resin are typically density and melt flow index.  The
        appropriate designations are ASTM D-1505 or D-792 and D-1238, respectively. Otlier
        in-house quality control tests should be encouraged and followed by the manufacturer.

      3. Typical densities for HDPE pipe resins are 0.950 to 0.960 g/cc. This is a Type III
        HDPE resin according to ASTM D-1248 and is higher than the density of the resin used
        in HDPE geomembranes and geonets.

      4. Carbon black can be added as a concentrate, as it customarily is, or as a powder. The
        type and amount of carbon black, as  well as the type of carrier resin if concentrated
        pellets are used, should be stated and certified by the manufacturer.

      5. The amount of additives used should be stated by the manufacturer.  If certification is
        required it would typically not state the type of additive, since they  are  usually
        proprietary, but should state that the additive package has successfully been used in the
        past and to what extent

8.1.3  High Density Polyethylene (HDPEI Corrugated Pipe

      Corrugated high density polyethylene (HDPE), also called "profiled" pipe, has been used in
waste containment systems for leachate collection and removal in a number of different locations
and configurations.  The pipe can be perforated or slotted depending on the site specific design.
The inside can be smooth lined or not depending on the site specific design. The pipes are often
supplied in 6.1 m (20 ft) lengths which are joined together by couplings made by the sam?
manufacturer as the pipe  itself.  Tnis is important since the couplings  are generally not
interchangeable among different pipe manufacturer's products. The HDPE material itself consists
of 97-98% resin, approximately 2% carbon black and up to 1% additives.  Figure 8.5 illustrates
HDPE corrugated pipe.

      Regarding the contract specification or MQA document on HDPE corrugated pipe and
fittings, the following items should be considered:

      1. The  basic material should be  made of HDPE resin and should conform to the
        requirements of ASTM D-1248. Details are contained therein.

      2. Quality control tests are typically density and melt flow index. Their designations are
        ASTM D-1505 or D-792 and D-1238, respectively. Other in-house quality  control tests
        are to be encouraged and followed by the manufacturer.

      3. Typical densities  for HDPE pipe resins are 0.950 to 0.960 g/cc. This is a Type III
        HDPE resin according to ASTM D-I248 and is higher than the resin density used  in
         HDPE geomembranes.

      4.  Carbon black can be added as a concentrate as it customarily is, or as a powder.  The
         type and amount of carbon black, as  well as the type of carrier resin if concentrated
         pellets are used, should be stated and certified by the manufacturer.
                                          257

-------
     5. The amount of additives used should be stated by the manufacturer. If certification is
        required it would typically  not state the type  of additive, since they are usually
        proprietary, Hut should state that the additive package has successfully been used in the
        past

     6. The lack of ASTM documents for HOPE corrugated pipe should be noted There is an
        AASHTO Specification available for corrupted polyethylene pipe In the 300 to 900 nun
        (12 to 36 in.) diameter range under the designation M294-90 and another for 75 to 250
        mm (? to 10 in.) diameter pipe under the designation of M252-90.
                                                          *•-"•
     Figure 8.3 - Photograph of HOPE Corrugated Pipe Being Coupled and After Installed.
g.1.4 Handling of Plastic Pipe

      As with all other geosynthetic materials a number of activities occur between the
manufacturing of we pipe and its final positioning in the waste facility.  These activities include
packaging, storage at the manufacturers facility, shipment, storage at the field site, confomtance
testing and the actual placement
                                          25g

-------
8.1.4.1 Packaging

      Both PVC pipe and HDPE pipe are manufactured in long lengths of approximately 6.1 m
(20 ft) with varying wall thicknesses and configurations. They are placed on wooden pallets and
bundled together with plastic straps for bulk handling and shipment. The packaging is such that
either fork lifts or cranes using slings can be used for handling and movement. As the diameter
and wail tlt;"kness increases, however, this may not be the case and above 610 mm  (24 in.)
diameter the pipes are generally handled individually.

8.1.4.2 Storage at Manufacturing Facility

      Bundles of plastic pipe can be stored at the manufacturing facility for relatively long periods
of time with respect to other geosynthetics.  However, if stored outdoors for over 12 months
duration, a temporary enclosure should be used to cover the pipe from ultraviolet exposure and
high temperatures.  Indoors, there is no defined storage time limitation. Pipe fittings are usually
stored in a container or plastic net

8.1.4.3  Shipment

      Bundled pallets of plastic pipe are shipped from the manufacturer's or their representative's
storage facility to the job site via common carrier. Ships, railroads and trucks have all been used
depending upon the locations of the origin and final destination.  The usual carrier from vithin the
USA, is truck. When using flatbed trucks, tht pallated pipe is usually loaded by means cf a fork
lift or a crane with slings wrapped around the entire unit. When the truck  bed is  closed, i.e., an
enclosed trailer, the units are usually loaded by fork lift. Large size pipes above 610 mm (24 in.)
in diameter are handled individually.

8. 1 .4.4 Storage at Field Site

      Offloading of palleted plastic pipe at the site and temporary storage is a necessary follow-up
task which must be done in an acceptable manner.

      Items to be considered for the contract specification or CQA document are the following:

      1 .   Handling of pallets of plastic pipe should be done in a competent manner such that
           damage does not occur to the pipe.

      2.   The location of field storage should not be in aretj where water can accumulate. The
           pallets should be  on  level ground and oriented so as not to form a dam creating the
           ponding of water.                               %.

      3.   The pallets should not be stacked more than three high. Furthermore,  they should be
           stacked in such a way that access for conformance testing is possible.

      4.   Outdoor storage of plastic pipe should not be  longer than 12  months.  For storage
           periods longer than 12 months a temporary covering should be placed  over the pipes,
           or they should be moved to within an enclosed facility.

8.1.5 Conformance Testing and Acceptance

      Upon delivery of the plastic pipe to the project site, and temporary storage (hereof, the CQA
engineer should see that conformance test samples are obtained. These samples are  then sent to the
                                          259

-------
CQA laboratory for testing to ensure that the pipe supplied conforms to the project plans and
specifications.

     Items to consider for the contract specification or CQA document in this icgard are the
following:

     1.  The pipe should be identified according to its proper ASTM standard:

         (a) for PVC Schedule 40,80 and 120: see ASTM D-1785

         
-------
8,1.6  Placement

      Plastic pipe is usually placed in a prepared trench or within other prepared subgrade
materials. If the pipe is to be placed on or near to a geomembrane, as in the leachate collection
system shown in Fig. S.I, the drainage sand or stone should be placed first There may be a
requirement to lightly compact sand to 90% relative density according to ASTM p-4254. Small
excavations of slightly greater than the diameter of the pipe are then made, and the pipe is placed in
these shallow excavation^. Thus a trench, albeit a shallow one, is constructed in all cases of pipe
placetrcnt in leachate collection sand or stone.

      Where plastic pipe is placed at other locations adjacent to the containment facility and the soil
is cohesive, compaction is critical if high stresses are to be encountered.  Compaction control is
necessary, e.g., 95% of standard Proctor compaction ASTM D-698 is recommended so as to
prevent subsidence of the pipe while in service.

      The importance of the density of the material beneath, adjacent and immediately above a
plastic pipe insofar as its load-carrying capability is concerned cannot be overstated. Figure 8.6
shows the usual configuration and soil backfill terminology related to the various materials and
their locations.

      Regarding a specification or CQA document for plastic pipe placement, ASTM D-2321
should be referenced. For waste containment facilities die following should be considered:

      1 . The soil beneath, around and above the pipe shall be Class LA, IB or II according to
         ASTM D-2321.

      2. The backfill soil should extend a minimum of one pipe diameter above the pine, or 300
         mm (12 in.) which ever is
      3. Other conditions should be taken directly according to ASTM D-2321.
      4. Pipe  finings  should
         recommendations.

8.2 S^ms. Manholes and Risers
                                  in accordance with the  specific  pipe  manufacturer's
      Leachate which migrates along the bottom of landfills and waste piles flows gravitationally
to a low point in the facility or cell wi.ere it is collected in a sump. Two general variations exist;
one is a  prefabricated sump, made either in-situ or off-site, with a manhole extension rising
vertically through the wane and final covt . the other is a tow area formed in the liner itself with a
solid wall pipe riser coming up the side slope where it eventually" penetrates the final cover.
Both variations are shown schematically in the sketches of Fig. 8.7.  In addition, the sump and
sidewall rise, of a secondary leachate collection system typically used in double lined facilities is
shown in the right sketch of Fig. 8.7(b). i.e., a leak detection system.  Each type of system will be
briefly described.

      Many existing landfills have bu»> constructed with primary leachate collection and removal
sumps and manholes constructed to ttu. tie specific plans and specifications as shown in the left
hand sketch of Fig. 8.7(a).  The vertical riser is either a concrete or plastic standpipe placed in 3 m
(10 ft) sections. It is extended as the waste is placed in the facility and eventually it must penetrate
the final cover.  Leachate is removed from this  manhole, on  an as demanded basis,  by a
submersible pump which is permanently located in the sump.
                                          261

-------
                                                                Secondary Backfill
                                                               Primary Backfill

                                                               Pipe Haunch Area
                                                                Bedding Soil
Figure 8.6 - A Possible Buried Pipe Trench Cross Section Scheme Showing Soil Backfill
            Terminology and Approximate Dimensions (for illustration purposes only).
      A more recent variation of the above removal system is an off-site factory fabricated sump
and manhole system wherein the leachate collection pipe network frames directly into the sump,
see the right hand sketch of Fig. 8.7(a). Various standardized sump capacities are available. This
type of system requires the least amount of field fabrication. The riser is extended in sections as
the waste is placed in the facility and eventually it must penetrate the ftoal cover.  Leachate is
removed from the manhole by a submersible pump which is permanently located in the sump.

      Quite a different variation for primary leachate removal is a well defined low area in the
primary geomembrane into which the leachate collection pipe network flows  This low area creates
a sump which is then filled with crushed stone and from which a pipe riser extends up the  side
slope.  The pipe riser is usually a  solid wall pipe with no perforations.  When the facility is
eventually filled with solid waste, the riser must penetrate the coyer as shown in the left hand
sketch of Fig. 8.7(b).  The leachate is withdrawn using a submersible pump which is  lowered
down the pipe riser on a sled and left in place except for maintenance and/or replacement, recall
Fig. 8.4.
                                          262

-------
               LMChate
               Removal
                                            Cover
                                          Solid
                                          wastt
                                          fff#f*r*
                                         X'X'X'X'S f
                 1
Ltachate
Removal
                                         iS
     EZS33s§&,
 ^liligBUfikHNE
                   Geomembrane
FOOtiOQ
      kvSHu Febrtcatton
Header Pipe               Geomembrana

          Factory Fabrication
      (a) Types of Primary Leachate Collection Sumps and Manholes with
            Vertical Standpipe Going through the Waste and Cover
 Cover
                       (.••chate
                       Removal
        Cover
                           Removal
              Stone
                                                      ^* Stone
       (b) Types of Primary (Left) and Secondary (Right) Leachate Collection Sumps
                       and Pipe Risers Going Up the Side Slopes
           Figure 8.7 - Various Possible Schemes for Leachate Removal
                                   263

-------
      In • similar manner as above, but now for secondary leachate removal, a sump can be
formed in die secondary liner system which is filled with gravel as shown in the  'ght hand sketch
of Fig. 8.7(b).  A solid wall pipe riser, perforated in its lower section, extend* up the sidewall
between the primary and secondary liner where it must penetrate both the primary liner, and
eventually the coyer system liner, see the right hand sketch of Fig. 8.7(b). This pipe riser is often
a solid wall pipe in the 100-200 (4 to 8 in.) diameter range with no perfotations. The leachate is
withdrawn and/or monitored using a small diameter sampling pump which is lowered down the
riser and left in place except for maintenance and/or replacement, recall Fig. 8.4.

      Some specification and CQA document considerations for the various sump, manhole and
riser schemes just described are as follows. Note, however, that there are other possible design
schemes that are available in addition to those mentioned above.

      1 .  In-situ  fabrication of sumps requires a considerable amount of hand labor in the field.
         Seams  for HDPE and VLDPE geomembranes are extrusion fillet welded, while PVC
         and CSPE-R geomembranes are usually bodied chemical seams (EPA, 1991). Careful
         visual inspection is necessary.

      2.  The soil support beneath the sumps and around the manhole risers of plastic pipes is
         critically important.  The specification should reference ASTM D-2321 with  only
         backfill types IA, IB and II being considered.

      3.  Riser pipes for primary and secondary leachate ismoval are generally not perforated,
         except for the lowest section of pipe which accepts the leachate.

      4.  Riser pipe joints for primary and secondary leachate removal require special visual
         attention since neither destructive nor nondestructive tests can usually be accommodated.

      * .  The sump, manholes and risers must be documented by the CQA  engineer before
         acceptance and placement of solid waste.

8.3
      Altnough the intention of most designers of waste containment facilities is to avoid liner
penetrations, leachate removal is inevitably required at some locarion(s) of the barrier system.
Recall Fig. 8.7 where the cover is necessarily penetrated for primary leachate removal. For leak
detection both the primary  liner and the cover liner must  be penetrated.  It should also be
recognized  that the  penetrations will include geomembranes, compacted clay liners and/or
geosynthetic clay liners. Figure 8.8 illustrates some details of pipe penetrations through all three
types of barrier materials.                                         x

      The following recommendations are made for a specification or CQA document:

      1 . Geomembrane pipe boots are usually factory fabricated to a size which tightly fits the
        outside diameter of the penetrating pipe. Unique situations, however, will require field
        fabrication, e.g., when pipe penetration angles are unknown until final installation.

      2. The skin of the pipe boot which flares away from the pipe penetration should  have at
        least 300 mm ( 1 2 in.) of gcomembrane on all sides of the pipe.

      3 . The skin of the pipe boat should be seamed to the base geomembrane by extrusion fillet
        or bo.,i £ chemical  seaming depending on the type of geomembrane (EPA, 1991 ).
                                         264

-------
                                           Cushioning Lay*r




                                           Stain!*** StMl Clamp
                                            Qtomwiifann*
                          Ptp.
                 (a) Geomembrane Penetration
                                                        FtoMSaam
          (b) Compacted Oty Liner (CCL) Penetration
   OCU
Dry Bwitonlt*
                                          Dry Bantontt*
          (c) GeosynUxMic Qay Liner (GCL) Penetration
Figure 8.8 - Pipe Penetrations through Various Types of Barrier Materials
                               26S

-------
      4. The nondestructive testing of the skirt of die pipe boot should be by vacuum box or air
        lance depending on the type of geomembrane. Refer to Section 3.6.2.

      5 . The pipe boot should be of the same type of geomembrane as that of the liner through
        which the penetration is being made.

      6. Pipe penetrations should be positioned with sufficient clearance to allww for proper
        welding ana inspection.

      ?. Stainless steel pipe clamps used to attach pipe t ooij to the penetrating pipes should be
        of an adequate size to allow for a cushion of compressible material to be placed between
        the inside surface of the clamp and that of t> j geomembrane portion of flic pipe boot.

      S. Location of pipe clamps should be as directed on the plans and specifications.

      9. Pipe penetrations through compacted day liners and geosynthetic clay liners should use
        an excess of hand placed dry bentonite clay as directed in the plans and specifications.
8.4
      Generally, the ; wsynthetics used to line or cover a waste facility end in an anchor trench
around die individual cell or around the entire si*.

8.4.1  GeomembranM

      The termination of a geomembrene at the perimeter of landfill cells or at the perimeter of the
entire facility generally ends in an anchor trench.  As shown in Fig. 8.9, the variations are
numerous.   Such details should be specifically  addressed  in the  construction plans and
specifications.

      Some general items that should be addressed in the specification or CQA documents
regarding geomembrane termination in anchor trenches ETC as follows:

      1 .  The seams of adjacent sheets of geomcmbranes should be continuous into the anchor
         trench to the full extent indicated in the plans and specifications.

      2.  Seaming of geomembrancs within the anchor trench can be accomplished by temporarily
         supporting the adjacent sheets to be seamed on a wooden support platform in order that
         horizontal seaming can be accomplished continuously to the end of the geomembrane
         sheets.  The temporary support  is removed after  the seam is  complete  and the
         geomembrane is then allowed to drop into the anchor trench.

      3.  Destructive seam samples can be taken while the seamed geomembrane is temporarily
         supported in the horizontal position,

      4.  Nondestructive tests can also be performed while the  seamed geomembrane  is
         temporarily supported in the horizontal position.

      S .  The anchor trench is generally backfilled after the geomembrane has been documented
         by the CQA engineer, but may  be at a later date depending upon the site specific plans
         and specifications.
                                         266

-------
                                                      mm

                                      Typical Anchor Trench
                                                                  800mm
                                     Horizontal Runout Anchor
                                                            300-400 mm
                                Shallow "V" Anchor Trench
                 Top of Stop*     p»|
                                       Sotted Aneher System
                                                  Polymer Bitten Strip
                              Concrete Anchor Bloc!'
Figure 8.9 - Various Types of Geomembrane Anchors Trenches (Dimensions are Typical and for
           Example Only).
                                       267

-------
      6,  The anchor trench itself should be made with slightly rounded comers so as to avoid
         sharp bends in the geomembrane. Loose soil should not be allowed to underlie the
         geomembrane in the anchor trench.

      7.  The anchor trench should be adequately drained to prevent ponding of water or softening
         of the adjacent soils while the trench is open.

      8.  Backfilling in the anchor trench should be accomplished with approved backfill soils
         placed at their required moisture content and compacted to the required density.

      9.  The plans and specifications should provide detailed construction requirements for
         anchor trenches regardless if soils or other backfill materials are used.

8.4.2  Ofter peosynihetics

      Since all geosynthetics, not only gepmembrancs, need adequate  termination, some
additional comments are offered for plans, specifications or CQA documents.

      1.  Gcotextiles, either beneath or above geomernbranes, usually follow their associated
         geomembrane into the same type of anchor trenches as shown in Fig. S.9.

      2.  Geonets may or may not terminate in the anchor trench.  Water transmission from
         beyond the waste containment may be a concern when requiring termination of the
         geonet within the geomembrane's anchor trench or in a separate trench by itself. Thus
         termination of a geonet may be short of the  associated geomembrane's anchor trench.
         Hits is obviously a design issue and must be clearly detailed in the contract plans and
         specifications.

      3.  When used by themselves, geosynthetic clay  liners (GCLs) will generally terminate in a
         anchor trench in soil of the type shown in Fig. 8.9. When GCLs are with an associated
         geomembrane, as in a composite liner, each component will sometimes end in a separate
         anchor trench. These are design decisions.

      4.  Double liner systems will generally have separate anchor trenches for primary and
         secondary liner systems. This is a design decision.

      S.  In all of the above cases, the plans and specifications should provide detailed dimensions
         and construction requirements for anchor trenches of all geosynthetic components.

      6.  The plans and specifications should also show details of how nafitral soil components,
         e.g., compacted clay liners and sand or gravel drainage layers, terminate with respect to
         one another and with respect to the geosynthetic components.

8.5 Aceesi Ramps

      Heavily loaded vehicles must enter the landfill  facility during construction activities and
during placement of the solid waste. Typical access ramps will be up to  5.5 m (18 ft) in width
and have grades up to 12%. The general geometry of an access ramp is shown in Fig. 8.10(a).
                                         268

-------
           (a) Geometry of a Typical Ramp
          Roadway
                                                         Leachate
                                                         Collection
                                                       Gcomcmbrane
                                                         Leak
                                                       Detection
                                                        Gcomcmbrane
      (b) Cross Section of Ramp Roadway
Figure 8.10 - Typical Access Ramp Geometry and Cross Section
                           269

-------
      The traffic loads on such a ramp can be extremely large and generally involve some degree
of dynamic force due to the constant breaking actio.i which drivers use when descending the steep
grades.  Note that the entire liner cross section must extend uninterrupted from the upper slope to
the lower slope and in doing so must necessarily pass beneath the roadway base course.  When
working with a double lined facility this can involve numerous geosynthetic and natural soil layers.
Further complicating the design issues is that drainage  from  the upper  side slopes  must
communicate beneath the roadway base course layer or travel parallel to  it and be contained
accordingly. A reinforcing dement (geotextile or geogrid) can be incorporated in the roadway base
course material. This can serve several purpose*; Le., to protect long-term integrity of underlying
systems, to minimize potential sliding failures, and to minimize potential  rutting and bearing
capacity failures.  These are critical design issues and must be well defined in the plans and
specifications.

      Regarding recommendations for the contract specifications or CQA document, the following
items apply:

      1. Many facilities will limit the number of vehicles on the access ramp at a given time.
         Such stipulations should be strictly enforced.

      2. Vehicle speeds on access ramps should be strictly enforced.

      3. Regular inspection should be required to observe if tension cracks open in the roadway
         base coarse soils. This may indicate some degree of slippage  of the soil and possible
         damage to the liner system.

      4. Ponding of upper slope runoff water against the roadway profile should be observed for
         possible erosion effects and loss of base course material.  If a drainage ditch or pipe
         system is indicated on the plans, it should be constructed as soon as possible after
         completion of the roadway subbase soils.

      5. The roadway base course profile should be fully maintained for the active lifetime of the
         facility.

8.6 Qeosynthen'c Reinforcement Materials

       For landfill  and waste pile covers with slopes greater than 3 horizontal to 1 vertical
(3H:1V), stability issues regarding downgradient sliding begin to be important Additionally, the
stability of primary leachate collection systems for landfill and waste pile liners with slopes greater
than 3H: IV is suspect at least until the solid waste material within the unit raises to a stabilizing
level. Such issues, of course, must be considered during the design phase and the contract plans
and specifications must be very clear on the  method of reinforcement, if any?" If reinforcement is
necessary it can be accomplished by using geotextiles or geogrids within the layer contributing .to
the instability to offset some, or even all, of the gravitational stresses. Refer to Fig. 8.11 (a) and
(b) for  the general orientation of such  reinforcement, which is sometimes called "veneer
reinforcement'.

       The concept of using geogrid or geotextile reinforcement to support a liner or liner system
when a new landfill is built above,  or adjacent to, an existing landfill has recently been developed.
The technique has been referrrd to as "piggybacking" when vertical expansions are involved, see
Fig. 8.11(c).  The main focus of the reinforcement is »r» provide stability against differential
settlement which can occur in the existing landfill
                                          270

-------
                             Reinforcement
                             (Geogridor
                             Geotextile)
                  Cover Soil
                                                                            Geomembrme
                                 (a) Cover Soil Veneer Stability
              Anchor
              Trenches
                                                                    Reinforcement
                                                                    (Geogridor
                          (b) Leachate Collection Soil Veneer Stability
                                                                       Reinforcement
                                                                       (Geogridor
                                                                       Geotextile)
                     (c) Liner System Reinforcement for "Piggy backing"
Figure 8.11 • Geogrid or Geotextile Reinforcement of (a) Cover Soil above Waste, (b) Leachate
              Collection Layer beneath Waste, and (c) Liner System Placed above Existing Waste
              TPiggybacking")
                                            271

-------
       Since geotextiles were described previously from a manufacturing standpoint and for
separation and filtration applications, they will be discussed here only from their reinforcement
perspective.  Geogrids will be described from both their manufacturing and reinforcement
perspectives.

8.6.1 Geotextiles for Reinforcement

       The manufacturing of geotextiles was described in section &2 along with recommendations
for MQC and MQA documents. Regarding CQC and CQA, the focus was on separation and
filtration applications. Some specific recommendations regarding reinforcement geotextiles for a
specification or CQA document are as follows:

       1.  A manufacturer's certification should be provided that the geotextile meets the property
          criteria specified for the  geotextile that was approved for use on the project via the
          plans and specifications.

       2.  CQA personnel should check that the geotextile delivered to the job site is the proper
          and intended material. This is done by verifying the identification label and its coding
          and by visual identification of the product, its construction and other visual details.

       3.  Conformance samples of the geotextile supplied to the job site should be obtained as
          per ASTM D-4759. Typically, the outer wrap of the rolls are used for such sampling.

       4.  Conformance tests should be the following.  Wide width tensile strength per ASTM D-
          4595, trapezoidal tear strength per ASTM D-4533 and puncture strength per ASTM D-
          4833.  Additional Conformance  tests which may  be  considered  are  polymer
          identification via thermogravimetric analysis (TGA) and grab tensile strength, via
          ASTMD-4632.

       5.  Field placement of geotextiles should be at the locations indicated on the contract plans
          and in the specifications.  Details of overlapping or seaming should be included.

       6.  Geotextile deployment is usually from the top of slope downward, so that the
          geotextile is taut before soil backfilling proceeds.

       7.  If the upper end of the geotextile should be anchored in an anchor trench, the details
          shown in the contract plans should be fulfilled.

       8.  Soil backfilling should proceed from the bottom of the slope upward, with a minimum
          backfill thickness of 220  mm (9 in.) of cover using light ground contact construction
          equipment of 40 kPa (6 lb/in2) contact pressure or less.

       9.  Seams in geotextiles on  side slopes are generally not allowed. If permitted, they
          should be located as close to the bottom of the slope as possible.  Seams should be as
          approved by the CQA  engineer.  Test strips of seams should be requested for
          Conformance tests in the CQA laboratory following ASTM D-4884.
                                         272

-------
                                                                                               II
8.6,2 Gfopids

      Geogrids are reinforcement geosynthetics formed by intersecting and joining sets of
longitudinal and transverse ribs with resulting open spaces called "apertures". Two different
classes of geogrids are currently available, see Fig. 8.12(a).  They are the following:  (a) stiff,
unitized, geogrids made from polyethylene or polypropylene sheet material which is cold worked
into a post-yield state, and (b) flexible, textile-like geogrids made from high tenacity polyester
yams which are joined at their intersections and coated with a polymer or bitumen. Figure 8.12 (b)
shows geogrids being used as veneer reinforcement

      Some  recommended contract specification or CQA document items that should be
addressed when using geogrids as reinforcement materials are as follows:

       1.  A manufacturer's certification should be provided that the geogrid meets the property
          criteria specified for the geogrid that was approved for use on the project per the plans
          and specifications.

      2.  CQA personnel should check that the geogrid delivered to the job site is the proper and
          intended material. This is done by verifying the identification label and its coding and
          by visual identification of the product, its rib joining, thickness and aperture size.  If
          the geogrid has a primary strength direction it must be so indicated.

      3.  Conformance samples of the geogrid supplied to the job site should be obtained as per
          ASTM  D-4759.  Typically, the outer wrap of the rolls are used for such sampling.

      4.  Conformance tests  should be the following.  Aperture size by micrometer or caiiper
          measurement, rib thickness and junction thickness by ASTM EM 777, and wide width
          tensile  strength  by ASTM  D-4595 suitably modified for geogrids.  Additional
          conformance tests which may be considered are polymer identification via thermal
          analysis methods and single rib tensile strength, vkGRl GG1.

      5.  Field placement of geogrids should be at the locations indicated on the contract plans
          and in the specifications. Details of overlapping or seaming should be included.

      6.  Geogrid deployment is usually from the top of slope downward, so that the geogrid is
          taut before soil backfilling proceeds.

      7.  If the upper end  of die geogrids are to be anchored in an anchor trench, the details
          shown in die contract plans should be fulfilled.

      8.  Soil backfilling should proceed from the bottom of the slope upward, with a minimum
          backfill thickness of 22 cm (9.0 in.) of cover using light ground contact construction
          equipment of 40 kpa (6 lb/in*) contact pressure or less.

      9.  Connections of geogrid rolls on side slopes should generally be avoided. If permitted,
          they should He located as close to the bottom of the slope as possible.  Connections
          should be as approved by the CQA engineer. Test strips of connections should be
          requested for conformance tests in the CQA laboratory following ASTM D-4884
          (mod.) test method.
                                         273

-------
                       (a) Various Types of Geogrids
                  (b) Geogrids Used as Veneer Reinforcement
Figure 8.12 - Photographs of Geogrids Used as Soil (or Waste) Reinforcement Materials
                                  274

-------
8.7 Qeosynjhetic Erosion Control Materials
       Often on sloping solid waste landfill covers - J loss in the form of rill, gully or sheet
erosion occurs in the topsoil and sometimes exte  * down into the cover soil.  This requires
continuous maintenance until the phenomenon is halted and the long-term vegetative growth is
established. Alternatively, the design may call for a temporary, or permanent, erosion control
system to be deployed within or on top of the topsoil layer. Additional concerns regarding erosion
control  are on perimeter trenches, drainage ditches, and other surface water control structures
associated with waste containment facilities. Listed below are a number of alternative erosion
control  systems ranging from the traditional hand distributed mulching to fully paved cover
systems. They fall into two major groups; temporary de^adable and permanent nondegradabie.
Temporary Sosion Control and Revegetation \fots fTERi»||)
       * Mulches (hand or machine applied straw or hay)
       • Mulches (hydraul'cally applied wood fibers or recycled paper)
       • Jute Meshes
       • Fiber Filled Containment Meshes
       • Woven Gcotexale Erosion Control Meshes
       * Fiber Roving systems (continuous fiber systems)
          Erosion Contro  arid Reveetato  tats fPERMs
       * Geosynthen'c Systems
              *  njrf ranfonxinent and rtvegetation mats (TRMs)
              *  erosion control and revegetation mats (ECRMs)
              *  geomatting systems
              *  geocelluiar containment systems
       • Hard Armor Systems
              *  cobbles, with or without geotextiles
              *  rip-rap, with or without geotcxtiles                   x
              •  articulated concrete blocks, with or without geotextiles
              *  grout injected between geotextiles
              *  partially or fully paved systems
       Temporary degrsdable systems are used to enhance the establishment of vegetation and
 then degrade leaving the vegetation to provide the erosion protection required. Challenging sites
                                          275

-------
that require protection above and beyond what vegetation can provide need to use a permanent
nondegradaupn system, i.e., high flow channels, over steepened slopes etc.  Of these various
alternatives, jute meshes, containment meshes and geosynthetic systems are used regularly on
landfill and waste pile cover systems, see Fig. 8.13.

       Some items which are recommended for contract specifications or CQA document for these
particular systems are as follows:

       1.  The CQA personnel should check the erosion control material upon delivery to see
           that the proper materials have been received.

       2.  Water and ultraviolet  sensitive materials should be stored in dry conditions and
           protected from sunlight

       3.  If the erosion control material has defects, tears, punctures, flaws, deterioration or
           damage incurred during manufacture, transportation or storage it should be rejected of
           suitably repaired to the satisfaction of the CQA personnel.

       4.  If the material is to be repaired, torn or punctured sections should be removed by
           cutting a cross section of the material out and replacing it with a section of ur... •imaged
           material. The ends of the new section should overlap the damaged section by 30 cm
           (12 in.) and should be secured with t,,x»und anchors.

       5.  All ground surfaces should be prepared so that the material  lies in complete contact
           with the underlying soil

       6.  Ground anchors, called "pins", should be at least 30 cm (12  in.)  long  with an
           attached oversized washer 50 mm (2.0 in.) in diameter, or "staples" number 8 gauge
           "If* shaped wire at least 20 cm (8.0 in.) long. For less severe temporary applications
           e.g., TERMS's, one may consider 15 cm (6 in.) number 11  gauge "IT* shaped wire
           staples.

       7.  Adjacent rolls of erosion control material shall be overlapped a minimum of 75 mm
           (3.0 in.). Staples should secure the overlaps at 75 cm (2.5 ft) intervals. The roll
           ends should overlap a  minimum of 45 cm (18 in.) and be  shingled downgradient.
           The end overlaps should be stapled at 45 cm (1.5 ft) intervals, or closer, or as
           recommended by the manufacturer.

       8.  If required on the plans and specifications, the erosion control material should be
           filled with  tcpsoil, lightly raked or brushed into the mat to either fill it completely or
           to a maximum depth of 25 mm (1.0 in.).

       9.  For geosynthetic materials used in drainage ditches, their overlaps should always be
           shingled downgradient with overlaps as recommended by the manufacturer or plans
           and specifications whichever is the greatest.

       10. If required by the plans and specifications, the manufacturer  of the erosion control or
           drainage ditch material should provide a qualified and experienced representative on
           site to assist the installation contractor at the start of construction. After an acceptable
           routine  is established, the representative should be available on an  as-needed basis, at
           the CQA engineer's request
                                          27«

-------
I J
-I
-I
    c
    3
    •ac
a

j
3
5-
    •
    o
    3
     M^ii:---,  .-'•'
                  ?*l/*''
                   '?•& '• •

-------
                                Figure 8.13 - Continued
8.8 Floating Geomembrane Covers for Surface^ Impoundments

      In concluding this Chapter, it was felt that a short section on geomembrane floating covers
for liquid wastes contained in surface impoundments is appropriate.  These floating covers are
gcomembranes of the types discussed in Chapter  3.  Hence all details such as polymer type,
production, conformance testing, etc., are applicable here as  well.  The uniqueness of the
application is that the geomembrane is always exposed to the atmosphere, thus subject to sunlight,
heat, damage, etc., and furthermore ii must be rigidly anchored to a concrete anchor trench or other
similar structure, surrounding the perimeter of the facility, see Fig. 8,4.4.

      Some items in addition tp those mentioned in Chapter 3 on  geomemhranes  that are
recommended for a contract specification or a CQA document are as follows:

      1.  Acceptance of the geomembrane should  have some verification as to its weatherability
          characteristics. The iests most frequently referenced are ASTM D-4355 and ASTM G-
          26,  There is also a growing body of data being developed under the ASTM G-53 test
          method.

      2,  Other conformance tests,  e.g.. physical and mechanical property tests, are product
          specific and have been described in Chapter 3.
                                         278

-------
Figure 8.14 - Surface Impoundments with Geomembranc Floating Covers along with Typical
             Details of the Support System and/or Anchor Trench and Batten Strips
                                        279

-------
      3.  The anchorage detail for floating covers is critically important. Construction plans and
          specifications must be followed explicitly.  To be noted is that there are very different
          anchorage schemes that are currently available. Some use concrete anchor blocks with
          embedded bolts which attach the geomembrane under a batten strip. Other anchorages
          are patented systems consisting of tensioned geomembrenes attached to movable dead
          weights riding inside of stationary columns. Additional schemes are also possible. In
          each case the manufacturer's recommendations should be cited in the contract
          documents and must be followed completely.
      4.  The manufacturer/fabricator cf the floating cover should provide a qualified and
          experienced representative on site to assist the installation contractor at the start of
          construction. After an initial start-up point, die representative should be available on an
          as needed basis, at the CQA engineer's request
8.9 References
AASHTO M232-90, "Corrugated Polyethylene Drainage Tubing"
AASHTO M294-90, "Corrugated Polyethylene Pipe, 12- to 36-in. Diameter"
ASTM D-698, "Moisture Density Relations of Soils and Soil/Aggregate Mixtures"
ASTM D-792, "Specific Gravity and Density of Plastics by Displacement"
ASTM D-1238, "Flow Rates of Thermoplastics by Extrusion Plastomer"
ASTM D-1248, "Po^thylene Plastics and Extrusion Materials"
 ASTM D-1505, "Density of Plastics by the Density-Gradient Technique"
ASTM D-1755, "Poly (Vinyl  Chloride) (PVC) Resins"
ASTM D-1777, "Measuring Thickness of Textile Materials"
ASTM D-178S, "Poly (Vinyl Chloride) (PVQ Plastic Pipe, Schedules 40,80 and 120"
ASTM D-2122, "Determining Dimensions of Thermoplastic Pipe and Finings"
ASTM D-2241, "Poly (Vinyl  Chloride) (PVC) Pressure Rated Pipe (SDRsSeries)"
ASTM D-2321, "Underground Installation of Thermoplastic Pipe for Sewers and Other Gravity -
    Flow Applications"
ASTM D-2412, "External Loading Properties of Plastic Pipe by Parallel Plate Loading"
ASTM D-2444, "Impact Resistance of Thermoplastic Pipe and Fittings by Means of a Tup (Falling
    Weightr
ASTM D-3034, 'Type PSM Poly (Vinyl  Chloride) (PVC) Sewer Pipe and Fittings"
ASTM D-4254, "Maximum Index Density of Soils and Calculation of Relative Density"
                                         280

-------
ASTM D-4355, "Deterioration of Geotexdles from Exposure to Ultraviolet Light and Water
       (Xenon-Arc Type Apparatus)
ASTM D-4533, 'Trapezoidal Tearing Strength of Geotextiles"
ASTM D-4595, 'Tensile Properties of Geotextiles by Wide Width Strip Method"
ASTM D-4632, "Breaking Load and Elongation of Geotextiles (Grab Method)"
ASTM D-4759, "Determining the Specification Conformance of Geosynthetics"
ASTM D-4833, "Index Puncture Resistance of Geotextiles, Geomembranes and Related Products"
ASTM D-4884, "Seam Strength of Sewn Geotextiles"
ASTM F-714, "Polyethylene (PE) Plastic Pipe (SDR-PR) Based on Outside Diameter"
ASTM G-26, "Operating Light-Exposure Apparatus (Xenon-Arc Type) With and Without Water
       for Exposure of Nonmetallic Materials"
ASTM G-53, "Operating Light- and Water-Exposure Apparatus (Fluorescent UV - Condensation
       Type) for Exposure of Nonmetallic Materials"
GRIGG1, "Geogrid Rib Tensile Strength"
U.S. Environmental Protection Agency (1991), "Inspection Techniques for the Fabrication of
    Geomembrane Field Seams," Technical Resource Document. U.S. EPA, EPA/S30/SW-
    91/051.
                                        281

-------
                                   Appendix A
                                 List of Acronyms

 AASHTO    American Association of State Highway and Transportation Officials
 API         American Petroleum Institute
 ASTM      American Society for Testing and Materials
 ATV        All-Tenain Vehicle
 CB         Cement-Bentonite
 CERCLA    Comprehensive Etwtronrneffial Response, Compensation, and Liability Act
 CH         Fat day (ASTM D-2487)
 CL         Lean day (ASTM D-2487)
 CPE         Chlorinated Polyethylene
 CQA        Construction Quality Assurance
 CQC        Construction Quality Control
 CSPE        ChJorosulfonated Polyethylene
 CSPE-R      CMorosuIfonatedPolyethylene (Scrim Reiiiforced)
 ECRM       Erosion Control and Revegetabon Mat
 HA         EthylenelnterpolymeT Alloy
 EIA-R        EthytenefateipolymefAUoy-Reinforced
 SPA         Environmental Protection Agency
 EPDM        EthytenePropyleneDiene Monomer
 FCEA        FuUyCrosslinkedElastoineric Alloy
 FML         Flexible Membrane Uner
 FTB         Rim Tear Bond
 FTM         Federal Test Method
GCL        Geosynthetic Cky Liner
GRI        Geosynthetic Research Institute
                                      282

-------
HDPE       High Density Polyethylene
EFAI         Industrial Fabrics Association International
LL          Liquid Limit
LLDPE       Linear Low Density Polyethylene
MARV       Mimimum Average Roil Value
MQA,        Manufacturing Quality Assurance
MQC        Manufacturing Quality Control
NOT         Nondestructive Testing
NICET       National Institute for Cemficarion in Engineering Technologies
PE          Professional Engineer or Polyethylene
PERM       Permanent Erosion Control and Revegetation Mat
PI           Plasticity Index
PL          Plastic Limit
PP          Polypropylene
PVC         Polyvinyl Chloride
CA          Quality Assurance
QC          Quality Conoid
RCRA       Resource Conservation and Recovery Act
SB          Soil-Bentonite
SC          Clayey Sand (ASTMD-2487)
SCB         Soil-Cement-Bentonite
SDR         Standard Dimension Ratio
TERM       Temporary Erosion Control and Revegetation Mats
TGA         Thernwgravirnetric Analysis
TRM         Turf Reinforcement and Revegetation Mat
USCS        Unified Soil Classification System
                                        283

-------
USP
VLDP1
U.S. Pharmaceutical
Very Low Density Polyethylene
                                       284

-------
United States
Environmental Protection
Agency
Solid Waste
and Emergency Response
(5306W)
EPA530-F-97-002
July 1997
 Geosynthetic  Clay  Liners
 Used   in  Municipal
 Solid  Waste  Landfills
      his fact sheet describes new and innovative technologies and products
      that meet the performance standards of the Criteria for Municipal Solid
      Waste Landfills (40 CFR Part 258).
  Geosynthetic clay liners (GCLs) represent a relatively new technology (devel-
oped in 1986) currently gaining acceptance as a barrier system in municipal solid
waste landfill applications. Federal and some state regulations specify design stan-
dards for bottom liners and final covers. Alternative technologies are allowed,
however, if they meet federal performance standards. GCL technology is an alter-
native that performs at or above standard federal performance levels.
  GCL technology offers some unique advantages over conventional bottom
liners and covers. GCLs, for example, are fast and easy to install, have low
hydraulic conductivity (i.e., low  permeability), and have the ability to self-repair
any rips or holes caused by the swelling properties of the bentonite from
which they are made, GCLs are cost-effective in regions where clay is not read-
ily available.  A GCL liner system is not as thick as a liner system involving the
use of compacted clay, enabling engineers to construct landfills that maximize
capacity while protecting area ground water.
  Before using a GCL in a landfill barrier system, remember there currently are
no standard methods for comparing GCL products or installation systems. In
addition, GCL performance properties, including the ability of GCL liner systems
to effectively prevent landfill  leaching, have not yet been firmly established.
  This emerging technology is currently in use at a number of sites across the
nation. This fact sheet provides information on this technology and presents
case studies of successful applications.
 GCL Technology

 Materials
 A GCL is a relatively thin kyer of processed
 clay (typically bentonite) either bonded to a
 geomembrane or fixed between two sheets of
 geotextile. A geomembrane is a polymeric sheet
 material that is impervious to liquid as long as
 it maintains its integrity. A geotextile is a woven
 or nonwoven sheet material less impervious to
 liquid than a geomembrane, but more resistant
 to penetration damage. Both types of GCLs are
 illustrated in Figure 1. Although die overall
 configuration of die GCL affects its perfor-
          mance characteristics, the primary performance
          factors are clay quality, amount of clay used per
          unit area, and uniformity.

            Bentonite is an extremely absorbent, granu-
          lar clay formed from volcanic ash. Bentonite
          attracts positively charged water particles;
          thus, it rapidly hydrates when exposed to liq-
          uid, such as water or leachate. As the clay
          hydrates it swells, giving it the ability to "self-
          heal" holes in the GCL. In laboratory tests on
          bentonite, researchers demonstrated that a
          hole up to 75 millimeters in diameter will seal
          itself, allowing the GCL to retain die proper-
          ties that make it an effective barrier system.
    ) Printed on paper that contains at least 20 percent postconsumer fiber.

-------
  Figure 1. General Configurations of GCLs
                    Bentonite Sandwiched Between Two Geotextiles
      Geotextile-

      Bentonite-

      Geotextile-

                     Bentonite Glued to Geomembrane
   Bentonite is affixed to synthetic
materials in a number of ways to form
the GCL system. In configurations
using a geomembrane, the clay is
affixed using an adhesive. In geotextile
configurations, however, adhesives,
stitchbonding, needlepunching, or a
combination of the three, are used.
Although stitchbonding and
needlepunching create small holes in
the geotextile,  these holes are sealed
when the installed GCL's clay layer
hydrates. Figure 2 shows cross-section
views of the three separate approaches
to affixing bentonite to a geotextile.

Properties and
Characteristics
An important  criterion for selecting an
effective landfill barrier system is
hydraulic conductivity. Before choosing
a barrier system, the landfill operator
should test the technology under con-
sideration to ensure that its hydraulic
conductivity, as well as other character-
istics, are appropriate for the particular
landfill site.
Hydraulic Conductivity
GCL technology can provide barrier
systems with low hydraulic conductivi-
ty (i.e., low permeability), which is the
rate at which a liquid passes through a
material. Laboratory tests demonstrate
that the hydraulic conductivity of dry,
unconfined bentonite is approximately
1 x 106 cm/sec. When saturated, how-
ever, the hydraulic conductivity of ben-
tonite typically drops to less than
1 x 109 cm/sec.
  The quality of the clay used affects a
GCL's hydraulic characteristics. Sodium
bentonite, a naturally occurring com-
pound in a silicate clay formed from
volcanic ash, gives bentonite its distinct
properties. Additives are used to
enhance the hydraulic properties of
clay containing low amounts of sodium
bentonite.
  Hydraulic performance also relates
to the amount of bentonite per unit
area and its uniformity. The more ben-
tonite used per unit area, the lower the
system's hydraulic conductivity.
Although the amount of bentonite per
                                        unit area varies with the particular
                                        GCL,  manufacturers typically use 1
                                        pound per square foot. As a result, the
                                        hydraulic coaduomty of most GCL
                                        fModucfs range$:6o|8 about 1 x 16s
material, add the r>wduct 
-------
 Testing
 GCL configurations for barrier systems
 are based on the design specifications
 of each specific project. The American
 Society for Testing and Materials
 (ASTM) developed standardized labora-
 tory tests for assessing mass per unit
•,,W€^ fASTM                         '
Figure 2. Affixing Bentonite to Geotextiles
•shear (ASTM,:
   Researchers at the Geosynthetlc
 Research Institute at Drexel University
 (in Philadelphia, Pennsylvania) and the
 Geotechnical Engineering Department
 at the University of Texas (in Austin)
 developed tests to measure shear
 strength, as well as confined swelling,
 rate of creep, and seam overlap perme-
 ability. These test methods have been
 adopted by ASTM. Additionally, the
 bentonite industry developed a test to
 measure free swell (USP-NF-XVII).
   Test values for hydraulic conductivity
 depend on the degree of effective over-
 burden stress around the GCL during
 testing. The higher the effective overbur-
 den stress, the lower the hydraulic con-
 ductivity. When comparing two different
 bentonite products, both must be sub-
 jected to the same degree of effective
 overburden stress.


 Available  GCL

 Products

 Product Types
 The following types of GCL products
 are currently available:
 • Geotextile type:
   —  Bentofix® (activated sodium
       bentonite as primary ingredient
       and affixed by needlepunching
       to a woven or nonwoven upper
       geotextile and a nonwoven lower
       geotextile).
   —  Bentomat® (sodium bentonite
       as primary ingredient and affix-
       ed by needlepunching to a
       Clay Bound With Adhesive to
       Upper and Lower Geotextiles
                  CLAY AND ADHESIVE
                                                         Upper Geotextile
                                                                                                 Lower Geotextile
                                               Clay Stitchbonded Between
                                               Upper and Lower Geotextiles
           ftf CLAY AND(ADHESIVE'oR1		'	j
           '•f*r*r*f*f***f*m fuvf**ff*ftf*fr f*ftf*f
                                                                                               Upper Geotextile

                                                                                               Stitchbonded in
                                                                                               Rows

                                                                                               Lower Geotextile
                                               Clay Needlepunched Through
                                               Upper and Lower Geotextiles
                                                                                                 Upper Geotextile
                                                                                                 Needlepunched
                                                                                                 Fibers
                                                                                                 Throughout

                                                                                                 Lower Geotextile
                                              woven or nonwoven upper geo-
                                              textile and a nonwoven lower
                                              geo textile).
                                          —  Claymax® (sodium bentonite as
                                              primary ingredient mixed with
                                              water-soluble adhesive and bond-
                                              ed or Stitchbonded to a woven
                                              upper and lower geotextile).
                                          Geomembrane type:
                                          —  Gundseal® (sodium bentonite as
                                              the primary ingredient mixed with
                                              an adhesive and bonded to a blend
                                            of high density polyethylene and
                                            very low density polyethylene).

                                         Table 1 lists information on varia-
                                       tions of these product types by manu-
                                       facturer, and Figure 3 presents
                                       cross-section views of these product
                                       configurations.
                                         In general, manufacturers ship GCL
                                       products in rolled sheets ranging from
                                       13 to 18 feet wide and from 100 to 200
                                       feet long. GCLs range in thickness from
                                       0.2 to 0.3 inches.

-------
  Table 1. Principal GCL Products Available in the United States
  Manufacturer &
  Product Name
Upper           j Lower          [  Bonding Method
Geosynthetic*    Geosynthetic3
Standard Roll
Width x Length
(feet)
Bentofix NS
Bentofix WP

woven
woven
nonwoven
nonwoven
b
needlepunched
needlepunched
(15.2x100)
(15.2x100)

                       nonwoven
                                         nonwoven
                                    needlepunched
(15.2 x 100)
Claymax 200R
Claymax 500SP
Claymax 506SP
Bentomat "ST"
Bentomat "N"
woven
woven
woven
woven
nonwoven
woven
woven
woven
nonwoven
nonwoven
adhered
adhered and stitchbonded
adhered and stitchbonded
needlepunched
needlepunched
(13.83x 150)
(13.83x150)
(13.83x150)
(15.3x125)
(15.3x125)
Gundseal HD 20
Gundseal HD 30
Gundseal HD 30
Gundseal HD 60
Gundseal HD 80
Gundseal HD 40
Gundseal HD 60
Gundseal HD 80
d
none

-------
                                                                                                                            1
Installation
Landfill operators can install all available
GCL products much fester and more easily
dtan etanpacted clay finers, Unlike com-
pacted clay liners, however, GCLs are
         during and after installation to
                                  '
. iihcoji fifieS ,
 and causes tf*e geetetSc layers to puB:
 apart, undermining the integrity of the
 GCL configuration,
   Manufacturers usually specify indi-
 vidual GCL installation procedures,
 Basic procedures, however, call for
 rolling out the large GCL sheets onto
 the site subgrade, which should be
 smooth (e.g., free of stones and grade
 stakes), well compacted, and dry. Once
 installers cover the GCL with soil, the
 GCL hydrates by drawing moisture
 from the soil. As a result, when laying
 out the GCL, installers must allow
 enough seam overlap at adjoining
 sheets to guard against the potential
 opening of the barrier system.
 Currently, the recommended amount
 of seam overlap  and other seaming con-
 siderations vary  with the  particular
 GCL product. Thus, installers should
 follow the manufacturers instructions
 for the particular product.
   GCL manufacturers, and some pri-
 vate engineering firms, provide training
 for GCL installers. Among other con-
 siderations, instructions typically
 emphasize techniques for minimizing
 potential damage to the GCL during
 installation. The National Institute for
 Certification of Engineering
 Technologists in Alexandria, Virginia,
 offers a certification program in quality
 assurance and quality control inspec-
 tion of GCL installations.
                                          Figure 3. Available GCL Products
                                          Bentofix and Bentomat
                                                      SODIUM BENTONITE
                                         !iiiiL__
                                         s »*•« %vL * & ^•v«'K%V'%^sl»sVC*"C'» Oi *S* %** %\ <• ^» s*« ^ V« *& *** "i* %*• ^-'
                                         mSmfmfmJmfmJmJmfmfmSmSmJmSmfmJmJmfmS*f»fmfmf*fmJ*J*f*fmf
                                         ^•ooo^^.^v^.^.v.^.sV^.oJ^.vt^.^v^vor^.^.<.oo%v^Vsv
                                                        -Woven Geotextile

                                                        _Needlepunched
                                                         Fibers
                                                        .Nonwoven
                                                         Geotextile
  Claymax 200R
                                          V«V'V"S'S'W.»V«^«S«S'S«Sl1,»%l1i"^'^»^'^'%"%"11-"1^"^''^'*-''*.
                                          ggffi'!:#?'ff*!*f+**:+!f!*!+**>+f*ff!'f!f!*t+f+f*lJfl**fff-

                                          •iSSSS1?1
  t
                SODIUM BENTONITE
            MIXED WITH AN ADHESIVE
                                                        • Woven Geotextile
                                                        . Open-Weave
                                                         Geotextile
  Claymax 500SP
                                                        SODIUM BENTONITE
                                                    MIXED WITH AN ADHESIVE
                                                         Woven Geotextile

                                                         Sewn Stitches

                                                         Woven Geotextile
  Gundseal
                SODIUM BENTONITE
             MIXED WITH AN ADHESIVE
                                            ^".V".V«>*3
                                                         . Polyethylene
                                                          Geomembrane
Costs
As of 1994, the cost of an installed GCL
ranged from $0.42 to $0.60 per square
foot. Factors affecting the cost of a GCL
include:
• Shipping distance
" Size of the job
                                                                               " Market demand
                                                                               " Time of the year

                                                                                  In general, GCL barrier systems are
                                                                               especially cost-effective in areas where
                                                                               clay is not readily available for use
                                                                               as a liner material.

-------
Issues To  Be

Addressed

This emerging technology requires addi-
tional field and laboratory testing to
further assess its effectiveness as a
landfill baxrier system in terms of the
key performance factors discussed
below. Improved product design and
installation standards must also be
established,

Performance Factors
Further research is needed into the
following key performance factors of
GCLs:

Hydraulic Conductivity
Available data on the hydraulic con-
ductivity of various GCL configura-
tions  are gathered exclusively under
laboratory conditions. Data from
field tests should be collected to
establish product design values.

Bearing Capacity
A study by the Geosynthetic
Research Institute provides the basis
for allaying  some concerns about the
bearing capacity of hydrated GCLs,
but more research is needed. The
study demonstrated that an adequate
layer  of cover soil (according to the
product manufacturers' recommen-
dations), placed on GCLs during
installation, prevents a decrease in
liner thickness with the application
of a load. Without a sufficient soil
layer, GCLs  become compressed,
raising their hydraulic conductivity
(i.e., making them more permeable)
and reducing their effectiveness as a
barrier.
Slope Stability
Research is ongoing on the slope stabili-
ty of GCLs used in landfill sidewall
applications to determine whether this
use of GCLs provides sufficient resis-
tance to internal shear and physical dis-
placement. Additional data are needed
to support the preliminary results of a
U.S. Environmental Protection Agency
field study indicating good stability of
GCL technology following capping
operations. This study mimicked the
construction stresses all four GCL prod-
ucts (see Figure 3) are subjected to dur-
ing capping. Constructed in November
1994, the study site used five plots of
GCL placed at a 3 to 1 slope and eight
plots placed at a 2 to 1  slope. All plots
had a 3-foot-thick soil cap. Researchers
collected information on the soil and
clay moisture  of die GCL using internal
probes, and they measured the GCL for
physical displacement. Results to date
indicate good slope stability for all plots.

Long-Term Reliability
The geotextile or geomembrane in
GCL products remains durable for
long periods of time.

Freeze and Thaw Cycles
Freeze and thaw cycles do not affect
GCLs used in landfill bottom liner
applications because these systems are
installed below me frost line. Limited
laboratory data indicate that die
hydraulic conductivity of GCLs is not
affected by freeze and thaw cycles.
Laboratory tests performed on a
bentonitic blanket indicate that
hydraulic conductivity before freezing
of 2 x 10'" cm/sec was unaltered after
five freeze and thaw cycles. Full-scale
field tests still must be conducted, how-
ever, to  corroborate the laboratory data,
especially for GCL technology used as
an infiltration barrier in landfill caps.
Design and  Installation
Standards
The following issues must be
addressed to encourage the further
development of GCL technology as a
landfill barrier system:

Material Properties and Additional
Testing Methods
To allow design engineers to develop
more precise site specifications, a list
of important performance properties
for  materials used in GCL products,
as well as minimum performance val-
ues, must be established. Additional
testing procedures must be developed,
and all methods  should be standard-
ized to facilitate  the realistic compari-
son of different GCL products.

Construction and Installation
Procedures
Standardized practices must be devel-
oped to address GCLs' vulnerability to
the following:

• System stress from inclement weather
  after installation.

" Potential for lack of hydration of
  bentonite clay in arid regions.

• Punctures in the barrier system
  (reducing the barrier potential of
  both die clay and the geosynthetics).

• System decay caused by biological
  intruders, such as burrowing animals
  and tree roots  (potentially affecting
  both the clay and the geosynthetics).
  Additionally, a standardized quality
assurance and quality control  program
must be developed.

-------
Case Studies
    The following case studies illustrate some of the
    uses of GCL technology as a barrier system in
    landfills. Currently available information from
    these sites relates to installation only; long-term
    performance is still being assessed. Only one of
    the studies concerns the use of GCL technology
    in bottom liner applications, because this use is
    relatively new. The other two studies focus on cap
    system applications, which represent a slightly
    more established use of the technology. The case
    studies represent sites in three different geograph-
    ic regions and involve three different GCL
    products.

    GCL Landfill Liner;
    Broad Acre Landfill
    Pueblo, Colorado

    Broad Acre Landfill installed a liner system in 1991
    that included:
    • A 60-mil Gundseal GCL
    • 1 foot of compacted clay
       According to landfdl operators, the Gundseal
    was easy to work with. They installed 200,000
    square feet in 1 week. Workers installed the liner
    with the bentonite side down (i.e., the geomem-
    brane side up). As of February 1996, landfill
    officials reported that the liner was functioning
    effectively. No releases of leachate have been
    detected by the ground-water monitoring
    system.
GCL Landfill Cap:
Whyco Chromium Landfill
Thomaston, Connecticut
During July 1989, Whyco Chromium Landfill
installed a Claymax 200R GCL in a cap system chat
included the following (from top to bottom):
• 6 inches of topsoil
• 24 inches of earthen material
" Geogrid (for tensile strength)
• Geotextile
• Polyvinyl chloride geomembrane (30-mil thickness)
• Claymax
• Geotextile
  The landfill site occupies 41,000 square feet, and
workers installed the Claymax product in 1 day.
Thus far, the cap is functioning well.

GCL Landfill Cap:
Enoree Landfill
Greenville, South Carolina
In August 1994, the first phase of closure at the
Enoree Landfill involved installing the following
cap system:
• 6 to 12 inches of new and native soil
• 18 inches of compacted clay
• Bentofix GCL
  Enoree staff capped approximately 26 acres of the
landfill in 6 weeks. Landfill officials report that the
cap is functioning effectively.
                                                                        ie mention or
                                                                                        us I act slice i uoo
                                                                      not constitute or imply endorsement or
                                                                      approval tor UM.' by ihc U.S.
                                                                      FnvironnK'iual Protection Aircncv.

-------
                                                                                                                                                    I
References
Daniel, D.E., and R.B. Gilbert, 1994. Geosynthetic Clay Liners for Waste
Containment and Pollution Prevention. Austin, Texas: University of Texas at
Austin. February.

Koerner, R.M., and D, Narejo. 1995. Bearing capacity of hydrated
geosyntheric clay liners. J. Geotech. Eng., January:82-85.
Shan, H.Y., and D.E. Daniel.  1991, Results of Laboratory Tests on a
Geotextile/Bentonite Liner Material. Proceedings, Geosynthetics 1991,
Industrial Fabrics Association International, St. Paul, MN, vol. 2,
pp. 517-535.
U.S.  EPA. 1995. Effect of Freeze/Thaw on the Hydraulic Conductivity of
Barrier Materials; Laboratory and Field Evaluation. EPA600-R-95-118.
Prepared by Kraus, J.E, and C.H. Benson for the Risk Reduction Engineering
Laboratory, Cincinnati, OH.


Sources of Additional

Information

ASTM. 1994. ASTM Standards and Other Specifications  and Test Methods
on the Quality Assurance of Landfill Liner Systems. ASTM, 1916 Race Street,
Philadelphia, PA. April.

Daniel, D.E. 1992. Compacted Clay and Geosynthetic Clay Liners. American
Society of Civil Engineers National Chapter Section: Geoteehnical Aspects of
Landfill Design. National Academy of Sciences, Washington, DC, January.
Daniel, D.E,, and R.M, Koerner. 1993. Geotechnical Aspects of Waste
Disposal (ch. 18). In: Daniel,  D.E., ed., Geotechnkal Practice for Waste
Disposal. Chapman and Hall, London.
Elth, A.W., J. Boschuk, and R.M. Koerner. Prefabricated Bentonire Clay
Layers. Geosynthetic Research Institute, Philadelphia, PA.
Estornell, P. 1991. Bench-Scale Hydraulic Conductivity Tests of Bentonitic
Blanker Materials for Liner and Cover Systems. University of Texas at Austin.
August.

Fang, H.Y,  1995. Bacteria and Tree Root Attack on Landfill Liners: Waste
Disposal by landfill, Balkema,  Rotterdam, pp. 419-426.

Fang, H.Y., S, Pamukcu, and R.C. Chancy. 1992. Soil-Pollution Effects on
Geotextile Composite Walls. American Society for Testing and Materials.
Special Technical Publication 1129:103-116.

Grube. W.E., and D.E. Daniel. 1991. Alternative Barrier Technology for
Landfdl Liner and Cover Systems. Air and Waste Management Association,
84th Annual Meeting and Exhibition, Vancouver, British Columbia,
June 16-21.

Koerner, R.M. 1994. Designing with Geosynthetics. Third ed. Prentice Hall.

McGrath, L.T., and P.D. Creamer. 1995. Geosynthetic clay liner application.
Waste Age  Magazine, May;99-104.

Schubert, W.R. 1987. Bentonite Matting in Composite Lining Systems.
Geotechnical Practice for Waste Disposal. American Society of Civil
Engineers,  New York, NY, pp. 784-796.

U.S. EPA,  1990, Compilation of Information on Alternative Barriers for Liner
and Cover  Systems. EPA600-R-91-002. Prepared by Daniel, D.E., and P.M.
Estornell for Office of Research and Development, Washington, DC. October.

U.S. EPA.  1992. Construction Quality Management for Remedial Action and
Remedial Design Waste Containment Systems. Technical Guidance
Document. EPA540-R-92-073. Risk Reduction Engineering Laboratory,
Cincinnati, OH.

U.S. EPA.  1993. Reporr of Workshop on Geosyntheric Clay Liners. EPA600-
R-93-171,  Office of Research  and Development, Washington, DC. August.

U.S. EPA,  1993. Quality Assurance and Quality Control for Waste
Containment Facilities. Technical Guidance Document. EPA6QO-R-93-182.
Risk Reduction Engineering Laboratory, Cincinnati, OH. September.

U.S. EPA.  1996. Report of 1995 Workshop on Geosynthetic Clay Liners.
EPA600-R-96-149. Washington, DC. June
     United States
     Environmental Protection Agency
     (5306W)
     Washington, DC 20460

     Official Business
     Penalty for Private Use
     $300

-------
                                  PB91-204354
                                  EPA/530/SW-91/054
                                  June 1991
     TECHHICAL RESOURCE DOCUMENT



DESIGN, CONSTRUCTION, AHD OPERATION


                  OF

  HAZARDOUS AND NON-HAZARDOUS WASTE


        SURFACE IMPOUNDMENTS
          Robert P.  Hartley
        Cincinnati,  OE  45230

 EPA Purchase Order No. 1C6081 NATX
            Project officer

          Robert E.  Landreth
Waste Minimisation, Destruction,  and
      Disposal Research Division
Risk Reduction Engineering  laboratory
        Cincinnati,. OE  45268
RISK REDUCTIOK ENSIHEERIMG LABORATORY
  OFFICE OF  RESEARCH AND DEVELOPMENT
 0.S. EHVIROHMEMTJO, PROTECT2O1I A61HCY
        CINCINNATI, OEIO  45268
  REPRODUCED BY                   —
  U.S. DEPARTMENT OF COMMERCE    ^§ Printed on Recycled Paper
         NATIONAL TECHNICAL
        INFORMATION SERVICE
        SPRINGFIELD. VA 22161

-------
                           DISCLAIMER


     The information in this document has been funded wholly or
in part by the United States Environmental Protection Agency
under Purchase Order No. 1C6081 if ATX to Robert P.  Hartley.  It
has been subjected to the Agency's peer and administrative
review, and it has been approved for publication as an EPA
document.  Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
                                ii

-------
                            FOREWORD


     Today's rapidly developing and changing technologies and
industrial products and practices frequently carry with them the
increased generation of solid and hazardous wastes.  These
materials, if improperly deaj-t with, can threaten both public
health and the environment.  Abandoned waste sites and accidental
releases of toxic and hazardous substances to the environment
also have important environmental and public health implications.
The Risk Reduction Engineering Laboratory assists in providing an
authoritative and defensible engineering basis for assessing and
solving these problems.  Its products support the policies,
programs and regulations of the Environmental Protection Agency,
the permitting and other responsibilities of State and local
governments, and the needs of both large and small businesses in
handling their wastes responsibly and economically.

     This report is a Technical Resource Document, summarizing
the state-of-the-art in the design, construction and operation of
hazardous waste and non-hazardous waste surface impoundments.
Details are generally left to referenced materials.  Some of the
information, presented in more detail, has not been previously
published.  Most of the information has been gathered in the
course of hazardous waste research, in accord with past
regulatory emphasis.  However, it is believed that most of the
technical information will also be applicable to non-hazardous
waste surface impoundments.


                          E. Timothy Oppelt, Director
                     Risk Reduction Engineering Laboratory
                               111

-------
                             PREFACE


     Subtitle C of the Resource Conservation and Recovery Act
(RCRA) requires the O, S. Environmental Protection Agency (EPA)
to establish a Federal hazardous waste management program.  The
program must ensure that hazardous wastes are handled safely from
generation until final disposal.  EPA issued a series of
hazardous waste regulations under Subtitle C of RCRA that are
published in Title 40 Code of Federal Regulations (CFR)  Parts 260
through 265 and Parts 122 through 124.

     Parts 264 and 265 of 40 CFR contain standards applicable to
owners/operators of all facilities that treat, store, or dispose
of hazardous wastes.  Wastes are identified or listed as
hazardous under 40 CFR Part 261.  Part 264 standards are
implemented through permits issued by authorized States or EPA
according to 40 CFR Part 122 and Part 124 regulations.  Land
treatment, storage, and disposal (LTSD) regulations in 40 CFR
Part 264 issued on July 26, 1982, and July 15, 1985, establish
performance standards for hazardous waste landfills, surface
impoundment*, land treatment units, and waste piles.  Part 265
standards impose minimum technology requirements on the
owners/operators of certain landfills and surface impoundments.

     EPA is developing three types of documents to assist
preparers and reviewers of permit applications for hazardous
waste land disposal facilities.  These are RCRA Technical
Guidance Documents  (TGDs), Permit Guidance Manuals (PGMs), and
Technical Resource Documents (TRDs).  Although emphasis is given
to hazardous waste facilities, the information presented in these
documents may be used for designing, constructing, and operating
non-hazardous waste LTSD facilities as well.

     The RCRA TGDs present design, construction, and operating
specifications or evaluation techniques that generally comply
with or demonstrate compliance with the design and operating
requirements and the closure and post-closure requirements of
Part 264.  The Permit Guidance Manuals are being developed to
describe the permit application information the Agency seeks and
to provide guidance to applicants and permit writers in
addressing information requirements.  These manuals will include
a discussion of each step in the permitting process and a
description of each set of specifications that must be considered
for inclusion in the permit.
                                xv

-------
     The TGDs and PGMs present guidance,  not regulations.  They
do not supersede the regulations promulgated under RCRA and
published in the CFR.  Instead, they provide recommendations,
interpretations, suggestions, and references to additional
information that may be used to help interpret the requirements
of the regulations.  The recommendations  of methods, procedures,
techniques, or specifications in these manuals and documents is
not intended to suggest that other alternatives might not satisfy
regulatory requirements.

     The TRDs present summaries of state-of-the-art technologies
and evaluation techniques determined by the Agency to constitute
good engineering designs, practices, and  procedures.  They
support the RCRA TGDs and PGMs in certain areeas by describing
current technologies and methods for designing hazardous waste
facilities or for evaluating the performance of a facility
design.  Whereas the RCRA TGDs and PGNs are directly related to
the regulations, the information in the TRDs covers a broader
perspective and should not be used to interpret the requirements
of the regulations.

     This document is a Technical Resource Document.  It reflects
the considerable research that has been performed in the area of
waste containment and the experience that has been gained in this
technology.  It incorporates responses to many comments received
in the peer review of the draft document.

     Comments on this publication will be accepted at any time.
The Agency intends to update these TRDs periodically based on
comments received and/or the development  of new information.
Comments on any of the current TRDs should be addressed to Docket
Clerk, Room S-269(c), Office of Solid Waste and Emergency
Response (WH-562), U. S. Environmental Protection Agency, 401 M
Street, S.W., Washington, D.C., 20460.  Communications should
identify the document by title and report number.

-------
                            ABSTRACT


    i This Technical Resource Document provides current
information on the design, construction, and operation of surface
impoundments used for the treatment, storage, or disposal of
hazardous and non-hazardous wastes.  The pertinent regulations
under the Resource Conservation and Recovery Act (RCRA) are
summarized.  Surface impoundment structures that will meet the
regulatory requirements are described.  RCRA's "minimum
technology requirements11 specify double-lined structures with a
leak collection and removal layer between the two liners.  Dikes,
liners, and leak collection layers may be constructed of
combinations of soils and synthetic materials in multilayer
systems.  Other components, also described in the document,
include leak detection systems, secondary containment, and liquid
level monitoring systems.  Methods for closing surface
impoundments, either in-place or by waste removal, are described.
In-place closure requires waste treatment and stabilization and
the installation of a landfill cover that also must meet minimum
technology regulatory requirements.  Cover technology is also
summarized.  Details of the technologies summarized in this
document may be found in the many references cited.

     This report was submitted in fulfillment of EPA Purchase
Order No. 1C6081 NATX by Robert P. Hartley under sponsorship of
the u. s. Environmental Protection Agency.  This report covers a
period from October 1990 to February 1991, and work was completed
as of February 15, 1991.
                                vi

-------
                        TABLE OF CONTENTS
SECTION                                                     PAGE

DISCLAIMER	ii
FOREWORD	iii
PREFACE	iv
ABSTRACT	vi
LIST OF FIGURES	xiii
LIST OF TABLES	xvi
ACKNOWLEDGMENTS	xvii

CHAPTER 1.  INTRODUCTION 	 1

     1.1  PURPOSE AND SCOPE	1
     1.2  BACKGROUND INFORMATION AND PREVIOUS STUDIES	1
     1.3  SURFACE IMPOUNDMENT REGULATIONS	4

          1.3.1  Design and Operating Requirements 	 4
          1.3.2  Monitoring and Inspection Requirements. ... €
          1.3.3  Emergency Repairs and Contingency Plans ... 6
          1.3.4  Response Action Plans 	 7
          1.3.5  Closure and Post-closure Care	8

CHAPTER 2.  PRE-DESIGN CONSIDERATIONS	9

     2.1  TOPOGRAPHY	9
     2.2  SURFACE AND SUBSURFACE HYDROLOGY 	 9
     2.3  GEOLOGY AND SUBSURFACE SOIL CONDITIONS	11
     2.4  LAND USE	14
     2.5  CLIMATE	16

          2.5.1  Flooding	16
          2.5.2  Precipitation vs. Evaporation *	18
          2.5.3  Soil Freezing and Thawing	.  .  . 19

     2.6  AIR QUALITY	19

CHAPTER 3.  DESIGN	21

     3.1  SELECTION OF BASIC CONFIGURATION	21

          3.1.1   Impoundment Type	...21
          3.1.2   Number, Size, and Position	21
          3.1.3   Impoundment Surface Area	24
          3.1.4   Impoundment Depth	25

                  3.1.4.1  Normal  Operating Level	26
                  3.1.4.2  Maximum Operating Level without
                          External Runoff Input	26
                                via.

-------
                  TABLE OF CONTENTS (continued)

SECTION                                                     PAGE

                            Water Budget Approach	27
                            Design Storm Approach	30

                 3.1.4.3  Maximum Operating Level with
                          External Watershed	31
                 3.1.4.4  Freeboard Determination	32

     3.2  STRUCTURAL COMPONENT DESIGN	33

          3.2.1  Foundation Analysis 	 ......34

                 3.2.1.1  Settlement	34
                 3.2.1.2  Bearing Capacity	35

          3.2.2  Dike Design	35

                 3.2.2.1  Shear Strength .......... .37
                 3.2.2.2  Slope Stability Analysis ..... .39

                            Minimum Factory of Safety. ... .42

          3.2.3  Liner Systems	44

                 3.2.3.1  Regulatory Constraints and
                          Guidance	44
                 3.2.3.2  Geomembrane/Composite Double Liner .46
                 3.2.3.3  Geomembrane/Compacted Soil
                          Double Liner	47
                 3.2.3.4  Double Composite Liner	48
                 3.2.3.5  Multiple-Layer Liner
                          Materials and Specifications ... .49

                            Geomembranes	49
                            Geomembrane Protective
                            Layers	52
                            Low-Permeability Soil
                            Liners	54
                            Leak Detection and
                            Collection Systems	57
                            Geotextile Filter	59
                            Gas-Venting Layer	60

     3.3  LIQUID LEVEL CONTROL	60
     3.4  SECONDARY CONTAINMENT.	 .65
     3.5  LEAK DETECTION SYSTEMS	65
     3.6  SURFACE WATER MANAGEMENT	68
     3.7  CONTROLS FOR VOLATILE ORGANIC COMPOUND EMISSIONS  . .71
     3.8  CONSTRUCTION QUALITY ASSURANCE (CQA) PLAN. . . .  . .73

                               viii

-------
                  TABLE OF CONTENTS (continued)

SECTION                                                     PAGE

CHAPTER 4.  CONSTRUCTION	 .74

     4.1  SITE PREPARATION	 .74
     4.2  CUT-SLOPE AND FOUNDATION CONSTRUCTION	74

          4.2.1  Cut-slopes.	75
          4.2.2  Dike Foundation	 .76

     4.3  DIKE AND SOIL LINER CONSTRUCTION	77

          4.3.1  General Construction Process. .  .  t, . . . , .77
          4.3.2  Pre-placement Soil Preparation	79
          4.3.3  Soil Material Placement ...........80
          4.3.4  Soil Compaction	.80

     4.4  GEOMEMBRANE LINER INSTALLATION 	  ..... .82

          4.4.1  Storage of Materials and Equipment. .... .83
          4.4.2  Construction Quality Assurance/Inspection . .83
          4.4.3  Subgrade Preparation	83
          4.4.4  Geomembrane Liner Placement	...84
          4.4.5  Sealing Around Structures and Anchoring
                 the Geomembrane	86

     4.5  LEAK COLLECTION AND REMOVAL SYSTEMS	88
     4.6  TESTING THE LINER SYSTEM	90
     4.7  PROTECTIVE COVERINGS	91

          4.7.1  Liner Protection	.91
          4.7.2  Dike Protection	91

     4.8  LIQUID LEVEL CONTROL SYSTEMS	92

          4.8.1  Active Liquid Level Control	 . .92
          4.8.2  Passive Liquid Level Control	93

     4.9  SECONDARY CONTAINMENT.	93

CHAPTER 5.  OPERATION, MAINTENANCE, AND MONITORING ..... .94

     5.1  OPERATION AND MAINTENANCE ACTIVITIES	94

          5.1.1  Facility Start-up	94
          5.1.2  Routine Inspections and Maintenance  .... .95

                 5.1.2.1  Regulatory Inspections of
                          Facility	 .95

                                ix

-------
                  TABLE OF CONTENTS (continued)

SECTION                                                     PAGE

                 5.1.2.2  Operator Inspections of Dike
                          Slopes,  Faces,  and Crest	97
                 5.1.2.3  Operator Inspections of Ancillary
                          Site Facilities	97
                 5.1.2.4  Liner Systems	99

                            Detecting and Measuring
                            Liner  Leakage	99
                            Determining the Cause of
                            Liner  Leakage	102
                            Liner  Repair	102
                            Solids and Liquid Removal. .  . . 103

          5.1.3  Record-keeping	104


     5.2  SAMPLING AND ANALYSIS MONITORING ACTIVITIES. .  . . 104

          5.2.1  Hazardous Waste Monitoring	104
          5.2.2  Air Monitoring	107

                 5.2.2.1  Estimating Emissions from
                          Surface  Impoundments 	 107
                 5.2.2.2  Air Sampling and Analyses	109

          5.2.3  Ground-water Monitoring 	 110
          5.2.4  Soil-vapor Monitoring 	 112
          5.2.5  Leak Collection and Removal System
                 Monitoring	113

CHAPTER 6.  CONTINGENCY PLANNING	 115

     6.1  LIQUID-LOSS RESPONSE PLANS 	 115

          6.1.1  Contingency Plan	115
          €.1.2  Response Action Plan	116
          6.1.3  Corrective Action Program	118

     6.2  TYPES OF FAILURE	119
     6.3  RESPONSE PLAN IMPLEMENTATION 	 120

          6.3.1  Contingency Plan  Implementation	120

                 6.3.1.1  Immediate Actions	121
                 6.3.1.2  Contamination Assessment	122
                 6.3.1.3  Selection and Implementation of
                          Remedial Actions 	 123

-------
                  TABLE OF COHTEHTS (continued)


SECTION                                                     PAGE

                 6.3.1.4  Cleanup Verification 	 124

          6.3.2  Implementation of Response Action Flan. . . 125

                 6.3.2.1  Leak Correction Verification . . . 127

          6.3.3  Implementation of Corrective Action
                 Program	127

                 6.3.3.1  Ground-water Cleanup
                          Verification 	 128

     6.4  PERSONAL SAFETY DURING REMEDIAL OPERATIONS . . . .128

CHAPTER 7.  CLOSURE AND POST-CLOSURE CARE	130

     7.1  ASSESSMENT OF CLOSURE OPTIONS.	132

          7.1.1  Waste Characteristics ........... 134
          7.1.2  Site Location Features.	134
          7.1.3  Cost	134
          7.1.4  Intended Future Site Use	135
          7.1.5  Environmental Bisk	 135

     7.2  CLEAN CLOSURE (CLOSURE BY REMOVAL) 	 135

          7.2.1  Free Liquids.	 136
          7.2.2  Residual Sludges. .	137
          7.2.3  Subsoils, Liners, and Other Contaminated
                 Materials		137
          7.2.4  Verification Sampling 	 137

                 7.2.4.1  Sampling Schemes  ....  	 138
                 7.2.4.2  Indicator Parameters ....... 139
                 7.2.4.3  Quality Assurance/Quality Control. 139

          7.2.5  Regulatory Variance ............ 140
          7.2.6  Backfilling	141

     7.3  IN-PLACE CLOSURE ...... 	  .  	 141

          7.3.1  Removal of Free Liquids	142
          7.3.2  Sludge Dewatering 	 142
          7.3.3  Waste Residuals	 143

                 7.3.3.1  Stabilization. 	 143
                 7.3.3.2  Treatment of Residues.  ...... 145

                                xi

-------
                  TABLE OF COHTEHTS (concluded)


SBCTIOM                                                     PAGE

                            Extraction	146
                            Immobilization	146
                            Biodegradation 	 148
                            Chemical Degradation 	 148

          7.3.4  Final Cover System	149

                 7.3.4.1  Protective Surface Layer 	 151

                            Vegetation	 151
                            Topsoil Layer	151
7.3
7.3
7.3
7.3
7.3
                       2  Drainage Layer .......... 152
                       3  Biotic Barrier .......... 153
                       4  Hydraulic Barrier Layer ...... 153
                       5  Gas-Vent Layer .......... 154
                       6  Hydraulic Barrier Support Layer. . 155
     7.4  POST-CLOSURE ACTIVITIES. .	155

          7.4.1  Monitoring	157
          7.4.2  Maintenance	157
          7.4.3  Use of the Site	158
          7.4.4  Delisting	158

REFERENCES	161
                                • •
                               XII

-------
                         LIST OF FIGURES
FIG. HO.
1.  Area of net evaporation in the United States	19
2.  Maximum anticipated depths of freezing	20
3.  Components of maximum operating level for (a) treatment,
    (b) surge, and (3) evaporation/disposal .. 	 ..22
4.  Example of one large vs. four small impoundments. ... .22
5.  Components making up impoundment design depth	25
6.  Example of frequency distribution of monthly
    climatic data	29
7.  Percent confidence that design return period will
    not be exceeded during design life. ...........30
8.  Example of the effect of differential foundation
    compressibility on a surface impoundment dike ..... .34
9.  Forces and displacements in bearing capacity analysis . .35
10. Surface impoundment dike and liner interfaces
    and layers.  .. 	 ....36
11. Mohr-Coulomb failure envelopes for clays and sands.  . . .38
12. Typical compaction curves showing  (a) dry unit
    weight - water content relation and (b) variation        J
    of shear strength with water content for a
    cohesive soil	  . .  . . ;39
13. Types of slope instability.	 .40
14. Factor-of-safety contours for slope stability 	43
15. Cross section of double liner with composite
    bottom liner	.45
16. Cross section of double liner with soil-only
    bottom liner	.45
17. Cross section of double-composite bottom liner.  .... .45
18. Impoundment  dike cross section showing optional
    protective layers  ..........  	53
                               Xlli

-------
                   LIST OF FIGURES (continued)

FIG. HO.                                                   PAGE

19. Junction of side-wall geonet and bottom granular
    layers of leak detection, collection, and removal
    system	,	58

20. Example access for leak collection system and
    liquid removal	59

21. Liner system showing gas vent layer and exit
    through top of dike	61

22. Dike spillway with protective apron ..........  .62

23. Impoundment overflow discharge pipe through dike	62

24. Example of liquid-level monitoring setupj manually
    read staff gauge	..63

25. Example of liquid-level recording and alarm system. . .  .63

26. Primary and secondary containment dikes	65

27. System to detect and locate leaks in top
    (primary) liner	66

28. System to detect leakage through top (primary) liner. .  .67

29. System to detect leakage to the vadose zone	67

30. Lysimeter for leak detection beneath bottom liner  ...  .68

31. Runoff diversion past surface impoundment ........69

32. Typical diversion ditch and berm cross sections	69

33. Wind diversion fences for VOC control	72

34. Cut-slope, dike, and side-wall cross section	75

35. Idealized schematic showing effects of slope
    on compactive effort.	...76

36. Dike cross section showing lifts and final slope cuts  .  .78

37. Compactor foot designs	82

38. Geomembrane liner panel layout	.......85

39. Geomembrane anchor designs at top of dike	87

                               xiv

-------
                   LIST OF FIGURES (concluded)

FIG. NO.                                                   PAGE

40. Seals at geomembrane penetrations 	88

41. Example of leak collection system layout	90

42. Example of geotextile use in leak collection
    system drain layer. ...................90

43. Cutaway view of emission sampling apparatus	 109

44. Example layout of ground-water monitoring wells .... Ill

45. Schematic of soil-gas sampling probe	113

46. Flow chart of closure options and requirements	131

47. USEPA-reconnnended landfill cover design	150

48. USEEA-recommended landfill cover design with
    optional layers	152

-------
                         LIST OF TABLES

TABLE HO.                                                   PAGE

1.  Distribution of Surface Impoundment Applications.  .... 2

2.  Geotechnical Soil Properties Used to Characterize
    Surface Impoundment Site Soils	  .  . . . 14

3.  Geotechnical Soil Properties Used to Characterize
    Borrow Material Sources	15

4.  Sources of Climatic Data Used in Surface Impoundment
    Design and Analysis	17

5.  Cost Comparisons of Different Surface Impoundment
    Positions with Respect to Grade
    (Geomembrane/Composite Liner)	...24

6.  Current Procedures for Stability Analysis	41

7.  Typical Optimum Soil Liner Design Specifications.  ... .56

8.  Average Leak Rates (mVyr) from Different Size
    and Shape Flows in 0.08-cm HOPE Liner over Gravel
    at Two Liquid Beads	101

9.  Calculated Leak Rate (mVyr)  for a Range of Hole
    Sizes in Geomembrane Liners over Soils of
    Different Conductivities and for Three Heads	101

10. Recommended Air Emission Models for Hazardous
    Waste Disposal Facilities 	 108

11. Outline of Contingency Plan Response Data Sheet .... 117

12. Innovative Investigation Technologies to Assess
    Site Contamination	123

13. Advantages and Disadvantages of Closure Options .... 133

14. Test Procedures for Stabilized Wastes ......... 145

15. Compatibility of Surface Impoundment Features and
    Various Site Uses for In-place Closure	.159

16. Compatibility of Surface Impoundment Features and
    Various Site Uses after Hazardous Waste Removal :  .  .  . 160
                               xvi

-------
                         ACKNOWLEDGMENTS


     This document was prepared in draft form by P1I Associates,
Cincinnati,  OH,  K. W. Brown and Associates,  College Station, TX,
and E. C. Jordan Co., Portland, ME.  After updating and peer
review, it was rewritten and prepared in final fora under EPA
Purchase Order No. 1C6081 NATX with Robert P. Hartley.  The
document was prepared under the supervision of Robert E. Landreth
of the Risk Reduction Engineering Laboratory.  Dr. Robert M.
Koerner, Dr. Gregory N. Richardson, Dr.  David E. Daniel, and Dirk
Brunner were especially helpful in the final review of the
document.
                               xvii

-------
                            CHAPTER 1

                           INTRODUCTION


     A surface impoundment is an excavation or diked area.
typically used for the treatment, storage, or disposal of liquids
(e.g., wastewater) or materials containing free liquids (e.g.,
sludges).  The hydraulic barriers in surface impoundments are
usually constructed of low-permeability soil or polymeric
membranes or both.

     Liquids and solids typically separate in a surface
impoundment by gravity settling.  Liquids can be removed by
draining, evaporation, or flow from an outlet structure.
Accumulated solids may be removed by dredging during impoundment
operation or when it is closed.  Alternatively, solids may be
left in place, as a landfill, when the surface impoundment is
closed.
1.1  PURPOSE AND SCOPE

     This document summarizes and supplements existing sources of
information on the state-of-the-art in the design, construction,
operation, and closure of  surface impoundments used for waste
containment.  In addition, relevant regulations are briefly
summarized, and post-closure activities are discussed.  The
document reflects the fact that most of the available
information, ongoing research, and pertinent regulations deal
with surface impoundments  used to contain hazardous waste.
However, much of the technical information should be applicable
to non-hazardous waste impoundments, which comprise the majority
of waste impoundment sites.


1.2  BACKGROUND INFORMATION AMD PREVIOUS STUDIES

     Two major baseline  studies by the U.S. Environmental
Protection Agency  (USEPA)  have gathered background information on
surface impoundments.  The first study  (USEPA,  1983a) was a
national survey to determine the number, size, and location of
hazardous and non-hazardous waste surface impoundments and their
potential influence on ground-water quality.  The survey
characterized over 180,000 surface impoundments.  Nearly 30,000
are used by industry, including chemical manufacturers, food

-------
processors, oil refineries, primary and fabricated metals
manufacturers, paper plants, and commercial waste facilities.
Most of the remainder are used in sewage and wastewater
treatment.  Organic chemical manufacturers are the largest
industrial contributors to impounded hazardous waste.   Most
surface impoundments are not used for waste disposal but rather
for waste treatment processes (i.e., neutralization, settling,
anaerobic or aerobic digestion, pH adjustment, and polishing).
The industrial surface impoundments ranged in size from less than
0.1 acre (29 percent) to greater than 100 acres (1 percent), with
the majority less than 5 acres.  (One acre = 0.405 hectare.)

     The EPA national survey categorized surface impoundment
applications into five groups with the percentages in each group
used for storage, disposal, or treatment.  The results are shown
in Table 1.  Note that the majority of agricultural surface
impoundments were used for waste storage, the majority of oil and
gas surface impoundments for disposal, and the majority of
municipal, industrial, and mining impoundments were used for
treatment.


 TABLE 1.  DISTRIBUTION OP SURFACE IMPOUNDMENT APPLICATIONS.*
                      Storage         Disposal       Treatment
Agricultural
Municipal
Industrial
Mining
Oil & Gas
55
5
17
18
29
26
31
31
27
67
19
64
52
56
4
"EPA (1983a).


     The  second major study  (USEPA, 1984a), reviewed the design
of surface  impoundments  and  operating and maintenance practices,
based on  nine  case  studies and interviews with technical experts.
Topics included site selection, liner material selection and
performance, construction quality assurance (CQA) programs,
operation and  maintenance programs, and leak detection systems.
USEPA concluded that,  in many cases, inappropriate design
criteria, CQA  programs,  and  operation and maintenance procedures
have resulted  in  release of  impounded waste constituents to

-------
ground water and/or surface water.  The study identified areas in
need of further research and development,  including evaluation of
new design concepts, compatibility of wastes and liner materials,
and monitoring systems.

     USEPA (1982a) has identified typical  industrial surface
impoundment applications as follows:

     1.   Mining and Milling Operations — Surface impoundments
are used to contain various wastewaters such as acid mine water,
solvent wastes from solution mining, and wastes from dump
leaching.  Surface impoundments may also be used for separation
settling, washing, sorting of mineral products from tailings, and
recovery of valuable minerals by precipitation.

     2.   Oil and Gas Industry ~ one of the largest users of
surface impoundments.  Surface impoundments may contain salt
water associated with oil extraction and deep-well repressurizing
operations, oil-water and gas-fluids to be separated or stored •
during emergency conditions, and drill cuttings and drilling
muds.

     3.   Textile and Leather Industries — Surface impoundments
are primarily used for wastewater treatment and sludge disposal.
Organic species impounded include dye carriers such as
halogenated hydrocarbons and phenols; heavy metals impounded
include chromium, zinc, and copper.  Tanning and finishing wastes
may contain sulfides and nitrogenous compounds.

     4.   Chemical and Allied Products Industries — Surface
impoundments are used for wastewater treatment, sludge disposal,
and residuals treatment and storage.  Waste constituents are
process-specific, including phosphates, fluoride, nitrogen, and
assorted trace metals.

     5.   Other Industries — Surface impoundments are found at
petroleum-refining, primary metals production, wood-treating, and
metal-finishing facilities.  Surface impoundments are also used
for the containment and/or treatment of air pollution scrubber
sludge and dredging spoils sludge.

     Waste impoundments may contain complex mixtures of
materials, often aggressive to lining materials, capable of rapid
migration to ground water, and producing harmful emissions to the
surrounding air.  The design engineer must understand and
minimize these potentials.

     Much of the information needed in the design and
construction of surface impoundments can be found in Technical
Resource Documents published by the USEPA that deal with
structural components common to both landfills and surface
impoundments.  The more significant of these include "Lining of

-------
Haste Containment and Other Impoundment Facilities" (USEPA,
1988a), "Design, Construction, and Evaluation of Clay Liners for
Waste Management Facilities" (USIPA,  1988b),  and "Technical
Guidance Document: Construction Quality Assurance for Hazardous
Waste Land Disposal Facilities" (USEPA, 1986a).   The USEPA's
Office of Solid Waste has also issued technical  guidance relevant
to hazardous waste surface impoundments.  Included are "Draft
Minimum Technology Guidance on Double Liner Systems for Landfills
and Surface Impoundments — Design, Construction, and Operation"
(USEPA, 1967a) and "Technical Guidance Document: Final Covers on
Hazardous Waste Landfills and Surface Impoundments" (USEPA,
1989).  The cited documents are supported by a significant amount
of research and field experience,  most of which  is referenced in
this document.  Some of the later information has not yet been
published, but has been cited here for completeness.


1.3  SURFACE IHPOUKDMEHT REGULATIONS

     The design, construction, operation, and closure of
hazardous waste surface impoundments are regulated under
authority of the 1984 Hazardous and Solid Waste  Amendments (HSWA)
to the Resource Conservation and Recovery Act of 1976 (RCRA).
The most pertinent regulations are contained in 40 CFR 264,
Subpart K, dealing specifically with new hazardous waste surface
impoundments.  They are summarized below.


1.3.1  Designaad Operating Requirement*

     The rules for design and operation of liner systems for
hazardous waste surface impoundments, provided in 40 CFR 264.221,
are summarized as follows:

(1)  A new or expanded surface impoundment must have a liner
     system that prevents migration of wastes to the adjacent
     subsurface soil, ground water, or surface water at any time
     during the active life of the surface impoundment.  The
     liner system must be:

     *  constructed of materials with appropriate chemical
        properties and sufficient strength and thickness to
        prevent failure;

     *  placed on a foundation or base capable of supporting the
        liner and resisting pressure gradients above and below
        the liner to prevent failure; and

     •  installed to cover all surrounding earth likely to be in
        contact with the waste or leachate.

-------
(2)   New surface impoundments,  and replacement or  lateral
     expansion of existing surface impoundments, must  have  two  or
     more liners and a leachate collection system  between the
     liners.   The liner and leachate collection system must
     include  the following:

     •  a top (primary)  liner that will prevent migration of any
        chemical constituent into such liner during  the  active
        life  and post-closure care period;

     *  a lower (secondary) liner, that will prevent the
        migration of any chemical constituent through  such  liner
        during the active life and post-closure care period,
        (this requirement may be satisfied by a layer  at least  3
        feet  [0.91 meter]* thick of recompacted clay or  other
        natural material with a permeability equal to  or less
        than  3.9 x 1CT8 in./sec [1 x 10"7 cm/sec])?  and

     •  a leachate collection system that must detect  and remove
        liquids that leak through the primary liner  (without
        clogging) during the active life and post-closure care
        period.

(3)   Double-liner requirements may be waived by the  Regional
     Administrator for any monofill (i.e.,  surface impoundment
     containing one type of waste) if the following  requirements
     are met:

     *  the wastes are from foundry furnace emission controls or
        metal-casting molding sand, and contain no constituents
        that  would render them hazardous for reasons other  than
        IP Toxicity characteristics;

     *  the monofill has at least one liner for which  there is  no
        evidence of leakage;

     •  the monofill is located more than 0.25 miles (0.4
        kilometers) from an underground source of  drinking  water;

     *  the monofill is in compliance with generally applicable
        ground-water monitoring requirements for RCRA-permitted
        facilities; and
     Throughout this document, English units of measurement are
     used, followed by metric equivalents in parentheses.
     English units are primary because nearly all cited work was
     originally done using English units and because most
     engineers still use the English system, except in referring
     to hydraulic conductivity.

-------
     *  the monofill is located, designed,  and operated so as to
        assure that there will be no migration of hazardous
        constituents into ground water or surface water at any
        future time.

(4)   Surface impoundments must prevent overtopping resulting from
     normal or abnormal operations; overfilling;  wind and wave
     action; rainfall; runon; malfunctions of level controllers,
     alarms, and other equipment; and human error.

(5)   Surface impoundments must have dikes with sufficient
     structural integrity to prevent massive failure; dike
     stability should be analyzed with the presumption that
     liners will leak during the active life of the unit.

(6)   The Regional Administrator will specify in the permit, the
     design and operating practices that are necessary to ensure
     compliance with the requirements of this section.


1.3.2  Monitoring and Inspection Requirements

     The rules for monitoring and inspecting surface
impoundments, provided in 40 CFR 264.226, are summarized as
follows:

     •  During construction and installation, liners and cover
        systems (including synthetic, soil-based, and admixed
        systems) must be inspected for uniformity, damage, and
        imperfections.  (Detailed construction quality assurance
        requirements have been proposed in 40 CFR 264.19 and
        264.20 covering most structural features of a surface
        impoundment [USEPA, 19 8 7 i]).

     *  During operation, surface impoundments must be inspected
        weekly and after storms for evidence of overtopping,
        sudden drops in liquid level, and deterioration of dikes
        or other containment devices.

     •  Before a permit is issued or after an extended period of
        downtime, a qualified engineer must certify the
        structural integrity of the surface impoundment dike.
        (Note: it is assumed that a "qualified engineer" is a
        registered professional engineer.)


?.3.1  FMuiencv Repairs and Contingency Plans
     The rules pertaining to emergency repairs and contingency
plans for hazardous waste surface impoundments, provided in 40
CFR 264.227, are summarized as follows:

-------
     »  Surface impoundments must be removed from service when
        the liquid level suddenly drops for reasons  other than
        changes in the inflow or outflow (e.g., when the dike or
        liner(s) leak).

     •  To remove a surface impoundment from service,  the
        owner/operator must cease adding wastes to it; contain
        surface leakage and stop the leak;  stop or prevent
        catastrophic failure; empty the impoundment if the leak
        cannot be stopped; and notify the Regional Administrator
        within seven days of detecting the  problem.

     •  The owner/operator must provide a contingency plan, as
        required in 40 CFR 264, Subpart D,  that specifies
        procedures for complying with the requirements for
        removing a surface impoundment from service.

     •  Surface impoundments removed from service in accordance
        with these requirements may be restored to service only
        if the failed portion is repaired and the following steps
        are taken, if appropriate:  the structural integrity of
        the dike is recertified; a new liner is installed; or the
        repaired liner system is certified  by a qualified
        engineer to comply with the permitted design
        specifications.

     *  Surface impoundments removed from service and not
        repaired must be closed in accordance with the closure
        and post-closure care provisions of 40 CF1 264.228.


1.3.4  Response, Action Plans

     USEPA has proposed rules in 40 CFR 264.222 and 265.222
detailing the requirements for a response action  plan to be
implemented in case of leakage from a hazardous waste impoundment
into the leak collection system.  The proposal is summarized as
follows:                                                      j
                                                              f
     •  Two action leakage levels are proposed:  (1) an "action
        leakage rate," (ALR) of between 5 and 20  gallons/acre/
        day (47 and 187 liters/hectare/day), which requires
        Agency notification and correction, but does not
        constitute an emergency; and (2) "rapid and extremely
        large leakage rate"  (RLL) which must be determined for
        each site, and is the rate that exists when the fluid
        head is greater than the thickness  of the leak collection
        layer.

     *  Response to a RLL must be immediate and major, because of
        the increased risk of escape of contaminants in a swamped
        system.  Response to an ALR may be less  immediate,

-------
        because the leak collection system can still handle the
        leak rate.

     •  The Response Action Plan for a RLL must accompany the
        permit application for a new surface impoundment.  It may
        be submitted for ALR at a later time, but not later than
        the leak event.

     USEPA (19871) in 40 CFR 264.301 and 265.301 has also
proposed that leak collection systems be capable of detecting a
leak in the top liner as low as one gallon/acre/day (9.35
liters/hectare/day}.


1.3.S  Closi

     Rules for closure and post-closure care of hazardous waste
surface impoundments are provided in 40 CFR 264.228.  There are
currently two closure options available:

     *  clean closure, which may be accomplished by removing or
        decontaminating all waste residues, contaminated
        containment system components, contaminated subsoils, and
        contaminated structures and equipment; or

     *  in-place closure (landfill or disposal unit closure),
        which may be accomplished by dewatering and stabilizing
        the wastes and placing an impermeable cover over the
        surface impoundment to minimize infiltration.

     If clean closure is accomplished, no post-closure care is
required.  If in-place closure is performed, some waste residues
or contaminated materials are left in place at final closure
because it is technically or economically impractical to remove
all waste materials; a 30-year (or other appropriate period)
post-closure program is required.  This program includes
monitoring and maintaining the cover; monitoring the ground water
and maintaining the ground-water monitoring system; monitoring,
maintaining, collecting, and removing liquids in the leachate
collection system; and preventing runon and runoff from damaging
cover integrity.

     Also included in 40 CFR 264.228 are rules that stipulate if
clean closure of an surface impoundment is intended without
satisfaction of the liner requirements of 40 CFR 264.221(a), then
the closure plan must include both a plan for clean closure and a
contingent plan for in-place closure, in case all contaminated
subsoils cannot be practicably removed at closure.
                                 8

-------
                            CHAPTER 2

                    PRE-DESIGN CONSIDERATIOHS
     Factors affecting impoundment design include topography,
geology, surface and subsurface hydrology, soil conditions, land
use, and climate.  Some environments should simply be avoided
without further consideration of other factors.  These
environments include populated areas, flood-prone areas, karst
terrain, unstable soils, wetlands, and aquifer recharge areas.


2.1  TOPOGRAPHY

     Topographically, an ideal surface impoundment site is one
that requires minimal physical modification to accommodate
construction.  This generally means an area of low relief.
Wherever possible, the site should be above the 100-year flood
elevation to minimize the height of dikes and the threat of
flooding and washout.

     A detailed topographic map is fundamental to site selection
and impoundment design.  The topographic survey can be conducted
by aerial or field surveying methods; however, all sites will
require some field surveying.  Although published topographic
maps may not provide sufficient detail for design, they may be
used as base maps for the topographic survey and as sources of
information on land use of the surrounding areas.


2.2  SURFACE AND SUBSURFACE HYDROLOGY

     The proximity of surface and subsurface water is directly
related to the risk of contamination from surface impoundment
releases.  High water tables, in addition to being high risk, may
interfere with surface impoundment construction if they are above
the bottom of the impoundment.  The subsurface risk is higher in
porous materials and less in dense clays.  Desirable subsurface
conditions include great depth to the uppermost aquifer and a
massive clay soil zone enclosing the impoundment.

     The direction of ground-water movement must be known, since
downstream property is at risk if a release should occur.  The
flow velocity is also important as it may be directly related to
the time available for remediation before significant
environmental damage.

-------
     Information to be obtained during the hydrogeologic
investigation includes the following:

        presence of a perched water table;
        depths to the uppermost saturated zones;
        ground-water flow direction;
        ground-water flow rates;
        vertical components of ground-water flow;
        potentiometric surface attitude;
        effects of seasonal and temporal factors  on ground-water
        flow; and
     *  aquiclude characteristics.

     Existing descriptions of the occurrence, location,  hydraulic
characteristics, and temporal and spatial behavior of shallow
aquifers and aguicludes are unlikely to be sufficient.  Gathering
the necessary hydrogeologic information may require exploratory
borings, installation of piezometers and observation wells, and
aquifer testing (e.g., pumping tests, slug tests,  and tracer
studies).  In most cases, placement of borings and sampling
efforts can be coordinated with those conducted for the geology
and subsurface soil survey.  Interpretation of the data obtained
from this investigation can be presented as potent iometric maps,
geologic cross sections, boring logs, geologic maps, field and
analytical test results, and a narrative description of the
hydrogeologic conditions.  This information is useful in the
design of a ground-water monitoring system (see Section 5.2.3).

     During the drilling of test borings, the location of
saturated layers should be recorded.  Piezometers and/or
monitoring wells should be installed in selected boreholes to
enable the determination of horizontal and vertical ground-water
flow directions and for aquifer pumping tests.  Core samples
collected from saturated zones should be analyzed in the
laboratory for particle-size distribution, porosity, and
permeability.

     The surface drainage pattern and flows and downstream water
uses also must be known and related to potential releases from a
surface impoundment.  The associated risks should be considered
and minimized in the design.

     The surface water hydrology investigation should include
determining stream flow characteristics, peak flood stages, and
water quality.  If sufficient design-related information is not
available in existing sources, a field investigation should be
conducted.  Stream-flow characteristics can be measured with
current meters and weirs or flumes.  These devices allow
measurement of stream discharge under weather conditions at the
time of the survey.  However, to estimate flow rates and volumes
for 25-, 50-, and 100-year flood events, historical flood and
precipitation data must be used.                                _

                                10

-------
     The scouring and sedimentation effects of surface drainage
nay affect the stability of a surface impoundment.  Any evidence
of recent nearby stream meandering should be noted.  If
impoundment construction involves rerouting of surface drainage
networks, the potential for scouring and/or sedimentation must be
carefully evaluated.

     The surface-water hydrology investigation should also
include collecting samples of runoff and nearby streams, during
periods of high and low flow, for water quality analysis.
Analysis of these samples will be used to provide background data
to determine the effect of potential releases to the surface
water.
2.3  GEOLOGY AND SUBSURFACE SOIL COHDITIOHS

     The geologic makeup (rock lithology and structure) of a
potential site should be determined in detail, for it can
significantly influence the design of a surface impoundment and
the environmental risk associated with it.  Seismically active
(earthquake) and unstable (sliding) areas should be avoided.
Areas of differential subsidence, which may be suggested by
intensive subsurface mining activity/ are likely to be
unattractive impoundment locations.  Porous rock could allow
rapid migration of any released liquids from an impoundment.
Larger voids, either natural or man-made, would be of similar
concern.  Limestone, for example, is often characterized by
solution-caused natural voids and channels.  On the other hand,
thick, dense, undisturbed shales may be nearly impermeable, a
desirable characteristic.

     The depth to bedrock is important.  Fractured, porous rock
near the surface may eliminate a site from consideration, while
it might be acceptable at deeper levels under dense soil
overburden.

     The characteristics of the soil underlying the site affect
the strength and long-term integrity of a surface impoundment
structure and the risk associated with impoundment operation.  As
noted earlier, great thicknesses of clay and great depths to the
upper aquifer, say 20 feet (6.1 meters) or more beneath the
impoundment in both cases, are ordinarily most desirable.  When
such is true, impoundment releases will be naturally slowed and
attenuated, thus adding to the time available for remedial
actions before significant ground-water contamination occurs.

     At the other extreme, coarse-grained, highly porous soils
and shallow water tables are generally the least desirable site
conditions for a surface impoundment.  Perhaps most often,
subsurface soil conditions will lie between the extremes of dense
clay and highly permeable coarse materials.  Layering and lensing

                                11

-------
are likely in many, if not most,  situations.   It is important to
determine the subsurface soil characteristics in three-
dimensional detail, so that the design may correct for
deficiencies and take advantage of soil attributes.  Several
types require special design treatment, if present at the surface
impoundment site.  These include expansive and sensitive clays,
acid sulfate soils, sodic soils,  poorly consolidated soils,
saturated soils, and wind-erosive soils.

     On-site or near-site soil materials suitable for dike and
liner construction are desirable.  Ordinarily, soils with a high
clay content are sought that can be compacted to a high density
and low permeability.  Measurement and interpretation of several
engineering properties generally reveal the soil's adequacy for
construction purposes.  The properties include grain size, water
content, plasticity, permeability, bearing capacity, and shear
strength.  Since the soils will be used to retain waste liquids,
chemical characteristics of the soils should also be determined.
At a minimum, the measured properties should include pH, cation
exchange capacity, and resistivity.

     A comprehensive subsurface soils and rock investigation
should assess the following:

     •  character, distribution, and thickness of the soil and
        the surficial geologic units;

     •  zones of saturation and ground-water surface elevations;

     •  pertinent physical and engineering properties of site
        soils and their horizontal and vertical variability;

     •  unstable conditions, including slopes, active or
        potentially active faults, seismic activity, and heave
        potential;

     •  ground response to excavation and the effect of stress
        increases from the facility loading; and

     •  suitability of on-site soil materials for construction of
        dikes, berms and low-permeability liners.

     Several methods of soils and rock investigation are
available.  Prior to constructing a surface  impoundment that
covers a  large area of land or is located in an area with
difficult access, geophysical methods  (e.g., seismic refraction
surveys)  can be  used to obtain general information  on  subsurface
conditions.  Test pits can be used to  observe shallow  (i.e.,  zero
to  15 feet  [zero to 4.6 meters]) soil  and ground-water conditions
and to obtain soil and rock samples for laboratory  testing.   Test
borings provide  a means for assessing  subsurface conditions at
deeper depths  (i.e., down to  100 feet  [30 meters]  or more), in-

                                12

-------
situ testing, collection of soil samples for laboratory testing,
and installation of ground-water observation wells.   Test pits
and borings should be excavated or drilled on a grid pattern
across the site with grid density depending on site  uniformity.

     Test borings made to assess foundation conditions for
structures, dikes, or fills should extend to a depth where the
anticipated increase in stress from the proposed facilities will
be less than 5 percent of the estimated overburden pressure.
Large-diameter borings are sometimes preferable to smaller ones,
especially when drilling in expansive soils or known earthslide
areas, because it is often possible to locate the slide plane by
examining the recovered soil samples.  Thin-walled tube samples
(i.e., "undisturbed" samples) should be taken for laboratory
analysis at selected depths in strata suitable for such sampling.

     Soil samples collected from test borings or test pits should
be analyzed to determine engineering properties (e.g., hydraulic
conductivity, compressibility, and shear strength) and to
classify the soils.  Shear strength analysis is necessary for
surface impoundment design and is discussed in Subsection
3.2.2.1.  Classification of soil should be done in accordance
with the Unified Soil Classification System (USCS) (ASTM D2488-
84).  Representative samples from borrow material sources should
be analyzed to determine if they meet design specifications.
Soil test methods to characterize the geotechnical properties of
site soils and potential borrow materials are listed in Tables 2
and 3, respectively.  Laboratory test methods are also described
in a number of geotechnical references (Lambe, 1951; U.S. Army
Corps of Engineers [USAGE], 1970a; Head, 1980j 1982? and 1986).

     During the pre-design phase, the soils investigation should
enable the designer to make the most economical use  of the
available on-site or nearby borrow materials.  The following
factors should be considered in selecting the borrow materials:

     *  expansion of soils of medium to high plasticity when
        placed under low applied pressure*,

     *  compaction difficulties of plastic soils with high
        natural moisture;

     *  necessity of extensive mixing and/or separation of
        naturally stratified soils to be used for borrow
        material;

     *  design specifications for soils to be used for low-
        permeability soil liners, dikes, foundations, etc.; and

     »  available quantity of usable borrow material.
                                13

-------
  TABLE 2.  GEOTECHNICAL SOIL PROPERTIES USED TO CHARACTERIZE
            SURFACE IMPOUNDMENT SITE SOILS.*
Property
  Method
Soil Type
  Visual-Manual Procedure
  Particle Size Analysis
  Atterberg Limits
  Soil Classification
Strength
  Unconfined Compressivc Strength
  Triaxial Compression

Moisture Content

Moisture-Density Relations
  Standard Proctor
  Modified Proctor

Hydraulic Conductivity (lab)
  Fixed Wall
  Flexible Wall
Hydraulic Conductivity (field)
  Large Diameter Single-Ring and
    Double-Ring Infiltrometers
  Sai-Anderson Infiltrometer
ASTM D 2488
ASTM D 422
ASTM D 4318
ASTM D 2487
Soil Survey Staff (1975)


ASTM D 2166
ASTM D 2850

ASTM D 2216
ASTM D 698
ASTM D 1557
EPA, SW-870  (1983b)
Daniel et al  (1985)
EPA, SW-846  (1982b),
Method 9100
Daniel & Trautwein  (1986)

Anderson et al  (1984)
'Minimum of 3 tests per soil type.
2.4  LAHD USE

     Hazardous waste surface impoundments are not normally
desirable features of the landscape.  They have a high potential
for interfering with other land uses.  Land use considerations
can be the most significant controlling factors in siting a. new
surface impoundment.  For this reason, most new surface
impoundments are located in heavy industry areas or at existing
waste treatment-disposal areas.  They are most often constructed
in areas where impoundments already exist.  Outside these areas,
remoteness from dense population is not only desirable but
probably mandatory.
                                14

-------
   TABLE 3.  GEOTECHNICAL SOIL PROPERTIES USES TO CHARACTERIZE
             BORROW MATERIAL SOURCES.
Property
Frequency of
Measurement
  Method
Soil Type
  Visual-Manual Procedure
  Particle Size Analysis
  Atterberg Limits
  Soil Classification
Strength
  Unconfined Compressive
    Strength
  Triaxial Compression

Moisture Content

Moisture-Density Relations
  Standard Proctor
  Modified Proctor

Hydraulic Conductivity
Laboratory-Undisturbed Cores
  Fixed Wall

  Flexible Wall
each 1000 yd3
   (836 m3)
each 5000 yd3
   (4180 m1)

each 1000 yd3
   (836 m3)

each 5000 yd3
   (4180 m3)
each 5000 yd3
   (4180 m1)
Volume Change
  Consolidometer
  COLE
Susceptibility to
Frost Damage
  Classification
each 5000 yd3
   (4180 m1)
ASTM D 2488
ASTM D 422
ASTM D 4318
ASTM D 2487
Soil Survey  Staff
 (1975)
ASTM D 2166
ASTM D 2850

ASTM 2216
ASTM D 698
ASTM D 1557
EPA, SW-870
 (1983b)
 Daniel et al
 (1985)
 EPA,  SW-846
 (1982b)
 Method 9100
Bolz, 1965
 Grossman et al
 (1968)
                 Chamberlain (1981)
     Future  land  use may be a consideration at an existing or
proposed hazardous waste surface impoundment.  Uses will probably
be restricted,  depending upon the degree of contaminant removal
                                15

-------
and the character of the waste remaining at the time of closure.
Restrictions are likely to parallel post-closure monitoring
requirements, with intensively monitored sites being the most
restricted.  Future use considerations should be integrated with
planned uses of adjacent properties.


2.5  CLIMATE

     Climate (e.g., excessive rainfall or low evaporation rate)
may eliminate a particular type of surface impoundment as a
waste-handling option.  Otherwise, climate will seldom directly
influence the selection of a surface impoundment site.

     Climate plays a direct role in sizing a surface impoundment.
The unit must be large enough to contain the planned waste volume
plus some amount of precipitation.  Other climatic factors may
affect the operation of the impoundment.  For example, air
circulation  (wind), temperature, and radiation affect waste
volatilization rates and dispersion of emissions.  These factors
also affect the evaporation rate from the unit and thus the
unit's size and need for periodic pumpout and treatment.

     The most extensive, continuous, and reliable climatic data
are available from National Weather Service reporting stations.
If a station is not close enough to the planned impoundment,
there may be other data sources, as shown in Table 4.

     Factors directly related to climate that may influence site
selection are discussed in the following subsections.


2.S.I  Flooding

     The potential for flash floods and large-scale floods should
be considered during the design of a surface impoundment.  A
flash flood  results from localized, very heavy rainfall in a
short time acting on a relatively small watershed.  In this case,
it would be  the watershed containing the surface impoundment.
Flash flooding occurs very quickly after, or perhaps even during,
the precipitation event, and the flow may be high-velocity.  It
can occur in watersheds not subject to large-scale, longer-term
flooding.  Predicting peak flow from a flash flood is necessary
for the design of runon control and diversion structures.
Hydrologic analysis, using methods such as those developed by  the
Soil Conservation Service  (SCS)  (Kent, 1973), can be used to
predict the  maximum discharge and flood levels.

     Large-scale flooding is the result of precipitation, often
combined with snowmelt, acting on a regional watershed.  Lag
times are greater and flood duration longer than for  flash
floods.  The events causing large-scale flooding usually occur

                                16

-------
TABLE 4.  SOURCES OF CLIMATIC DATA USED IN SURFACE IMPOUNDMENT
          DESIGN AND ANALYSIS.
Parameter
           Type
  Reference
Rainfall
Atlas of U.S. - Duration to
1 day and return periods to
100 years

Atlas of Western U.S.  - Dura-
tions to 1 day and return
periods to 100 years

Monthly totals and normals
Hershfield (1961)
                                                Miller et al
                                                (1974)
                                                NOAA (no date)
Evaporation
Atlas of U.S. - Average
monthly Class A Pan Evaporation

Class A pan evaporation -
Monthly totals

Atlas of U.S. - Class A pan
and pond evaporation

Atlas of Western U.S. - Class
A and pond evaporation

Mean annual Class A pan-to-
lake coefficients

Monthly mean Class A pan
coefficients

Monthly totals
Brown et al
(1983)

NOAA  (no date)
                                                Kohler et al
                                                (1959)

                                                Nordenson (1962)
                                                NOAA  (1968)
                                                Kohler  (1954)

                                                Jenson  (1973)
                                                State
                                                Climatologist
                                                or Agricultural
                                                Experiment
                                                Stations
 ============:=:==::
over a  sizable  area  that may  include the locale of the  surface
impoundment,  but  they  may  take place completely outside it.

     While  flash  flooding  may occur almost anywhere near a
stream,  large-scale  flooding  ordinarily occurs in the valleys  of
larger,  slower-moving  streams.  Units  in areas prone to large-
                                17

-------
scale floods must be designed to prevent washout from at least a
100-year flood.

     Several sources of flood data are available (see Table 4).
Because watershed characteristics may be altered by natural
forces or by man's activities, flood maps may be inaccurate or
outdated, and direct field evidence.should be acquired to
supplement or revise data from published sources.

     Siting an impoundment within the 100-year flood plain or in
an area prone to flash floods is generally not advisable and
makes the acquisition of a permit more difficult.  However, it
does not preclude the construction of a surface impoundment that
is designed accordingly.  The erosive action of flowing flood
waters or wave action, in addition to the potential for dike
overtopping (externally and into the impoundment},  must be
addressed in the surface impoundment design in these situations,
using estimates of flood depth and flow rate.  Additionally, dike
stability analyses should consider rapid drawdown effects, which
occur when flood waters on outer dike slopes quickly recede.
Larger safety factors are likely to be required.


2.5.2  Precipitationv«. Evaporation -

     Precipitation adds an unwanted liquid volume to a surface
impoundment.  In effect, it reduces the impoundment's waste-
holding capacity.  On the other hand, evaporation has the
opposite effect, and is generally desirable.  Precipitation and
evaporation rates rank with the rate of waste input in design
consideration.  Together, they control the design capacity of the
surface impoundment, and the design of the berms, discharge
pumps, inflow/outflow structures, secondary containment
structures, and surface water controls.

     In humid regions, and in cases where the impoundment's
purpose is to collect contaminated storm runoff, provisions for
the storage and periodic discharge must be included in the
freeboard design and in the design of inflow/outflow structures.
Long-life, non-discharging surface impoundments  (evaporation
ponds), are feasible in the drier western U.S. where the
evaporation rate exceeds precipitation  (see Figure 1).  Even
there they are feasible only when storage capacity is adequate to
handle extremes of precipitation and seasonal periods of excess
precipitation.  Short-life (say, less than five years), non-
discharging storage impoundments may be designed in excess
precipitation areas, with input being terminated when a pre-
determined level is reached.  Further discussions on surface
impoundment water budgets and sizing considerations are included
in Subsection 3.1.
                                18

-------
     Figure 1.  Area (shaded) of net evaporation in the U.S.
                (derived from HOAA data).


2.5.3  Soil Freezing and Thawing

     Cycles of freezing and thawing, particularly in the northern
states, can change the chemical and physical properties of soil
from the surface down to the maximum freezing depth.  Critical
properties, such as permeability and shear strength, of
constructed soil dikes and liner components may be adversely •
affected.  Therefore, in evaluating some sites, particularly in
northern areas, it may be necessary to add a protective cover
over the liner or dike to the depth of maximum frost penetration.
                                                             i
     Figure 2 shows the maximum anticipated depth of freezing1for
the contiguous United States.  Local sources should be sought1to
determine the soil freezing depth for a specific site.

    The freeze-thaw phenomena, as they affect soil-based
structures, are summarized in an EPA publication on clay liner
construction  (USEPA, 1988b).

2.6  AIR QUALITY

     Background air guality should be tested before surface
impoundment construction to avoid potential distortion by later
site activities.  Background air measurements are particularly
important in  industrial areas where other emission sources are

                                19

-------
 Depth contours are in feet.
 (1 ft = 30.5 cm)
        Figure 2.  Maximum anticipated depths of freezing.
                   (Spangler  and Handy,  1982)
common.  Testing should be  conducted under a variety of weather
conditions.  Without background data, it may be difficult or
impossible to confine  or  defend against later accusations of
emissions violations.  Several air sampling techniques have been
described by the USIPA (1985f).
                                20

-------
                            CHAPTER 3

                              DESIGN


     Fundamental impoundment design decisions include the basic
configuration and the selection, justification, and arrangement
of the structural components and ancillary facilities.  Certain
components of the configuration (e.g., a double-liner system) are
now required by minimum technology regulations.


3.1  SELECTION OF BASIC COHFIGURATIOH

     Among the first design decisions are the type of impoundment
based on specific use; the number, size, and position of
impoundment units; the impoundment shape and areal dimensions;
and the selection of liner type.


3.1.1  Impoundment Type

     The type of surface impoundment required depends on waste
composition, waste-generation rate, and the purpose of the
impoundment.  A surface impoundment can be classified as one of
three generic impoundment types:   (1) treatment; (2) surge or
equalization  (i.e., storage); and  (3) non-discharging
(evaporation or disposal).  Figure 3 depicts the three types.
The greatest number in use are of the treatment type.

     Waste inputs and treated waste discharges from treatment
impoundments may be steady, fluctuating, or intermittent.  Except
for some surge or equalization impoundments that are intended to
collect runoff, the only external water input is direct
precipitation on the impoundment surface and interior dike
slopes.  Non-discharging surface impoundments generally rely
strictly on natural evaporation to maintain liquid level.


3.1.2  Wn«tK*r. Size, and Position

     More than one impoundment may be required where several
incompatible  liquid wastes are to  be  stored.  Multiple
impoundments  may also be desirable for  single or compatible
wastes in some situations.  Unger  et  al.  (1985) compared a large
single surface impoundment to multiple  small impoundments, as
depicted in Figure 4, with respect to cost and the implications

                                21

-------
         waste inflow
               -^=
precipitation
    O
   o
evaporation
discharge
   (a)
        waste inflow
   
-------
of dike and liner failure.  Bach was optimized for the highest
"fill efficiency ratio," which is defined as the most cost-
effective dimensions for a given volume.   The side slopes were
3H:1V in each case.  It was found that four small impoundments
would cost nearly twice as much as a single large impoundment.
Other factors were also considered.  For  example, there is a risk
of greater leakage due to more geomembrane liner seam length and
greater bottom surface area in multiple impoundments.  On the
other hand, higher head, and higher hydraulic gradients in clay
liners, make can increase the risk in a single large impoundment.
The risk of a dike failure occurrence is  greater in multiple
impoundments due to more dike length, but greater environmental
damage would ensue from dike failure of a single large
impoundment.  All things considered, it appears that a single
large impoundment and multiple small impoundments, constructed to
serve the same purpose, are approximately equal in environmental
risk.

     According to the USEPA (1983b) the most common and most
economical shape for a surface impoundment is rectangular with
straight-sloped sides.  Other shapes increase the cost of
grading, dike construction, and liner installation.  The
rectangular shape was confirmed in the later study by Unger et al
{1985).  That study did not consider the  cost differences between
impoundments excavated entirely below grade, those diked entirely
above grade, and combinations of excavation and above-grade
dikes.  Unger's study assumed the combination type.

     Orientation of a surface impoundment with respect to compass
direction is probably relatively unimportant as a design
consideration.  However, if the primary geomembrane liner is
exposed (not covered with a protective soil layer), weathering
effects are likely to be the most severe  on south-facing slopes
(those slopes exposed to direct sunlight  for the longest periods)
(USEPA, 1985j).  The designer may want to consider this potential
in orienting a rectangular impoundment, if he has the freedom to
do so.  Another possible consideration in larger impoundments is
the fetch, or the orientation of the long dimension compared to
higher-velocity prevailing winds.

     The position of the surface impoundment with respect to
natural grade can vary from one excavated below-grade, to one
entirely above grade and surrounded by containment dikes, to a
combination of the two.  The most common is the combination type,
probably because it is the most economical to construct.  For
this type, the excavated material can often be used as dike
material.  Below-grade surface impoundments have the
environmental advantage of being less prone to catastrophic
failure, since no dike is involved.  Above-grade  impoundments, on
the other  hand, could lose all their waste to the surrounding
area if the dike were breached.  Table 5 is a cost comparison for
a surface  impoundment with respect to the three positions.

                                23

-------
  TABLE S.   COST COMPARISONS OF DIFFEREITT SURFACE IMPOUNDMENT
            POSITIONS WITH RESPECT TO GRADE
            (GEOMEMBRAHE/COMPOSITE LINER)
    Operation
Above Grade
  Cost

Below Grade
 Liner
Leak Detection
  Main
  Lateral
  Riser
Pump
Sump
Riprap
Level Controller

TOTAL
      421
    1,234
        0
    8,955
    4,964
    8,186
    5,000

 $321,349
      372
      728
       82
    8,955
    4,964
        0
    5,000

 $307,624
Combination
Geotechnical Invest.
Clear and Grub
Excavation
Foundation Preparation
Bern (Construction
Soil Liner
Sand Drain Layer
Geosynthetic Drain Layer
Geotextile Layer
Primary Geomembrane Liner
Geotextile Layer
Geomembrane in Composite
$13,805
2,325
0
1,670
41,095
126,982
26,928
27,182
6,313
18,151
6,313
21,825
$13,805
2,325
40,562
3,463
0
123,657
26,170
28,478
5,869
16,873
5,869
20,452
$13,805
2,325
20,602
2,708
8,556
125,734
26,639
27,432
6,175
17,754
6,175
21,397
     408.
   1,196
     425
   8,955
   4,964
   3,314
   5,000

$303,564
Note:  All impoundments are designed to hold 9,940 m3  (2.63
       million gallons) of waste liquid (Unger et al,  1985).
       Construction materials assumed to be the same in all 3
       cases.  Actual site-specific costs of construction may be
       grossly different.
     Standard reservoir tables (Kays, 1977) can be used to
determine volume and liner areas for various surface impoundment
configurations.  Usually, the surface area and storage capacity
requirements are known beforehand or are determined during pre-
design.  The surface area may be limited by available land area.
If so, it can usually be compensated, within reason, by
increasing the impoundment depth*
                                24

-------
     Surface area  is  one  of the more important factors in
impoundments used  for evaporation.   The rate of evaporation must
equal or exceed the rate  of precipitation plus the liquid
(aqueous waste) inflow rate.  If,  for example, evaporation is 40
inches (102 cm) per year  and precipitation is 30 inches (76 cm),
then 10 inches  (25 cm)  per year is  available for waste liquid
input.  Thus, the  larger  the surface area, the larger the amount
of waste liquid that  can  be accommodated.

     The potential for slope damage by wave action increases with
impoundment size and  is directly related to both area and depth.


3.1.4  Impoundment Depth

     The surface impoundment depth is a function of the waste
volume to be handled, its eventual  disposition, and the increased
depth necessary to contain foreseeable and unforeseen events.
Assuring adequate  depth by accurately predicting liquid levels
for all circumstances is  a very important design operation.

     The liquid level in  a surface impoundment will change due to
waste inflow and outflow, storm surges, precipitation, runon (if
applicable), wind  speed,  and dike slope.  Figure 5 is a cross-
sectional representation  of a surface impoundment, showing the
components which cumulatively make up its design depth.  The
bottom liquid level shown on the figure is the normal operating
level which takes  into account only waste inflows and discharges.
The next higher liquid level on the figure is the maximum
operating level.   The maximum operating level is the resultant
level caused by the addition of water from a major climatic event
to the normal operating liquid level.  The major climatic event
may be, for example,  the  wettest month in 25 years, or a 100-year
24-hour rainstorm. Additional impoundment depth allowances
(freeboard) are included  for wind set-up, wave run-up, and
finally a safety factor.
                         spillway level ~
                                     "*S^     f"           J__   «JM*
                       max. wave run-up level —^	   '— *—•	•   *   waves

                          max. wind set-up level
                           maximum operating level —«*N
                                               . maximum precipitation event
                                normal operating level —"^
                                                 \
                                                          liquid volume
    Figure 5.   Components making up impoundment design depth.

                                 25

-------
3.1.4.1  Horaal Operating Level --

     For flow*through impoundments, normal operating liquid level
can be determined from waste inflow and outflow rates, liquid
surface area, and detention time of wastes in the impoundment,
using the following formula:
where:

     d - normal liquid level or depth in ft. (m)
     t « average detention time (hr)
     Qt « maximum rate of inflow during t  in ftVhr (mVhr)
     Qo » average rate of outflow  during t in ftVhr (mVhr)
     A » liquid surface area in ft2  (mj)

     For non-flowthrough  (storage) impoundments,  where inflow
exceeds evaporation over the long term, the liquid level will
rise.  Reaching the design level would call for an end to
operations, or for cleanout and resumption of operations.  In
this case, the above equation would be solved for t,  for a given
design depth (d) and surface area (A).

     The normal operating level in a surface impoundment used for
evaporation may be considered, in some cases, to be the highest
liquid level reached during a normal year, for example during a
"rainy season," when precipitation exceeds evaporation.  If the
waste inflow is variable, the highest expected waste inflow would
be added to the higher precipitation.


3.1.4.2  MaxiBUB Operating Level without External Runoff Input —

     Designing the surface impoundment's maximum operating level
is typically done using one of three methods:  hydrologic models;
water budget approach; or design storm approach.

     The additional storage capacity calculated as being
necessary to contain climatic extremes will vary depending upon
the method used.  Geographic areas having high temporal vari-
ability in rainfall amounts (e.g., the southwestern U.S.),
generally have larger storm surcharges that are calculated using
a recorded rare return-period storm rather than using a water
budget or hydrologic simulation approach.  The choice of approach
for determining the maximum operating level will depend on the
availability of climatic  data and hydrologic models, and the type
of impoundment operation, waste characteristics, and the risks
associated with an uncontrolled release of hazardous waste from
the surface impoundment.

                                26

-------
     Kat^r Budo»t Approach — The water budget can determine the
feasibility of using a surface impoundment at the given locale
and can help determine the needed storage volume for non-
discharging disposal or evaporation units.  Storage requirements
are a function of precipitation and evaporation and waste inputs.
A basic assumption in a water budget for most non-discharging
surface impoundments is that there is no net change in the volume
of liquid in the impoundment over the long term.

     Regardless of the intended impoundment use, the water budget
for determining the maximum operating level (and storage
capacity) can be derived from the following expression:

               S = P + W-E-D-L

        where:

          S - change in storage
          P « precipitation
          W « waste input
          E = evaporation
          D = discharge
          L = leakage (negligible for a functional liner)

     The water budget can be computed using the climatic record,
watershed properties (for units that collect runoff), waste input
rates, and discharge rate, if applicable.  For the purpose of
these calculations, it is assumed that adequate volume will be
available to contain all events (i.e., no overtopping of passive
level controls, such as spillways).

     The approach taken is a frequency analysis to determine the
amount of excess precipitation (i.e., rainfall minus evaporation)
that can accumulate in the impoundment in a given period, at
least equal to the active life of the impoundment.  Long-term
records  (preferably more than 20 years, but at least equal to the
impoundment life) of precipitation and lake evaporation are
required for these water budget calculations.  Measurements of
pan evaporation can also be used, although pan-to-lake
coefficients will have to be applied  (Saxton and McGuinness,
1982).  The data should be obtained from National Oceanic and
Atmospheric Administration (NOAA) weather stations or other local
weather stations located as close to the impoundment site as
possible, and from hydrologically similar areas.  Long-term
precipitation records are available for weather stations across
the U.S. (NOAA, 1983).  While stations collecting evaporation
data are fewer in number, evaporation rates are reasonably
similar over regional areas (Saxton and McGuinness, 1982);
therefore, evaporation data collected at the nearest weather
station recording that parameter can often be used.
                                27

-------
     Usually an empirical probability distribution is used with
observations of precipitation and lake evaporation data to
determine excess precipitation amounts.  Monthly totals are
commonly used to simplify data requirements, although data
compiled for shorter increments (e.g., the design retention time)
may lead to a more accurate estimate of storage requirements.

     The occurrence frequency of excess precipitation amounts is
determined by ranking the observed data, computing a plotting
position, and plotting the excess precipitation amounts and
positions on probability paper.  A widely used, consistent
plotting position formula is the Weibull Relationship (Haan,
1977).  Haan (1977) and Osborne et al (1972) present information
on frequency analysis and probability plotting.  The Weibull
Relationship is expressed as follows:


        P(y)
               N + 1

     where:  y = the excess precipitation value
             P(y) « the plotting position
             m = the rank of the excess precipitation value
             N « the number of observations

     The return period, or recurrence interval, is given by the
inverse of P(y).  There is no general agreement as to which
statistical distribution should be used for this type of
frequency analysis.  The National Weather Service (NWS) used a
modified version of the Gumbel extreme value distribution to
develop a rainfall frequency atlas (Miller et al., 1974) for the
U.S.  Pruitt and Snyder (1984) recommend using the normal
distribution for determining surface impoundment storage needs
for reclaimed wastewater irrigation projects.

     Regardless of the distribution chosen, if a data set plots
as a straight line on the probability paper, the data are said to
be distributed as the distribution corresponding to the
probability paper.  Because it is rare for a set of data to plot
exactly on a line, a decision must be made as to whether the
deviations from the line are random or true deviations,
indicating that the data do not follow the given probability
distribution.  Often, several types of frequency plots must be
evaluated to select the appropriate distribution.  Figure 6 is an
example probability plot for excess precipitation which follows a
normal distribution.

     In this example, excess precipitation was greater than six
inches (15 cm) in 10 percent of the months recorded.  This
additional six-inch  (15-cm), one-month storage requirement
corresponds with a 10-year return period.
                                28

-------
       8

     £ c
     •—•• o
     c
     o  .
     i2
     S  0
     1-2

       *
     D.-6
     o
      -10  U-
          I  I
      -12
               1.05
 return period (yrs.)
1.4   2     4    10  25  100
                                   MONTHLY DATA
                                                90%
         12   5102030   50   70 80
                           percent less than
                90  95  98 99
99,9
         Figure 6.  Example of frequency distribution of
                    monthly cliaatic data.
    Once the probability plot of excess precipitation  is
constructed, the next step in establishing the maximum operating
level is selecting the maximum climatic event (monthly
precipitation minus evaporation) that can be tolerated,  and  the
associated precipitation amount.  Because surface impoundments
may contain hazardous waste, the probability of occurrence of a
climatic event that results in exceedance of the maximum
operating level  (and possible release) must be very small.   The
maximum probability recommended is a 1 in 25, or 4%, probability
that the maximum operating level will be exceeded in any year.
This probability level represents a return period of 25  years for
the maximum monthly excess precipitation calculated using the
water balance approach.  However, this does not mean that the
maximum operating level will not be exceeded during a  25-year
design life.

    Figure  7 shows the design return period required  so  that the
maximum operating level will not be exceeded during the  expected
facility life for various degrees of confidence.  As shown on the
graph, to be 90% confident that the maximum operating  level  will
not be exceeded  in a surface impoundment with a  10-year  design
life, the impoundment must be designed using a 100-year  climatic
event,
                                29

-------
                1(
                 600

                 400

                 200


               ? 100

               T  50
               I
               !  25

               |

               |  10

                   .
                   5
               »
x

                       2     5   10    25   50  100
                          design life - "L (yrs.)
  Figure 7.  Percent  confidence that design return period will
             not be exceeded during design life.
    Design Stora Approach — The simplest approach for
determining the maximum operating level is to make it equivalent
to the additional storage capacity, over the normal operating
level, that is needed to contain precipitation from a single
design storm.  The minimum acceptable probability of a discharge
for hazardous waste surface impoundments is the 100-year, 24-hour
storm.  This value represents a 1-percent probability that the
given amount will be equaled or exceeded in any given year.  It
also says that if 1,000 surface impoundments are in operation
throughout the U.S., using the 100-year, 24-hour storm would
result in about 10 units (1% x 1,000) equaling or exceeding this
amount each year.

    Figure 7 should be consulted to evaluate the risk or percent
chance of a precipitation event in excess of the design  storm
during the facility design life.  For the 100-year return period
and 10-year design life, there is a about a 10% chance that the
maximum operating level will be exceeded.

    Adding to the confidence that the maximum operating  level
will not be exceeded is the fact that the storm surge amount is
                                30

-------
added to the normal operating liquid level, which already
accounts for extremes in waste inputs.  The likelihood that the
extreme events would coincide is slight.  The further inclusion
of freeboard that accounts for wind set-up, wave run-up, and a
safety factor adds still more assurance that wastes can be safely
contained under any conditions.


3.1.4.3 Maximum Operating Level with External Watershed —

     Computing the maximum operating level of a surface
impoundment having an external watershed (e.g., an impoundment
used for detention of runoff) requires adding watershed runoff to
the direct precipitation on the impoundment surface.  The volume
contributed by runoff will be directly related to the watershed
area and is likely to be much larger than the volume contributed
by rainfall directly on the impoundment surface.

     Available hydrologic models were not developed for surface
impoundment design, but can be modified to meet the need.  Most
rainfall models are designed to predict runoff and are suited for
estimating storm surges to surface impoundments that act as
catchment basins.  To date, none of the many available models has
been widely used.  One of the better suited models is STORM
(Storage, Treatment, Overflow, Runoff Model), developed for the
USAGE Hydrologic Engineering Center (EEC).  STORM is a stochastic
rainfall model used to predict thunderstorm activity and to
generate a synthetic rainfall record for watersheds (Corotis,
1976).  Stochastic rainfall models such as STORM are particularly
useful in geographic areas where the weather record may not be
sufficiently long for use in impoundment design.

     The Soil Conservation Service developed a method for
estimating volume and rate of runoff from U.S. agricultural
watersheds  (Kent, 1973).  A more detailed description of the.JSCS
method has been provided by McCuen  (1982).  The "Rational
Formula" is another widely used method  for predicting peak runoff
from a small watershed for a given design storm intensity:   !

        Q - CIA                                              |
     where
        Q = runoff volume  (cubic feet per second)
        C = runoff coefficient  (dimensionless); the percentage
            of rainfall that is surface runoff
        I « average rainfall intensity  (inches per hour)
        A = watershed area  (acres)

     Introductions to the SCS and Rational Formula methods and
others have been provided by Whipple et al  (1983) and Osborne et
al  (1972).  The hydrologic models discussed earlier will commonly
estimate runoff volumes directly or provide synthetic rainfall
records to be used in a runoff model.

                                31

-------
3.1.4.4 Freeboard Determination —

    Freeboard is the distance from the water line on a structure
to the top of the structure.  In the case of a surface
impoundment it is the distance between the maximum operating
level and the liquid level (e.g., spillway level)  which would
result in the release of stored liquid.  Determining necessary
freeboard in impoundment design requires considering wind speed,
fetch, maximum liquid depth, and the slope and roughness of the
embankment.  These factors influence the development of waves,
wave run-up, and wind set-up.  Wave run-up and wind set-up, plus
a safety factor, are accounted for in the calculation of
freeboard.

    Wind speeds used in freeboard calculations should be based
on local historical data for maximum sustained winds.  In the
absence of such data, a value of 75 mph (120 km/hr) is suggested
for areas that are not subject to hurricanes, and a value of 100
mph (161 km/hr) for areas that are hurricane-prone.  In this
document, a "hurricane-prone area" is any area within 50 statute
miles (80 km) of a coast subject to hurricanes (USEPA, 1986b).

    Fetch is the distance that the wind blows over the water  (or
waste liquid) surface.  For a surface impoundment, it will vary
with the wind direction vs. the orientation and dimensions of the
impoundment.  In a rectangular impoundment, the maximum fetch
will be along the diagonal.

    Depth of the impoundment will ordinarily be uniform, or
nearly so, over most of the liquid area, with relatively small
slope areas along the edges, if the total area is more than, say,
an acre.  For calculation of wave height and set-up, the depth
can be considered essentially uniform.  For small impoundments,
where the slopes make up most of the area, fetch, and thus wave
action and runup, are likely to be insignificant.

    Techniques  for predicting wave action in shallow water are
provided by USAGE (1984).  Maximum wave height and wave length
are directly related to fetch and water depth.  Using a fetch
(distance that the wind blows across the impoundment) of 300 feet
(91 m), a liquid depth of 10 feet (3 m) and a wind speed of 75
mph (120 km/hr), the maximum wave height would be 0.9 feet (0.3
m).  Waves would not be expected to break under these conditions,
or in any normally designed surface impoundment.

    Wind  set-up is the  rise  in  liquid  level  on the downwind  side
of a surface impoundment caused by the wind pushing  the liquid  to
that side  (Figure 5).  A corresponding drop in level occurs on
the upwind side.  In the past, wind set-up has largely been
ignored in determining  freeboard requirements for  surface
impoundments.  Two methods considered acceptable for set-up
calculation have been presented by Keulegan  (1951) and Van Dorn

                                32

-------
(1953),  Given the same wind conditions and impoundment
dimensions shown above, the set-up would be 1.11 feet (0.34 m).

     Wave run-up is the vertical height to which approaching
waves force water above the still water level on a sloped
embankment (see Figure 5).  Wave height, embankment slope and
slope roughness affect wave run-up height.  Typically, the
highest run-up occurs on low smooth slopes.  Host investigations
conducted to accurately define the parameters that control run-up
have been directed toward run-up on beaches, seawalls, and
breakwaters (USAGE, 1984| Machemehl and ierbich, 1970; Saville,
1956 j Toyoshima et al, 1966).  Roughness factors will vary from a
value of 1.0 for smooth synthetic liners to 0.45 for coarse
surfaces (e.g., riprap) (USAGE, 1984).  Curves that relate slope
roughness to other freeboard parameters are convenient for
estimating roughness effects and can be found in Saville (1956)
and USEPA (1986b).  Saville (1956) looked at run-up on smooth
surfaces, which applies to defining run-up on smooth synthetic
liners used in many surface impoundment designs.  In surface
impoundments, the highest run-up can be expected on slopes of 3sl
to 5:1 lined with impermeable synthetic membranes (Saville,
1956).  Therefore, the designer should consider, as protective
cover, rough-surfaced materials such as riprap, to reduce run-up,
or he should consider, generally as a lesser option, designing
steeper siopes.

     More detail on freeboard calculation can be found in
references by Herbich  (1986) and those cited above.


3.2  STRUCTURAL COMPONENT DESIGN

     The following subsections discuss the structural
considerations necessary in the design of a surface impoundment.
References are presented that may be consulted for detailed
information on specific mechanics of the design process.  The
design components discussed include the following:

        dikes and foundations
        liner systems
        liquid level controls
        inflow and outflow
        protective coverings
        secondary containment
        leak detection systems
        surface water  management
        controls for emissions of volatile organic compounds
         (VOC)
        construction quality assurance
                                33

-------
3.2.1  Foundat
    The foundation underlying a  surface  impoundment is the
native soil.  The soil mist be capable of supporting the added
load of an overlying dike.  It also must be capable of supporting
the load imposed by the impounded waste.  Although it is not
likely, that load may exceed the load removed by excavation.  The
potential for soil compression and the soil's bearing capacity
are the properties of principal concern.


3.2.1.1  Settlement —

    Foundation compressibility is an important consideration  if
sections of a dike will be placed on soils that are soft or
loose.  Such soils will consolidate (compress) under the
additional load imposed by the dike.  If the dike height,
foundation thickness, and/or foundation compressibility vary,
differential settlement and perhaps cracking of the dike and
liner could result as shown in Figure 8.
                                                   dike settlement
                                                   and cracking
                                         compressible zone
                                         of foundation soil
   Figure 8.  Example of the effect of differential  foundation
              compressibility  on  a surface  impoundment  dike.


    The compressibility of fine-grained  soils can best  be
determined in the  laboratory by consolidation tests  on
undisturbed tube or block  samples obtained  during field
exploration.  Methods for  predicting  consolidation settlement are
presented in several introductory geotechnical texts (Spangler
and Bandy, 1982; Terzaghi  and  Peck,  1967; Bowles, 1977).
Settlement of granular  soils can  be determined by a  method
devised by Schmertmann  (Schmertmann,  1970;  Winterkorn and Fang,
1975).
                                34

-------
    If the foundation soils are granular, loose, uniformly
graded, and saturated, and could be subjected to earthquake
loading, liquefaction (loss of foundation support)  could result.
Liquefaction-potential assessment is also discussed in geo-
technical texts {Winterkorn and Fang, 1975).  Low-density
granular soils are best identified during field exploration,
using standard penetration tests or other in-situ testing and/or
field density determinations made in test pits.


3.2.1.2  Bearing Capacity --

    The foundation soil that will underlie a dike or other
massive structure associated with the surface impoundment should
be evaluated for its bearing capacity.  The bearing capacity of a
foundation is the maximum loading to which the foundation soil
may be subjected without permitting shear displacements
detrimental to the function of the structure (Spangler and Handy,
1982).  Figure 9 shows the forces and displacements that occur
when the bearing capacity is exceeded.  The design bearing
capacity should have a safety factor of 2.5 to 3.0.  Methods of
determining a soil's bearing capacity are presented in several
introductory geotechnical texts (Spangler and Handy, 1982;
Terzaghi and Peck, 1967? Bowles, 1977).  Input variables include
the dike and foundation geometry, and the weight and shear
strength of the soil materials.
                                LOAD
     Figure 9.  Forces and displacements in bearing capacity
                analysis  (adapted from Scott, 1980)
3.2.2  Pike Design

    Surface  impoundments may range  from completely  above  grade
to completely  below grade,  but most will be a combination with an
excavated  portion below grade and a diked portion above grade.
The excavated  materials may be used to  build the dike.  Figure 10
is a cross-section  of one side of the combination type showing
the various dike and liner interfaces.

                                35

-------
                  geomembrane liners
                  with protective
                  geotextiles
     protective soil layer
     leak collection
     layer
                               cut-slope
                     low-permeability
                     soil layer
  *
          Figure 10.   Surface impoundment dike and liner
                     interfaces and layers.


    A surface impoundment dike and the  associated double liner
comprise a relatively complex structure.  Stability  of  the native
foundation soil and internal stability  of the dike proper must be
considered, along with the stability of  the double liner
components on the inside slope of the dike and excavation.  Each
of the liner slope interfaces may be a  plane of weakness.

    Dike  and foundation stability must  be evaluated  for the
construction and operating conditions,  using expected in situ
engineering properties of the foundation and berm materials,
pertinent geologic information (see Section 3.3),  and planned
dike slope and  height.  Side slopes of  2H:1V or flatter are
usually considered adequate  for maintaining dike  stability.
However, the slope limits for the soil  liner component may be
controlling for the dike.  USEPA's Minimum Technology Guidance
(USEPA, 1987a)  for surface impoundments recommends a
demonstration to show that the low-permeability soil liner on the
side walls can  be compacted  effectively at the maximum slope used
in the design (USEPA, 1987a).  Soil liner slopes  steeper than
3H:1V are not recommended due to the difficulty in constructing
them to meet current USEPA performance  standards.

    Dike  stability is determined primarily by the ability of the
dike and  foundation soils to resist shear stress.  Shear stresses
result from externally applied artificial loads  (e.g., the
impoundment: contents), the internal body forces caused by the
weight of the soil and dike  slopes, and external  forces  (e.g.,
earthquakes).   When the ability  of the  soils  to resist shear
stresses  is exceeded, the result can be (1) slope failure,
characterized by a failure plane or surface along which a portion

                                36

-------
of the embankment slides; (2)  excessive settlement; or (3) a
bearing capacity failure.  The following sections describe design
against these failure modes.


3.2.2.1  Shear Strength —

     The principal parameter in analyzing dike and foundation
stability is soil shear strength (or shearing resistance).  It is
generally determined by field or laboratory tests.  To obtain
meaningful shear strength results from a laboratory test program,
structurally undisturbed samples, representative of the in «* situ
soils or compacted soils proposed for the project, must be
obtained for testing.  For native, cohesive foundation soils,
samples of "undisturbed" materials are typically used.  (Soil is
always disturbed to some extent when the sampling operation is
performed.  "Undisturbed" samples are those taken in a way to try
to retain as many of, or as near to, the in-place physical
properties as possible.)  To determine the shear strength of
soils to be used as compacted fills (dikes and liners), the soils
must simulate as-constructed field conditions, including method
of compaction, dry unit weight, and water content.  These samples
may be prepared in the laboratory, or, preferably, they are
"undisturbed" samples taken from a field test fill prepared under
construction conditions.

     Test methods for determining shear strength in the
laboratory include triaxial tests, unconfined compression tests,
direct shear tests, and laboratory vane shear tests.  Procedures
and descriptions for such testing, and interpretation of the test
results, are provided by several sources (Lambe, 1951; USAGE,
1970; Head, 1980; 1982; and 1986).

     Methods for determining in-situ shear strength include
standard penetration, vane shear, pressure meter, cone
penetrometer, and dilatometer.  In-situ testing is described in
several sources  (ASCE, 1986; FHWA, 1980).

     The shearing resistance of cohesive soils differs from that
of cohesionless  soils due to the following factors:

     •  particle size — cohesive soils are finer
     •  particle shape — cohesive soils are platy; cohesionless
        are rounded to angular
     *  permeability — cohesive soils are less permeable and
        drain more slowly
     •  internal friction angle -- lower in cohesive  soils
     *  plasticity — cohesive soils more plastic

     The Mohr-Coulomb failure envelopes for clays  and sands are
shown in Figure  11 for comparison.  The value C indicated on the
shear stress axis, also referred to as the "cohesion  intercept,"

                                37

-------
        CD
        CD

        1  •
                  Mohr-Coulomb
                  envelope for days
                                     Mohr-Coulomb
                                     envelope for sands
                         Normal Stress, a
Figure 11.  Mohr-Coulaab failure envelopes for clays and sands.


is due to the attractive forces between the plate-shaped
particles forming cohesive soils such as clays.

    Dikes are ordinarily constructed of compacted soil.  A
typical soil compaction curve showing the dry-unit-weight  to
water-content relationship for a cohesive soil is given in Figure
12a.  The variation of shear strength with water content is shown
in Figure 12b.  Although a cohesive soil can be compacted  at the
same dry weight at points A and B in Figure 12a, the water
content at these points would differ considerably.  The shear
strength of the compacted soil at A would be higher than that  at
B.  Thus, it is necessary to prepare samples in the laboratory at
values of water content and dry unit weight comparable to  those
of the field-compacted soils.

    The structure of  a compacted clay is highly dependent  upon
the method of compaction and molding water content, and structure
has a significant influence on both shear strength and
permeability in compacted soil.  It is important to use a
laboratory compaction procedure that would yield compacted test
samples with a structure similar to that of the field-compacted
soils.

    Shear strength parameters appropriate to a particular
analysis depend on drainage condition, stress  condition, and  rate
of shear.  The parameters fall into two general categories:   (1)
total stress  (undrained strength parameters);  and  (2) effective
stress (drained strength or undrained strength with pore pressure
measurement parameters).  In addition, pore water pressure, the
                                38

-------
                 V)
          (a)
                Oi
                 ra
                 e
                 to
                           water content, w
                                compaction
                                curve
                           water content, w
    Figure 12.
 Typical compaction curves showing (a) dry unit
weight-water content relation and  (b) variation
of shear strength with water content  for cohesive
soil.
hydrostatic pressure present  in the water contained in the  soil
voids, affects shear strength determined by  effective stress
parameters.  During stress  application to a  soil mass, the
applied stress tends to compress the  soil, causing an increase in
the pore water pressure.  This pore water pressure buildup  may
reduce shear strength.

    The low permeability  of clays  causes pore-water pressure  to
build up during laboratory  or field tests, thereby controlling
the shear strength properties that are measured and used in
design computations.  Therefore, it becomes  necessary to
understand the mechanisms of  pore-water pressure changes in
cohesive soils when making  an attempt to understand the  shear
strength behavior of cohesive soils.  Discussions of pore
pressure and shear strength relationships can be found in most
soil mechanics texts  (e.g.  Bowles, 1984j Scott, 1980; Sowers,
1979, etc.).


3.2.2.2  Slope Stability  Analysis  —

    A dike that  includes  a  conservatively  specified height  and
slope may require only a  simple  stability chart analysis or a
                                39

-------
basic computer  analysis.   Guidelines for such conservative design
embankment heights  and slopes are provided by the U.S. Bureau of
Reclamation  (1974).   A dike design that includes relatively high
embankments  and steeper slopes,  however, may require a complex
computer analysis.

    Slope  stability analysis techniques are well-developed.
They are described  in most soil  mechanics text books and various
references (Perloff and Baron, 1976; USAGE, 1970b? Wright, 1969).
The conventional stability analysis failure modes can be
classified as either "circular arc" (rotational slide)  or
"sliding wedge" (translational slide).   A third type of
instability  is  soil "flow."  The three  types are depicted in
Figure  13.
                                        a. translational slide
                                            (sliding wedge)
                                        b. rotational slide
                                            (circular arc)
                                        c. flow
              Figur* 13.  Type* of slop* instability.
                                 40

-------
     Stability analysis should be performed considering the
following load cases:

     •  end-of-construction or short-term
     •  steady seepage or long-term
     •  rapid drawdown
     »  seismic

Descriptions of these load cases and appropriate strength
parameters for each are presented in several sources (USAGE,
1970b? Sherard, 1963; Perloff and Baron, 1978).  Depending on the
subsurface conditions, embankment geometry, operational
characteristics, and regional location of the impoundment, one or
more of the noted load cases may be eliminated from the analysis.

     More than 30 methods for performing slope stability analysis
exist.  Among these, Bishop's Modified Method and the Ordinary
Method of Slices are the most commonly used for circular failure
surfaces, and Spencer * s Method for non-circular failure surfaces.
Johnson (1975) recommends using the simplest suitable procedure,
because even the most detailed of the conventional or limiting
equilibrium methods is relatively crude and neglects stress-
strain relationships.  Several of the methods available and
applicable stability conditions are listed in Table 6.


       TABLE 6.  CURRENT PROCEDURES FOR STABILITY ANALYSIS.
     ________________________________________________
           Equilibrium Conditions Satisfied  Shape of  Computation
Procedure  Overall  Slice  Vertical  Horiz.  Sliding        by
           Moment   Moment  Force    Force   Surface  Computer Band
Ordinary
method of
slices - Yes No No
Swedish
method

No Circular Yes Yes


Simplified
Bishop       Yes      No     Yes      Ho    Circular    Yes    Yes

Janbu        Yes     Yes     Yes     Yes      Any       Yes    Yes

Morgenstern
and Price    Yes     Yes     Yes     Yes      Any        No    Yes

Spencer      Yes     Yes     Yes     Yes      Any       Yes    Yes

Sarma        Yes     Yes     Yes     Yes      Any       Yes    Yes
 ======= = :
                                  41

-------
    These methods are applied to dike stability evaluations
using computer programs and stability charts.   Generally,
embankments less than 10 to 15 feet (3 to 4.6  m)  high can be
analyzed effectively using stability charts.   For higher
embankments or more sensitive situations, computer analysis may
be recommended and verified by the design engineer.

    Software programs for personal computers are available from
USIFA and'other sources (e.g., Jeyapalan, 1986; Geo-Slope
Programming, Ltd., 1985; and USEPA, 1986c).  These programs
typically include various options to handle  even the most
difficult site conditions.

    "Geotechnical Analysis for Review of Dike  Stability  (GARDS)"
software, developed for the USEPA (1986c),  is  a menu-driven
software package for design review of earth  dike stability.  It
can be used to analyze site hydraulic conditions, dike slope and
foundation stability, dike settlement, and soil liquefaction
potential.  The program was designed expressly for the evaluation
of dikes at hazardous waste disposal facilities.   It is intended
to help determine whether a design meets minimum safety criteria
under service conditions.

    Slope stability can be analyzed quickly using stability
charts.  The charts generally assume smooth,  straight-line slopes
and uniform soil conditions.  They can be used to obtain
approximate safety factors for most complex problems if irregular
slopes are approximated by a straight line,  and average values of
unit weight, cohesion, and friction angle are  used.  Stability
charts are presented in a number of geotechnical references
(USAGE, 1970b; Naval Facilities Engineering Command, 1982;
winterkorn and Fang, 1975).

    Minimum Factor of Safety — The recommended safety factor
for a slope depends on the following:

    *    the degree of uncertainty  in  the shear strength
         measurements, slope geometry, and other conditions;
    »    the costs of  flattening or lowering the slope to
         increase  stability;
    *    the costs and consequences of a  slope  failure; and
    *    whether the  slope is  temporary or permanent.

    When detailed analyses of slope stability  are performed, a
number of circles must be examined to locate the most critical
circle, that is,  the one with the lowest factor of safety.  The
factors may be plotted, on a dike cross-section, at the  circle
centers and then  contoured to find the location of the minimum.
Figure 14 shows two examples of the results of this procedure.
                                42

-------
                                              all circles tangent
                                              to elevation -20 ft.
           (a)  Contours ofHor circles tangent to elevation -20 ft.
                  +60 -

                  +50 -

                  +40 -

               &  +30 -

               c  +20 -

               |  +10 -

               *o
                  -10  -

                  -20  -
                                              all circles tangent
                                              to elevation-10 ft.
           (b)  Contours offfor circles tangent to elevation -10 ft.
Figure 14.
                 Pact or-of -safety contour* for slop* stability.
                 (10 feet =3.05 meters)
     Factors of  safety for design slopes  may range  from 1.4 to
2.0.  The typical  range for  a hazardous  waste surface impoundment
is 1.75  to 2.0, with the actual value depending on consideration
of the influence of the four items noted above.
                                  43

-------
     The liner system includes the barrier layer,  the leak
collection layer, and any soil protective layers.


3.2.3.1  Regulatory Constraints and Guidance —

     Minimum technology requirements have not been specified
under the Resource Conservation and Recovery Act for non-
hazardous waste surface impoundments.  USEPA is currently
proposing minimum technology for non-hazardous waste landfills,
but not surface impoundments, based upon a single composite liner
or its performance equivalent.  Similar technology would seem
reasonable for non-hazardous waste impoundments, but it has not
been officially proposed at this time.

     The "minimum technology requirements" specified in HSWA
Section 3004(o) for new hazardous waste landfills and surface
impoundments, require a double liner with a leak detection and
collection layer placed between the liners.  The double-liner
system must prevent the migration of hazardous constituents
through the liner during the impoundment's active life and post-
closure care period.  USEPA's guidance on the minimum technology
requirements (USEPA, 1987a) identified two acceptable double-
liner system designs that would meet the HSWA requirements:

     •  Geomembrane/Composite Double-liner System (Figure 15)
     •  Geomembrane/Soil Double-liner System (Figure 16)

     A double-composite liner may also be used  (Figure 17).

     A "composite" liner in the guidance is a two-component
liner, a geomembrane installed directly on a smooth-surfaced low-
permeability soil component, with the two components acting as a
unit.  The two-component composite liner is the bottom secondary
liner of a double-liner system.  The top primary liner of the
double-liner system is a geomembrane.  The drainage  (leak
detection) layer separates the two liners.

     Careful consideration must be given to the relatively low
friction angle between the geomembrane and the  low-permeability
soil, especially for the possibility of moisture collecting
between the layers.  At the same time, attention must be given to
the contact between the geomembrane and the low-permeability
soil.  The soil surface should be smooth and the contact with the
geomembrane as continuous as possible to minimize any potential
flow along the interface.

     The double-liner system must prevent  leakage through the
secondary liner of the system for the active life of the
impoundment and the post-closure care period.   More

                                44

-------
         protective soil layer
                       rr^afcr-  compacted toil
          composite liner
 Figure 15.  Cross section of  double  liner with composite
                bottom  liner.
        protective sol! layer
geomembran* liner
with protective oeotextiles
                               . _ _ .'-^ . _
                      Lr^r^rLr-/compacted sol
                geotexHIe fitter
                                                   leak detection/collection layer
  Figure  16.   Cross section of double  liner with  soil-only
                bottom liner.
        protective soil layer
aeomembrane liners
with protective gaotactHes
                      r^->£-izt_— / OOfKMCtod SON
                                                    leak detection/collection layer
          composite liners
Figure  17.   Cross  section  of double composite  bottom  liner.
                                    45

-------
specifically,the double-liner system must prevent migration of
contaminants into the primary liner and through the secondary
liner during the active life and post-closure monitoring period.

    The hydraulic head of liquid above the secondary liner
(within the drainage layer) of a double-liner system typically
has a non-zero value (Giroud, 1985).  This indicates that the
primary liner does, in fact, leak.  A leak-collection (drainage)
layer between the primary and secondary liners, that allows
liquids to drain freely to a pumpout unit, will minimize the
hydraulic head on the bottom liner and thus will lessen any
potential leakage through it.

    USEPA has described the interaction of the leak-collection
system and the composite bottom liner in the process of
detecting, collecting, and removing any contaminant leakage
through the top liner (USEPA, 1987g) .  It was shown that the
composite (geomembrane/clay) liner significantly improved the
leak detection capability for small leaks (leaks less than 1
gal/acre/day [9.35 liters/hectare/day).  Even with small holes in
the geomembrane component, the collection efficiency is not
significantly reduced.  On the other hand, the IPA report
suggests that a clay liner alone, meeting the 3.9 x 1CT8 in./sec
(1 x 10"7 cm/sec) criterion, would absorb a leak on the order of
80 gal/acre/day (748 liters/hectare/day).  It is possible then
that only a leak in excess of that rate would be detected in the
sump of the leak-collection system above a clay liner.  Further,
USEPA believes that the liquid absorbed in a soil liner will
eventually be released to the environment.

    Even though a soil liner may not be effective in preventing
contamination from eventually escaping the facility, it will
minimize leakage through a breach in the adjacent geomembrane
component.  In addition, it will attenuate migrating waste
constituents, provide protective bedding for the geomembrane, and
provide a long-lasting stable foundation for the impoundment.

    Careful attention must be given to the sliding potential of
the geomembrane on the low-permeability soil layer.  It is not
appropriate to include a geotextile at the interface, because it
would destroy the compression seal intended between the
geomembrane and soil.


3.2.3.2  GeoBumhrana/Coatposite Double Liner —

    As noted above,  the geomembrane/composite  double-liner
system consists of two liners — a top primary geomembrane liner
and a bottom secondary geomembrane/soil composite liner,
separated by a leak-collection system  (Figure  15).  USEPA"s
minimum technology guidance  (USEPA, 1987a) recommends that the
primary geomembrane  be at least 30 mils  (0.76 mm) thick where

                                46

-------
covered by a protective soil and/or geotextile layer.   For an
uncovered geomembrane, a' thickness of at least 45 mils (1.14 mm)
is recommended.  The guidance suggests that thicknesses of 60 to
100 mils (1.52 to 2.54 mm) may be required to resist various
stresses.  In any case, the design engineer should recognize that
some geomembrane materials may require greater thicknesses to
prevent failure or to accommodate unique seaming requirements.
It may be possible to use combinations of geotextiles and
geomembranes in lieu of increasing the geomembrane thickness in
some instances.

    All liner and leak-collection components must be chemically
resistant to the waste handled at the facility.  Chemical
compatibility of geosynthetic materials should be tested using
EPA Method 9090.  Past test data or actual performance data may
be used in lieu of new testing if the materials and wastes can be
shown to be the same in both the past and new applications.

    As noted above, in the composite secondary liner, the
geomembrane upper component and a low-permeability-soil lower
component should be in direct and uninterrupted contact.  The
interface should be smooth and designed and constructed to
provide a. compression connection between the two components to
minimize lateral flow between them, should a hole develop in the
geomembrane.  Attaining the desired contact is difficult because
of the near impossibility of completely flattening all
imperfections on the soil surface.  It requires careful
compaction and construction quality control to minimize the
imperfections (Giroud & Bonaparte, 1989).

    The low-permeability-soil component of the composite liner
should be at least 36 inches (90 cm) thick and have an in-place
hydraulic conductivity of 3.9 x 10"8 inches/sec (1 x 10~7 cm/sec)
or less.  These characteristics have been chosen by US1PA as
those that will prevent contaminant breakthrough during the
active and post-closure period, when the liner is constructed
ideally (USEPA, 1987a).

    The leak-collection system is generally a layer of highly
permeable sand or synthetic material extending across the bottom
and up the sides of the impoundment between the liners.  The
hydraulic conductivity of the leak-collection system should be at
least 2 ft/min  (1 cm/sec) for sand.  The transmissivity of a
geosynthetic layer should be at least 3.3 x 10"2 ftVsec (3 x 10"3
m2/sec).   A series of drain pipes,  a sump,  and pumps are also
included in the system to facilitate rapid removal of liquid.


3.2.3.3  Geomembrane/Compacted Soil Double Liner --

    In accord with USIPA's minimum technology guidance, the
geomembrane/compacted soil double-liner design consists of a

                                47

-------
primary geomembrane liner placed above a secondary low-
permeability-soil liner, separated by a leak-collection system
(Figure 16).  It is very similar to the geomembrane/composite
double-liner system described above, except that it does not
contain the geomembrane component in the secondary liner.  The
primary geomembrane liner has the same recommended minimum
thickness and chemical compatibility as the primary geomembrane
liner of the geomembrane/composite liner system.  The low-
permeability-soil secondary liner thickness depends on site- and
design-specific conditions, but should not be less than 36 inches
(90 cm) thick, according to the guidance.

   .Although USEPA (1985b) recommends a 36-inch (90-cm) minimum
thickness, the Agency believes that a low-permeability-soil liner
of that thickness with a hydraulic conductivity of 3.9 x 10"8
in.'/sec (1 x 10~7 cm/sec) may not prevent constituents from
migrating through the liner prior to the end of the post-closure
care period.  Note here again that USEPA has concluded that a
well-constructed soil liner will allow contaminated liquid to be
absorbed, and eventually released, at a rate on the order of 80
gal/acre/day (750 liters/hectare/day) or more (USIPA, 1987g).
Therefore, documentation supporting the use of a particular
thickness is required and should include responses to the
reservations expressed by USEPA (1987a) that such a design
meeting the no-breakthrough requirement is neither economically
or technically feasible.  USEPA states that conservative assump-
tions should be used to estimate the necessary low-permeability-
soil liner thickness, due to the lack of precision with which
such estimates can be made (USEPA,  1987a).

    The leak-collection  (drainage)  layer in the geomembrane/soil
double liner system is essentially the same as the leak-
collection layer in the geomembrane/composite double-liner system
described above.  In this case it is placed directly on the low-
permeability-soil layer, preferably with a geotextile separating
them.
3.2.3.4  Double Ccaq>oslt.* Lin«r —

    A further variation  of the double  liner  is  a system
comprised of  two composite geomembrane/compacted soil  layers
 (Figure  17).   USEPA  (1987h) recently noted that many owners and
operators of  landfills and surface impoundments have indicated
that  they planned to use a composite liner for  the  top liner  as
well  as  the bottom liner.

     In a double-composite-liner  system,  the  liners  are separated
by a  leak-collection layer, no different than those of other
double-liner  designs, except  for the insertion  of a separation
layer at the  top of  the  leak-collection  layer.   The separation


                                48

-------
layer would prevent the micrration of the urinary  soil liner

-------
geomembrane as the principal barrier (USEPA, 1989a).  The
geomembrane need not necessarily be of the same composition in
both the liner and cover.

     Geomembranes are manufactured using low-permeability
synthetic polymers, either non-reinforced or reinforced with
fabric material.  Geomembranes are made with various base
polymers and additives, thus varying considerably in their
physical and chemical properties and their interactions with
different wastes.  No geomembrane is applicable to all wastes; a
particular type must be selected for each application.

     The base materials of geomembranes are high-molecular weight
compounds (polymers).  Some common polymers presently in use as
base products for geomembranes follow:

     *  thermoplastics (e.g., polyvinyl chloride [PVC])
     •  crystalline thermoplastics (e.g., high density
        polyethylene [HDPE])
     *  thermoplastic elastomers (e.g.,  chlorosulfonated
        polyethylene [CSPE])
     *  elastomers (e.g., butyl rubber)? limited availability

     Kays (1987), Giroud (1985), and OS1PA (1983b) detail the
processing and manufacturing techniques used in producing
geomembranes.  USEPA has provided detailed technical information
about geomembrane materials used to contain specific types of
hazardous wastes (USEPA, 1983b and 1988a).  Geomembranes and
their polymers continue to improve as their manufacturers and
users gain experience in waste containment.  Indestructability
and construetability are two general goals difficult to achieve
in the same material.

     The compatibility of a liner material with a specific waste
is an important consideration in planning a surface impoundment.
Liquid wastes in surface impoundments may vary from complex
mixtures of variably concentrated constituents to highly
concentrated contaminants contained in a relatively simple
matrix.  Additionally, waste constituents may change over time.
During the liner design process, a representative sample of the
liquid waste should be analyzed for waste properties that could
potentially cause damage to liner material.  Several methods for
obtaining samples of hazardous wastes have been discussed by
USEPA (1982).  Potentially detrimental waste properties include
the following:

     •  acidity;
     *  alkalinity (greater than pH 10);
     *  temperature extremes; and
     *  hydrocarbons.
                                50

-------
      For hazardous waste surface impoundments, effective waste
containment is required for an extended length of time, including
the active impoundment life plus (if the unit is closed in place)
a 30-year post-closure period.  Testing nay be necessary to
determine the long-term effect of the impounded wastes on the
liner.  Performance history and existing test data may preclude
site-specific testing in those cases where the waste has not
changed characteristics and the same geomembrane material
(essentially identical formulation) is used.

     Test and/or performance data must be provided with the
permit application.  Tests should be conducted using
representative samples of waste and leachates to which the liner
is to be exposed.  USEPA has developed Test Method 9090,
"Compatibility Test for Wastes and Membrane Liners."  This test
is intended to facilitate the estimation of the long-term
compatibility of liner material with wastes by immersing liner
samples in the waste for 30, 60, 90, and 120 days at 23 ± 2°C (73
± 3.6°F)  and at 50 ± 2°C  (122 i 3.6°F).  A comparison of specific
physical properties of the geomembrane determined before and
after immersion is then made to determine suitability of the
liner.  This test does not address all relevant variables, but
does provide a uniform method of testing geomembrane materials.
USEPA has published other compatibility test methods and a list
of liner manufacturers and material sources (USEPA, 1988a).

     Geomembranes used in hazardous waste surface impoundments
are susceptible to failure during facility operation.  Giroud and
Ah-Line  (1984) discuss two basic types of geomembrane failure:
(1) excessive geomembrane displacement (i.e., vertical or lateral
movement) and  (2) an unacceptable leak.  Geomembrane liner design
should specify material components and construction practices to
prevent failures by:

     *  protecting the geomembrane from puncture, scratching,
        abrasion, or other damage (from above and below);

     *  demanding that great care be taken to prevent damage to
        the geomembrane sheet during all installation processes;

     •  providing gas venting in the drainage layer;

     •  avoiding bridging, rippling, stretching, or other
        stressed conditions;

     *  allowing slack for shrinkage;

     *  avoiding nonessential penetrations;

     *  eliminating tensile stresses as much as possible, by
        design;


                                51

-------
     •  providing detailed and rigorous seaming instructions;

     •  providing slip-preventing anchorage the tops of slopes;

     •  requiring well-trained and experienced installation
        personnel; and

     •  providing a detailed quality assurance plan.

     Design details for geomembranes and other geosynthetics to
resist various stresses have been provided by USEPA (1987c) and
Koerner (1990).  Protective layers such as soil or geotextile can
help guard against damage to the geomembranes.

     Geoaenbrane Protective Layers —   A protective layer of
soil or a soil /geotextile layer may be used on the surface of the
liner system to protect the underlying geomembrane from
construction damage during installation, loads imparted by the
waste, weathering, erosion and abrasion, to increase friction,
and to dampen potential chemical attack.  Above the liquid level
of the impoundment, coarser-grained (sometimes rubble) material
placed over a geotextile protective layer is often used.  The
coarse material will generally be more stable on steeper slopes
and will dampen wave action and run-up.  An even coarser layer
(e.g., riprap) may be applied over the geomembrane-protective
layer to prevent erosion in larger impoundments.  Below the
liquid level, sand is often used to protect the geomembrane
against puncture, and to dampen the effects of strong chemical or
high -temperature waste inputs.  A cross section showing some
protective layer options is shown in Figure 18.

     Based on field experience, USEPA  (1983b) recommends that
protective soil on a liner be placed at no greater than a  3H:1V
slope and that it be at least 18 inches (45 cm) thick before
allowing any construction equipment over the liner.  Slope
stability may be a problem with soil placed directly on the
geomembrane.  Mitchell et al  (1989) have examined geomembrane-
soil interfacial stability and have concluded that the design
should incorporate the effect of pore-water pressure.  A
geotextile can reduce pore-water pressure at this interface.  The
recommended general equation  for determining the stability is:

               [( i z cos (3) - n] tan ^
          FS =
                      7 z sin /3

     where FS = factor  of safety
             Y = soil density (including moisture)
             z = soil depth perpendicular to the geomembrane
             /3 = slope angle
             ^ = interfacial friction angle
             j/ = pore-water pressure

                                52

-------
    A factor  of  safety of  at  least 1.5 is recommended.

    Geotextiles  may  be used at various places in the liner
system as protective layers and, where appropriate,  as slide-
resistant interface  materials between  soil and geomembranes.   On
the top (primary) geomembrane liner, a geotextile may be used to
shield the geomembrane from any larger,  sharper particles in the
overlying soil protective  layer.   A geotextile layer can be used
to reduce the thickness of the soil protective layer from 18
inches (45 cm) to perhaps  12 inches (30 cm).   Geotextiles may
also be used between that  geomembrane  and the drain layer
material, and again  at the bottom  of the drain layer atop the
second geomembrane as  shown in Figure  18.   Nonwoven geotextiles
ordinarily have  excellent  protection properties, making them
ideal for these  applications.  In  general, the greater the mass
per unit area, the greater the protection afforded by the
geotextile.

    The exterior dike  slope may be protected by riprap,
vegetation, or other stabilizing material to prevent erosion.
Vegetation requires  planting a suitable plant species and
maintaining it.  Several references are available to assist in
establishing vegetation (e.g., USEPA,  1979b and 1983g; Lee et al,
                                geotextiles
  riprap armor


  soil protective
  layer
xxxxx/;
 native soil
 foundation
                                compacted low-
                                permeability soil
        leak collection layer
        Figure 18.  Impoundment dike cross  section showing
                    optional protective layers.
                                53

-------
1985).  Riprap consists of graded rock courses and a geotextile
filter placed directly on the dike.  Riprap has many advantages,
including flexibility, ease of repair, simple construction, and
salvageability.  Battelle Pacific Northwest Labs (1982), the U.S.
Bureau of Reclamation (1974f, and Barren (1977)  discuss the
design of riprap coverings.  FHWA (1985 and 1988)  and Koerner
(1986) have reviewed geotextile design for erosion protection.

    Although not common, other materials may be used for dike
protection.  They include concrete, asphalt, and soil cement.
Weather resistance, and resistance to the contained waste where
in contact with it, must be carefully considered before designing
with these materials.

    Low-Permeability Soil Liners — The purposes of the low-
permeability soil component of the secondary liner are (1) to
minimize the migration of hazardous liquids through the
geomembrane component if a breach in the geoinembrane should occur
and (2) to attenuate constituents that might leak through the
geomembrane.

    Soil liners are not impervious, but they do control seepage
and have been used, because of that and their low cost, in the
past as the sole liner in surface impoundments and landfills.
Soil may be treated, remolded, and/or compacted to achieve
prescribed flow-impeding and contaminant-attenuating
specifications.  However, the USEPA does not believe that a soil
liner can ordinarily be constructed to meet, by itself, the
requirement of no contaminant breakthrough during the active life
and post-closure monitoring period for a surface impoundment.

    Despite USEPA's wariness of the impermeability of soil
liners, they are still recommended in USEPA's minimum technology
guidance as backup to the geoinembrane component of the secondary
liner in the required double-liner system.  Further, low-
permeability soil may be used as the secondary liner without the
geomembrane if it can be shown that it will provide equivalent
performance and not allow contaminant breakthrough during the
active life and post-closure monitoring period.  The Agency
believes that such a showing will be very difficult.

    Waste fluids or leaked fluids may substantially increase the
overall permeability of a soil liner, which then will not meet
the design requirements (USEPA, 1983b).  USEPA (1983b) provides
guidance on evaluating the effects that waste fluids may have on
the permeability of soil liners.  Selection of soil liner
material should include testing the material with a standard
aqueous permeant (e.g., site ground water or a 0.01 Normal
solution of CaSO4)  and a representative sample of the waste
liquid to be impounded.  Distilled water is not a good aqueous
                                54

-------
penneant because it tends to react with clay particles, resulting
in an unnaturally low permeability (Olsen and Daniel, 1981).

    The material chosen for the low-permeability soil component
should be compactable to uniform hydraulic conductivity values no
greater than 3.9 x 10"8 in./sec (1 x 10"7 cm/sec).  Elsbury et al
(1985) found from a survey of hazardous waste disposal facilities
that liner soils are generally selected on the bases of grain-
size distribution, liquid limit (LL),  plasticity index (PI), and
laboratory hydraulic conductivity.  In that survey,  soils used to
construct low-permeability soil liners consisted predominantly of
the clay groups designated by CL (low-plasticity clay) and CH
(high-plasticity clay) in the USC5.  LL and PI values were
generally in the range of 20 to 45 percent and 5 to 30 percent,
respectively.  Table 7 lists typical soil liner design
specifications.

    The USEPA uses the term "low-permeability soil" so as not to
imply that there is a narrow restriction on the type of soil that
may be used.  The terra "clay liner" is used casually to refer to
any soil liner, but the liner may, in fact, be comprised for the
most part of materials other than clay.  Also, a sandy soil may
be made to meet very low permeability requirements with the
addition of only a small percentage of bentonite.

    The performance of the soil selected for the design can be'
verified in a test fill.  The test fill data should demonstrate
whether a structurally stable soil liner component with uniform
hydraulic conductivity of 3.9 x 10"8 inches/sec  (1 x  10"7 cm/sec)
or less is attainable using the soil, construction equipment, and
methodology specified in the design.  USEPA (1988b)  describes
some of the techniques used in constructing and testing a test
fill.  Data from test fills can be use"d to evaluate the
relationships between density, water content, and hydraulic
conductivity, and to validate design and construction procedures
(Mundell and Bailey, 1985).

    USEPA's minimum technology guidance  (1987a) recommends that
the soil component of the secondary liner have a thickness ofi at
least 36 inches  (90 cm) and be chemically resistant to the   I
impounded waste.  The guidance also states that the soil should
be compacted in lifts of 6 inches  (15 cm) or less, after
compaction, and be free of rocks, roots, and rubbish.

    The  finished  surface of the soil component must  be as smooth
as possible to facilitate continuous contact with the overlying
geomembrane.
                                  55

-------
   TABLE 7.  TYPICAL OPTIMUM SOIL LINER DESIGN SPECIFICATIONS.
Property
    Specification
                                             Test Method/Reference
Thickness
Grain Size
Hydraulic
Conductivity
(laboratory)
Hydraulic
Conductivity
(field)

Liquid Limit

Plasticity
Index

Compaction
Maximum
Clod Size

Frost
Susceptibility

Volume Change

General
                Minimum 36 in.  (90 cm)
                measured perpendicular
                to slope and compacted
                in 6-in. (15-cm)  lifts.

                >50% by weight  passing
                200 meshi >25%  clay
                content (0.002  mm)

                <1 x 10'7 cm/sec (3.9 x
                10"8 in./sec) when
                compacted to anticipated
                field density and moisture
                content.

                <1 x 10"7 cm/sec
                (3.9 x 10"8 in./sec)


                24 - 60

                10 - 30
>95% Standard or
>90% modified
Proctor density

< 1 in. (2.5 cm)
0.039 - 0.31 in./day
  (1-8 mm/day)

<10%

Liner should be free of
all organic matter (e.g.,
vegetation, wood), large
stones, and construction
debris.
                             EPA (1987a)
                             ASTM D 422-63
                             Daniel et al  (1985)
                             EPA (1983c)
                             Day and Daniel  (1985)
                             Anderson et al  (1984)
                             ASTM D 4318-84

                             ASTM D 4318-84
                                             ASTM D 698-78 or
                                             ASTM D 1557-78
                                             Observational
                                             Chamberlain et al
                                             (1982)

                                             Holtz (1965)
                                  56

-------
    Leak Detection and Collection System — 40 CFR 264.221
requires a leak detection,  collection and removal system between
the liners, including the sidewalls,  of the double-liner system.
It must perform its function during the active life and post-
closure care period of the impoundment.

    The leak collection system should incorporate a granular or
synthetic drainage layer designed to rapidly detect, collect, and
remove liquid that migrates into the space between the liners.
The system should be designed so that little or no liquid head
will ever be present to impinge on the bottom (secondary) liner.

    Excessive head in the leachate collection system can result
in liquid "whales" in the upper liner (USEPA, 1989b).  The head
would be caused, at least partially,  by ground water infiltrating
from a high water table.  After the collection system along the
bottom has been saturated and pumpout is inadequate, water can
rise rapidly along the side slopes.  Whales may also be caused by
gas buildup beneath the primary liner.  Either type is less of a
concern with double liners than with older single liners.

    A minimum bottom slope of 2% should be specified for the
secondary liner to facilitate drainage in the collection layer to
the pumpout sump.  That layer should have a minimum hydraulic
conductivity of 2 ft/min (1 cm/sec) and should operate without
physical or biological clogging.  Granular materials should be
USCS-classified as "SP", without fines that may clog the layer,
and with grain sizes no larger than 3/8-inch (0.95 cm) if placed
directly on the secondary geomembrane liner.  A geotextile layer
can be placed between the drainage layer and geomembrane to
facilitate placement of the granular layer and to reduce the
potential for damage to the geomembrane.

    The drainage  layer materials should be chemically resistant
to the waste contained in the surface impoundment, and of
sufficient strength and thickness to prevent collapse from the
expected overburden pressures.  This makes granular drainage
especially attractive.  Drain pipes of appropriate size and
spacing must be located in the bottom of the unit to remove
liquids that may  collect there.  Larger diameter pipes have been
recommended  (e.g., 6 inches  [15.2 cm]) because they offer greater
protection against clogging and are simpler to inspect and
maintain.  Guidance concerning appropriate drain-pipe sizing and
spacing has been  provided by USEPA (1983a).

    Geosynthetic  drainage  layer materials, available  in  a
variety of configurations, can be used in place of the granular
layer.  Several of the available materials have been described by
Koerner  (1990).   If used, they must be demonstrated as equivalent
in performance to the conventional granular system with pipes.
Equivalence  in flow capacity equates to a transmissivity of  3 x

                                57

-------
1Q~3 mVsec (0.032 ftVsec) for the geosynthetic material.
Geosynthetic drainage materials must also be chemically  resistant
to the waste and leachates,  have no detrimental effect on  the
geomembrane liners, and withstand the designed loads.

    Geosynthetic drainage and filter materials may creep and
deform under load, resulting in gradual restriction of flow
capacity over time.  Design should be based upon laboratory
transmissivity tests of the proposed material over time  under
loads equivalent to and exceeding the expected overburden.   A
minimum design ratio (factor of safety) of 3.0 for loading has
been suggested by USEPA (1987c)  to maintain the required
transmissivity.

    Some impoundment designs use synthetic (e.g., "geonet")
drain layers on the side slopes and granular layers on the bottom
section (Figure 19).  Geosynthetic material is easier to install
and may be more stable (with anchorage) on the side slopes and,
because it is thinner than a granular layer, allows greater waste
containment capacity within an impoundment of the same overall
dimensions.  A geotextile layer may be placed on either  side of
the drain layer to prevent creep of the geomembrane into the
drainage material, lowering the capacity (see Figure 42).
Anchorage for geosynthetic drainage material is usually  similar
to the anchorage used for the geomembranes and may perhaps be the
same trench used for one of them.
                         soil protective
                               layer
     geomembrane liners
     with protective
     geotextiles
 exuiea 	-7\
 L, /rC\.. ......
 r^    ••
I— granular drain layer
                                              low-permeability
                                              soil layer
   Figixr* 19.   Junction of «id«-wall geonet and bottom granular
              layers of leak detection, collection,  and removal
              system.
                                58

-------
    The system design should include an appropriately sized
sump, with a depth of at least 12 inches (30 cm)  below the
drainage layer grade, to collect and remove liquids.  If
possible, the  pumpout and sampling system should be routed up the
side walls and not penetrate the top liner until well above the
liquid level (Figure 20).  Again, all system components should be
designed with  materials resistant to the waste being contained in
the impoundment.   Provision should be made for measuring and
recording the  liquid volumes collected in the sump, and for
disposal of the liquid.  Operating procedures for handling
liquids collected in leak detection/col lection systems are
discussed in Section 5.1.2.

    Access to  all parts of the system should be provided to
facilitate inspection and maintenance of components.  Manholes
and cleanouts  should be placed so that maintenance equipment can
reach any pipe section.  Bass (1985) summarizes the state-of-the-
art information on system failure, concerning leak collection
system design, construction, inspection, maintenance, and repair.

    Geotextile Filter — Geotextiles are used as filters to
prevent fine-grained soils and solids from the waste from
entering and clogging the leachate collection system.
                      teak collection system   	
                      pumpout and sampling access
       geomembrane liners
       with protective
       geotextiles
                                               tow-permeability
                                               soil layer
            leak collection layer
      Figure 20.  Example access for leak collection system
                   and liquid removal.

-------
Geotextiles, as filters, are placed adjacent to both synthetic
and granular drainage layers.  Adequate performance will depend
on proper selection of the geotextile in relation to its
compatibility with the material that is to be filtered.   Detailed
design of geotextile filters is described by USEPA (1987c), FHWA
(1985), and Koerner (1990).  Design basically consists of
selecting a geotextile whose largest opening, as measured by its
"apparent opening size" (AOS), is smaller than the larger of the
particles to be filtered, as measured by the d(S of that material
times a retention factor (\)r i.e.,

    AOS * d85/factor of  safety
    i
    The flow rate through the geotextile, as measured by its
permittivity, f , is then selected to be greater than the flow
from the soil times a factor of safety, usually 10 or greater.
The clogging potential of the filter is then evaluated based on
the fines in the material to be filtered and the type of
geotextile.  Clogging potential is best evaluated through long-
term flow tests using a simulated cross section and leachate.

    Other important considerations include physical durability
(it must survive construction activities) and chemical and
biological resistance to the environment of the impounded waste.
One reference for the minimum physical requirements for
separation and drainage applications is the AASHO-AGC-ARBTA Task
Force 25 set of specifications, as published in "Geotextile
Design and Construction Guidelines" (FHWA, 1990).

    Gag-Venting Layer — A gas-venting layer should be
considered in the design beneath the bottom (secondary)  liner,
between it and the foundation.  It would be appropriate where the
foundation material contains biodegrading organic material or
other sources of gas.  The material used for the venting layer
should be similar to that used for the leak collection layer.  It
may be either granular or geosynthetic material.  Thick
gee-textiles, with the capacity for in-plane transmission, may be
considered where gas generation is anticipated to be low.  Geonet
strips may be inserted at regular intervals to increase flow
capacity and facilitate venting.  Gas exits from the venting
layer should not penetrate the liners.  Instead, wherever
possible, they should be brought to the surface outside the side-
slope liner (Figure 21).


3.3  LIQUID LEVEL COHTROL

    Perhaps the greatest environmental risk in surface
impoundment operation is unintentional liquid overflow.
Reliable liquid level controls are critically important.  Two are


                                60

-------
                                         gas vent
                         soil protective
                         layer
   geomembrane liners
   with protective
   geotextiles  —A
           /\
                          ^
                         -»*»Z*

                     ^
                   tr3:»V   low-permeability
                jjj& '     soillayer
              ^jl^     gas vent layer
              "^'*         (geosynthetic or granular)
         , _.  ,
       collection layer
                              r*-—-^
                                       geotextile fitters
       Figure 21.
Liner system showing gas-vent layer  and
exit through top of dike.
required, passive  (spillway or overflow pipe)  and active (level-
actuated pump or valve).   For detailed information on liquid
level controls, the following references are recommended: Linsley
and Franzini  (1979),  Considine (1974), and Hicks (1972), and
Shiver et al  (1985).

    Unless  a  passive outfall is part of an intended flowthrough
process, it should be designed only as an emergency outlet to
prevent a breach in the dike in case of an uncontrolled liquid
rise.  Spillways or weirs are preferred (Figure 22).  Pipe or
conduit outlets through the dike (Figure 23) are not generally
recommended because seepage may occur around the conduit, leading
to soil displacement and possible dike failure.  A leak plate,  as
shown in Figure 23, may be placed within the dike to interrupt
possible flow and  channel creation along the pipe-soil interface.

    Liquid  releases from a hazardous waste surface impoundment
via a passive outfall operating as a safety valve constitute an
emergency situation.  Potential releases should be considered in
the design.  Containment for such releases, such as a secondary
perimeter dike,  should be provided.

    The  surface impoundment design must also include a  system to
monitor liquid  level.  The system may vary from a manually-read
guage  (Figure 24)  to a liquid-level detection, monitoring, and
control system  designed to operate as an automated system that
detects, monitors, and controls the liquid level with little
manual assistance, other than routine maintenance and
calibration.
                                 61

-------
  impoundment
Figure 22.  Dike spillway with protective toe apron.
                  overflow
                  discharge pipe
                                          apron
     Figure 23.
Impoundment  overflow discharge
pip* through dike.
                          62

-------
                    gauge support and access
                                               staff gauge
      Figure 24.
Example of liquid-level monitoring setup;
manually-read staff gauge.
     In an automated system,  the attainment of a specified liquid
level may trigger  a device to stop input to the impoundment or to
open a discharge line,  or both.  Another option is to trigger a
warning signal  indicating that some other action is needed.
Other options,  or  combinations of options, may be activated with
an automatic  level-sensing system.  Figure 25 is a sketch of a
liquid-level  recorder and alarm system.

                        automatic
                        recorder and alarm
                        gauge support and access
                   DIKE
                                              float
 Figure  25.   Example of liquid-level recording and  alarm •ysten.

                                 63

-------
    The major components governing the flow into and out of a
surface impoundment are the inflow and outflow structures.
Normally, the inflow structure will be a pipe equipped with a
flow valve.  Typical outflow structures are pumps, weirs,
spillways, and pipes.

    The valves used at hazardous waste surface impoundment
should be capable of being integrated into the level control
system, and acting as a manual override for adjusting flow if the
automated system fails.  Valves used in hazardous waste service
must be constructed of materials that are compatible with the
hazardous liquids to which they are exposed.  Considine (1974)
provides a complete review of the types of valves available.

    The selection of a specific pump type should be based on the
characteristics of the liquid to be moved (e.g., corrosivity,
ignitability, specific gravity, density, and total solids
content).  Aid in selecting a pump can be found in a reference by
Karassik et al (1986).

    A protective membrane or apron must be designed to prevent
erosion at outfall and inflow structures where the discharge is
onto a surface.  The design of the protective apron will largely
depend on the types of inflow and outflow structures, flow rates,
and liner protection.  The construction material for the apron is
often concrete, but may be geomembrane, or, in some cases,
velocity-impeding riprap placed over a geotextile.

    When passive outfalls are used, the protective apron should
extend from the outflow structure to some distance away from the
toe of the dike slope  (see Figures 22 and 23).  If the outfall is
designed for regular process use rather than for emergency
overtopping prevention, apron design must prevent erosive flow
and reduce terminal velocities.  Common methods for reducing flow
velocities and erosive flow employ baffled aprons (U.S. Bureau of
Reclamation, 1974) and properly designed sloping aprons (Linsley
and Franzini, 1979).

    In many cases, pipes will be used for both inflow and
outfall structures.  When used for inflow, the primary concern is
to prevent erosion in  soil liners and tears in geomembranes
caused by impinging liquid.  Enlarging the outlet of the pipe to
a size several times larger than the average diameter of the
inflow pipe will act to dissipate energy, reducing the potential
for causing liner failure  (Kays, 1977).  The outlet may be
extended out to the deeper part of the impoundment and located
near the liquid surface so that the energy is dissipated into the
liquid.  If a pipe is  used as an impoundment outfall structure,
the preferred design is to place the pipe so that the outlet is
at or near the same elevation as grade, and extends beyond the
outside toe of the berm  (see Figure 23).


                                64

-------
3.4  SECONDARY CONTAINMENT

    Secondary containment structures are not specifically
required under current RCRA regulations.  However, they can
prevent widespread damage if the impoundment should be breached,
particularly if the impoundment has been constructed entirely
above grade.

    Secondary containment structures usually consist of low
dikes with a much larger plan area than the primary containment
dikes (see Figure 26),  The secondary containment structure
should be able to withstand a sudden surge of released liquid and
should retain the released waste for a period sufficient to allow
detection and recovery of the material.  Engineering
considerations (e.g., soil stability) used in the design,
construction, and operation of the primary surface impoundment
dike are applicable to secondary containment structures.
Hydrologic considerations also should be included in the
secondary containment design.
  primary impoundment dike
                                                  secondary
                                                  containment dike
    Figure 26.  Primary and secondary containment dike layout.
3.5  LEAK  DETECTION SYSTEMS

     "Leak  detection system"  is a term applied  to  a monitoring
device or  technique which monitors  the integrity  of an
impoundment  liner in a non-destructive manner.  In a  sense,  the
conventional leak collection system serves that purpose.   It can
detect the presence of liquid in the leak collection  system and
thus the occurrence of a leak in the primary liner.   If a leak
collection system is installed beneath the secondary  liner,  it
could serve  the same purpose for the secondary liner*  However,
detection  does not occur in  either  case  until  the liquid  reaches
the  sump.  Nor can the leak  location be  pinpointed.

     A  leak detection system, as the term is used  here,  should be
capable of detecting liquid  moving  through the liners essentially
as soon as a hole appears.  In addition, it is desirable  that it
have the capability of locating very small leaks  to within an
area of less than 1 ft2  (0.09  n?) (USEPA, 1984b).

                                 65

-------
    Leak detection techniques employ either sensors  (1)  located
within the potentially leak-affected zone or (2) remotely located
outside the impoundment area.  These are referred to as  in-situ
and remote systems, respectively.  An in-situ system is  arrayed
at the time of construction, while a. remote system may be
installed at an existing site.  Remote sensors may be located at
the surface or in a borehole at some distance from the leak. The
output of either a remote or in-situ system would be recorded at
some convenient surface site.  In situ instrumentation can be
continuously monitored throughout the active life of the
impoundment and during the post-closure period.  Remote  systems
are ordinarily activated periodically during routine monitoring
or when liner failure is suspected.

    Leak detection techniques that may be adaptable to the
surface impoundment situation include (1) electrical resistivity,
(2) electromagnetic techniques, (3) acoustic emissions,  and  (4)
seismic methods.  Of these techniques, electrical resistivity
appears to show the most promise with respect to performance and
reliability (US1PA, 1984h).

    A leak detection system may be arrayed to detect  leakage
into the leak collection system and the leak location in the
primary liner (Figure 27).  This is usually a system that is
operated intermittently, because it uses a transducer floating
across the impoundment surface.  A system may be designed to
detect leakage into the leak collection system, but without  the
capability of locating the leak (Figure 28).  One may also be
arrayed to detect leakage through the geomembrane component  of
the secondary liner, or through the entire composite  liner into
the vadose zone (Figure 29).  All of these systems are based on
electrical resistivity.


 r— recording station
 «                                           ^__ geomembrane liner
                                     leak collection layer
  Figur* 27.   System to detect and locate leaks in top (primary)
             geooMmbran* liner.
                                66

-------
   recording station
conductors

   geomembrane liner
       leak collection layer
   Figure  28.   System to detect leakage through top  (primary)
               geomembrane liner.
   recording station
                               geomembrane liner
        conductor
        well
        leak collection layer
      Figure 29.  System to detect leakage into vadose  zone.


    Lysimeters  (soil-pore liquid samplers)  have been used
successfully to monitor  leachates migrating through the
unsaturated zone  at  land treatment  facilities and other disposal
sites  (USEPA, 1988b).  Lysimeters have  also been placed below
liners at surface impoundments  to collect samples of leaking,
waste  liquid  (Figure 30).   The  advantage of placing lysimetere in
the unsaturated zone beneath a  liner system is that liquid   i
samples can be  collected and analyzed for the presence of    '
hazardous constituents,  their rate  of degradation, and       '
decomposition by-products (USEPA, 1988b).  Unfortunately,
lysimeters cannot be used to determine  the rate or absolute
amount of leachate moving through the soil.

    Using a liquid mass  balance to  identify leaks in some
situations may  be a  reasonable  backup method for detecting
surface impoundment  leaks (USEPA, 1984b).  It is not proposed as
a primary leak  detection method. The liquid mass balance process
for leak detection requires long-term measurements of waste  input
and environmental parameters.  Leaks will not be detected in a
period of days, weeks, or possibly  even months, unless the leak
is fairly large.
                                67

-------
lysimeler manhole

geomembrane


?*

r— — protective layer
— ^ ^_^_^_^_^
	 <>

| tysimpTfir

                                                    double liner
                                                 low-permeabilrty
                                                 soil
 Figure  30.  Lysiateter  for leak detection beneath  bottom liner.


     The liquid mass balance method is suited for covered surface
impoundments with known, or controllable input and output
parameters*  In open systems, where the parameters cannot be
controlled, it is considerably more difficult.  Nearly all
surface impoundments are open systems where evaporation,
precipitation, wind speed, and temperature have not been recorded
or have been measured only at low levels of accuracy.

     Perhaps the largest source of error in the liquid-mass
balance is the evaporation estimate.  Due to the many
uncertainties in applying the mass balance to an open impoundment
and the potentially large errors in the parameters, the mass
balance should not be considered as a primary leak detection
method for open impoundments (USEPA,1984b).
3.6  SURFACE WATER MAHAGEMEHT

     Diversion structures (berms and ditches) intercept and
redirect flow of surface water away from a surface impoundment
(Figure 31).  Surface impoundments, unless they are intended to
collect runoff, need to be hydrologically isolated from the
surrounding terrain to function properly.  Diversion structures
must be designed to prevent flow into the impoundment from at
least the runoff from a 25-year storm (USEPA, 1982a).

     Rather than moving water away from an impoundment, diversion
structures can also be used to divert runoff, where contaminated,
into a holding pond or surface impoundment.

     Runoff diversion is accomplished by constructing a berm of
moderately compacted soil or by excavating ditches to divert
water around the impoundment.  Excess material from constructing
the surface impoundment or from ditch excavation is often
suitable for constructing diversion berms.   Figure 32 shows three
different ditch and berm cross sections.

                                68

-------
                  surface water flow
                        direction \ \
               diversion ditch
               andberm
                                            ^secondary
                                               containment dike
      Figure 31.  Runoff diversion past surface iaqpoundnent.
     design flow depth —
                           freeboard
     design flow depth —
                           freeboard
     design flow depth
                          -freeboard
   Figure 32.  Typical  diversion ditch and ber» crois lections.
     Berms are embankments of earth or other suitable materials
constructed  to protect a surface impoundment from spreading
surface waters.  They are designed to provide short-tern
protection by intercepting storm runoff  and diverting the flow to
natural or manmade drainageways.  Sufficient freeboard should be
provided to  prevent overtopping of the benn by the design flood.
Berm sloughing can be reduced by using proper side slopes and
construction methods.  A benn should have a minimum top width of
4 feet  (120  cm) and a freeboard of 3.5 to 12 inches (9 to 30 cm)
(USIFA,1982a).

                                 69

-------
     Benns should be placed across the slope at an angle to the
horizontal so that water moving downslope is intercepted and
moved laterally around the impoundment at a low velocity to
minimize erosion.  Longer slopes may need more than one berm.

     A diversion ditch should be designed to accommodate flows
from a 25-year frequency storm (USEPA, 1982a), and must intercept
and convey the flows at non-erosive velocities.  Diversion
ditches are .sized based on the Mannings Formula (USEPA, 1982h):
        Q -

     where s

        Q -
        n »
        A »
        R *
        S -
              1.49
    n
        AR2/JS1/2
discharge in cfs (mVsec)
coefficient of roughness
cross-sectional area of channel in ft2
hydraulic radius of the channel in ft (m)
longitudinal slope of the channel in ft/ft (m/m)
(m2)
     Design and construction criteria to use in diversion ditch
design follow:

     »  Location should be based on outlet conditions,
        topography, soil type, slope length, and grade.

     *  Capacity should be sufficient to carry the peak discharge
        from a 25-year design storm using an assumed Mannings*
        coefficient (n).

     *  Channel should be trapezoidal, V-shaped, or parabolic,
        with side slopes no greater than 3H:1V.

     »  Flow velocity must not exceed the flow for the assumed
        Mannings coefficient  (n) during the storm period.

     The USDA (1979) provides information on maximum allowable
velocities for an unlined drainage ditch excavated in various
soils.  However, channels must be grass-lined or riprapped over a
geotextile to minimize erosion.  Riprapped channels are
appropriate for sites with heavy runoff volumes.  The maximum
allowable velocity for a grass-lined channel can also be found in
the SCS manual (USDA, 1979).  Channel shape may be based on
construction and maintenance equipment available (e.g., graders
are suitable for V-shaped channels).

     Erosion is a function of velocity, depth, and time.  The
runoff peaks for the site must be determined to find which of
these are key design parameters.  Manuals (e.g., Urban Drainage
and Flood Control District, 1979) can be used to find the design
flow depth of channels.
                                70

-------
3.7  CONTROLS FOR VOLATILE ORGAHIC COMPOUND (VOC)  EMISSIONS

    The loss of VOCs from hazardous waste surface impoundments
has been the focus of recent research.  Investigated methods for
reducing VOC emissions include complete enclosure, surface
barriers, wind diversion fences, and adsorbents (!PA, 1985e).

    Complete enclosure of the impoundment involves the use of an
air-supported structure, or an inflated membrane,  covering the
entire impoundment (USEPA, 1985e).  The membrane would be sealed
around the edges, and the vapors would be bled off for treatment
or incineration.  This method could be expected to be nearly 100%
effective in preventing the escape of VOCs.  Compatibility of the
vapors with the synthetic cover material must be considered.

    Isolating volatile liquids from the air using a surface
barrier involves floating a cover of foam-like material, oil,
geomembrane, or other material, on the liquid surface.  The
objective is to eliminate, or at least reduce, the exposed liquid
surface area, thereby reducing VOC emissions.  Among the many
potential problems associated with this method are dislocation of
the surface barrier by wind and wave action, thereby uncovering
the waste.  In the case of an oil barrier, VOC emission from the
oil itself may occur.  Valsaraj et al (1985), in laboratory
simulations, examined the ability of a mineral oil layer in
controlling emissions of acetone, n-propanol, ether, and ben2ene.
It was much more effective in controlling the first two.

    If a floating geomembrane is to be used, its compatibility
with the waste liquid and vapors must first be considered.

    Winddiversion fences have been researched in laboratory
simulations by Thibodeaux et al  (1985).  Study results indicate
that various arrangements of perimeter and network grid fences
can reduce VOC emissions up to 80 percent (Figure 33).  The
research is based on the premise that reducing the flow of air
across a surface impoundment will cause a subsequent reduction in
VOC emissions.  This is because the volatilization process which
increase with wind speed.  (Wind fences are ineffective at zero
wind speed.)  Because the resistance on the liquid side of the
interface is typically much greater than the resistance of the
gas side, the liquid controls the volatilization process.

    Adsorbents  (e.g., activated  carbon) appear to be the  least
attractive of the methods.  They are expensive and must be
replaced or regenerated.  The use of adsorbents is based  on the
adsorbent "trapping" a VOC, which is then no longer available to
volatilize.  Increasing the amount of adsorbents in the
impoundment reduces the effective concentration of VOCs in the^
depends on chemical gradients across the liquid/air interface,

                                71

-------
                        perimeter wind fence
                       network wind fence

             I   I   I   I  I  i   I  I  II  III
       Figure 33.
Wind diversion fences for VOC control.
(from Thibodeaux  et al,  1985)
liquid phase, thereby decreasing the emission rate.  Further
information on this technique is available in Cudahy and Sandifer
(1980).

    It is unlikely that these methods will be perfected to a
point that they can reduce VOC emissions from surface
impoundments by 100%.  Therefore, an alternate procedure to
reduce the VOCs reaching the impoundment (e.g., by pretreating
the waste stream) may become important if regulatory guidelines
are adopted that impose stringent VOC emission standards.  At the
least, VOC monitoring is likely to become a requirement for
surface impoundments handling hazardous wastes.  Waste-stream
pretreatment is a design alternative which should be considered.

    Waste pretreatment methods to control VOC emissions may
include stripping the waste stream with air or steam, using
adsorbents (i.e., activated carbon) or chemical oxidation.  Each
system has its advantages and disadvantages, and none is best
suited for use in all situations.  Information on stripping
methods, adsorption technology, and chemical oxidation is
available in the literature (USEPA, 1984c and 1984d).

    Much of the research into methods to reduce VOC  emissions
has resulted in models that attempt to estimate VOC  losses.
Unfortunately, these models involve complex mathematical
equations and are not easily adapted for practical use.  One
exception is a procedure suggested by Shen  (1982), which provides
practical engineering solutions and offers a simple  technique for
estimating the volatilization rate of VOCs from waste lagoons.
Shen emphasizes that this is an estimate and should  only be used
to evaluate the potential for VOC emissions.
                                72

-------
     Other methods and.related information available for
estimating VOC emissions include the following:   USEPA (1985c and
1985f), Mackay and Yeun (1983), and Thibodeaux et al. (1982).


3.8  CONSTRUCTION QUALITY ASSURANCE (CQA) PLAN

     A Construction Quality Assurance Plan should be developed as
part of the surface impoundment design process.   The plan
describes the programs and principles employed to control quality
of construction.  Quality Assurance (QA) principles include the
following:

     •  sufficient autonomy and authority for the designated QA
        personnel or QA/Quality Control (QC)  organization;

     •  establishing procedures to control activities affecting
        construction quality;

     •  documenting evidence that these activities are performed
        satisfactorily; and

     •  systems to identify, correct, and report nonconforming
        items.

     USEPA (1986a) has provided guidance for preparing a CQA plan
to ensure, with a reasonable degree of certainty, that completed
disposal facilities meet or exceed design criteria, plans, and
specificatipns.  The guidance covers CQA for the following
structural components of surface impoundments:

        foundations
        dikes
        low-permeability soil liners
        geomembranes
        geotextiles
        leak collection systems
        final cover systems

     CQA activities must be coordinated with construction
activities to ensure that specifications and performance
standards are achieved.  This coordination requires that the
contractors establish and modify on a routine basis a schedule of
construction activities.  An accurate schedule is essential if
CQA goals are to be achieved.
                                73

-------
                            CHAPTER 4

                          CONSTRUCTIOH


     The surface impoundment construction phase includes the
following tasks:  site preparation; excavating and preparing
foundations and cut-slopes;  constructing dikes, soil liners, and
secondary containment structures; installing geomembrane liners;
constructing leak collection systems; placing protective
coverings over dikes and liners; and installing liquid level
control systems.  This section discusses each of these tasks.


4.1  SITE PREPARATION

     The surface impoundment site is prepared for construction by
clearing trees, vegetation,  topsoil, stones, and unwanted
structures.  Support facilities for work crews and equipment are
then installed.  They include a management office, field
laboratory, and equipment storage and maintenance areas.  Safety
and spill containment devices are installed.  Finally, borrow
areas are cleared and established as sources of material for
constructing dikes, liners,  and other earthen structures.

     Surface drainage at the site should be diverted or otherwise
controlled to prevent runon to the borrow and construction areas.
Runon can adversely impact the workability of clay soils and the
time required to complete construction.

     Attention to site security is generally required to prevent
theft, vandalism, or unauthorized entry.  The degree of security
depends on the type and amount of equipment and materials stored,
the amount of time personnel are off-site, the remoteness of the
site, and other site-specific factors.  Measures that may be
taken include the installation of fencing and locked storage
facilities, and providing security personnel.


4.2  CUT-SLOPE AHD DIKE FOUHDATIOT CONSTRUCTIOH

     Dikes and cut-slopes are constructed so that they form the
continuous interior slopes of a surface impoundment (Figure 34).
Dikes are situated above grade and are constructed of compacted
fill material.  Cut-slopes are excavation side slopes cut into
the native soil.  The final interior surface of dikes and cut-
slopes, along with the excavated bottom surface, will act as a

                                74

-------
                    geomembrane liners
                    and protective
                    geotextites
           protective soB layer
         teak collection
         layer
                                 cut-stope
                           tow-perme ability
                           soS layer
     Figure 34.  Cut-slope, dike, and side-wall cross  section.


uniform foundation for the impoundment liner, and, therefore,
must have constructed strength and eompressive properties
compatible with foundation support requirements.


4,2.1  Cut-slopes

     Cut-slopes are required where all or part of the  impoundment
is to be below grade.  They are constructed by removing the soil
with standard earth-moving equipment.  After the cut forming the
sidewall is completed, the cut-slope may require moisture content
adjustment and/or compaction to bring the soil to desired
specifications.  If the moisture content is too high,  simply
allowing the slope to dry may be adequate.  The cut-slope should
not be exposed to rain unless runoff will not adversely affect
the foundation characteristics of the slope soil.

     Some sensitive soils can actually lose strength as a result
of attempted compaction.  If the design engineer considers
compaction appropriate, it can be performed using the  same
equipment and procedures used to compact dikes and soil liners.
Most equipment can adequately compact side-wall slopes up to
3H:1V.  Steeper slopes or wet conditions may require that towed
compactors be used, or even that towing assistance be  provided
for self-propelled equipment.  Towing may require another piece
of heavy equipment or a winch positioned at the top of the slope.

     The steepness of the side slope also affects the  soil
compactive effort.  As the slope increases, the effect of the
compacting weight applied to the soil decreases due to the
increasing force vector directed down the slope  (see Figure 35).

                                75

-------
                               W = weight of compactor = 2000 Ibs (90i kg)
                               E = effective compacting
                                   weight = 1414 Ibs (644 kg)
                               P = weight parallel to
                                   siope = 1414 Ibs (644 kg)
          effective compaction on 45°slope
                               W = 2000 Ibs (909 kg)
                               E = 1897 Ibs (862 kg)
                               P = 63S Ibs (289 kg)
          effective compaction on 18.5 °slope
         Figure 35.   Idealized schematic showing effects
                      of slope on compactive effort.
This effect is overcome by increasing the number  of  equipment
passes required to  reach the specified soil density.

     The impoundment bottom is also a cut surface that should
possess the same characteristics as the side  slopes  when
completed.  Host of the same considerations apply.
4.2.2  Dike Foundation

     The dike  foundation is the base upon which  the dike is con-
structed.  The condition of this foundation  is as  important to
dike stability as the strength of the dike material itself.  As
impoundment construction begins, the foundation  area is cleared
and the topsoil  is removed.

     If the foundation soil consists of  soft or  sensitive fine-
grained soil,  final excavation to subgrade elevation should be
done using a straight-edged excavator bucket (no teeth) to avoid
remolding of the subgrade soils and the  resulting  strength loss.
If zones of weak or undesirable soil are suspected, proof-rolling
may verify their existence.  Proof-rolling consists of passing a
piece of heavy compaction equipment, such as a pneumatic roller,

                                 76

-------
over the soil.  Poor or weak areas will be revealed by the
development of rutting or ground heaving.  If weak areas are
exposed, poor soils should be removed and the area filled and
recompacted with acceptable material.

     Scarification of the foundation soil may be required to
provide the soil with an adequate bonding surface for placing the
initial soil lift during dike construction (Creager et al, 1945 j
Hammer and Blackburn, 1977).  Scarification is accomplished using
a ripper attachment on a bulldozer, disc harrow, or similar
implement.  Foundations composed of firm soil should be easily
scarified, while scarifying soft soils may be less successful.
The correct combination of moisture content, density, and
scarification should promote bonding between the foundation soil
and the initial lift of dike soil by facilitating intermixing of
the two materials during compaction.  This action will minimize
horizontal permeability at the foundation-dike interface.


4.3  DIKE AND SOIL LINER CONSTRUCTION

     The construction of dikes and the construction of soil
liners have many similarities.  The following discussion
addresses them together where appropriate.


4.3.1  General Construetion	Process

     Dikes and low~permeability soil liners are constructed of
soil which is placed in lifts and compacted at a specified
moisture content to a specified density.  Successive lifts are
placed and compacted until the design height or total thickness
is achieved.  Standard recommended lift thicknesses are 9 inches
(23 cm) for loose liftsj and 6 inches  (15 cm) for compacted
lifts, when compacting with penetrating-foot rollers on cohesive
soils  (Church, 1981; Creager et al., 1945; Hammer and Blackburn;
1977).  The general sequence for dike construction is as  follows:

      (1)  Scarify and adjust moisture content (if required) of
          the surface on which the lift is to be compacted; first
          lift placed on foundation soil.

      (2)  Place the loose soil over the surface and level to  the
          appropriate thickness.

      (3)  If the soil has not been processed in the borrow area
          or if further processing is required, process before
          compaction.

      (4)  Compact the soil to the specified density using an
          appropriate number of passes with the compaction
          equipment.

                                77

-------
     (5)  Repeat the above steps until a sufficient number of
          lifts has been placed.

     A crest should be maintained along the dike centerline to
promote surface drainage (Figure 36).
                                                  final dike slope
           foundation soil       \       X     foundation soil
   Figure 36.   Dike cross  section showing lifts  and  final  slope
              cuts.


     Proper soil compaction should provide the dike with
sufficient strength and sufficiently low permeability to perform
satisfactorily.  Further information on soil compaction is given
in Section 4.3.2.1.3, as compaction procedures for soil liners
and dikes are essentially identical.  CQA is critical to properly
constructing the dike and other surface impoundment components.
CQA monitors the construction process and ensures that design
specifications are achieved.  The elements of the CQA Plan are
discussed in Section 3.11.

     Constructing a low-permeability soil liner involves many of
the same procedures as dike construction:  preparing the soil for
compaction, placing the material in loose lifts, compacting the
soil to the required density and thickness, and testing to ensure
that specifications are met.  The soil-liner construction process
has been described by USEPA (1986b) in its Technical Resource
Document addressing clay liners.

     Technical guidance provided by USEPA (1986a) on construction
quality assurance recommends constructing a representative test
fill before constructing the full-scale soil liner to verify that
performance standards can be consistently achieved using
specified materials and equipment.  A test fill is also
recommended as a prerequisite to dike construction.  Potential
scheduling delays during dike and liner construction due to
compaction or material problems can be avoided  if a test fill  is
constructed and problems corrected during the surface impoundment
design phase.  The test fill should be constructed using the same
soil, equipment, procedures, and specifications to be used in
constructing the full-scale facility.                          —


                                78

-------
     Strict control of the test fill construction process and
documentation of the test fill data are required for useful
results.  The test fill data can be compared with data obtained
during impoundment construction to indicate acceptable criteria.
USEPA (1986a) provides guidance on constructing and utilizing
test fills.
4.3.2  Pro-placementSoil Preparation

     Preparing or processing soil materials prior to soil liner
or dike construction improves compaction efficiency.  Processing
activities include reducing clod size, removing unwanted
material, and adjusting moisture content.  Removing unwanted
material and adjusting moisture content are often performed in
the borrow areas, while clod size reduction may be performed in
the construction area.  Bomogenization thoroughly mixes the soil
to dilute small quantities of undesirable materials (e.g., sand
and gravel) which tend to increase the hydraulic conductivity of
soil liners.  Clod-size reduction is best accomplished at water
contents at or near the plastic limit-  At low water contents,
the clod strength is often highj at high water contents, the soil
may become sticky and have reduced workability*

     Mechanical equipment, (e.g., rototillers, pugmills, and
pulverizers) may be used to reduce clod size, homogenize the
soil, and mix amendments or moisture prior to soil placement and
compaction.  These processes may also be accomplished using a
disc, blade, or other implement towed by a tractor or dozer.
Proper soil processing enhances compaction and uniformity of
desired liner or dike properties.

     Moisture-content adjustments are .often required prior to
soil compaction.  The moisture content is usually specified with
respect to optimum, as determined from standard Proctor (ASTM D
698-78) or modified Proctor (ASTM D 1557-78) compaction tests
(ASTM, 1986c).  A specific acceptable moisture content range
(e.g., 1% less to 3% greater than optimum for liner construction;
drier, perhaps up to 2% less than optimum, for dike construction)
is generally stated in the design specifications.  Moisture is
either added or removed to maintain the recommended moisture
content.  Moisture reduction is normally performed by air-drying
the soil or by adding and mixing drier soil.  Moisture addition
is usually performed using water trucks or sprinklers to add
water to the soil.  Dozers, disk harrows, or rototillers are then
used to mix and incorporate the water into the soil.  The
preferred procedure suggests that water content be adjusted in
the storage or borrow area before the soil is placed for
compaction.

     Moisture may also be added or removed during construction,
or between specific activities, as required.  For example, slight

                                79

-------
desiccation of the surface of a compacted soil lift may occur due
to delays, and CQA may require that moisture be added before
continuing construction.  Moisture is added or removed in the
same or similar manner as in the borrow area before placement for
compaction.  The moisture content of the soil during and after
compaction is monitored according to CQA procedures using
equipment such as a neutron moisture-density probe.  Actions
should be taken to prevent or minimize moisture changes due to
construction delays or other factors.  For example, a temporary
protective cover of soil may be applied to reduce evaporation, or
the surface may be sealed (rolled smooth) to hasten runoff and
lessen the potential for moisture addition from rainfall.


4.3.3  Soil Material Placement

     A survey of liner construction sites and personnel by
llsbury et al (1985) indicates that soils are generally placed in
parallel strips over the liner area using earth movers or trucks,
and then leveled with bladed equipment to the desired loose lift
thickness.  Strips may be stepped up the side slopes in liner
construction in order to maintain a horizontal working surface
and facilitate compaction.

     Control of lift thickness is important because it influences
compaction effectiveness and the resulting hydraulic conductivity
of the compacted soil liner.  Horizontal and vertical survey
control points should be established to inspect the areal extent
and thickness of liner lifts throughout construction.  Loose lift
thickness is commonly specified at 9 inches (23 cm), which
results in a compacted lift of approximately 6 inches (15 cm),
whether a dike or a liner is being constructed.  However,
compacted lift thickness depends on soil, equipment, and
operating characteristics (Church, 1981j Creager et al, 1945j
Elsbury et al, 1985? Hammer and Blackburn, 1977).


4.3.4  Soil. Compaction  -

     Effective compaction imparts strength to dikes, foundations,
and other earth structures, and reduces the permeability of the
low-permeability soil component of the liner.  A proper
compaction procedure densities the loose lifts and bonds the
upper lift to the underlying lift.  Bonding of layers is
essential for eliminating zones of higher horizontal hydraulic
conductivities at the lift interfaces.  Bonding is achieved by
-scarifying (roughening) the upper surface of a compacted lift
before placing and compacting the next lift, and by compaction
equipment that will penetrate through the loose lift into the
compacted lift and a zone of intermixed material between the two.
                                80

-------
     The critical performance standard for surface impoundment
soil liner components is low hydraulic conductivity.  This
characteristic is obtained by efficient compaction at specified
water contents, which are generally wetter than optimum (e.g., 1
to 3 percent, dependent on soil type and compactive effort).
Preferred compaction induces a homogeneous dispersed or
unoriented structure within the soil material which, in turn,
provides a lower hydraulic conductivity.

     Stability, rather than permeability, is the prime concern
for surface impoundment dikes.  Thus, dike construction strives
for soil strength, which is accomplished by compacting dry of
optimum to maximum density.

     After a number of passes are completed, in-situ moisture and
density measurements should be made with a moisture-density probe
or similar equipment, according to CQA requirements.  Compaction
should continue if measurements show that density specifications
have not been achieved.  In-place compaction density is normally
required to be in the range of 90 to 95 percent of standard
Proctor maximum density obtained in laboratory testing.

     Soil compaction must be accomplished with equipment that is
best suited for the particular soil type and situation.  Several
equipment types are available, including the following:

     (1)  Penetrating-foot rollers:
          • sheepsfoot
          • pad foot
          • tamping foot
     (2)  Pneumatic or rubber-tired rollers
     (3)  smooth wheel or steel drum rollers
     (4)  grid rollers

     Compaction equipment is manufactured in both towed and self-
propelled models.  The better-suited model will be based on soil,
site, and design characteristics, and equipment model
availability.  Compactor choice depends on site-specific
compaction needs and on soil grain sizes.  Dikes and liners are
most often compacted with penetrating-foot rollers  (Figure 37).
Recommendations are provided by Holtz and Kovacs (1981).
Equipment can often compact satisfactorily outside the range of
prescribed use.  Additional information on applicability of
compaction equipment is given by USEPA  (1988b and 1989b).  (1975).

     Several older but still useful references discuss compaction
with penetrating-foot or kneading compactive force  (Bilf,  1975;
Johnson and Sallberg, 1960; Lambe, 1958; Mitchell et al, 1965;
and Sowers and Gulliver, 1955).  The theory of compaction  of
cohesive soils was developed by R. R. Proctor and presented in a
series of articles published in 1933 (Proctor, 1933ajb;c;  and d).


                                81

-------
             x--	-^
                      r—^
                      K,	^A  sheepsfoot
                      - "/^T-"--— i  —*"r
      pad foot
               Figure 37.  Compactor foot designs,
     Steel-wheel (also known as steel drum) rollers are effective
in final preparation of the low-permeability soil surface for
installing the geomembrane liner.  The equipment provides the
soil with a smooth, even surface, free of clods that can damage
the geomembrane.  The smoothest possible surface is required for
installation of the overlying geomembrane in a composite liner
(Giroud and Bonaparte, 1989).


4.4  GEOMEMBRAHE LIBER INSTALLATIOH

     Geomembrane liner installation involves placing and seaming
panels of the synthetic material over a prepared subgrade to form
a barrier against liquid migration.  The geomembrane must be
inspected and tested throughout installation to ensure that the
seams and panels are free of wrinkles, blemishes, holes,
inadequate seams or other defects which may allow the escape of
liquid.  Workmanship, experience of installers, unstinting care
in installation, and a commitment to quality control are all
critical to a successful, leakproof installation.  Giroud (1985)
and USEPA (1989b) present information on constructing liners with
geomembranes.  All facets of geomembrane selection, testing, and
installation have been comprehensively addressed by USEPA
(1988a).  More specific information is being prepared by USEPA on
the important topic of geomembrane seaming; the first of these
addresses the seaming of polyethylene geomembrane (USEPA, 1989c).
                                82

-------
4.4.1  Storage of Materials and Eouipaent
     Geomembrane materials and installation equipment delivered
to the site should be stored in a secure location to protect them
from vandalism and theft.  An existing secured area or a
temporary storage area can be used to provide this protection.
The impoundment area should also be protected from intrusion and
tampering that could damage the geomembrane during construction.

     Provisions should be made for appropriate equipment to
unload and transfer the membrane rolls or panels.  The rolls are
heavy and may require special or modified equipment to move them
without damaging the material.

     Unless the geomembrane is applicable for exposed service, it
should be stored out of the sunlight to prevent ultraviolet
degradation and to minimize blocking.  Blocking occurs when liner
materials are heated, causing them to stick together.  The
material may then tear when unrolled on the subgrade.


4.4.2  Construction Quality Assurance/Inspection

     CQA inspection during installation is essential if the
geomembrane is to be an effective barrier against waste migration
to the underlying soils.  CQA assures planned review and tracking
of construction activities.

     The subgrade condition, geomembrane placement and seaming,
and sealing of penetrations through the liner require
considerable CQA.  A CQA Officer representing the surface
impoundment owner/operator is required to assure that
installation specifications and the contractual obligations of
the installing contractor are met.

     An adequate CQA program results in surface impoundments
being built as designed.  Guidance for preparing CQA programs is
given by USEPA  (1986a).  More information on CQA plan preparation
is provided in Section 3.11 of this document.


4.4.3  Subctrade Preparation

     Subgrade preparation provides a firm base for the
geomembrane liner.  In the DSEPA-recommended double-liner design,
the drain or leak collection layer is the subgrade for the
primary or top geomembrane, while the low-permeability soil
component is the subgrade for the geomembrane component of the
composite secondary liner.  Rocks, clods, or irregularities with
sharp edges should be eliminated from the finished subgrade by
fine-finishing  (rolling  smooth) the surface before installing the

                                83

-------
geomembrane.  Fine-finishing of sand layers is accomplished using
vibratory rollers and drags on a slightly wet surface.  Smooth-
wheel rollers may be required to finish the surface in both the
low-penneability-soil and sand subgrades.  During the fine-
finishing stage, grasses and other vegetation must be removed
from the subgrade layer, or vegetation killer applied, to prevent
their penetrating the geomembrane.  Care must be taken to attain
a smooth and flawless surface on the low-permeability soil layer
of the composite liner to facilitate continuous contact with the
geomembrane installed directly on it.

     Proper timing of construction activities is essential to
maintaining proper moisture content of the subgrade.  The
geomembrane should be placed on the finished subgrade soon after
the finishing  process is completed.  Uncovered, fine-finished
subgrades can be easily disturbed by rain or wind.  If rainfall
occurs during or after fine-finishing work on a slope, rills,
ruts, and ravines may be eroded into the surface. Sealing or
covering the fine-finished subgrade with a protective layer will
prevent soil erosion by surface runoff.  A protective covering is
also required to prevent moisture loss and desiccation of the
soil layer during extended dry periods.  The protective covering,.
if on the low-permeability soil layer, must be removed before
applying the geomembrane.


4.4.4  Geomembrane LinerPlacement

     The geomembrane liner is secured at the top of the
impoundment dike, usually in an anchor trench, extending around
the impoundment perimeter.  The trench should be excavated and
ready to receive the geomembrane before the panels are brought to
the installation location.  More information on geomembrane
anchoring is given in Section 4.4.5.

     The geomembrane panels will be placed in a pre-determined
pattern, an example of which is shown in Figure 38.  Placement
begins by arranging (or spotting) the geomembrane rolls or folded
panels at the top of the dike or on the impoundment floor.
Special or modified equipment may be required to move and locate
the geomembrane without damage.  The panels are unfolded or
unrolled down the side slope or across the floor.  The panels are
then placed in the proper position and secured with sandbags to
protect them from wind damage or displacement.  The geomembrane
should lie relatively flat and smooth after placement on a smooth
compacted subgrade.  Sufficient slack should be left in the
material to accommodate shrinkage due to temperature changes.
Depending on the characteristics of the geomembrane material,
shrinkage can occur by temperature increases and decreases.
Temperature increases can cause shrinkage by loss of volatiles
and by release of manufacturing stresses (Giroud and Peggs,
1990).  Shrinkage also occurs by contraction induced by lowering

                                84

-------
                       perimeter anchor trench
                toe of slope
      Figure 38.
Example of geomembrane liner panel layout
Other patterns may be used.
the temperature.  Temperature increases can also expand a
geomembrane" during installation, creating wrinkles where cracking-
can later occur.  It is a very important aspect of design and
construction to accommodate all shrinkage- and expansion-causing
factors so  that failure-inducing stresses, particularly at seams,
are minimized.

     The success or failure of a geomembrane installation depends
largely on  seam integrity and continuity.  USEPA (1983b) lists a
number of seaming methods and factors which affect field seaming,
not the least of which are the skill and experience of the
installer.  Seaming methods vary among geomembrane materials.
Common field seaming methods include the following:

     *  Thermal (hot air gun, hot wedge and dielectric)      J
     •  Heat (extrusion fillet or flat welding, and
        extrusion/wedge welding)                            f
     •  Chemical  (cement, solvents and vulcanizing adhesive) I
     •  Adhesive Tapes                                       :
                                                             I
     Recommended panel overlap, where the seam is located, varies
from 4 to 12 inches  (10 to 30 centimeters).  lowever, the
installation contractor must follow the surface impoundment
design specifications, which, in turn, should include the
manufacturer's recommended overlap and bonding systems.  The
following factors must be considered in constructing field seams
of high integrity:

     *  Manufacturers' guidelines for adhesives must be followed.
        The adhesive system must be compatible with the
        geomembrane  and be applied under acceptable ambient
        conditions.
                                85

-------
     *  Cold temperatures can prevent successful bonding of
        panels.  Many manufacturers recommend that adhesive
        bonding take place only when temperatures are above 15°C
        (59*F).

     •  Preparation of panel edges for seaming must be carefully
        done to prevent damage.  For example, grinding can create
        deep scratches that may result in  cracks (see USEPA,
        1989c).

     »  The seam surface should be clean and dry. The presence of
        moisture interferes with the curing and bonding
        characteristics of the adhesive, while the presence of
        dust creates voids that provide a  path for fluid
        migration through the seam.

     *  The geomembrane should rest on a dry, hard, and flat
        surface to facilitate applying pressure rollers.

     *  Panels  should be placed and seamed on the same day to
        minimize the risk of geomembrane damage by wind, and soil
        erosion under the geomembrane due  to rain; unfinished
        panels  should be anchored by bags  of sand.

     The finished seams should be free of  large wrinkles and the
surface should  be rolled flat.  Many manufacturers recommend that
field seaming begin at the panel center, lengthwise, and continue
to each seam end; this procedure minimizes the potential for
large wrinkles.

     As in soil liner compaction, proper placement of the
geomembrane on  the impoundment side slope  is essential to
successful liner construction.  Generally, the panels should be
of sufficient length so that, when placed  the field, seams run up
and down the side slopes, with no horizontal seams on the slopes
(Figure 38). This orientation reduces gravitational stress on
field seams, particularly desirable when seams have not yet
cured.  Corner  patterns should be cut to fit where necessary, and
in a way that minimizes horizontal seams on the slope.


4.4.5  Sealing  Around Structures and Anchoring the Geomembrane

     The geomembrane is anchored at the top of the dike in a
trench, V-shaped or U-shaped in cross section (Figure 39(a) and
(b)).  This technique is often recommended by manufacturers, due
to its simplicity and economy.  Excavating the anchor trench is
done using a trenching machine or a bulldozer blade tilted at an
angle.  The excavated soil should be spread away from the trench
and smoothed to facilitate unrolling and  spotting of panels.
                                86

-------
                              trench
         (a)
                                                geomembrane
                                                liner
                            trench
         (b)
                                               geomembrane
                                               liner
                       bolted anchor
         (c)
                                                geomembrane
                                                liner
                            anchor beam
      Figure  39,   Geomembrane anchor designs at top of dike.

     The panels should be anchored in the trench following  the
field-seaming operation unless other components  (e.g.,  synthetic
drainage layer) are also to be anchored in the trench.  Once the
seams are completed for an individual panel, the trench should be
backfilled with soil to anchor the panel.  The trench should not
be backfilled until the panels have been seamed, to  allow
positioning for optimum seaming.

     The geomembrane can also be anchored to concrete structures
along the benn top by securing the liner with batten strips
attached to anchor bolts embedded in the concrete  (Figure 39(c))»
This technique can also be applied to bonding the  geomembrane to
metal structures  (e.g., pipes).  A common method places the
anchor bolts on 6- to 12-inch (15- to 30-cm) centers. The
geomembrane is placed over the bolts, an adhesive  is applied to
the membrane, and the batten strip is secured and  bolted in
place.  Compatibility of the adhesive/sealant with the  synthetic
and the impounded liquid must be determined during impoundment
design to assure the integrity of the seal.  Rn  extruded polymer
strip may be cast in the concrete and the liner welded  to it as
an alternative to the batten strip and bolts.  Details  of
anchoring techniques are discussed by USEPA (1984d)  and Kays
(1987).

     Manufacturers' recommendations  for using  specific  materials
and procedures must be followed to establish an  effective seal
around penetrations through  the geomembrane.   Bonding  synthetic
materials  (e.g., the geomembrane and a plastic pipe) is typically
accomplished using solvents, adhesive, or welding  techniques.
Where components of different materials are to be  bonded, it must
                                87

-------
be verified before installation that the materials can be
successfully bonded using one of these techniques.  For example,
the thermal-expansion characteristics of each material must  be
evaluated to determine if temperature-driven expansion and
contraction cycles will permit an effective bond.

     Wherever possible, pipes, level control equipment, and  other
structures should be placed entirely above or beneath the
geomembrane to avoid penetration. If penetrating structures  are
included in the impoundment design, sealing the geomembrane
effectively around the structure is critical to liner integrity.
Standardized designs include pre-engineered seals installed  in
the plane of the geomembrane, and pre-fonned boots for
installations around penetrations (Figure 40).  If pipes are
installed in the impoundment through a concrete structure, the
seal can usually be made in the plane of the liner.
                     steel clamp

           geomembrane
.
mastic


 welds
                       boot seal at geomembrane liner
                      gasket
              geomembrane
                       flange seal at geomembrane liner
       Figur* 40.  Seals at geonmnbrane liner penetration*,
4.S  LEAK COLLECTIOH AHD REMOVAL SYSTEMS

     The leak  collection and removal system is designed to drain
and pump out liquids accumulating in the liner system.
Typically, the leak collection system consists of a granular
material (sand)  immediately overlying a hydraulic barrier
(liner).  The  system's ability to drain away moisture is enhanced
by constructing the system at a minimum bottom slope of 2
                                88

-------
percent, using highly permeable drainage media and engineered
geotextile filters, where needed, to prevent migration of fine-
grained soils into the media, and by spacing drain pipes
properly.

     The leak collection layer is normally constructed to extend
up the side slopes.  The development of synthetic drainage nets
(geonets) has resulted in many surface impoundments being
constructed with side-slope leak collection layers of synthetic
materials and bottom drainage layers of granular material and
drain pipes.  Synthetic drainage net material is used on side
slopes instead of granular systems because it is more easily
installed on steep side slopes.  Steep side slopes cause granular
drainage material to slump down, while the synthetic drainage
material tends to remain in place if anchored properly at the top
of the slope (refer to Figure 18).

     A typical leak collection and removal system is installed as
follows.  A layer of sand (about 2 inches [5 cm] thick) is spread
over the underlying layer (e.g., a geomembrane) as a protective
soil covering.  It should consist of material which is free of
clods, stones, or other sharp objects that can puncture the
geomembrane.  Optionally, a geotextile layer may be used.  The 2-
inch  (5-cm) protective layer will also provide bedding for the
drain pipes according to the design layout.  If a geotextile
protective layer is used, a bedding layer is placed only along
the pipe alignments for pipe support.  Typically, perforated
pipes of 4 to 6 inches (10 to 15 cm) in diameter are used.  Pipe
perforations are placed face down to prevent clogging by the
drainage media.  After the pipes are placed, the remaining
granular material is spread over the area in a single loose lift
to the required thickness, and compacted with a vibratory roller
into a firm base for the primary geomembrane liner.  If a
gravelly drainage layer is used, a geotextile protective layer
should be placed over the surface prior to placing the primary
geomembrane.  Figure 41 is an example of a leak collection system
layout.

     If synthetic drainage materials (e.g., geonet and
geotextiles) are used, they should be unrolled and spotted as in
geomembrane installation, except that the panels are not
overlapped and seamed.  The panels are placed edge-to-edge and
connected according to the manufacturer's suggested procedures,
so that the lower portion of the side-slope panel extends into
the granular or other bottom layer to enhance continuity between
the drain layers  (refer to Figure 18).  A geotextile is placed on
both  sides of the drain panels to prevent intrusion of the
geomembranes due to compression or creep phenomena  (Figure 42).
The synthetic drain system is then secured in the anchor trench
as in the geomembrane liner installation.
                                89

-------
                            ^r^r€^>
           perforated drain pipes
granular leak collection layer
        Figure 41.   Exaapl* leak collection syatan layout.
                                           geomembrane
                                           liners
            geotextiles
                                               geotextiles
                     leak collection sump
     Figur* 42.  0«o>3fntli«tic  u««« in leak collection layer.
4.6  TESTIBG THE LIMKR SYSTEM

     After the liner  system has been installed, it should be
tested for leaks due  to pin holes,  inadequate seams, and
punctures.  Methods for liner testing prior to operational start-
up include filling the impoundment  with water and monitoring the
leak collection system for  liquid.   This method or procedures
that use in-situ and  remote (e.g.,  electrical resistivity) leak
detection techniques  permit the identification and repair of
leaks in the primary  liner  before waste is placed in the surface
impoundment.  Unless  a leak detection system has also been
                                90

-------
installed below the secondary liner, the integrity of the
secondary liner cannot be tested.


4.7  PROTECTIVE COVERIHGS

     As discussed in Section 3.7, and shown there in Figure 21,
protective coverings are constructed or installed as preventive
measures against erosion, weathering, or other degradation of the
impoundment's structural components by mechanical and
environmental forces.  Coverings may consist of natural material
(e.g., soil, stone, and vegetation) or synthetic materials-
(geotextiles). The following discussion covers construction and
installation of protective coverings for liner systems and dikes.


4.7.1  Liner Protection

     Protection of the primary liner is often provided by a soil
cover of sufficient thickness to prevent environmental or
mechanical damage.  If the liner is a geomembrane, placement of
the soil cover must be accomplished so that the synthetic
material is not damaged.  Geotextile bedding may be placed
between the soil and geomembrane (Figure 21).  The soil cover
should be placed using low-contact-pressure equipment (to avoid
puncturing the liner) in a thickness of no less than 18 inches
(45 cm).  Though not necessary, placing the soil at or near
optimum moisture allows slight soil compaction by equipment
traffic.  The cover soil material must be free of clods, stones,
or other sharp objects which can puncture the geomembrane.

     Synthetic materials may also be used in some cases for liner
protection instead of soil; for example, the liner may be covered
with a geotextile for short periods of time.  The synthetics must
be placed according to manufacturers' suggested methods, and the
engineering plans and specifications for the facility.  The
synthetic materials are generally less thick than a soil cover
and, therefore, may be layered (e.g., several layers of
geotextile, drainage layer, or both) to provide a thickness which
will protect the liner from weathering or puncture.  As noted
earlier, a reduced-thickness soil protective layer may be placed
on the geotextile for added protection.


4.7.2  PikeProtection

     Dike protection consists primarily of vegetation and/or
riprap  (Figure 21).  Establishing a vegetative cover on the dike
exterior and top surfaces is typically accomplished by
hydroseeding, broadcasting, grass or grain drills, blowers, or
hand planting.  The best-suited  seeding method depends on
topography, type of vegetation,  soil condition, and equipment

                                91

-------
availability (Lee et al., 1985).  The soil usually requires
physical and chemical treatment before vegetation can be
established.  Physical treatment includes tilling and adding
topsoil.  Chemical treatment includes applying fertilizer or
lime.  After seeding, the soil should be watered and mulch
applied to prevent erosion until the vegetation is established.
Seeding the dike slope may also be accomplished by placing
degradable nets on or just below the soil surface to provide soil
stability until the vegetation becomes established.

     Riprap may be used as dike protection on the inside slope or
on both sides (Figure 21).  Riprap placement is discussed by the
Bureau of Public Roads (1967) and USAGE (1977),  Riprap should be
placed over a geotextile filter selected to prevent washout of
the underlying protective soil layer.  The stone may be placed
using a dragline or similar equipment, or dumped on the dike
slope by transport vehicles, spread into place, and compacted
with a grid roller if required.  Riprap should not be placed
directly on a geomembrane, but should be separated from it by a
buffer material of soil as in Figure 21.  Other materials that
may be used for dike protection include synthetics or geogrids,
concrete, and asphalt.  These should be placed according to
manufacturers' accepted methods, and the facility engineering
plans and specifications.


4.8  LIQUID LEVEL CONTROL SYSTEMS

     Active liquid level controls are installed equipment that
control daily level changes within the impoundment.  Passive
controls are engineered and constructed into the dike.  Passive
controls function in emergencies to prevent liquid overtopping
and catastrophic failure.


4.8.1  Active Liquid LevelControl

     Construction associated with active liquid level control
systems consists primarily of erecting mounting structures  (e.g.,
piers or pilings) needed to suspend a sensor over the liquid
surface.  The preferred arrangement is to locate the mounting
structure outside the impoundment and extend it over the liquid
surface  (see Figures 24 and 25).  This arrangement has no contact
with liner-support components and no liner perforations as
potential leak sources.  An alternative design is to construct a
mounting structure that rests on the impoundment bottom, but does
not penetrate or abrade the liner.  Both approaches preserve
liner system integrity and are preferable to mounting structures
that penetrate the liner.  If the design requires that the
structure penetrate the liner, QC should be diligently exercised
to maintain the integrity of the seal at the liner penetration.


                                92

-------
     Regardless of the mounting structure selected, installing
the liquid level sensing system properly, according to the
manufacturer's specifications, is essential. The system will be
integrated with facility inflow and/or outflow devices,
monitoring devices, and alarms to control the liquid level in the
impoundment.  The installed system should be tested to verify
that its components function properly.  Testing should include
deliberate attempts to test the fail-safe aspects of the system.
A record of testing procedures should be maintained for reference
during routine system maintenance.


4.8.2Passive Liquid Level Control

     Passive outfall structures allow the release of liquid when
the freeboard level is exceeded.  Release is accomplished by one
of two general means: (1) through an opening at the dike crest
(e.g., weir, spillway, or flume); or (2) through a conduit (e.g.,
pipe) extending through the dike and liner system  (see Figures 22
and 23).  Generally, outfalls provided by an opening in the dike
crest are preferred for a surface impoundment because no liner
system penetration is required.

     The primary concern associated with constructing spillways,
weirs, flumes, or conduits is ensuring an effective seal between
the outfall structure and the liner.  The surface impoundment
liner system generally includes synthetic materials which require
bonding to the outfall.  The outfall may possibly be constructed
from the same synthetic material used to line the impoundment,
thereby minimizing the bonding problem.  lowever, the material
used for the outfall will often differ and a special bonding
procedure will be required.


4.9  SECONDARY CONTAINMENT

     The secondary containment structure for a surface
impoundment is typically a low dike or berm surrounding the
facility (Figure 25).  The secondary containment berm is
constructed in the same manner as the primary dike and consists
of compacted lifts of soil.  Given the inherent stability of its
typically low design, a secondary containment dike can often be
compacted effectively with several passes of transportation or
dozer equipment.  However, field density tests should be
conducted to determine that compaction using -this method is
adequate.

     The surface of a secondary dike should not require armored
protection  from erosion, unless it is in an area of potentially
rapid-flowing flood waters.  Vegetation  should be  adequate to the
task, similar to vegetation on the primary dike, as discussed in
Sections 4.7.2 and 3.2.3.2.4.

                                93

-------
                            CHAPTER 5

              OPERATION, MAINTENANCE, AMD MOHITORIHG


    This section presents guidance on routine operation and
monitoring of surface impoundments.  Host aspects are similar for
all surface impoundments.
    I
    I
5.1  OPERATIOH AMD MAIHTEHAHCE ACTIVITIES

    The information in this section emphasizes current industry
practice for operating and maintaining surface impoundments for
waste treatment, storage, and disposal.  The discussions expand
upon previous work (e.g., USEPA, 1983a and 1984a).


5.1.1  Facility Start-no

    The start-up procedures for a new surface impoundment should
lay the framework for future facility inspections.  The initial
inspection should document baseline conditions against which
future operating decisions are made.

    A dormant period usually occurs between the time that
construction is completed and the time that final approval for
use is obtained from regulatory agencies.  During this period,
several "dry-run" inspections can be made by the operating
personnel to perfect procedures, to train personnel conducting
the inspections, and to become familiar with various structures
and equipment before use.  Before the surface impoundment is
placed in service, structures and equipment should be maintained
and protected from weather damage.  Storms present the potential
for erosion of dikes and side walls, and to cause flow, level,
and volume gauges to operate.  These components must be checked
and maintained until the facility is ready to accept waste.  A
small pool of water should be maintained in the impoundment to
prevent the protective soil layer covering the liner from drying
out.  Dike side slopes and top surfaces should be watered
regularly to prevent desiccation and to facilitate the initial
growth of newly planted vegetation.

    A final inspection of the entire facility must be made
before start-up.  If inflow will enter the impoundment through
pipes, all piping, pumps, valves, controls, gauges, and manual

                                94

-------
controls must be inspected and shown to be in working order
before they are used to accept waste.  If discharge to the
surface impoundment is directly from tank trucks, the discharge
area or ramp must be thoroughly inspected before use.  Emergency
discharge pipes or spillways must also be inspected to ensure
that they are free of debris and ready for use.   These items
should be covered in the CQA Plan discussed in Section 3.11.


5.1.2  Routine

     The RCRA Part B permit application describes the frequency
and detail of inspection for each hazardous waste surface
impoundment system component.  When conducting an inspection,, the
inspector should document the nature and extent of problems
noted.  The inspector should follow up the inspection with a
report noting remedial actions taken to correct deficiencies, the
time required for correction, special problems encountered, and
suggestions for preventing recurring problems.  Inspections can
be expected to be conducted periodically by the regulatory agency
and more frequently by the impoundment's operating personnel.


5.1.2.1  Regulatory Inspections of Facility —

     Agencies responsible for inspecting surface impoundments
usually have an inspection checklist which emphasizes those areas
that serve as "trigger" points? that is, a problem noted in a
particular area which usually reflects the existence of a larger
problem, which may be less apparent, and triggers a more in-depth
investigation.  Prudent facility operators will include a similar
checklist for routine inspections so that problems can be
corrected before more troublesome conditions develop.

     Particular questions to be answered during routine
inspections include, but are not limited to, the following:

*    Does the surface impoundment appear to provide at least the
     design freeboard?

        To determine this for a nearly full impoundment, the
        inspector should look for erosion on protective soil
        cover due to wave action.  On geomembranes, the inspector
        should look for lines of algae growth, salt deposits, and
        oil and grease marks.  If evidence exists of erosion, or
        markings are evident on the geomembrane, less than the
        design freeboard vertically below the lowest point on the
        dike summit  (e.g., a spillway), then adequate freeboard
        is not being maintained.  The inspector should also
        examine outfall structures for evidence of overtopping.
        The inspector should ask for records of impoundment
        inflow and discharge volumes and compare these to level-

                                95

-------
   recorder data to obtain an impoundment volume estimate
   based on a simple mass balance*   Large discrepancies in
   the balance volume estimate may  suggest possible leakage
   and should be followed by a more detailed evaluation,
   including checks for possible leaks.

Do the dike's crest, side slopes, abutment, and toe areas
have protective cover (e.g., rock or grass) to minimize wind
and water erosion?

   In addition to inspecting these  components for signs of
   erosion, the inspector should look for dead vegetation,
   oil stains, salt or mineral deposits,  seep areas,
   slumping, irregular settling, sink holes, slides,
   sloughs, bulges, cracks, animal  burrows, roots of shrubs
   or trees, and other visible evidence  of problems.
   Although some of these items may reflect merely the need
   for simple maintenance, most indicate the potential for
   leakage through the dike.

Is the emergency spillway or overflow structure clean?

   The inspector should look for animal  nests, collections
   of leaves or other detritus that could impede flow, and
   signs of an overflow (e.g., oil  and rust stains).

Do the emergency (manual) shutoff valves on the influent
line work properly?

   The inspector should determine if these valves can be
   shut off quickly in an emergency.

Does the emergency and safety equipment  work?  Is it clearly
labeled and accessible?

   This is normally a fairly quick check and is an
   indication of the impoundment management's attitude
   toward overall safety and emergency preparedness.

Is the leak collection/detection system  operational?

   Pumps and detection devices should be checked to assure
   that they function.  Routine operational records should
   be examined for frequency of operation and operational
   results.  Sumps should be checked for liquid level.

Are gecstembrane liner problems apparent?

   If the geomembrane liner is visible (no protective soil
   cover), the inspector should look for signs of abrasion
   or tearing along the top edge due to vehicular traffic or
   other activities.  The inspector must also determine  if

                           §6

-------
        w
         whales" are present; that is, gasses forming under the
        geomembrane which cause it to rise.  If the liquid is
        sufficiently clear, the inspector should look for ripples
        on the impoundment bottom, or stretch folds along the
        sfde slopes.  This unevenness is frequently accompanied
        by liquid beneath the geomembrane.  Such liquids should
        generally be intercepted by the leak collection/detection
        system.  Visible seams and geomembrane anchoring should
        be checked for signs of weakness.


5.1.2.2  Operator Inspections of Dike Slopes, Faces, and Crest —

     The dike system should be inspected by operating personnel
weekly and after rain, ice, or wind storms for the following:

     •  liquid level exceeding the design freeboard;

     *  wave erosion on the interior embankment;

     •  erosion due to precipitation or irrigation runoff;

     •  seepage along toe areas; evidence includes wet or spots,
        mineralization spots, discoloration or dying vegetation;

     •  soil movement; evidence includes irregular bank
        alignment, slides, sloughs, bulges, or depressions;

     •  animal burrows and shrub or tree growth on dikes;

     •  fractures or cracks in the crest, embankment slopes, toe
        areas, or around structures such as spillways and
        leachate sump pipes;

     •  damage to crest and interior wall due to traffic when
        discharge to the impoundment is from vehicles;     .-
                                                             I
     *  spillage on the crest or outside the embankment at   !
        unloading dock or ramp; and                          !

     *  damage from vehicles or other agents to geomembrane
        anchorage along the crest.


5.1.2.3  Operator Inspections of Ancillary Site Facilities --

     The surface impoundment design normally includes a liquid
level indicator and hard copy recorder; leak collection/detection
system; emergency spillway or other discharge structure; a system
to discourage birds and other animals from using the impoundment;
                                97

-------
survey reference points? a weather station (if needed)? inlet
pipes and valves and/or a vehicle unloading area? security
equipment; and health, safety, and emergency response equipment.

     A routine inspection and maintenance program for these
facilities should be incorporated into the owner/operator's
operation and maintenance protocol.  The following should be
included in the inspection program:

     •  The liquid-level indicator should be checked for freedom
        of movement and response.  Grease or scum accumulations
        can adversely affect response.  The chart recorder should
        be checked for calibration and ink supply, as specified
        in the manufacturer's guidelines.

     •  Leak collection systems should be checked for component
        deterioration (e.g., pipes and locks).  Pumps should be
        inspected for proper operation.  The sump should be
        checked for presence of liquid and, if any is present, a
        sampling program should be initiated to characterize the
        material.  Liquid-level indicators or recorders, if
        present, in leak-collection sumps should be inspected and
        maintained as necessary.

     •  The emergency discharge outfall structure should be
        checked for cracks, and damage due to mowing or other
        vehicular contact.  Debris or excessive plant growth that
        may impede its operation should be removed.

     *  Devices to ward off animals (e.g., plastic banners,
        fences, and electrically charged wires) should be
        inspected routinely.

     •  Elevation reference points or survey markers should be
        inspected for damage.  These reference points should be
        surveyed annually to monitor soil stability.

     *  Heather stations  (if present) should be inspected per the
        manufacturer's operation and maintenance protocol.

     *  Impoundment inflow structures must be inspected to assure
        that valve systems and recorders work properly, and that
        piping is not deteriorating.  If an unloading dock is
        present, the liner immediately below the unloading point
        should be checked for impingement or other damage*  The
        crest area and outside dike wall should also be checked
        for waste spills.

     *  Security equipment  (e.g., locks and/or locking caps on
        leak collection sump lines, monitoring wells, gates, and
        fencing) should be checked to verify that the equipment
        operates and tampering has not occurred.

                                98

-------
     •  Health, safety, and emergency response equipment (e.g.,
        air sampling systems, safety showers,  eye-wash stations,
        respiratory protective equipment,  fire-fighting
        equipment, caustic and acid neutralization materials,
        protective clothing, first-aid kits,  air horns or
        flashing lights, and radio systems)  should be inspected
        and maintained according to the manufacturer's
        recommended procedures.  In addition,  personnel must
        receive periodic training in proper equipment use.


5.1.2.4  Liner Systems —

     Routine facility operations pertaining to liners include
detecting and locating leaks, determining causes of liner
failure, maintaining and operating collection systems and pumps,
and periodically removing accumulated solids.   These topics are
discussed in the following paragraphs.

     Detecting and Measuring Liner Leakage —     USEPA (1987i)
has proposed that the impoundment operator have the capability of
detecting a leak through the primary liner as small as 1
gal/acre/day (9.35 liter s /hectare/day).  Also proposed has been
an "action response level" and a "rapid and extremely large
leakage" rate.  The first is a low rate (to be established)
between 5 and 20 gal/acre/day  (47 and 187 liters/hectare/day) and
the second is the rate, when exceeded, that leads to a hydraulic
head greater than the thickness of the leak collection layer.
These rates are used as bases for certain leak response actions,
as described in the proposed requirements for Response Action
Plans, called for under RCRA (USEPA, 1987i).

     Continuous monitoring of the leak-collection system should
be considered good practice.  This can be accomplished by
continuous liquid-level recording in collection sumps.  An
increase in level would be prima facie evidence of a leak in the
primary liner.  Monitoring for leaks can also be accomplished by
electronic leak detection methods as described in Section 3.5,
and backed up by sump level monitoring.

     If leakage through the primary liner is suspected, one
procedure for establishing its existence and magnitude is to pump
the leak collection system dry, measure the amount of recovered
liquid, and note the rate at which the system refills.  Recovery
of all the material leaked is not necessarily expected because
the drainage layer will retain some liquids (up to the system's
"field capacity").  If the underlying secondary liner is
compacted soil, it, too, will  absorb and retain some liquid.  "By
estimating the total volume of leak-saturated material, the
liquid volume retained by the various soil components can be
estimated.
                                99

-------
    If leakage occurs in a small area, its location may be
determined by a remote or in-situ leak detection system before
the leak-detection layer becomes fully saturated.   It is thus
possible for the leak to be detected before it reaches the
collection sump.

    The liquid pumped from the leak-collection system can be
managed as a "pump back" problem by returning the  leaked liquid
to the surface impoundment.  This allows the operator additional
time to plan and devise appropriate remedial actions.  Criteria
for determining the type of remedial action required when leakage
occurs through surface impoundment liner systems is discussed in
Section 6.3.3.

    If the amount of recovered leakage indicates a significant
loss, or if pump-back rates and leakage rates differ, then the
secondary liner may also be leaking and contaminants escaping to
the environment.  This may be cause for cessation of operations
and emptying (and perhaps closing) the impoundment.

    Leak-detection technologies should be used to locate the
source point of small leaks in the primary geomembrane liner.
Similar technologies may be used for early detection of releases
to the subsurface soil beneath the secondary liner.  Several
reliable methods are available to detect and locate geomembrane
leaks larger than 0.1 to 0.2 inches (2 to 5 mm), but smaller pin-
hole leaks are more difficult to find.  Remote and in-situ
techniques to locate leaks are discussed in Section 3.5.

    The flow of liquids through flaws in geomembranes depends on
the size and shape of the flaw, the liquid head, and the
hydraulic characteristics of the sub-base.  In a recent study of
the characteristics of leaks through geomembrane flaws, Brown et
al (1986) found that leakage rates through slits and seam flaws,
as might logically be expected, were much more variable than
those through round holes.  This is due to the variable hole
sizes which can result when a seam or one side of a slit becomes
displaced relative to the other side.  The magnitude of leak
rates for various geomembrane flaws is shown in Table 8.

    Brown et al  (1986) also investigated the effects of subbase
hydraulic conductivity on the leakage rate through a defective
geomembrane liner.  Less-permeable soils containing greater
amounts of clay had lower leakage through the geomembrane
component because a better seal was formed between the
geomembrane and soil components, allowing less lateral flow of
liquids at the interface.  Table 9, taken from Brown et al
(1986), gives predicted leakage rates from various hole sizes in
geomembranes overlying soils of varying hydraulic conductivities.
                               100

-------
TABLE 8.  AVERAGE LEAK RATES (*Vyr)  FROM DIFFERENT SIZE AND
          8EAPE FLAWS IN 30-mil (0.08-on) HOPE LINER OVER GRAVEL
          AT TWO LIQUID READS. (1 B3  = 261.2 gal)
                                             Bead  ( cm )
Hole Shape and Size
                                   50
                                                            100
Round - 0.16 cm (.06 in.) diam.
Round - 0,64 cm (.25 in.) diam.
Round - 1.27 cm (.5 in.) diam.
Slit - 5 cm (1 in.) long
Slit - 15 cm (6 in.) long
Seam - 5 cm (1 in.) long
Seam - 15 cm (6 in.) long
110
1482
4257
—
3866
404
4702
145
2208
6780
79
5623
325
7244
TABLE 9.  CALCULATED LEAK SATES  (nVyr) FOR A RANGE OF BOLE SIZES IS
          GEOMEMBRANE LINERS OVER SOILS OF DIFFERENT  CONDUCTIVITIES
          AND FOR THREE READS  (H) .   (I •* = 264.2 gal)
— ~S:CJS«*SBSWB«»—
KMt ( cm/ s )

=:======:=:=::==:
0.08
(1/32 in


.)

Hole Diameter (cm)
0.16 0.64
(1/16 in.) (1/4 in. )
I = 0.3 m (1 ft)

1.
(1/2


27
in.

)
3.40 x 10"*
3.40 x W5
3.40 x 10'6
3.40 x lO""1
19.30
4.30
0.54
0.066
31.50
4.88
0.60
0.072
43.20
6.28
0.77
0.095
§0.60
7.30
0.89
0.107
3.40 x 10'
3.40 x 10-
3.40 x 10
3.40 x 10'
         's
167.00
 84.60
 14.30
  1.80
                              H
                                  1.0 m (3 ft)
3.40 x 10'*
3.40 x 10'5
3.40 x 10'*
3.40 x W7
42.30
12.80
1.66
0.20
87.80
14.80
1.83
0.22
128.00
18.70
2.29
0.28
147.00
21.40
2.61
0.32
                              H =  10.0 m  (3.1  ft)
438.00
123.10
 15.60
  1.90
1030.00
 153.50
  18.80
   2.30
1170.00
 171.30
  21.00
   2.60
                                101

-------
    Determining the Causa of Liner Leakage — If the cause of
liner leakage is known, it may affect the type of repair that is
made (i.e., strengthen a weak area).  In addition it can provide
information toward preventing a repetition of the same failure.

    An unacceptable geomembrane leakage rate may result from
imperfect seaming; rips, punctures, and tears that occur during
installation, or later as a result of operational error; failures
resulting from subsidence or shear failure of the supporting soil
after installation; or exposure to incompatible wastes (e.g.,
concentrated organic solvents), which can dissolve either the
plastic or plasticizer (Brown et al, 1986) of a geomembrane.
Solvent attack can also permanently alte.r the fabric of a low-
permeability soil liner should the leak extend to that component.

    Leakage does not necessarily result from some events that
may be considered liner failures.  For example, holes above the
liquid level caused by equipment misuse or by tension separation
at seams, or anchor pullout, are failures, without immediate
leakage.  Leakage may be expected later in these situations as
such areas of weakness propagate.  The cause of liner failure
above the liquid level may be observable and relatively obvious,
especially if there is no protective soil layer.  Below the
liquid level, pumpout may be required, followed by a thorough
inspection of the primary geomembrane liner.  If the failure
appears to be chemical-caused, an evaluation of the waste liquids
and sludges contained in the impoundment and their compatibility
with the liner may be in order.

    When a liner is repaired or replaced, it goes without saying
that the cause of the failure should not be allowed to repeat
itself.  In other words, repair or replacement should always be
accompanied by elimination of the cause of failure before the
operation of the impoundment is resumed.

    Lipar Repair  — Since liners meeting minimum technology
requirements will have a geomembrane as the top or primary liner,
a failed (leaking) system will certainly have a leak in that
geomembrane.  In the recommended system, the secondary liner will
also have a geomembrane top component.  Thus, the first two
barriers are geomembranes.  If the secondary liner is leaking,
the primary liner is likely to be leaking also.  With both
leaking, repair becomes much more complicated and much less
practical, than if only the primary liner is involved.

    The  options available  for  repairing the  primary  geomembrane
liner depend on the type and cause of failure.  Mechanical
failures are often repairable by patching, re-seaming, or
returning the geomembrane to its original position.  If the
primary  geomembrane liner has not been significantly affected  by


                                102

-------
the waste, such as above the liquid level, or even below the
liquid level if the waste and geomembrane are compatible, repair
may be very practical.  However, if the damage is below the
liquid level, and particularly if the primary geomembrane liner's
integrity or chemical structure has been compromised by the
waste, repair may not be practical or even possible.  In this
case, the correction is to cease operations and close the
impoundment or replace the entire liner.  Repair and replacement
and the difficulties encountered have been explored (USIPA,
1987d).

    Solids and Liquid Removal — Sludges  that accumulate in the
surface impoundment will need to be removed periodically, and
disposed or temporarily stored.  Also, solids removal is likely
to be required following leak detection because maintenance or
repair of leaking liners cannot be performed with liquid sludges
and solids present.

    The method selected to remove sludge  from the surface
impoundment must be non-destructive to the geomembrane liner and
underlying leak detection/collection system, particularly if the
impoundment is to be put back into operation.  The use of
shovels, scrapers, backhoes, etc. would not ordinarily be
advisable for the removal of sludge in contact with the liner.
Pumping would be more appropriate.

    Several sludge pumping systems are available.   If the liquid
contained in the surface impoundment is to be removed and treated
beforehand, the intake line should be mounted on a float to
prevent pumping of bottom sludge.  Sufficient liquid must be
retained, however, to form a pumpable sludge slurry.  Some of the
liquid may be stored in tanks for washing sludge from the
geomembrane.  This eliminates the need to obtain and use
additional water for liner washing, which would then need to be
treated before disposal or discharge.

    After sludge removal, a solidification-stabilization agent,
such as fly ash, lime, port land cement, or a combination, may be
used to stabilize the sludge.  A pug mill or other nixing systen
should be installed in the immediate vicinity and sufficient
space provided to allow easy vehicle access.  Methods for
stabilizing and/or solidifying hazardous wastes have been
described in a handbook published by US1PA  (1986g).

     If  sludge cannot be disposed of on-site, the owner/operator
has two disposal options.  Some sludges can be "delisted"  (re-
classified as non-hazardous) following mixing with Type II fly
ash.  If this can be achieved, the sludge nay gain approval for
disposal at a sanitary landfill.  If the  sludge cannot be
delisted, it must be disposed in an approved hazardous waste
disposal facility.  In any case, loaders  and transportation
equipment will be required to handle the  solidified material.

                                103

-------
    Waste liquid removed from an impoundment must be treated, or
perhaps pre-treated and discharged to a municipal wastewater
treatment system.  The treatment must be consistent with the
liquid composition and will vary depending on that composition.
Treatment may be expected to produce more sludge that may require
handling and disposal similar to that of the impoundment bottom
sludge.  Techniques to dewater and solidify sludges are discussed
in Section 7.3.

    Repairing the secondary liner/ or replacing it, will require
removal of waste, all or most of the primary liner, and all or
most of the leak collection layer.  Thus, in many, if not most or
all, cases of secondary liner leakage, the entire liner system
will require re-building.  In these cases, closure may be the
most practical solution.


5.1.3  Record-keeping

    A complete inspection and maintenance program requires
complete, comprehensive records of operating and maintenance
actions.  The date and nature of the activity, the details of the
action taken, the time required to achieve the desired result,
and the name of the person responsible should be recorded.

    Documenting problems and solutions provides valuable
knowledge for future designs and for solving recurring problems
that may plague a waste disposal facility.  Recurring problems
indicate that a change in equipment, personnel training,
supervisory practices, operational procedures, or some
combination thereof may be needed.

    Inspection and maintenance records should be  kept in a  safe
place.  Such records are required to be maintained and available
for inspection by federal and state regulators.


5.2  SAMPLIHO AHD AHALYSIS MOHITORIHO ACTIVITIES

    Routine monitoring  activities applicable to a surface
impoundment include sampling and analysis of hazardous waste,  air
emissions, ground water, and liquid from the leak collection and
detection systems.  The following sections discuss those
activities required under current regulations, and those that  are
either recommended or pending with regulatory agencies.


5.2.1  Hazardous Waste Monitoring

    Monitoring waste  streams  and stored wastes  is an essential
part of hazardous waste surface  impoundment  operating procedures.

                                104

-------
Because hazardous wastes vary in their composition, degree of
hazard, and treatability, the impoundment owner/operator must be
familiar with the industrial processes that generate the waste;
the physical, chemical, and biological properties of the waste?
and any pretreatment processes used on the waste.  This knowledge
helps the owner/operator determine potential hazards associated
with handling and storing the wastes, and aids in developing
proper waste handling, isolation, and storage practices.

     Federal regulations require that the impoundment
owner/operator provide a list to the regulating agency of
hazardous wastes that are stored, treated, or disposed of in the
impoundment.  This list must include the following information
for each waste received (USEPA, 1984e):

     *  common name of each waste;

     •  chemical analysis of each waste;

     «  USEPJk ID number of each waste;

     *  location of each waste at the facility;

     •  volume of each waste received per month (estimates from
        inflow structures or from transport vehicles such as
        trucks, trains, and barges);

     •  physical form of each waste received (i.e., liquid,
        sludge, or slurry);

     •  approximate moisture or solids content and other
        significant characteristics of each waste; and

     *  special handling requirements for each waste.

     In addition, the owner/operator must record the physical
characteristics of each waste according to the following
classification (USEPA, 1984e):

     *  aqueous:  inorganic and aqueous-organic (water is the
        solvent and inorganics or organics are the solutes);

     *  organic: the predominant liquids are organic in
        composition and the solutes are organic compounds
        dissolved in the organic solvent; or

     *  solids, sludges, and slurries.

     After the waste is placed in the surface impoundment r
routine monitoring  (sampling and analysis) should continue to
assess the physical and chemical behavior of the waste over time.

                               105

-------
If the impoundment serves to temporarily store and physically,
chemically, or biologically treat the waste, a routine sampling
and analysis program enables the owner/operator to evaluate
current waste status and to monitor the treatment process.  This
monitoring program is essential when an effluent is to be
discharged under the facility's National Pollutant Discharge
Elimination System (NPDES) discharge permit.  Any effluent from a
surface impoundment complex must be sampled and analyzed before
and during its release into a stream, river, lake, or ocean via
spillway or discharge pipe.

     Collecting representative samples from a surface impoundment
is o^ten difficult due to inaccessibility.  If the impounded
waste is well-mixed, samples collected from any depth will be
representative.  However, if the wastes are stratified (i.e.,
waste density increases with depth), samples should be collected
at different depth intervals at each sample location and the
sample containers labeled accordingly.  The number and locations
of samples should be determined by a statistically valid random
sampling plan.  USEPA has established a protocol for
statistically valid sampling (USEPA, 1986a).

     A variety of sampling devices exists for collecting
hazardous waste samples  (USEPA, 1980a and 1986a; Franson, 1985).
Liquid or semi-liquid waste samplers include the following:

     •  pond sampler or rod-and-clamp sampler (beaker attached to
        a telescopic aluminum pole);

     •  Kemmerer sampler  (vessel with a trigger-activated closure
        at either end);

     •  subsurface grab  sampler (bottle with a closing lid
        attached to an aluminum pole); and

     •  weighted bottle  sampler (weighted bottle with corked
        lid).

     The type of equipment selected depends on site-specific
conditions.  USEPA recommends using a pond  sampler for sampling
impounded liquid.  If the physical form of  the waste is a solid,
sludge, or slurry, sampling equipment such  as spoons, scoops,
shovels, hand augers, or small-diameter push tubes should be  used
(USEPA, 1986).  A boat or crane should be used to sample areas
beyond reach from dikes  (Crawley et al, 1985).

     Sampling safety procedures should be tailored to the
specific situation and documented  in the  sampling plan.  Sampling
may require special safety equipment, clothing,  and personal
precautions to guard against injury from  the impounded waste.


                                106

-------
    Due to the complexity of wastes often handled in a surface
impoundment, waste analyses nay be complex.  The owner/operator
will usually have a written waste analysis plan from the RCRA
Part B Permit (standard operating procedures) which describes
sampling schedules, sampling methods, analytical methods, and a
QA/QC procedure.  Information on analytical procedures for
analyses of wastes and wastewaters is given by Considine (1974),
USEPA (1979a and 1982b), Franson (1985), and 40 CFR 261.


5.2.2  Air Monitoring

    Hazardous VOCs, odors, and particulates emitted into the
atmosphere during the handling and storage of hazardous wastes
can adversely affect air quality.  The influent liquid must be
properly screened so that the emission potential can be predicted
and the appropriate management plans developed to detect and
control the emissions.  Air quality over the impoundment must be
continuously monitored to detect and quantify the VOCs being
emitted.  Current regulations do not specify which techniques
should be employed for air monitoring.


5.2.2.1  Estimating Emissions iron Surface Impoundments —

    Several predictive mathematical models exist for estimating
emission rates of VOCs at hazardous waste disposal facilities
(USEPA, 1985c).  Table 10 summarizes the recommended air
emissions models for various types of facilities.  The two models
that apply to surface impoundment sites are the Mackay and
Leinonen Model  (ML) (1975) and the Thibodeaux, Parker, and Beck
Model (TPH) (1982).  Of these, the TPH model is most widely
recognized by regulators.  This model applies to surface
impoundments under steady-state conditions (i.e., inlet rates,
biodegradation rates, and waste constituent concentrations remain
constant) and assumes that different waste species do not
interact.

    USEPA  (1985f) determined rates of gaseous emissions  from
surface impoundments using several sampling techniques and
compared the results to predictions from the TPH Model.  The
study concluded that the TPI Model appears to be generally
applicable to several classes of compounds contained in surface
impoundments that have no oily surface films or mechanical
spraying devices.  The presence of oily films on the impoundment
surface caused the model to grossly overestimate the rate of
gaseous emissions for several VOCs.
                               107

-------
 TABLE 10.  RECOMMENDED AIR EMISSIONS MODELS FOR HAZARDOUS WASTE
            DISPOSAL FACILITIES.*
Source
                Models
Landfill
Land Treatment
Surface Impoundment
Open Tank
Storage Pile
Fixed Roof Tanks
Floating Roof Tanks
Farmer et al (1978)  - for covered landfills
Thibodeaux (1980)  -  landfill equation w/o
  internal gas generation
Thibodeaux (1981)  -  landfill equation with
  internal gas generation

Hartley model (1969)
Thibodeaux - Hwang (1982)

Mackay & Leinonen (1975)  - Unsteady-state
  predictive model for nonaerated surface
  impoundment s
Thibodeaux, Parker & Heck (1981)  - Steady-
  state predictive model  for nonaerated and
  aerated surface impoundments

Thibodeaux (1980)  -  Aerated surface
  impoundment (ASI)  model
Hwang (1970) - Activated sludge surface
  aeration (ASSA)  model
Freeman (1980) - Diffused air activated
  sludge (DASS) model

Midwest Research Institute emission factor
  equations for storage piles

API (1962), modified by TRW/EPA ~ Fixed-
  roof tank breathing losses
API (1962) - Fixed-roof tank working losses

API (1980) - Evaporation loss from external
  floating-roof tanks
EPA/API (1981) - Standing storage losses
  from external floating-roof tanks
EPA/API (1981) - Standing storage losses
  from internal floating-roof tanks
'from EPA (1984c)

     Emission models for surface impoundments  have  been described
in some detail by USEPA (1987e)  in a volume dealing with
treatment, storage  and  disposal facilities.   Estimation models
are  described for quiescent, mechanically aerated, diffused-air,
and  oil-film impoundments  for  flow and non-flow situations.

                                108

-------
5.2.2.2  Air Sampling and  Analysis  —

     Air sampling generally involves collecting  air  over a given
area via detect ion/concentration measurement devices or
collection instruments.  Direct  sampling using an  emission
isolation flux chamber in  contact with the liquid  surface (Figure
43) has been used by researchers to measure emission fluxes of
sulfur and VOCs from surface impoundments and other  waste-
handling facilities  (Schmidt et  al, 1982).  USEPA  (1985f)
reported that the variability in emission rates  using the flux
chamber was typically much less  than the variability using
indirect sampling techniques, or predicted emission  rates.  The
flux chamber appears to  be the sampling device best  suited to
measuring emission rates from surface impoundments,  provided a
sufficient number of sampling stations are placed  in appropriate
locations.
          temperature
            readout
                                         sample collection
                                         and/or analysis
     Figure  43
                                         \- flow controls
                                            grab sample
                                            port
                                            — plexiglass
                                             dome
                                      stainless
                                      steel collar
 Cutaway view of emission sampling apparatus
(USEPA,  1985f).
     The  flux  chamber determines gaseous emission rates by  ! |
passing clean,  dry-sweep air through the chamber at a fixed  j
controlled  rate.   The carrier gas flow carries with it VOCs  *
present in  the collection chamber.  The carrier gas volumetric
flow rate through the chamber is recorded and the concentration
of the species of interest is measured at the chamber exit
(USEPA, 1985f).

     While  the flux sampler is in operation, total hydrocarbon
concentrations (THC) are monitored continuously in the chamber
outlet gas  stream using a photoionization detector (PID) or a
portable  flame ionization detector (FID).  Once steady-state
emission  rates are obtained from the sampling chamber, gaseous
samples can be collected for subsequent gas chromatographic (GC)
separation  and analyses.
                                109

-------
     Sampling containers using various solid sorbents can be used
to trap low-molecular weight VQCs, while heavier VOCs are sorbed
by Florisil, glass fiber filters, or polyurethane foam (USEPA,
1985f).  Teflon bags and stainless steel containers have also
been used to collect air (emission) samples.  If the air
monitoring program requires sampling of liquids or solids from
the impoundment, the collection containers should be designed to
minimize headspace to prevent loss of VOCs from the liquid.

     Analyses of collected samples may be performed on- or off-
site, depending on site conditions and available instrumentation.
Two analytical methods described by USEPA (1985f) include the use
of (1) a field-portable GC-FID, and (2) a laboratory-based
capillary column GC-FID/PID/Electron Capture Detector (GC-
FID/PID/ECD) with cryogenic concentration and subambient
temperature programming.  When liquid or solid waste samples are
collected in conjunction with air samples, the analyses are
normally conducted using purge and trap techniques followed by
GC-FID/1ID/ECD.
5.2.3  Ground-water Monitoring

     Under current RCRA regulations, a ground-water monitoring
and protection program must be implemented at all surface
impoundment sites.  The ground-water monitoring program has two
parts:

     *  routine collection and analysis of ground-water samples
        to detect contaminants that may have leaked from the
        surface impoundment;

     *  if releases are detected, the collection and analysis of
        more ground-water' samples over a wider zone to evaluate
        concentrations of contaminants, plume characteristics,
        and contaminant migration rates.

     Intensive monitoring is implemented when contaminants have
entered the ground water.  It is continued through the course of
remediation in order to evaluate the progress of that activity.
More information on this type of monitoring is given by Canter
and Knox  (1985) and USEPA (1982e; 1983b; and 1986d).

     The  design of a ground-water monitoring program is based on
an evaluation of site-specific conditions.  These conditions
include the characteristics of the wastes handled by the
facility, and the subsurface geology and hydrology and the
potential fate of the hazardous constituents.  Guidance may be
found on  a ground-water monitoring design in a recently published
handbook  on ground water by USEPA (1987f).
                               110

-------
    A minimum of four monitoring wells  is required to meet the
objectives of a detection monitoring program  (EPA,  1985a)»  One
well is located hydraulically upgradient of the  surface
impoundment and three wells downgradient.  Due to  the often
complex nature of a facility layout  (e.g., more  than one
impoundment unit) and the complexity of subsurface hydrogeology,
it may be necessary to install additional wells  (Figure 44).
                            ground-water
                            flow direction
           upgradient well H
a upgradient well
                   D D D D
                                      impoundment
                                          B
                           downgradient
                     0       WellS 	•»" D D D D D
                     4         *
                     L— optional downgradient wells —^- •
   Figure 44.   Example layout of ground-water monitoring wells.


    As with leak  detection in the impoundment liner system, the
main objective of ground-water monitoring  is  the early detection
of leakage after  a  failure of the liner  system.   Because many
existing  surface  impoundments are not equipped with a sensory
leak detection system,  the monitoring wells located directly
downgradient of the impoundment  act  as the "early warning"
system.   The number and location of  downgradient wells, and
placement of the  screened  intervals  in each well, depend directly
on the location of  ground-water  pathways along which migrating
contaminants are  transported.  General regulatory guidelines for
determining the number  and spacing of wells have been provided by
USEPA (1985a).

    Guidance in selecting  the best-suited  drilling method for a
given application (e.g., air rotary, water rotary, cable tool,
etc.) has been provided by USEPA (1980bj 1985a;  Ii87f), Minning
(1982),  and Scalf et al (1981).

    The  type of construction materials  used  for well casings and
screens  depends on  local hydrologic  conditions and the chemical
properties of the ground water and suspected  pollutants (USEPA,
1985a; Lewis,  1982). Pettyjohn  et al  (1981)  discuss well casing
                                111

-------
materials and their suitability in the presence of various
organic contaminants.  Stainless steel, black steel,  galvanized
steel, polytetrafluoroethylene (teflon), PVC, polyethylene, epoxy
biphenol, and polypropylene are commonly used materials.

     In most cases, 2- to 4-inch (5- to 10-cm) inside-diameter
(ID) well casing is sufficient for monitoring purposes.   However,
designing for possible future pumping activities (e.g.,  ground-
water recovery) requires a larger-diameter casing.  Some large-
volume pumping units require up to 36-inch (90-cm) diameter
casing.

     Once constructed, a well must be developed by overpumping,
surging, or jetting.  The objective is to produce a clean,
debris-free environment in the well bore.  At the time of
completion, the well should contain only formation water.

     For all ground-water monitoring wells, documentation of
design, construction, and completion is required by enforcement
officials.  USEPA (1985a) discusses the information that must be
presented as a written record.

     The RCRA-required Part B Permit Application for a hazardous
waste surface impoundment includes a comprehensive written
ground-water sampling and analysis plan, which discusses the
following topics:

        sampling equipment
        sampling techniques
        sampling schedules
        sample handling and preservation
        sample analysis procedures
        field and laboratory QA/QC programs
        data collection and statistical analysis

Information on these topics is contained in references by USEPA
(1984f and 1985a), Gillham et al (1983), Scalf et al (1981), and
USGS  (1985).

     Each component of the written sampling and analysis plan
must be closely followed to ensure sample integrity and data
quality.  Ground-water sampling and analyses must be conducted by
trained personnel, and procedures should follow the techniques
recommended by USEPA ground-water monitoring regulations.


5.2.4  Soil-vapor Monitorlog

     Monitoring soil vapors in the vadose zone just above the
water table nay be useful in detecting the leakage of volatile
contaminants from the impoundment.  Use of soil vapor probes is
an  inexpensive and rapid means to collect data that may  indicate

                               112

-------
a problem.  Soil-vapor probes are  designed to  collect organic
vapors that accumulate in, or migrate  through,  unsaturated soil
zones (Figure 45).  Field studies  using  soil-vapor sampling and
analysis at ground-water contamination sites have  indicated that
the concentrations of VOCs in soil gases correlate strongly with
the concentrations of those compounds  in the ground water
(Kerfoot et al, 1986).  Soil-gas sampling should be conducted at
surface impoundment sites by experienced operators of soil-gas
equipment, where ground-water contamination has occurred or is
suspected, provided suitable geologic  conditions exist for using
the soil-gas technique.
         vacuum pump
         sampler
                              D access
          or
               NOT TO SCALE
         evacuated sample vial _
         may be lowered to here
small-diameter hole
1 inch (2.5 cm) or less
                                 disposable tip (hand-driven)
        Figure 45.   Schematic of soil-gas sampling probe.
5.2.5  Leak Collection and Removal System Monitoring

    A leak collection system collects leakage accumulating above
the secondary liner  and channels the flow through a porous medium
(i.e., sand and  gravel)  and piping networks to an outflow sump or
collection basin.  Primary liners tend to allow some release, and
the intent is to have that waste liquid migrate through the
collection system  rather than enter the underlying secondary
liner.  The leak collection system serves another purpose, and
that is to monitor the volume and chemical composition of liquid
that may enter it  (USEPA, 1983a).

    The leak collection system generally consists of sampling
and pumpout sumps  along lateral lines or one sump at the system
outflow.  For sampling,  small-diameter tubing may be inserted
                                113

-------
into a sump and extended to the ground surface through a
protective riser pipe (Figure 20).  Samples are collected by
pumping liquid up through the tubing.  If excessive leakage
occurs in the primary liner and several sumps are available, the
general point-source area may be identified from increases in
liquid volumes in one or more lateral sumps.

    The sump at the system outflow, if any, can be sampled in a
similar manner, although some outflow sumps are connected to the
ground surface by a manhole and samples can be collected using
bailers, pond samplers, grab samplers, or similar devices.

    The sampling and analysis schedule for leak-collection
systems depends largely on site conditions and the liquid volume
moving through the collection system..  If the system also serves
as the primary method to detect liner leakage, liquid volumes
should be monitored at least on a weekly basis.  Continuous
monitoring of the liquid level is preferred.  Sampling for
chemical analyses of the liquid should be performed periodically.
Samples should be taken immediately upon the occurrence of
changes in pumping rates that indicate possible leaks.  In cases
where the liquid is to be treated and either returned to the
impoundment or released under an NPDES discharge permit, frequent
analyses must be conducted on samples before and after treatment.
Sampling techniques, sample handling and preservation, and
analytical procedures are the same as those used in ground-water
and waste monitoring.
                               114

-------
                            CHAPTER 6

                      COHTIHGEHCY PLANHIHG


     Owners and operators of surface impoundments must take
precautions to protect human health and the environment from
impoundment failures.  In general,  RCRA regulations and/or
proposed RCRA regulations address three kinds of events that call
for corrective or remedial actions.   These are (1) sudden drops
in impoundment level or dike leaks;  (2) leaks through the top
liner into the leak detection, collection, and removal system;
and (3) whenever a ground-water protection standard is exceeded
at a compliance monitoring point.


6.1  LIQUID-LOSS RESPOHSE PLAHS

     All three release event possibilities must be addressed as
parts of the facility permit and approved by the Regional
Administrator.  Sudden drops in impoundment level, and dike
leaks, must be addressed as part of the facility's "contingency
plan."  Liner leaks are addressed in a "response action plan,"
and ground-water standards violations by "corrective action
program" descriptions in the permit.  Summaries of the pertinent
regulations are provided in Subsection 1.3 of this document.


6.1.1  Contingency Plan

     Contingency plans are required for all hazardous waste
facilities.  They must be submitted and approved as part of the
facility permit.  The general requirements regarding the plan
content are provided in 40 CFR Parts 264 and 265, Subpart D.
Additional requirements, specific to surface impoundments, are
given in 40 CFR 264.22? and 265.227.  Contingency plans address
potential unintended, large, sudden releases of hazardous
materials.  These are the most threatening releases, measured in
perhaps tens of liters or more per second, that require immediate
response actions to minimize risk to human health and the
environment.  Immediate response actions might include evacuation
of the affected area and mobilization of emergency response teams
and spill mitigation equipment.  The decision-making process and
response should proceed according to the contingency plan.
                               115

-------
     The general components of a contingency plan are as follows:

     *  names, addresses, and phone numbers of emergency
        coordinators;

     •  locations, amounts, and characteristics of impounded
        wastes;

     »  potential hazards to human health or the environment
        caused by fire, explosion, or uncontrolled release of
        impounded wastes;

     *  list of emergency equipment, including descriptions and
        locations;

     •  evacuation plan for the surface impoundment facility;

     *  emergency response procedures (preplanned and discussed
        with local authorities);

     »  procedures to reduce and prevent exposure caused by
        sudden releases, non-sudden releases, fires, and
        explosions;

     •  procedures for removing the surface impoundment from
        service, containing leakage, shutting off inflow,
        preventing catastrophic failure, and emptying the
        impoundment;

     *  potential remedial actions, associated health hazards,
        decontamination procedures, and methods to provide
        personal safety for personnel carrying out remedial
        actions;

     •  location of contingency plan copies; and

     *  procedures for updating the contingency plan in the event
        of system modifications.

     Table 11 shows an outline for a "response data sheet,"
summarizing many of the above points, which should be prepared
for each impoundment and made a part of the contingency plan.
     In proposed regulations, 40 CFR 264.222 and 265.222,
published May 29, 1987, the USEPA proposed requirements for
response action plans as part of the permit process for hazardous
waste surface impoundments.  These plans would deal with leaks of
waste liquid through the top liner and into the leak collection
and removal system.                                             —

                               116

-------
SABLE 11.  OUTLIKE OF CONTINGENCY PLAN RESPONSE DATA SHEET.
1.  Emergency Coordinator{s)

           Name:
           Address;
           Phone:

2.  Surface Impoundment Design

           Volume :
           Flow Controls:
           Flood Routing Data:

3.  Properties of Impounded Hazardous Substances

           CIS Class:
           CERCLA Categorization (class and subclass):
           Physical/Chemical Properties:
           Hazardous Characteristics:
           Protection Level Required:

4.  Spill Countermeasures

           Physical Countermeasures:
           Chemical Countermeasures:
           Land/Soil:
           Surface Water:
           Ground Water:

5.  Emergency Equipment

           Type:
           Capabilities:
           Limitations:
           Location:
     USEPA has proposed two types of response action plans.  The
first,  required  to be  submitted with the Part B permit
application, would address a  "rapid and extremely large leak."  A
rapid and extremely  large leak is defined as the maximum that the
leak collection  and  removal system can remove without the fluid
head on the bottom liner exceeding one foot in a granular leak
collection system.   The head  limit in a geosynthetic leak
collection system would be equivalent to the geosynthetic
thickness.  The  actual leakage rate for a rapid and extremely
large leak would be  site-specific, but might be expected to be
several hundreds or  thousands of gallons per acre per day.

                               117

-------
     The second type of response action plan,  that may be
submitted at a later date, even as late as the leak event, would
address a leak through the top liner exeeding  an "action leakage
rate."  USEPA has not determined that rate, but suggests in the
preamble to the proposed regulation that it may be established at
between 5 and 20 gallons/acre/day (47 and 187  liters/hectare/
day).  Based upon measurements of actual leakage through top
liners at facilities that have been built under rigid quality
control, Bonaparte and Gross (1990)  have suggested an action
leakage rate of 200 liters/hectare/day or about 21
gallons/acre/day for landfills.  For surface impoundments, they
found little leakage through a top geomembrane liner, and they
suggest that an action leakage rate of 200 liters/hectare/day can
be met if ponding tests or leak location surveys are carried out.

    I The proposed regulations would require that a response
action plan include the following!

     *  description of the unit and the planned method of
        closure j

     »  hazardous constituents in the waste;

     •  events that may lead to a rapid and extremely large leak;

     •  factors affecting leakage into the leak collection
        system;

     »  mechanisms to prevent leakage out of the unit;

     *  assessment of responses that decrease  leakage into leak
        collection system; responses may include limiting waste
        receipt, expeditious repair, or operational changes; and

     *  correlation of a range of leakage rates with different
        responses, indicating why others were  not chosen.

     For leaks above the action leakage rate but less than rapid
and extremely large, the plan must include similar information,
but the range of responses could be broader, depending on the
severity of the situation.  The options could  include
continuation of liquid removal, treatment and  ground-water
monitoring, or the maintenance of current operations.


6.1.3
     Either a spill or a liner leak could result in contamination
of the ground water in the vicinity of the impoundment.  In order
for such a liner leak to occur in an engineered double-liner
system, the response action plan would not have been effective,
and both liners will have leaked.

                               118

-------
system, the response action plan would not have been effective,
and both liners will have leaked.

     When a release has escaped a surface impoundment,  and a
ground-water protection standard has been exceeded at a
compliance monitoring point, a corrective action program must be
implemented.  The corrective action program is required under 40
CFR 264.100 and must be initiated and completed "within a
reasonable period of time."  The terms of the program are
specified by USEPA's Regional Administrator in the facility
permit.  In the permit, he establishes the ground-water
protection standard for the site.

     Corrective actions, such as removal or treatment of the
ground-water contaminants, must be adequate to bring the ground-
water at and downgradient of the compliance point to levels that
meet or exceed the standard.
6.2  TYPES OP FAILURE

     A surface impoundment failure is manifested by a loss of
impounded waste liquid.  Ordinarily, the loss will be one of two
types, (1) a sudden catastrophic loss, or spill» or (2) a liner
leak.  As noted above, the two types are reflected in the
different requirements for contingency plans and response action
plans, respectively.  The spill type would be characterized by a
large volume lost in a short time, with the flow perhaps measured
in tens of liters per second, whereas the leak type might be
measured in cubic centimeters or smaller per second.

     A surface impoundment spill situation would be expected to
be associated with a foundation or dike failure.  Causative
factors might include unstable soil on dike slopes, very high
precipitation and dike overtopping, or subsurface weakness
(solution channel or sink hole).  The last of these might result
in a spill to ground water, whereas the first two would result in
surface spills.  Spills, because of inherently higher volumes,
are usually emergency situations.  The real possibility of spills
justifies the contingency plan.

     The hazard associated with a sudden release or spill
involves immediate, short-term exposure to hazardous liquids.  In
this situation, the appropriate emergency response involves
attention to the safety of persons, immediate spill control,
cessation of process operations, notification of regulatory and
local authorities, and implementation of remedial activities for
treating, storing, or disposing of contaminated soil and water.

     Leaks can be associated most often with weaknesses in the
impoundment liner, such as pipe penetrations and liner attachment
to structures.  Leaks may also be attributed to accidental

                               119

-------
puncture, parted seams, tensional tears, and low chemical
resistance of the liner.  Most impoundment failures appear to be
of this type without a discernible sudden onset.  Leaks would be
expected to permeate to ground water, but they could also
permeate the dike to surface water.  Leaks generally constitute
serious, but not emergency, instant-response situations.

     The risk associated with an impoundment leak, or non-sudden
release, involves the potential for long-term exposure to low
concentrations of hazardous substances in drinking water.
Additionally, a non-sudden release may result in potential long-
term, lower-level exposure to aquatic organisms in surface water
or terrestrial biota in contact with contaminated soil.


6.3  RESPONSE PLAH IMPLEMENTATIOH

     The primary difference in response to a spill and response
to a leak lies in the amount of time, and therefore the degree of
detail, available for investigating the distribution of
contamination and selecting and implementing the remedial
actions.  In general, response to a sudden release requires quick
application of containment and protective measures.  Responses to
a non-sudden release (a leak detected in the leak collection
system or at a ground-water compliance point) are generally less
urgent and allow more time to assess the contamination and select
appropriate remedies.


6.3.1  Contingency Plan I«pl«a*ntation

     When a large sudden release or spill is detected at a
surface impoundment, the contingency plan is implemented by the
Emergency Coordinator (designated in the plan), who has
independent authority to initiate and carry out the response.
Immediate response actions include controlling the source,
containing the leaked liquid, and notifying the authorities.
When immediate threats have been controlled, the response process
proceeds with the contamination assessment; determining the
necessary response objectives; screening available remedial
options; selecting and implementing an appropriate remedy; and
sampling to verify cleanup effectiveness.

     Quick implementation of the contingency plan before
contaminants are detected in the ground water could minimize the
magnitude of contamination and necessary remedial action.  Just
as importantly, it could help reduce owner/operator liability.

     A hazard classification of over 150 compounds, which
provides information on the toxieity of these chemicals to human
health and the environment, was published by Clements and
Associates (1985).  Hazardous chemicals can also be categorized

                               120

-------
flammability/combustibility, persistence, and corrosiveness.  The
information from the response data sheet and the general
contingency plan will enable the coordinator to quickly evaluate
the emergency situation, review the available countenneasures,
and assess the feasibility of each approach.  The four basic
steps in a contingency plan are:

     1. Immediate Actions
     2. Contamination Assessment
     3. Selection and Implementation of Remedial Actions
     4. Cleanup Verification

     Each of these steps is discussed in the following sections.


6.3.1.1  Immediate Actions —

     In the event of a sudden release, initial precautionary
emergency measures must be implemented to minimize the risk to
workers and the general public, and to prevent additional fires,
explosions, or uncontrolled releases.  Basic steps to reduce the
potential exposure are as follows:

        notify on-site facility personnel
        control the source of hazardous release
        mobilize emergency response teams and equipment
        isolate the sudden release area
        eliminate ignition sources
        restrict water use
        prevent commingling of reactive wastes
        control fires

     The Emergency Coordinator must also initiate efforts to
contain the released waste.  Temporary berms may be constructed.
The Emergency Coordinator has the authority to stop process ...
operations, if necessary.  If facility operations are halted,
monitoring should commence in order to detect leaks, pressur'e^
buildup, gas generation, or ruptures in valves, pipes, or other
components of the surface impoundment.                       i

     The local police, fire departments, and hospitals must be
notified of the release, and advised on the magnitude of
potential health hazards and the potential need for evacuation of
local  areas.  The Emergency Coordinator must notify either the
government official designated as on-scene coordinator for that
geographic area or the National Response Center.  These response
actions contained in 40 CFR 264.56 are consistent with RCRA
guidelines for emergency procedures at hazardous waste
facilities.  Prompt actions during the initial phase of emergency
response can minimize potential hazards to workers and the
general public and reduce the overall cost of containment and
clean-up operations.

                                121

-------
6.3.1.2  Contamination

     The purpose of the contamination assessment is to determine
the relative severity of the release and the potential safety
hazard to workers and the public.  This is done by evaluating the
fate and transport of the contaminants, identifying the exposure
pathways, and assessing the risks posed to public health and the
environment.  Additionally, an assessment conducted immediately
after a sudden release from a surface impoundment may allow
reconstruction of the events that caused the release, and
possibly prevention of future releases.  The assessment also
provides information needed for screening and selecting remedial
actions, and defines resources available for countenneasure
activities  (Melvoid et al, 1984).  The contamination assessment
should result in defining the response objectives for
remediation.  Factors to consider in the contamination assessment
follow:

     •  surface impoundment(s) involved;

     •  human exposures and/or injuries;

     •  potential for fire, explosion, or continued release;

     •  name and class of substance(s) released;

     •  physical and chemical characteristics of the released
        substance;

     •  approximate volume and concentration of released
        substance;

     •  media affected by release;

     •  permeability of affected soils;

     •  anticipated direction and speed of migration;

     •  proximity to natural barriers;

     •  local terrain/topography;

     •  proximity to environmentally sensitive areas and
        likelihood of impact; and

     •  weather conditions (i.e., wind speed and direction,
        air/ground/water temperature, and precipitation).

     Some of the innovative technologies that may be used in
assessing site contamination are listed in Table 12.
                                122

-------
   TABLE 12.  IHHOVATIVE INVESTIGATION TECHNOLOGIES TO ASSESS
              SITE CONTAMINATION.
Technology/Instrument
Information Provided
Soil Vapor Detector/Analyzer
(Nadeau et al, 1985)
Aerial Photographs
(Finkbeiner & O'Toole, 1985)
Backscatter Adsorption Gas
Imaging
(McRae, 1984)

Acoustic Mapping
(Meyer et al,  1984)
Ground-Penetrating Radar
(Stanfield & McMillan,  1985)
Migration of volatile organic
compound (VOC)  plume in soil or
ground water.

Documentation of site conditions,
assessment of potential hazards,
assessment of off-site impacts,
progress of clean-up operations,
and evaluation of post-clean-up
conditions.

Detection and tracking of VOCs.
Delineation of the location and
thickness of contaminant layers
in ground-water aquifers or
surface water sediments.

Location of subsurface obstacles,
contamination, faults, and
fracture zones in clay liners.
6.3.1.3  Selection and  Implon«ntation of R«awdial Actions —

     The Emergency Coordinator will use information from the
contamination  assessment  and knowledge of facility operations to
evaluate the available  remedial response alternatives.  This
evaluation  should be  based on performance, practicality, and
effectiveness  of the  response alternatives (Melvold et al, 1984).
Availability of the required resources, speed of application, and
effectiveness  of the  response methodology in different weather
and  topographic conditions are also considered.

     Several remedial action technologies are available for use
in the event of a sudden  release from a surface impoundment.
Available technologies  can be divided into three categories:  (1)
removal and disposal; (2) in-place treatment; and  (3)
containment.   For preparation of the contingency plan, resource
documents describing  remedial action technologies  should be
reviewed.   The contingency plan should contain a thorough
                                123

-------
description of each response strategy or alternative in order to
screen and select the appropriate response  measures.  Information
on each remedial alternative should include,  at a minimum,  the
following (Unterberg et al,  1984):

     •  technical description;

     •  applicability for field use;

     •  mitigation of human and environmental risks;

     •  environmental, demographic, and legal constraints;

     •  requirements for, and availability  of,  adequate power
        supply, manpower, special equipment,  and supplies;  and

     •  cost.

     Several important references describe  available response
strategies for hazardous substance releases (USEPA,  1985k;  1983d;
1984j; and 1986e).


6.3.1.4  Cleanup Verification —

     Once a cost-effective remedial action  alternative is
selected and implemented, the Emergency Coordinator should begin
evaluating the effectiveness of site cleanup efforts.  Factors
that may affect the performance of response efforts include
weather, equipment malfunctions, detection  of additional hazards,
and changing response objectives.  The remedial action program
should continue until the following criteria have been met,
documented, and verified:

     •  risk to human health and the environment posed by the
        release of hazardous substances has been eliminated or
        reduced to an acceptable level;

     •  potential for recurring hazards (e.g., fire, explosion,
        and chemical release) has been eliminated; and

     •  response equipment and facilities have been
        decontaminated.

     Environmental sampling should be conducted to determine if
the released contaminants have been successfully treated or
removed.  Although the purposes differ, the methods for
verification sampling of environmental media for facility closure
are similar to those for response action; they are discussed in
Section 7.2.4.  Decontamination techniques for emergency response
equipment are presented by Meade and Ellis (1985).


                               124

-------
6.3.2  Implementation of Response Action Plan

     When a leak through the top liner of a surface impoundment
is detected, its severity should be assessed.  If it exceeds the
"rapid and extremely large leakage rate," the owner/operator
should implement the response action plan.  A smaller leak, but
exceeding the "action leakage rate," also calls for
implementation of the response action plan, but the response will
be of less immediacy.  The response action plan should have
foreseen, and included in its range of potential events, the type
of leak and appropriate responses.  Any leak greater than the
"action leakage rate" in a double-lined impoundment requires a
response and written notification to the USEPA Regional
Administrator.

     One advantage of a double liner system is that there is
likely to be more time available to respond to a leak than to one
from a single-liner system, because the former allows for early
detection and collection of leaks without contaminant loss to the
environment.

     If the leak is through both liners, or through a single .
liner as exists in some, especially older, impoundments, the
response may require quicker reaction, with fewer options
available.  In this case, a response action plan and a ground-
water corrective action program may overlap, because the leaked
contaminants may have migrated to the ground water before
detection.

     If the leakage is greater than the "action leakage rate" but
less than the "rapid and extremely large leakage" rate, the leak
collection and removal system should remain effective, allowing
time to choose among the corrective options available.  One of a
variety of responses may be appropriate.  At one extreme the
response may be simply a continuation of normal operations with a
greater alertness to contamination release potential.  At the
other extreme, impoundment closure may be implemented.  If the
response action plan has been prepared and approved, it should
contain the range of options available.  It should also indicate
the preferable response option for the type of leak that has
occurred.  If the continuation of normal operations is chosen,
increased monitoring is likely to be undertaken.  Even though a
leak above the action leakage rate may be relatively easily
handled by an early decision to continue operations, perhaps with
increased monitoring, at some point leak correction will probably
be undertaken as part of the response plan.  The correction may
be to repair the leak, change operations to reduce leakage, or
terminate receipt of waste.  The choice may be dependent on the
rate of leakage.  The choice of response must be approved by the
Regional Administrator.
                               125

-------
     The first reponse may be similar for a leak greater than the
"rapid and extremely large leakage"  rate.  Because it is a leak
of much greater severity,  the response must be more immediate and
decisive.  By definition,  the leak collection and removal system
alone cannot handle a rapid and extremely large leak, although
its operation should continue.  The proposed regulations suggest
three options: (1)  limiting or terminating waste receipts; (2)
expeditious repair; or (3) operational changes to reduce leakage
to less than rapid and extremely large.  At the same time,
mechanisms must be implemented to prevent leakage out of the
unit.  In all cases, leaked liquids must continue to be collected
and removed.

     When liquids are found in the leak collection system between
two liners of a double-liner system, several possible courses of
action are available, depending on the potential site-specific
impact.  In contrast, when leakage occurs in a conventional
single-liner impoundment,  the only appropriate response is to
identify and eliminate the source of leakage, and clean up the
contamination.  It is apparent that by increasing the level of
protection afforded by the surface impoundment components, the
potential for contamination of the environment decreases, the
number of possible remedial response alternatives increases, and
the flexibility of selecting and implementing the response
actions increases.

     Finding the location of a leak in the top liner is likely to
be advisable if liner repair is contemplated.  When found, and it
is a relatively small, discrete leak, the liquid may be drawn
down to the leak location, and the damage patched.  This may
prove to be difficult if the liner has been exposed to waste for
some time and waste absorption has occurred.

     If it is a large leak, or one caused by general
deterioration of the liner, it may be decided to cover the old
liner with a new one.  In most cases, this may require emptying
the impoundment.  In some, it may be possible to emplace a new
liner without drawing down the waste liquid (Cooper and Schultz,
1983; Shultz et al, 1985).  The emplacement would involve pulling
an assembled liner across the impoundment beneath the liquid.  An
untried method might be to pull an assembled liner across the
impoundment's surface and then sink it to the bottom.

     If both liners of a double liner are leaking sufficiently to
demand action, repair of the system may not be practical.  The
bottom liner will likely not be accessible without removal of the
top liner and at least part of the leak detection and collection
layer.  It may be most cost-effective to close the impoundment.

     USEPA conducted a nationwide study  (Cochran et al, 1983) of
the remedial actions being used at uncontrolled hazardous waste
sites.  Remedial actions at surface impoundments are also

                               126

-------
discussed in other USEPA documents (USEPA, 1985k and 1983d).
Modeling of a remedial technology using site-specific data may be
used to determine the applicability of a remedial strategy for a
particular site.  General guidance in selection of models for
evaluating remedial actions is provided by 0SEPA (1985g).


6.3.2.1  Leak Correction Verification —

     The owner or operator of the surface impoundment must submit
a report (as proposed in 264.222) on the effectiveness of the
leak response action to the Regional Administrator within 60 days
after the action has been taken.  Verification of the
effectiveness, if the impoundment has not been closed, requires
the leak collection system to be operating under pre-leak
conditions, that contaminants are not entering the collection
system above the action response level, that contaminants are not
being released to the soil and ground water, and that the liner
system has been returned to USEPA-acceptable design conditions.
One of the leak-detection methodologies described in Section 3.5
may be used to confirm that a leak no longer exists.

     If the site has been closed, ground-water sampling must
verify that the ground water does not contain hazardous
constituents exceeding health-based standards.


6.3.3  Implementation of Corrective Action Program

     A corrective action program is generally implemented after a
release of contaminants to ground water.  The program is directed
at the cleanup, by removal or treatment, of the contaminated
ground water.

     In general, a specific sequence of events will have occurred
that triggers a corrective action program.  First, the presence
of a monitored hazardous constituent will have been detected at
the compliance monitoring point.  When that occurs, a compliance
monitoring program must be implemented.  If the constituent(s)
exceed the ground-water protection standard for the constituent,
or a concentration limit established by the Regional
Administrator, a corrective action program must be implemented.
The owner or operator may be relieved of program implementation
only if he can show that the source of the hazardous
constituent(s) is not from his impoundment.

     Several reporting requirements are involved in detection and
compliance monitoring and during and after completion of a
corrective action program.  All reports are required in writing
to the Regional Administrator.  Compliance monitoring and
corrective action programs, once implemented, must be completed
and the contamination cleaned up, regardless of the activity or

                               127

-------
inactivity stage of the impoundment.  Cleanup must include ground
water both on and off the site, to the extent necessary to bring
the ground water to within the limits of the established
concentration limits or the ground-water standards.

     Cleanup of ground water may be a long and difficult task.
The available options include: (1) extraction (may include
flushing) and removal of the contaminated water from the site,
(2) pumping, treating, and returning the contaminated water to
the aquifer, (3) treating the aquifer in place (may include
flushing after treatment).  Treatment alternatives include
immobilization and degradation techniques.  Many of the
techniques have been described by USEPA (1984m and 1985k).

     Ground-water cleanup may be facilitated by the construction
of barrier walls.  Barrier walls may confine the contaminated
zone to a more manageable area, prevent the spread of
contaminated water, and reduce the need to treat uncontaminated
water that would otherwise enter the zone.  Slurry walls are
commonly used for this purpose (USEPA, 1984n).


6.3.3.1  Ground-water Cleanup Verification --

     A corrective action program must include ground-water
monitoring throughout and after completion of the program.
Monitoring may be a continuation of the compliance monitoring
program initiated before the corrective action program.  The
corrective action program may be terminated when the hazardous
constituent concentration has receded to the compliance
concentration limit.  When the corrective action is taken after
the impoundment's active life, the action may not be terminated
until the owner/operator can demonstrate that the ground-water
protection standard has not been exceeded for three years.

     The owner-operator must report periodically on the progress
of the the corrective action.  In addition, he must attest to the
outcome of the program upon its completion.  The reports are made
to the Regional Administrator.  During the action, if it is not
showing success, the Regional Administrator may require a change
in the program including a change in the permit conditions.


6.4  PERSONAL SAFETY DURING REMEDIAL OPERATIONS

     The contingency plan for remedial actions at surface
impoundment sites should identify health hazards and potential
risks and give appropriate methods to provide personal safety.
In light of the possible health hazards that may occur during a
remedial investigation and subsequent cleanup operation, it is
important that the contingency plan provide the means for
assuring worker protection.  The  required level of worker

                               128

-------
protection during remedial actions will depend on site and waste
characteristics which affect the fate and transport of
contaminants, toxicology of the waste, environmental features of
the site, and physical hazards.  Additional information on worker
health and safety hazards has been provided by NIOSH (1985).
There are four basic parts to worker protection (Lippitt et al,
1984):

     *  site management procedures to control access and minimize
        exposure;

     »  engineering safeguards to contain the waste and isolate
        workers from hazardous areas;

     •  personal protective clothing and equipment to minimize
        direct contact and inhalation; and

     •  decontamination procedures and practices to remove and
        control the spread of contamination.

     Bach of these facets should be discussed fully in the
contingency plan, and all of the necessary equipment and personal
safety gear should be available and accessible at all times at
the site.
                               129

-------
                            CHAPTER 7

                  CLOSURE AND POST-CLOSURE CARE


     Owners/operators are required by  40 CFR 264.111 to close
hazardous waste surface impoundments in a manner that minimizes
the need for further maintenance and prevents threats to human
health and the environment.   Closed impoundments must control,
minimize, or eliminate the post-closure escape of hazardous
waste/ hazardous constituents,  contaminated runoff, and waste
decomposition products to ground water, surface water, and the
atmosphere.  Closure regulations have  not been promulgated or
proposed by USEPA for non~ha zardgus waste surface impoundments.

     Two basic closure options are available for a hazardous
waste surface impoundment under RCRA regulations:  clean closure
(complete removal), and in^pjlace closure (40 CFR 264.228[a][l] &
[2]).  Figure 46 presents a flowchart  for each of the two
options.

     Clean closure of a hazardous waste surface impoundment
includes (1) removal or decontamination of all waste residues and
contaminated liner system components and subsoils; (2) sampling
to verify decontamination; and (3) backfilling.  USEPA and RCRA
regulations tend to encourage clean closure, because the site is
essentially restored to its pre-impoundment condition, and
further attention is not required.  However, the removed
hazardous materials must still be treated and disposed of at some
location, which may be a landfill.

     In-place closure of a hazardous waste surface impoundment is
similar to closure of a landfill or other disposal unit.  It
involves the solidification and/or treatment of all contaminated
media in the surface impoundment  (i.e., liquids, sediments, and
subsoils) by chemical, physical, and/or biological techniques.
No free liquid may remain in the unit after closure.  Hazardous
wastes must be treated to minimize their toxicity.  The treated
wastes left in place must have a final cover structure that meets
the minimum technology requirements for a hazardous waste
landfill cover (USEPA, 1989a).

     The primary goal of in-place closure is stabilization of all
residuals, providing the material with a bearing capacity to
support a final cover.  While treatment to reduce the impounded
contaminants may be required, this approach does not necessarily

                               130

-------
petrton lor "ctean ctoeure-
 warver and establish
alternative concentration
      firnits
bttdcfll impoundrnsnt
•HOi unoontvninalBd
matonal
1






»
pcBt-ctooure care
parlod
j




__
petition for post^iosuTB
variance H applicable



   Figure  46.   Flow chart of  closure  options  and  requirements.
                                       131

-------
eliminate all hazardous constituents.  Thus, a 30-year post-
closure care period is required.  Post-closure activities are the
same as for a closed hazardous waste landfill.  They include
long-term operation and maintenance of the final cover, leachate
collection system operation and maintenance, and ground-water
monitoring.

     If the impoundment is older and does not have a non-leaking
liner, all contaminated materials (waste, liner, soils, etc.) may
be removed and managed as hazardous waste.  An alternative is to
eliminate free liquids, stabilize the remaining waste, provide an
effective cover, and provide monitoring and maintenance
throughout the post-closure care period.

     A critical aspect of in-place closure is constructing the
final cover system with an hydraulic conductivity at least as low
as that of the liner system.  If the liner is ineffective, USEPA
may still require a cover that meets the performance objectives
of a "minimum technology" cover.  The final cover acts to
minimize infiltration and leachate formation by diverting surface
water with slopes, drainage layers, and low-permeability barriers
(USEPA, 1986a).  USEPA (1989a) has published minimum technology
guidance for hazardous waste landfill covers.  The guidance
recommends a top soil layer with a vegetated or armored surface,
a drainage layer, and a two-component (geomembrane and compacted
soil) barrier layer.  Other layers, such as geotextiles, may be
required for specific purposes.


7.1  ASSESSMENT OF CLOSURE OPTIONS

     The anticipated use of the surface impoundment (i.e.,
treatment, storage, or disposal) influences the design and,
eventually, the type of closure option implemented.  Surface
impoundments that are used to store and treat wastes usually
require removal and disposal elsewhere of all contaminated
material following the active life of the facility.  Disposal
impoundments, on the other hand, are usually  (but not always)
closed in place and require a post-closure care period.  Table  13
presents advantages and disadvantages of the two closure options.

     Several factors should be evaluated in choosing the closure
option for a particular site:

     •  waste characteristics, including toxicity, mobility,
        leachability, reactivity, biodegradability, and
        degradation by-products;

     •  site location features, including topography, geology,
        climate, geohydrology, and proximity  of highly populous
        or environmentally sensitive areas;


                                132

-------
   TABLE 13.   ADVANTAGES AND DISADVANTAGES OF CLOSURE OPTIONS.
          Advantages
          Disadvantages
  No post-closure monitoring
  and care period

  Simpler in concept,  requires
  less administrative  time

  May be suited for environ-
  mentally sensitive areas

  May be best option for
  highly toxic materials

  Land may be used for other
  purposes after clean closing
                           -Removal-
•  Current regulations require
  removal of all contaminated
  liner and subsoils containing
  levels of waste constituents
  above background conditions.

•  Transportation costs often high

•  Risks associated with trans-
  portation

•  Generator keeps long-term
  liability for waste removed to
  other management facilities
                .—_———_-——Contai nment——-*——————.-.
                 —Stabilization/Solidification—
•  Stabilization technology
  available for most wastes

•  Work completed on site,
  generator keeps control

•  May offer generators a cost-
  effective option, especially
  those not near a commercial
  RCRA facility

•  Hay petition for shortened
  post-closure period
  Work completed on site,
  generator keeps control

  May offer generators a cost-
  effective option, especially
  those not near a commercial
  waste-disposal facility.

  May file delisting petition

  May petition for shortened
  post-closure period
• 30-yr post-closure monitoring

• Liability associated with
  potential threat to environment
  as a result of leaking landfill

• Land use is limited

* State administration fees for
  hazardous waste programs

• Financial assurance required
  for post-closure period r
                           Treatment-
  30-yr post-closure monitoring
  if no variance is received

  Land use limited if monitor-
  ing variance not received

  Financial assurance required
  for post-closure period

  State fee for administration
  of hazardous waste program
                                133

-------
     *  cost?

     *  intended future site use; and

     •  environmental risk.


7.1.1  Waste Characteristics

     The removal option is advantageous for surface impoundments
with subsoils or residual materials that cannot be treated
sufficiently, or within reasonable cost, to remove environmental
threats.  For example, a surface impoundment containing leachable
or volatile hazardous wastes or liquids that cannot be solidified
or stabilized to yield consolidated wastes of sufficient strength
to support the final cover might be better closed by materials
removal (USEPA, 1982a).  Inorganic wastes easily treated to a
low-mobility state might be better suited to in-place closure.


7.1.2  SiteLocation Features

     In-place closure of surface impoundments near
environmentally sensitive areas (e.g., wetlands, fault zones,
flood-prone areas, areas with shallow ground water, or areas with
highly permeable subsoils) may pose a significant environmental
risk.  Similarly, in-place closure of surface impoundments near
major population centers may increase the future risk to human
health.  In these cases, removal and treatment or remote disposal
are preferred.  On the other hand, some hazardous wastes (e.g.,
containing a significant radiation component) may pose less of a
threat if left in place and secured there.


7.1.3  Cost

     Closure of a surface impoundment involves immediate closure
expenses, plus (if it is in-place closure) longer-term costs  for
maintenance and monitoring.  Closure costs depend on factors  such
as availability and complexity of in-place treatment
technologies; labor and equipment costs; and proximity to an
alternative off-site hazardous waste treatment facility.  If  both
removal and in-place closure are viable alternatives, the closure
decision may be based on a comparison of cost estimates.

     Complete removal or clean closure eliminates long-term
liability and the need for post-closure maintenance and
monitoring costs.  However, if soils have been contaminated,  the
necessity for removing them may increase the closure costs
dramatically.  In-place closure requires a 30-year post-closure
monitoring and care period and associated costs.  A post-closure-,
financial assurance bond is required at the beginning of the

                               134

-------
post-closure care period (40 CFR 264.140).   In addition, further
costs will result if remedial actions are required due to
improper in-place closure operations (e.g.,  faulty design or
construction of the cover system), which lead to migration of
hazardous leachate from the impoundment.


7.1.4  Intended Future Site Use

     The potential future use of the site should be considered
during the assessment of closure options.  Clean closure is the
better choice if unlimited use is desired.   If in-place closure
is an option being considered, the planned site use should be
such that it poses no problems with maintenance of the final
cover or functioning of monitoring systems.   Compatibility of
impoundment and cover system components with various site uses is
discussed in Section 7.4.3.
7.1.5  Environmental Risk

     Both short-term and long-term environmental risks are
associated with surface impoundment closure.  Short-term risks
occur during closure and involve handling and transporting
wastes, worker safety, and mobilizing hazardous constituents by
disturbing the wastes.  These risks are potentially greater for
closure through removal than for in-place closure.

     Long-term risks are generally associated with in-place
closure, and include the potential for slow release of hazardous
constituents into ground water or nearby surface water through
cover or liner deterioration.  Long-term risks may also be
associated with off-site disposal which may be involved during
clean closure.  Because of the long-term costs and liabilities
associated with surface impoundment closure, it is important to
evaluate the potential environmental risk which may result from
cover failure during or after in-place or clean closure.


7.2  CLEAN CLOSURE  (CLOSURE BY REMOVAL)

     Clean closure  of a surface impoundment requires removal or
decontamination of  all wastes, contaminated liner system
components, structures, subsoils, and equipment (USEPA, 1984a).
As stated earlier,  this option requires the removal of all
hazardous constituents to background levels.  The owner/operator
of a RCRA-permitted surface impoundment may find that it is not
possible to comply  with the clean closure plan because waste
constituents have migrated to great depths and removal to
background levels is not feasible.  In that case, current federal
policy requires that the surface impoundment comply with in-place
closure requirements  (US1FA, 1984e).  For clean closure of

                               135

-------
interim status facilities, it may not be necessary to remove all
subsoils containing hazardous constituents at concentrations
above background levels, if the levels can be shown to be
nonhazardous via 40 CFR Part 261.3(d).  A plan describing the
sampling and analysis procedures to verify decontamination of the
site should be submitted as part of the closure plan.


7.2.1  Free Liquids

     Regulations (40 CFR 264.314) disallow the disposal of free
liquids in a hazardous waste landfill.  This has special
significance for surface impoundments since landfilling is
ordinarily the ultimate disposal method, regardless of whether
clean closure or on-site closure is chosen.

     Several options may be considered in dealing with the
impounded hazardous liquid during clean closure.  The options
include (1) natural evaporation? (2) forced evaporation? (3)
other on-site treatment (e.g., chemical neutralization); (4)
solidification or stabilization; (5) off-site treatment; and (6)
recovery and process re-use.  Some consideration should have been
given to treatment methodology in the impoundment planning and
design process, in anticipation of closure.

     Criteria used to select a treatment option include technical
feasibility, reliability, effectiveness, regulatory acceptance,
and cost.  Methods used to evaluate these options include
literature review of treatment technologies and vendor
interviews, bench-scale studies, pilot tests, cost estimation,
and consultation with regulatory agencies  (Stevens, 1986).

     On-site liquid treatment can use technologies developed by
the wastewater treatment industry in recent years, including
mobile or temporary treatment units (USEPA, 1982f).  In many
cases, the treated liquids are released either to a publicly
owned treatment works (POTW) or directly to surface waters if
they meet effluent requirements of federal, state, and local
agencies (Crawley et al,  1984; Hale et al, 1983).

     Disposing of liquids through natural  evaporation is perhaps
the least expensive method, but requires a suitable climate and
can be subject to air emission regulations.  However, the
impoundment often will have been operating as an evaporation
impoundment through its active life.  With the cessation of waste
liquid input, it would be allowed to dry up.

     Solidification and stabilization by the addition of a
solidification agent will ordinarily be one of the less expensive
alternatives.  After solidification, the waste material can be
removed and transported to a permitted  landfill.


                                136

-------
     Other on-site and off-site treatment can be very costly,
with off-site usually being the most expensive.   Much of the cost
can be allotted to transport.  Transport of hazardous liquid
wastes is controlled by the U.S. Department of Transportation,
which requires the use of proper containment vessels, manifests,
placards, and other handling procedures (40 CFR Part 172).

     Alternatives to treatment and disposal of free liquids
(e.g., reuse or waste reduction) are often more cost-effective;
however, these technologies must be implemented during or prior
to the facility's active life rather than during closure.


7.2.2  Residual Sludges

     Residual sludges in a hazardous waste surface impoundment
will always be considered hazardous materials and require removal
or treatment.  The primary removal option for residual solids is
transport to an off-site, RCRA-approved disposal facility (e.g.,
an incinerator or landfill).  To reduce transportation and
disposal costs, the sludges may be dewatered or stabilized before
off-site transport, provided the separated liquids can be treated
and properly disposed.  Alternatively, residual solids can be
transported and stabilized at the receiving facility.  Methods to
remove surface impoundment sludges for stabilization or
dewatering are discussed in Section 5.1.2.4.  Stabilization
techniques are discussed in Section 7.3.3.1.


7.2.3  Subsoils, Liners, and Other Contaminated Materials

     Contaminated liner system components and subsoils must also
be treated or removed and transported to a permitted hazardous
waste disposal facility.  This is the final step in waste removal
from a surface impoundment.  Documentation showing that all
contaminated material has been removed is essential.

     All equipment and support structures that are not removed
must be decontaminated (USEPA, 1986f; Meade and Ellis, 1986).
These structures include unloading areas, outfalls, leak
collection sumps, pumps, level detection stations, and other
structures that have contacted the waste.
7.2.4  Verification Sampling

     The owner/operator  should provide analytical data to verify
that closure was  completed  satisfactorily and that no significant
environmental  hazard  remains.  Therefore, a plan to sample and
analyze the residual  materials, subsoils, adjacent areas, and
equipment  should  be developed and submitted as part of the
closure plan.   This plan should include sampling procedures,

                               137

-------
location and depth of soil samples, indicator parameters,
analytical procedures, and sampling methods to document the
decontamination of equipment and site soils.

     Owners/operators who plan to close an impoundment using the
removal option should submit procedures in the sample plan to
establish chemical constituent levels for soils and ground water
at or below which no significant threat is anticipated.
Normally, these are background levels.  For new surface
impoundments, background samples should be collected before
construction.  For existing surface impoundments,  background
values for chemical constituents can be established by collecting
samples in an adjacent area with soil conditions similar to those
at the impoundment site before construction.   Efforts should also
be made to establish constituent background levels for borrow
materials used to construct liners, foundations, and dikes.
Establishing constituent background values is important because
the clean-closure option may use these values as standards to
verify adequate treatment or removal.

     Soil and ground-water background sample data should be
statistically evaluated to assess the data's spatial variability.
The closure plan should contain a method for comparing the
background data set with the post-closure data set.  Information
on this subject can be found in publications by Keith et al
(1983), Mausbach et al (1980), and Beckett and Webster (1982).

     In most cases, verifying clean closure is limited to
collecting and analyzing samples from the liner materials and
immediately adjacent subsoils.  For most surface impoundments,
ground-water monitoring wells have already been installed and are
being monitored routinely at the time of closure.  Ground water
should be sampled and analyzed during verification sampling.


7.2.4.1  Sampling Sch«n*s —

     Verification sampling schemes applicable to closed surface
impoundments include simple random sampling, stratified random
sampling, systematic sampling, and judgmental sampling (USEPA,
1983b).  Each scheme requires developing protocols for sampling,
sample handling and analyses, and adherence to a QA/QC program.
Each sampling scheme must also include provisions for collecting
soil samples at a number of specified depths to develop a
constituent concentration profile with depth.

     The systematic sampling approach uses grids or transects to
provide sample locations.  Because surface impoundments typically
are regularly shaped structures, this procedure is an effective
sampling method and is recommended for use in the process of
delisting hazardous wastes at surface impoundments  (USEFA,
1985h).  The procedure requires that the entire facility be

                                138

-------
divided into unit areas of equal size which do not exceed a
specified surface area (e.g., approximately 10,000 ft2 [930  m2]).
Each unit area is further divided using a grid, representing
potential sampling locations, which are selected using random
numbers.  The number of samples collected from each unit quadrant
depends on the degree of spatial variability within the facility.
The samples collected from within a unit area are composited to
form one homogeneous sample for analyses.  During the planning
stage, it is recommended that a reference be consulted (e.g.,
USBPA, 1983f and USEPA, 1984k).


7.2.4.2  Indicator Parameters —

     Ground-water and soil samples suspected of being
contaminated should be analyzed for certain basic parameters
indicative of contamination.  Analysis for indicator parameters
screens samples and identifies those samples requiring more
detailed testing to determine the type and concentration of
specific contaminants.  Choice of indicator parameters depends on
the wastes stored, but commonly includes the following:

     »  total organic carbon  (TOC): indicator of organic
        compounds;

     •  electrical conductivity: indicator of soluble ions;

     •  pH: indicator of acidic or basic conditions;

     »  oil and grease: indicator of petroleum-based products;

     »  total organic halogens (*TOX):  indicator of halogenated
        organic compounds; and

     •  toxicity characteristic leaching procedure  (TCLP):
        indicator of leachability of hazardous organic and
        inorganic compounds.

     Soils and ground water suspected of metals contam.1 nation
should be analyzed for total quantities of the specific metals.


7.2.4.3  Quality Assurance/Quality Control —

     The QA/QC program is implemented to minimize or eliminate
error associated with the sampling and analytical programs.  Each
item outlined in the sampling protocol must be defined clearly,
especially those concerning specific sampling locations.
Sampling protocol should include the following points:

     »  methods used to determine the number of samples to  be
        collected and the sampling point locations;          ~

                               139

-------
     •  depth(s) for sample collection;

     •  type(s) of equipment for sample collection?

     •  methods for decontaminating sampling equipment;

     •  compositing procedures; and

     •  a list of all analytical parameters, preservation
        methods, and handling and shipping procedures.

     Sampling error may be introduced into the sampling procedure
regardless of the precautions taken.  Thus, auditing the sampling
procedures may be worthwhile to determine the magnitude of
sampling error; for example, replicate samples can be analyzed to
monitor sampling procedures.  Detailed information on preparing a
QA/QC program has been presented by USEPA (1980e).

     The QA/QC program includes laboratory procedures for
chemistry analyses, analytical verification techniques,  data
handling and analysis, and personnel responsibilities.


7.2.5  Regulatory Variance

     In some if not most cases, removing all subsoil materials
containing hazardous constituents above background levels is not
feasible.  This situation is likely to occur in the case of an
existing surface impoundment which either does not have a liner
or has a liner that has been ineffective.  In such a situation,
closure leaving the remaining materials in place is necessary to
comply with federal regulations.  In some instances, the
materials remaining after free liquids, sludges, contaminated
liners, and subsoils are removed may have hazardous constituent
concentrations higher than background values, but below levels
that present a hazard to the environment or human health.

     Current RCRA guidance on facility closure  (USEPA, 1982a)
does not provide specific guidance on closure options for surface
impoundments with hazardous waste constituents present in
subsoils or liners at less than hazardous levels.  USEPA
proposed on March 19, 1987, in 40 CFR 265.310(c), conditions for
allowing "alternate closure requirements."  If the conditions are
met, the requirement could be waived for removing all materials
having contamination above background levels.  A waiver would
require that the owner/operator show that the closure performance
standard in 40 CFR 265.111 would still be met.  It may require
that "alternate concentration limits"  (ACL) be established for
the waste constituents in the soils at the site.

     Selecting ACLs for inorganic compounds may follow guidelines
for determining whether materials are considered hazardous, such

                               140

-------
as EP Toxicity testing (40 CFR 261.24) or past experience at
Superfund remediation (CERCLA) sites.  Selecting acceptable
values for organic compounds may be extremely difficult due to
the lack of information concerning the persistence or long-term
environmental hazard presented by these chemicals.  In situations
where some level of organic compounds is proposed to be left in
place, it is likely that the responsibility lies with the
owner/operator to demonstrate acceptable limits.


7.2.6  Backfilling

     The area can be backfilled after verification that the
closure area has been decontaminated.  No specific guidelines are
available for backfilling materials? however, a few
considerations are advisable.  Soil used for backfilling should
not be contaminated, and the selected soils should be analyzed if
contamination is possible.  Soils placed into the impoundment
area should be compacted to a dry bulk density at least as high
as the surrounding soils to prevent subsidence and differential
settlement.  Additionally, the backfilled soils should be
compatible with intended future site use.  Soil characterization
and engineering tests on the soils may be required before use.

     The area should be contour-graded to enhance and control
site runoff and reduce percolation and erosion.  Surface grading
and compacting should result in a surface sufficiently sloped to
prevent ponding.  Rishel et al (1984) present cost estimates for
contour-grading at surface impoundments.

     The surface soil (12 inches [30 cm] or more) should be of
sufficient nutritive capacity to support vegetation.  The area
should then be seeded with a suitable.grass to prevent erosion.


7.3  IK-PLACE CLOSURE

     In-place closure means closing the surface impoundment as a
landfill with a subsequent 30-year post-closure care period.
Facilities with intact liners at closure can petition the USEPA
Regional Administrator to reduce the post-closure period.  The
in-place closure option consists of four basic phases:

     •  solidification or removal, treatment, and disposal of
        free liquids;

     •  treatment and stabilization of residual sludges and
        contaminated soils;

     •  decontamination of equipment; and

     »  construction of a final cover;

                               141

-------
US1PA recommends that permit applicants address the following
items in their surface impoundment closure plans (USEPA,  1984e);

     •  detailed information on how free liquids will be  removed
        or solidified;

     *  detailed plans describing how the wastes will be
        stabilized;

     •  determination of the unconfined compressive strength and
        consolidation characteristics of the stabilized waste;
        and

    | •  analysis results showing that the stabilized wastes will
        provide sufficient, permanent support for the cover and
        other expected loadings.
     j

7.3.1Removal of Free Liquids

     Treatment and removal of free liquids for in-place closure
is similar to that for clean closure (discussed in Section
7.2.1).  An alternative to removal is solidifying the liquids in
place, and eliminating the need for off-site discharge or
removal.  Care must be taken during this activity to prevent
damage to the impoundment liner.  Solidification is discussed in
Section 7.3.3.1.
7.3.2  Sludge Dewaterinq

     Remaining sediments may require temporary removal from the
impoundment and dewatering to meet consistency requirements and
to ensure adequate handling characteristics for landfilling.  In
some situations, portable processing systems (e.g., clarifiers,
centrifuges, thickeners, or belt-filter presses) may be installed
at the closure site to accomplish active dewatering.

     Solidification agents may also be added instead of or in
addition to dewatering.  In arid or semi-arid climates, sediments
can be dewatered using passive means (evaporation or free
drainage).  This technique usually involves constructing a drying
area within the impoundment or drying beds outside the berms.
Collected water requires proper treatment and disposal {see
Section 7.2.1).  Once sufficient liquids are removed so that the
sediments have a semi-solid consistency (i.e., will not flow),
the sediments can be mechanically transported (by clamshell,
auger, dragline, or dozer) and spread in the surface impoundment.
                               142

-------
7.3.3  Waste Residuals

     Surface impoundments closed in-place require stabilization
or treatment of the residual solids.  The residual solids nay be
a combination of the impoundment's accumulated bottom sediment
and the residual from treating the liquid portion.  Stabilization
of these solids increases the bearing capacity to a degree
sufficient to support the final cover.  Treating certain waste
residuals can render them nonhazardous,  thereby increasing the
flexibility of the post-closure care program.


7.3.3.1  Stabilization —

     Sludges and other solids in surface impoundments are likely
to possess poor physical and structural properties; therefore,
stabilizing them to a semi-solid consistency is usually required
to provide adequate containment of waste constituents.
Stabilizing waste residuals improves their physical properties by
two primary means:  (1) increasing the density, which reduces the
compressibility of waste residuals and the resultant potential
for settlement of the final cover; and (2) decreasing the
permeability of the waste residuals and thus the mobility of
pollutants in the residuals (Anderson and Jones, 1982).

     Stabilization techniques have become important remedial
operations, largely due to CERCLA (Superfund)  remediation
activities.  The goal in applying these techniques is to produce
a solid, chemically non-reactive material.  Several methods exist
for mixing the wastes with stabilizing agents, including in-drum
mixing, in-situ mixing, mobile mixing plants,  and area mixing.
These methods and techniques have been outlined by USEPA (1986g).

     Most stabilization techniques are proprietary processes that
involve adding absorbents and solidifying agents to the residual.
These processes generally involve one of the following:

     *  sorption
     *  pozzolan formation
     *  encapsulation

Detailed discussions of these processes are contained in reports
by USEPA (1982b, 1982f, 1986g, and 1989e), Spooner (1985), and
Anderson and Jones (1982).

     Sorption involves adding a dry, solid substance to a liquid
or semiliquid waste to take up free liquid and improve waste-
handling characteristics.  Common sorbents include fly ash, lime
kiln dust, cement kiln dust, zeolites, and soil.  This method can
be implemented using readily available equipment at relatively
low cost.  The disadvantages include increased material for
disposal and relatively high leaching loss of potential

                               143

-------
contaminants from the stabilized wastes,  thus requiring a secure
disposal (Spooner, 1985).

     Lime/fly ash pozzolan and Portland cement pozzolan processes
use solidifying agents (e.g.,  hydrated lime or cement)  along with
pozzolans to increase strength and durability.  The suspended
solids in a waste slurry become incorporated into the hardened
concrete matrix, increasing the strength and decreasing the
permeability of the mixture*  lowever, a number of compounds or
materials (e.g., oil and grease, calcium sulfate, sodium borate,
and organic matter) can weaken the waste/cement bond and decrease
the physical strength.

     Encapsulation isolates wastes by completely surrounding them
with a durable, impermeable coating.   Containment of the waste is
complete and assured for the life of  the coating material.
Thermoplastic microencapsulation involves mixing dried wastes
with materials (e.g., asphalt, paraffin,  polyethylene,  and
polypropylene) and placing the mixture in a mold.  Some of these
processes are adaptable to highly soluble toxic substances which
are not amenable to lime or cement-based techniques.  The
disadvantages of this process include the cost of materials, and
the need for specialized equipment, skilled labor, and energy.

     A bench-scale study,  in which the wastes are mixed with
various stabilizing agents, is usually performed to select the
appropriate stabilization technique.   Standards for testing the
stabilized wastes have not been developed; however, a testing
scheme should be developed for each individual situation.  The
tests should be conducted on representative waste residual
samples that have been subjected to the proposed stabilization
technique.  Various tests are described by USEPA  (1989e).

     The two most important engineering properties of stabilized
wastes that must be addressed in the closure plan are unconfined
compressive strength and compressibility.  These engineering
properties are used in engineering calculations to demonstrate
that the stabilized waste has sufficient strength to support the
maximum anticipated loadings that nay result from overburden,
cover, and equipment that will be used to close the facility
(USEPA, 1984e).  Further, these properties are used to estimate
the magnitude of long-term settlement of the cover as a result of
consolidation of the stabilized waste.  Unconfined compressive
strength and consolidation properties, as well as other
characteristics of the stabilized waste, should be determined
using procedures listed in Table 14.

     Other tests may be required for  specific situations  (e.g.,
closing existing surface impoundments with inadequate liners).
These tests include leaching rate, permeability,  free liquid
content, and biological consolidation (see Table  14).


                                144

-------
   TABLE 14.  TEST PROCEDURES FOR STABILIZED WASTES.
Parameter
Reference
      Comments
Unconfined
Compressive
Strength
Consolidation
Biological Oxygen
Demand
ASTM D 2166
Leaching Potential
Hydraulic
Conductivity
tests
ASTM D 2435
American Public
Health Assoc.
(1985)
Chan et al (1978)
Means & Parcher
(1963)
Adjust water content
and density to the same
as dewatered and
compacted sludge.

Adjust water content
and density to the
same as dewatered and
compacted soil.

Biological consoli-
dation estimated by
assuming 1 unit of
organic matter will be
destroyed for each 2
units of oxygen demand.

EPA considers this
shake test to be the
best currently
available.

Laboratory K tests
typically give lower
values than field.
This method also allows
measurement of
compressibility and
rate of settlement.
Free Liquid Content    Spooner (1985)
—=:==—= ====;
7.3.3.2  Treatment of Residues —

     The primary objective of treating residues is to render them
nonhazardous, thus limiting potential long-term liability and
possibly the duration of the post-closure care period.  In-situ
treatment uses chemical, biological,  or physical mechanisms to
degrade, remove, or immobilize contaminants.   Treating the
residual sediments is similar to treating soils.  Therefore, the
following discussions apply to dried sludges  and contaminated
soils.
                               145

-------
     Contaminated materials that are treated to render them
nonhazardous must be managed as hazardous wastes, unless they can
be shown to be exempt (delisted) via 40 CFR 261.3(d).  Wastes
considered characteristically hazardous must be formally delisted
via petition to the USEPA Regional Administrator (see Section
7.4.4).

     Extraction (Soil-flushing) — Extraction entails washing
contaminants from the soil with a suitable solvent (e.g., water,
surfactants, dilute acids or bases, or chelating agents).  During
elutriation, sorbed contaminants are solubilized, emulsified, or
reacted chemically with the flushing solution and are mobilized.
The washing solution is injected into the area of contamination,
and the contaminated elutriate is pumped to the surface for
removal or on-site treatment and reinjection.  This technique
probably has limited applicability for treating sediments in
lined surface impoundments, because of the difficulty of
establishing a subsurface circulation in the shallow confines of
the impoundment.  Care must be exercised in using dilute acids
and bases to prevent the undesired result of dissolving already-
fixed contaminants.  This technology is derived from the mining
industry, where it has been used for in-place extraction of
metals from ores (USEPA, 1984b; Wagner and Kosin, 1985? Ellis et
al, 1985).

     Immobilization — Immobilization techniques are designed to
reduce the rate of contaminant release from the soil so that
concentrations along exposure pathways are maintained within
acceptable limits.  The primary techniques include precipitation,
sorption, ion exchange, and solidification (USEPA, 1984f).

     Precipitation is primarily applicable to heavy metal
contaminants.  It can be accomplished by adding carbonates,
sulfides, phosphates, and hydroxides.  Within the correct pH
range, these compounds are very insoluble.  Potential additives
include calcium phosphate, calcium carbonate, calcium sulfide,
sodium sulfide, iron hydroxide, alum, and ferrous sulfate.  The
two major considerations influencing metal precipitation are pH
and rate of complexing agent application.  The chemistry of the
potential compounds must be evaluated to prevent creating soluble
compounds that will leach, and to prevent changes in the
chemistry (e.g., pH) over the long term that may lead to leaching
(USEPA, 1984f; Wagner and Kosin, 1985; Malone et al, 1983).

     Chemical neutralization is a type of immobilization which
lends itself to acidic, metal-containing wastes.  Neutralization
lowers corrosivity and immobilizes metals, rendering the waste
nonhazardous.  Calcium and magnesium hydroxides are the additives
most commonly used.
                                146

-------
     Application of additives to waste sediments can often be
accomplished with agricultural equipment.  In some cases, it is
possible to treat the sludges during the dewatering process.
Where application into deeper subsoil layers is required,
grouting equipment may be used to inject the additive.  Grouting
is a technique in which an aqueous suspension containing the
additive is pressure-injected into the subsoil, where it forms a
gel or solidifies in place.  This procedure has been used
successfully to treat acidic, metal-contaminated soils beneath
surface impoundments (Crawley et al, 1984).

     Sorption is potentially applicable to organic and inorganic
contaminated sediments.  A variety of natural and synthetic
materials can be added to soils to increase their natural ability
to sorb ions.  Suitable materials that have potential for sorbing
metals include various agricultural products and by-products
(e.g., peanut hulls, straw, bark, sawdust, and wastewater
sludge).  The pH of the residuals should be maintained above 6.5
for maximum adsorption efficiency.  Tetraethylenepentamine
(tetren) has been used successfully to form metal chelates.
These complexes are strongly sorbed by soil clays and are not
sorbed by organic matter.  Tetren must be applied to a soil
relatively high in clay to be effective.  The long-term stability
of tetren-metal complexes against decomposition or degradation is
not yet known (Wagner and Kosin, 1985| USEPA, 1984j).

     Certain natural clays, zeolites, synthetic resins, and other
colloids have the capacity to adsorb ions preferentially through
exchange reactions or to capture certain ions through stearic
hindrance in the crystal lattice.  The ability of a colloid to
sorb metals chemically is related to its cation exchange capacity
(CEC) and surface area.  Cation exchange capacity is an
electrical property of colloids and is defined by the number of
positively charged ions  (cations) which can potentially be
adsorbed by the colloid.  Naturally occurring zeolites typically
have the highest CEC values among natural earthen materials,
followed by high organic (peat) soils, and soils high in
vermiculite or smectic clay content.  Zeolites behave as
"molecular sieves," capturing certain ions in the crystal
framework while allowing others to pass through.  Increasing the
clay content or CEC of a dried sludge can increase capacity to
immobilize metallic cations  (USEPA, 1984J).

     Synthetic resins that can adsorb or chelate both cations and
anions have been developed.  However, resins are generally
expensive and have limited availability  (USEPA, 1984j? Wagner and
Kosin, 1985).

     Techniques used to  stabilize residues can be used to render
them nonhazardous if the waste constituents are adequately


                                147

-------
immobilized.  Sorption, pozzolan formation,  and encapsulation
processes are discussed in Section 7.3.3.1,

     Biodecrradation — Biodegradation is the breakdown of organic
compounds by microorganisms.  Biodegradation techniques include
land treatment and composting sludges {USEPA,  1983a;  Overcash and
Pal, 1979). Important factors to be considered in biodegradation
by land treatment include (1) soil conditions (i.e.,  pH, redox
potential, moisture content, texture, nutrient supply, and
temperature); (2) presence of toxic compounds; (3)
biodegradability of the contaminant; and (4) breakdown products.
Certain hazardous wastes may have limited data available on
degradation rates or by-products.  Therefore,  a laboratory or
greenhouse treatment demonstration study may be required to
demonstrate treatability of the waste and to determine optimum
degradation conditions.  If materials high in organic matter are
closed in place without adequate treatment,  biodegradation will
occur during post-closure.  This can lead to reduction of the
total mass of solids, consolidation, excessive production of
gases, and differential settling of the cover.

     Chemical Degradation ~ Chemical treatment of soils and .  .
other solids in situ should be carefully considered beforehand.
It is very difficult to treat the materials homogeneously.  The
mass may contain zones where treatment has been very effective
while containing other zones where virtually no reaction has
taken place.  The following discussion should be taken in light
of these probabilities.

     Oxidation and reduction reactions may be carried out in
place to transform soil contaminants into less toxic or less
mobile products.  Introducing chemical oxidants (e.g., ozone and
hydrogen peroxide) into the soil system will often promote
oxidation of organics.  These agents are applied as aqueous
solutions directly onto the soil surface or injected into vadose
zone subsoils or ground water.  Laboratory and field treatability
studies will be required to assess the reactions that occur and
to develop data for design of a full-scale treatment system if
results are favorable  (USIPA, 1984J).

     In-*8Jtu reduction may be accomplished by adding chemical
reducing agents  (USEPA, 1984j).  This process has been shown to
degrade toxic organic constituents.  Organic wastes that are
amenable to treatment include chlorinated organics, unsaturated
aromatics and aliphatics, and other organics  susceptible to
reduction  (USEFA, 1984j).  Possible reducing  agents include
catalyzed metal powders of iron, zinc, or aluminum.  These agents
can be applied to the  soil surface and mixed with contaminated
soil using conventional agricultural equipment.  Alternatively, a
grout slurry of the metal powder can be injected into the
subsurface via closely spaced wellpoints.


                                148

-------
     Certain inorganic constituents can be immobilized or
transformed to less toxic compounds through reductions.  For
example, hexavalent chromium and selenium have been successfully
treated in soils using reduction technology (USEPA, 1984J).  The
resulting reduced compounds are trivalent chromium, elemental
selenium, and selenite (Se[IV]).  These reduced compounds are
less toxic and less mobile in soil systems than their oxidized
analogues.


7,3.4  Final Cover System

     Federal regulations require constructing a cover system over
a hazardous waste disposal impoundment upon final closure.
Regulatory statutes (40 CFR 264.228) require that the cpver:

     "  provide long-term minimization of liquid migration
        through the closed facility;

     »  function with minimum maintenance;

     »  promote surface water drainage and minimize erosion or
        abrasion of the cover;

     •  accommodate settling and subsidence so that cover
        integrity is maintained; and

     •  have a hydraulic barrier with a permeability less than or
        equal to that of the lowest-permeability bottom liner or
        natural subsoil layer present immediately beneath the
        facility.

     USEPA  (1989a) provides guidance on designing and
constructing a final cover system that will exhibit these
characteristics.  The following discussion of cover design is
condensed from that publication.

     The recommended cover design is a multi-layered system
intended to minimize leachate formation by promoting surface
water drainage using slopes, water-retaining topsoil layers,
geotextile  filters, and porous drainage layers.  The ultimate
resistance  to water percolating into the waste is provided by a
composite hydraulic barrier layer, which consists of a compacted,
low-permeability soil component in direct contact with an
overlying geomembrane.  For underlying waste materials of low
strength, a geotextile or geogrid stabilization/ reinforcement
layer may also be required to support the hydraulic barrier
layer.  Figure 47 provides a schematic of the recommended cover
design, which consists of the following components, top to
bottom;
                               149

-------
      vegetation/soil
          top layer
      drainage layer

     low-permeability
 geomembrane/soil layer
            waste
                     \\// \\// \\h \\h \\l/
• ° • e-»*0° 1*0 °«* «•*« ° * O * • o
= . . .,".-. .•  • »?.°<>«.o »
*e°»o.°'e°*<>«4 ». o.o
                     0
                     03
                                   O
                               O
                        0
60cm
— .^— granular or geotextile filter
30 cm
— ~«— 20-mil (0.51 mm) geomembrane
60 cm   w/overlying protective geotextile

	•«— geotextile separation layer
       (for low-bearing-strength waste)


    (30 cm = approx. 1 ft)
       Figure 47.   USEPA-recommended  landfill cover design.


     •  protective surface  layer (vegetation component and
        topsoil layer)

     •  geotextile or granular filter         f

     •  drainage layer

     •  geotextile or sand  protection layer

     •  hydraulic barrier layer (geomembrane and compacted soil
        component)

     •  gas vent layer  (granular or geosynthetic)

     •  stabilization layer (if required)

The cover  should be constructed with a convex  (high in the
center) surface topography  that has gentle slopes, preferably
between 3  and 5 percent, after full settlement.   This shape
should effectively drain surface waters from the cover without
erosion of the protective surface layer (USEPA,  1979b).
Geosynthetic materials, such as geotextiles, may be incorporated
in and on  the surface layer to prevent erosion,  particularly
where slopes are necessarily steeper or where  water flows
persistently or at higher velocities.
                                 150

-------
7.3.4.1  Protective Surface Layer —

     The upper layer of the final cover system is a protective
surface layer composed of vegetative and topsoil components.  A
layer of armoring material, such as cobbles, may be substituted
under some (e.g., arid) conditions (Figure 48).  An armoring
layer is designed to prevent erosion and abrasion of the
underlying cover components, while functioning with minimum
maintenance.  If cobbles are used as an armor layer, a geotextile
or granular layer placed below it will aid in preventing erosion.

     Vegetation — The uppermost component of the protective
surface layer, the vegetative cover (see Figure 47), reduces
percolation into the cover system, shields the topsoil from rain,
stabilizes the soil against erosive and abrasive forces, binds
and anchors the soil to form a stable mass, increases evaporation
rates, and enhances site aesthetics.

     Selecting the vegetation species is important and depends on
factors such as climate, site characteristics, and soil
properties.  Deep-rooted plant species (especially shrubs and
trees) should be avoided to minimize root extension into the
drainage and hydraulic barrier layers, thus compromising cover
integrity.  The vegetation selected should provide a self-
maintaining cover that does not require perennial fertilization
or irrigation, and minimizes the maintenance required at the
closed facility.  Several references discuss available plants and
site selection criteria (USEPA, 1979b; 1983d? 1983g? and 1985i),
and Lee et al (1985).  A local extension service, consulting
agronomist, or SCS agent should be contacted for recommendations
on locally adaptable plant species and information on area crop
cultivation.

     In some regions of the country (e.g., those with arid and
semi-arid climates), establishing a vegetative cover is difficult
or impossible.  In these areas, a rock, cobble, or other armor
mulch layer approximately 5 to 10 centimeters  (2 to 4 inches)
thick may be substituted for the vegetation  (see Figure 48).
Again, geotextiles should be considered under rock or cobble
layers to aid in preventing soil erosion.

     Topsoil Layer — The topsoil component  (see Figure 47)
should be designed and constructed to support the vegetative
cover by allowing sufficient surface water to infiltrate the
topsoil and by retaining sufficient water to sustain plant growth
through drought periods.  Particle-size distribution, structure,
and organic matter content influence the quantity of available
water a given soil can supply and should be considered in
selecting the topsoil material.  In general, medium-textured
soils  (e.g., loam) have the best overall characteristics for seed
germination and plant root system development.


                               151

-------
        cobbles/soil
           top layer
        biotic barrier
          (cobbles)
       drainage layer
     low-permeability _
geomembrane/soil layer

       gas vent layer
             waste

V« ". .* **.*.*« ".* _ * • *
*»•  ••*,-.•*»*••*.»*,
• «•••«»•!• •.«.•• . *•'
. . . » . « '. • «*.»'« . .• • .
                         00
P 0
 0  O

                 60cm
                 30cm
30cm
                 60cm
                 30cm
geotextile fitter
geotextile filter

    20-mil (0.51 -rnm) geomembrane
    w/overlying protective geotextile

geotextile filter

geotextile fitter
                                       (30 cm = approx. 1 ft)
           Figure 48,
    USEPA-recommended  landfill cover
   design with optional layers.
     The  USEPA-recommended minimum thickness for the topsoil
component is 24 inches (60 cm).  Some geographic regions and
climatic  conditions may require a thicker component to provide
adequate  plant-available water.  A professional soil scientist
experienced in designing cover systems  should be consulted for
guidance  on soil material selection.

     The  topsoil is placed in uncompacted layers over a graded
granular  or geotextile filter overlying the drainage layer.  The
filter  reduces penetration of soil particles into the drainage
layer and,  therefore, clogging of the layer.


7.3.4.2  Drainage Layer —

     The  final cover system includes a  drainage layer (see Figure
47), located below the protective topsoil layer, which intercepts
percolating water.  When the precipitation rate is sufficient,
the percolating water migrates downward through the topsoil  layer
and into  the drainage layer.  When it meets the hydraulic barrier
layer,  it flows horizontally through the drainage medium to  an
outlet  located at the cover perimeter.

     If the drainage layer is constructed of soil, the USEPA
recommends a minimum thickness of 12 inches (30 cm) with a
minimum hydraulic conductivity of 3.9 x 10"3 inches/sec  (1 x  10~2
cm/sec) and a minimum bottom slope of 2 percent after allowance
for settlement.  The layer may be constructed of granular
drainage  material classified by the USCS as SP  (poorly graded
sand).
                                 152

-------
     Alternatively, the drainage layer may be constructed of
synthetic materials, such as geonets or prefabricated drainage
boards.  The layer must have a transmissivity equivalent to, or
greater than, the soil drainage layer described above.

     Constructing the drainage layer consists of placing a 12-
inch (30-cm) layer of granular material or an equivalent
synthetic directly over the geomembrane component of the
hydraulic barrier layer.  Protection against damage to the
geomembrane by equipment or personnel should be provided.  Fines,
if present, should be removed from granular material before
construction to prevent clogging of the drainage layer.  A
synthetic drainage layer may require a geotextile filter between
it and the overlying cover soil.


7.3.4.3  Biotic Barrier —

     In some locations, although not required, a biotic barrier
(Figure 48) may be installed to reduce potential intrusion by
burrowing animals or plant roots, which may damage the hydraulic
barrier layer and increase percolation through it.  Hokanson
(1979) found that a biotic barrier of 28 inches (70 cm),
consisting of cobbles, overlain by 12 inches (30 cm) of gravel
was effective.  The cobbles had sufficient mass to deter
burrowing animals and the large void spaces, which lacked water
and nutrients, acted as a barrier to downward plant root
development.

     Information is not yet available on an optimum thickness for
a barrier layer.  Until then, a biotic barrier thickness of 24
inches (60 cm) may be considered sufficient in most
circumstances, unless evidence is available to justify a
different thickness.  The biotic barrier is located between the
drainage layer and overlying topsoil.


7.3.4.4  Hydraulic Barrier Layer —

     The USEPA-recommended hydraulic barrier layer of the final
cover  system  (Figure 47) consists of two components:  a smooth-
surfaced, compacted soil component with a maximum field hydraulic
conductivity of 3.9 x  10"*  inches/sec  (1 x  10"7 cm/sec)  overlain
by a geomembrane.  The  geomembrane is placed in direct contact
with the soil, a compression seal is created by the overburden
pressure, and the two  components form a composite barrier to
percolating liquid flow.

     The recommended minimum thicknesses of the two components
are 24 inches  (60 cm)  for the compacted soil and 20 mils  (0.51
mm) for the geomembrane.  The actual thicknesses can be

                                153

-------
considerably greater, based on site characteristics,  soil,
synthetic material, and expected external forces (e.g.,
settlement and overburden pressures).   Constructing the  compacted
soil component and installing the geomembrane are analogous to
the practices used in liner construction (see Section 4.5.1).
Similar techniques, along with appropriate CQA procedures,  should
be used to construct the hydraulic barrier.   Additional
recommendations on barrier design and  construction are given by
USEPA (1986aj 1987bj and 1989), and information on developing a
CQA program is also given by USEPA (1986a).

     The hydraulic barrier geomembrane component may be  one of
several available synthetic materials.  It need not be of the
same material as the impoundment liner.  It is placed directly
above the compacted and smoothed soil  component and at the bottom
of the drainage layer.  The compacted  soil acts as a buffer and
foundation for the geomembrane, and the drainage layer provides
protection from overlying materials.   The drainage layer should
be inspected for materials that may damage the geomembrane.  If
the drainage layer contains gravel-sized particles, a geotextile
should be used to protect the geomembrane.  Appropriate  CQA
procedures, as discussed in Section 3.12, should also be
maintained to ensure the integrity of  the geomembrane liner
installation.

     Care in placing and compacting the soil barrier component is
essential to achieving a low-permeability hydraulic barrier.  The
compacted soil component serves as the hydraulic barrier to
leakage through the geomembrane.  Appropriate construction proce-
dures {see Sections 4.4.1 and 4.4.2)  and the CQA Plan (see
Section 3.12) must be followed to ensure that the completed two-
component hydraulic barrier achieves performance standards.

     As outlined in the CQA Plan, field-testing of the completed
soil component of the barrier layer is also strongly recommended.
Because these tests are time-consuming and can cause construction
delays, constructing a test barrier section analogous to the test
fill for liner construction (see Section 4.4.1.1), using the same
soil material and equipment to be used in the actual barrier, is
recommended.  Field and laboratory hydraulic conductivity can be
measured on the test section.  If the results are satisfactory,
the number of field measurements required on the actual  barrier
may be reduced.


7.3.4.5  Ga»-V«nt Layer —

     The gas-vent layer (Figure 48) is recommended in facilities
where gases may be generated by decay of biodegradable organic
matter buried within the closed facility.  The gases produced in
biodegradation are usually methane and carbon dioxide and,
without a venting system, may present a fire or explosion hazard.

                               154

-------
Other gases emanating from volatile materials may be malodorous
or toxic.  The recommended gas vent consists of a 12-inch (30-cm)
layer of porous granular or geosynthetic material similar to that
used to construct the drainage layer.

     Horizontal, perforated pipes placed laterally within the
layer collect the gas and vent it to the surface through vertical
riser pipes.  The gas may be vented to the atmosphere or
collected for disposal or treatment.  Adequate soil compaction
and geomembrane seals must be provided around the vertical riser
pipes.  Design and guidance information for gas-vent layers has
been provided by USEPA (1979b; 1985i? and 1989a).

     The gas-vent layer provides an exit for gases generated
within the facility, and should provide a buffer between the
waste and compacted soil barrier.  The vent layer granular
material may be placed over the waste and brought to design
elevation (minimum thickness of 12 inches (30 cm) before placing
and compacting the soil component of the hydraulic barrier layer.

     A vent layer of geosynthetic drainage material may be
substituted for a granular layer, if it can provide equivalent
performance.  Performance requirements should be based on actual
measured gas generation times a factor of safety.


7.3.4.6  Hydraulic Barrier Support Lay«r (Optional) --

     In some applications, the waste material may not provide a
suitable foundation, by itself, for the construction of the
hydraulic barrier.  In these cases, a geotextile may be used to
help support construction equipment and to provide increased
structural support for the hydraulic barrier and overlying
layers.  Several failure conditions require analysis for the
design of a geosynthetic support system.  Design guidance may be
found in several references (Bonaparte & Christopher, 1987?
Christopher, 1990; Christopher & Holtz, 1989; Fowler, 1982?
Fowler & Loemer, 1987; Koerner, 1990; and Koerner, 1988)


7.4  POST-CLOSURE ACTIVITIES

     Post-closure care and monitoring of the facility are
required if closure does not remove or decontaminate all wastes,
waste constituents, and contaminated components contained in the
surface impoundment.  Regulations in 40 CFR 264.117 require the
post-closure care period to continue for 30 years after final
closure.  The USEPA Regional Administrator may reduce or extend
the post-closure care period where conditions warrant.
Throughout the post-closure care period, the owner/ operator is
required (40 CFR 264.228[b]) to:


                               155

-------
     •  maintain the integrity and effectiveness of the final
        cover, including repairing the cover as necessary to
        correct the effects of settling,  subsidence, erosion, or
        other events;

     •  maintain the ground-water monitoring system and comply
        with other requirements of 40 CFR,  Subpart F;

     *  maintain and monitor the leak detection/collection
        system, where such a system is present in a double liner;
        and

     «  prevent runon and runoff from eroding or otherwise
        damaging the final cover.

Therefore, post-closure care at a minimum includes a periodic
inspection and maintenance program, especially of the final
cover, to assure the integrity of the system and prevent
migration of hazardous constituents into the environment.
Additionally, a ground-water monitoring and leak detection
program, which is generally a continuation of monitoring
activities from the active life of the facility, is required to
detect the migration of waste constituents and/or their presence
in the ground water.  Financial assurance will also be required
by the regulatory agency to ensure that funds are available for
the 30-year post-closure care activities.  Financial assurance
may be provided by means of a trust fund, surety bond, or letter
of credit (Rodensky, 1985).

     A written post-closure care activities plan must be
submitted to the regulatory agency prior to beginning post-
closure care.  The plan must include, at a minimum  (USEPA,
1984e) :

     *  a list of the components to be inspected and frequency of
        inspections;

     *  description of remedial action to repair damaged facility
        components;

     •  frequency of monitoring device sampling and analysis; and

     »  the design of runon controls to prevent erosion of final
        cover.

     The plan  should achieve a balance between the  specific  site
and design limitations and the need for post-closure care  (e.g.,
sufficient monitoring and maintenance to ensure the integrity of
the closed unit).  The climate, soil type, and cover design
should be considered in planning maintenance activities for  the
final cover system.  Further discussion on the post-closure  care
plan has been  provided by USEPA  (19B2a and 1984e).

                               156

-------
7.4.1  Monitoring

     Monitoring activities during the post-closure period include
sampling and analysis of ground water from wells located
upgradient and downgradient of the closed impoundment (see
Section 5.2.3).  Sampling methodology, sample-handling protocol,
frequency of collection, analysis, and analytical parameters
should be outlined in the post-closure care plan, based on
regulatory requirements and waste characteristics.  USEPA has
published guidance on ground-water monitoring well installation,
sample collection, and preservation (USEPA, 1985a).  Methods for
sample analysis are given by USIPA (1982b).

     Additional monitoring is required for leak-detection and
collection systems or other components designed to detect
releases from the facility.  Specific monitoring requirements for
these components should also be detailed in the written post-
closure plan.


7.4.2  Maintenance

     Maintenance during post-closure care consists of scheduled
inspections, general site maintenance (e.g., mowing vegetation),
and remedial repairs in the event of disturbances to the cover or
other components.  Inspections of the following components of the
surface impoundment will be required to ensure the continued
effective performance of the facility?

     *  the final cover system for signs of surface erosion,
        settlement and subsidence, vegetation damage, and clog-
        ging of drainage outlets;

     »  the leak-detection and collection system, gas vents, -and
        other facility components for potential performance
        problems and evidence of leachate;

     *  ground-water monitoring wells for evidence of damage,j
        tampering, or subsidence;                            •
                                                             I
     *  runon diversion structures for effectiveness; and

     *  the overall facility condition.

     Maintenance of the vegetation is required to ensure that it
remains sufficient to protect the cover from erosion.  Mowing,
watering, revegetation, fertilizer application, and herbicide or
insecticide application may be needed to promote plant growth.

     Remedial activities are required if components deteriorate.
They include repairing eroded areas, damage from settlement or
subsidence, stressed vegetation, runon control structures, or

                               157

-------
other facility components damaged by wind, water, and weathering.
Armoring with cobbles or with geosynthetic materials may often be
appropriate to repair and prevent further damage.

     Leak collection systems must be maintained in operating
condition.  These systems require pump maintenance and continued
liquid detection and disposal.


7.4.3  Use of thm Sit«

     Post-closure site use must be limited to activities that
will not disturb the integrity of the final cover, liner, or
other components of the closed facility.  Cover system design
must consider site conditions, surrounding land use, and
compatibility of the intended land use with maintenance of the
closed unit.  A discussion of post-closure site use is given by
USEPA (1982c).  Tables IS and 16 summarize the compatibility of
features with various future site uses for impoundments closed
in-place and by waste removal (clean closure), respectively.


7.4.4
     Delisting of the waste contained in the closed impoundment
is a potential option for eliminating or reducing the
requirements for post-closure care and monitoring.  If the wastes
contained in the surface impoundment prior to closure were listed
wastes  (as defined in 40 CFR 261) and closure operations rendered
them nonhazardous (also defined by 40 CFR 261), the waste may be
delisted or otherwise considered nonhazardous.  If so, post-
closure requirements may be eliminated or reduced, as well as the
requirement to notify local land authorities.  Delisting of
wastes, however, is considered on a site-by-site basis and USEPA
delisting authorities should be consulted before submitting the
delisting petition.  Guidance for submitting a delisting petition
is provided by USEPA (1985h).

     If hazardous waste or constituents remain in the closed unit
after closure  (i.e., in-place closure), a notice must also be
given to the local land use authority indicating the location and
dimensions of the unit.  The notice must contain a survey plan
prepared by a professional land surveyor, with a note that states
the owner's obligation to restrict disturbance of the site (40
CFR 264.116).  Additionally, a notice must be given in the deed
to the property stating that the land has been used to manage
hazardous waste, that land use is restricted, and that a survey
plan and record have been filed with local land authorities and
regulatory agencies  (40 CFR 264.119).
                                1S8

-------
   TABLE 15.  COMPATIBILITY OF SURFACE ZXPOUHDMEHT FEATURES AMD
              VARIOUS SITE USES FOR IH-PLACE CLOSURES.*
Design Features
Subsurface Water Control
Extraction
Well point system
Cut-off walls
Subsurface drainage
Surface Water Control
Cover
Grading
Surface water diversion
Levees / f loodwalls
Drainage /erosion control
Air Factors
Passive gas control
Active gas control
Control of bird hazard
to aircraft
Surface Area Factors
Covers
Access
Land Buffers

Bldgs

C
C
C
C

NC
NC
C
NC
RDA

NC
NC
C


C
C
NC
Site
Recre-
ation
Areas

RDA
RDA
C
C

C
C"
C
NC
RDA

C
C
C


C
C
C
Uses Uoon Closure
Parking
Areas

RDA
RDA
C
C

C
C
NC
NC
RDA

C
C
C


C
C
C
Agri-
culture

C
C
C
C

C
C
C
NC
C

C
NC
C


C
C
NC
Open
Spaces

C
C
C
C

C
C
C
C
C

C
C
C


C
C
C
C = compatible, NC - not compatible, RDA = requires design alteration

* from EPA (1982a).
  except ballparks
                               159

-------
   TABLE 16.  COMPATIBILITY OF SURFACE IMPOUNDMENT FEATURES AND
              VARIOUS SITE USES AFTER HAZARDOUS WASTE REMOVAL.*
                                    Site Uses Upon Closure
Design Features
       Recre-
       ation  Parking
Bldgs  Areas   Areas
 Agri-    Open
culture  Spaces
Subsurface Water Control
Wells
Subsurface drainage
Surface Water Control
Cover
Surface water diversion
Levees /f loodwal Is
Air Factors
Surface Area Factors

CIR
C

RDC
RDC
CIR
C
RDC

CIR
C

C
C
CIR
C
C

CIR
C

C
NC
CIR
C
C

CIR
C

C
C
CIR
C
C

CIR
C

C
C
CIR
C
C
CIR = compatible if removed & dismantled, C
    requires consideration.

*  from EPA (1982a).
                   compatible, RDC =
                               160

-------
                            REFERENCES


Allis, J. A., B. Harris, and A. L. Sharp.  1963.  A Comparison of
     Performance of Five Rain-Gauge Installations.  In Journal of
     Geophysical Research Vol. 68.  pp. 4723-4729.

American Public Health Association.  1985.  Standard Methods for
     the Examination of Water and Wastewater.  Washington, DC.
     1268 pp.

American Society of Testing and Materials.  1986a.  Annual Book
     of ASTM Standards Construction.  Section 4, Vol. 04-08.
     Philadelphia, PA.

American Society of Testing and Materials.  1986b.  Standard Test
     Method for Unconfined Compressive Strength of Cohesive Soil.
     D 2166-85.  ASTM Committee D-18.  In Annual Book of ASTM
     Standards.  Philadelphia, PA.

American Society of Testing and Materials.  1986c.  Standard Test
     Method for One-Dimensional Consolidation Properties of
     Soils.  D2435-80.  ASTM Committee D-18.  In Annual Book of
     ASTM Standards.  Philadelphia, PA.

American Society of Civil Engineers.  1960.  Research Conference
     on Shear Strength of Cohesive Soils.  New York.

American Society of Civil Engineers.  1986.  Use of In-situ Tests
     in Geotechnical Engineering.  In In-situ '86.  New York.

Anderson, D. C., J. O. Sai, and A. Gill.  1984.  Surface
     Impoundment Soil Liners: Permeability and Morphology of a
     Soil Liner Permeated by Acid, and Field Permeability Testing
     for Soil Liners  (Draft Report - unpublished).  Prepared by
     K.W. Brown and Associates under EPA Contract 68-03-2943.

Anderson, D. C. and S. G. Jones.  1982.  In-Place Closure of
     Hazardous Waste Surface Impoundments  (Unpublished report).
     USEPA Contract 68-03-2943, WA 8.  97 pp.

Anderson, D. C. and S. G. Jones.  1983.  Clay Barrier-Leachate
     Interaction.  In Proceedings of the Fourth National
     Conference on Management of Uncontrolled Hazardous Waste
     Sites.  Washington, DC.


                               161

-------
Anderson, R. D. et al.  1978.  Soil Improvement History,
     Capabilities, and Outlook.  ASCE, New York.

Barren, R. A.  1977.  The Design of Earth Dams.  In Handbook of
     Dam Engineering, edited by R. Golze.  Van Nostrand Reinhold
     Co., New York.  pp. 291-318.

Bass, J. M.  1985.  Leachate Collector Systems - Summarizing the
     State-of-the-Art.  In Proceedings of the Eleventh Annual
     Research Symposium.  EPA/600/9-85/013.  Cincinnati, OH.

Battelle Pacific Northwest Laboratory.  1982.  Rock Riprap Design
     Methods and Their Applicability to Long-Terat Protection of
     Uranium Mill Tailings Impoundments.  NUREG/CR-2684.
     Prepared for the Nuclear Regulatory Commission, Washington,
     DC.  62 pp.

Beckett, P. I. T. and R. Webster.  1982.  Soil Variability: A
     Review.  In Soils and Fertilizers, vol. 34.  pp. 1-15.

Black, R. F.  1954.  Precipitation at Barrow, Alaska, Greater
     than Recorded.  Eon. Trans.  AGU 35s 203-207.

Bonaparte, R. and Christopher, B* R.  1987.  Design and
     Construction of Reinforced Embankments over Weak
     Foundations.  Record 1153.  Transportation Research Board,
     Washington, DC.

Bonaparte, R. and Gross, B. A.  1990.  Field Behavior of Double-
     Liner Systems.  In Waste Containment Systems; Construction.
     Regulation, and Performance, Edited by R. Bonaparte.
     Geotechnical Publ. 26, ASCE.  pp. 52-83.

Bowles, J. E.  1984.  Foundation Analysis and Design.  McGraw-
     Hill, New York.  578 pp.

Brady, N. C.  1974.  The Nature and Properties of Soils,
     MacMillan, New York.

Brown, K. W., G. B. Evans, Jr., and B. D. Frentrup.  1983.
     Hazardous Waste Land Treatment.  Butterworth Press, Woburn,
     MA.

Brown, K. W., J. C. Thomas, R. L. Lytton, P. Jayawickrama, and S.
     C. Bahrt.  1986.  Quantification of Leak Rates through Holes
     in Landfill Liners  (Unpublished report).  Prepared under
     USEPA Cooperative Agreement CR810940.

Bryant, S. M., J. M. Duncan, and E. B. Seed.   1983.  Application
     of Tailings Flow Analysis to Field Testing.  Report
     submitted to U. S. Bureau of Mines.
                                162

-------
Canter, L. W. and R. C. Knox.  1985.   Ground Water Pollution
     Control.  Lewis Publishers, Inc.   529 pp.

Chamberlain, E. J.  1981.  Comparative Evaluation of Frost-
     Susceptibility Tests.  Transportation Research Record 809.

Chamberlain, E. J., P. N. Gas kin, D.  Esch, and  R. L. Berg.  1982.
     Identification and Classification of Frost-Susceptible
     Soils.  ASChE Spring Convention,  Las Vegas,  NV.

Chan, K. ¥., B, G. Davey, and H. R.  Geering.  1978.  Interaction
     of Treated Sanitary Landfill Leachate with Soil.  Jour, of
     Environmental Quality, vol. 7.   pp.  300-310

Christopher, B. R. et al.  1990.  Design and Construction
     Guidelines for Reinforced Soil  Structures.  2 volumes.
     Federal Highway Administration,  Washington,  DC.

Christopher, B. R. and Holtz, R. D.   1989.  Geotextile
     Engineering Manual.  HI-89-050.   Federal Highway
     Administration, Washington, DC.

Church, H, K.  1981.  Excavation Handbook.  McGraw-Hill, New
     York.

Clements and Associates.  1985.  Chemical, Physical, and
     Biological Properties of Compounds Present at Hazardous
     Waste Sites.  EPA Unpublished Report.

Cochran, S. R., M. Kaplan, P. Rogoshewski, and C. Furman.  1983.
     Survey and Case Study Investigation of Remedial Actions at
     Uncontrolled Hazardous Waste Sites.  In Land Disposal of
     Hazardous Waste - Proceedings of the Ninth Annual Research
     Symposium.  EPA-600/9-83-018.

Considine, D. M.  (Ed.)  1974.  Process Instruments and Control
     Handbook.  Second Edition.  McGraw-Hill, New York.  pp. 6-
     162 to 6-175.

Cooper, J. W. and D. W. Shultz.  1983.  Development of Liner
     Retrofit Concepts for Surface Impoundments.  In Land
     Disposalof Hazardous Waste; Proceedings of the NinthAnnual
     Research Symposium,  EPA-600/9-83-018.  Municipal
     Environmental Research Laboratory, Cincinnati, OH.  pp. 211-
     218.

Corotis, R. B.  1976.  Stochastic Considerations in Thunderstorm
     Modeling.  Jour. Hydraulics Div., ASChE, vol. 102.  pp. 865-
     869.
                               163

-------
Crawley, W. W., K. W. Brown,  and D.  C.  Anderson.   1984.   Inplace
     Closure of Previously Backfilled and Active  Surface
     Impoundments.  In Proceedings of the 5th National Conference
     on Management of Uncontrolled Hazardous Waste Sites.
     Washington, DC.  pp.  185-188.

Crawley, W. W., R. L. Shiver, and D. C. Anderson.  1985.  In-situ
     Sampling of Hazardous Waste Surface Impoundments.  In
     Proceedings of the 6th National Conference on Management of
     Uncontrolled Hazardous Waste Sites,  pp. 80-83.

Creager, W. P., J. D. Justin, and J. Hinds.   1945.  Earth, Rock-
     fill, Steel, and Timber Dams - Vol. III.  In Engineering for
     Dams.  John Wiley & Sons, New York.  pp. 749-769.

Cudahy, J. J. and R. L. Sandifer.  1980.  Emissions Control
     Options for the Synthetic Organic Chemicals  Industry: Second
     Emissions Report.  USEPA Contract 69-02-2577.

Daniel, D. E., D. C. Anderson, and S. S. Boynton.  1985.  Fixed-
     Wall vs. Flexible-Wall Permeameters.  In Hydraulic Barriers
     in Soil and Rock.  ASTM STP 874.  329 pp.

Daniel, D. E. and S. J. Trautwein.  1986.  Field Permeability
     Tests for Earthen Liners.  In ASCE Specialty Conference on
     Use of In-Situ Tests in Geotechnical Engineering.  Virginia
     Polytechnical Institute and State University.

Davis, J. L., R. Singh, M. J. Waller, and P. Gower.  1983.  Time-
     Domain Refleetometry and Acoustic-Emission Monitoring
     Techniques for Locating Liner Failures.  Final Draft Report
     to EPA by EarthTech Corp.

Day, S. R. and D. E. Daniel.  1985.  Hydraulic Conductivity of
     Two  Prototype Clay Liners.  Jour. Geotech.  Eng., Vol III.

Dobbs, D. M. and A. B. Selber.  1986.  Risk Management:
     Personnel, Equipment and Indemnity.  In Proceedings of the
     National Conference on Hazardous Wastes and Hazardous
     Materials.  Atlanta, GA.  pp. 346-347.

Ellis, W. D., J. R. Payne, and G. D. McNabb.  1985.  Treatment of
     Contaminated Soils with Aqueous Surfactants.  EPA/600/2-
     85/029.  Hazardous Waste Engineering Research Laboratory,
     Cincinnati, OH.

Elsbury, B. R., J. M. Norstrom, D. C. Anderson,  J. A. Rehage, J.
     O. Sai, R. L. Shiver, and D. E. Anderson.   1985.  Optimizing
     Construction Criteria for Hazardous Waste Soil Liners.
     Phase I Interim Report to USEPA Hazardous Waste Engineering
     Research Laboratory.  USEPA Contract 68-02-3250.   89 pp.


                               164

-------
Farmer, W. J., M. S. Yang, and J. Letey.  1978.  Land Disposal of
     Hazardous Waste:  Controlling Vapor Movement in Soils.  In
     Proceedings of the 4th Symposium on Hazardous Waste.
     EPA/600/9-78-016.

Federal Highway Administration.  1980.  Proceedings of a
     Symposium on Site Exploration in Soft Ground Using In-situ
     Techniques.  U.S. Dept. of Transportation, Washington, DC.

Federal Highway Administration.  1985.  Geotextile Engineering
     Manual.  U.S. Dept. of Transportation, Washington, DC,

Federal Highway Administration.  1990.  Geotextile Design and
     Construction Guidelines.  U.S. Dept. of Transportation,
     Washington, DC.

Ferguson, H. L. and V. A. Znamensky.  1981.  Methods of
     Computation of the Water Balance of Large Lakes and
     Reservoirs: Vol. 1, Methodology.  In Studies and Reports in
     Hydrology No. 31.  UNESCO, Paris, France.  120 pp.

Finkbeiner, M. A. and M. M. O'Toole.  1985.  Application of
     Aerial Photography in Assessing Environmental Hazards and
     Monitoring Cleanup Operations at Hazardous Waste Sites.  In
     Proceedings of the Sixth National Conference on Management
     of Uncontrolled Hazardous Waste Sites.  Washington, DC.  pp.
     116-124.

Fowler, J.  1982.  Theoretical Design Considerations for Fabric
     Reinforced Embankments.  In Proceedings of the Second
     International Conference on Geotextiles.  Industrial Fabrics
     Association International.  Las Vegas, NV.  pp. 665-676.

Fowler, J. and Loemer, R. M.  1987.  Stabilization of Very Soft
     Soils Using Geosynthetics.  In Proceedings:  Geosynthetics
     '87.  Industrial Fabrics Association International.  New
     Orleans, LA.  pp. 289-300.

Franson, M. A. H.  1985.  Standard Methods for the Examination of
     Water and Wastewater:  16th Edition.  American Public Health
     Association.  1268 pp.

Gallant, R. W.   1965-1969.  Physical Properties of Hydrocarbons.
     In Hydrocarbon Processing Journal. Vols. 46-48.

Gillham, R. W., M. J. L. Robin, J. F. Barker, and J. A. Cherry.
     1983.  Groundwater Monitoring and Sample Bias.  American
     Petroleum Institute.   206 pp.

Giroud, J. P.  1985.  Geotextiles and Geomembranes - Definitions,
     Properties  and Design.  Industrial Fabrics Association
     International, St. Paul, MN.

                               165

-------
Giroud, J. P. and C. Ah-Line.  1984.  Design of Earth and
     Concrete Covers for Geomembranes.  In Proceedings of the
     International Conference on Geomembranes.  Vol.  II.  Denver,
     CO.  pp. 487-492.

Giroud, J. P. and R. Bonaparte.  1989.  Leakage through Liners
     Constructed with Geomembranes — Part II.   Composite Liners.
     In Geotextiles and Geomembranes.  Elsevier Science
     Publishers, England,  pp. 71-111.

Geo-Slope Programming, Ltd.  1985.  PC-Slope.  Calgary, Alberta,
     Canada.

Goldman, L. J., R. S. Truesdale, and C. M. Northeim.  1985.  Clay
    I Liner Systems: Current Design and Construction  Practices.
     In LandDisposal of Hazardous Waste, Proceedingsof the
     1leventh Annual Research Svmpo sium.  EPA/600/9-85/013.

Grossman, R. B., B. R. Brasher, D. P. Franzmeier,  and J. L.
     Walker.  1968.  Linear Extensibility as Calculated from
     Natural Clod Bulk Density Measurements.  In Soil Science


Haan,  C. T.  1977.  Statistical Methods in Hydrology.  Iowa State
     University Press, Ames, IA.  378 pp.

Hakonson, T. E.  1986.  Evaluation of Geologic Materials to Limit
     Biological Intrusion into Low-Level Radioactive Waste
     Disposal Sites.  Los Alamos National Laboratory Report No.
     LA-10286-MS.

Hale,  F. D., C. B. Murphey, Jr., and R. S. Parrat.  1983.  Spent
     Acid and Plating Haste Surface Impoundment Closure.  In
     Proceedings of the 4th National Conference on Management of
     UncontrolledWaste Sites.  Washington, DC.  pp. 195-201.

Hammer, D. P. and E. D. Blackburn.  1977.  Design and
     Construction of Retaining Dikes for Containment of Dredged
     Material.  Tech. Report D-77-9, U.S. Army Engineer Waterways
     Experiment Station, Vicksburg, MS.  181 pp.

Harbeck, G. E., M. A. Kohler, and G. E. Coberg.  1958.  Water
     Loss Investigations: Lake Mead Studies.  USGS Prof. Paper
     298.

Haxo,  H. E., H. A. Nelson, and J. A. Mi edema.  1985.  Solubility
     Parameters for Predicting Membrane-Waste Liquid
     Compatibility.  In Proceedings of the llthAnnual Research
     Sympoaiim on Land Disposal of.. Hazardous_Waste.  EPA/600/9—
     85-013.  Cincinnati, OH.


                                166

-------
Head, K. I.  1980.  Manual of Soil Laboratory Testing.  Vol. 1,
     Soil Classification Compaction Tests.  HaIsted Press, New
     York.

Head, K. H.  1982.  Manual of Soil Laboratory Testing.  Vol. 2,
     Permeability, Shear Strength, and Compression Tests.
     Halsted Press, New York.

Head, K. H.  1986.  Manual of Soil Laboratory Testing.  Vol. 3,
     Effective Stress Tests.  Halsted Press, New York.

Herbich, J. B.  1986.  Guidelines for Surface Impoundment
     Freeboard Control.  Consulting and Research Services, Inc.
     Report No. JBfl-867.

Hershfield, D. M.  1961.  Rainfall Frequency Atlas of the U.S.
     for Durations from 30 minutes to 24 hours and Return Periods
     from 1 to 100 Years.  Tech. Paper 40, U.S. Weather Bureau.
     61 pp.

Hicks, T. G.  1972.  Standard Handbook of Engineering
     Calculations.  McGraw-Hill, New York.

Hilf, J. W.  1975.  Compacted Fill.  In Foundation Engineering
     Handbook by Hinterkorn and Fang (Eds.).  Van Nostrand
     Reinhold Co., New York.  pp. 244-311.

Holtz, W. G.  1965.  Volume Change.  In Methods of Soil Analysis,
     Part I, ed. by Black.  American Society of Agronomy, Madison
     WI.

Holtz, R. D. and W. D. Kovacs.  1981.  An Introduction to
     Geotechnical Engineering.  Prentice-Hall, Inc., Englewood
     Cliffs, NJ.  733 pp.

Huang, Y. H.  1983.  Stability Analysis of Earth Slopes.  Van
     Nostrand Reinhold Co., New York.  305 pp.

Hunt, G. E.f V. E. Hodge, P. M. Wagner, and P. A. Spooner.  1983.
     Collection of Information on the Compatibility of Grouts
     with Hazardous Wastes.  In Proceedings of the Ninth Annual
     Research Symposium.  EPA/600/9—83-018.

Jenson, M.  E. (Ed.)  1973.  Consumptive Use Water and Irrigation
     Water  Requirements.  ASCE, New York.  215 pp.

Jeyapalan,  J. K.  1980.  Analyses of Flow Failures of Mine
     Tailings Impoundments.  Dissertation presented to the
     University of California, Berkeley, CA.
                               167

-------
Jeyapalan, J. K.  1983.  An Engineering Manual  for the Evaluation
     of Stability of Dikes (Unpublished).   USEPA Permit Writers
     Training Program.

Jeyapalan, J. K.  1986.  Personal Communication.   Programs
     available for Wisconsin Waste Management Training Center.
     University of Wisconsin, Madison,  WI.

Johnson, A. W. and J. R. Sallberg.  1960.   Factors that Influence
     Field Compaction of Soils.   Bulletin  272,  Highway Research
     Board, Washington, DC.  206 pp.

Johnson, S. J.  1975.  Analysis  and Design Relating to
     Embankments J A State-of-the-Art Review.  Soils and Pavement
     Laboratory, U.S. Army Engineer Waterways Experiment Station,
     Vicksburg, MS.

Johnson, V. R.  1981.  Leak Detection Systems for Landfills and
     Impoundments.  Third National Conference on Hazardous
     Materials Management, Anaheim, CA.

Karassik, I. J., W. C. Krutzsch, W. H.  Fraser,  and J.  P. Messina
     (Eds.)  1986.  Pump Handbook: 2nd Edition.  McGraw-Hill, New
     York.

Kays, W. B.  1987.  Construction of Linings for Reservoirs, Tanks
     and Pollution Control Facilities.   Second  Edition.  John
     Wiley & Sons, New York.  379 pp.

Keijman, J. Q. and R. W. R. Koopmans.  1973.  A Comparison of
     Several Methods of Estimating the Evaporation of  Lake Flevo.
     In Hydrology of Lakes.  Helsinki Symposium,  International
     Association of Hydrological Sciences.   Vol.  109.   pp. 225-
     232.

Keith, L. H., W. Crummett, J. Deegan, Jr.,  R. A.  Libby, J. K.
     Taylor and G. Wentler.  1983.  Principles  of Environmental
     Analysis.  In Analytical Chemistry, vol. 55.  pp. 2210-2218.

Kent, K. M.  1973.  A Method for Estimating Volume and Rate of
     Runoff in Small Watersheds.  Soil Conservation Service TP-
     49.  62 pp.

Kerfoot, H. B., J. A. Kohout, and E. N. Amick.   1986.   Detection
     and Measurement of Groundwater Contamination by Soil-Gas
     Analysis.  In Proceedings of the National  Conference on
     Hazardous Wastes and Hazardous Material.   Atlanta, GA.  pp.
     22-26.

Keulegan, G. H.  1951.  Wind Tides in Small Closed Channels.
     Jour, of Research of the National Bureau of Standards.  Vol.
     46.

                               168

-------
Kingsbury, R. P.  1986.  Personal Communication.   American
     Colloid Co., Skokie, IL.

Knipschild, F. W., R. Taprogge, and H. Schneider.  1979.  Quality
     in Production and Installation of Large Area Sealing
     Sections of High Density Polyethylene.  Shlegel Engineering,
     Chelmsford, Essex, United Kingdom.

Koerner, R. M. and J. P. Welsh.  1980.  Construction and
     Geotechnical Engineering Using Synthetic Fabrics.  John
     Wiley & Sons, New York.  267 pp.

Koerner, R. M., Editor.  1988.  Soft Soil Stabilization Using
     Geosynthetics.  Conference Proceedings.  Elsevier Applied
     Science Publishing Company.

Koerner, R. M.  1990.  Designing with Geosynthetics.  Second
     Edition.  Prentice-Hall, Englewood Cliffs, NJ.  1990.

Kohler, M. A.  1954.  Lake and Pan Evaporation: Water Loss
     Investigations: Lake Hefner Studies.  USGS Prof. Paper 269.
     pp. 127-148.

Kohler, M. A., T. J. Nordenson, and D. R. Baker.   1959.
     Evaporation Maps for the United States.  Tech. Paper No. 37,
     U.S. Weather Bureau, Washington, DC.

Kurtyka, J. C.  1953.  Precipitation Measurements Study. Report
     of Investigations No. 20, Illinois State Water Survey.  178
     pp.

Lambe, T. W.  1951.  Soil Testing.  John Wiley & Sons, New York.

Lambe, T. W.  1958.  Compacted Clay: Engineering Behavior.
     Award-winning ASCE papers in Geotechnical Engineering, 1950-
     1959.  Published 1977.  ASCE, New York.  pp. 668-706.  f(

Larson, L. W.  1971.  Precipitation and its Measurement: A State
     of the Art.  Water Resources Series No. 24, Water Resources
     Research Institute, University of Wyoming.  74 pp.

Larson, L. W. and E. L. Peck.  1974.  Accuracy of Precipitation
     Measurements for Hydrological Modeling.  Vol. 10, Water
     Resources Research,  pp. 857-863.

Lee, C. R., J. G. Skogerboe, K. Eskew, R. A. Price, and N. R.
     Page.  1985.  Restoration of Problem Soil Materials at Corps
     of Engineers Construction Sites.  DACW39-80-C0098.  USAGE,
     Dept. of Army.
                               169

-------
Lewis, R. W.  1982.  Custom Designing of Monitoring Wells for
     Specific Pollutants and Hydrogeologic Conditions.   Second
     Annual Symposium on Aquifer Restoration and Ground Water
     Monitoring.  National Water Well Association,  Worthington,
     OH.

Lindsay, W. L.  1979.  Chemical Equilibria in Soils.  John Wiley
     & Sons, New York.  449 pp.

Linsley, R. K., M. A. Kohler, and J. L. H. Paulhus.  1982.
     Hydrology for Engineers: 3rd Edition.  McGraw-Hill, New
     York.  508 pp.

Linsley, R. K. and J. B. Franzini.  1979.  Water Resources
     Engineering.  McGraw-Hill, New York.

Lippitt, J. M., T. G. Prothero, W. F. Martin, and L. P. Wallace.
     1984.  An Overview of Worker Protection Methods.  In
     Proceedings of Hazardous Material Spills Conference,  pp.
     354-361.

Lowe, J. and L. Karafiath.  1960.  Stability of Earth Dams upon
     Drawdown.  In Proceedings of the First Pan American
     Conference on Soil Mechanics and Foundation Engineering,
     Volume 2.  Mexico City.  pp. 537-552.

Machemehl, J. L. and J. B. Herbich.  1970.  Effects of Slope
     Roughness on Wave Run-up on Composite Slopes.   TAMU-SG-70-
     222.  Texas A&M University Sea Grant Program.

Mackay, D. and Leinonen.  1975.  Unsteady State Predictive Model
     for Non-aerated Surface Impoundments.

Mackay, D. and A. Yeun.  1983.  Mass Transfer Coefficient
     Correlations for Volatilization of Organic Solutes from
     Water.  In Environmental Science and Technology. Volume 17.
     pp. 211-217.

Malone, P. G., L. W. Jones, and J. P. Burkes.  1983.  Application
     of Solidification/Stabilization Technology to Electroplating
     Wastes.  In Proceedings of the 9th Annual Research Symposium
     for Land Disposal of Hazardous Wastes.  EPA-600/9-83-108.

Mather, J. R.   1978.  The Climatic Water Budget in Environmental
     Analysis.  D. C. Heath and Co., Lexington, MA.  239 pp.

Mausbach, J. J., B. R. Brasher, R. D. Yeck, and W. D. Nettleton.
     1980.  Variability of Measured Properties in Morphologically
     Matched Pedons.  Soil Science of America Journal, vol. 44.
     pp. 358-363.
                               170

-------
McCuen, R. H.  1987.  A Guide to Hydraulic Analysis Using SCS
     Methods.  Prentice-Hall Inc., Bnglewood Cliffs, NJ.

McGuinness, J. L.  1966.  A Comparison of Lysimeter Catch and
     Rain Gauge Catch.  ARS 41-124.  USDA Agricultural Research
     Service. 9 pp.

McGuinness, J. L. and G. H. Vaughan.  1969.  Seasonal Variations
     in Rain Gauge Catch.  In Water Resources Research. Vol. 5.
     pp. 1142-1146.

McRae, T.G.  1984.  Bagi: A New Concept for Detecting and
     Tracking Hazardous Gases.  In Proceedings on Hazardous
     Materials Spills Conference,  pp. 165-173.

Meade, J. P. and W. D. Ellis.  1985.  Decontamination Techniques
     for Mobile Response Equipment Used at Haste Sites.
     EPA/600/2-85/105.

Means, R. E. and J. V. Parcher.  1963.  Physical Properties of
     Soils.  Charles E. Merrill Books, Inc., Columbus, OH.  464
     pp.

Melvold, R. W., J. D. Byroade, and L. T. McCarthy, Jr.  1984.
     Emergency Response Procedures for Control of Hazardous
     Substances Releases.  In Proceedings of Hazardous Materials
     Spills Conference,  pp. 141-150.

Metcalf and Eddy, Inc.  1979.  Wastewater Engineering,
     Treatment/Residue/Reuse, 2nd Ed.  McGraw-Hill, New York.

Meyer, R. A., J. E. Brugger, and D. J. Lawrence.  1984.  Mapping
     Sunken Pollutant Pools with Depth Finders.  In Proceedings
     of Hazardous Materials Spills Conference,  pp. 214-219.

Miller, D. E.  1969.  Flow and Retention of Hater in Layered
     Soils.  Conservation Report No. 13, Agricultural Research
     Service, USDA.  684 pp.

Miller, J. P., R. H. Frederick, and R. J. Tracey.  1974.
     Precipitation Frequency Atlas for the Coterminous Western
     United States.  NOAA Atlas 2  (11 vols.}.

Minning, R. C.  1982.  Monitoring Hell Design and Installation.
     In Second Annual Symposium on Aquifer Restoration and Ground
     Water Monitoring. National Hater Well Association.
     Worthington, OH.

Mitchell, D. H., M. A. McLean, and T. E. Gates.  1988.  Stability
     of Lined Slopes at Landfills and Surface Impoundments.
     EPA/DOE Interagency Agreement DW 89930846-01-0, Battelle
     Pacific Northwest Laboratories, Richland, HA.

                               171

-------
Mitchell, J. K., D. R. Hooper, and R.  G.  Campanella.   1965.
     Permeability of Compacted Clay.   Jour,  of Soil Mechanics and
     Foundation Div., ASCE Proceedings,  vol. 91.   pp.  41-65.

Mundell, J. A. and B. Bailey.  1985.   The Design  and Testing of
     Compacted Clay Barrier Layer to Limit Percolation through
     Landfill Covers.  In Hydraulic Barriers in Soil and Rock.
     ASTM STP 874.  Philadelphia, PA.   pp. 246-262.

Nadeau, R. J., J. P. Lafornara, G. S.  Klinger, and T.  Stone.
     1985.  Measuring Soil Vapors for Defining Subsurface
     Contaminant Plumes.  In Proceedings of  theSixth National
     Conference on Management ofUncontrolled Hazardous Waste
     Sites.  Washington, DC.  pp. 128-129.

National Institute of Occupational Safety and Health.   1985.
     Occupational Safety and Health Guidance Manual for Hazardous
     Waste Site Activities.  NIOSH Pub.  No.  85-115.

Naval Facilities Engineering Command.   1982.  Design Manuals 7.1,
     7.2, and 7.3.  Dept. of the Navy, Alexandria, VA.

Neff, E. L.  1977.  How Much Does a Rain Gauge Gauge?  Jour, of
     Hydrology, Vol. 35.  pp. 213-220.

Neuwirth, F.  1973.  Experience with Evaporation Pans at a
     Shallow Steppe Lake in Austria.  In Hydrology of Lakes,
     Helsinki Symposium.  International Association of
     Hydrological Sciences.  Vol. 109.  pp.  209-297.

NOAA.  Climatography of the United States No. 81.  National
     Climatic Center, Asheville, NC.

NOAA.  1968.  Climatic Atlas of the United States.  National
     Climatic Data Center, Asheville, NC.  80 pp.

NOAA.  1983.  Selective Guide to Climatic Data Sources: Key to
     Meteorological Records Documentation No. 4.11.  National
     Climatic Data Center, Asheville, NC.  338 pp.

Nordenson, T. J.   1962.  Evaporation from the 17 Western States.
     USGS Prof. Paper No. 272-D.

Olson, G. W.  1976.  Criteria for Making and  Interpreting  a Soil
     Profile Description.   In Kansas Geological Survey Bulletin
     No. 212.   University of Kansas Press, Lawrence, KS.   47  pp.

Olson, R. Z. and  D. E. Daniel.   1981.  Measurement of the
     Hydraulic  Conductivity of Fine-Grained Soils.  In
     Pennei»t>ility and Groundwater Contaminant Transport.   ASTM—
     STP 746.   pp.  18-64.

                               172

-------
Osborne, H. B., L. J. Lane, and J. F. Hundley.   1972.   Optimum
     Gauging of Thunderstorm Rainfall in Northeastern  Arizona.
     In Kater Resources, vol. 8.  pp. 259-265.

Overcash, H. R. and D. Pal.  1979.  Design of Land Treatment
     System for Industrial Wastes - Theory and Practice.  Ann
     Arbor Science, Ann Arbor, MI.  684 pp.

Perloff, W. H. and W. Baron.  1976.  Soil Mechanics.   Ronald
     Press Co., New York.

Pettyjohn, W. A., W. J. Dunlap, R. L. Cosby, and J. W. Keeley.
     1981.  Sampling Ground Water for Organic Contaminants.  In
     Ground Water, vol. 19.  pp. 180-189.

Proctor, R. R.  1933a.  Fundamental Principles of Soil
     Compaction.  In Engineering News-Record.  August  30.  pp.
     348-351.

Proctor, R. R.  1933b.  Description of Field and Laboratory
     Methods.  In Engineering News-Record.  September  7.  pp.
     286-289.

Proctor, R. R.  1933c.  Field and Laboratory Verification of Soil
     Suitability.  In Engineering News-Record.  September 21.
     pp.   348-351.

Proctor, R. R.  1933d.  New Principles Applied to Actual Dam-
     Building.  In Engineering News-Record.  September 28.  pp.
     372-376.

Pruitt, W. 0. and R. L. Snyder.   1984   Crop Water Use.  In
     Irrigation with Reclaimed Municipal Wastewater.  P.eport 84-
     1, California State Water Resources Control Board,
     Sacramento, CA.

Reades, D. W. and C. D. Thompson.   1984.  Quality Control Testing
     and Monitoring of  Performance  cf Clay Till Liner.  CEO-CGS
     Seminar on Design  and Construction of Municipal and
     Industrial Waste Disposal Facilities.  Malton, Ontario.

Rishel, H. L., T. M. Boston,  and  C.  J. Schmidt.  1984.  Costs  of
     Remedial Response  Actions at Uncontrolled Hazardous Haste
     Sites.  Pollution  Technical  Review, no. 35.  Noyes
     Publications, Park Pidge, NJ.

Rodda,  J.  C.   1968.  The  Rainfall Measurement Problem.   In
     Geochemistry. Precipitation. Evaporation. Soil Moisture,
     Hydrometry.  General  Assembly  of Bern.  International Assoc.
     of Scientific Hydrology  Publication 78.  pp.  215-231.


                                173

-------
Rodensky,  Robin.  1985.   Closure and Post-Closure Care
     Regulations: Implications of Recent  Amendments.   In
     Proceedings of National Conference on Hazardous  Wastes and
     Environmental Emergencies,  pp. 29-32.

Sai, J. 0. and J. D. Zabcik.  1985.   Estimate of Surface
     Impoundment Construction Cost Under  RCRA Amendments of 1984
     (Draft report).  Prepared by K. If. Brown and Associates
     under USEPA Contract 68-03-1816.

Saint-Venant, B. D.  1871.  Theory of Unsteady Water  Flow with
     Application to River Floods and to Propagation of Tides in
     River Channels.  In Computes Rendus. Vol. 73. Academy of
     Sciences, Paris, France.

Saville, T.  1956.  Wave Run-up on Shore  Structures.   Jour, of
     Waterways and Harbors Division, ASCE, Vol. 82.  pp. 925-1
     through 925-14.

Saxton, K. E. and J. L.  McGuinness.   1982.  Evaporation.  In
     Hvdrclogic Modeling of Small Watersheds. Monograph No. 5,
     American Society of Agricultural Engineers, St.  Joseph, MI.
     533 pp.

Scalf, M.  R., J. F. McNabb, W. J. Dunlap, R. L. Cosby, and J.
     Fryberger.  1981.  Manual of Groundwater Sampling
     Procedures.  National Water Well Association/EPA Series.  93
     pp.

Schmertmann, J. H.  1970.  Static Cone to Compute Static
     Settlement over Sand.  Jour. Soil Mechanics and Foundations
     Divisions, vol. SM3.

Schmidt, C. E., W. D. Balfour, and R. D.  Cox.  1982.   Sampling
     Techniques for Emissions Measurements at Hazardous Waste
     Sites.  -In Proceedings of the 3rd National Conference on
     ManagementofUncontrolled Waste Sites.  Washington, DC.

Scott, C. R.  1980.  An Introduction to Soil Mechanics and
     Foundations.  Third Edition. Applied Science Publishers,
     Ltd., London England.  406 pp.

Seed, H. B.   1979.  Considerations in Seismic Design of Earth and
     Rockfill Dams.  19th Rankine Lecture.  London, England.

Shen, T. T.   1982.  Estimation of Organic Compound Emissions front
     Waste Lagoons.  Jour. Air Pollution Control Assoc,, vol. 32.
     pp. 79-82.

Sherard, J. L. et al.  1963.  Earth and Earth-Rock Dams.  John
     Wiley &  Sons, New York.


                               174

-------
Shiver, R., S. Johnson, and J. D. Zabcik.  1985.  Methods for
     Detecting Liquid Level and Maintaining Minimum Freeboard at
     Hazardous Waste Surface Impoundments (Draft report).
     Prepared by K. W. Brown and Associates under USEPA Contract
     68-03-1816.

Shultz, D. W., D. L. Laine, and J. W. Cooper.  1985.  Development
     of a Technique for Retrofitting Impoundments with
     Geomembranes.  In Land Disposal of Hazardous Waste;
     Proceedings of the Eleventh AnnualResearch Symposium.
     EPA/600/9-85/013.  Hazardous Waste Engineering Research
     Laboratory, Cincinnati, OH.  pp. 321-331.

Smith, G. N.  1982.  Elements of Soil Mechanics for Civil and
     Mining Engineers, 5th Edition.  Granada Publishing.  London,
     England.  493 pp.

Smith, R. L., D. T. Musser, and T. J. McGrood.  In.-sj.tu
     Solidification/Fixation of Industrial Wastes.  In
     Proceedings of 6th National Conference onManagementof
     Uncontrolled Hazardous Wastes.  Washington, DC.  pp.231-233.

Soil Survey Staff.  1975.  Soil Taxonomy.  U.S. Government
     Printing Office, Washington, DC*  754 pp.

Sowers, G. F.  1979.  Introductory Soil Mechanics and
     Foundations: Geotechnical Engineering.  Fourth Edition.
     McMillan Publishing Co., Inc., New York.  621 pp.

Sowers, G, F. and J. G. Gulliver.  1955.  Effects of Varying
     Tamping-Foot Width on Compaction of Cohesive Soil.  Highway
     Research Board, Proceedings of the 34th Annual Meeting,
     Washington, DC.  pp. 598-601.

Spangler, M. G. and R. L. Bandy.  1982.  Soil Engineering  (4th
     Edition).  Harper and Row, New York.  819 pp.

Spencer, W. F. and M. M. Cliath.  1977.  Transfer of Organic
     Pollutants Between the Solid-Air Interface.  In Fate_of
     Pollutants in the Air and Water Environments edited by H.
     Saffett.  John Wiley & Sons, New York.  pp. 107-109.

Spooner, P. A.  1985.  Stabilization/Solidification Alternatives
     for Remedial Action.  In Proceedings of the 6th National
     Conference on Management of Uncontrolled HazardousWastes.
     Washington, DC.  pp. 214-220.

Spooner, P. A., G. E. Hunt, V. E. Hodge, P. M. Wagner, and I. R.
     Melyk.   1984.  Compatibility of Grouts with Hazardous
     Wastes.  EPA-600/2-84-015.  Municipal Environmental Research
     Laboratory, Cincinnati, OH.


                                175

-------
Springer, C., P. D. Lunney, and K. T. Valsaraj.   1984.   Emission
     of Hazardous Chemicals from Surface and Near-Surface
     Impoundments to Air.  USEPA Cooperative Agreement  808161-02.

Stanfield, D. F. and K. S. McMillan.  1985.  Inspection of
     Hazardous Waste Sites Using Ground-Penetrating Radar (Ci'R) .
     In Proceedings of the National Conference on Hazardous
     Wastes and Environmental Emergencies,  pp.  244-249.

Stanhill, G.  1970.  The Heat and Water Balance of a Fishpond in
     the Upper Galilee of Israel.  Israel Jour,  of Agricultural
     Research, vol. 20.  pp. 21-39.

Stevens, D. K.  1986.  Optimization of Free Liquid Removal
     Alternatives in the Closure of Hazardous Waste Surface
     Impoundments.  In Proceedings of the National Conference on
     Hazardous Wastes and Hazardous Materials.  Atlanta, GA.  pp.
     148-153.

Stewart, W. S.  1978.  State-of-the-Art Study of Land Impoundment
     Techniques.  EPA/600/2-78-196.  USEPA, Cincinnati, OH.

Terzaghi, K. and R. B. Peck.  1967.  Soil Mechanics in
     Engineering Practice.  John Wiley & Sons, New York.

Thibodeaux, L. J.   1981.  Estimating Air Emissions from Hazardous
     Waste Landfills.  Journal of Hazardous Materials,  Vol. 4,
     pp. 235-244.

Thibodeaux, L. J.,  C. Springer, and R. S. Parker.  1985.
     Volatile Organic Emissions Reductions from Surface
     Impoundments by the Use of Wind Fences and Wind Barriers.
     In Land Disposal of Hazardous Waste. Proceedings of the
     Eleventh Annual Research Symposium.  EPA/600/9-85/013.

Thibodeaux, L. J.,  D. G. Parker, and H. H. Heck.   1982.
     Measurement of Volatile Chemical Emissions from Wastewater
     Basins.  EPA600/2-82-095.

Thibodeaux, L. J. and S. T. Hwang.   1982.  Land Farming of
     Petroleum Wastes — Modelling the Air Emissions Problem.
     Environmental  Progress, Vol.  1, pp.  42-46.

Topp, G. C. and J.  L. Davis.   1981.  Detecting Infiltration of
     Water Through  Soil Cracks by Time-Domain Reflectometry.   In
     Geoderma. vol. 26.  pp.13-23.

Toyoshima, O., N.  Shuto, and H.  Hashimoto.   1966.  Wave Run-up on
     Coastal  Structures.   In Coastal Engineering  in Japan, vol.
     9.   pp.  119-126.
                                176

-------
Tratnyek, J., P. Costas, and W. Lyman,  1985.  Test Methods for
     Determining the Chemical Waste Compatibility of Synthetic
     Liners.  EPA/600/2-85-029.  Hazardous Waste Engineering
     Research Laboratory, Cincinnati, OH.

linger, M., J. O. Sai, R. Shiver, J. Jeyapalan, G. B. Ivans, Jr.,
     and  D. C. Anderson.  1985.  Comparison of Multiple Small
     versus Single Large Surface Impoundments (Draft Report).
     Prepared by K. W. Brown and Associates under EPA Contract
     68-03-1816.

U.S. Army Corps of Engineers.  1957.  Effect of Lift Thickness
     and  Tire Pressure: Soil Compaction Investigation Report No.
     8.   Tech. Memorandum 3-271.  Waterways Experiment Station,
     Vicksburg, MS.

U.S. Army Corps of Ipgineers.  1970a.  Laboratory Soils Testings
     Engineering Manual 1110-2-1906.

U.S. 'Army Corps of Engineers.  1970b.  Engineering and Design
     Stability of Earth and Rock-Fill Dams.  US6PO.

U.S. Army Corps of Engineers.  1977.  Construction Control for
     Earth and Rock-fill Dams.  Engineer Manual EM 1110-2-1911.
     Dept. of the Army, Washington, DC.

U.S. Army Corps of Engineers.  1984.  Shore Protection Manual:
     Volumes 1 and 2  (4th Edition).  Waterways Experiment
     Station.  USGPO.

U.S. Bureau of Public Roads.   1967.  Use of Riprap for Bank
     Protection.  Hydraulic Engineering Circular no. 11.  Federal
     Highway Administration.

U.S. Bureau of Reclamation.  1974.  Design of Small Dams.  Dept.
     of the Interior, Washington,  DC.  816 pp.

U.S. Bureau of Reclamation.  1981.  Ground Water Manual.  U.S.
     Government Printing Office, Denver, CO.  480 pp.

U.S. Dept. of Agriculture.   1979.  Engineering Field Manual for
     Conservation Practices  (3rd Edition).  Soil Conservation
     Service.

USEPA.   1978.  Liners for Sanitary Landfills and Chemical and
     Hazardous Waste  Disposal  Facilities.  EPA/600/9-78-005.

USEPA.   1979a.  Methods  for  Chemical Analysis of Water and
     Wastes. EPA-600/4-79-020.  Office of Research & Development,
     Cincinnati, OH.
                                177

-------
US1SPA.  1979b.  Design and Construction of Covers for Solid Waste
     Landfills.  EPA-600/2-79-165.  Municipal Environmental
     Research Laboratory, Cincinnati, OH.   250 pp.

USEPA.  1980a.  Samplers and Sampling Procedures for Hazardous
     Waste Streams.  IPA-600/2-80-018.  NTIS PB 80-135353.  69
     pp.

USEPA.  1980b.  Procedures Manual for Groundwater Monitoring at
     Solid Waste Disposal Facilities.  3W-611.  Office of Water
     and Waste Management, Washington, DC.  269 pp.

USEPA.  1980c.  Interim Guidelines and Specifications for
     Preparing Quality Assurance Project Plans.  QAMS-005/8Q.
     Prepared by the Office of Monitoring Systems and Quality
     Assurance.

USEPA.  1982a.  Closure of Hazardous Waste Surface Impoundments.
     SW-873.  Office of Solid Waste and Emergency Response.  92
     pp.

USEPA.  1982b.  Test Methods for the Evaluation of Solid Waste:
     Physical/Chemical Methods.  SW-846.  Office of Solid Waste
     and Emergency Response.  USGPO No. 055-00281001-2.

USEPA.  1982c.  Hazardous Waste Management Systems,  Permitting
     Requirements for Disposal Facilities.  FederalRegister,
     Vol. 47, No. 43.  July 26, 1982.

USEPA.  1982d.  Draft RCRA Guidance for Surface Impoundments,
     Liner Systems, Final Cover, and Freeboard Control.  Office
     of Solid Waste and Emergency Response.

USEPA.  1982e.  Groundwater Monitoring Guidance for Owners and
     Operators of Interim Status Facilities.  SW-963.  Office of
     Solid Waste and Emergency Response, Washington, DC.  176 pp.

USEPA.  1992f.  Guide to the Disposal of Chemically Stabilized
     and Solidified Waste.  Municipal Environmental Research
     Laboratory, Cincinnati, OH.  114 pp.

USEPA.  1982g.  Evaluating Cover Systems for Solid and Hazardous
     Waste.  SW-867.  Office of Solid Waste and Emergency
     Response, Washington, DC.

USEPA.  1982h.  Design Manual - Dewatering Municipal Wastewater
     Sludges.  EPA-625/1-82-014.  Municipal Environmental
     Research Laboratory, Cincinnati.

USEPA.  1983a.  Surface  Impoundment Assessment National Report.
     EPA/570/9-84-002.
                               178

-------
USEPA.  1983b.  Lining of Haste Impoundments and Disposal
     Facilities.  Office of Solid Waste &  Emergency Response.
     SW-870.  448 pp.

USEPA.  1983c.  Landfill and Surface Impoundment Performance
     Evaluation.  SW-869.  Office of Solid Haste and Emergency
     Response.

USEPA.  1983d.  Handbook for Evaluating Remedial Action
     Technology Plans.  EPA/600/2-83-076.   Office of Research and
     Development, Washington, DC.  439 pp.

USEPA.  1983e.  Land Treatment of Hazardous Wastes.  SW-874 .
     Office of Solid Waste and Emergency Response.

USEPA.  1983f.  Preparation of Soil Sampling Protocol:
     Technologies and Strategies.  EPA-600/4-83-020.

USEPA.  1983g.  Standardized Procedures for Planting Vegetation
     on Completed Sanitary Landfills.  EPA-600/2-83-055.
     Municipal Environmental Research Laboratory, Cincinnati, OH.

USEPA.  1984a.  Assessment of Hazardous Haste Surface Impoundment
     Technology Case Studies and Perspectives of Experts.
     EPA/60C/2-84-173.  300pp.

USEt.i.  1984b.  Use of Water Balance to Quantify Seepage from
     Hazardous Waste Surface Impoundments.  Prepared by K.W.
     Brown and Associates under EPA Contract 68-03-2943,  WA 14.
USEPA.  1984c.  Preliminary Assessment of Hazardous
     Pretreatment as an Air Pollution Control Technique (Draft
     Report).  Prepared by Research Triangle Institute under EPA
     Contract 68-03-3149.

USEPA.  1984d.  Field Studies of Liner Installation Methods at
     Landfills and Surface Impoundments.  EPA/600/2-84-168.
     Municipal Environmental Research Laboratory, Cincinnati, OH.
     58 pp.

USEPA.  1984e.  Permit Applicant's Guidance Manual for Hazardous
     Waste Land Treatment, Storage, and Disposal Facilities.
     EPA/530-SW-84-OC4.  Office of Solid Waste and Emergency
     Response, Washington, DC.

USEPA.  1984f.  A Guide to the Selection of Materials for
     Monitoring Well Construction and Groundwater Sampling.

USEPA.  1984g.  Assessment of Innovative Techniques to Detect
     Waste Impoundment Liner Failures.  EPA-600/2-84-041.
     Municipal Environmental Research Laboratory Cincinnati, OH.
     139 pp.

                               179

-------
USEPA.  I984h.  Electrical Resistivity Technique to Assess the
     Integrity of Geomembrane Liners.  EPA-6QO/2-84-180.
     Municipal Environmental Research Laboratory, Cincinnati, OB.
     63 pp.

USEPA.  1984j.  Review of In-Place Treatment Technology for
     Contaminated Surface Soils, Vol. 1: Technical Evaluation.
     EPA-540/2-84-003a.  Office of Research and Development.

USEPA.  1984k.  Soil Sampling Quality Assurance User'* Guide.
     EPA-600/4-84-043.

USEPA.  1985a.  ECRA Groundwater Monitoring Technical Enforcement
     Guidance Document (Draft).  Office of Solid Waste and
     Emergency Response, Washington, DC.

USEPA.  I985b.  RCRA Preliminary Assessment/Site Investigation
     Guidance (Draft).  Office of Solid Waste, Washington, DC.

USEPA.  1985c.  Air Emissions for Quiescent Surface Impoundments
     Emissions Data and Model Review (Draft Technical Mote).
     Prepared by GCA Corp. under EPA Contract 68-01-68'.'1, WA 49.

USEPA.  1985d,  Field Evaluations of Hazardous Waste Pretreatment
     as an Air Pollution Control Technique (Draft Report).
     Prepared by Research Triangle Institute and Associated
     Technologies, Inc. under EPA Contract 68-02-3992.

USEPA.  1985e.  In-situ Methods for the Control of Emissions for
     Surface Impoundments and Landfills (Draft Final Report).
     Prepared by the University of Arkansas and Louisiana State
     University under EPA Cooperative Agreement CR810856.

USEPA.  1985f.  Evaluation of Air Emissions from Hazardous Waste
     Treatment, Storage, and Disposal Facilities.  EPA 600/2-85-
     057.  Municipal Environmental Research Laboratory,
     Cincinnati, OH.  NTIS PB 85-203792.

USEPA.  1985g.  Modeling Remedial Actions at Uncontrolled
     Hazardous Waste Sites.  EPA/540/2-85/001.  Office of Solid
     Waste and Emergency Response, Washington, DC.

USEPA.  1985h.  Petitions to Delist Hazardous Wastes: A Guidance
     Manual.  EPA/530-SW-85-003.  Office of Solid Waste and
     Emergency Response, Washington, DC.  54 pp.

USEPA.  1985i.  Covers for Uncontrolled Hazardous Waste Sites.
     EPA/540/2-85/002.  Hazardous Waste Engineering Research
     Laboratory, Cincinnati, OH.
                                180

-------
USEPA.  1985j.  Liner Materials Exposed to Hazardous and Toxic
     Waster.  EPA/600/2-84/169.  NTIS PB 85 121333.   Hazardous
     Haste Engineering Research Laboratory, Cincinnati,  OH.   256
     pp.

USEPA.  1983k.  Handbook - Remedial Action at Waste  Disposal
     Sites (Revised).  EPA-625/6-85-006.  Hazardous  Haste
     Engineering Research Laboratory, Cincinnati,  OK and Office
     of Solid Waste and Emergency Response, Washington,  DC.

USEPA.  1986a.  Technical Guidance Document.  Construction
     Quality Assurance for Hazardous Waste Land Disposal
     Facilities.  EPA/530-SW-85-021.  Office of Solid Haste  and
     Emergency Response, Washington, DC.  100 pp.

USEPA.  1986b.  Minimum Technology Guidance Document on Freeboard
     Control for Surface Impoundments (Draft).  Prepared by  K.H.
     Brown and Associates under EPA Contract 68-03-1816.

USEPA.  1986c.  Geotechnical Analysis for Review of  Dike
     Stability (CARDS) (Draft).  Prepared by Dept. of Civil
     Engineering, University of Cincinnati under EPA Contract 68-
     03-3183.

USEPA.  1986d.  Revised Draft Protocol for Groundwater
     Inspections at Hazardous Waste Treatment, Storage, and
     Disposal Facilities.  USEPA Hazardous Waste Ground-Water
     Task Force.

USEPA.  1986e.  Reference Manual of Countermeasures for Hazardous
     Substances Releases: Draft Report.  Prepared by Combustion
     Engineering/Rockwell International Corporation under EPA
     Contract S8-03-3014.

USEPA.  1986f.  Decontamination Techniques for Mobile Response
     Equipment Used at Waste Sites  (State-of-the-Art Survey).
     EPA/600/2-85/105.  Hazardous Waste Engineering Research
     Laboratory, Cincinnati, OH.

USEPA.  1986g.  Handbook for Stabilization/Solidification of
     Hazardous Wastes.  EPA/540/2-86/001.  Hazardous Waste
     Engineering Research Laboratory, Cincinnati, OH.

USEPA.  1987a.  Minimum Technology Guidance on Double Liner
     Systems  for Landfills and Surface  Impoundments - Design,
     Construction, and Operation.  EPA/530-SW-85-014.  Office of
     Solid Waste and  Emergency Response, Washington, DC.  71 pp.

USEPA.  1987b.  Design, Construction, and  Maintenance of Cover
     Systems  for Hazardous Waste: An Engineering Guidance
     Document.  EPA-600/2-87-039.   Hazardous  Waste Engineering
     Research Laboratory, Cincinnati, OH.

                                181

-------
USEPA.  1987c.  Geosynthetic Design Guidance for Hazardous Waste
     Landfills and Surface Impoundments.   EPA-600/2-87-097.
     Hazardous Haste Engineering Research Laboratory,  Cincinnati,
     OB.

USEPA.  198?d.  Assessment of Techniques  for In-»itu Repair of
     Flexible Membrane Liners.  EPA-600/2-R7-038.  Hazardous
     Haste Engineering Research Laboratory,  Cincinnati,  OH.

USEPA.  1987e.  Hazardous Haste Treatment, Storage and Disposal
     Facilities (TSDF) — Air mission Models.   EPA-450/3-87-026.
     Office of Air Quality Planning and Standards, Research
     Triangle Park, NC.

USEPA.  1987f.  Handbooki  Ground Hater.   EPA/625/6-87/016.
     Robert S. Kerr Environmental Laboratory, Ada, OK.

US El'A.  1987g.  Background Document on Bottom Liner Performance
     in Double-Lined Landfills and Surface Impoundments.
     EPA/530-SW-87-013.  Office of Solid  Haste, Washington, DC.

USEPA.  1987h.  Hazardous Haste Management System:  Minimum
     Technology Requirements.  Federal Register, Vol.  52, No. 74,
     April 17, 1987.  pp. 12566-12575.

USEPA.  1987i.  Liners and Leak Detection for Hazardous Haste
     Land Disposal Units.  Federal Register, Vol. 52,  No. 103,
     May 29, 1587.  pp. 20218-20310.

USEPA.  1988a.  Lining of Waste Containment and Other Impoundment
     Facilities.  EPA/600/2-88/052.  Hazardous  Haste Engineering
     Research Laboratory, Cincinnati, OH.

USEPA.  1988b.  Design, Construction, and Evaluation of Clay
     Liners for Haste Management Facilities.  EPA/530/SW-86/OQ7f.

USEPA.  1988c.  Guidance for Conducting Remedial Investigation
     and Feasibility Studies under CERCLA (Interim Final).  EPA
     540/G-89-004.  OSWER Directive 9355.3-01.   Office of
     Emergency and Remedial Response, Washington, DC.

USEPA.  1988d.  Guide to Technical Resources for the design of
     Land Disposal Facilities.  EPA/625/6-88/018.  Risk Reduction
     Engineering Laboratory, Cincinnati,  OH.

USEPA.  1989a.  Technical Guidance Document: Final Covers on
     Hazardous Haste Landfills and Surface Impoundments.
     EPA/530-SW-89-047.  Office of Solid Haste and Emergency
     Response, Washington, DC.  39 pp.
                               182

-------
USEPA.  1989b.  Seminar Publications  Requirements for Hazardous
     Waste Landfill Design,  Construction,  and Closure.
     EPA/625/4-89/022.  Center for Environmental Research
     Information, Cincinnati,  OB.   127 pp.

USEPA.  1989c.  The Fabrication of Polyethylene Flexible Membrane
     Liner Field Seams.  EPA/530/SW-89/069.   Office of Solid
     Haste and Emergency Response, Washington, DC.  42 pp.

USEPA.  1989d.  Hazardous Waste TSDF — Fugitive Particulate
     Matter Air Emissions Guidance Document.   EPA/450/3-89-019.
     Office of Air Quality Planning and Standards, Research
     Triangle Park, NC.

USEPA.  19393.  Stabilization/Solidification  of CERCLA and RCJtA
     Wastesi  Physical Tests,  Chemical Testing Procedures,
     Technology Screening, and Field Activities.  EPA/625/6-
     89/022.  Center for Environmental Research Information,
     Cincinnait, OH.  70 pp.

U.S. Geological Survey.  1985.  Quality of Water.  Branch
     Technical Memorandum no.  85-09.   Reston, VA.  26 pp.

Unterberg, W., R. W. Mel void,  L. M. Flaherty, and L. T. McCarthy.
     1984.  Procedures for Selection of Countenneasures for
     Hazardous Substance Releases.  In PrgcegdinqB of the
     HagardouB Materials Spills,, Conference.   pp. 151-161.

Urban Drainage and Flood Control District*  1979.  Urban Storm
     Drainage Criteria Manual, vols.  1 and 2.  In Engineering
     Field.Manual for Conservation Practices,3rd Edition.  U.S.
     Dept. of Agriculture, Soil Conservation  Service.

Valsaraj, K. T., C. Springer,  T. Nguyen, and  L. J. Thibodeaux.
     1985.  Investigation of Floating Liquids to Control Volatile
     Organic Chemical  (VOC) Emissions from Surface Impoundments.
     In Land  Disposal ofHazardous Waste.  Proceedings of the
     Eleventh Annual Research Symposium.  IPA/600/9-85/013,

Van Dorn, W.  G.  1953.  Wind Stress on an Artificial Pond.  Jour.
     of Marine Research, vol.  12.

Waddell, J. J.   1985.  Construction Materials Ready-Reference
     Manual.  McGraw-Hill, Hew York.  395 pp.

Wagner, K. and Z. Kosin.  1985.   In-situ Treatment.  In
     Proceedings of the6th National Conference on Management of
     Uncontrolled Hazardous Waste Sites.  Washington, DC.  pp.
     221-230.
                               183

-------
Warnick, C. C.  1956.  Influence of Hind on Precipitation
     Measurements afr-High Altitudes.  University of Idaho
     Engineering Experiment Station Bulletin No. 10.  63 pp.

Whippie, W.,  N. S. Grigg, T. Grizzard, C. W. Randall, R, P.
     Shubinski, and L. S. Tucker.  1983.  Stormwater Management
     in Urbanizing Areas.  Frentice-Hall, Inc., Englewood Cliffs,
     NJ.  234 pp.

White, "L. A., D. F. Doyle, and W. M. Kaschak.  1985.  Remedial
     Action Planning at Hazardous Waste Sites.  In Proceedj.nfs..of
     the Sixtlj National Conference on Management of Uncontrolled
     Hazardpus Was t e Si t e s.  Washington, DC.  pp. 281-284.

Winter, T. C.  1981.  Uncertainties in Estimating the Hater
     Balances of Lakes.  Jn Hater Resources Bulletin, vol. 17.
     pp. 82-115.

Winterkorn, H. F. and H. Fang.  1975.  Foundation Engineering
     Handbook.  Van Nostrand Reinhold, New York.

Wright, S. G.  1969.  A Study of Slope Stability and the
     Undrained Shear Strength of Clay Shales.  PhD dissertation,
     University of California, Berkeley.
                                184
                                              U.S.

-------
Reproduced by NTIS
National Technical Information Service
U.S. Department of Commerce
Springfield,  VA 22161
This report was printed specifically for your
order from our collection of more than 2 million
technical reports.
For economy and efficiency, NTIS does not maintain stock of its vast
collection of technical reports.  Rather, most documents are printed for
each order.  Your copy is the best possible reproduction available from
our master archive. If you have any questions concerning this document
or any order you placed with NTIS, please call our Customer Services
Department at (703)487-4660.

Always think of NTIS when you want:
« Access to the technical, scientific, and engineering results generated
by the ongoing  multibillion dollar R&D program of the U.S. Government.
• R&D results from Japan, West Germany, Great Britain, and some 20
other countries, most of it reported in English.

NTiS also operates two centers that can provide you with valuable
information:
• The Federal Computer Products Center - offers software and
datafiles produced by Federal agencies.
* The Center for the Utilization of Federal Technology - gives you
access to the best of Federal technologies and laboratory resources.

For more information about NTIS, send for our FREE NTIS Products
and Services Catalog which describes how you can access this U.S.  and
foreign Government technology.  Call (703)487-4650 or send this
sheet to NTIS, U.S. Department of Commerce, Springfield, VA 22161.
Ask for catalog, PR-827.
Name
Address.
 Telephone
              - Your Source to U.S. and Foreign Government
                 Research and Technology.

-------
*PBi1204354*
BIN:
IIWQICE:
SHIPTO;
PAYMENT;
M292
423381
1*2BBB10
CSH*CPQRNG
04-17-97

-------
          United States
          Environmental Protection
          Agency
           Office of Research and
           Development
           Washington DC 20460
EPA/600/R-95/051
April 1995
&EPA
RCRA Subtitle D (258)
Seismic Design
Guidance for Municipal
Solid Waste Landfill
Facilities

-------
                                EPA/600/R-95/051
                                April 1995
            RCRA SUBTITLE D (258)
          SEISMIC DESIGN GUIDANCE
                     FOR
MUNICIPAL SOLID WASTE LANDFILL FACILITIES
                      by

              Gregory N. Richardson
           G.N. Richardson & Associates
           Raleigh, North Carolina 27603

                      and

              Edward Kavazanjian, Jr.
                      and
                 Neven Matasovi
              GeoSyntec Consultants
         Huntington Beach, California 92647
             Contract No. 68-C3-0315
              PROJECT MANAGER

                 Robert Landreth
         Waste Minimization, Destruction and
             Disposal Research Division
        Risk Reduction Engineering Laboratory
              Cincinnati, Ohio 45268
  RISK REDUCTION ENGINEERING LABORATORY
    OFFICE OF RESEARCH AND DEVELOPMENT
   U.S. ENVIRONMENTAL PROTECTION AGENCY
            CINCINNATI, OHIO 45268
                                                Printed on Recycled Paper

-------
                                   DISCLAIMER
        The information in this document has been funded wholly or in part by the United States
Environmental Protection Agency (EPA) under Contract No. 68-C3-0315 WA #05 to Hardine
Uwson Associates.  It has been subjected to the Agency's peer and administrative review, and
it has been approved for publication as an EPA document.  Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
                                       u

-------
                                     FOREWORD
        Today's rapidly developing technologies, industrial products, and practices frequently
carry with them generation of materials that, if improperly dealt with, may threaten both human
health  and the environment.  The United States Environmental Protection Agency  (EPA) is
charged by Congress with protecting the Nation's land, air, and water resources.  Under a
mandate of national environmental laws, the Agency strives to formulate and implement actions
leading to a compatible balance between human activities and the ability of natural resources to
support and nurture life.  These laws direct  the EPA to conduct research to  define  our
environmental problems, measure the impacts, and search for the solutions.

        The Risk Reduction Engineering Laboratory is responsible for planning, implementing,
and managing research, development, and demonstration programs.  These programs provide an
authoritative, defensible engineering basis in support of the policies, programs, and regulations
of the  EPA with respect to drinking water,  wastewater, pesticides, toxic substances,  solid  and
hazardous wastes, and Superftmd-related  activities.  This publication presents information on
current research efforts and provides a vital communication link between the researcher and the
user community.

        Recent RCRA Subtitle D regulations (40 CFR Part 258) establish the requirements  that
MSW landfills must not be sited where they  can be damaged by active ground faulting (258.13)
and  that they must be designed to resist the effect of regional earthquakes (258.14). This
document is intended to provide technical guidance to regulatory reviewers and landfill designers
to ensure these objectives  are accomplished.   It is meant to be a  practical design document
applicable to the vast majority of MSW landfills.

        Further  information  relative to  this document  may   be obtained  by  writing
Robert Landreth, Risk Reduction Engineering Laboratory, Cincinnati, OH,  45268.

                                     E. Timothy Oppelt, Director
                                     Risk Reduction Engineering Laboratory
                                           m

-------
                                     ABSTRACT
        On October 9, 1993, the new RCRA Subtitle D regulations (40 CFR Part 258) went into
effect.  These regulations are applicable to landfills receiving municipal solid waste (MSW) and
establish minimum Federal criteria for the siting, design, operation, and closure of MSW landfills.
These  regulations apply to the entire waste containment system, including liners, leachate
collection systems, and surface water control systems. This document presents field and design
procedures to  satisfy  the  earthquake (or seismic) related criteria contained  within these
regulations.  Sample analyses are provided to evaluate the Subtitle D seismic requirements for a
range of site and facility conditions.

        Section 258.13 of  the regulations requires that new or lateral expansions of existing
landfills cannot be sited within 200-feet of a fault that has been active during the Holocene Epoch
(past 11,000 years) unless it can be demonstrated that a lesser setback is safe. This  document
presents field identification methods used to identify active faults.  Additionally, the  document
reviews general tectonic and seismological considerations that strongly suggest that movement of
faults during the Holocene Epoch is very rare east of the Rocky Mountains.

        Section 258.14 of the regulations identifies seismic impact zones within the United States
based on earthquake probability maps prepared by the United States Geological Survey (USGS).
Seismic impact zones are defined in the new regulations  as those regions having a peak bedrock
acceleration exceeding 0.1 g based on a 90% probability of non-exceedance over a 250 year time
period.  Within seismic impact zones, the regulations require that the waste containment system
for new MSW landfills and for lateral expansions of existing MSW landfills be designed to resist
the maximum horizontal acceleration in Minified earth material (MHA).  The MHA is defined as
the maximum expected horizontal acceleration either depicted on a seismic  hazard map with a 90
percent probability of non-exceedence in 250 years or  based upon a site-specific seismic risk
assessment.

        This document presents analysis procedures to evaluate the ability of the site subgrade
to resist liquefaction and of the waste mass/subgrade  to resist slope failure where subjected to the
MHA.   Sample calculations  are provided to demonstrate the analysis techniques for liquefaction
and slope stability. Additional discussion is provided regarding more sophisticated deformation
analysis methods that may be required for MSW landfills in highly seismic regions.

        This report was submitted in ralfillment of Contract No. 68-C3-0315 WA #05 under the
sponsorship of the United States Environmental Protection Agency. This report covers a period
from November, 1993  to May, 1994, and work was completed as of May, 1994.
                                           IV

-------
                              TABLE OF CONTENTS
Number                                                                   Page No.

Disclaimer		ii
Foreword	 iii
Abstract	 iv
List of Tables	vii
List of Figures  	  viii
Abbreviations and Acronyms	xii
Acknowledgements   	  xiii

1.0   Introduction  	1

      1.1     Introduction to Subtitle D Seismic Criteria	1
             1.1.1     Part 258.13 Fault Zone Siting Criteria  	1
             1.1.2     Part 258.14 Seismic Impact Zones	 . 2
      1.2     Scope of this Document	2
      1.3     Limitations of this Document  	4
      1.4     References	4

2.0   258.13 Fault Area Considerations	7

      2.1     Regional Fault Characteristics	7
      2.2     Site Fault Characterization	9
      2.3     Defining Fault Movement in Holocene Epoch	 11
      2.4     Comments on Fault Considerations East of the Rockies  	12
      2.5     References	12

3.0   258.14 Seismic Impact Zones: U.S.G.S. Probabilistic
      Bedrock Acceleration	25

      3.1     Development of Design Earthquake	26
      3.2     Interpretation of Peak Bedrock Accelerations	27
      3.3     References	28

4.0   258.14 Seismic Impact Zones: Site Specific Seismic Design
      Ground Motion		 40

      4.1      General Methodology	42
              4.1.1    Simplified Analysis	43
              4.1.2    One-Dimensional Site Response Analysis	47

-------
                        TABLE OF CONTENTS (continued)
Number                                                                   Page No.

             4.1.3     Two- and Three-Dimensional Site Response Analysis	51
      4.2     Selection of Earthquake Time History	52
      4,3     References	55

5.0   258.14 Seismic Impact Zones:  Liquefaction Analysis	. . 74

      5.1     Initial Screening	74
      5.2     Liquefaction Potential Assessment	76
      5.3     Liquefaction Impact Assessment 	80
      5.4     Liquefaction Mitigation	81
      5.5     References  	,	82

6.0   258.14 Seismic Impact Zones:  Slope Stability and Deformation Analysis  ....  102

      6.1     Key Material Properties	  103
             6.1.1     Unit Weight	  103
             6.1.2     Interface Shear Resistance 	  104
             6.1.3     Low Permeability Soil	 .  104
             6.1.4     Granular Soil Shear Strength	  105
             6.1.5     MSW Shear Strength	  105
             6.1.6     Sensitivity Studies  	  106
      6.2     Seismic Stability and Deformation Analysis	  106
      6,3     Additional Considerations  .	  109
      6.4     References	  110

Appendices

A - Seismic Design Examples - Liquefaction		  125
B - Seismic Design Examples - Slope Stability	  132
                                        VI

-------
                                 LIST OF TABLES


Number                                                                    Page No.

2.1    Significant Earthquakes in Eastern North America (Adams and Busham, 1994)  ..15

2.2    Sources of Information  .	16

2.3    Addresses of State Geological Survey Offices Source (Geotimes, 1980)	17

2.4    Detailed Seismic Event Data Available from USGS National Earthquake Information
       Center	18

3.1    Parameters for Seismic Source Zones (USGS, 1982)	31

4.1    Parameters for the Empirical Relationship to Estimate Gmax (after Imai and
        Tonouchi, 1982)	59

5.1    Estimated Susceptibility of Sedimentary Deposits to Liquefaction During Strong
       Seismic Shaking (Youd and Perkins, 1978)  	87

5.2  '  Recommended "Standardized" SPT Equipment (after Seed, et al., 1985 and Riggs,
        1986)	88

5.3    Correction Factors for Nonstandard SPT Procedure and Equipment	89

5.4    Improvement Techniques for Liquefiable Soil Foundation Conditions  (NRC,
        1985)	90

6.1    Unit Weight Data for MSW (Fassett et al., 199.4)	  114

6.2    Compilation of the Available Shear Strength Data on MSW (GeoSyntec, 1993)  .  116

6.3    Lower Bound Friction Angles Backfigured from Observations of Steep Landfill
        Slopes 0 (Kavazanjian et al., 1995)	  117
                                         vn

-------
                                 LIST OF FIGURES
Number                                                                    Page No.

1.1     Seismic Impact Zones (Areas with a 10% or Greater Probability that the Maximum
        Horizontal Acceleration will Exceed 0.10 g in 250 Years) (EPA, 1993) ,.,.,.,, 6

2.1     The Six Major Tectonic Plates and their Approximate Linear Velocity Vectors
        (Adapted from Park,  1983)  	19

2.2     Seismic Source Areas in the United States (Krinitsky, et al., 1993)	20

2.3     Isoseismal Contours for Intra-Plate vs. Edge-Plate Events of Similar Magnitude
        (Nuttli, 1974)	21

2.4     Epicenters for Earthquakes M 2.5 in the Southeastern United States (July 1977 -
        December 1984) (Sibol et al., 1984)	22

2.5     Characteristics of Hay ward Fault as Exposed in Five Trenches at Fremont Site
        (CMf et al., 1972)	23

2.6     Detail of West Fault Trace Exposed in Trench  "G" at Fremont Site (see Fig, 2.5)
        (Cluff et al., 1972)	24

3.1     Basic Elements of the USGS Probabilistic Hazard Calculations: (a)  Typical Source
        Areas and Grid of Points at which the Hazard is to be Computed; (b)  Statistical
        Analysis of Seismicity Data and Typical Attenuation Curves; (c) Cumulative
        Conditional Probability Distribution of Acceleration; (d) The Extreme Probability,
        Froax,t  (a) for Various Accelerations and Exposure Times (T) (USGS, 1982) .... 35

3.2     Seismic Source Zones in the Contiguous United States (USGS, 1982)   	36

3.3     Seismic Source Zones in the Central United States (Johnson and Nava, 1994)  ... 37

3.4     Time-Dependent Fluctuations in Seismic Ground Response  Parameters (17 January
        1994 Northridge, California Earthquake, Oil Site, Longitudinal Component)
        (Hushmand Associates, 1994)          	38

3.5     Contribution of Various Magnitudes and Distances to the Seismic Hazard
        (Moriwaki et al., 1994)	39

-------
                           LIST OF FIGURES (continued)
Number                                                                   Page No.

4.1    Soil Conditions and Characteristics of Recorded Ground Motions, San Francisco
       M 5.7 Earthquake of 22 March 1957 (Seed, 1968)	60

4.2    Development of Acceleration Response Spectrum for Damped Single Degree of
       Freedom System	61

4.3    Tripartite Representation of Response Spectra	62

4.4    Relationship Between Maximum Acceleration on Rock and Other Local Site
       Conditions: (a) Seed and Idriss (1982); (b) Idriss (1990)	63

4.5    Observed Variations of Peak Horizontal Accelerations on Soft Soil and MSW
       Sites in Comparison to Rock Sites (Kavazanjian and Matasovi,  1995)	64

4.6    Approximate Relationship Between Maximum Ground Accelerations at the Base and
       Crest for Various Ground Conditions (Singh and Sun, 1995)	65

4.7    Variation of Maximum Average Acceleration Ratio with Depth of Sliding Mass
       (Kavazanjian and Matasovi, 1995)  	66

4.8    Normalized Maximum Horizontal Equivalent Acceleration versus the Normalized
       Fundamental Period of the Waste Fill (Bray et al., 1995)	  67

4.9    Modulus Reduction and Damping Curves for Soils of Different Plasticity Index (PI)
       (Vucetic and Dobry, 1991)	68

4.10   Shear Wave Velocity of MSW (Kavazanjian et al., 1995)  	69

4.11   Modulus Reduction and Damping Curves for MSW (Earth Technology, 1988) ...  70

4.12   Modulus Reduction and Damping Curves for MSW (Singh and Murphy, 1990) . .  71

4,13   Modulus Reduction and Damping Curves for MSW (Kavazanjian and Matasovi,
       1995)	72

4.14   Comparison of Oil Landfill Response to Results of Equivalent Linear Analysis
       (Kavazanjian et al., 1995) .	73
                                         IX

-------
                            LIST OF FIGURES (continued)


Number                                                                     Page No.

5.1     Grain Size Distribution Curves of Liquefied Soils (Ishihara et al., 1989)	92

5.2     Variation of q^^ Ratio with Mean Grain Size, D50 (Seed and De Alba,  1986)  .  . 93

5.3     Stress Reduction Factor,  rd (Seed and Idriss, 1982)  	94

5.4     Correction Factor for the Effective Overburden Pressure, CN (Seed et al., 1983)   . 95

5.5     Relationships Between Stress Ratio Causing Liquefaction and (N^ Values for Sand
        for M 7.5 Earthquakes (Seed et al., 1985)	96

5.6     Curve for Estimation of Magnitude Correction Factor, kM (after Seed et al., 1983) 97

5.7     Curves for Estimation of Correction Factor k (Harder 1988,  and Hynes 1988, as
        Quoted in Marcuson et al., 1990)	   98

5.8     Curves for Estimation of Correction Factor k (Harder 1988,  and Hynes 1988, as
        Quoted in Marcuson et al., 1990)	   99

5.9     Curves for Estimation of Post-Liquefaction Volumetric Strain using SPT  Data and
        Cyclic Stress Ratio (Tokimatsu and Seed, 1987)  	  100

5.10    Relationship Between Corrected "Clean Sand" Blowcount (N^ and Undrained
        Residual Strength (Sr) from Case Studies (Seed et al., 1988)  	  101

6.1     Fundamental Principles of the Newmark Seismic Deformation Analysis (after Bray et
        1994)	  118

6.2     Unit Weight Profile for MSW (Kavazanjian et al., 1995)	  119

6.3     Bi-Linear Shear Strength Envelope for MSW (Kavazanjian et al., 1995)	  120

6.4     Yield Acceleration as a Function of Shear Strength Parameters for the Oil Landfill
        (Siegel et al.,  1990)   	121

6.5     Hynes and Franklin Permanent Seismic Displacement Chart (Hynes and Franklin,
        1984)	122

-------
                         LIST OF FIGURES (continued)






Number                                                               Page No.



6.6    Makdisi and Seed Permanent Displacement Chart (Makdisi and Seed, 1978) ... 123>




6.7    Modes of Instability of a MSW Landfill   	124
                                       XI

-------
                       ABBREVIATIONS AND ACRONYMS
ASTM        American Society for Testing and Materials
ATC          Applied Technology Council
CPT          Cone Penetration Test
CSR          Critical Stress Ratio
EERC         Earthquake Engineering Research Center
EERI         Earthquake Engineering Research Institute
EPA          United States Environmental Protection Agency
FSAR         Final Safety Analysis Report
MFZ         Mendocino Fracture Zone
MHA         Maximum Horizontal Acceleration
MM          Modified Mercali (Intensity Scale)
MSW         Municipal Solid Waste
MSWLF       Municipal Solid Waste Landfill Facility
NAPP         National Aerial Photographic Program
NCEER       National Center for Earthquake Engineering Research
NRC          National Research Council
Oil           Operating Industries, Inc. (Landfill)
PSAR         Preliminary Safety Analysis Report
RCRA        Resource Conservation and Recovery Act
SDOF         Single Degree of Freedom (System)
SPT          Standard Penetration Test
SSA          Seismological Society of America
USGS         United States Geological Survey
                                       xn

-------
                              ACKNOWLEDGEMENTS
The authors wish to express their sincere appreciation to the following individuals who reviewed
and critiqued drafts  of this manuscript:   Dr. Rudolph Bonaparte, GeoSyntec  Consultants,
Professor Jonathan Bray, University of California of Berkeley, Professor Ron Chancy, Californis
State University at Humbolt, Professor Robert Koerner, Drexel University,  Professor Gerald
Leonards, Purdue University, Mr, Robert Phaneuf, New York State Department of Environmental
Conservation, and Mr. Robert Landreth, United States Environmental Protection Agency,  The
authors also thank the New York Department of Environmental Conservation for providing case
studies used in the sample calculations.

The authors also gratefully acknowledge the many individuals, too numerous to name here, who
over the years have shared their experiences and recommendations regarding seismic probability
studies, liquefaction analysis, and dynamic stability evaluation.
                                          xm

-------
                                      SECTION 1

                                   INTRODUCTION

On October 9, 1993, the RCRA Subtitle D regulations (40 CFR Part 258) went into effect. These
regulations are applicable to landfills receiving municipal solid waste (MSW) and establish
minimum Federal criteria for the site location, design, operation, ground-water monitoring, and
closure/post closure care of MSW landfills. This document focuses on the earthquake (or seismic)
siting and facility design criteria contained within Subtitle D.  The document is intended for use
by both designers of MSW landfills and the regulatory community that reviews such designs.
Where possible, actual landfill situations have been used in the development of example problems
to demonstrate the various analysis procedures.  Emphasis is placed herein on simple analysis
methods that are within the technical capabilities of the general engineering community.  The
range of applicability and the limitations of these methods are reviewed and more rigorous
analysis methods are briefly summarized.

1.1    Introduction to Subtitle D Seismic Criteria

Subtitle D regulations address the potential for damage to a MSW landfill resulting from relative
ground displacements (e.g., fault displacement) and from strong ground motions (e.g., ground
accelerations) that can accompany an earthquake. Limiting the potential for fault displacement-
induced damage is accomplished by siting criteria (258.13)  that may preclude the use of a given
site for a MSW landfill. The impact of earthquake-induced strong ground motions on a MSW
landfill must be addressed by the design engineer. Subtitle D does not specify the required
evaluation process but establishes (258.14) a lower value for the maximum horizontal acceleration
(MHA) in lithified earth material (e.g. the peak bedrock acceleration) that must be considered in
the design of landfill containment structures.  The MHA may be based upon either a probabilistic
map such as those published by the United States Geological Survey (USGS) or upon the results
of a site-specific analysis.  Landfill containment structures are defined to include liners, leachate
collection systems, and surfaces water control systems.

1.1.1  Part 258.13 Fault Zone Siting Criteria

The Federal Subtitle D regulations state that a new MSW landfill or a lateral expansion of an
existing landfill may not be located within 200 feet (60 meters) of a fault that has experienced
displacement in the Holocene time unless the owner or operator demonstrates to the Director of
an approved State Program that an alternative setback distance of less that 200 feet (60 meters)
will prevent damage to the structural integrity of the landfill unit and will be protective of human
health and the environment.  Within the regulations, a fault  means a fracture or zone of fractures

                                            1

-------
along which strata from one side have been displaced with respect to strata on the other side.  The
Holocene time means the most recent epoch of the Quaternary period, e.g. within the last 10,000
to 12,000 years. This requirement means that MSW landfill site suitability studies must both
identify  potential fault  zones that impact the proposed  site and then evaluate whether fault
displacement has occurred during the past 10,000 to 12,000 years. Section 2.0 of the document
presents  the technical methodology for identifying  fault zones and for  complying with  the
regulatory criteria.

1.1.2  Part 258.14 Seismic Impact Zones

A seismic impact zone is defined in the Subtitle D regulations as an area having a 10% or greater
probability  that the peak horizontal acceleration in lithified earth material, expressed as  a
percentage of the earth's gravitational pull (g), will exceed 0. 10 g in 250 years.  These zones may
be defined using seismic probability maps prepared by the  USGS (USGS, 1982 and USGS, 1991)
or by more detailed regional or site specific studies.  The USGS maps present
accelerations and velocities reflecting a 90% probability that the acceleration will not be exceeded
over 10, 50 and 250 year interval periods.  Seismic impact zones in the United States, defined by
application of the  Subtitle  D criteria to  the  USGS  seismic probability maps, are shown in
Figure 1.1.

Section 3.0 of this document provides general background information on the development of the
USGS seismic probability maps, and a simple method for interpretation and use of the peak
bedrock acceleration from these maps.  Section 4.0 provides methodologies for calculating the
peak ground surface acceleration at a landfill site and the peak surface acceleration and peak
average acceleration of the waste mass based on site characteristics and peak bedrock acceleration.
These peak ground accelerations are then used in Section 5,0 for evaluating the liquefaction
potential of a site and in Section 6.0 for evaluating the stability of a landfill foundation, waste
mass, and waste slopes.   Sections 5.0 and 6.0 present simplified seismic  analysis procedures  that
can typically  be performed without the need for supplemental field  investigative programs,
expensive specialized laboratory testing, or sophisticated dynamic analyses to evaluate compliance
of the design of MSW landfills with Subtitle D regulation  for seismic design.

1.2   Scope of This Document

Damage to landfills  from  earthquake  may be due  to the  primary seismic hazard  of fault
displacement or to  secondary hazards such as slope instability or liquefaction of the foundation
induced by strong ground motions.  Potential modes of damage MSW landfills associated with the
primary seismic hazard include:

-------
       •       disruption of liner and cover systems;
       •       disruption of the landfill gas control system; and
       •       disruption of surface water and drainage control systems.

Secondary modes of damage to the containment systems of MSW landfills that are subject to
strong ground motions include:

       •       damage due to liquefaction and lateral spreading of the foundation;
       •       damage due to seismically-induced settlement of the foundation; and
       •       damage due to seismically-induced landslides.

In general, MSW landfills have performed extremely well in earthquakes. Observations of the
performance of solid waste landfills subject to strong ground motions (Anderson and Kavazanjian,
1995;  Matasovic, et al., 1995) indicate that minor cracking of cover soils at the waste/natural
ground interface and disruption of landfill gas control  systems due to loss of power and breaking
of vertical wells and headers are the most common types of damage experienced by MSW landfills
subject to strong ground shaking.  Neither of these effects is considered to present a significant
environmental hazard. However, experience with the performance of modern landfills conforming
to Subtitle D requirements is limited.  Of the three landfills designed in accordance with Subtitle
D standards subject to the strongest shaking in the Northridge earthquake of 17 January 1994, one
experienced two tears in the liner, one of which was approximately 75 ft (23 m) in length, along
an anchor trench above the waste.  Furthermore, no landfill with a geosynthetic cover is known
to have been subjected to strong shaking in an earthquake and no solid waste landfill is known to
have experienced fault displacement or liquefaction in the foundation during an earthquake (even
though there are solid waste landfills known to  be sited on active faults  and liquefiable soils).
Therefore, caution is warranted in concluding  unconditionally that landfills will continue to
perform well  in earthquakes and investigations  and analyses are required  to demonstrate that
landfills are properly sited to avoid active faults and are properly designed to resist the effects of
strong ground motions and liquefaction.

This document presents a set of simplified analyses for seismic performance analysis of the waste
mass,  liner and cover systems, and foundation of a MSW landfill within a seismic impact zone.
The analyses presented herein include analyses of the  impact of instability of the waste mass and
cover soil on the integrity of geosynthetic  liner and cover systems of a landfill subject to strong
ground motions. Analyses of the potential for liquefaction-induced lateral spreading of the waste
mass and seismically-induced settlement of the foundation soil  are also presented herein.  The
simplified analyses presented  herein provide a minimum standard for design of MSWLF in
accordance with Subtitle D standards.  These analyses are not intended for analysis of the seismic
performance of landfills containing hazardous or toxic substances or large amounts of liquids.

-------
Neither should they be applied to landfills that do not conform to Subtitle D siting restrictions.
Such landfills must be considered on a case-by-case basis and may require a higher standard of
care dependency on the potential consequences of seismically-induced damage.

1.3    Limitations of this Document

The simplified analyses described in this document are presented as an example of one way in
which such a seismic  performance  assessment may be conducted.  The simplified analyses
presented herein are designed to produce an expedient assessment of the seismic resistance of the
landfill containment systems.  If such simplified analyses indicate potential seismic problems
(e.g., results in unacceptable factors of safety), then more sophisticated analysis methods may be
required to demonstrate satisfactory performance of the facility.

This document addresses the seismic design of the landfill waste  mass, liner and cover systems,
and foundations, only.  With the exception of the liquefaction analyses, this  document does not
provide guidance on assessing the impact of geologically unstable  terrain on landfill performance.
A demonstration that the MSW landfill unit will not be disrupted by geologic instability, including
seismically unstable areas,  is required under Section 258.15 of Subtitle D. Guidance for satisfying
the provisions of Section 258.15 of Subtitle D is provided by EPA (1993) and is not included in
this document.  Additional seismic analyses may be required to assess the performance of other
components of the landfill containment systems,  including the  leachate  collection system and
surface-water control systems.

Seismic analysis and design of landfills is a rapidly developing field.  Even as this document was
being completed, new and important studies on the seismic  behavior of landfills were appearing
in print and/or being presented at conferences and other professional meetings. It is essential that
the designer and regulator involved in seismic design of MSW landfills keep abreast of current
developments  in the field.  As new information and techniques  become available, they will
supersede the information and methods presented herein.

1.4   References

Anderson, D.G.,  and  Kavazanjian,  E. Jr. (1995) "Performance of Landfills Under Seismic
Loading,"  Proc., Third International Conference on Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, University of Missouri, Rolla,  Vol. 3, 2-7 April.

EPA (1993),  "Technical Manual:   Solid Waste Disposal Facility Criteria," United States
Environmental Protection Agency, EPA 530-R93-017, Washington, District of Columbia.

-------
Matasovic, N., Kavazanjian, E., Jr., Augello, A.J., Bray, J.D., and Seed, R.B. (1995), "Solid
Waste Landfill Damage Caused by 17 January 1994 Northridge Earthquake," In: Woods, Mary C.
and Seiple, Ray W., Eds., The Northridge, California, Earthquake of 17 January 1994, California
Department of  Conservation, Division  of  Mines  and  Geology  Special Publication  116,
Sacramento, California, pp. 43-51.

USGS  (1982), "Probabilistic Estimates of Maximum Acceleration and Velocity in Rock in the
Continuous United States," United States Geological Survey, Open-File Report 82-1033.

USGS  (1990), "Probabilistic Earthquake Acceleration and Velocity Maps for the United States
and Puerto Rico," United States Geological Survey, Miscellaneous Field Studies Map MF-2120.

-------
Figure 1.1
Seismic Impact Zones  (Areas With a  10% or Greater Probability that
Maumnm Horizontal Acceleration Will Exceed 0.10 g in 250

                             6

-------
                                      SECTION 2

                      258.13 FAULT AREA CONSIDERATIONS

Locating a landfill in the vicinity of faults that have experienced relative movement in recent times
poses significant risk to the integrity of the landfill containment system. The impact to the landfill
from a seismic event can result directly from ground surface rupture or from deformation,
liquefaction,  lateral spreading, and differential  settlement induced by ground  shaking  that
accompanying the event.  The fault area location restrictions imposed by Subtitle D restrict siting
of new MSW units or lateral expansions of existing unite within 200 feet (60 meters) of a fault that
has displaced in Holocene time.

This section of the guidance document reviews methods for evaluating both the presence of faults
on-site and the possible movement of a fault  within the Holocene Epoch.  The section concludes
with a discussion regarding the difficulty in  applying such methodology to  faults located east of
the Rocky Mountains.

2.1    Regional Fault Characteristics

Faults are created when the stresses within geologic materials exceed the ability of those materials
to withstand the stresses.  An understanding  of such stresses is aided by a review of current plate
tectonics theory. Figure 2.1  shows the major tectonic plates that form the earth's continents and
their directions of movement. Along the west coast, earthquakes are the result of several different
fault systems that occur along the edge of the Pacific and North American plates. South of the
Mendocino Fracture  Zone  (MFZ)  approximately 200  miles (320 kilometers) north of San
Francisco, the San Andreas  fault system (strike-slip) controls earthquakes.  North of  the MFZ,
earthquakes are controlled by the Cascadia Subduction Zone. In between these two major fault
systems lie the ridges, rifts, and subduction zones associated with the Juan de Fuca and Gorda
plates.  The complex interactions between these zones create stresses in the crust away from the
plates that generate earthquakes. Earthquakes may occur along faults in the crust in Washington,
Oregon, and  California adjacent to the plate boundaries or in the interior of the continent away
from the plate boundaries.   In the interior of the North American plate, tectonic stresses have
created fault systems that are known to have generated major earthquakes in Utah, South Carolina,
New England, Oklahoma,  and Missouri/Tennessee in Holocene time.  The sense of the fault
displacement within these fault systems range from horizontal (strike-slip) to vertical (dip-slip) to
combinations of these components. These major fault systems and one suggested representation
of the major seismic  source areas in the United States are shown in Figure 2.2.  The Roman
numerals on this figure represent the maximum  observed (historic) seismic intensity in  the region
as measured  by the Modified Mercali (MM) intensity scale (Richter,  1958).

-------
In contrast to the west coast, earthquakes east of the Rocky Mountains cannot be associated with
the relative displacements of edge-plate faults (active margin).   Intra-plate (passive margin)
earthquakes occur less frequently than the edge-plate associated earthquakes of the west coast but
impact a significantly larger geographic area. Table 2.1 (Adams and Busham, 1994) presents one
summary  of significant earthquakes in Eastern North America  in historic time,  ordered by
decreasing magnitude.  As this table shows, the pattern of significant earthquakes east of the
Rocky Mountains is broadly dispersed both geographically and temporally.

The differences in the sizes of affected areas may be caused by the differences in stress conditions
in the basement rock structure.  In the west, the stress condition is predominantly tension, while
in the east, stresses  in the basement rock are primarily compressional.  Whatever the mechanism,
the rate of attenuation of earthquake ground motions east of the Rocky Mountains appears to be
significantly slower than in the western United States (Nuttli, 1974; 1981), resulting in a much
larger impacted area in the eastern U.S.  than the  western U.S. for  earthquakes of the same
magnitude.  For equivalent historical earthquakes, Figure 2.3 shows the isoseismal contours of
MM VI and YE for an event at the plate boundary in the western United States and for two intra-
plate events in the eastern United States. Note the small geographic area impacted by the western
edge-plate event as compared to the two intra-plate event; one that shook a large portion of the
central United States centered around  New Madrid and one that shook far beyond Charleston,
even into Canada.  Another observed difference between earthquakes in the eastern and western
U.S. is that eastern earthquakes appear to be enriched in high frequency components compared
to western earthquakes (Atkinson, 1987).

The significance of the differences between western edge-plate earthquakes and the intra-plate
events that occur east of the Rockies with respect to identification of surface faulting  is discussed
subsequently within this section. The significance of the differences between western and eastern
earthquakes with respect to frequency content is discussed in subsequent sections.

Characterization of the seismicity of the eastern and central U.S. is a topic of much current study
and discussion (Applied Technology Council (ATC), 1994).  Due to the many ongoing studies,
our understanding of the seismicity of the central and eastern United States is evolving rapidly.
Prudent investigators  should consult current sources of information on local and regional
seismicity at the  initiation  of any project.   Sources of current information are  discussed
subsequently in this section.

-------
2.2    Site Fault Characterization

The principal factors controlling the general characteristics of surface faulting are: (a) the type of
fault (reverse,  normal, or strike-slip), (b) the  inclination of the fault plane, (c) the amount of
displacement on the fault, (d) the depth and geometry of the surficial earth materials, and (e) the
nature of the overlying earth materials.  Strike-slip faults that are not fairly linear may produce
complex surface features. Step-over zones where fault displacement is transferred from adjacent
strike-slip faults may  be particularly complex. Dip-slip faults, with either normal or reverse
motion, typically produce multiple fractures within rather wide and irregular fault zones. These
zones generally are confined to the hanging-wall side of the fault leaving the footwall  side little
disturbed.  With respect to fault impacts on a structure, setback requirements for such faults may
be rather narrow on the footwall side, depending on the quality of data available, and  larger on
the hanging wall side  of the zone.  Some fault zones may contain broad deformational features
such as pressure ridges and sags rather than clearly defined fault scarps or shear zones (Hart,
1990).

An investigation to identify faulting at a given site must rely on a review of available data and
field  geologic  reconnaissance methods.   Available  data  may  include pertinent  technical
publications, unpublished reports, maps, aerial photographs, and interviews with experts familiar
with the region under study.  Pertinent technical publications include maps prepared by the USGS
identifying young  faults in the western states, publications of the Seismological Society of
America, and regional reports from the seismological networks and state geological surveys. A
detailed summary  of available sources of engineering geologic information  is presented by
Trautmann and Kulhawy  (1983).   General  sources  for  such information are indicated on
Table 2.2.  Table 2.3 provides a listing of addresses of the geological survey offices for all
50 states.

Studies performed for siting of nuclear power plants can be  a useful source of information on
regional seismicity and geology. All applications for construction permits for nuclear generating
stations are required to submit documentation on regional geology, including known  faults and
observed seismicity, within a 200 mile (320 kilometer) radius of the site.  This information can
be found in the Preliminary Safety Analysis Report (PSAR) and the Final Safety Analysis Report
(FSAR) for the project.  These reports are available through the National Technical Information
Service (see Table 2.2) for all existing and many proposed nuclear generating stations.  However,
as  may of these reports are over 20 years old, more recent sources of information on regional
seismicity and tectonics should be consulted.

Existing seismic networks provide very detailed identification of recent earthquakes within seismic
impact regions.  Such information includes the  magnitude  and  epicentral location of all identified

-------
events and is commonly available plotted in map form as shown on Figure 2.4.   A detailed
evaluation of each detected event is also available as shown on Table 2.4. Note that while the
presence of micro-seismic activity can be used to infer the location of a subsurface fault, it cannot
be directly interpreted as evidence that surface displacement of the fault has taken or will take
place.  To date, the only known earthquake east of the Rocky Mountains in historic time that has
been accompanied by observations of surface fault rupture is the 1989 Ungava, Quebec earthquake
of magnitude 6.3.   The Meers fault  in Oklahoma, where evidence  points to a magnitude 7+
earthquake within the past 1,100 to 1,400 years, and the Reelfoot fault in Tennessee, the source
of the 1811/1812 New Madrid earthquake sequence, are  the only other generally recognized
Holocene faults east of the Rocky  Mountains.

An interpretation of available stereo aerial photographs is useful in identifying and locating
potentially active  faults. One source of such photographs is provided in Table 2.3.  Other sources
are discussed by Trautmann and  Kulhawy (1983).   Active faults may be indicated in aerial
photographs by geomorphic features such as fault scarps, triangular facets, fault scarplets, fault
rifts, fault slice ridges, shutter ridges, and fault saddles (Cluff, et al. , 1972). Additional evidence
can be provided by ground features such a open fissures, offsets in fence lines, landscape features,
mole tracks and furrows, etc., rejuvenated streams, folding or warping of young deposits,  ramps,
ground water barriers in recent alluvium,  echelon faults in alluvium, and fault paths on young
surfaces. Usually a combination of such features is  generated by recent fault movements at the
surface. Note that many of the fault movement  indicators require the presence of undisturbed
surface soils at the site. Regions that have limited surface soils due to past geologic mechanisms
or man's activities can provide a significant challenge in demonstrating the recent activity of
existing faults. The aerial photo analysis should include an area within a five-mile radius of the
site.

Initial field reconnaissance should be performed at a minimum for the area within approximately
1-kilometer (3300-feet) of the proposed unit (EPA, 1993).  This initial field reconnaissance can
include the following;

        •      walking portions of the site within 1-kilometer (3300 feet) of the unit to identify
              possible geomorphic or ground features that indicate faulting;

        «      preparation and interpretation of special aerial photographs  such  as low sun angle
              photographs that use  shadows  to accentuate topographic differences, infrared
              photos that indicate differences in surface moisture content,  and color photographs
              to study slight color changes.

Section 2.3 discusses the field methods for establishing fault movement.

                                            10

-------
2.3    Defining Fault Movement in Holocene Epoch

If the site fault characterization study indicates the potential presence of faults on the proposed
landfill site, then a detailed geologic reconnaissance may be required for the site of the proposed
unit.   A  detailed geologic  surface  reconnaissance should  be  made  to  identify the best
approximation of the fault location on site and the amount  and sense of past fault movements.
The detailed field site characterization can include the following:

       *      using geophysical methods such as resistivity,  seismic  refraction,  or magnetic
              methods to identify specific fault locations;

       •      excavation of exploratory trenches at  an angle to faults  identified on the site to
              allow the detailed examination of the trench walls  for evidence of recent fault
              displacements; and

       «      subsurface drilling exploration to locate fault zones.

The depth of the  subgrade investigated should be sufficient to represent activities  within the
Holocene Epoch.   Radiocarbon dating of carbonaceous material  encountered can be used to
constrain the age of most recent fault offsets.  A detailed description of soil-stratigraphic dating
techniques is presented by Shlemon (1985).  Sieh et.  al (1984) describe the application of high-
precision radiocarbon analyses for chronological analysis of active faulting. Note that establishing
that recent displacement has occurred is greatly  complicated if a limited soil profile over rock
exists at the site, e.g., glacially polished regions, or if the Holocene zone of the alluvium is absent
or disturbed.

Trenching across a fault through overlying alluvium  and colluvium has been the most common
tool used to establish both the existence of fault displacement  and  for dating the displacement.
Trench geologic sections established by trenching for portions of the Hay ward fault  in California
are shown on Figure 2.5. These trench sections established that the western trace of the Hay ward
fault was active (e.g., the fault displacements projected up through the overlying alluvium) and
that the  eastern trace was not active.  These observations are  shown on Figure 2.6.  Note the
distinct stratigraphic displacements that identify the west fault trace.  Thus, this  study would lead
to the requirement that all proposed MSW landfills be located more than 60 meters (200 feet) from
the western trace of the Hay ward fault.  Since the eastern trace of the Hayward fault was found
not to be active, there are no regulatory constraints  that would exclude  siting a MSW landfill
adjacent to the eastern trace of the Hayward  fault on  the basis of faulting.
                                            11

-------
2.4    Comments on Fault Considerations East of the Rockies

In recent years, seismologists have expressed significant concern regarding  our lack of
understanding of the source of earthquakes, referred to as seismogenesis, in the eastern United
States. Plate tectonic theories are not adequate to explain the mechanisms associated with intra-
plate earthquakes.  Recent workshops and seminars on the seismogenesis and seismicity of the
eastern United States (SSA,  1988; ATC, 1994) have shown that some widely accepted views on
earthquake origins are inconsistent with recent observations and that a global perspective may be
required to understand intra-plate seismogenesis.  Obviously such concerns are beyond the scope
of this document.   It is, however,  important to realize the following observations regarding
earthquake/fault considerations east of the Rocky Mountains:

       •       Earthquake source zones do appear to be related to subsurface crustal structure.
              However, these source zones do not appear to be related to surface expressions of
              faulting (ATC, 1994).

       «       The relationship between  intra-plate earthquakes and the potential for surface
              faulting remains in question. This is in part  due to the lack of either accumulated
              strain or recorded significant seismic events in the eastern and central United
              States.

       •       Detailed comparison of earthquake hypocenters and known surface fault locations
              have failed to  indicate a correlation (Himes et al., 1988).

       »       Only two faults that  have  experienced ground  surface displacement during the
              Holocene Epoch have been identified east of the Rocky Mountains.

Current  understanding of seismogenesis east of the Rocky Mountains strongly  suggests that
significant efforts to define seismically active faults in this region may not be useful. The region
of capable faults identified on Figure 2.2 reaches to from the West Coast the Meers fault in
Oklahoma but clearly excludes most of the Midwest and all of the Eastern United States.

2.5    References

Adams,  J., and Basham,  P.W. (1994),  "New Knowledge of  Northeastern North  American
Earthquake Potential,"  Proc. Seminar on New Developments in Earthquake Ground Motion
Estimation and Implications for Engineering Design Practice,  Applied Technology Council,
ATC35-1, Redwood City, California, pp. 3-1 - 3-20.
                                           12

-------
ATC (1994), "Seminar on New Developments in Earthquake Ground Motion Estimation and
Implications for Engineering Design Practice," Applied Technology Council, ATC35-1, Redwood
City, California.

Atkinson, G.M.  (1987), "Implications of Eastern Ground Motion Characteristics for Seismic
Hazard Assessment in Eastern North America," Proc. Symposium on Seismic Hazards, Ground
Motions, Soil-Liquefaction and Engineering Practice in Eastern North America,  Tuxedo, New
York, NCEER Technical Report No. NCEER-87-0025.

Cluff, L.S, Hansen, W.R., Taylor, C.L., Weaver, K.D., Brogan, G.E., Idriss, I.M., McClure,
F.E., and Blayney, J.A. (1972),  "Site Evaluation in Seismically  Active Regions - An
Interdisciplinary  Team Approach," Proc.  International Conference on Microzonat ion for Safety
Construction, Research and Application, Seattle, Washington, Vol. 2, p. 9-57 - 9-87.

Engdahl, E.R., and Rinehard, W.A. (1988), "Seismicity Map of North America," Geologic
Society of America, Centennial Special Map CSM-4, Scale 1:5,000,000.

EPA  (1993),  "Technical  Manual:  Solid Waste Disposal Facility Criteria"  United States
Environmental Protection Agency, EPA 530-R93-017, Washington, District of Columbia.

Geotimes (1980), "A Directory of Societies in Earth Science," Vol. 25, No. 7, pp. 21-29.

Hart, E.W. (1990), "Fault-Rupture Zones in California," Special Publication  42, California
Division of Mines and Geology, Sacramento, California.

Himes,  L., Stauder, W., and Herrmann,  R.B. (1988), "Indication of Active Faults in the New
Madrid  Seismic  Zone  from  Precise Location  of  Hypocenters,"  National  Workshop  on
Seismogenesis in the Eastern  United States, A.C. Johnston, et al, Eds.  (also presented  in
Seismological Research Letters, Vol. 59,  No. 4, Seismological Society of America).

Johnston, A.C., and Nava,  S.J.  (1994), "Seismic Hazard Assessment in the Central United
States," Proc. Seminar on New Developments in Earthquake Ground Motion Estimation and
Implications for Engineering Design Practice, Applied Technology Council, ATC35-1, Redwood
City, California, pp. 2-1 - 2-12.

Krinitzsky, E.L., Gould, J.P. and Edinger, P.H. (1993), "Fundamentals of Earthquake-Resistant
Construction," John Wiley & Sons, New York, New York.
                                          13

-------
Nuttli, O.W. (1974), "Seismic Hazard of the Rocky Mountains," Preprint 2/95, American Society
of Civil Engineers, National Structural Engineering Meeting, Cincinnati, Ohio.

Nuttli, O.W. (1981), "Similarities and Differences Between Western and Eastern United States
Earthquakes, and their Consequences for Earthquake Engineering," Earthquakes and Earthquake
Engineering:  The Eastern United States, Vol. 1, Assessing the Hazard - Evaluating the Risk, I.E.
Beavers, Ed., Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan, pp. 25-51.

Park, R.G. (1983), Foundations of Structural Geology, Blackie Publishing, Chapman and Hall,
New York, New York.

Richter,  C.F. (1958), Elementary Seismology, W.H. Freeman and Company, San Francisco,
California.

Shlemon, R.J. (1985), "Application of Soil-Stratigraphic Techniques to Engineering Geology,"
Bulletin of the Association of Engineering Geologists, Vol. XXII, No. 2, pp. 129-142.

Sibol, M.S., Bellinger,  G.A., and Mathena, E.G. (1984),  "Seismicity of the Southern United
States," Southeastern U.S.  Seismic Network Bulletin No, 15, Seismologic Observatory, Virginia
Polytechnic Institute and  State University, Department  of Geological Sciences,  Blacksburg,
Virginia.

Sieh, K., Stuiver, M., and Brillinger, D.  (1989), "A More  Precise Chronology of Earthquakes
Produced by the San Andreas Fault in Southern California," Journal of Geophysical Research,
Vol. 94, No. Bl, pp. 603-623.

SSA (1988), "National Workshop on Seismogenesis in the Eastern  United States," A.C. Johnston,
A.C. et  al., Eds. (presented in Seismological Research Letters, Vol. 59, No. 4, Seismological
Society of America).

Trautmann, C.H., and Kulhawy, F.H. (1983), "Data Sources for Engineering Geologic Studies,"
Bulletin  of the Association of Engineering Geologists, Vol. XX, No. 4, pp.  439-454.
                                           14

-------
                     ERRATA SHEET


        RCRA Subtitle D(258) Seismic Design Guidance for
             Municipal Solid Waste Landfill Facilities


                      EPA/600/R-95/051
SECTION 4 258.14 SEISMIC IMPACT ZONES:
          SITE SPECIFIC SEISMIC DESIGN GROUND MOTION
             V (2n - 1)
                       "  =0.2,3.-)                        (4.1)
                   = P ' V?                               (4.2)
                Gmm[ = cW                               (4.3)
               N = 0.833(AT60)                              (4.4)

-------
SECTION 5        258.14 SEISMIC IMPACT ZONES:

                   LIQUEFACTION ANALYSIS



                 rd = 1  - 0.015 D                                      (5.1)



                      ^  max'(2)depth =D                                      , - —,
             _
             ~
               __
           rf ~
               *•  o'@depth=D   *•  max °> ©surfa
             CSREQ = 0. 65 (amai/g) rd (o0/o0 ')                              (5.3)



                   ^60 = N ' C60                                       (5-4)



                    CN = (I/O,')*                                        (5.5)


       where o0' equals the vertical . . .
                                                                     •

                    W« = N«>-CN   (D*3m)                         (5.6a)


                    (N,)M  = ' Nao • (€„), (D < 3 m)                      (5.6b)
             CSRL = CS/Z ' *w ' *0 ' *a                                 (5.7)
                                                                       (5.8)

 Estimate the liquefaction-induced lateral displacement, AL .

                A  = 0.75 (ff)m  (S)V3                                   (5.9)

-------
SECTION 6       258.14 SEISMIC IMPACT ZONES:

                 SLOPE STABILITY AND DEFORMATION ANALYSIS
     c/(y-z- COS2P) + tan(J)[l -y (z~rf )
                                          '                   (6.1)
                         ks
        c/(y z- cos2p) + tan(|)fl -yw(z-w
        	L	i-                  (o.2)
                    1   +  tan • P tan (J)

-------
TABLE 2.1: SIGNIFICANT EARTHQUAKES IN
       EASTERN NORTH AMERICA
   SOURCE:  ADAMS AND BUSHAM (1994)
Earthquake
New Madrid Region
New Madrid Region
New Madrid Region
Baffin Bay
Grand Banks
Charlevoix, Que
Charleston, SC
Nahanni, N.W.T.
Charlevoix, Que
Ungava, Que
Charleston, MO
Timiskaming, Que
Charlevoix, Que
Cape Ann, offshore
Charlevoix, Que
New Madrid Region
Charlevoix, Que
Franklin L., N.W.T.
Saguenay, Quebec
Giles County, VA
Massena/Cornwall
Miramichi, N.B.
Attica, NY
Year
1812
1811
1812
1933
1929
1663
1886
1985
1870
1989
1895
1935
1925
1755
1791
1843
1860
1992
1988
1897
1944
1982
1929
M
8.7
8.6
8.4
7.3
7.2
7.0
6.9
6.9
6.5
6.3
6.2
6.2
6.2
6.1
6.0
6.0
6.0
6.0
5.9
5.8
5,8
5.7
5.5
Comments
largest stable craton eq.


largest Arctic earthquake
27 dead from tsunami
earliest large earthquake
devastating
prior M 6.6 event

10 km surface rupture

Quebec/Ontario border

might be larger




shaking equivalent to M 6.!

Ontario/NY border
shallow, three M 5 aftershock
western NY
                     15

-------
                   TABLE 2.2: SOURCES OF INFORMATION
Aerial Photographs
Young Fault Maps
National Seismicity
 National Technical
 Information Service
National Aerial Photographic Program (NAPP)
National High Altitude Program (NHAP)
USGS EROS Data Center
Sioux Falls, South Dakota
(605) 594-6151

United States Geological Survey (USGS)
USGS National Center
H800) USA-MAPS or
USGS Map Sales Center
(303) 236-7477

National Earthquake Information Center
United States Geological Survey (USGS)
P.O. Box 25046, Denver Federal Center, MS 967
Denver, CO 80225
(800) 525-7848

Earthquake Engineering Research Institute (EERI)
499 14th Street, Suite 320
Oakland, California 94612
(510) 451-0905

National Center for Earthquake Engineering Research (NCEER)
State University of New York at Buffalo
Red Jacket Quadrangle
Buffalo, New York 14260
(716) 645-3391

Seismological Society of America (SSA)
National and Eastern Section
El Cerrito, California
(415) 525-5474

5285 Port Royal Road
Springfield, VA  22161
(703) 487-4650
FTS 737-4650
                                           16

-------
            TABLE 2-3:  ADDRESSES OF STATE GEOLOGICAL
                                      SURVEY OFFICES
                              SOURCE:  GEOTIMES  (1980)
   •facial]
Drawer 0. University. Ala.. 3S486
3001 Porcupine Onvc. Anchorage, Alaska,
Arteona Bureau of Geology  * Mbwtal
T.ch.nlogT
UrMversity of Aruona. Tucson. Ariz.. 85721

Arkansas Geological Commission
Vardette  Parham Geology  Center. 381S
West  Roosevelt Road. Little Rock.  Mi..
72204

Ca6l«mia Ormioa of Mine* * Geology
*m Ninth  St.. Room 1341. Sacramento.
Calrf.. 95814

Colorado Geological Survey '
1313  Sherman St., Room  715. Denver,
Coto.. 80203

Cormtcticut Natural Resources Center
Room S5S. SUM Office Building. 165 Capi-
M< Awt.. Hartford, Conn.. 06I1S

Odiware Ceologkal Survey
University  of  Delaware. Newark.  Od..
tsm

flofMa BurcM o< Geology
903 West Tennessee St.. Tallahassee. *Uu.
12304

Georgia Department of Natural Kesovrcef
Earth A Water Division. 19 Hunter Si. SW.
Room -403, Atlanta, Cl., 30334

Hawaii Oniiioo of Water & Land Oevdop-
tnefil
 Bo* 373. Honolulu. Hawaii. 96609

 tdiho lureau of Mines & Geology
 MOKOW. Idaho. £3843

 llfinon Ctologkil Survey
 121 Natural Resources Building. Urbana.
 III.. 61801

 Indiana Geological Survey
 611  North Walnut Grove. Btoonwngion.
 Ind.. 47«f1

 Iowa Geological Survey
 123 North Capitol $«.. Iowa  City. Iowa.
 S2242

 Kansas Geological Survey
 1930 Avenue A. Campu* West. University
 o< Kansas. Lawrence. Kan.. 66044

 Kentucky Geological Survey
 311 Breckinridge Mall. University o<  Ken-
 tucky. Lexington. Ky.. 40506
Louisiana Geological Survey
ion C, Univefsily Station. Salon Kou«e.
La.. 79003

Maine tunau o4 Geology
State OWc* Bwiding. Room 211. Augusta.
Me.. 04330

MaryUwl Gcotogiul Survey
Menyman Han, lohns Kopkinf Unrvenny.
ftaltimor*. Met.. 21218

MaiMdHHtttf  Department  of Environ-
ntenlal Quafily Engineering
OrvHion o< Waterways.  100 Nashua St..
Room S12. Boston. Mass.. 02114

Michigan Geoiogkal Survey Division
•ox 30028. Lansing. Mich., 48909

Minnesota Geological Survey
1613 Cusiis St., St Paul. Minn., 55106
MUtiuippi Cealogk. (co«omk A Topo-
       "4H«Cf
        , iackson. MJsj., 39216
grapMc Survey
60x4915,1
 Miuouri DMcioii  of Geological & Land
 Survey
 Box ISO. Holla. Mo.. 65401
 Montana Bureau of Mines 4, Geology
 Montana College of Mineral Science
 Technology. Butte, Mont., 59701
 Nebncka Conservation * Survey OMsion
 University al Nebraska. Lincoln. Neb..
 68506

 Nevaila turciu of Mines & Geology
 University of Nevada, Reno, Nev.. 89S57

 New HampsMfC Oepartment of Resourcei
 A Iconomic Development
 James Hall, University ol New Hampshire.
 Durham. N.H.. 03824

 New lersey tureau of Geology I> Topogra-
 phy
 Box 1390, Trenwn, N.|., 08625

 New Mexico Bureau of Mines & mineral
 Resources
 Socorro, N.M.. 87J01

 New York Slate Geological Survey
 New York Slate Education Building, Room
 973. Albany. N.Y., 12224

 North Carolina Department of Natural Re-
 source* A Community Development
 Geological Survey  Section.  Box 27687.
 Raleigh. N.C. 27611

 North Dakota Geological Survey  •
 University Station.  Grand  Forks. N.O.,
 S8202
Cmlo Division «f Geological Survey
fountain Square. Building 6. CokMnbus,
Ohio. 43224

Oklahoma Geological Survey
830 Van Vteet Oval. Room HI. Norman.
OUa.. 73069

Oregon Oepartment of Geology & Mineral
Industries
1069 State Office Building. 1400 SW fifth
Ave.. Portland. Ore.. 97201

FerMMytvania Bureau of Topographic a,
Geologic Survey
Department of Environmental Resources.
Box 23S7, Harrisburg, Pa.. 17120

Puerto Rico Service Ceotogico
AfMrtado 5487. Pueru de Tierra, San |uan,
P.R., 00906

 South Carolina Geological Survey
 State Development Board, Harbison  for-
 est Road. Columbia. S.C.. 29210

 South Dakota Geological Survey
 Science Center. Utwversitv of South Dako-
 ta, Vermillion. S.D...S7069

 Tennessee Division of Geology
 C-S  Stale   Office Building.  NajhviHe.
 Term.. 37219

 Texas Bureau of Economic Geology
 Unrversity of Texas. Box X. Uruversrty Sta-
 tion. Austin, Tex.. 76712

 Utah Geological * Mineral Survey
 606 Black Hawk Way. Salt Lake Cry. Utah,
 64106

 Vermont Geological Survey
 Agency of  environmental Conservation.
 Montpelier. Vt.. 05602

 Virginia Division of Mineral Resources
 Box 3667. Charlottesville. Va., 22903

 Washington Division of Geology 4  Earth
  Resources
  Olympia. Wash.. 98S04

  Weil Virginia Geological & Economic Sur-
  vey
  Bon 879, .Morgantown, W. Va.. 26505

  Wisconsin Geological  & Natural History
  Survey.
  1815 University Ave.. Madison. Wis..  53706

  Wyoming Geological Survey
  Box 3008. University Station. University of
  Wyoming. Laramie, Wyo.. 82071
                                                        17

-------
TABLE 2.4: TYPICAL SEISMIC EVENT DATA AVAILABLE FROM
     USGS NATIONAL EARTHQUAKE INFORMATION CENTER
C OAT
C
r
4)1

• 1

• 1


• 1

• 1



• 1
• t
41
41

41

41
4)1
• 1
• 1

• 1

4)1
• 1
4t
• t
• 1

• 1
• 1

• 1
• 1

•'
• 1
• I
• 1
«1

' *1




• 1
• 1
• 1
• 1
• 1
• 1
• 1
OKICIN TIMC
VIC
Mt UN SCC
M 14 29.9*

• 1

4)1


4)1

• 1



42
•2
•2
•2

• 3

•3
• 4
•4
•4

•9

•4
4)4
44
4)4
•*

•*
4*

4*
44

4*
4*
14
14
11

11




12
»2
17
13
19
14
14

13

14


34

3*



•3
12
4)4
9*

44

47
3*
3*
93

13

23
J7
Y4>
47
•2

12
21

14
39

41
13
37
34
1*

34




•4
14
9*
31

•3.4T

17.1


19. *X

11 .4



4*. 3
4*.*
9* .9
1*.1«

44. •*

92.3.
93.*
97.3
34.9.

4)7.3-

14.*.
•7. It
47.4.
SS-f)
2*. 4

14.4*
••.9

94.3.
43.4

94,4.
31.4.
34.37
34.4.
43.7

4*. 3




44.DI
43. Zt
43.41
J7.S7
33-27.7.
43
12
4*.*7
34.*
OOMOINA1C*
IAT LONC
34.*4«*4tl«»l.
MCA* S. COAST Or HONSHU. JAPAN. f«ll (II JUA) «« A|
•nj (1 JUA) •! MI«kl4M. T«kr» 4>«41 4MI O«fcl4M.
MCA* S. COAS1 Or HONSHU. JAPAN, fell (IV JUA) •« A|
(III JMA) «t T4tl4y4«4. Ml4fcl4»4. T4lL4lb4MM« 4«l4 4HI O*ll
(II JUA) •( M4*4tl««4 4)«4 T«kr«: «. ' '
CHILC-4)OLIVIA *O«OC« IICCION
fKANCC. ML 1.9 (CCN). 1.3 (tOC).
flJI IStAMOS *CCION
MCA* S. COAST Or HONSHU. JAPAN. f«lt (II JUA) •«
OtftlM: (1 JUA) «l AJIr«.
•It** S. COAST Or HONSHU. JAPAN. f«lt (II J«A) •« A|
(1 JUA) «l Ml ••>>•« •"<• •« Othlo*.
OCA* CAST COAST Or KAMCHATKA
rui ISLANDS RCCION
MOLUCCA PASSACC
MCA* S. COAST or HONSHU. JAPAN. f«l« (1 JUA) •( AJI
Ml»hl4M ««4 •* Otni4M.
MCA* S. COAST Of HONSHU. JAPAN. Mil (II JUA) 4*
04HIM: (1 JUA) «l A)lf4. Ul>kl«« «Ml Tl1i«n r«llf. Ue«l«««.
TlWlCCT
•ANOA StA
r«AHCC. UL 2.7 (IDC).
11 MO*
• URUA
WCSI CHILC *ISt
NCAIt S. COAST Or HONSHU. JAPAN. t»\\ (II JUA) «l A)
                           18

-------
          I     ,	,	.	,	,	.	,
60* -
      Figure 2.1    The Six Major Tectonic Plates and Thek  Approximate Linear Velocity
                   Vectors (adapted from Park, 1983).

                                              19

-------
Figure 2.2    Seismic Source Areas in the United States (Krinitzsky et ah, 1993).

-------
SAN FRANCISCO

 1906
  SAN FERNANDO
       1971
Figure 2.3    Isoseismal Contours for Intra-Plate vs. Edge-Plate Events of Similiar
             Magnitude (Nuttli, 1981)

                                     21

-------
       40 F
       30
              -88   -86   -84   -82   -80    -78    -76
                           Longitude  (Deg.)
Figure 2.4    Epicenters for Earthquakes M £ 2.5 in the Southeastern United States (My
            1977 - December 1984) (Sibol et al., 1984).
                                    22

-------
                                        ^ - ^    "\AV^"^  fc
                                                                 MAIN ZONE OF
                                                                FAULT EVIDENCE
                           TOTAL WIOTV  OF >AULT ZONE 78 FtET
Figure 2.5    Characteristics of Hayward Fault as Exposed in Five Trenches at Fremont
             Site(Cluffetal., 1972).

                                         23

-------
                  Hill  •



Figure 2.6   Detail of West Fault Trace Exposed in Trench "G" at Fremont Site (see
            Fig. 2.5) (Cluff et al.,  1972).

                                      24

-------
                                      SECTION 3

                          258.14 SEISMIC IMPACT ZONES:
                USGS PROBABILISTIC BEDROCK ACCELERATION

Subtitle D provides for two alternative methods to determine the maximum (peak) horizontal
acceleration (MHA) for design of MSWLF's.  The prescriptive method of determining the MHA
is from a seismic hazard map depicting the peak horizontal ground acceleration in Minified earth
(bedrock) with a probability of 90 percent (or greater) of not being exceeded in a 250 year period.
Either USGS Map Sheet MF-2120 (USGS, 1982; USGS, 1990), as presented in Figure 1.1 and
used to define  the extent of seismic impact zones  in the United States, or an equivalent map
acceptable to  the director of a USEPA-approved state or tribal regulatory program may be used.

Alternatively, the MHA may be based upon a site specific seismic hazard analysis.  The details
of what constitutes an acceptable site specific hazard analysis  are not provided in Subtitle D.
Rather, these details are left to the discretion of the director of a  USEPA-approved state or tribal
regulatory program.  Many states simply provide for site specific analysis that determine the peak
horizontal ground acceleration in lithified earth with a  90 percent (or greater) probability of
exceedance in 250 years.  Figure 3.1 outlines the steps required for such an analysis.   Other states
allow for use of deterministic analysis to determine the largest or most damaging earthquake
expected to impact the site.

Many experts consider the use of site specific analyses preferable to use of generic seismic hazard
maps for assessing the peak ground acceleration for engineering analyses  due to the ability to
achieve a greater  degree of precision and to incorporate up-to-date information  on regional
seismology and tectonics in a site specific analysis (Anderson  and Kavazanjian, 1995).  Other
experts maintain not only the superiority of site specific analyses over seismic hazard but also the
superiority of deterministic analyses over probabilistic seismic hazard  evaluation (Krinitzsky,
1993).  Discussion of these issues is beyond the scope of this guidance document.

The USGS probabilistic seismic hazard map presented in Figure 1.1 provides the  prescriptive
means of determining the MHA for MSWLF. The  latest available version of the map should be
used in the analyses.  If knowledge and understanding of local and regional seismology and
geology or attenuation of earthquake ground motions have changed sufficiently since development
of the map to invalidate the map, or if it is believed that the map is otherwise inappropriate, a site-
specific analysis may be warranted. The details of an acceptable site specific analysis are left up
to the discretion of a USEPA-approved state or tribal regulatory program. However, considering
the relative low probability and large return period associated with the peak acceleration evaluated
from the USGS map, in most situations landfills designed using the peak acceleration from the

                                           25

-------
most recent version of the map will afford a high degree of protection to the environment against
earthquakes reasonably expected over the life and post-closure care period of the MSWLF.

The USGS  map presents the estimated peak ground acceleration for a hypothetical bedrock
outcrop at a project site.  If bedrock is not present at or near the ground surface, the peak
acceleration from the USGS map may need to be modified to account for local site conditions. The
primary difficulty associated  with the USGS or  other seismic hazard map is that such maps
typically do not provide  information on the magnitude or duration of the earthquake associated
with the map acceleration values. In fact, the acceleration values provided on such maps are
typically composed of contributions of earthquakes  of many different magnitudes at many different
distances.  For most geotechnical analyses, peak acceleration and  magnitude are necessary.
Distance and/or duration may also be required for certain geotechnical evaluations.

This section of the guidance document provides background on the methodology used to generate
the USGS seismic probability map and discusses interpretation of the acceleration value obtained
from the map in order to obtain site-specific seismicity parameters for design.

3.1    Development of  Design Earthquake

The USGS map is shown on Figure 1.1 in a reduced size format.  The original map generated
by  USGS is sufficiently  large that individual counties within the states are shown for ease in
locating a  particular landfill site.  Selection of a peak horizontal ground acceleration from this map
is a straight forward process.  However, association of a magnitude with this peak acceleration
requires interpretation and judgement.

An acceleration value from the USGS maps for any particular site is composed of contributions
from a family of earthquakes of different magnitudes and distances.  Figure 3.2 (Moriwaki et al.,
1994) shows the distribution of magnitudes and distances from a hypothetical probabilistic seismic
hazard analysis for a 10  percent probability of exceedance (90 percent probability of not being
exceeded) over a 50 year exposure period  (this corresponds  to a 475  year return period).
Selection of a representative magnitude from this data  might be based upon either a 90 percentile
criterion or the mean (expected) value at the discretion of the design engineer and regulatory
agency.

The information on the distribution of magnitudes is  generally not available for USGS or other
regional seismic hazard studies, and most common seismic hazard programs must be modified to
yield this  data. As an alternative, the maximum magnitude assigned to the  seismic source zones
which contribute to the seismic hazard (the zone the site is in plus all adjacent zones) may be
conservatively taken as the representative magnitude.  The source zones used to develop the 1982

                                            26

-------
USGS map are presented in Figure 3.2.  No supplemental documentation was presented with the
1990 map.  Maximum magnitudes are presented in Table 3.1 for the USGS source zones shown
in Figure 3.3, and Figure 3.4 presents maximum magnitudes for much of the central United States
from a more recent study (Johnston and Nava, 1994).  Knowledge and understanding of seismic
source zones is continually advancing. The prudent investigator should consult current sources
of information on local and regional seismicity to identify the source zones impacting the site and
the maximum magnitudes assigned to these source zones.

In performing a seismic performance analysis  for a MSW landfill, it should be recognized by the
engineer and regulator that the peak ground acceleration at the site is not always the acceleration
associated with the most damaging earthquake. Large magnitude earthquakes at large distances
can generate ground motions at a site that are  of lower intensity but greater damage potential than
a small magnitude nearby event associated with the MHA.  Use of the maximum magnitude from
all contributing source zones as the magnitude associated with the MHA combined with the low
probability occurrence and large return period associated with the MHA will generally provide
a design event of sufficient damage potential to provide a high degree of environmental protection
over its' active life and post closure care period.

It is anticipated that use of the maximum magnitude event from all contributing source zones in
combination with the USGS map MHA will often produce a very conservative assessment of the
design event. However, in some situations, most notably when the site is on the fringe of a major
seismic source zone capable of generating a great earthquake (e.g., San Andreas, New Madrid,
Charleston, Cascadia Subduction Zone),  the large distant event with a lower PGA may be the
most damaging earthquake and an event-specific analyses for the MHA associated with the great
earthquake may be warranted in addition to the use of the MHA from  the seismic hazard map.
If such an event-specific analysis is conducted, the magnitude from the great earthquake source
zone  need not  be  considered  in evaluating the magnitude associated with the  USGS  map
acceleration.

3.2    Interpretation of Peak Bedrock  Accelerations

The attenuation  relationships used to establish the USGS seismic probability maps are based on
ground motions  recorded at bedrock sites. Bedrock is commonly defined in engineering practice
as material having a shear  wave velocity greater than 2,500 feet per second (750 meters per
second).  This is referred to as lithified  earth within Subtitle D.  Lithified earth is defined in
Subtitle D as all  rock, including all naturally occurring and naturally formed aggregates or masses
of minerals or  small particles  of older  rock that formed by  induration of loose sediments.
Lithified  earth  does not include man-made materials such as fill, concrete and  asphalt,  or
unconsolidated earth materials, soil, or regolith (saprolites) lying at or  near the ground surface.

                                           27

-------
It is important to realize that the accelerations presented on the USGS maps are not the peak
ground surface accelerations unless bedrock is exposed at the ground surface.  Section 4.1 of this
guidance document reviews methods for calculating the peak ground surface acceleration based
on the site specific subgrade profile that exists above the top of Minified earth (rock) and the peak
bedrock acceleration from the USGS map.

The peak acceleration is only one characteristic of the earthquake ground motion at a site.  The
damage potential of seismically-induced ground motions also depends upon the duration of the
motion, the frequency content of the motion, and the intensity of the motion at times other than
when the peak acceleration occurs.  Acceleration,  velocity, and  displacement time-histories
recorded at the top  of the Oil landfill in Los Angeles during the 17 January 1994 Northridge
earthquake (Hushmand Associates,  1994) are shown on Figure 3.5.  Note that the peak
acceleration occurs only once during the  record  and  that motions  approaching the peak
acceleration exist for only a small fraction of a  second.  Use of this peak acceleration for
traditional geotechnical stability analyses is very conservative in most cases.  Section 6.2 of this
guidance  document discusses the  reduction of this peak  ground surface acceleration to  an
equivalent pseudo-static acceleration for use in slope stability analyses.

3.3    References

Algermissen, S.T.,  Perkins, D.M. Thenhaus, P.C., Hanson, S.L., and Bender, B.L. (1982),
"Probabilistic Estimates of Maximum Acceleration and Velocity in Rock in the Contiguous United
States," U.S. Geological Survey Open-File Report 82-1033, 99 p.

Anderson, D.G., and Kavazanjian, E., Jr. (1995),  "Performance  of Landfills Under Seismic
Loading," Proc. Third International Conference on Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, University of Missouri, Rolla,  Vol. 3, 2-7 April.

Bonilla,  M.G.,  Mark  R.K.,  and Lienkaemper,  JJ. (1984),  "Statistical  Relations Among
Earthquake  Magnitude, Surface Rupture Length and Surface Fault Displacement, Journal of
Geophysical Research, Vol. 74, No. 6, pp. 2379-2411.

Boore, D.M., and Joyner, W.B. (1994), "Prediction of Ground Motion in North America," Proc,
of the Seminar on New Developments in Earthquake Ground Motion Estimation and Implication
for Engineering Design Practice,  Applied Technology Council Publication No.  ATC 35-1,
Redwood City, California, pp. 6-1 - 6-41.

Cornell, C.A. (1968), "Engineering Seismic Risk Analysis," Seismological Society of America
Bulletin, V. 58, pp. 1583-1606.

                                           28

-------
Hanks, T.C., and Kanamori, H. (1979), "A Moment Magnitude Scale," Journal of Geophysical
Research, Vol. 84, No. B5, pp. 2348-2350.

Heaton, T.J., Tajima, F., and Mori, A.W. (1986), "Estimating Ground Motions Using Recorded
Accelerograms" Surveys in Geophysics, Vol. 8, pp. 25-83.

Hushmand Associates (1994), "Landfill Response to Seismic Events," Report prepared for the
USEPA Region IX, Hushmand Associates, Laguna Niguel, California.

Idriss,  I.M. (1985),  "Evaluating  Seismic  Risk in Engineering Practice,"  Proc. Eleventh
International Conference on  Soil  Mechanics and Foundation Engineering,  San  Francisco,
California, Vol.  1, pp. 255-320.

Johnston, A.C., and Nava, S.J. (1994),  "Seismic Hazard  Assessment in the Central United
States," Proc. Seminar on New Developments in Earthquake Ground Motion Estimation and
Implications for Engineering Design Practice, Applied Technology Council, ATC35-1, Redwood
City, California, pp. 2-1 - 2-12.

Krinitzsky, E.L. (1993),  "Earthquake Probability in Engineering - Part 2: Earthquake Recurrence
and  Limitations of Gutenberg-Richter b-values for  the  Engineering of Critical Structures,"
Engineering Geology, Vol. 36, pp. 1-52.

Moriwaki, Y., Tan, P. and Somerville, P. (1994); "Some Recent Site-Specific Ground Motion
Evaluations - Southern California Examples and Selected Issues," Proc.  Seminar  on New
Developments in Earthquake Ground Motion Estimation and Implications for Engineering Design
Practice, Applied Technology Council ATC35-1, Redwood City, California, pp. 14-1  - 14-25.

Nuttli, O.W. (1981), "Similarities and Differences Between Western and Eastern United States
Earthquakes, and their Consequences for Earthquake Engineering," Earthquakes and Earthquake
Engineering: The Eastern United States, Vol. 1, Assessing the Hazard - Evaluating the Risk, J.E.
Beavers, Ed., Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan, pp. 25-51.

USGS (1982),  "Probabilistic Estimates of Maximum Acceleration and Velocity in Rock in the
Contiguous United States, United State Geological Survey, Open-File Report 82-1033.

Wesnousky, S.G. (1986), "Earthquakes, Quaternary Faults and Seismic Hazard in California,"
Journal of Geophysical Research, Vol. 91, No. B12, pp.  12587-12631.
                                          29

-------
Woodward-Clyde Consultants (1979),  "Evaluation of Maximum Earthquake and Site Ground
Motion  Parameters  Associated with the Offshore  Zone of Faulting, San  Onofre Nuclear
Generating Station," Report prepared for Southern California Edison Company, 241 p.

Wyss, M, (1979),  "Estimating Maximum Expectable Magnitude of Earthquakes from Fault
Dimensions," Geology Vol. 7, pp. 336-340,
                                         30

-------
Table 3.1:   Parameters for Seismic Souice Zones (USGS, 1982).
Zone
No.*
pQOl
(002
p003
pOOA
pOOS
p006
p008
p009
pOlO
pOll
p012
pO!3
pO!4
pOlS
p016
pQl?
pOJ8
p019
cOOl
C002
c003
c004
c005
C006
c007
c008
c009
0010
cOll
0012
c013
0014
c015
d016
cO!7
C018
cO!9
CD20
c021
0022
c023
c024.
c025
0026
No. of Modified
Mercalli Maxima
Intensity V'a
per year
0.1 1010
0.43510
0.12440
0.34840
0.12390
0.02831
0.01642
0.20850
0.4 5200
0.96370
0.37090
0.69020
0.10940
0.34480
0.04926
0,87860
0.18810
0.04090
0.62770
0.1S700
0.31960
0.31960
0.04843
0.15700
0.15700
0.04740
0.04843
0.18190
0.77010
0.19050
0.3 5840
0.91990
1.49200
0.22560
0.02760
1.09200
0.3 1980
0.19280
0.10880
0.02422
0.11650
1.97000
0.0 5085
0.09145
b
-0.40
-0.40
-0.54
-0.62
-0.62
-0.62
-0.42
-0.28
-0.28
-0.28
-0.28
-0.28
-0.42
-0.42
-0.42
-0.28
-0.54
-0.54
-0.42
-0.4*2
-0.42
-0.42
-0.42
-0.42
-0.42
-0.42
-0.42
-0.42
-0.42
-0.42
-0.42
-0.66
-0.45
-0.51
-0.48
-0.49
-0.42
-0.42
-0.42
-0.42
-0.37
-0.43
-0.55
-0.55
Maximum
Magnitude
K**
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.9
7.9
7.9
7.9
7.9
7.3
7.3
7.3
7.9
7.3
7.3
7.3
7.3
7.3
7.3
6.1
7.3
7.3
6.1
6.1
6.1
7.3
7.3
7.3
7.9
7.9
7.9
7.3
7.3
6.7
6.1
6.1
6.1
7.9
8.5
7.3
7.3
                                        31

-------
Table 3.1:    (continued)
Zone
No.*
0027
c028
c029
c030
c031
c032
c033
c034
c035
dD36
e037
c038
cQ39
d)40
cQ41
001
002
003
004
005
006
00?
008
009
010
Oil
012
013
014
015
016
017
018
019
020
022
023
024
025
026
027
029
030
031
Ko. of Modified
Me rcall! Maximum
Intensity V's
per year
0.03437
0.13010
0.02350
0.03630
0.47580
0.55190
0.23070
0.67120
0.02325
0.35220
0.8 1950
0.82680
0.35810
0.15820
0.08448
0.22700
0.03600
0.08800
0.22700
0.09100
0.13500
0.41900
0.21100
0.19400
0.20800
0.55100
0.34900
0.05500
0.49000
0.01800
0.14600
0.69300
0.26100
0.11717
1.84900
0.19600
0.15350
0.27400
0.16800
0.47700
0.1 1100
1.31900
0.58800
1 .82685
b
-0.37
-0.37
-0.37
-0.42
-0.51
-0.45
-0.37
-0.51
-0.60
-0.59
-0.51
-0.54
-0.45
-0.42
-0.37
-0.73
-0.73
-0.73
-0.54
-0.73
-0.73
-0.73
-0.73
-0.54
-0.54
- 0.164
-0.64
-0.64
-0.73
-0.73
-0.73
-0.59
-0.54
-0.54
-0.64
-0.64
-0.54
-0.64
-0.64
-0.64
-0.64
-0.64
-0,64
-0.54
Maxima
Magnitude
H**
7.3
7.3
7.3
6.7
6.7
7.9
7.9
7.9
7.3
6.7
6.1
7.9
7.9
6.1
7.9
7,3
7.3
6.1
7,3
7.3
7.3
7.3
6.1
6.1
7.3
7.3
7.3
7.3
7.3
6.7
6.1
7.3
7.3
7.3
7.3
6.1
7.3
7.3
6.1
6.1
5.5
7.3
7.3
7.3
                                     32

-------
Table 3.1:    (continued)
Zone
No.*
032
033
034
035
036
037
038
039
040
041
042
043
044
045
046
047
048
049
050
051
052
053
054
055
056
057
058
059
060
061
062
063
064
065
066
067
068
069
070
071
072
073
074
075
No. of Modified
Mercalli Maximo
Intensity V'e
per year
0.48114
0.08557
0.62380
0.20070
0.01800
0.05100
0.8 0600
0.12000
0.29100
0.24400
0.01800
0.04600
0.11300
0.45600
0.0 1274
0.00427
0.00329
0.01663
0.17000
0.01706
0.19000
0.03600
0.01800
0,67300
0.17700
0.66200
0.19800
0.19200
0.03600
0.08900
0.03600
0.12900
0.34400
0.15200
0.01800
0.07715
0.02894
0.00588
0.03552
0.01176
0.02026
0.02353
0.00270
0.06510
b
-0.54
-0.54
-0.54
-0.54
-0.58
-0.58
-0.58
-0.58
-0.58
-0.73
-0.73
-0.73
-0.73
-0.73
-0.73
-0.73
-0.73
-0.73
-0.73
-0.73
-0,58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.58
-0.73
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
Maxioun
Magnitude
M**
6.1
6.1
7.3
7.3
6.1
7.3
7.3
7.3
7.3
7.3
6.1
7.3
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
7.3
7.3
6.1
7.3
6.1
7.3
7.3
6.1
6.1
7.3
6.1
6.1
7.3
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
                                    33

-------
Table 3.1:    (continued)
Zone
No.*
076
077
078
079
080
081
082
083
084
085
086
087
088
089
090
091
092
093
094
095
096
097
098
099
100
101
102
103
104
106
107
108
109
110
111
112
113
114
115
116
117
118
No. of Modified
Mercalll Maxima
Intensity V's
per year
0,14742
0.03469
0.04389
0.03082
0.02987
0.02044
0.03552
0.00996
0.04117
0.03802
0.04626
0.29865
0.09703
0.15689
0.06103
0.00644
0.02661
0.02680
0.10835
0.05901
0.02675
0.01156
0.01215
0.24830
0.42290
0.18720
0.09532
0.33150
0.05544
0.01952
0.19100
0.29390
0.10650
0.30220
0.32430
0.01532
0.07432
0.00754
0.05834
0.06783
0.03950
0.01334
b
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.46
-0.50
-0.50
-0.50
-0.50
-0.59
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
-0.50
Maximum
Magnitude
M**
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
8.5
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
7.3
7.3
7.3
7.3
7.3
7.3
6.7
7.3
6.7
7.9
7.9
7.9
6.7
6.7
6.7
7.3
6.7
7.3
7.3
*The zones are shown In Figure  3.2
**See text for definition of M
                                    34

-------
       (A)
                       Source  1 (FAULT)
                        *  NI
   Source 2
   
-------
                   475-Year Average Return Period
                                                       0-15 km
                                                   15-50 km
         5.0   5.5   6.0  6.5   7.0   7.5   8.0
                     Magnitude
Figure 3.2   Contribution of Various Magnitudes and Distances to the Seismic Hazard
          (Moriwald et al., 1994).
                                 36

-------
UJ
                          Figure 3.3   Seismic Source Zones in the.Contiguous United States (0SGS, 1982).

-------
                         Superior Oat on
            iround
       ||||||NefiKihaf
                          Rcelfoot
                            M f

                           S.Cootta
                           A   Mh<5.O

  Tront-
  f%co«
  ST Mb6,6
Figure 3.4   Seismic Source Zones in the Central United States (Johnston and Nava,
                               38

-------
   -30
                   10
                   10
20
30
50
60
20         30         UO
   TIME - seconds
                       50
            60
Figure 3.5   Time-Dependant Fluctuations in Seismic Ground Response Parameters (17
            January 1994 Northridge,  California Earthquake, OH Site, Longitudinal
            Component) (Hushmand Associates, 1994).

                                      39

-------
                                     SECTION 4

                          258.14 SEISMIC IMPACT ZONES:
               SITE SPECIFIC SEISMIC DESIGN GROUND MOTION

The USGS map discussed in the previous section provides values for the peak ground acceleration
of a hypothesized bedrock outcrop at a MSW landfill site. This section of the guidance document
discusses methods for calculation of: (1) a peak acceleration in the free field at the ground surface
at the project site that reflects the soil stratigraphy and (2) a peak acceleration at the top of the
landfill that reflects the properties of the waste. These accelerations are used in later  sections of
this guidance document in evaluation of the  seismic response of the landfill waste mass, the
seismic performance of the liner and cover systems,  and subgrade liquefaction potential.

Qualitative reports of the influence of local soil conditions on the intensity of shaking  and on the
damage induced by earthquake ground motions  date back to at least the 1906 San Francisco
earthquake (Wood, 1908). Reports of localization of areas of major damage within the same city
and of preferential damage to buildings of a certain  height within the same local area from the
Mexico City earthquake of  1957, the  Skopje, Macedonia earthquake of 1963, and the Caracas,
Venezuela earthquake of 1967 focused the attention of the engineering community on local soil
effects.

Back-analysis  by Seed  (1975) of  accelerograms from the magnitude M  5.7  San Francisco
earthquake of 22 March 1957, presented in Figure 4.1, demonstrate the  influence of local soil
conditions on site response.  Peak accelerations  and the frequency contents  of ground motions
measured at six sites approximately the same distance from the earthquake source were dependent
on the soil profile beneath each specific site.

Figure 4.1  shows  peak acceleration,  the acceleration and velocity response spectra,  and soil
stratigraphy data at the six San Francisco sites from the 1957 earthquake.  A response spectrum
presents the maximum response of a damped single degree-of-freedom  (SDOF) linear elastic
system to the accelerogram  recorded at a site. The maximum response of the SDOF system is
calculated for  a range of system natural frequencies to plot the response spectrum.  Response
spectra are typically calculated for  several levels of system damping, as  shown on Figure 4.2.
Acceleration data generated in response spectra analysis is commonly plotted on the tripartite plot
shown on Figure 4.3. In addition to peak acceleration, the tripartite presentation also provides
approximate values of peak  velocity and peak displacement for the response of the SDOF.

At the sites shown in Figure 4.1, the local soil deposits attenuated the peak ground acceleration
by a factor of approximately two compared  to the  bedrock  sites.  However, the acceleration

                                          40

-------
response spectra clearly show amplification of spectral accelerations at longer periods (periods
greater than 0.25 sec).  If the bedrock motions had greater energy at these  longer periods,  a
characteristic of larger magnitude events and of events from a more distant source,  or if the
natural period of the local soil deposits more closely matched the predominant period of the
bedrock motions, amplification of the peak acceleration could have occurred at the soil sites.

Amplification of long period bedrock motions  by local soil deposits is now  accepted as an
important phenomenon that can exert a significant influence on the damage potential of earthquake
ground motions. Significant structural damage has been attributed to amplification of both peak
acceleration and spectral acceleration by local soil conditions. Amplification of peak acceleration
occurs when the resonant frequency of the soil deposit is close to the predominant frequencies of
the bedrock earthquake motions (the frequencies associated with the peaks of the acceleration
response spectra). The resonant frequencies, /„, of a soil layer (deposit)  of thickness H can be
estimated as a function of the average shear wave velocity of the layer, Vs, using the following
equation:
                    V (In    1)
              /   -	  n    (1,2,3,..)                                       (4.1)
where/, is the resonant frequency for the first mode of vibration, /2 is the resonant frequency for
the second mode of vibration, f3 is the resonant frequency for the third mode of vibration, and so
on.   At most soil sites, amplification  of seismic motions  is most important  for the first
(predominant) mode of vibration and rapidly decreases  in significance with increasing mode
number.

Spectral amplification may occur at soil sites in any earthquake at frequencies around the resonant
frequency of the soil deposit.  Spectral amplification causes damage when the resonant frequency
of the soil deposit matches the resonant frequency of the structure. Some of the most significant
damage in recent earthquakes (e.g., building damage in Mexico City in the 1985 earthquake and
damage to freeway structures in the Loma Prieta earthquake of 1989) has occurred in situations
where the predominant frequencies of the bedrock motions and the resonant frequencies of both
the local soil deposit and the overlying structure all  fell within the same range.

Observations of ground motions generated in recent earthquakes at the Oil landfill, a solid waste
landfill in Los Angeles composed of both industrial and municipal wastes, have demonstrated that
amplification of both spectral acceleration and peak acceleration can occur at the top of solid waste
landfills. Anderson et al. (1992) report spectral amplification of greater than 10 at Oil for low
amplitude (less than 0.1 g) ground motions from small magnitude (less than M 5.0) earthquakes.

                                           41

-------
Hushmand Associates (1994) report a peak horizontal acceleration amplification factor of 3.0 at
OH during the M 7.4 Landers earthquake in 1992.

Considering the landfill facility as an engineered structure built upon a local soil deposit, there
are clearly two different sources of local site effects that must be considered in a  seismic impact
analysis. First, the influence of the local soil conditions on the bedrock motions must be evaluated
to determine the free field ground surface motions at the project site.  Second, the influence of the
landfill on the free  field ground surface motions must be  evaluated.  While it is convenient
conceptually to separate these two effects, in practice they may be inter-dependent and a coupled
analysis of the interaction between the response of the  foundation soil and the response of the
landfill may be warranted.

This section of the guidance document presents simplified and detailed methods for evaluating
both the free field ground response and the response of the landfill mass. The free field ground
motions are used to evaluate the liquefaction potential of the foundation.  The response analysis
of the landfill mass provides input for seismic performance analyses of the landfill liner and cover
systems.

4.1    General Methodology

The influence of local soil conditions on seismic ground  motions is usually addressed using one-
dimensional site response analyses. Conventional one-dimensional site response  analyses  are
based upon the assumption of a horizontal shear wave propagating vertically upwards through
horizontal soil layers of infinite  lateral extent.  The influence of vertical motions,  compression
waves, and laterally non-uniform soil conditions are  typically  not accounted for in a one-
dimensional site response analysis.  Similarly, geotechnical engineering analyses of liquefaction
potential and seismic stability consider only  the horizontal component of the seismic motions.
This reliance solely on the horizontal  component is consistent with common design and code
practices.

The  most common  analytical method  used for one-dimensional site response analyses is  the
equivalent linear method, wherein a layered vertical soil column is treated as a linear visco-elastic
material characterized by an elastic modulus and a viscous damping ratio.  To account for the non-
linear, strain-dependent behavior of soil, the equivalent linear modulus  and damping ratio  are
evaluated from the modulus and damping  measured in  uniform  cyclic  loading  at  the
"representative" shear strain.  Based on comparison of observed seismic site response with site
response predicted using equivalent linear analysis, the representative shear strain is usually taken
as 65 percent of the maximum shear strain calculated in the site response analysis.  Because the
maximum shear strain is not known prior to the start of an analysis, equivalent linear response

                                           42

-------
analyses are performed in an iterative manner.  The maximum shear strain from one run is used
to evaluate the equivalent modulus and damping for the next run and continuing to convergence.

Input to one-dimensional equivalent linear site response analyses typically includes the shear wave
velocity and mass density or small-strain shear modulus for each soil layer, curves relating the
representative shear strain to a modulus reduction factor and the fraction of critical damping for
each soil type (modulus reduction and damping curves),  the representative shear strain factor (the
fraction of the maximum shear strain assumed to correspond to the representative shear strain) and
an input acceleration-time history.  Other input parameters include the density and  shear wave
velocity of the underlying bedrock.  The acceleration-time history may be input as the motion at
a hypothetical bedrock outcrop or at the bedrock-soil interface at  the base of the soil column.
Results of the analysis provide shear stress- and acceleration-time histories for each layer within
the soil profile.

An alternative to the equivalent linear  method of site response analysis is truly non-linear site
response  analysis (Lee and Finn, 1978; Matasovi and Vucetic, 1993).  In a truly non-linear
analysis, the actual hysteretic stress-strain behavior of each element of soil (or waste) is calculated
in the time domain.  Equivalent linear analysis are typically performed in the frequency domain,
employing the principal of superposition to calculate the time history of ground motions. Non-
linear site response analyses require a description of the hysteretic stress strain behavior of the soil
(or waste), the mass  density profile of the material, and an input acceleration time history.  Truly
non-linear site response analyses hold the promise of a more accurate representation of the  seismic
behavior of soil deposits and solid waste landfills.  However, at the present time, truly non-linear
site response analyses  are still primarily a research tool and have yet to be widely employed in
engineering practice.

4.1.1  Simplified Analysis

Whereas  structural  analyses typically  require information on the spectral content of  ground
motions, and thus require a complete time history to characterize the design motion, geotechnical
analyses frequently  only require knowledge  of either the peak ground acceleration or the peak
ground acceleration and the earthquake magnitude.  Several investigators have related the peak
ground acceleration from a hypothetical bedrock outcrop, such as presented on the USGS maps,
to the peak ground acceleration at a specific  site as a function of the local soil conditions based
upon the  results of one-dimensional site response analysis and observations of ground motions
during earthquakes.  The top plot on Figure  4.4 shows an early relationship developed by Seed
and Idriss (1982) for a variety of local soil conditions.  This plot was developed using SHAKE,
a computer program for equivalent linear one-dimensional site response analyses developed at the
University of California, Berkeley (Schnabel et al.,  1972).

                                            43

-------
Experience from recent earthquakes has shown that the curves in the top plot of Figure 4.4 can
significantly under-predict site amplification effects in many situations. The plot on the bottom
of Figure 4.4 shows a recent curve developed by Idriss (1990) for soft soil sites.  This plot was
developed from both SHAKE analysis and from field observations of soft soil site response in two
recent earthquakes.

Observations of the response of the Oil landfill, Los Angeles,  in recent earthquakes (Hushmand
Associates, 1994) and the results of truly non-linear one-dimensional seismic response analyses
of landfills  (Kavazanjian and Matasovi,  1995) indicate  that  the Idriss  (1990) soft soil-site
amplification curve may also provide an appropriate  representation of the potential for typical peak
acceleration amplification at solid waste landfills.  Data obtained at the OH landfill during four
recent earthquakes is plotted in Figure 4.5 along with the soft soil site field data and recommended
curve from Idriss (1990).  Also plotted on this figure are the results of non-linear analyses of
landfill seismic response performed by Kavazanjian  and Matasovi (1995) using waste parameters
backfigured from strong motion records obtained at  the Oil landfill in the 17 January 1994 M 6.7
Northridge earthquake (peak acceleration at the landfill crest equal to 0.24 g).

Some of the non-linear landfill response analyses results plotted on Figure 4.5 at 0.3 g and 0.5 g
bedrock acceleration fall significantly above the Idriss (1990) curve.  However, the results that
fall above the Idriss curve  are from low amplitude (less than 0.1  g) accelerograms recorded at
large distances from the earthquake source (greater than 50 kilometers) that were scaled up to
large accelerations representative of near field conditions. Therefore, the large amplification
factors computed for these cases may not be representative of the amplification potential from real
earthquakes.  On this basis,  Kavazanjian and Matasovi (1995) concluded mat the Idriss (1990) soft
soil amplification curve provides a reasonable representation of the average peak acceleration
amplification potential at the top of solid waste landfills.

Figure 4.6,  from Singh and Sun (1995), present the results of their theoretical analyses for the
amplification potential of a 100 ft (30 m)- high refuse fill along with a summary of the upper
bound for observations of amplification at  the crest  of earth dams in earthquakes by Harder
(1991). These investigators suggest that the observational data on earth dams may also provide
an upper bound on the amplification potential of waste fills.  The earth dam curve corresponds
closely to the upper range of the analytical  results of Kavazanjian and Matasovi (1995) presented
in Figure 4,5.

The soft soil site curve  developed by Idriss and  presented in Figure 4,5,  the analytical data
developed by Kavazanjian and Matasovi (1995) and presented in Figure 4.5, and the observatinal
data presented in Figure 4.6 may  be used in a three-step simplified procedure  developed by
GeoSyntec (1994) to perform a simplified site response analyses for the purpose of adjusting the

                                           44

-------
peak acceleration from the USGS map (or from a site-specific analysis for the MHA in lithified
earth) for the influence of local soil conditions  (to obtain the free field peak acceleration at a
project site) and for the influence of the landfill (to obtain the peak acceleration at the crest of the
landfill). The three-step procedure is as follows:

Step 1:  Classify the Site.  Classify the site as special study, soft, medium stiff, stiff, or rock on
         the basis of the average shear wave velocity for the top 30 meters  (100 feet) of soil and
         the following table (Borcherdt, 1994):

        CLASSIFICATION             AVERAGE SHEAR WAVE VELOCITY

          Special Study                  Less than 100 m/s       (330ft/s)
          Soft                           100 to 200 m/s           (330 to 660 ft/s)
          Medium Stiff                  200 to 375 m/s           (600 to 1,230 ft/s)
          Stiff                          375 to 700 m/s           (1,230 to 2,300 ft/s)
          Rock                          Greater than 700 m/s    (2,300 ft/s)

          Note that special study  soils also include liquefiable soils, quick and highly sensitive
          clays, peats, highly organic clays, very high plasticity clays (PI>75%), and soft soil
          deposits more than 37 meters (120 feet) thick.

Step 2:   Estimate the Free field Acceleration.  Estimate the potential amplification  of the
          bedrock motions by the local soil deposit based upon the soil profile classification. For
          soft soils, use the curve in Figure 4.5 recommended by Idriss (1990) to estimate the
          free field peak ground acceleration from the peak bedrock acceleration.  For medium
          stiff soils, use an acceleration equal to the average of the rock site acceleration and the
          soft soil site  acceleration from  Idriss1 curve  in  Figure  4.5 for peak  bedrock
          accelerations less than or equal to 0.4 g.   For medium  stiff soils when the peak
          bedrock acceleration exceeds 0.4 g and for stiff sites for all acceleration levels, assume
          the  free field  peak ground acceleration at  the  site is equal  to the peak  bedrock
          acceleration. For Special Study soil sites, Figure 4.5 should not be used.  Instead, site
          specific seismic response analyses such as those described in the next section  of this
          guidance document should be conducted.

Step 3:   Estimate the Peak Acceleration at the Top of the Landfill.  Estimate the potential
          amplification of the peak acceleration of the landfill mass using the  analytical data in
          Figure 4.5 and the earth dam amplification curve in Figure 4.6. The decision as to
          whether to use  the upper bound of the  analytical data and/or the  earth  dams
          observations or to use a value closer to the median of the analytical data in Figure 4.5

                                           45

-------
          is a matter of engineering judgment. As a general rule, if the predominant period of
          either the design earthquake or the foundation soil matches the predominant period of
          the landfill, the upper bound data should be used.  The predominant period of the
          landfill and the foundation may be evaluated using equation 4.1.  For relatively thin
          fills (less than 33 ft (10 m) founded on soft soils or subject to long period motions
          from distant earthquakes and for thick  fills (over 165 ft (50 m)) founded on rock
          subject to high frequency motions from nearby earthquakes, the median analytical data
          (corresponding roughly to the Idriss soft soil curve) may be appropriate. The free field
          ground acceleration  developed in Step 2  is used in place of the peak bedrock
          acceleration on the abscissa of Figures  4.5 and 4.6, and the acceleration at the top of
          the landfill is obtained from the ordinate of the appropriate figure.

The three-step procedure  presented  above is a simplified,  decoupled analysis  that  ignores
interaction between the waste mass and the  ground.  Analyses of the coupled response of landfills
and foundation soils indicates that this simplified, decoupled analysis will yield a conservative
upper bound estimate  of the combined amplification potential of a  landfill and its foundation
(Bray, etal., 1995; GeoSyntec,  1994).

The peak acceleration at the top of  the landfill  estimated  in  Step 3 may be used in  seismic
performance analyses of the landfill cover  and surface water drainage systems and in evaluation
of other facilities constructed on top  of the landfill (e.g., flare station or storage tanks).  The
acceleration calculated in Step 3 is not, however, the  appropriate peak acceleration for use in
seismic stability and deformation potential calculations of the waste mass.  For seismic stability
and deformation potential evaluations of the waste  mass, the average acceleration  of the assumed
failure mass, and  not the acceleration at the top of the landfill, is the relevant response quantity,
as the average acceleration is directly proportional to the seismically-induced inertia forces and
to the seismic shear stresses induced at the base of the failure mass (Repetto et al., 1993).

Makdisi and  Seed (1978)  developed a  "typical" curve relating the  ratio of peak average
acceleration to peak ground acceleration to  the depth of the failure surface for earth dams founded
on rock. Kavazanjian and Matasovi (1995) demonstrated that the Makdisi and Seed (1978) earth
dams curve provides a reasonable representation of the profile of average acceleration versus
depth in solid waste landfills over 50 ft (15 m) thick.  Figure 4.7 (Kavazanjian and Matasovi,
1995)  compares  nine different solid  waste  landfill  non-linear seismic  response  analyses,
encompassing waste fills from  50  to 300 ft (15 to 90 m) thick to the representative profile
developed by Makdisi and Seed.  Based upon a maximum average acceleration ratio at the base
of the landfill of 0.45,  as indicated by Figure 4.7, and upon a maximum amplification factor of
2.0 from Figure 4.5 for a peak bedrock acceleration of 0.1 g  or greater, Kavazanjian and
Matasovi (1995) concluded that the free field peak ground acceleration calculated for the landfill

                                           46

-------
site in Step 2 provides a conservative estimate of the peak average acceleration at the base of the
landfill for use in analyses of base liner stability and seismic deformation potential.

Bray, et al. (1995) provide a means of quantifying the limitations of the Kavazanjian and Matasovi
(1995) caveat  of fill thickness and of more precisely evaluating the average acceleration of the
waste mass as  a function of the peak ground acceleration. Figure 4,8, from Bray, et al. (1995),
presents a normalized plot of the peak acceleration at the crest of the landfill divided by the peak
average acceleration of the waste mass versus the ratio  of the fundamental period of the waste
mass to the fundamental period of the design earthquake from results of a large number of landfill
response analyses with peak ground accelerations up to 0.35 g.  The fundamental period of the
waste fill is the reciprocal of the fundamental frequency  evaluated using Equation 4.1.

Based upon the mean plus two standard deviation curve, Figure 4.8 indicates that the peak average
acceleration of the waste mass is equal to or less than the peak ground  acceleration when the
fundamental period of the waste mass is at least 1.2 times greater than the  fundamental period of
the design earthquake.  Based upon typical shear wave velocities for solid waste  and typical
predominant periods for earthquake motions, Figure 4.8 suggest that the peak average acceleration
of the  waste mass can be assumed to  be less than the free field  peak ground acceleration for
nearby earthquakes for waste thicknesses greater than 50 ft (15 m) and for  distant earthquakes for
waste thicknesses greater than 100 ft (30 m).  For larger waste thicknesses and/or high frequency
(short period) earthquakes, Figure 4.8 indicates the peak average acceleration of the  waste mass
can be as little as 20 to 40 percent of the free field peak  ground acceleration.

4.1.2     One-Dimensional Site Response Analysis

For Special Study soil sites, for major projects,  and when an analysis more accurate than the
simplified one  presented in the previous  section is desired, a one-dimensional seismic site response
analysis can be performed.  The site response analysis can be performed for the foundation soils
only, for the waste mass only, or for the coupled response of the foundation soil and waste mass,
depending on the needs and desires of the design engineer.

The computer  program, SHAKE,  originally developed by Seed and his  co-workers (Schnabel et
al., 1972) and recently updated by Idriss and Sun (1992) is perhaps the most commonly used
computer program for one-dimensional equivalent linear seismic  site response analysis,  Basic
input to SHAKE includes the soil profile, soil  properties, and  the input time history.  Soil
properties include the maximum  (small strain) shear wave velocity or shear modulus and unit
weight for each soil layer plus curves relating the reduction in modulus and damping ratio to shear
strain for each soil type.
                                           47

-------
Modulus reduction and damping curves can be specified by the user based upon laboratory testing
or upon recommendations from previous investigations.  Laboratory data on soil modulus and
damping at small strains (shear strains less than 10"*%) can be obtained from resonant column
tests. At larger strains, cyclic simple shear, cyclic triaxial, and cyclic torsional shear tests can
be used.  American Society for  Testing and Materials (ASTM) standards exists for resonant
column testing (ASTM D-3999) and cyclic triaxial testing (ASTM D-4015).  Small strain modulus
can also be determined from field measurements of shear wave velocity.  Shear wave velocity can
be measured in the field using geophysical methods such as down-hole and cross-hole velocity
testing,  seismic refraction, and spectral analyses of  surface waves.  Field measurements are
generally considered more reliable than laboratory measurements of shear wave velocity or small
strain modulus. Field techniques for measurement of the dynamic modulus at large strains and
of the damping ratio are not currently available.  Shear wave velocity is related to small strain
shear modulus, G^, by the equation:
                                                                                 (4.2)
As  an alternative to laboratory or field measurement of soil properties, dynamic moduli and
damping for soils may be estimated as a function of soil type based upon recommendations for
typical values from previous investigations. One set of practical recommendations for estimating
modulus and damping of typical soils are summarized in Figure 4.9 and Table 4.1   Figure 4.9
presents typical modulus reduction and damping curves as a function of the plasticity index of the
soil, PI, from Vucetic and Dobry (1991). (Note that the curve for PI = 0 represents sands and
cohensionless soils.) These curves are for all soil types for a broad range of overconsolidation
ratios. Table 4.1 presents coefficients and exponents for evaluating the small strain shear modulus
for different soil types using the Standard Penetration Test blow count, N, and the following
equation from Imai and Tonouchi (1982):


                       GmaK   c(JV)«                                              (4.3)

where N is in blows per foot of penetration and c and a are coefficients from Table 4.1. Equation
4.3 was developed using Japanese data.  Therefore, a blowcount corresponding to hammer
efficiency of 60 percent, Ngo, as used in U.S. practice (described in Section 5.3), needs to be
converted to Japanese standards by multiplying N^ by 0.833  before input to Equation 4.3:

                      N  0.833(Ar60)                                             (4.4)
                                           48

-------
Other correlations for these properties are available and may be used.  Due to the uncertainty
involved in using these types of empirical correlations, considerable judgement is required in
interpreting results of analyses that employ them and sensitivity studies on the values of these
parameters are recommended.

Unit weight,  shear modulus, and  damping values  are  also required for MSW if the MSW is
included in the response analysis.  Measurement of the dynamic properties of MSW in the
laboratory is considered neither practical nor reliable due to the difficulties inherent to sampling
and testing MSW.  Back calculation of MSW properties from field observations  is generally
considered to  be the most reliable means of evaluating these properties at this time (Kavazanjian
et al., 1995).  Evaluation of the density of MSW from reported field measurements is discussed
in Section 6.1.

At present, the shear wave velocity of MSW has been  measured in-situ at a limited number of
locations. Cross-hole shear  wave velocity measurements at the Puente Hills MSW Landfill in
southern California reported by Earth Technology  (1988) varied from 213 m/s (700 ft/s) at the
ground surface to 278 m/s (920 ft/s) at a depth of approximately 14 meters (45 feet).  Sharma et
al. (1990) report an average shear wave velocity of 198 m/s (455 ft/s) for MSW at depths between
0 and 15 meters (0 and 50 feet) at a landfill in Richmond, California from downhole shear wave
velocity measurements. Singh and Murphy (1990) cite an investigation by others at the Redwood
Landfill in the San Francisco Bay area where an average shear wave velocity of 91 m/s (300 ft/s)
was reported for the refuse.  Shear  wave velocities backfigured using assumed values of Poisson's
ratio and waste density from Young's Modulus  values developed by Carey et al. (1993) from
cross-hole shear wave velocity measurements vary from 185 m/s (610 ft/s) near the surface to 478
m/s (1,580 ft/s) at a depth of 30 meters (100  feet) at the Brookhaven landfill on Long Island in
New York (actual shear wave velocity measurements were not reported). Measurements at 8
MSW landfills in southern California made using Spectral Analysis of Surface Waves (SASW)
were reported by Kavazanjian et al. (1994). Shear wave velocities varied from 78 to 170 m/s (260
to 560 ft/s) near the ground surface, and from 150 to 300 m/s (500 to 990 ft/s) at a depth of 20
meters (66 feet).  Shear wave velocity was reported to increase steadily with depth in the waste
at all 8 sites,

Hushmand Associates (1994) report that seismic refraction surveys performed by others at the Oil
landfill yielded a shear wave velocity of between 200 to 240 meters per second (660 to 800 feet
per second).  Hushmand Associates (1994) also report that measurements of micro tremors from
small earthquakes and of ambient vibrations  at a strong motion instrumentation station located
over an estimated 75 meters (250 feet) of waste at the Oil site indicate a predominant period of
between 0.8 and 1.2 seconds (corresponding to a predominant frequency of between 1.25 and 0.83

                                          49

-------
cycles per second) for the waste mass. Using Equation 4.1, this corresponds to an average shear
wave velocity of between 240 and 360 meters per second (800 and 1200 feet per second) for the
assumed 75 meter (250 foot) waste column.  While  the Oil  landfill is  composed  of mixed
industrial and municipal waste, the portion of the landfill at which the strong motion station is
located is believed to be composed primarily of MSW (personnel communication, Professor R.B.
Seed, U.C. Berkeley, to Dr. Edward Kavazanjian, Jr., GeoSyntec Consultants).

Based upon the data cited above, Kavazanjian et al.  (1995) developed a "representative" shear
wave velocity profile for  MSW landfills.   Figure 4.10  (Kavazanjian et al.,  1995) presents a
composite plot of the available MSW shear wave velocity data along with the shear wave velocity
profile developed by these investigators for use in the absence of site-specific data.  In developing
this shear wave velocity profile, the seismic refraction data from the Oil  site was assumed to
represent the average velocity over the top 30 meters (100 feet) of waste and the data derived from
cross-hole measurements was considered unreliable due to the potential for  "short-circuiting" of
the wave travel path by layers of daily and intermediate cover soils.

Modulus reduction and damping curves for MSW have never been measured in the laboratory.
Prior to the 17 January 1994 Northridge earthquake, no data was available to back-calculate MSW
modulus and damping from the observed seismic response of landfills.  In the absence of special
measurements, most investigators based the modulus reduction and damping  curves for MSW
upon those of clay and peat soils (Earth Technology, 1988; Singh and Murphy, 1990; Sharma and
Goyal,  1991; and Repetto et al.,  1993).   Figure 4.11 presents recommendations from Earth
Technology (1988)  for modulus reduction and damping curves for MSW.  These curves are
reported to be based upon modulus reduction curves for peat and damping curves for clay. Figure
4.12 presents recommendations for modulus reduction and damping in MSW from Singh and
Murphy (1990). The "recommended" curves are described by Singh and Murphy as the "average"
of typical modulus reduction and damping curves for peat and clay that are used in engineering
practice.

The  strong motion recordings captured at the Oil landfill in the M 6.7 Northridge earthquake
represent the first (and currently the only) direct measurement of the seismic response of a solid
waste landfill.  In the Northridge event, the peak ground acceleration at the monitoring station on
the rock outcrop adjacent to the landfill was 0.25 g, while  the peak ground acceleration at the top
of the landfill was 0.24 g (Hushmand Associates , 1994). Time histories of acceleration, velocity,
and displacement recorded at the top of the landfill for one horizontal component of motion were
previously presented in Figure 3.4.

Kavazanjian and Matasovi (1995) developed the MSW modulus reduction and damping curves
shown in Figure 4.13 from the observed response of the Oil landfill in the Northridge event.

                                           50

-------
Using the representative shear wave velocity profile shown in Figure 4,8 and the "typical" unit
weight profile developed by Kavazanjian et al. (1995), presented subsequently in Figure 6.3 of
this document, Kavazanjian and Matasovi (1995) back-calculated parameters describing the cyclic
behavior of MSW for a non-linear site response model from the observed landfill response. Then,
these investigators used the non-linear model to predict the response of MSW to uniform cyclic
loading and compute the  modulus reduction and damping curves for MSW shown in Figure 4.13.

The modulus  reduction  and damping curves for MSW  shown in Figure 4.13 were used  by
Kavazanjian and Matasovi (1995) in the program SHAKE to perform an equivalent linear response
analysis of the response  of the Oil landfill in the Northridge earthquake.  By trial and error, a
representative shear strain factor of 0.8 (representative shear strain equal to 0.8 times the
maximum cyclic shear  strain) was  found to give the best agreement between observed and
predicted behavior.

Figure 4.14 compares the Oil landfill response observed in the Northridge earthquake to the
response predicted by Kavazanjian and Matasovi (1995) using SHAKE, the modulus reduction and
damping curves in Figure 4.12, and a representative shear strain factor of 0.8.   This figure also
shows  landfill response predicted by Kavazanjian et al.  (1995) using SHAKE and various
combinations of modulus reduction and damping curves for  peat and clay along with the best fit
representative shear strain factor.  Based upon this comparison,  Kavazanjian et  al.  (1995)
suggested that the modulus reduction and damping curves shown  in Figure  4.13 be used in
equivalent linear  seismic response analysis of MSW landfills until additional information on the
cyclic response of MSW becomes available.

The use of a representative shear strain factor of 0.8 with the Kavazanjian and Matasovi modulus
reduction curve  indicates that the Oil landfill behaved relatively elastically in the Northridge
earthquake.  Furthermore, the maximum cyclic shear strain induced in the Oil landfill during the
Northridge event was on the order  of 2 x 10"2 percent.  Therefore, the shape of the modulus
reduction and damping curves shown in Figure 4.13 must be considered speculative for values of
shear strain greater than 2 x 102 percent.  Furthermore,  there  is some concern over geological
conditions at the location where the base motions of the Oil landfill were recorded.  For these
reasons, caution  is warranted in using the modulus reduction and damping curves presented in
figure 4.13.

4.1.3     Two-  and Three-Dimensional Site Response Analysis

Computer programs are available for equivalent linear  and truly non-linear two-  and three-
dimensional seismic site response analyses. However, such programs are not commonly used in
landfill engineering practice.  The programs for two- and three-dimensional site response analyses

                                          51

-------
are not particularly "user-friendly".  Furthermore, experience with two-dimensional site response
analyses of earth dams has shown that one-dimensional site response analyses of vertical columns
within the embankment typically yield accelerations and stresses within ten percent of the results
of the more sophisticated two- and three-dimensional analyses (Vrymoed and Calzascia, 1978).
Two- and three-dimensional effects may logically be expected to  be even less significant with
respect to the seismic response of landfills compared to earth dams, as landfills tend to be massive
structures with broad decks.  Considering the level of uncertainty associated with  material
properties of solid waste, two- and tree-dimensional seismic response analyses do not appear to
be warranted for most municipal solid waste landfill projects at this time.

Once the soil profile and material properties have been specified, the only remaining input is the
input earthquake motion.   Selection of representative time histories for the  input motion is
discussed in Section 4.2.

4.2       Selection of Earthquake Time History

Earthquake time histories may be required for input to SHAKE seismic response analyses or, if
a simplified seismic response analysis is employed, for input to the seismic deformation analyses
described in Section 6. Time histories can be developed either by selecting a representative time
history  from the catalog of acceleration time  histories recorded in previous earthquakes or by
synthesizing an artificial accelerogram.  Time histories should be developed for each significant
souce impacting the  site.

Selection of a representative time history from the catalog of available strong motion records and
scaling  it to the appropriate peak acceleration is, in general, a preferable approach to use of a
synthetic time history. However, due to limitations in the catalog  of available records, it is not
always  possible to find a  representative  time history from the catalog of available records,
particularly for the eastern and central United States.

In selecting a representative time history from the catalog of available records, an attempt should
be made to match as many of the relevant characteristics of the design earthquake as possible.
Important characteristics that should be considered in selecting a time history from the catalog
include:

                 earthquake magnitude;
                 source mechanism (e.g., strike-slip, normal, or reverse faulting);
                 focal depth;
                 site to source distance;
                 site geology; and

                                            52

-------
                 peak ground acceleration.

These factors are ranked in a general order of decreasing importance.  However,  the relative
importance may vary from case to case.  For instance, if a bedrock record is chosen for use in a
SHAKE analysis of the influence of local soil conditions, site geology will not be  particularly
important in selection of the input bedrock tune history.  However, if a soil site record scale to
a peak ground acceleration that already  includes consideration of the potential for amplification
of motions by local soil conditions is to be used in the response analysis, site geology can be a
critical factor in selection of an appropriate time history.

Scaling  of the peak acceleration of a record by a factor of more than two is not recommended,
as the frequency characteristics of ground motions can be directly and indirectly related to the
amplitude of the motion. Leeds (1992) and Naeim and Anderson  (1993) present summaries of
available strong motion records and their characteristics.

Due to uncertainties in the selection of a representative earthquake time history, response analyses
should never  be performed using only a single time history. The use of a suite  time histories is
recommended for purposes of evaluating seismic site response.  Engineers commonly use three
and sometimes as many as five time histories to represent each significant seismic source in a site
response analysis.   For earthquakes  in the western United States,  it should be possible to find
three to five representative time histories that satisfy the above criteria.  However, at the present
time, there are only two bedrock strong motion records available from earthquakes of magnitude
M 5.0 or greater in the central and eastern North America:

                 the Les Eboutements record with a peak horizontal acceleration of  0.23 g from
                 the 1988 Saguenay, Quebec earthquake of magnitude M 6.0; and

                 the Loggie Lodge record with a peak horizontal acceleration of 0.4 g from the
                 1981 Mirimichi, New Brunswick earthquake of magnitude M 5.0.

Therefore, for analysis of sites east of the Rocky Mountains, at least one record from  a western
United States site,  an international recording site, or a synthetic accelerogram is  required to
compile a suite of three records  for analysis. For  the new Madrid seismic  zone, where neither
the Mirimichi nor Saguenay record is of appropriate magnitude, all three records must be from
either the western United States, an international site, or synthetically generated.

One of the primary  differences anticipated between earthquakes in the eastern and central United
States and those in the western United States is frequency content (Nuttli, 1981; Atkinson,  1987).
There may also be a difference in duration due to the different rates of acceleration  attenuation.

                                           53

-------
For liquefaction analyses which depend only on peak acceleration, use of a western United States
earthquake record of appropriate magnitude and intensity for analysis of a site in the eastern or
central United States should be acceptable.   However, for analysis of seismic deformation
potential at an eastern or central United States site, the appropriateness of using a western United
States earthquake record is uncertain. The greater energy at lower frequencies in typical western
U.S. records could result in a conservative estimate of deformation potential at an eastern or
central United States site. On the other hand, the potential for a longer duration on the east coast
compared to the west coast for an earthquake of the same magnitude and distance could have the
opposite effect.

Due to the difference in the anticipated depths of the causative faults, when using a western United
States record to analyze a site in the eastern United States precedence should be given to matching
hypocentral distance over peak acceleration.  Hypocentral distance is the distance from the site
to the center of energy release for the earthquake.   Hypocentral distance includes the effect of the
depth of the earthquake in the distance measure.

Computer programs are available to generate a synthetic seismic accelerogram to meet peak
acceleration, duration, and frequency content requirements (Gasparin and Vanmarcke, 1976; Ruiz
and Penzien, 1969; Silva and Lee, 1987).  Synthetic earthquake accelerograms for many regions
of the country  are currently being  compiled by Dr. Klaus  Jacob  at the Lamont-Doherty
Observatory of Columbia University under the auspices of the National Center for Earthquake
Engineering Research (see Table 2.2).  However, at the time of preparation of this guidance
document this compilation was not yet available.  The generation of synthetic acceleration time-
histories is not generally within the technical expertise of civil engineering firms and should not
be undertaken without expert consultation.  For this reason, generation of synthetic earthquake
acceleration time-histories is beyond the scope of this manual.  However, appropriate synthetic
accelerograms may be available to the engineer from previous studies and may be used if they are
shown to be appropriate for the site.

Comparison of acceleration of response  spectra  from candidate accelerograms  to acceleration
response  spectra deemed representative of the design event provides useful means of determining
whether the selected  accelerograms are  indeed  representative.  Each accelerogram from the
selected suite of accelerograms should fall primarily within the two-sigma boundaries  of the
statistically-derived response spectra and the suite of accelerogram should average out to close to
the near spectra.  In this manner, a representative suite of tune histories can be developed.
                                           54

-------
4.3       References

Anderson, D.G., Hushmand, B., and Martin, G.R. (1992), "Seismic Response of Landfill Slopes,
" Proc. Stability and Performance of Slopes and Embankments - II, Vol. 2, ASCE Geotechnical
Special Publication No. 31, Berkeley, California, pp. 973-989.

Atkinson, G.M. (1987), "Implications  of Eastern Ground Motion Characteristics  for Seismic
Hazard Assessment in Eastern North  America," Proc. Symposium on Seismic Hazards, Ground
Motions, Soil-Liquefaction and Engineering Practice in Eastern North America, Tuxedo, New
York, NCEER Technical Report No. NCEER-87-0025.

Borcherdt, R.D. (1994) "New Developments in Estimating Site Effects on Ground Motion," Proc.
Seminar on New Developments in Earthquake  Ground Motion Estimation and Implications for
Engineering Design Practice, Applied Technology Council, ATC35-1, Redwood City, California,
pp. 2-1 - 2-12.

Bray, J.D., Augello, A.J., Leonards, G.A., Repetto, P.C., and Byrne, R.J. (1995), "Seismic
Stability Procedures for Solid-Waste Landfills," Journal of the Geotechnical Division, ASCE,
Vol. 121, No. 2, pp. 139-151.

Carey, P.J., Koragappa, N. and Gurda, J.J. (1993) "A Case Study of the Brookhaven Landfill
-Long Island, New York," Proc.  Waste Tech '93, Marina del Rey, California, 15 p.

Earth Technology  (1988), "In-Place Stability  of Landfill Slopes,  Puente Hills Landfill,  Los
Angeles, California," Report No. 88-614-1, The Earth Technology Corporation, Long Beach,
California.

Gasparin, D.A. and Vanmarcke, E.H.  (1976) "SIMQKE - A Program for Artificial Motion
Generation,"  Department of  Civil Engineering,  Massachusetts  Institute  of Technology,
Massachusetts.

GeoSyntec (1994),  "Non-Linear Seismic Response Analysis of Solid Waste Landfills," Internal
Research Report, GeoSyntec Consultants, Huntington Beach, California.

Harder, L.S., Jr. (1991), "Performance of Earth Dams During the Loma Prieta Earthquake,"
Proceedings of the Second  International  Conference on  Recent Advances  in Geotechnical
Earthquake Engineering and Soil Dynamics, University of Missouri, Rolla, 11-15 March.
                                          55

-------
Hushmand Associates (1994), "Landfill Response to Seismic Events," Report prepared for the
USEPA Region IX, Hushmand Associates, Laguna Niguel, California.

Idriss, I.M. (1990), "Response of Soft Soil Sites During Earthquakes," Proc. Symposium to Honor
Professor H.B. Seed, Berkeley, California.

Idriss,  I.M.  and Sun, J.I. (1992) "User's Manual for SHAKE91,"  Center for Geotechnical
Modeling, Department of Civil and Environmental Engineering, University of California, Davis,
California, 13 p. (plus Appendices),

Imai, T. and Tonouchi, K. (1982), "Correlation of N Value with S-Wave Velocity," Proc. 2nd
European Symposium on Penetration Testing, Amsterdam, The Netherlands, pp. 67-72.

Kavazanjian, E., Jr., and Matasovi, N.  (1995), "Seismic Analysis of Solid Waste Landfills,"
Proceedings of the Geoenvironment 2000 Specialty Conference, ASCE, Vol. 2, pp. 1066-1080,
New Orleans, Louisiana,  24-26 February .

Kavazanjian, E., Jr., Snow, M.S., Matasovi,  N.,  Poran, C., and Satoh, T. (1994a), "Non-
Intrusive Rayleigh Wave Investigations at Solid Waste Landfills."   Proc.  1st International
Congress on Environmental Geotechnics, Edmonton, Alberta, pp. 707-712.

Kavazanjian, E., Jr., Matasovi, N., Bonaparte, R., and Schmertmann, G.R. (1995), "Evaluation
of MSW Properties for  Seismic Analysis," Proceedings of the Geoenvironment 2000 Specialty
Conference, ASCE, Vol. 2, pp. 1126-1141, New Orleans, Louisiana,  24-26 February .

Lee, M.K.W.,  and  Firm, W.D.L.  (1978), "DESRA-2, Dynamic Effective Stress Response
Analysis of Soil Deposits with Energy  Transmitting  Boundary Including Assessment  of
Liquefaction Potential," Soil Mechanics Series No. 36, Department of Civil Engineering,
University of British Columbia, Vancouver, Canada, 60 p.

Leeds, D.L. (1992)  "State-of-the-Art for Assessing Earthquake Hazards in the United States:
Report 28, Recommended Accelerograms for Earthquake Ground Motions," Misc. Paper S-73-1,
Geotechnical Laboratory, U.S Army Waterways  Experiment Station,  Vicksburg, Mississippi, 171
p. (plus Appendices).

Matasovi, N. and Vucetic, M. (1993), "Cyclic Characterization of Liquefiable Sands," Journal
of Geotechnical Engineering, ASCE, Vol. 119, No. 11, pp. 1805-1822.
                                         56

-------
Makdisi, F.I. and  Seed, H.B.  (1978),  "Simplified  Procedure for Estimating Dam  and
Embankment Earthquake-Induced Deformations," Journal of the Geotechnical Division, ASCE
Vol. 104, No. 4 , pp. 849-867.

Naeim, F. and Anderson, J.C. (1993), "Classification and Evaluation of Earthquake Records for
Design," Report No.  CE 93-08, Department of Civil Engineering,  University of Southern
California, Los Angeles, 288 p.

Nuttli, O.W. (1981), "Similarities and Differences Between Western and Eastern United States
Earthquakes, and their Consequences for Earthquake Engineering,"  Earthquakes and Earthquake
Engineering: The Eastern United States, Vol. 1, Assessing the Hazard - Evaluating the Risk,  J.E.
Beavers, Ed., Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan, pp. 25-51.

Repetto, P.C., Bray, J.D., Byrne, R.J.  and Augello,  A.J., (1993),  "Applicability of Wave
Propagation Methods to the Seismic Analysis of Landfills," Proc.  Waste Tech '93, Marina Del
Rey, California, pp. 1.50-1.74.

Ruiz, J.  and Penzien, J., (1969) "PSEQGN - Artificial Generation of Earthquake Accelerograms,"
Report No.  EERC  69-3,  Earthquake Engineering Research Center, University of California,
Berkeley, California.

Schnabel,  P.B.,  Lysmer,  J. and Seed,  H.B.  (1972)  "SHAKE:  A Computer Program for
Earthquake Response  Analysis of Horizontally Layered  Sites."   Report No.  EERC 72-12,
Earthquake Engineering Research Center, University of California, Berkeley, California.

Seed, H.B. (1975) "Earthquake Effects on Soil-Foundation Systems," In Foundation Engineering
Handbook,  H.F. Winterkorn and H.Y. Fang Eds., Van Nostrand Reinhold, New  York,
pp. 700-732.

Seed, H.B. and  Idriss, I.M.,  (1982), "Ground  Motions and Soil  Liquefaction During
Earthquakes,"  Monograph No.  5,  Earthquake Engineering  Research Institute, Berkeley,
California, 134 p.

Sharma, H.D., Dukes, M.T., and Olsen, D.M., (1990), "Field Measurements of Dynamic Moduli
and Poisson's Ratio of Refuse and Underlying Soils at a Landfill Site," In Geotechnics of Waste
Fill - Theory and Practice, ASTM STP 1070, pp. 57-70.
                                          57

-------
Sharma, H.D. and Goyal, H.K. (1991), "Performance of a Hazardous Waste and Sanitary Landfill
Subjected to Loma Prieta Earthquake," Proc, 2nd International Conference on Recent Advances
in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis, Missouri, pp. 1717-1725.

Silva, W.D. and Lee, K. (1987) "State-of-the-Art for Assessing Earthquake Hazards in the United
States: Report 24, WES RASCAL Code for Synthesizing Earthquake Ground Motions," Misc.
Paper S-73-1, Geotechnical Laboratory, U.S. Army Waterways Experiment Station, Vicksburg,
Mississippi.

Singh, S.  and Murphy,  B.J., (1990),  "Evaluation of the Stability of Sanitary Landfills," In
Geotechnics of Waste Fills - Theory and Practice, ASTM STP 1070, pp. 240-258.

Singh, S.  and Sun, J.I.  (1995),  "Seismic Evaluation of Municipal  Solid  Waste Landfills,"
Proceedings of the Geoenvironment 2000 Specialty Conference, ASCE, Vol.  2, pp. 1081-1096,
New Orleans, Louisiana, 24-26 February,

Vrymoed, J.L. and Calzascia, E.R.  (1978), "Simplified Determination of Dynamic Stresses in
Earth Dams," Proc. Earthquake Engineering and Soil Dynamics, ASCE, Pasadena, California,
pp. 991-1006.

Vucetic, M. and Dobry, R. (1991) "Effect of Soil Plasticity on Cyclic Response." Journal of the
Geotechnical Engineering, ASCE, Vol. 117, No. 1, 89-107.

Wood, H.O.  (1908),  "Distribution of Apparent Intensity in San Francisco," in The California
Earthquake of April  18,  1906,"  Report of the State Earthquake  Investigation Commission,
Carnegie Institution of Washington,  D.C.,  pp. 220-245.
                                         58

-------
   TABLE 4.1:  PARAMETERS FOR THE EMPIRICAL RELATIONSHIP
                             TO ESTIMATE G^
                       (After Imai and Tonouchi, 1982)
SOIL TYPE
Peat
Clay
Sand
Gravel
c
(kg/cm2)
53.7
176.0
125.0
82.5
a
(-)
1.08
0.607
0.611
0.767
Notes: (1) G^, = Small strain shear modulus; G,,,,,, = c(W)'; G^ in kg/cm2.
      (2) N = (uncorrected) SPT blowcount according to Japanese standards. Multiply NM from
         U.S. practice by 0.833 to estimate a comparable blow count.
      (3) Correlation applies only for soils of alluvial origin.  For soils of other origin, the original
         reference should be consulted.
                                           59

-------
Os
O
                       0  0.5  1.0  1.5 20
                          Period-Seconds
0  O5  TO  152.0
   Pertod-Seeendt
0  05  10  1.5 2OO  0.5to15 200  oi   10  WiO  0  05  10  15 2.0
   Pefied-Swondi     Perted-Secondi     Aried-Sieondi        Med-Steendt
                                                                                                                                   400
                           Figure 4.1    Soil  Conditions and Characteristics of Recorded  Ground  Motions, San
                                          Francisco M 5.7 Earthquake of 22 March 195? (Seed, 1975).

-------
          ACCELERATION
            RESPONSE
             SPECTRA
PEAK GROUND ACCELERATION
                                                           TIME IN
                                                          SECONDS
                                                   SYSTEM
                                                  RESPONSE
                                                    ATT,
          PERIOD IN
          SECONDS
Figure 4.2   Development of Acceleration Response  Spectrum for Damped  Single
            Degree of Freedom System.
                                       61

-------
   400
   too r
          SO
               to
     FREQUENCY. HERTZ



*  4     2     1 A A A
                                                    .1  J09 J04
O
O
Ul
<0

to

s
O
O

gi
        .04 .06  .1
       ^4  .6    1     2    4   6   10    20



      PERIOD. SECONDS
            SRN FERNflNOO EflRTHOURKE   F£6 9. 1911 - 0800 PSI



         IlKCTB  7I.O08.0  S3W CKIOH «.VO, I3J fUMB. 1.05 WCCLCS. Ol.  O5W HOOK



                 OBWJNC VBLUCS (Kit 0. 1. S. 10 B«0 M POCtNl OT OttTiCAL
  Figure 4.3    Tripartite Representation of Response Spectra.
                                     62

-------
§
1
8
o
E
1
x
                   o.e
                   0.5
                   O.4
                                                          Soft to M«dtum
                                                          Stiff Ct«y «f*d Sarul
             O.t
                                    O.2     O.3     O.4     O.S    O.6
                                  Maximum Acceleration in Rock (g)
                               0.1
                        Q.2      0.3      Q.4

                      Acceleration at Rock Sites - g
O.S
o.e
Figure 4.4    Relationship Between Maximum Acceleration on Rock and Other Local Site
              Conditions: (a) Seed and Idriss (1982); (b) Idriss (1990).
                                            63.

-------
   L8
too
IO/M
   0.6 -
   0.0
          A  OH Landfill - Recorded
          •*™ Landfills -  NOIL—linear Analyses
          D  Soft  Ground Sites  — Recorded
          	Idrisg (1990) -  Recommended
                               1979 Montenegro Eq.
                       —-1989 Loma Prieta Eq.
                 1985  Mexico City Eq.
         TIIIJ1III|IIIIjIIIIJ!IIIT~~Tl7
     0.0     0.1     0.2     0.3     0.4     0.5     0.6

       PEAK OUTCROP ACCELERATION  -  aomax (g)
Figure 4.5   Observed Variations of Peak Horizontal Accelerations on Soft Soil and
          MSW Sites in Comparison to Rock Sites (Kavazanjian and Matasovid,
          1994).
                              64

-------
                                               Harder (1991)
                                               Upper Bound for Observed
                                               Motions at Earth Dams
                                                        100 ft Height
                                 Seed • Idriss (1982) for
                                 Deep Coheaionless Soil
                          0.2    0.3   0.4   0.5   0.6   0.7

                          Max. Acci. at Base, g
Figure 4.6     Approximate Relationship Between Maximum Accelerations at the Base

              and Crest for Various Ground Conditions (Singh and Sun, 1995)
                                        65

-------
         Average Acceleration. Ratio, a«^ (z)/a«« (z=0)

            0.0     0.2     0.4     0.6
         0.0
         0.2
Figure 4.7   Variation of Maximum Average Acceleration Ratio with Depth of Sliding
           Mass (Kavazanjian and Matasovie, 1995)
                                   66

-------
              0    mean +2 std. dev.
                                          mean -2 std. dev
                     246
                     Normalized Fundamental Period (Tw/Tp-«q)
Figure 4.8     Normalized Maximum Horizontal Equivalent Acceleration versus the
             Normalized Predominant Period of the Waste Fill (Bray, et al., 1995)
                                    67

-------
  1.0
  0.8
  0,6
8
  0.4
  0.2
  0.0
              |OCR = 1-151
                                    PN200
                                   1001
    0.0001
0.001        0.01          0.1            1

    CYCLIC  SHEAR  STRAIN,7C(%)
10
    25-
  o15

  i—
  <
  OL


  I10
  _

  21
  <
  O
                  locR=i-el
     0.0001
 0.001         0.01          0.1           1

     CYCLIC SHEAR STRAIN , Tc (%)
  10
Figure 4.9   Modulus Reduction and Damping Curves for Soils of Different Plasticity

           Index (PI)  (Vucetic and Dobry, 1991).
                                    68

-------
a
H
Pu,
     30-
60
     90-
   120
                                               This Study
                                         >
          Kavazanjian et al. (1994)  (8 Bites) \
          after Carey et al. (1993)           v
          Sharma et al. (1990)
          Woodward-Clyde Consultants (1987)
          after Hushmand Associates (1994)
          • after Earth Technology (1988) (crosshole)
          after Earth Technology (1988) (downhole)
      I  i I  I  | I  I  I  i j  i  i i  i  \ i  i  i
                                             1 I  t  I I  I  I  1
        0      100     200    300     400    500
                         WAVE  VELOCITY  (m/s)
                                                     600
     Figure 4.10  Shear Wave Velocity of MSW (Kavazanjian et al., 1995)
                            69

-------
                                                            PEAT (SEED AND IDRISS, 19701
                                       MODIFIED FOR THIS PROJECT
 10-*
10-3
10-2             tO-1

 SHEARING STRAIN <%l
100
                                CLAY 
-------
                  PRA(1987)
                  Upper/Lower
                 : Bound for SHAKE
                                               Landfill Refuse Material
                                            Assumed to Correspond to the
                                           Statistical Average of Peat & Clay
                   EMCON (1988)
                   Richmond Fill
               \EMCON (1988)
               Redwood fill
                 I  Note: Data Points are eaima»adev«rao«
                 lvalues torn shear wave data far Su-143 KNAn2
      CD
        — , - i - 1

         0.001         0.01         0.1
                CYCLIC SHEAR STRAIN IN %
                                                             1.0
T
 10
         25-1
         20-
      O  15-
      jr
      O 10-

      CL


      1  8
    Landfiti Refuse Material
 Assumed to Correspond to the
Statistical Average of Peat & Clay
                         I            1             1            T

                       0.001         0.01         0.1          1.0

                                 CYCLIC SHEAR STRAIN IN %
                                                              10
Figure 4.12   Modulus Reduction and Damping Curves for MSW (Singh and Murphy,
             1990).
                                         71

-------
                                -f-H- i tt	H-i-H-H+H	t-H-H-HH
                                     ft       «•         41
        o.oooi
                           mill
0.001     0.01     0.1       1
CYCLIC  SHEAR  STRAIN (55)
       0.0001   0.001    0.01     0.1       1
                 CYCLIC  SHEAR STRAIN  (SB)
                                      10
Figure 4.13  Modulus Reduction and Damping Curves for MSW (Kavazanjian and
          Matasovic, 1995).

                               72

-------
       1.2
                              RECORDED (On-Nortliridge; Longitud.)
                              MSW (Kavazanjian & Matasovic, 1995)
                              CIAY (PI=15)lVucetic  Ic Dobry.  1991)
                              CIAY (Seed & Idriss, 1970b)
                              PEAT (Seed & Idriss. 1970a)
                                        Damping =  5%
           0.0   0
1.0    1.5   2.0  2.5
   PERIOD  (sec)
Figure 4.14  Comparison of Oil  Landfill Response to Results of Equivalent Linear
           Analysis (Kavazanjian et al., 1995).

-------
                                      SECTION 5
                           258.14 SEISMIC IMPACT ZONES:
                             LIQUEFACTION ANALYSIS

During strong earthquake shaking, loose, saturated cohesionless soil deposits may experience a
sudden loss  of strength and stiffness, sometimes resulting in large, permanent displacements of
the ground.  This phenomenon is called soil liquefaction.  Liquefaction beneath and in the vicinity
of a municipal solid waste landfill facility (MSWLF) can have severe consequences with respect
to the integrity of the landfill containment system.  Localized bearing capacity failures, lateral
spreading, and excessive settlements resulting from liquefaction may damage landfill liner and
cover systems.  Liquefaction-associated lateral spreading and flow failures can also affect the
global stability of the landfill.  Therefore, a liquefaction potential assessment is a key element in
the seismic design of landfills.

This Section outlines the current  state-of-the practice for evaluation of the potential for soil
liquefaction and the consequences of soil liquefaction (should it occur) as it applies to the seismic
design of a MSWLF.  Initial screening criteria to determine whether or not a liquefaction analysis
is needed are presented in Section  5.1.  The simplified procedure for liquefaction potential
assessment commonly used in engineering practice is presented in Section 5.2.  Methods for
performing a liquefaction impact assessment are presented  in Section 5.3.  Methods for mitigation
of liquefaction potential and the consequences  of liquefaction are discussed  in Section 5.4.
Advanced methods for liquefaction potential assessments, including one- and two-dimensional
fully-coupled effective stress site response analyses, are also discussed in Section 5.4.

5.1    Initial Screening

The first step in any liquefaction evaluation is to assess whether the potential for liquefaction of
cohesionless soils exists at a site. A variety of screening techniques exists to distinguish sites that
are clearly safe with respect to liquefaction from those sites that require more detailed study (e.g.,
Dobry et al., 1980). The following five  screening criteria are most commonly used to make this
assessment:

              Geologic age and origin.  Liquefaction potential decreases  with increasing age of
              a soil deposit.  Pre-Holocene age soil deposits generally do not liquefy, though
              liquefaction has occasionally been observed in Pleistocene-age deposits. Table 5,1
              presents the  liquefaction susceptibility of soil  deposits as a function of age and
              origin (Youd and Perkins, 1978).
                                            74

-------
             Fines content and plasticity index.  Liquefaction potential decreases with increasing
             fines content and increasing plasticity index, PI.  Data presented in Figure 5.1
             (Ishihara et al., 1989) show grain size distribution curves of soils known to have
             liquefied in the past. This data serves as a rough guide for liquefaction potential
             assessment of cohesionless soils.  Soils having greater than 15 percent (by weight)
             finer than 0.005 mm, a liquid limit  greater than 35 percent, and an in-situ water
             content  less than 0.9  times the liquid limit generally  do not liquefy (Seed  and
             Idriss, 1982).

             Saturation.  Although partially saturated soils have been reported to liquefy, at
             least 80 to 85 percent saturation is generally deemed to be a necessary condition
             for soil  liquefaction.  In many locations, the water table is  subject to seasonal
             oscillation.  In general, it is prudent that the highest anticipated seasonal water
             table elevation be considered for initial screening.

             Depth below ground surface.  While failures due to liquefaction of end-bearing
             piles resting on sand layers up to 100 ft (30 m) below the ground surface have been
             reported, surface effects from liquefaction is generally not likely  to occur more
             than 50  ft  (15 m) below the ground surface.

             Soil Penetration Resistance. According to the data presented in Seed and Idriss
             (1985),  liquefaction has not been observed  in soil deposits having normalized
             Standard Penetration Test (SPT) blowcount, (N^ larger than 22. Marcuson, et
             al. (1990) suggest a normalized SPT value of 30 as the threshold  value above
             which liquefaction will not occur.  However, Chinese experience, as quoted in
             Seed et  al. (1983), suggests that in extreme conditions liquefaction is possible in
             soils having normalized SPT blowcounts  as high as 40.  Shibata and Teparaska
             (1988), based on a large number of observations,  conclude that no liquefaction is
             possible if normalized Cone Penetration Test (CPT) cone resistance, qc, is larger
             than 157 tsf (15  MPa). This CPT resistance corresponds to normalized blow
             counts between 30 and 60, depending on the grain size of the soil (see Figure 5.2).

If three or more of the above criteria indicate that liquefaction is not likely, the potential for
liquefaction may be considered to be small .  If, however, based on the above initial screening
criteria, the potential for liquefaction of a cohesionless soil layer beneath the site of a planned
landfill (new construction or lateral expansion) cannot be dismissed,  more rigorous analysis of
liquefaction potential is needed.
                                            75

-------
Liquefaction susceptibility maps, derived on the basis of the some (or all) of the above listed
criteria, are available for many major urban areas in seismic zones (e.g., Kavazanjian et al., 1985;
Tinsley et al., 1985; Hadj-Hamou and Elton, 1988; Hwang and Lee, 1992). However, as most
new MSWLF's are sited outside of major urban areas, these maps may not be available for many
landfill sites. Furthermore, most areas have not been mapped in sufficient detail to be useful for
site-specific studies.

There have been several attempts to establish threshold criteria for values of seismic shaking that
can induce  liquefaction (e.g.,  minimum  earthquake magnitude, minimum peak horizontal
acceleration, maximum distance from causative fault).  Most of these criteria have eventually been
shown to be misleading,  since even  low intensity bedrock  ground  motions  from distant
earthquakes can be amplified by local soils to intensity levels strong enough to induce liquefaction,
as observations  of liquefaction in the 1985 Mexico City and 1989  Loma Prieta earthquakes
demonstrate.

Most soil deposits known to have liquefied are sand deposits.  However, gravel deposits are also
susceptible to liquefaction. Discussion of the liquefaction potential of gravel deposits is beyond
the scope of this document.  The reader is referred to Harder (1988) for a discussion of methods
for evaluation of the liquefaction potential of gravels.

5.2    Liquefaction Potential Assessment

Due to the difficulties in obtaining undisturbed representative  samples from most liqueflable soil
deposits and to the difficulties and limitations of laboratory testing, the use of in-situ test results
to evaluate  liquefaction potential is generally the preferred method for liquefaction potential
assessment among most practicing engineers.   Liquefaction potential assessment procedures
involving both the SPT and the  CPT  are widely used in practice  (e.g., Seed and Idriss, 1982;
Ishihara, 1985; Seed and De Alba, 1986; Shibata and Teparaska, 1988).  For gravelly soils, the
Becker Hammer penetration test is  commonly used to evaluate liquefaction potential (Harder,
1988).

The most common procedure used in practice for the liquefaction  potential assessment of sands
and silts, the Simplified Procedure, was originally developed by Seed and Idriss (1982).  As used
in engineering practice  today, the Simplified Procedure has been progressively revised, extended
and refined (Seed et al., 1983; Seed et al.,  1985; Seed and De Alba,  1986; Liao and Whitman,
1986). The Simplified  Procedure may be used with both CPT and SPT data. Recent summaries
of the various revisions to the Simplified Procedure are provided by Marcuson et al., (1990) and
by Seed and Harder (1990). Based on the recommendations from these two studies, the Simplified
                                           76

-------
Procedure for evaluating liquefaction potential at the site of a MSWLF can be carried out using
the following steps:

Step 1:       From in-situ testing and laboratory index tests, develop a detailed understanding
              of site conditions:  stratigraphy,  layer  geometry, material properties  and their
              variability,  and the areal extent of potential problem zones.  Establish the most
              critical zones to be analyzed and develop simplified sections amenable to analysis.
              The data should include location of the  water table, either SPT blowcount, N, or
              tip resistance of a standard CPT cone, qc, and mean grain size, D50, the unit weight
              of the soil,  and the percentage of fines (percent by weight passing the No. 200
              sieve) for the materials involved in the  liquefaction potential assessment.

Step 2:       Evaluate the total vertical stress, 0, and vertical effective stress, 0',  in the deposit
              at the time of exploration and for design.  Design values should include  the
              overburden stress  due to  the landfill.   Outside of  the  waste footprint,  the
              exploration and design values may be the same if the design ground water level is
              at the same elevation as the ground water  was during sampling, or they  may be
              different due to temporal fluctuations in the water table.

Step 3:       Evaluate the stress reduction factor, rd.  The stress reduction factor is a soil
              flexibility factor defined as the ratio of the peak shear stress for the soil column,
              (maJd.to mat °f a rigid body, (raax)r. There  are several ways to determine  rd. For
              depths less than 40 ft (12 m), the average value from Figure 5.3 (Seed and Idriss,
              1982) can be used.  Alternatively, the following equation proposed by Iwasaki et
              al. (1978) can be used:

                      rd   \    0.015 D                                              (5.1)


              where D is depth in meters.

              If results of a site response  analyses (e.g., a SHAKE analysis) are available, rd  can
              be determined directly from results of such analysis, as:
                   	I mtx)@depth D	                                       ., _,

                   *• o'@depth D   *• max %'@surface
                                            77

-------
              where a^ is the peak ground surface acceleration and g is the acceleration of
              gravity.

              Use of the results of a site response analysis to evaluate rd is considered to be
              generally more reliable than either of the two simplified approaches and is strongly
              recommended for sites that are marginal with respect to liquefaction potential (sites
              where the factor of safety for liquefaction is close to  1.0).

Step 4;       Calculate the critical stress ratio induced  by the design earthquake, CSREQ, as:

                              CSRm = 0.65 (a^/g)  rd (//)                         (5.3)

Step 5:       Evaluate the standardized SPT blowcount, N^. N^ is the standard penetration test
              blowcount for a hammer with an  efficiency of 60  percent  (60 percent of the
              nominal SPT energy is delivered to  the rods).  The "standardized" equipment
              corresponding to  an efficiency of 60 percent  is specified in  Table 5.4,   If
              nonstandard equipment is used, N^ is determined as;


                      ^60   *   C60                                               (5-4)

              where C^ is  the product  of various correction factors.   Correction  factors
              recommended  by various investigators  for  some common non-standard  SPT
              configurations are provided in Table 5.3.  Alternatively, if CPT data are used, Nw
              can be obtained from the chart relating N& to qc and D50 shown in Figure 5.2 (Seed
              and De Alba, 1986).

Step 6:       Calculate the normalized  standardized SPT blowcount,  (N^.   (ty ^  is  the
              standardized blow count normalized to an effective overburden pressure of 1 tsf
              (2000 psf or 950 kPa) in order to eliminate the influence of confining pressure.
              The most commonly used way to normalize blowcount is via the correction factor,
              CN, shown in Figure 5.4 (Seed et al., 1983). However, the closed-form expression
              proposed by  Liao and Whitman (1986) may also be used:

                                      CN  = (1/0')*                                  (5.5)

              where „' equals the vertical effective stress at the sampling point in tons/ft2.

              As illustrated in Figure 5.4, Equation 5.5, and the correction factor curves are
              valid only for depths greater than 3 m (10 ft).  For depths of less than 3 m (10 ft),
                                           78

-------
              Seed et al. (1983) suggested that a correction factor of  (CN)3  = 0.75 be applied.
              According to these recommendations, the normalized standardized blowcount is
              calculated as:

                                   eo = Neo • CN   (D  3m)                          (5.6a)

                                   = • N60 • (CN)3 (D <  3  m)                        (5.6b)

              There is some indication  that other factors such as grain size distribution can
              influence CN (Marcuson and  Bieganousky, 1977).  However, considering  the
              uncertainties involved in the Standard Penetration test itself, the above correlations
              should be adequate for engineering purposes.

Step 7:       Evaluate the critical stress ratio (CSR) at which liquefaction is expected to occur
              during an earthquake of magnitude M 7.5 as a function of (N^.  Use the chart
              developed by Seed et al. (1985), shown in Figure 5.5, to find CSR (= av/0').

Step 8:       Calculate the  corrected  critical stress ratio  resisting  liquefaction,  CSRL.
              Corrections applied to the CSR calculated in Step 7 include:  kM, the correction
              factor for magnitudes other than 7.5; k, the correction factor for stress levels larger
              than 1 tsf (2000 psf); and k, the correction factor  for the driving static shear stress
              (this is a correction for non-level ground conditions). CSRL is therefore calculated
              as:

                CSRL  CSR   ku   k    k                                          (5.7)


              kM can be determined from chart given in Figure 5.6, developed by interpolation
              through tabular data presented by Seed et al.,  (1983).  k can be determined from
              the chart presented in Figure 5.7 (Harder, 1988; Hynes,  1988).  k depends on the
              relative density of the soil, Dr, and can be determined from Figure 5.8, originally
              proposed by Seed and modified by Harder, (1988), and Hynes (1988).

Step 9:       Calculate the factor of safety against liquefaction, FSL, as:

                    FSL   CSRL I CSREQ                                            (5.8)
                                            79

-------
There is no general agreement on the appropriate factor of safety against liquefaction (NRC,
1985).  There are cases where liquefaction-induced instability has occurred prior to complete
liquefaction, i.e., with a factor of safety greater than 1.0.  However, when the design ground
motion  is extreme or conservative, most geotechnical engineers are satisfied with a factor of
safety, FSL, greater than or equal to 1.0.  It should be noted that the Simplified Procedure is
aimed primarily at moderately strong ground motions (0.2 g  <  a^^ < 0.5 g).   If the peak
horizontal acceleration is larger than 0.5  g, more sophisticated, truly non-linear effective stress-
based analytical approaches should be considered. Computer programs for non-linear evaluation
of liquefaction potential described in the technical  literature include DESRA-2 (Lee and Finn,
1978) and its derivative codes DESRAMOD (Vucetic, 1986) and D-MOD (Matasovi, 1993),
DYNAFLOW (Prevost, 1981), TARA-3 (Finnetal., 1986), LINOS (Bardet, 1987), DYSAC2
(Muraletharan et. al.,  1991), and FLAG (Itasca Consulting Group, Inc.,  1992).

An example of liquefaction analysis using the Simplified Procedure is presented in Appendix A.

5.3   Liquefaction Impact Assessment

For the soil layers for which the factor of safety  against the liquefaction is  unsatisfactory, a
liquefaction impact analysis should be conducted. A liquefaction impact analysis may consist of
the following steps:

Step 1;       Calculate the magnitude  and distribution of liquefaction  induced settlement by
             multiplying  the post-liquefaction volumetric strain, v, by thickness of the liquefiable
             layer, H. v can be estimated from chart presented in Figure 5.9 (Tokimatsu and
             Seed, 1987).    An alternative chart has recently  been  proposed by  Ishihara
             (Ishihara, 1993). However,  application of Ishihara's chart requires translation of
             normalized SPT blowcount (Nt)^ values determined in Section  5.1 to Japanese-
             standard N! values (N, = 0.833 (N,)^;  after Ishihara, 1993). The magnitude of
             seismic settlement should be calculated at each boring or CPT sounding location
             to evaluate the potential variability in seismic settlement across the site.

Step 2:       Estimate the liquefaction-induced lateral displacement, L. The empirical equation
             proposed by Hamada et al. (1987) may be used to estimate L in meters:


                    L  0.75 (H)m (S)1/3                                           (5.9)

             in which H is the thickness of the liquefied layer in meters and S is the ground
             slope in percent.

                                           80

-------
             The Hamada et al.  (1987) formula is  mainly based on  Japanese data on
             displacements of very loose sands for soil deposits having a  slope, S, less than
             10%.  Therefore, Equation 5.9 should be assumed to provide only as a rough
             estimate of lateral displacement. Since the equation does not  reflect the density,
             or (N^go value of the soil, or the depth of the liquefiable layer, it may provide a
             conservative estimate of lateral displacement for denser sands or for cases where
             the soil  liquefies at depth.  Note that estimate of  lateral displacement by this
             equation predicts  large liquefaction-induced lateral displacements in areas  of
             essentially flat ground conditions.  More complex methods for assessment of the
             potential for lateral spreading are available and can be used where appropriate
             (Youd, 1995).

Step 3:      In areas of significant ground slope, or in situations when a deep failure surface
             may pass through waste and through underlying liquefied layers, a flow slide can
             occur following liquefaction.  The potential for a flow slide  to occur should be
             checked using conventional limit equilibrium approach for slope stability analyses
             (discussed in Section 6 of this document) together with residual shear strength in
             zones in which liquefaction may occur. Residual shear strength can be estimated
             from the penetration resistance values of the soil using the chart proposed by Seed
             et al. (1988) presented in Figure 5.10.  Seed and Harder (1990) and Marcuson et
             al.  (1990) present a further guidance for performing a post-liquefaction stability
             assessment using residual shear strengths.

The  above liquefaction-associated deformation  phenomena,  if too great  in magnitude, can
adversely impact the  integrity of the landfill containment structures.  The question the engineer
must answer is "What magnitude of deformation is excessive?"  The magnitude of acceptable
deformation should be determined by the design engineer on a case-by-case basis.   Seed and
Bonaparte (1992)  report that calculated seismic deformations along the liner-waste interface on
the order of 0.15 to 0.30 m (0.5 to 1.0 ft)  are  generally deemed to be acceptable in current
practice in California. As cover deformations are readily  observable and damage to the cover is
repairable, larger  deformations are typically considered acceptable along interfaces in the cover
system than along liner system interfaces.   At the current time,  determination  of allowable
deformations remains a subject requiring considerable engineering judgement.
5.4    Liquefaction Mitigation
                                           81

-------
If the  seismic impact analysis  presented  in Section  5.3 yields  unacceptable  deformations,
consideration may be given to performing a more sophisticated liquefaction potential assessment
and to liquefaction potential mitigation measures.  Generally, the design engineer has  the
following options:  (1) proceed with a more advanced analysis technique; (2) design the facility
to resist the anticipated deformations; (3) remediate the site to reduce the anticipated deformations
to acceptable levels; or (4)  choose an alternative  site.  These options may require additional
subsurface investigation, advanced laboratory testing, more sophisticated numerical modeling,
and, in rare cases, physical modeling.  Discussion of these techniques is beyond the scope of this
study.

Design to resist anticipated  deformations could  include the use of reinforced earth,  structural
walls, or buttress fills keyed into non-liquefiable strata to resist the effects of lateral spreading.

A variety of techniques exist to remediate potential liquefiable soils and mitigate the liquefaction
hazard. Table 5.4 presents a summary of available methods for improvement of liquefiable soil
foundation conditions (NRC, 1985).  The cost of foundation improvement can vary over an order
of magnitude, depending on site conditions (e.g., adjacent sensitive structures) and the nature and
geometry of the liquefiable soils. Remediation costs can vary from as low as several thousand
dollars per acre for dynamic compaction of shallow layers of clean sands in open areas to upwards
of $100,000. per acre for deep layers  of silty soils  adjacent to sensitive structures.
5.5    References

Bardet, J.P.  (1987),  "LINOS, a Nonlinear Finite Element Program for Geomechanics and
Geotechnical Engineering,"  University of Southern California, Los Angeles.

Dobry, R., Powell, D.J., Yokel, F.Y., and Ladd, R.S.  (1980),  "Liquefaction Potential  of
Saturated Sand - The Stiffness Method," Proc. 7th World Conference on Earthquake Engineering,
Istanbul, Turkey, Vol. 3, pp. 25-32.

Finn, W.D. Liam,  Yogendrakumar, M., Yoshida, N.  and Yoshida, H. (1986), "TARA-3:  A
Program   for Nonlinear Static and Dynamic Effective Stress Analysis,",  Department of Civil
Engineering, University of British Columbia, Vancouver, British Columbia, Canada.
Hadj-Hamou, T., and Elton, D.J. (1988), "A Liquefaction Potential Map for Charleston, South
Carolina," Report No. GT-88-1, Toluane University, New Orleans, Louisiana, 67 p.
                                           82

-------
Hamada, M.  Towhata, I., Yasuda S., and Isoyama, R. (1987), "Study on Permanent Ground
Displacements Induced by Seismic Liquefaction," Computers and Geomechanics 4, pp. 197-220.

Harder, L.F., Jr. (1988)*, "Use of Penetration Tests to Determine the Cyclic Loading Resistance
of Gravelly Soils During Earthquake Shaking," Ph.D. Dissertation, University of California,
Berkeley, California.

Hwang, H. and Lee, C.S. (1992), "Evaluation of Liquefaction Potential in Memphis Area, USA."
Proc. 10th World Conference on Earthquake Engineering, pp. 1457-1460.

Hynes,  M.E. (1988)*, "Pore Pressure Generation Characteristics of Gravel Under Undrained
Cyclic Loading," Ph.D. Dissertation, University of California, Berkeley, California.

Ishihara, K. (1985), "Stability on Natural Deposits During Earthquakes," Proc. llth International
Conference on Soil Mechanics and Foundation Engineering, San Francisco, California, Vol.  1,
pp. 321-376.

Ishihara,  K.,  Kokusho, T., and  Silver,  M.L. (1989), "Recent Developments in Evaluating
Liquefaction  Characteristics of Local Soils," State-of-the-Art Report, Proc. 12th International
Conference on Soil Mechanics  and Foundation Engineering, Rio de Janeiro, Brazil, Vol.  4,
pp. 2719-2734.

Ishihara (1993), "Liquefaction and Flow Failure During Earthquakes," Geotechnique 43, No.  3,
pp. 351-415.

Itasca Consulting  Group (1992),  "FLAG, A Finite Difference Computer Program for Two-
Dimensional  Static and Dynamic Analysis."

Iwasaki, T., Tatsuoka, F., Tokida, K-I and Yasuda, S. (1978), "A Practical Method for Assessing
Soil Liquefaction  Potential  Based on Case  Studies at Various  Sites in Japan," Proc.  2nd
International Conference on Microzonation, San Francisco, California, Vol. 2. pp.  885-896.

Kavazanjian,  E. Jr., Roth, R.A., and Echezuria, H. (1985),  "Liquefaction Potential Mapping for
San Francisco," Journal of Geotechnical Engineering, ASCE, Vol. Ill, No. 1, pp. 54-76.

Lee,  M.K.W., and Finn, W.D.L.  (1978),  "DESRA-2, Dynamic Effective Stress Response
Analysis  of  Soil  Deposits  with Energy Transmitting Boundary  Including Assessment  of
* Available through University Microfilms International, (313) 761-4700 Ext, 3879, or (800) 521-0600 Ext. 3879

                                           83

-------
Liquefaction  Potential," Soil Mechanics  Series No.  36, Department  of Civil Engineering,
University of British Columbia, Vancouver, Canada, 60 p.

Liao, S.S.C.  and Whitman, R.V.  (1986), "Overburden Correction Factors for SPT in Sand,"
Journal of Geotechnical Engineering, ASCE, Vol.  112, No. 3, pp. 373-377.

Marcuson, W.F., III, and Bieganousky, W.A. (1977),  "Laboratory Standard Penetration Tests
on Fine Sands," Journal of the Geotechnical Engineering Division, ASCE, Vol. 103, No. GT-6,
June, pp. 565-588.

Marcuson, W.F., m, Hynes, M.E. and Franklin, A.G. (1990), "Evaluation and Use of Residual
Strength in Seismic Safety  Analysis  of Embankments,"  Earthquake Spectra, Vol. 6, No. 3,
pp. 529-572.

Matasovi, N. (1993)*,  "Seismic Response of Composite Horizontally-Layered Soil Deposits."
Ph.D. Dissertation, Civil and Environmental Engineering Department, University of California,
Los Angeles, 452 p.

Muraleetharan, K.K., Mish, K.D., Yogachandran C., and Arulanandan, K., (1991),  "User's
Manual for DYSAC2: Dynamic Soil Analysis  Code for 2-Dimensional Problems,"  Report,
Department of Civil Engineering, University of California, Davis, California.

NRC (1985),  "Liquefaction of Soils During Earthquakes." National Research Council, Committee
on Earthquake Engineering, Washington, DC.

Prevost, J.H. (1981), "DYNAFLOW: A nonlinear transient finite element analysis program."
Department of Civil Engineering and Operational Research, Princeton University, (Last update
January 1994).

Riggs, C.O. (1986), "North American Standard Penetration Test Practice," In Use oflnSitu Tests
in Geotechnical Engineering, ASCE Geotechnical Special Publication No. 6,  pp. 949-965.

Seed,  H.B.  and Idriss, I.M.,  (1982),  "Ground Motions and  Soil Liquefaction  During
Earthquakes," Monograph No.  5,  Earthquake Engineering  Research  Institute, Berkeley,
California, 134 p.
* Available through University Microfilms International, (313) 761-4700 Ext. 3879, or (800) 521-0600 Ext.  3879
                                          84

-------
Seed, H.B., Idriss, I.M.  and Arango, I. (1983), "Evaluation of Liquefaction Potential Using
Field Performance Data," Journal of GeotechnicalEngineering, ASCE, Vol. 109, No. 3, pp. 458-
482.

Seed, H.B.,  Tokimatsu,  K., Harder, L.F.   and Chung,  R.M.  (1985),  "Influence of SPT
Procedures in Soil Liquefaction Resistance Evaluations," Journal of Geotechnical Engineering,
ASCE, Vol. Ill, No. 12, pp. 1425-1445.

Seed, H.B. and De Alba, P. (1986), "Use of SPT and CPT Tests for Evaluating the Liquefaction
Resistance of Sands." In Use ofln-Situ Tests in Geotechnical Engineering, ASCE Geotechnical
Special Publication No. 6, pp. 281-302.

Seed, H.B., Seed, R.B., Harder, L.F., Jr., and Jong, H.-L. (1988), "Re-Evaluation of the Slide
in the Lower San Francisco Dam in the Earthquake  of  February  9,  1971."  Report No.
UCB/EERC-88/04, University of California, Berkeley, California.

Seed, R.B.,  and Harder, L.F., Jr.  (1990),  "SPT-Based Analysis of Cyclic Pore Pressure
Generation and Undrained Residual Strength," Proc. H.B. Seed Memorial Symposium, Vol. 2,
pp. 351-376.

Seed, R.B., and Bonaparte, R. (1992),  "Seismic Analysis and Design of Lined  Waste Fills:
Current Practice," In Proceedings of Stability and Performance of Slopes and Embankments - II,
Vol. 2, Berkeley, California, ASCE Geotechnical Special Publication No. 31, pp.  1521-1545.

Shibata, T., and Taparaska, W. (1988),  "Evaluation of  Liquefaction Potentials of Soils Using
Cone Penetration Tests," Soils and Foundations, Vol. 28, No. 2, pp. 49-60.

Skempton, A.W. (1986),  "Standard Penetration Test  Procedures  and the Effects in Sands of
Overburden  Pressure, Relative  Density,  Particle  Size,  Ageing  and  Overconsolidation,"
Geotechnique, Vol.  36, No. 3, pp. 425-447.

Tinsley, J.C., Youd, T.L., Perkins, D.M., and Chen, A.T.F. (1985),  "Evaluating  Liquefaction
Potential," in J.I. Ziony, ed, Evaluating Earthquake Hazards in the Los Angeles Region, an Earth
Science Perspective,"  U.S. Geological Survey Professional Paper 1360, pp. 263-315.

Tokimatsu, K. and Seed, H.B. (1987), "Evaluation of Settlements in  Sands due to Earthquake
Shaking," Journal of Geotechnical Engineering, ASCE,  Vol. 113, No. 8, pp. 861-879.
                                          85

-------
Vucetic, M.  (1986)*, "Pore Pressure Buildup and Liquefaction of Level Sandy Sites During
Earthquakes," Ph.D. Thesis, Rensselaer Polytechnic Institute, Troy, New York, 616 p.

Youd, T.L.,  and Perkins, D.M.  (1978),  "Mapping Liquefaction-Induced Ground  Failure
Potential," Journal of Geotechnical Engineering, ASCE, Vol. 104, No. GT4, pp. 433-446.

Youd, T.L. (1995) "Liquefaction-Induced Lateral Ground Displacement," Proceedings of the
Third International Conference on Recent Advances in Geotechnical Earthquake Engineering and
Soil Dynamics, University of Missouri, Rolla, 2-9 April.
' Available through University Microfilms International, (313) 761-4700 Ext. 3879, or (800) 521-0600 Ext. 3879
                                           86

-------
Table 5.1     Estimated Susceptibility of Sedimentary Deposits to Liquefaction During
              Strong Seismic Shaking (Youd and Perkins, 1978).
Type of
deposit
0),
General dis-
tribution of
cohesionless
sediments
in deposits
(2}
Likelihood that Cohesionless Sediments.
When Saturated, Would Be Susceptible
to Liquefaction (by Age of Deposit)
<§00yr
<3)
Holocene
(4)
Pleis-
tocene
(S)
Pre-
pleis-
tocene
(6)
                          (a) Continental Deposits
River channel
Flood plain
Alluvial fan and
plain
Marine terraces
and plains
Delta and fan-
delta
Lacustrine and ^
playa
Colluvium
Talus
Dunes
Loess
Glacial till
Tuff
Tcphra
Residual soils
Sebka
Locally variable
Locally variable

Widespread

Widespread

Widespread

Variable
Variable
Widespread
Widespread
Variable
Variable
Rare
Widespread
Rare
Locally variable
Very high
High

Moderate

—

High

High
High
Low
High
High
Low
Low
High
Low
High
High
Moderate

Low

Low

Moderate

Moderate
Moderate
Low
Moderate
High
Low
Low
High
Low
Moderate
Low
Low

Low

Very low

Low

Low
Low
Very low
Low
High
Very low
Very low
f
Very low
Low
Very tow
Very low

Very low

Very low

Very low

Very low
Very low
Very low
Very low
Unknown
Very low
Very low
?
Very low
Very low
                                  Coastal Zone
Delta
Esturine
Beach
High wave
energy
Low wave
energy
Lagoonal
Fore shore
Widespread
Locally variable


Widespread

Widespread
Locally variable
Locally variable
Very high
High


Moderate

High
High
High
High
Moderate


Low

Moderate
Moderate
Moderate
Low
Low


Very low

Low
Low
Low
Very low
Very low


Very low

Very low
Very low
Very low
                                 (c) Artificial
Uncompacted fill
Compacted fill
Variable
Variable
Very high
Low
—
—
—
                                   87

-------
   TABLE 5,2: RECOMMENDED "STANDARDIZED" SPT EQUIPMENT
                  (After Seed et al.,  1985, and Riggs, 1986)
Sampler:
Drill Rods:
Hammer:
Rope:

Borehole:
Drill bit:
Blowcount Rate:

Penetration Resistance
Count:
Standard split-spoon sampler with:   (a) O.D. =
2.00 in., and (b) LD.  — 1.38 in. (constant - i.e. no
room for liners in the barrel)

A or AW for depths less than 50 ft; N or NW for
greater depths

Standard (safety) hammer with: (a) weight = 140 Ib;
(b) drop =  30  hi. (delivers  2,520 in.-lbs which is
60% of theoretical fteefail)

Two wraps of rope around the pulley

4 to 5-in. diameter rotary borehole with bentonite mud
for borehole stability (hollow stem augers where SPT
is taken through the stem)

Upward deflection of drilling mud (tricone or baffled
drag bit)

30 to 40 blows  per minute

Measured over  range of 6 to  18  in. of penetration
into the ground.
 Note:   If the  equipment meets the above specifications, N  = NM  and only a
        correction for overburden is needed.
                                         88

-------
            TABLE 5.3;  CORRECTION FACTORS FOR NONSTANDARD SPT PROCEDURE AND EQUIPMENT
CORRECTION FOR
Nonstandard Hammer Type
(DH = donnut hammer; ER = energy ratio)
Nonstandard Hammer Weight or height of fall
/il — hpicrhf AT* ftill ifi tnHhfC" \V — u/^i&ht in 1Hc\

Nonstandard Sampler Setup (standard samples with
room for liners, but used without liners)
Nonstandard Sampler Setup (standard sampler with
room for liners, and liners are used)
Change in Rod Length
Nonstandard Borehole Diameter
CORRECTION FACTOR
CHT » 0.75 for DH w/rope and pully
COT - 1.33 for DHjv/trtp/auto & ER = 80
c - H'w
HW 140*30
GSS « 1.10 for loose sand
GSS *» 1.20 for dense sand
Csg « 0.90 for loose sand
Cjs « 0.80 for dense sand
CM, «• 0.75 for rod length 0-10 feet
CBD » 1.05 for 6 in. diameter
CBD « 1.15 for 8 in. diameter
REFERENCE
• Seed et al. (1985)
calculated per Seed et. al
f1QQ*\
(Ly03)
Seed et al. (1985)
Skempton (1986)
Seed et al. (1983)
Skempton (1986)
CO
Note:
          = C
             ST
             'HW
'SS
"BD

-------
  Table 5.4         Improvement  Techniques for  Liquefiable Soil  Foundation  Conditions (NRC,  1985).
Ktcihod
                                                StfiUfcle So3
MaxMMflB Effect**
        * Depth
                                                                                     TuencdAra
                                                                                                          MCI! ftopenief 40m                Any size
20 m routinely (Jntf*
  fectm aboW 3-4
  m depth); >» m
  sometime*; Vib~
      f. 40m
                                                                                                          C«i obtain rrl«trv<
                                           •0%;
                                           *bk dct
                                                                                    C«n obmin rel*(iv«
                                                                                      denMtki of 90% or
                                                                                      more. IfOMrflTeciive in
                                                                                      some Hands.
                                                                                                          Induce UquciJkction in   2
                                                                                                            coflUoikJ and lim-   3
                                                                                                            feed stages «ad in-
                                                                                                            erasc relative den.
                                                                                                                                 range. Hus hecn
                                                                                                                                     n effective m
Induce liqueCftetioo in  2
  controlled and Um-   3
  rted Xacct and in-
  ercaic rcbtivc den-
  ftily to potcnli&lly
Moderate
    2<3 m > 1,000 m3
placement of pfle partiy saturated
volume and by vi- daytry sods; loess.
braiion daring driv-
ing, incre*** m lat-
eral effective earth
pressure.







<5} Heavy tamping Repeated application Cohestonlcss soils 30 m (possibly >J,300 m;
(dynatnk compac- of high-intensity best, other types deeper!
tion) imriacts al surface. can also be tnv
proved.










(63 Displacement/ Highly viscous grout AH soiH. Unlimited Small
compaction acts as radial hy-
erout draulicjack when
pumped in under
high pressure.


CBJI obtain high den-
SiUc*, good ttoi-
fomtliy. Relative
densities of more
than 16%.









Can obtain high rela-
tive densities, rea-
sonable uniformity.
Relative densities
of 80% of more .









Groirt bwlbs within
compressed soil
matrix. Soil mass
as a whole is
Strengthened,


Useful in toils with I
fines. Increases reS- 2
ative densttie* lo J
nonUque&abte
range. Is ttsed Ms
prevent liquefac-
tion. Provides shear
rcsisUAce isfeori-
zonlal and inclined
diroetiocn. Useful
10 stabilize slopes
*r»d strengthen po-
lential failure sur-
faces or dip circles.
Suitable for some 2
soils with fines; MS* 3
able above and be-
low water. In cobe*
sionless coils.
induces liquefaction
in controlled and
limited stages and
increases relative
density lo poten-
tially nonli^ueSabie
range. Is used to
prevent liquefac-
tion.
Increase in soil rcla- 1
live density and 2
horizontal effective 3
Stress. Reduce lique-
faction potential.
Stabilize the ground
against movement.
Moderate to
feugh












LowtSO.40-
l&OQ/m1}












Low to moder-
ate {53.00-
SlS.Q&m1)




    "SP. SW. or SM soils that have average relative density equal to or greater than 85 percent and the minimum relative density not less than 80 percent are m general not susceptible to tiquefecuon
CTM 3-4ntM). D'Appolonla (19TO) stated! thai for soil within the zone of influence and confinement of the structure foundation, the relaltve density should not be less than 70 percent. Therefore, a criteria
may be med that relative density increase into the 70-90 percent range is in general considered to prevent liquefaction. These properties of treated! materials and applications occur onfy undfr ftttal
conditions of soil, moisture, and method application. The methods and properties achieved are IKM applicable and will not occur in all soils.
    *Applic3UQns and results of the Improvement methods are dependent oa: C*l soil profiles, types, and conditions, {bj site conditions, (c) earthquake loading. 
-------
  Table  5.4:  (continued)
M«t«xl Prim** "£**!££%? *•««••« fa**** E«««kSiK^ Wcril^pmic..*
	 — 	 — 	 *-a*an<»«-*«»«*^\Hm Unknown ^ Tf^ . ••
t*m crcat€ ^ objtf^. m unknown SoWed
to confine an area
of fiquenabfc SCK| so
that liquefied soil
e&utld not How oyt
                                                                            SOIL KKJWOKCCMKNT
30 m {limited by vi* >IXtt m*: fine* Increased vertkal and Provides (a} vertical S
hntlory equipmeotl grained «wli, horizontal ioad car- support, (b) drains 2
>IQOO m- rying capacity. lo relieve pore A'
Density increase in water pressure, and
cohesKjnkrss soils. 
-------
         0.001
           0.01
     0.1           1

  Grain  size  (mm)
  10
                    : Boundary for most liquefiable soil       1 /Tsuchida\

           /—(§>-/ : Boundary for potentially liquefiable soil J '  1570   /

              /.*•/  : Tailings slime (Ishihara. 1985)

              ,f /  : Liquefied sand in             / /  Liquefied sand in
             ^^     Chiba-Toho-Oki Eq. (1987)    / /  : Nihonkai-Chubu

              f  f   : Liquefied sand in                 &
             ' t      Niigata Eq. (1964)
    100
 bo
 '5
 ^  50

 .£
 <«^
 -4->
 c
 o
Extruded grave!
and sand at an
excavated site
at Biwa lake
bottom
                                                         Liquefied gravel
                                                         at Borah Peak Eq.
          0.01
           0.1
    1.0          10

Particle size  (mm)
100
1000
Figure 5.1   Grain Size Distribution Curves of Liquefied Soils (Ishihara et al.,  1989).
                                      92

-------
 2
 u
cr
tc
e
o
S 3
41
c
O)
Q-
     -   0
            JL
                                         O JomioUowlkl «1 «l («98S1
                                         0 MtKomocSi and KobayosN «962)
                                         & tihihofd o«d Koqo (19611
                                         •to Robertson (19621
                                         V Milch* H (19651
                                         fl Hardtt  €101(1994}
                                                            Mowt/fool
                    JL
                                                      JL
   O/Dt    OO2      COS     O.I       O.2        O,5      I
                               Mean Groin Size, Oso -mm
Figure 5.2
Variation of
1986).
                                Ratio with Mean Grain Size, DSO (Seed and De Alba,
                                           93

-------
                              \  max)d
  0


 10


 20


 30
                                   \
                                maxlr
          0.1  0.2  0.3  O.A  0.5  0.6  0.7   0.8 0.9 1.0
a.
tlJ
 50


 60


 70


 80


 90


100
           I     I    I    I     I     I     I     I     I
                           AVERAGE VALUES
             RANGE  FOR DIFFERENT
             SOIL PROFILES
           I	I
I	I
         Figure 5.3   Stress Reduction Factor, rd (Seed and Idriss, 1982).
                              94

-------
                 I  _
             r   2
             Sf
*:
 I

 *t
&.   5
 c
 »_
I   6
 V
a
I   7
 o
 «*
bl   8
                IO
                        O.2
                  O.4
                  nr
0.6
                        qc-volue« by CPT-
                        (Df
                     O.8
I.O
 12
T"
                                               -Of »6O !o8O%
                                               -Or » 40 to 60%
                                 N-volu«s by SPT
                           J.
        1
                                                                   1.4
1.6
0,2    a4    as    as
                                                     IX)
                                                       1.4
                                     1.6
Figure 5.4     Correction Factor for the Effective Overburden Pressure, CN (Seed et al., 1983).
                                             95

-------
0.6
0.5
0.4
0.2
0.1
            PERCENT FINES - 35   IS
                                      FINES CONTENT 2  5%

                                     ® MODIFIED CHINESE CODE
                                   PROPOSAL (CLAY CONTENT . S%]

                                                    MARGINAL
                                                  LIQUEFACTION
                             LIQUEFACTION
                   NO
              LIQUEFACTION
             PAN-AMERICAN DATA
             JAPANESE DATA

             CHINESE DATA
            	I	
                                                                     O
                                                                     o
                10
                                    20
30
40
50
Figure 5.5    Relationships Between Stress Ratio Causing Liquefaction and
             for Sands for M 7.5 Earthquakes (Seed et aL, 1985).
                                                                  Values
                                96

-------
   1.6
   1.5 -
55 1.3
O
I-H
E-.
u £ 2

OS
w
<•"	»

§ 0.9 H
   0.8
           i i i  i i i i i   it ii i i ii
                    Se|d et al.  (1J983)
                    Intjerpolated
                    6789
                  EARTHQUAKE  MAGNITUDE  (M)
Figure 5.6    Curve for Estimation of Magnitude Correction Factor,
           1983).
                                                 (after Seed et al.,
                              97

-------
                 0.4
X
                       G
                       O
                       %
RCtO 6EOFORO CANO, Of m 40.
MCW JERSEY eAOKTVUL. ffi RC .
	 _ 1	I	1
                                       2.0
                                                »-«
                                                                   6.0
                                                                                      7.0
                                          £FFECT1VE OVERBUROCH PRESSURE. KSf
Figure 5.7    Curves for Estimation of Correction Factor
              1988, as Quoted in Marcuson et al., 1990)
                                                                            (Harder 1988, and Hynes
                                                             98

-------
       2.S
       2.0
Of - 55 - 70%
       1.S
       1.0
       0.5
                                                      D, -  35%
             NOTE:  RANGES NOT PROVIDED FOR
                     Df -  40% AND 35% DUE TO
                     LIMITED DATA
             	I	l	
                      0.1          0.2          0.3
                        INITIAL STATIC STRESS RATIO. 
-------
        O.G
         05
         O.4-
         O.3
         O2
         O.I
             Volumetric Strain-%
            1054 3      2
           0
10
2O
30
40
50
Figure 5.9   Curves for Estimation of Post-Liquefaction Volumetric Strain using SPT
            Data and Cyclic Stress Ratio (Tokimatsu and Seed, 1987).
                                       100

-------
     2OOO
      I6OO
  UJ
  f£
  tr
  <
  ui
  I
  en
  o
  z
  O
  in
  UJ
  oc
      I2OO
       80O
       40O
     4|    EARTHQUAKE - INDUCED LIQUEFACTION AND SLIDING CASC HISTORIES VVMERE
          SPT  CMkT* *NO nCSIOUAL STRENCTM PARAWCTtffS HAVE BEEM MEASURED.



     O    EARTHQUAKE -INDUCED LWUEFACTION AND SUOIMG CASE MISTOA
-------
                                      SECTION 6

                          258.14 SEISMIC IMPACT ZONES:
                SLOPE STABILITY AND DEFORMATION ANALYSIS

The potentially large accelerations associated with seismic events can induce significant forces that
may lead to permanent deformations within a MSW landfill. These deformations potentially can
lead to impairment of the functions of the containment system. However, reports of significant
seismic-related damage to MSW landfill containment systems are relatively rare.  Several studies
dealing with landfill behavior during the 1989 M 7.1 Loma Prieta earthquake report only minor
damage to landfills, even for the landfills located in the epicentral region or founded on relatively
weak San Francisco  Bay mud (Orr  and Finch, 1990; Buranek and Prasad, 1991; Sharma and
Goyal, 1991; Johnson et al., 1991). Damage was mostly limited to cracking of earthen cover soils
and disruption of surficial piping systems.  No geomembrane-lined landfills were impacted by the
Loma Prieta event. At least three modern, geosynthetic-lined landfills were impacted during the
1994 M 6.7 Northridge Earthquake.  While preliminary  studies indicate that no major damage
occurred, the geosynthetic liner system was torn in at least two locations above the limit of waste
placement at one of the landfills (EERC, 1994; (Matasovi, et al., 1995).

Numerous methods  and  procedures are  currently available  to  evaluate static slope stability
(Duncan, 1992).  Most of the methods available are, in some form, suitable for seismic stability
analyses.  They can be used in conjunction with several different approaches for seismic analysis,
of which the following two conventional methods represent the current state-of-practice: (1)
pseudo-static factor of safety approach, and (2) permanent seismic deformation approach.  Both
of these conventional approaches to seismic stability assessment are based on the  principles of
limit equilibrium analysis.

In the pseudo-static factor of safety approach, a seismic coefficient is specified to represent the
effect of the inertial forces imposed by the earthquake upon the potential failure mass and a factor
of safety is defined in the conventional manner as the ratio of the ultimate shear strength of the
slope elements to the maximum shear stresses induced in those elements by seismic and static
loadings. The main  drawback of the pseudo-static factor of safety approach lies in its inability
to rationally relate the  value of the  seismic coefficient to  the  characteristics of  the design
earthquake.  Use  of the  peak acceleration (expressed as a fraction of gravity) as the seismic
coefficient in conjunction with a pseudo-static factor of safety of 1.0 has been shown to give
excessively conservative assessments of slope performance in earthquakes.

In contrast to the  pseudo-static factor of safety  approach, the permanent seismic deformation
approach involves the calculation of cumulative seismic deformations.  The most commonly used

                                          102

-------
method for calculating the permanent seismic deformation of slopes is  termed the Newmark
method (Newmark, 1965).  In this approach, the potential failure mass is treated as a rigid body
on a yielding base.  The acceleration time history of the rigid body is assumed to correspond to
the average acceleration time history of the failure mass. Deformations accumulate when the rigid
body acceleration exceeds  its yield acceleration.   The yield acceleration  is the  horizontal
acceleration that results in a factor of safety of 1.0 in a pseudo-static limit equilibrium analysis.

The calculation of permanent seismic deformations using the Newmark approach is depicted in
Figure 6.1. Acceleration pulses in the time history that exceed the yield acceleration are double
integrated  to calculate cumulative relative displacement.  In a Newmark  analysis,  relative
displacement is often assumed to accumulate in only one direction, the downslope direction. With
this assumption, the yield acceleration in the other (upslope) direction is implicitly assumed to be
larger than the peak acceleration of the failure mass being analyzed.

In practice, both the pseudo-static factor of safety and permanent seismic deformation approaches
are often combined in a unified seismic slope stability and deformation analysis. Such an analysis
outlined in Section 6.2 of this guidance document.

6.1    Key Material Properties

To perform seismic  slope stability  analyses, estimates of the unit weight and (dynamic) shear
strength parameters  of various components of the  landfill are needed.  Unfortunately, a large
amount of uncertainty often  exists as to  appropriate values for some of these parameters.
Evaluation of the material properties of MSW required  for a slope stability  analysis can be a
difficult task. This is due to the paucity of field and laboratory measurements of MSW properties,
the cost and difficulty in making project-specific field measurements, and the heterogeneous nature
of MSW.  The following sections summarize the information currently available for estimating
key material properties of MSW.

6.1.1  Unit Weight

Values of unit weight for MSW reported in the literature are summarized in Table 6,1  (Fassett
et al., 1994).  Initial values of MSW unit weight can be estimated from landfill gate receipts and
survey elevations of the waste face.  Current regulations in California require operators to achieve
an initial density of at least 1,250 Ibs per cubic yard (7.3 kN/m3 or 46 lb/ft3). Average values of
MSW unit weight can also be estimated based upon the total gate receipts over the life of a landfill
and survey data.  Average values for MSW unit weight cited by landfill operations  and used in
practice for landfill capacity estimates typically vary from 8.6 to 10.2 k/m3 (55 to 65 Ibs/ft)
(Kavazanjian et al.,  1995).  Landfill-specific values of MSW unit weight will depend upon actual
                                       *
                                           103

-------
operational practice.  For instance, significantly higher MSW unit weights have been reported for
a landfill that used an unusually high percentage of daily cover soil (Richardson and Reynolds,
1991).

The MSW unit weights in Table 6.1 do not account  for the increase in density with depth that
occurs in MSW due to its compressibility or to changes that occur with time.  Kavazanjian et al.
(1995) have demonstrated that the variation of density  with depth can have a significant influence
on the results of static and dynamic stability and seismic response analyses.  The dashed line on
Figure 6.2 (Kavazanjian et al., 1995) shows the density-depth relationship  developed for one
southern California landfill on the basis  of field  measurements of density and laboratory
measurements of waste compressibility (Earth Technology, 1988). Based upon the density-depth
profile developed by Earth Technology (1988), the initial and  average unit weights cited above,
and representative compressibility values for MSW reported by Fassett et al. (1994),  Kavazanjian
et al. (1995) developed the MSW unit weight profile shown by the solid line on Figure 6.2 for use
in stability and seismic response analyses of MSW landfills in the absence of landfill-specific data.

6.1.2  Interface Shear Resistance

The interface shear resistance between geosynthetic components (e.g., a  geomembrane and
geotextile interface) and between soil and geosynthetic components (e.g., a geomembrane and low
permeability soil interface) from static laboratory tests are generally  used for dynamic stability
analyses.  Typical values for peak and residual interface friction angles have been reported by
Williams and Houlihan (1986, 1987), Seed and Boulanger  (1990), Koerner  (1991), and Byrne
(1994). Byrne (1994) recommended that the interface friction angle used in dynamic analysis be
evaluated on the basis of compatibility between the load-deformation curve from laboratory testing
and the calculated seismic deformation.

Some  investigations have reported slight  differences  between static and  dynamic  interface
strengths (Kavazanjian et al., 1991).  Others have reported that interface shear strengths appear
to be independent of the frequency content and number of cycles of motion (Yegian and Lahlaf,
1992). However, considering the uncertainties inherent to other material properties, it appears
that the shear strength measured in static tests may be reasonably used to represent the dynamic
interface shear strength in seismic stability and deformation analyses.

6.1.3  Low Permeability Soil

The dynamic shear strength of clay soils may be influenced by the amplitude of the cyclic deviator
stress, the number of applied cycles, and the plasticity of the clay (Makdisi and Seed, 1978).  In
many cases, the static shear strength may be the same as or  even greater than the shear strength

                                           104

-------
for dynamic loading.  For saturated, normally consolidated soft clays, the dynamic shear strength
can be assumed to be equal to at least 80% of the static undrained strength with a high degree of
confidence.  However, for sensitive and stiff clays, cyclic loading  can lead to reductions in
strength and accumulated deformations can exceed the strain at which peak strength is mobilized,
resulting in reduction of strength to residual values.

6.1.4  Granular Soil Shear Strength

The cyclic shear strength of a dry or unsaturated granular  soil (sand or gravel with degree of
saturation less than 80 percent) can be assumed to be equal to or greater than the drained ultimate
(large strain) static shear strength.  In saturated sands, seismic loading can significantly alter the
dynamic shear strength. Evaluation of the potential for shear strength reduction in a saturated
or almost saturated cohesionless soil (low plasticity silt, sand, or gravel) subject to dynamic
loading may require  sophisticated cyclic laboratory testing.  Alternatively, a residual shear
strength may be assigned to the soil based upon either undrained laboratory tests or in situ test
results.  Evaluation of residual shear strength from laboratory tests is not recommended due to
the difficulties associated with testing.  Use of residual strengths derived from in-situ testing is
in general considered  more reliable. However,  use of residual shear strengths in a pseudo-static
stability  assessment can result in a very conservative assessment of the pseudo-static factor of
safety and/or yield acceleration and is not recommended for most problems (Marcuson et al.,
1990).

6.1.5  MSW Shear Strength

The available data on MSW shear strength is relatively limited. Available data includes laboratory
test results on reconstituted samples and strength values backfigured from field load tests and case
histories of landfill performance.  Laboratory and field tests  have  consistently shown shear
strengths in excess of a cohesion of 200 psf (10 kPa) and a friction angle of 20 degrees (Landva
and Clark, 1990 and  Richardson and Reynolds, 1991).  Table 6.2 presents a compilation of the
available data of MSW developed by GeoSyntec  (1993).  Table 6.3 presents a compilation of
lower bound friction angles backfigured from observations of the satisfactory performance of steep
side slopes at existing landfills by GeoSyntec (1993) based upon the assumption of a cohesion of
100 psf  (5 kPa).  Figure  6.3 presents  a bi-linear strength envelope for MSW developed by
Kavazanjian et al. (1995) based upon evaluation of the data in Tables 6.2 and 6.3.

Observations of the satisfactory performance of landfill slopes in major earthquakes indicates that
the dynamic shear strength of MSW may be significantly greater than the static shear strength.
Figure 6.4 (Siegel et al., 1990) shows combinations of cohesion, friction angle, and yield
acceleration that resulted a pseudo-static factor of safety of 1.0 in  analyses of the slopes of the

                                            105

-------
Oil landfill. .  Observations of the satisfactory performance of the Oil landfill slopes in recent
earthquakes combined with this plot indicate that the dynamic shear strength of MSW mobilized
during seismic loading is greater than the values indicated in Figure 6.3.

6.1.6  Sensitivity Studies

It  is strongly recommended  that all stability analyses of MSW landfills be performed using
parametric studies to clearly identify the sensitivity of the performance of the landfill to the
material properties used  in  the analysis.   If  performance depends  significantly on a given
parameter, then additional laboratory or field testing may be required to better define appropriate
properties for design.

6.2    Seismic Stability and Deformation Analysis

A  prerequisite for performing a seismic slope stability and deformation analysis is performance
of a static slope stability analysis.  The seismic stability and deformation analysis is carried out
using the same basic model(s) of landfill (waste mass and foundation) and containment system
used in the static  analysis. The following steps are used in a typical seismic slope stability and
deformation analysis:

Step 1:      Reinterpret the cross-sections  analyzed in the static stability analysis and assign
             appropriate dynamic strength parameters.  In cases where it is not clear whether
             drained or  undrained shear strength parameters are appropriate for the dynamic
             analysis, follow guidelines presented in Duncan (1992).

Step 2:      Evaluate the seismic coefficient, ks. There are many different views on how to
             define ks (e.g., Seed and Martin, 1966; Seed,  1979; Marcuson, 1981; Hynes and
             Franklin, 1984). The most reasonable definition appears to be one that regards the
             seismic coefficient as an empirical  factor.   This definition  recognizes the
             limitations  of the pseudo-static slope stability analysis in representing the actual
             effects of an earthquake on the slope.  Unfortunately, this definition provides no
             guidance to selection of an appropriate value of ks. Seed (1979) reports that clay
             slopes and embankments designed with a minimum pseudo-static factor of safety
             of 1.15  using a seismic coefficient of 0.15  have experienced  "acceptable"
             deformations in earthquakes of magnitude less than 7.5 and intensity less than 0.75
             g.   However, Seed's definition of acceptable deformations appears  to  include
             deformations of over one meter in some cases.
                                           106

-------
             Figure 6.5 shows the results of Newmark seismic deformation analyses performed
             by Hynes and Franklin (1984) using 387 strong  motion records and 6 artificial
             accelerograms.  Based upon this data and their experience with seismic response
             analyses of slopes and embankments, Hynes and  Franklin (1984) concluded that
             slopes and embankments with a yield acceleration equal to half the peak ground
             acceleration would experience permanent seismic deformations of less than one
             meter ( 3 ft) in any earthquake, even for embankments where amplification of
             acceleration by  a factor of three occurs.   In the  absence of amplification, or if
             amplification is taken into account in determining the peak acceleration, the Hynes
             and Franklin data suggest that deformations will remain less than 0.3 m (1 ft)  for
             yield accelerations  less than or equal to   one-half  the  peak   acceleration.
             Therefore,  based upon the work of Hynes  and Franklin,  it appears that  the
             maximum value of ks may be determined as ks = 0.5 •  amax/g to limit permanent
             seismic deformations to less  than 0.3  m (1  ft), where amax is peak horizontal
             acceleration at the ground surface for analyses of the liner system and at the top of
             the landfill for analyses of the cover system, a^ can be  estimated either using the
             simplified methods presented in Section 4 of this guidance document or from the
             results of a seismic response analysis.

Step 3:      Perform the pseudo-static stability analysis. If the minimum factor of safety, FS^,
             exceeds  1.0 and 0.3 m (1  ft)  of deformation is acceptable, the seismic stability
             analysis is completed.

Step 4:      If the pseudo-static factor of safety is less than 1.0 or the acceptable deformation
             is less than 0.3 m (1 ft), perform a Newmark deformation analysis.  This is done
             with the following three steps:

       1)    Calculate the yield acceleration, ky.  The yield acceleration is usually calculated in
             pseudo-static analyses using a trial and  error procedure in which the seismic
             coefficient is varied until FSmin = 1.0 is obtained.  The lowest yield acceleration
             for all possible failure surfaces passing through the liner,  cover, and/or waste mass
             should be evaluated.

       2)    Calculate the permanent seismic deformation.  The permanent seismic deformation
             may be calculated using either simplified design charts (e.g.,  Hynes and Franklin,
              1984; Makdisi  and Seed,  1978) or a formal time-history analysis in which the
             excursions of the average acceleration time history above the yield acceleration are
             double integrated.
                                           107

-------
       3)     Compare the calculated permanent seismic deformation to the allowable maximum
             permanent displacement, u^.  Seed and Bonaparte (1992) report that umax values
             of 0.15 to 0.3 meters (0.5 to 1.0 feet) are typically used in practice for design of
             geosynthetic liner  systems.   For cover  systems,  where permanent  seismic
             deformations may be observed  in post-earthquake inspections and damage to
             components can be repaired, larger permanent deformations may be considered
             acceptable.  In fact, some regulatory agencies consider seismic deformations of the
             landfill cover system primarily a maintenance problem.

Several investigators have presented  simplified charts  based upon the  results  of Newmark
deformation analyses for estimating permanent seismic deformations.  Makdisi and Seed (1978)
developed the chart  shown in Figure 6.6 from the results of two-dimensional finite  element
analyses of embankments founded upon rock. This chart includes the effect of amplification of
seismic motions by an embankment and provides upper and lower bounds on the permanent
deformation as a function of magnitude.  Hynes and Franklin (1984) developed the chart shown
in Figure 6.5 from classical Newmark  "sliding block on a plane" analyses.  The Hynes  and
Franklin chart does not consider amplification or magnitude effects but includes time histories
which encompass a wide range of soil conditions.  Due to the uncertainties in using a simplified
design chart and the characteristics and limitations discussed above, the use of the upper bound
curves from Makdisi and Seed (Figure  6.6) for  simplified analysis of the permanent seismic
deformation potential of the waste mass and liner system. The mean +  curve from Hynes  and
Franklin (Figure 6.5) is recommended for simplified permanent seismic deformation analysis of
the cover system.

If a  seismic  response analysis  has been performed, a  formal Newmark seismic deformation
analysis can  be performed by  using the acceleration or shear stress time histories from the
response analysis.  Jibson (1993) describes the  analytical procedure for performing  such an
analysis. To  evaluate the permanent displacement of the landfill mass, the average acceleration
time history of mass above the critical failure plane (the failure surface with the lowest yield
acceleration) should be used.  The average acceleration time history may be calculated as the
average of the acceleration time history of each layer above the interface weighted according to
the unit weight and thickness of each layer.  Alternatively, the average acceleration time history
may be calculated from the shear stress at the interface divided by the total vertical stress above
the interface, as described by Repetto et al.  (1993).  To calculate the permanent deformation of
the landfill final cover, either the average acceleration time history of the cover or the shear stress
time history at the cover-waste interface divided by the total vertical stress at the interface should
be used in the Newmark analysis.  Particularly for landfills in the eastern United States, where
the earthquake acceleration time history may contain relatively enriched high frequencies, the
                                          108

-------
formal Newmark deformation analyses may yield significantly lower seismic deformations than
simplified Newmark analyses using Figures 6.5 and 6.6.

6.3    Additional Considerations

Stability of the underlying foundation soil is an important consideration in evaluating the overall
performance of the landfill, particularly if a layer (or layers) in the foundation is susceptible to
liquefaction, as illustrated in Figure 6.7.  The potential for a liquefaction induced flow failure may
be analyzed  using limit equilibrium  analyses by employing residual shear  strengths  in the
potentially liquefiable zones.  In this  type of post-earthquake stability assessment, the seismic
coefficient should be set equal  to zero  (Marcuson et al., 1990). If the residual shear strength is
conservatively assessed using minimum values of SPT blow counts (or CPT tip resistance) within
the potentially liquefiable  layer(s), a  factor of safety of 1.1  may  be considered as acceptable.
Evaluation of seismic settlement potential, as described in Section 5.3, still must be conducted to
assess the impact of liquefaction on the landfill.

In some situations, it may be convenient to treat the final cover of the landfill as an infinite slope.
In these situations, the pseudo-static factor of safety and yield acceleration for the cover may be
assessed using the following general equations for the stability of an infinite slope (Matasovi,
1991):
     cl(  z  cos2 )   tan  1    (z d )/(    z)   k   tan
                               "'"''*    	                          (6.1)
s
                              k   tan
                               s
          cl(   z   cos2 )   tan   1   w(z djl(    z)  tan
    y                    1     tan    tan
where FS = factor of safety, ky = yield acceleration, k,, = seismic coefficient,  =unit weight of
slope material(s), w= unit weight of water, c = cohesion,  = angle of internal friction of the
assumed failure interface or surface, z = depth to the assumed failure interface or surface, and
dw  = depth to the water table (assumed parallel to the slope).  The above equations yield the
factor of safety and yield acceleration explicitly for  both cohesive (c 0 ) and cohesionless soils (c
= 0). If there is no downslope seepage, the depth to the water table, dw, should be set equal to
the depth to the assumed failure plane, z.
                                            109

-------
6.4    References

Bray, J.D., Augello, A.J., Leonards, G.A., Repetto,  P.C., and Byrne, R.J.  (1995), "Seismic
Stability Procedures for Solid Waste Landfills,"  Journal of Geotechnical Division, ASCE, Vol.
121, No. 2, pp. 139-151.

Buranek, D. and  Prasad, S. (1991), "Sanitary  Landfill Performance During the Loma Prieta
Earthquake," Proc. 2nd International Conference on Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, St. Louis, Missouri, pp. 1655-1660.

Byrne, R.J. (1994), "Design Issues with Strain-Softening Interfaces in Landfill Liners," Proc.
Waste Tech '94 Landfill Technology Conference,  National Solid Waste Management Association,
Charleston, South Carolina, 26 p.

Duncan, J.M, (1992), "State-of-the-Art: Static Stability  and Deformation  Analysis," Proc.
Stability and Performance of Slopes and Embankments - II, Vol. 1, pp. 222-266.

Earth Technology (1988), "In-Place Stability of Landfill  Slopes, Puente Hills Landfill, Los
Angeles, California,"  Report No.  88-614-1, The Earth Technology Corporation, Long Beach,
California.

EERC (1994), "Preliminary Report  on the Seismological and Engineering Aspects of the
January 17, 1994 Northridge Earthquake,"  Earthquake Engineering Research Center, Report
No. UCB/EERC-94-01, Berkeley, California.

Fassett, J.B., Leonards, G.A. and Repetto, P.C. (1994), "Geotechnical Properties of Municipal
Solid Wastes and Their Use in Landfill Design," Proc. WasteTech 94 - Landfill Technology
Conference, National Solid Waste Management  Association, Charleston, South Carolina, 32 p,

GeoSyntec Consultants (1993), "Report of Waste Discharge,  Eagle Mountain Landfill and
Recycling Center," Supplemental Volume 1,  Report prepared  for the  Mine Reclamation
Corporation, GeoSyntec Consultants, Huntington Beach, California.

Hynes, M.E.  and Franklin, A.G. (1984),  "Rationalizing the Seismic Coefficient Method,"
Miscellaneous Paper GL-84-13, U.S. Army Engineer Waterways Experiment Station, Vicksburg,
Mississippi, 34 p.
                                          110

-------
Jibson, R.W. (1993), "Predicting Earthquake-Induced Landslide Displacements using Newmark's
Sliding Block Analysis, Transportation Research Record, 1411, Transportation Research Board,
National Research Council, Washington, D.C., pp. 9-17,

Johnson, M.E., Lew, M., Lundy, J. and Ray, M.E. (1991), "Investigation of Sanitary Landfill
Performance During Strong Ground Motion from the Loma Prieta Earthquake of October 17,
1989," Proc. 2nd International Conference on Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, St. Louis, Missouri, pp. 1701-1708.

Kavazanjian, E., Jr., Hushmand, B., and Martin, G.R. (1991), "Frictional Base Isolation Using
a Layered Soil-Synthetic Liner System," Proc. 3rd U.S. Conference on Lifeline Earthquake
Engineering,  Technical Council on Lifeline Earthquake Engineering Monograph No. 4, Los
Angeles, California, pp.  1140-1151.
Kavazanjian, E., Jr., Matasovi, N., Bonaparte, R., and Schmertmann, G.R. (1995), "Evaluation
of MSW Properties for Seismic Analysis,"  Proceedings of the Geoenvironment 2000 Specialty
Conference, ASCE, Vol. 2, pp. 1126-1141, New Orleans, Louisiana,  24-26 February 1995.

Koerner, R.M. (1991), Designing with Geosynthetics, 2nd Edition, Prentice Hall, Englewood
Cliffs, New Jersey, 652 p.

Landva, A.O. and Clark, J.L, (1990), "Geotechnics of Waste Fill," In Geotechnics of Waste Fill -
Theory and Practice, ASTM STP 1070, pp. 86-103.

Makdisi, F.I.,  Seed, H.B. (1978), "Simplified Procedure for Estimating Dam and Embankment
Earthquake-Induced  Deformations," Journal of Geotechnical Engineering Division,  ASCE
Vol. 104,  No. GT7, pp. 849-867.

Marcuson, W.F. (1981),  "Earth Dams  and Stability  of Slopes Under  Dynamic Loads,"
Moderators Report, Proc. 1st International Conference on Recent Advances in Geotechnical
Earthquake Engineering and Soil Dynamics, St.  Louis, Missouri, Vol. 3,  pp. 1175.

Marcuson, W.F., III, Hynes, M.E., and Franklin, A.G. (1990), "Evaluation and Use of Residual
Strength in Seismic Safety Analysis of Embankments," Earthquake Spectra,  Vol. 6, No. 3, pp.
529-572.
                                         Ill

-------
Matasovi, N. (1991),  "Selection of Method for Seismic Slope Stability Analysis," Proc. 2nd
International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil
Dynamics, St. Louis, Missouri, Vol. 2, pp. 1057-1062.

Matasovi, N., Kavazanjian, E. Jr., Auguello, A.J., Bray, J.D., and Seed, R.B. (1995) "Solid
Waste Landfill Damage Caused by 17 January 1994 Earthquake," In: Woods, M.C. and Seiple,
R.W., Eds. The Northridge, California, Earthquake of 17 January  1994, California Department
of Conservation, Division of Mines and Geology Special Publication 116, Sacramento, California,
pp. 43-51.

Newmark, N.M. (1965),  "Effects of Earthquakes on Dams and Embankments," Geotechnique 15,
No. 2, pp. 139-160.

Orr, W.R. and Finch, M.O. (1990), "Solid Waste Landfill Performance During Loma Prieta
Earthquake," Geotechnics of Waste/ills: Theory and Practice, ASTM STP 1070, pp. 22-30.

Repetto, P.C.,  Bray,  J.D., Byrne, R.J. and Augello, A.J. (1993),  "Applicability of Wave
Propagation Methods to  the Seismic Analysis of Landfills," Proc.  Waste Tech '93, Marina Del
Rey, California, pp. 1.50-1.74.

Richardson, G.  and Reynolds, D.  (1991), "Geosynthetic Consideration  in  a  Landfill on
Compressible Clays,"  Proc. Geosynthetics '91, Industrial Fabrics  Association, Minneapolis,
Minnesota.

Seed, H.B. and Martin, G.R. (1966), "The Seismic Coefficient in Earth Dam Design," Journal
of Soil Mechanics and Foundations Division, ASCE, Vol. 92, No. SM 3, pp. 25-58.

Seed, H.B. (1979), "Considerations in the Earthquake-Resistant Design of Earth and Rockfill
Dams,"  Geotechnique, Vol. 29, No. 3, pp. 215-263.

Seed, R.B. and Boulanger, R.W. (1991), "Smooth HDPE-Clay  Liner Interface Shear Strengths:
Compaction Effects,"  Journal of Geotechnical Engineering,  ASCE, Vol.  117, No. 4,
pp. 686-693.

Seed, R.B., and Bonaparte, R.  (1992), "Seismic Analysis and Design  of Lined Waste Fills:
Current Practice," Proc. Stability and Performance of Slopes  and Embankments - II, Vol. 2,
ASCE Geotechnical Special Publication No. 31, Berkeley, California, pp. 1521-1545.
                                         112

-------
Sharma, H.D. and Goyal, H.K. (1991), "Performance of a Hazardous Waste and Sanitary Landfill
Subjected to Loma Prieta Earthquake," Proc. 2nd International Conference on Recent Advances
in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis, Missouri, pp. 1717-1725.

Siegel, R.A., Robertson, R.J., and Anderson, D.G. (1990), "Slope Stability Investigations at a
Landfill in Southern California," Geotechnics of Waste Fills-Theory and Practice, ASTM STP
1070, Arvid Landva, G., David Knowles, editors, American Society for Testing and Materials,
Philadelphia, Pennsylvania, pp. 259-284.

Singh, S.  and Murphy, B.J. (1990), "Evaluation  of  the  Stability  of  Sanitary Landfills,"
Geotechnics of Waste Fills - Theory and Practice, ASTM STP 1070, pp. 240-258.

Williams, N.D.,  and  Houlihan,  M.F. (1986),  "Evaluation  of Friction Coefficients Between
Geotextiles, Geomembranes,  and Related Products," Proc.  Third Conference on Geotextiles,
Vienna, Austria, pp. 891-896.

Williams, N.D., and Houlihan, M.F. (1987), "Evaluation of Interface Friction Properties Between
Geosynthetics and Soils,  "Proc. Geosynthetics '87, New Orleans, Louisiana, Vol. 2, pp. 616-627.
P-

Yegian, M.K., and Lahlaf, A.M. (1992), "Dynamic Interface Shear Strength Properties of
Geomembranes and Geotextiles," Journal of Geotechnical Engineering, ASCE, Vol. 118, No. 5,
pp. 760-779.
                                          113

-------
Table 6.1
REFERENCE
Cambell, 1982















Earth Technolgy,
1988
Franklin &
Assoc., 1990
Galante et
al., 1991

Ham, et al.,
1978
Landva et
al., 1984






Merz & Stone,
1962








Unit Weight Data for MSW (Fassett et al., 191
LOCATION WASTE WASTE PLACEMENT MOISTURl
TYPE AGE METHOD CONTENT
London

*

Northhampton

*
Sussex

Northhampton
*
•
"

•

Los Angelet,
CA
U.S.

S.6. PA


Madison,
Wl
Calgary
Edmonion
*
Misiissauga
Red Oeer
Vancouver
Waterloo
Winnipeg
Pomona,
CA
CelH

Cell 2

Cell 3

Cell 4

MSW

MSW

MSW

MSW
MSW

MSW

MSW
MSW

MSW

MSW

MSW

79% MSW Fresh
16% Sludge
:% Misc.
MSW Fresh









SoiliMSW
(by vol.)
0:1 Fresh

1 :6 Fresh

1 ;4.8 Frash

0:1 Fresh

Thin layers;
Cat 81 8
Sftims
Cat 81 6
6,8 ft lifts
L«1D Loader
'Minimal'
Pulverised
'Minimal'
snirfts
Cat 81 6

Thin layers;
Cat 81 6
5 nuns
Cat 955 Dozer
30.8
5J.3


Cat 126;
8-1 Oft lifts

Cat 930B 38.2
4 n lifts










(I) 1 laftlKt; 167,1
wetted to refusal .
(ll)44ftlHtt; 81, 8
Stand, Compaet'n
{ih) tame M (11) 32.S
but no H2O added
()v) 1 18 ft lift; 78,5
mln compaction
W).
? UNIT WilQHT TEST
' ""TOTAL DHr~ METHOD
(pet) (pet)
63 Full-Seal* Field Test

41

41

28 "
43

48 "

49 "
58

42 "

80,6 61,8 Plastic Tubes
82,6 93.3
30 Estimate

83-70 Lined Test Pit*


41 30 Surveyed

78-81 Test PHs
6441
5748
684S
7342
54-77
•642
54-73


58.6 22 Surveyed
(Refuse only)
45,3 29,8 Surveyed

36,8 29.1 Surveyed

22.7 12.6S Surveyed
(Refuse only)
RELIABILITY COMMENTS
Fair Do not have detail* on
measurement technique.
*

*

t
«

«

t
t

•

Good Depth about 50 a

Poor Estimate based on comp-
onent densities & relative vol.
Very Good


Good Density values are (or waste
only; no dally cover Included,
Good Testing don* In 1883-84,









Good

Good Compaction Is tub-
standard today
Good

Good


-------
Table 6.1:    (continued)
REFERENCE LOCATION WASTE WASTE PLACEMENT MOISTURE
TYPE AQE METHOD CONTENT
m
Calls 1:4.1 Fr«sh, mixed (V) ltd HIM; 41.7
with toll Stand, eompaet'n
Natarajan & Bombay, MSW
Rao, 1977 India
OweiS & NJ MSW Fresh
Khera, 1986 MSW 'Older'
Pacey, 1 982 MI. vi«w, MSW Fresh 51
CA ' 4T
43
35
41
35
Pfeffer, MSW Fresh M ,s « lifts n.t
1992 D-9 Dozer 35.4
53.2
Richardson & central MSW i -2 yrs
Reynolds, 1991 Maine
Sargunan et Madras. MSW 10-SO Uneompaeted 30-48
al., 1986 India
Schumaker, 1972 Poor
(From Oweis & Moderate
Khera, 1986) Good
Sharma et Richmond, MSW s f -40
al., 1990 CA Liquids
Siegel et s. CA MSW s-40yrs io-
-------
TABLE 6.2: COMPILATION OF THE AVAILABLE SHEAR STRENGTH DATA ON MSW
                    SOURCE: GEOSYNT1C (1993)
REFERENCE
Oweis (1985)
Oweis (1985)
Oweis (1985) ind
Dvirnoff and Munion (1986)
Fagotto and Rimotdi
(1987)
Earth Technology
(1988)
Landva and Clark
(1990)
Singh and Murphy
(1990)
Seigel et al.
(1990)
Richardson and Reynolds
(1991)
TYPE OF DATA
Laboratory tests on baled waste
Field load test at landfill in southern
California
Back calculation based on failure of a
landfill foundation on marsh clays and
silts
Back calculation from the results of
plate bearing tests
Small-scale triaxial tests on wastes
from LACSD Puente Hills landfill
Laboratory direct shear tests on
municipal solid waste
Laboratory tests performed by others
Small-scale direct shear tests on wastes
from OH
Large direct shear tests performed
hi situ
RESULTS
4> - 15" to 25°
c - 1400psf(67kPa)
 = 10" and c = 1000 psf (48 kPa)
to
$ - 26* and c = 0
-4-
* - 10" and c - 500 psf (24 kPa)
to
 - 30" and c - 0
* - 22* and c » 600 psf (29 kPa)
4 - 24" and c = 2,600 psf (124 kPa)
to
* - 35'
 - 24* and c = 450 psf (22 kPa)
to
4 - 39" and c = 400 psf (19 kPa)
0 =» 0 and c = 700 psf (34 kPa)
to
4 - 42s and c = 0
4 « 39" to * = 53°
$ = 18" to 43"and
c - 200 psf (10 kPa)
COMMENTS
No data on waste types, density, or test methods
is provided. Results correspond to a limiting
strain of 15 to 20%.
Values represent lower-bound strengths as the
slope did not fail. Results ire for relatively low
normal stresses. Assumed unit weight of
MSWis45pcf(7kN/mJ).
Values unreliable due to uncertainty in the back
analysis and strain incompatibility between waste
and foundation.
No data on waste types, test procedures, or
test results are provided.
Strength with cohesive component associated with
cover soil dominant waste; coheskmless strength
for refuse dominated waste.
Normal stresses up to about 10,000 psf
(480 kPa). Shear box size was approximately
11 in. (275 mm) by 17 to. (425 mm). Lower
value corresponds to shredded waste, not used.
Strength values bracket a range of results
compiled from work of various researchers and
consulting organizations.
Confining pressures from 2,000 to 12,000 psf
(100 to 570 kPa). Lower value of # is based
upon conservative interpretation.
Normal stresses from 200 to 800 psf (10 to
40 kPa). Unit weight of waste and cover sofl
estimated to be 96 pcf (15 kN/m3).

-------
                       TABLE 6.3: LOWER BOUND FRICTION ANGLES BACKFIGURED
                             FROM OBSERVATIONS OF STEEP LANDFILL SLOPES
                                         SOURCE:  Kavazanjian et al., (1995)
LANDFILL
(Location)
Lopez Canyon
(California)
Operating Industries, Inc.
(California)
Town of Babylon
(New YoA)
Private Facility
(Ohio)
AVERAGE SLOPE
Height,
ft(m)
400
(120)
250
(75)
90
(30)
130
(40)
Angle,
H:V
2.5:1
2:1
1.9:1
2:1
MAXIMUM SLOPE
Height,
ft(m)
120
(35)
75
(20)
45
(10)
35
(10)
Angle,
H:V
1.7:1
1.6:1
1.25:1
1.2:1
WASTE STRENGTH
(4, c - 100 psf)
FS = 1.0
25'
28°
30°
30°
FS » 1.1
27°
30°
34°
34=
FS = 1.2
29"
34°
38°
37'
Notes: (1) FS = Assumed factor of safety for back-analysis to estimate $• Back-analyses were performed assuming c = 100 psf (5 kPa).
     (2) Data for Lopez Canyon, Town of Babylon, and Private Facility Landfills obtained from GeoSyntec Consultants project files. Data for Operating
        Industries, toe. Landfill obtained from Siegel et al. (1990).

-------
                 One-Way Sliding
                 Only
       Positive Direction
                     Time
                                           Vg(t)
                                           Time
            Yield
            Acceleration,.
                                   Block
                                   Acceleration
                               iround
                             Acceleration
                                     Tune
                  Ground Velocity,
                                tl
                                        Block
                                        VelocSty.vm(t)
                                     Time
       0)

       1
       JB
       8-
       Q
                     ti
Ground
Displacement
          Ferm.
*	'---' =Oispl.
    Block
  x Displacement
                                     Time
Figure 6,1    Fundamental Principles of the Newmark Seismic Deformation Analysis
            (after Bray et al., 1994).

                                     118

-------
&
w

50
     75-
           This Study
            	Earth Technology v^,,v,^
            — — Extreme Values (Fassett et al.f 1994)
            iM Range of Typical Average Values
           cin«z,ct7plclllln: Derived from Fassett et al. (1994)
    100  I I  I  I  I  I | I  I  I  I  I |  I  I  I  l l  |  l  l  l l  l  |  I  I  I I I
         0          3          6          9         12        15
                      UNIT  WEIGHT  (kN/m3 )
       Figure 6.2  Unit Weight Profile for MSW (Kavazanjian et al., 1995)
                                119

-------
   250
£200
      0
+
O
0
A
   Richardson & Reynolds (1991)
   Lopez Canyon
   Operating Industries (On)
   Town of Babylon
   Private Facility in Ohio
   Fagotto & Rimoldi (1987)
   Landva le dark (1990)
                 24 kPa
IT I I  | « II I  f I T I I  |l  II I  J T I I
        0      50    100   150   200   250   300   350
                      NORMAL  STRESS  (kPa)
   Figure 6.3  Bi-Linear Shear Strength Envelope for MSW (Kavazanjian et al., 1995)


                              120

-------
                                         Friction Angle, 0 = 64
                Whittier Narrows
                earthquake,
                Oct. 1,1987
                   = 5.9
                                                                  = 0.47 g (peak)
                                                               (Garvey Reservoir)
                                                            Pasadena earthquake
                                                            Dec. 3,1988
                                                               = 5.0
                                                            k.,=0.22g(peak)
                                                            (Bedrock)
                                                         kh = 0.1 Og (peak)
                                                         (Top of Landfill)
                                 k,, represents horizontal ground acceleration.
                20
160
180
                                       Cohesion (kPa)
Figure 6.4    Yield Acceleration as a Function of Shear Strength Parameters for the OH
             Landfill (Siegel et al., 1990).
                                          121

-------
   1OOO
 «


 «
 o

 £.
 a.
 *

 o
10O
                    Yield Acceleration / Maximum  Acceleration
Figure 6.5   Hynes and Franklin Pennanent Seismic Displacement Chart (Hynes and

            Franklin, 1984).
                                       122

-------
   IOOO
                         Acceleration
                     Maximum average acceleration
                     Displacement (cm)
                     Rtehter magnitude
   O.oi
                                                           0.8
Figure 6.6    Makdisi and Seed Permanent Displacement Chart (Makdisi  and Seed
             1978).
                                        123

-------
        on
slllllllllllllll'. iill. I < » t l t. i
/ruuuuuuumm^uiu
mftvumiiunnur^um
uuiuumuiuuvrguiu
mmnum-immreruui
|u \ \ u u u u i m.i nu u in. u
 VHJ1UUIU I iUlN.lu|lUU

 11 in u ui u 11 u i mu|i\ 11
 luuutmummmnvu
 i u\u \\uhnunun.\\\i
 '\iuuuniumuiuuuv
  anauuuniuiiumu
            \
u\in\n\\uunu
'ununiuwfhui
   u-mu«uum
       '
           \uuiuu
           W\UUU
            III/

                     \
                 _.


                             *o
1
rt

                                   1
                             a
                             • VH«
                             Pu,
          124

-------
APPENDIX A LIQUEFACTION POTENTIAL






        EXAMPLE 1 • MSWLF on Level Site - Initial Screening



        EXAMPLE 2 - MSWLF on Level Site - Global Stability
                         125

-------
APPENDIX A EXAMPLE PROBLEM - LIQUEFACTION POTENTIAL
EXAMPLE 1 : This example evaluates the potential for liquefaction at the site of a proposed
above grade MSWLF. The subgrade is composed of fine sands and the water
table is near surface. The analysis neglects the additional normal stresses that
will result from the landfill itself. This is conservative.










•i

f
1
£ti xiiW t
iil| -1 	 •••'Hr* 	 '
i
- : - • , i
: " j
4 ' TP&ftfUf. C
	 	 "~1 	 ***=¥*".*
| 4 tJif^fiJ.y
4^^..
j 114 -sUd tk




1
: n
•i-M
"Dte^nC. |£J»»*»^* ***? ^»
-> -. i i ; '• »
_j • » ' '
~l


_.-

1 M *
1 ^11 ! ! I
srr^i 1 (
\o S 1 
£
j\j

'S

IP



^
^2*
|
-j*

! i \ \
i ; !
	 i
1 L-
. ! . --T

i \J
- ! ii
1 ^
>nl
" *CJ
^x'
M* f 54
ivV i 	
fSPT "{ . 1fl
f i
eJ«*e »c- _Q
"i i n  •
MJ! L-

- — -* — -
j
; ; ; ] ; ; ; }
-^, 'CO'*!"^ . — ^ -. L — '*- — .-"" t 1
^f ^ \ S : c^^uC i^VZ^S.. A^%,UKA
'• ;• J * ;: •: •: " -
•: i ! | * f : 1

*;$X
T-
»T«


•ace
i
1
d
5
1C
IS
2C
2S
3C
;
uj ^
|
f
I

i i
[ |
-nfV 1
-. *rd<
A

^=fi
aw*
«<;«
! I 1
es.1- K.C
,^j

«4a



Log of
Etjuipni
"| £7ewfi
(.win
1
1
H3
'"<
J'B
X
>-
.* *»*.
"'''+• '
f .** *
. "
* % *
:- ;• :
$
:)A.V
.* » * «
• --V'
.^"v"-"
•;,-,:

! j | i
"SI ------- | -----
£t !^Ct"%ol«4 i ' £>« 0N e;..!y|»fi
i ; i

AJT AAAIMi f^Af I^AIH^tM MAMft MM
IMA«H 4«w* to 4«tM, tiw«ijm in «e*n«,
wtur«t«d.
i«oomM 6ght orry-tfown.
fnd of taring «! 2C.5 f««t
j : - ! i j 1 : j j
1 ^PT -^ ^0 1 "* ] i
1 •s 1 I \ i ' '
i
.. _ ,t j_. 5._. A 	 i .___.,__. .._! _ >
1 M|| l||!
i i i I i i i t i \ {



















-
	
ZHJ
1
SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS Pg.4_ ot^
126

-------
APPENDIX A   EXAMPLE PROBLEM - LIQUEFACTION POTENTIAL
EXAMPLE 1:    The resistance of the existing •oil strata is evaluated using the Simplified
                 Procedure as outlined in Section S. On this page tie CSR capacity of the
                 existing subgrade is calculated.
                                          127

-------
  APPENDIX A   EXAMPLE PROBLEM - LIQUEFACTION POTENTIAL
  EXAMPLE 1:   The CSR generated by the design earthquake is evaluated. The acceleration at
                 the surface of the site is estimated using the general amplification/attenuation
                 relationships from Figure 4.2.
SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS
                                       128

-------
APPENDIX A  EXAMPLE PROBLEM - LIQUEFACTION POTENTIAL
EXAMPLE 1:    The Factor of Safety against liquefaction calculated for die two critical depths
                is greater than 1.0, so the site has limited liquefaction potential.
fe
SEISMIC DESIGN OF MUNICIPAL SOLED WASTE LANDFILLS
                                                                       g J: offh
                                        129

-------
APPENDIX A  EXAMPLE PROBLEM - LIQUEFACTION POTENTIAL
EXAMPLE 2:  A potentially liquefiable sand lay er is located approximately 30 feet below the landfill. Analyses
              nave shown that in the  case of liquefaction neither the integrity of overlying soil will be
              disrupted nor excessive settlement will occur.  However, there is a concern dial global stability
              of the landfill might be disrupted, as shown in the enclosed diagram.   Evaluate the post-
              earthquake global stability of the landfill assuming that the underlying sand layer has liquefied.
                                                                 I
                   1—r
          S
          5
                                         (jw»i) NOUVATO
SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS
                                               130

-------
4	1	1	i  [4-I-U444
                                1  M  I  I   lJ_lJJ_l-4-l
                              _.| ~~,j.,...,,.j—|.,_. 4..	j~~~|—j—-f-—|—|~—f-*
 The residual shear strength for the liquefied soil was estimated on the basis of the empirical relationship
 between corrected SPT blow counts and residual shear strength for sands developed by Seed et al. (1988),
 and reiterated by Marcuson et al. (1990).  Based on average corrected SPT blow count of 12 for the loose
 layer (liquefied zone), the relationship indicated lower bound residual strength of 200 psf.  The shear strength
 parameters for the underlying sand layer were therefore taken as friction angle of zero and a cohesion of 200
 psf for the post-earthquake stability assessment.  Slope stability analysis performed using the residual shear

 strength yielded a factor of safety of 1.1 which was deemed acceptable.
                     2000
                      1600
                   I
                   LJ
(/)

tt

Ul
I


o
Ul


crie« *»» JIIOINC e»»t
              5ft DAT* AttO DttieUAk irdtMTH MK»u[TtM Ntvc itCN [ITIHATCO.


              cowni)ueriA«.mMiet«Li4ut»cii«N >HO uiotne cut HIITOTIO.
      "0       4        8       12      16      20

          EQUIVALENT CLEAN SAND  SPT BLOWCOUNT.  (N,


-------
APPENDIX B SEISMIC SLOPE STABILITY






       EXAMPLE 1 - MSWLF on Ridge




       EXAMPLE 2 - MSWLF Above Grade




       EXAMPLE 3 - MSWLF on Soft Subgrade




       EXAMPLE 4 - MSWLF Displacement Analysis
                   132

-------
             B  EXAMPLE PROBLEM - Sffl§MtC $L6t»B frTAblU'l'Sf	
   EXAMPLE 1:       Tliis enunpfe nvtewi (he •osmk sta^^
                     subgnde stratigraphy it the site has approximately 10-ft of Mil profile over the rock
                     forming the ridge. Two important aaumptioos are made at &tt point l-4he peak
                     bedrock •coelenttoa actc oo the bottom of flMfioer, and 2- due to ft» ridge, die
                     •eismk coefficKot is assumed to equal (he pe*k bedn>ck acoelentio
                                                I	+-i	H-M-i
                                                !  .   I     i   !  !   !
SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LAJNDHLLS
                                          133

-------
1 APPENDIX B EX
EXAMPLE 1:











K.



i
AU






jtt*^
. t.--L •



j.»-



\
'.
\
i i ! .
MJnel MiVfL
j
i
3 * -
i
- T ' n.



i
i
j
t i
'.
; i ^-**
-i--*^-"""
— . • .
r ! i L


i


!
1
-i
-<
<
J
•

i
•
|
I

l I
' •• ?
i i

AMPLE PROBLEM - SEISMIC SLOPE ST>
The seismic stability of «*» unlinvi hnrffill U mnitt!
coefficients. T1*" ieTAnr ~.~4~i ;
MSWdueto
general analyi
subgrmdesho)
1 '! !
J: "jSTfc^iLil
1 :
. i
:
^ J- ~Ui
-r f=^«.aAV5%«
i i
! j^K
^S^rr
i i
; 1 t
! i S
I j i
\ p r J
; : j »
1 \ f

! * i I ! ill;!
PROFILE
2XAMPLE #1 RIDGE TOP J
J 5
). 1818. 50. 1820. 2
50. 1820. 850. 2050. 1
350. 2050. 1000. 2050.
LOOO. 2050. 1250. 1980
L250. 1980. 1450. 1980
50. 1820. 350. 1870. 2
350. 1870. 900. 1945.
900. 1945. 1250. 1980.
SOIL
2
;


~-


65. 65. 200. 20. 0. 0.
120. 120. 500. 25. 0.
EQUAKE JT^— v
0.11 0. 0. Co-l\ a ]
BLOCK ^ — ~S
25 2 50.
50. 1820. 350. 1870. .
351. 1870. 900. 1945.


i

J T : r
the hi^ber strengths
is wiB confirm this
ilti not l>c discount
{ j
j^ja/4 UNCA


s setup to force ft
of the lutonl s«L
VBHJ]
itedfor
e&flur
Is and d

d without analysis.
: l|j
rfire^
i i :
l •
^ i| 1 5

M^^Ti '1 i
ki-^4
is
•'h'^T
\ i
1 i 1
4° ! i '
j !
! 1 ; •
1 i !
] ! '
4 j j ! j
SITE W/O LI

1
• i J.ZJ
"? *~
. z 2
U) -,tfi»
2 ^ S -
2_ c
2 ii _ ic
ID *>^
\? •
M i
o 1^°-
o. o J~ —
in


.NEK
<
A_ __J__
r f
i \ •
1 \ s
l \


i
L

I"


nr
«n
B«
he*




i
'
**s%«
-— •


i

•— *
— —

mgeof
c&oeto
Monody







^^-







Beismic
bottom in the
.Whits*
miough the
j



\
"~ « «

1 | ;
I i
.T -t^-rl

I
-: *

• (
| ..$__
! !
i
{
i
i
i
J 	





J
i
1
1
)'

1 1


t
i


. • i • -i.\ - ; 	
i ! K !

j '

- t
i i
| >
1 l i





i
i
i
{
Ill't-' IIIHIlll
i
I 1






S
i



i


! ! \

>
'_!_
t ? 'X1
! ! *
:i .


i ' ::
; ; j
*» ) t 1 *h
0 _ ;
n ! h*»«Jr»l^-
.O i i i
»,.:$}» : f
! 5 : ! I 1 : i
S J ; i i i • i
; 1 i ; i :• s ; : i
it : '. ; • i }
I
.ps
I
1 i4
^.•L
> 1
'• \





Hft



SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS

<
! ? l
! ?
s, ! ? '

i
1^5
^
}
•>- 	 	 : 	 "
,j— ..- . >.,.._
i !
^


|
t i
.£> I
j
|
s _^
1 ,

























FS" VSZ.







	 _ —
i
Pg^ Of^
134

-------
APPENDIX B EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
EXAMPLE 1: The STABL model vaa modified to allow Che addition of 'sofl' types to represent the
liner system. In the example shown, two liner types are included so that textured and
smooth sheet could be incorporated at the same time-The shear strength of the liner
'soils' should be obtained from interface friction testing of the liner interfaces.






i
eUa»**
i i

|


™ H9o*
.._
4$£/!> _






	 -4
5


s ^ i
xe *f"4*ll
i
{
i
— r~ k*
1 f
^^^
^^TI


I i
ufn *TO\I
|
i
I

1
r 3-p

i ' i


j M i Ml! ;
jcn vg/ iW&e! •wfs-hsi'A 1 j
;---f -j ' ''• * \

\ ill! 3 M !
i i i \^^\ \-rr^

^**j i >*"} i__i— -*— — *~~-' — i * '
*— ^J""""":1 !;'*".' ; !
i i Mi; ; ; i i
I *• '* •'•***!
' j iQhra. - ' ': i : J * ^^U >
|| : i :- • - ": I
TA£»\~ :LW>M.T fcJe i •• l • I
~ PROFILE
"EXAMPLE 11 RIl
-11 5
_0. 1818. 50. 11
50. 1820. 850.
~850. 2050. 100<
-




-
-
LOOO. 2050, 12!
L250. 1980. 14.
50. 1819.5 350
350. 1870. 900
900. 1945. 125
50. 1819. 350.
350. 1869.5 90
900. 1944.5 12
4
65. 65. 200. 2
120. 120. 500.
60. 60. 0. 10.
60. 60. 0. 20.
0.055 0. 0. <(
BLOCK
25 2 50.
50. 1819.5 350
351. 1869.5 90
5GE TOP !
320. 2
2050. 1
3. 2050.
50. 1980
50. 1980
. 1870.
. 1945.
0. 1979.
1869.5
0. 1944.
50. 1979
0. 0. 0.
25. 0.
0. 0.0
0. 0. C
. 1870.
0. 1945,


^ITE W/ liXNc-K ; , -, i
1 1 ; \> : ? ^
[1^ i\ ! i :
l 5 { - \ = ;- ' ••
« ic i V ' •-'
- 1 Hi15! \ 1 ; -' r -
* > • • : \i i J
4« ;| j^. _; 	 , 	 iWV , \
i* ^ l i i 'l\ !
5 ^ 4JU ' ! !X-
2 -d * ' : '• s'~Ss
tj 2 'V '• - ' '• t
J £. tn ' '
CO "J ^ » I <
•5 2 i f .g ; ,.* ; Hi i H*
: 1 - • F/*"^ «*• SM=t
o M : •: i : s i i
o. o : i ; ;
"> 4 " .» • %" " ",
C ^^^ it- PMOT£ THA*^" TU^O MOC(^-
J i "I '; - J : " ! i 1"
: «J*« •^Ju>i»T^ Atwc fre>(.-ni«T«

v4«rt,^v«&c, ^^-(|) Aic
• W '. '' * 	 * ', •' • L^
0 . 'l t : • ; ; ^ t"
1 (


""**.
— ^
—










-



i
i
. j


i

i

i
*

_/;'=*; — r-j
j j i
!
| { i '
I 1 ^
j ;
| ,
i | i
_....





•

* •
|

:












l



1 '
1 :
k
*1* *$ '.


': —

i

i

"^Qf^
a^*(

SST
k
U:
JL-p
C3-
-------
APPENDIX B EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
  EXAMPLE 2:       This landfill is built upon a relatively Aalfcm layer irf--tiff b«*d Iffl tai^ ffart «to ^
                      surface acceleration can be assumed to equal 4* jxak bctock Mcetemtioa. The cite is
                      approximately level to that the seismic «jrffta»it b wrsmnod to e^
                      peak ground taiftoe acceleration or 0.125g.
                                              136

-------
APPENDIX B EXAMPLE PROBLEM- SEISMIC SLOPfi SfAWilTY
  EXAMPLE 2:     The STABL model used in this example is much more detailed than that used in
                  Example 1. Each layer of the liner is included in the model. The model can be
                  amplified by replacing die multiple layer liner with a angle layer having me lowest
                  interface friction properties. This complete model allows evaluation of the influence of
                  pone water (or leachate) pressures within individual layers on me overall stability.
  392.5 126,
  164.5 50.  174.5 50.
  174.5 50.  230.5 22.
  230.5 22.  850.
 PROFILE
 -EXAMPLE #2 CLAY FOUNDATION
 .21  6
 0.  35.  119.5 35.  5
 119.5 35. 158.5 48. 5
 158.5 48. 163.9 49.8 4
 163.9 49.8 164.5 50. 3
 164.5 50. 392.5 126. 2
             850. 126. 2
                       3
                       3
      5 22. 850.  25.1 3
 163.9 49.8 174.5 49.8  4
 174.5 49.8 230.5 21.8  4
-230.5 21.8 850. 24.9 4
 158.5 48. 174.5 48. 5
 174.5 48. 222.5 24. 5
 222.5 24. 232.5 19. 6
 232.5 19. 850.  22.1 6
 0.  24. 222.5  24. 6
 0.  17. 246.5  17. 7
 246.5 17. 850.  16. 7
-0.  14. 450.  12. 8
-450.  12.  850.  10.5 8
SOIL
8
110.  110.
65. 75.  300.
.1  .1 .0 10. 0.  0.
           1000.  0.  0,
             0. 0
1800. 0.  0.  0. 0
600. 0.  0.  0. 0
0. 32.  0.  0. 0
0. 34.  0.
                                                             0
 125. 130,
 135. 138.
 120. 120.
 130. 138.
 135. 140.
 SQUARE
 0.25 0.
 LIMITS
 3 3
 0. 8. 300.
 300. 8.  500
 500. 6.  850
 BLOCK
 50 3 40.
 174.5 49.95 174.5  49.95 0.
-230.5 21.9 230.5  21.9 0.
 325.0 22.37 425.  22.87  .8
                                                  6.
                                                  4.
SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS
                                         137

-------
APPENDIX B
EXAMPLE 3:
M ! -;








2.

'



Ot?
u*

-^Me^-







V

r i
i ..
I - ! -
i

!
1
;
|
!e<«MMHaoit4
''• <1<-*-s< !
EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
The global seismic analysis shows that the governing failure during an earthquake is
the result of a global failure. Increasing die global factor of safety may require
reduction of the landfill exterior slope or reduction of the waste height

	 ;
1 "' [
M
i


[_




\


i
%



1
:
i
i
i i !


«



A*



&>
I

^*
^
~***^

-r^

m






t-]-


>



! j
i i
! ..
f-».
!
\ i
1 «K. i ix^


	 * ^T\
^ \
i ; *
t:
3 :-
t - i
[ ' t
?
< : 1
;
!

r- $>.4gLsi
r T



I
!
i

[
I
^s*'\




> «

._.
< MM^M ! M •
M 1 M M i i


1 1 1

• |^0 SI'CT
I f !
i 1 i
i l i
1 i !


0

f ^ r
1 f i
< i t


'

1

j
i
I
I
t 1

i
i


Item
KW~,~,S-«>W.«
V&14Mfi& £lM
^A^rtL.tfvjlfe ^\J^( C»0*
f/VrVJElKlf CvrVf *JLff'ttr&
L **•**?€„
••;«hTr ;





1
I
;
i
i



{


\
1 \
t---
s
i


\ \
-•• i
T
i


*
i


^
(B
6c
1
— -—
i
! !
! j
] "- - - -
i
! !



i 	
j
1 1

;
PL" ' T V
f^f6^ 1 i
j j : S "^
i Mi




1
fad
1
j
h«
1
i
' \


T




e*4^
i .


 i
\ f








^renM
TftfcH

1
I
1 '
! ! t
MATERIAL ENGINEERING PROPERTIES
Moist Unit
Weight
(pel)
<*0
\0<3
-.

<5o

SEISMIC DESIGN
Saturated
Unit Weight
«_ Stress Analysis
(Long Term) *
Cohesion
(psO
^0
v»oe>
too
"ISO
o
i }


(.
Friction Angle
(Degrees)
*o*
o
o
e>
\o


\ 1 ^

USCS 217^3
























	







OF MUNICIPAL SOLED WASTti LANDhJL,LS PgjL_ ot±
138

-------
APPENDIX B EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
EXAMPLE 3: This landfill is ailed on a thick deposit of normally consolidated marine clays. The
static slope stability must evaluated the impact of the rate of waste placement, e.g. (he
marine days increase in strength as they consolidate under the weight of the MSWLJF.
\\l\\\ 11
.- ; j^tpKJV^^LAiBgi, O?T*&V
i i | i ; j

i 1 i ! 1 XJ&AIM
Mil
1 S4*4 3
;; 1 i *
i i i 1
44 j _u
i t i t
If':!
i • i i
1 ! S 1
I "V I • *" s^_
1 Ll_
i rr
? i : i ? f
	 PROFILE
EXAMPLE
9 3
0. 100.
160. IOC
— 325. 16C
_160. 10(
182. 82
160. 99
182. 81
— 0. 80. !
04.fl '
:: : :' '•
\
"ik*
: c
j
_\fr

e^p
i»3*»j j
i "] i

ft-ab
*
T*^
I
rp. i
c»el
!



•W


»iT
*i"<
!
«4e|£si
i i
< !
i -^\^
^^\ \
\ : <



] 1 ^
L J 	 J
i s
\ %
\ — p
$
s
Aitu^Ms^uteF



u_
f , i

t 1
4_



#3 STATIC STABILI'
160. 100. 3 1 g
). 325. 160. 1 f s
). 500. 160. 1 4-g
). 182. 82. 2 ic
500. 82. 2 ! i
182. 81. 3 ! g
. 500. 81. 3 tg
500. 80. 4 tl
500. 40. 5 f-j
i i ; ( i \ "\ f]
:- "' "• : ^ : • i
; : } ; j '• •


i c
i 2
1
;

1
|
j


! i
•ST^Jzi -t>Vt»f
—
^T^
— -|ir
se»y
i


k*
-•
•n;

i ! *
-V,






VUM
-Tl . T



FY
OI
0.
0.
00
0.
0.
,IK
.
:IE
0
.20







	 |
j

F i
~*~


F



i
2y£» ^"
*:

i

i
!
— i —
i
^ ^^
'i
\
i


: i ',
i

i4ftkdcBU-f4KK^
U<| j 1 \
^|_|_|...^_-
S I
Ml.l^lvO £U>*.
' ' jii— ^. •"-"'•
-U»M"i. s»>'"^
| i
»4e^J |_
i j
? f
i I


i ^ J_ /
1 1 {/
	 ! 1 *^j 	
^^x^ i
i ^
i ; {
|
i f !
; ; \
rSovfeu'U^ :
L = • •> -r .
60. 200. 30. 0. C
50. 0. 10. 0. 0.
. 100. 1000. 0. 0,
90.<400^5 0.0.0
90. 400T\0. 0. 0.
UTS /
0. 500. 0. 0.
CLE
10
. 300. 0. 500. 0.
\
i

«nk^ Jwg
1 J 1 *

a»46^e
r l'4b>:
"**4 <» __ 4*
j j -&->•
\ \
/ S
/; i i
/ ; j
I 1 |_
t~r
t , i
j
\
\
\
f *
i i ! 1
DJ=^ - ]
i ' :: i
. -;... : ....:. ..^.
). 0 —
0,,
0. 0
0
0
:«***.
10. 0. 0.




FSt-1.14-


FlS-Vft


















	
	






SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDHULS PgjL of±
139

-------
APPENDIX  B EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
  EXAMPLE 3     Hw static analysis indicates that the critic*! global failure surface passes through (he
                   soft marine clays and is significantly influenced by the isle of waste placement This
                   means 1f"*t (he waste must be placed at a rate that will allow (he consolidation
                   generated pore water pressures to dissipate.
  SOIL
  5
  60.  60.  200.  30.  0.  0.  0
  50.  50.  0. 10.  0.  0. 0
  100. 100. 1000. 0.  0.  0
  SO.  90.  400.  0. 0.  0.  0
            750.  0,
90.  90
EQUAKE
0.20 0
LIMITS
1  1
0.  100
BLOCK
50 3 10.
80. 100.  160.  100.  0.
160.5 99.5 182.  81.5  0.
182.5 81.5 500.  81.5  0.
            160.  100.
  EISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS
                                         140

-------
APPENDIX B EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
EXAMPLE 3: The pseudo-static seismic analysis indicates a block sliding fmctor-of-safety,FS, of 1.1
assuming  = 30 degrees in the MSW. The influence of me assumed MSW internal
angle of friction on the FS is shown below. The assumed MSW properties can
influence the FS as significantly as the seismic coefficient.

•
f
s fSfEvc










f

!
*
f
1 j



i

1
f s







;
i i
| I

i



—



» f . f i ! - '
iff ': •
MV

<2T,




...,...^e
>





.-tJUM
I 1
i Ic"
1 1
i 1
J
5
!
J 1
i -,
^\^
^J —

i 1\j
! ! t *
1 J i

f ! ;
'= i -.
5T*y%<





\
t-clv
\

—

•i, cio&fG
... jf.
	
5


i UK

^*""

; l



l"---.

>


J
	

[ 	


i


\
i
MM

i 1 M
; L * >** •
MM








	 V — -
: i
j j
J ;
J ;.
^£




; ^,
^^r" •;


Ill MM i 1 1 i
PROFILE
EXftMPLE #3 Seismic Analysis -
9 3
0. 100. 160. 100. 3
160. 100. 325. 160. 1
325. 160. 500. 160. 1 SOIL
160. 100. 182. 82. 2 5
n R? «•? «;nn R2 2 60. 60
4 :: ;
MM
• i
i T
i j
• i s
-4 !
J; |
1 !
] j
i i i
: » i '•
r<*Wliir*<£ -jf '
b-«i iy
— - i i /
• ', \/\
, 	 ; 	 __£. 	 , 	
]/\ \ \
*^i ! ! i
1 1 i
Mil
: t I >
1 1 i
1 ^ *
Longterm -j — |— 4—

f -
. 200. 30. 0. 0.







/
/
f






' •,
i i
1 1 i
1 !
} 	 f
i





! 1

. 1,4-.
•: t
\ \
i
|
i
. , J—
\ | |
"; > \
: \ I
j | i
i i
1 '- *
i i t
- —
i.. 1 L.
! 5 '
i t |
i :. 5
n
i 1 ;


i

i
!
l

\




fki^iuM
^




FS«O-A<
H£
•=3.













160. 99. 182. 81. 3
182. 81. 500. 81. 3
0. 80. 500. 80. 4
0. 40. 500. 40. 5
50. 50. 0. 10. 0.
100. 100. 1000. 0,
90. 90. 400. 0. 0.
90. 90. 750. 0. 0.
EQUAKE
0.20 0. 0.
LIMITS
1 1
0. 0. 500
CIRCLE
                                           0.  0
                                            0.  0.
                                            0.  0
                                            0.  0
0
                                    0.  0.
                      —r- 20 10
                         120.  300.  0. 500.
                                          0. 10
   0.
EISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS
                                 141

-------
APPENDIX B EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
EXAMPLE 4: MSWLF Displacement Analysis
_ ! M
Back£roun<
- An MSW la
. velocity of:
The weak L
interface. *
layer interfa
~ equal to k,
- MapMF-21
- The regiona
to be M 6,4
! i
1 i
\ \ i
1 \ j

-. i, .. ,| ,..| ,
• i i
1 ' !
j j

i ; :
•rt^

s
Cjlf -\r\A
i |- r
5-1 •
1 >

!; "•- -;
i < <
i ) i
! i !
i !
i 1 1
:| 5 ::

i
-_ p4

ndfill is founded
200 m/s (660 ft/:
lyer in the base i
lie weak layer ii
ce. Stability anal.
.10 g for bas
20) shows the M
L seismicity sourc
• _j _L_L J 	 ,
111
1 1 !
; ! \
i i 1
U i i
i 1 i
i | |
i I i
I * ^
j>4-"'l "! 	
**""j ' I 	 | MSv<
: ! 4
* J
/""llV :: :; :: ^"^


•• -i- 4 J \ /
: i i . > m
< i r* i ^
	 • |MHA *
: j 1 !
i j j ,•
< i ? '
: i i i
> I
: i ! i

i 1 { i i ; ! i
i !)'''! 1 '
._:_.. : 1 j ! = i i
i i
on 40 m (130 ft)
s). The landfill b
iner is anticipate*
i the cover is and
yses show yield ac
e liner and k, — <
[HA for a hypotb
e zone map (Fig.
of silt and clay st
as a geomembrai
1 to be at the filti
cipated to be at t
celcration (horizo
0.07 g for the co
etical bedrock ou
3 .2) reveals then
	 j 	 i. ,1 .
t 1
1 1 !
u i i •
II M
._^._j 	 ij 	 L 	 1 	 L

i i i
t ;

j

fLF 	 "~J- -"T"^
;
i ' ^
1 ''•' i
'^l*^^ i 1 '
ilii i
j _ v — ?nn
- ! ^
- 1 \ i
k * "^ | _,
0-12 g : i


i 1
j i
}

I !
/ f -

^J^xpj
: * ;


i 1 i
1 M !
i - i
! I !
	 j ) j -|
: 1 . !
lil with an averag
te in the base lin<
sx geoteirtile / op<
ie filter geotextili
ntal acceleration f
ver. Algennissen
tcrop at the site o
laximum earthqua
.... ...J_ j. 1 	 t
T
1 t j

I
f
*• 5
\
1 |
I
i i
j | .
	 —••••—}• • 	 j 	 ij"-" —-tojjil^

^^ j Bedroc
: \ '' \
Vt> 700


: •" =: '.
'-'•:*'.
'•Ill
; ! j
'• \ ' i
1 | ; {
: j j

;r and cover.
irations layer
E / vegetative
orFS= 1.0)
imap (USGS
ke magnitude

__, p_j |_
1 i !
> i i
i
	 	 	 1. 1
i *
I ; '
i 1 1
i t J



















J MHA « D 12 R
' /At
^J-': J^ ^
f

£ j | >
1 1 i
n/s ""1 	 \ 	
• .' I !
: 1 	 \ 	


::
,, , ( , \ \
| j i
i "\
\ i i
i i !







	








SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDFILLS Pgj_ of_2
142

-------
  APPENDIX B  EXAMPLE PROBLEM - SEISMIC SLOPE STABILITY
  EXAMPLE 4:   MSWLF Displacement Analysis
   Liner Stability Evaluation

   Step 1  Find the free-field peak ground acceleration (PGA) at the ground surface using Seed and Idriss
          (1982) charts.
          From Fig. 4.4b (or from Fig. 4.5), for MHA  = 0.12 g, PGA  = 0.23 g.

_ Step 2  Compare PGA to k,.
~        kyis less than PGA.  Therefore, a deformation analysis is required.

— Step 3  Estimate liner deformation (i.e., the maximum permanent displacement, u^J, as function of PGA
          and k, using Makdisi and Seed (1978) charts.
          Using upper bound of the M 6.5 area in Fig. 6.6, for k^ / PGA = 0.41 and M 6.5, a value for u^
          of approximately 20 cm (8 in.) is estimated.

   Step 4  Determine if the estimated v^. value is acceptable.
          The landfill is an area fill with essentially a  flat base  and  no penetrations through liner.  The
          weakest interface (interface for which k^ is  calculated) is between the operations layer and
          geotextile.  Therefore, based upon engineering judgement, the calculated deformation of 20 cm (8
          in.) is considered acceptable.
 - Cover Stability Evaluation

   Step 1  Find the maximum acceleration (a,,,,) at the top of the landfill using the Idriss (1990) chart.
           Using the recommended median relationship from Fig. 4.4b (or from Fig. 4.5) and the PGA of 0.23
           g from Step 1 of the liner stability evaluation, an a,,^ value of approximately 0.32 g is estimated
           at the top of the landfill.

   Step 2  Compare a,,,,, to k,.
           Since k^for cover is less than 0.5 • a,,^,  a seismic deformation analysis is required.

   Step 3  Estimate cover deformation as function of a,^ and k^ using Hynes and Franklin (1984) charts.
           Using mean + a curve from Fig. 6.5, for ky / a^ = 0.22, the u^ value of approximately 20 cm
           (8 in.) is estimated.
   Step 4  Determine if estimated u^ value is acceptable.
           The critical (lowest yield acceleration) surface in the cover is anticipated at the filter geotextile /
           vegetative layer interface on the side slopes. Since all vertical gas wells that penetrate the composite
           cover are on the horizontal deck, the estimated cover deformation of 20 cm (8 in.) is considered
           acceptable.
 SEISMIC DESIGN OF MUNICIPAL SOLID WASTE LANDblLLS                 Pg_2^
                                                      143

                                                     •U.S. GOVERNMENT PRINTING OFFICE: 1 995-650-006 / 2 20 1 6

-------
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268

Official Business
Penalty for Private Use
$300
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand comer.

!(you do not wish to receive these reports CHECK HERE D;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
      BULK RATE
POSTAGE & FEES PAID
          EPA
   PERMIT No. G-35
EPA/600/R-95/051

-------