December, 1990
CRITERIA FOR SELECTING MONITORING DEVICES AND INDICATOR
PARAMETERS FOR DIRECT PORE-LIQUID SAMPLING OF PETROLEUM
             HYDROCARBON CONTAMINATED SITES
                           by
        L.G.  Everett,  S.J.  Cullen,  R.G.  Fessler,
             Institute for Crustal  Studies,
             The University of California,
            Santa Barbara,  California  93106

                     D.W.  Dorranee,
             ENSR Engineering & Consulting,
                  3000 Richmond Ave.,
                 Houston, Texas  77098

                     L. G. Wilson,
      Department of  Hydrology and Water Resources,
               The University of Arizona,
                 Tucson,  Arizona  85721
                    Project Officer

                 Mr.  Lawrence A.  Eccles
          Advanced Monitoring Systems Division
      Environmental Monitoring Systems Laboratory
          U.S. Environmental Protection Agency
                Las Vegas, Nevada 89119
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S.  ENVIRONMENTAL PROTECTION AGENCY

-------
                                              December, 1990
CRITERIA FOR SELECTING MONITORING DEVICES AND INDICATOR
PARAMETERS FOR DIRECT PORE-LIQUID SAMPLING OF PETROLEUM
             HYDROCARBON CONTAMINATED SITES
                           by
        L.'G.  Everett,  S.J.  Cullen,  R.G.  Fessler,
             Institute for  Crustal  Studies,
             The University of California,
            Santa Barbara,  California  93106

                     D.W. Dorrance,
             ENSR Engineering & Consulting,
                  3000 Richmond Ave.,
                 Houston, Texas  77098

                     L. G. Wilson,
      Department of  Hydrology and Water Resources,
               The University of Arizona,
                 Tucson,  Arizona  85721
                    Project Officer

                 Mr.  Lawrence A.  Eccles
         Advanced Monitoring Systems Division
      Environmental Monitoring Systems Laboratory
         U.S.  Environmental  Protection Agency
                Las Vegas, Nevada 89119
          OFFICE OF RESEARCH AND DEVELOPMENT
         U.S.  ENVIRONMENTAL PROTECTION AGENCY

-------
                             NOTICE

   Although the research described in this article has been
supported by the United States Environmental Protection Agency
through cooperative agreement No. CR 81335069 to the University
of California, Institute of Crustal Studies, it has not been
subjected to Agency review and therefore does not necessarily
reflect the views of the Agency and no official endorsement
should be inferred.
                               -i-

-------
                             CONTENTS
INTRODUCTION  	  1

   The Vadose Zone	1
   Pore-Liquid Samplers  	  2
   Indicator Parameters  	  3

SELECTING PORE-LIQUID SAMPLERS  	  6

   In Situ Pore-Liquid Sampler Categories 	  6
   Criteria for Selection of Pore-Liquid Samplers 	  7
   Suction Samplers (Unsaturated/Saturated Sampling)  	  9
   Vacuum Lysimeters  	  14
   Pressure-Vacuum Lysimeters 	  16
   High-Pressure Vacuum Lysimeters  	  21
   Filter Tip Samplers	25
   Experimental Suction Samplers  (Unsaturated/Saturated
    Sampling)	27
   Cellulose-Acetate, Hollow-Fiber Samplers 	  27
   Membrane Filter Samplers  	  27
   Barrel Lysimeter 	  29
   Vacuum Plate Samplers  	  32
   Operational Constraints of Suction Samplers  	  32
   Experimental Absorption Samplers (Unsaturated/Saturated
   Sampling)  	34
   Sponge Samplers  	  40
   Ceramic Rod Samplers	41
   Problems with Experimental Absorption Samplers ......  41
   Free Drainage Samplers (Saturated Sampling)  	  41
   Pan Lysimeters	43
   Glass Block Lysimeters 	  43
   Caisson Lysimeters 	  46
   Wicking Soil Pore-Liquid Samplers  	  46
   Trough Lysimeters  	  49
   Vacuum Trough Lysimeters  	  51
   Sand-Filled Funnel Samplers  	  51
   Perched Ground-water Sampling  (Saturated Sampling) ....  53
   Point Samplers	54
   Cascading Water Samplers  	  58
   Drainage Samplers	58

SELECTING INDICATOR PARAMETERS FOR PORE-LIQUID SAMPLING OF
PETROLEUM HYDROCARBON CONTAMINATED SITES  	  62

   Chemistry of Petroleum Products	64
   Factors and Considerations that Affect the Selection of
   Indicator Parameters 	  69
   Fate and Transport Characteristics of Indicator Parameters  76
   Selection of Indicator Parameters  	  81

CONCLUSION	87

                              -ii-

-------
                         LIST OF FIGURES


Figure                                                      Page

  1       Hydraulic contact between saturated porous
          element of a vacuum lysimeter and soil	10

  2       Vacuum lysimeter (adapted from ref. 22)	15

  3       Pressure-vacuum lysimeter (adapted from ref. 22)   .   17

  4       Tube pressure-vacuum lysimeter (adapted from
          ref. 48)  	19

  5       Casing lysimeter (adapted from ref. 30)	20

  6       Modified pressure-vacuum lysimeter (adapted from
          ref. 49)	22

  7       Knighton-Streblow type vacuum lysimeter (adapted
          from ref. 50)	23

  8       High-pressure vacuum lysimeter (adapted from
          Ref. 22)  	24

  9       Filter tip sampler (adapted from ref. 51)	26

  10      Cellulose-acetate hollow-fiber sampler  	   28

  11      Membrane filter sampler (adapted from ref. 55)   .  .   30

  12      Barrel lysimeter (adapted from ref. 58)	31

  13      Vacuum plate sampler installation (adapted from
          ref. 62)  	33

  14      Example of pan lysimeter (adapted from ref.  80)  ..   44

  15      Glass block lysimeter (adapted from ref. 80)  ...   45

  16      Example of caisson lysimeter (adapted from ref.  73)   47

  17      Wicking type soil-pore liquid sampler (adapted
          from ref. 58)	48

  18      Trough lysimeter (adapted from ref. 75)	50

  19      Sand filled funnel sampler installation (adapted from
          ref. 83)	52

  20      Examples of point sampling systems (adapted from
          ref. 83)  	55


                              -iii-

-------
                   LIST OF FIGURES (Continued)


Figure                                                      Page

  21      A monitoring well with the uppermost ground-water
          level intersecting the slotted well screen
          (adapted from ref. 94)	56

  22      A monitoring well installed to sample from the
          lower of two groundwater zones (adapted from
          ref. 94	57

  23      An open-hole ground-water monitoring well in rock
          (adapted from ref. 94)  	59

  24      Conceptualized cross section of a well showing
          cascading water from perched zone (adapted from
          ref. 78)  	60
                               -iv-

-------
                          LIST  OF  TABLES


Table                                                       Page

  1       Criteria for selecting pore-liquid samplers 	  8

  2       Suction sampler summary 	  13

  3a      Porous material interactions  	  35

  3b      References and notes on experimental techniques .  .  39

  4       Selected physical and chemical properties for major
          components found in a typical gasoline mixture  .  .  68
                               -v-

-------
                           INTRODUCTION

In recent years, there have been increasing requirements for
vadose zone monitoring.

Vadose zone monitoring devices in use include a variety of in
situ samplers to collect pore-liquids under saturated or
unsaturated conditions.  Pore-liquid samples containing petroleum
hydrocarbons are generally considered as direct evidence of
contaminant migration; pore-liquid sampling is required by
regulation for land treatment units.  This report describes these
samplers together with their advantages and disadvantages.  It
also describes the application of indicator parameters and tracer
chemicals to monitoring petroleum hydrocarbon contaminated soils.
THE VADOSE ZONE

The vadose zone is the hydrogeological region extending from the
ground surface to the principle water table.  Other commonly used
terms for this region are the  "unsaturated zone" and the "zone of
aeration."  These other terms  do not suggest the existence of
locally saturated conditions above the principle water table.
Saturated or near-saturated conditions can develop when pore-
liquids collect on low permeability lenses that are often called
perching layers.  Most pore-liquid flow through the vadose zone
is under unsaturated conditions and is primarily controlled by

                               -1-

-------
negative pore-liquid pressure gradients



(negative pore-liquid pressures are referred to as pore-liquid



tensions).







Chemical species released at or near the land surface generally



migrate to some degree through the vadose zone.  Analyses of



vadose zone liquids and gases can provide an early warning of



potential ground-water pollution from such releases.  This early



warning can provide a means to mitigate potential problems prior



to ground-water degradation (1,2)







PORE-LIQUID SAMPLERS







Vadose zone liquids from soil samples are usually extracted in



the laboratory.  Alternatively, the mobile pore-liquids may be



sampled directly and repeatedly from "undisturbed" soils using



permanently installed in situ pore-liquid samplers.  The most



obvious difference between these two techniques is that soil



sampling is a destructive process which prevents repetitive



sampling from the same location.  More importantly, the two



techniques do not sample the same types of liquid (3,4).  In situ



samplers are capable only of sampling pore-liquids held at



tensions of up to about 60 centibars (cb) (5).  Soil sampling



with subsequent pore-liquid extraction provides liquids which may



be held at tensions of up to several bars, depending on the



extraction technique.  Extraction under several bars of pressure





                               -2-

-------
may strip off cations preferentially sorbed in electrical double
layers, sorbed organics, and even components of the soil.  These
species may not be present in the same concentrations (absolute
or relative) in samples provided by in situ pore-liquid samplers.
In situ pore-liquid samplers are "point samplers" and can be
effectively used to indicate relative changes in the amount of
solute flux.  Caution must be exercised when quantitative results
are desired since the variability of these measurements must be
appropriately established.

INDICATOR PARAMETERS

The concept of using indicator parameters to identify complex
systems with simple but representative elements has been in
practice for a long time.  For example, pH measurements of water
samples can indicate the presence of an acid or a base while
specific conductance measurements can indicate the presence of
ionized salts in a water sample.  More recently, phenanthrene has
been used as a indicator for the presence of creosote
contamination in investigations at past and current sites of wood
treating facilities.  It is the single most abundant chemical in
creosote and the most water soluble, and, therefore, the most
mobile.

The indicator parameter itself can vary depending on the needs or
goals of a particular program.  As an example, in California,

                               -3-

-------
when a  site  is  suspected to be contaminated, the initial site



investigation typically samples and analyzes the site soils for



ten or  more  Title  22  (California Code of Regulations) metals.



These metals are then used as indicator parameters to determine



if the  site  soils  are contaminated.  If, as a consequence of



these findings, remedial action (particularly excavation and



removal), is found to be warranted, then as few as three of these



metals  can be used as indicator parameters for evaluating



potential work  health and safety hazards and for identifying



requirements that  would assure adequate worker protection.







Indicator parameters  have been powerful tools for assisting in



the measuring of complex systems using simple techniques.  It is,



therefore, natural to identify their potential usefulness for



monitoring recently recognized environmental issues, such as the



complex petroleum  hydrocarbon contaminated sites.







Reviewing the definitions of the terms "indicator" and



"parameter"  in  the precise dictionary sense will assist in



understanding the  factors which must be considered in selecting a



chemical or  group  of  chemicals that can properly be termed



"indicator parameters."  Webster's defines an "indicator" as a



substance that  is  so  strictly associated with particular



conditions that its presence is a direct indication of the



presence of  these  conditions.  The term "parameter" is defined as



a characteristic element or constant factor which describes a





                               -4-

-------
particular population.  Thus, an indicator parameter is a
substance, a chemical or group of chemicals, with characteristics
indicative of the presence of a particular population.  In simple
terms, where the indicator parameters are found, then a
particular population is known to be present.  These definitions
help to clarify what determines an indicator parameter.  However,
numerous factors must be considered in selecting indicator
parameters appropriate for direct pore-liquid sampling
applications.

This report reviews various in situ samplers and includes
relevant literature citations.  Some of the described samplers
are not commercially available at this time.  However, they may
have been available in the past and may be found at sites with
established vadose zone monitoring programs.  Some of the
samplers can be fabricated.  There are numerous qualifiers,
hints, and warnings which should accompany the description of
each sampler.  We depend on the reader to review cited references
to obtain complete descriptions of the covered samplers.  The
applications and limitations of many of the samplers presented
here were described in a series of articles by Everett (6),
Everett et al. (7), Wilson (7), Wilson (8), Everett et al. (9),
and Everett and McMillion (5).  This report extends and updates
these articles.
                               -5-

-------
                  SELECTING PORE-LIQUID SAMPLERS

This report also  presents  a discussion of factors that affect
selection of indicator parameters and tracer chemicals.  The
concepts developed  and explained are intended to aid in selecting
and identifying chemicals  that can be used as indicator
parameters for monitoring  petroleum hydrocarbon contaminated
sites.

IN SITU PORE-LIQUID SAMPLER CATEGORIES

In situ samplers  extract liquids from saturated and unsaturated
soils.  Most samplers designed to sample from unsaturated soils
also sample from  saturated soils.  This is useful in areas where
the water table fluctuates, resulting in alternating saturated
and unsaturated conditions.  In contrast, samplers designed for
sampling from saturated soils cannot be used in unsaturated
conditions.  This is because the negative pore-liquid pressures
in unsaturated soils prevent liquid from moving into air-filled
cavities at atmospheric pressures (Richard's Outflow Principle).
Also, the openings  in saturated samplers are too large to prevent
air from entering the samplers when suctions are applied.  Using
this distinction, the types of pore-liquid samplers have been
categorized as follows:
                               -6-

-------
          suction samplers  (unsaturated/saturated sampling)







          experimental suction samplers  (unsaturated/saturated



          sampling)







          experimental absorption samplers  (unsaturated/saturated



          sampling)







          free drainage samplers  (saturated sampling)







          perched ground-water samplers  (saturated sampling).







The term "pore-liquid" could be applicable to any liquid residing



in porous media, both saturated and unsaturated, ranging from



aqueous pore-liquids to oil.  However, all of the samplers



described in this report were designed to sample aqueous pore-



liquids only.  The abilities of these samplers to collect other



pore-liquids may be quite different than those described.







CRITERIA FOR SELECTION OF PORE-LIQUID SAMPLERS







The choice of appropriate sampling devices for a particular



location is dependent on various criteria (see Table 1). Well-



structured soils have two distinct flow regions including



macropores (e.g. interpedal openings, cracks, burrows, and root



traces) and micropores (e.g. interpedal openings between soil





                               -7-

-------
grains).  Under saturated conditions, liquids move more rapidly

through macropores than through micropores.  Because of this,

contaminants transported by free drainage may bypass the finer

pores.  Consequently, pore-liquids in macropores may have

different chemistries than those in micropores (10).  This

difference can be attributed to the fact that oxygen contents of

macropores can change in a matter of hours during an infiltration

event, whereas micropores may remain suboxic regardless of flow

conditions (11).  In addition, micropores are less susceptible to

leaching than macropores (2, 12, 13, 14).  Because of these

differences, sample chemistry can vary widely from location to

location and from time to time depending on the amount of liquid

drawn from these two flow systems.  Therefore, it is prudent to

consider using both unsaturated and free drainage samplers in a

sampling program, depending on site characteristics.



      Table 1.  Criteria for Selecting Pore-Liquid Samplers
      1.  Required sampling depths
      2.  Required sampling volumes
      3.  Soil characteristics
      4.  Chemistry and biology of the liquids to be sampled
      5.  Moisture flow regimes
      6.  Required durability of the samplers
      7.  Required reliability of the samplers
      8.  Climate
      9.  Installation requirements of the samplers
     10.  Operational requirements of the samplers
     11.  Commercial availability
     12.  Costs
                               -8-

-------
Specific guidelines for designing vadose zone pore-liquid
monitoring networks have been discussed by Wilson  (1), Wilson
(15), Everett  (7), Wilson  (8), Everett et al. (7), Morrison (2),
Wilson  (16), Everett et al.  (9), Robbins and Gemmel  (17), Merry
and Palmer (18), U.S. Environmental Protection Agency (19), and
Ball and Coley  (20).
SUCTION SAMPLERS  (UNSATURATED/SATURATED SAMPLING)

Table 2 presents  suction samplers and some of their operational
constraints.  In  general, a suction sampler consists of a hollow,
porous section attached to a sample vessel or a body tube (see
Figure 1).  Samples are obtained by applying a vacuum within the
sampler and collecting pore-liquid in the body tube.  Samples are
retrieved by a variety of methods.

The principles of suction sampler operation are as follows.
Unsaturated portions of the vadose zone consist of
interconnecting soil particles, interconnecting air spaces, and
interconnecting liquid films.  Liquid films in the soil provide
hydraulic contact between the saturated porous section of the
sampler and the soil (see Figure 1).  When a vacuum greater than
the pore-liquid tension is applied within the sampler, a
pressure-potential gradient is created toward the sampler.  If
meniscuses of the liquid in the porous segment are able to

                               -9-

-------
POROUS
SECTION
          PORE-LIQUID
        MOVING TOWARDS
           SAMPLER
                                          SATURATED
                                            POROUS
                                           MATERIAL   SOIL
                                                    PARTICLE

                                                       /  SOIL
                                                          GAS
                            ACCUMULATED
                                  SAMPLE
                                                  PORE-LIQUID
                                                 UNDER TENSION
      Figure l.   Hydraulic contact between saturated porous
                 element of a vacuum lysimeter and soil
                               -10-

-------
withstand the applied suction, liquid moves into the sampler.
The ability of the meniscuses to withstand a suction decreases
with increasing pore size and also with increasing hydrophobicity
of the porous segment.  This relationship is defined by the
capillary rise equation (21).  If the maximum pore sizes are too
large, and/or they are hydrophobic, the meniscuses are not able
to withstand the applied suction.  As a result, they break down,
hydraulic contact is lost, and only air enters the sampler.

The ability of a sampler to withstand applied suctions is gaged
by its bubbling pressure (19, 22).  The bubbling pressure is
measured by saturating the porous segment, immersing it in water,
and pressurizing the inside of the porous segment with air.  The
pressure at which air starts bubbling through the porous segment
into the surrounding water is the bubbling pressure.  The
magnitude of the bubbling pressure is equal to the magnitude of
the maximum suction which can be applied to the sampler before
air entry occurs (see air entry values in Table 2).  Because the
bubbling pressure is a direct measure of how a sampler will
perform, it is more useful than measurement of pore size
distributions.

As soil pore-liquid tensions increase (low soil-liquid contents),
pressure gradients toward the sampler decrease.  Also, the soil
hydraulic conductivity decreases exponentially.  These result in
increasingly lower flow rates into the sampler.  At pore-liquid

                               -11-

-------
tensions above  about  60 cb  (for coarse grained soils) to 80 cb



(for fine-grained  soils), the  flow rates are effectively zero and



samples cannot  be  collected (Everett and McMillion)  (5).



Samplers which  have air entry  values exceeding the 60-80 cb range



are preferred  (see Table 2).







Porous samplers with  the appropriate air entry values suitable



for pore liquid sampling are typically hydrophilic.  In



situations where aqueous and nonaqueous liquid phases are



present, water  will be preferentially drawn into  (along with



dissolved components) the pores of the porous element because of



its polarity and relatively high  surface tension.  Field



collection of nonaqueous phase organic liquids using water



saturated suction  samplers  is  not possible.  However, suction



samplers can provide  representative samples- of the aqueous phase,



including dissolved components.







As demonstrated by Neary and Tomassini (23), new  samplers may be



contaminated with  water-soluble cations during manufacturing.  In



order to reduce chemical interferences from these and other



substances on the  porous sections, a variety of pre-installation



procedures have been  developed, including acid flushing (24, 25,



26, 27, 28, 29) Debyle et al.  (27) recommend discarding the first



one or two sample  volumes when sampling dilute solutions with



newly acid-flushed, installed  samplers.  This allows cation



exchange between the  porous segment and the





                               -12-

-------
    TABLE 2. SUCTION SAMPLER SUMMARY
Sampler
Type



Porous
Section
Material


Max.
Pore
Size
( (xm)

Air
Entry
Value
(kPa)

Wetting
Charac-
ter-
istic

Opera-
tional
Suction
Range
(kPa)
Max.
Opera-
tional
Depth
(m)
Commercially Available Suction Samplers
Vacuum
Lysi-
meters

Pressure
Vacuum
Lysi-
meters
High-
Pressure
Vacuum
Lysi-
meters
Filter
Tip
Samplers
Ceramic
PTFE
Stainless
Steel
Ceramic
PTFE
Stainless
Steel
Ceramic
PTFE
Stainless
Steel

Ceramic


1.2-3.0
15-30
7

1.2-3.0
15-30
7

1.2-3.0
15-30
7


1.2-3.0


>100
5-10
20

>100
5-10
20

>100
5-10
20


>100


HL
HB
HL

HL
HB
HL

HL
HB
HL


HL


<60-80
<5-10
<20

<60-80
<5-10
<20

<60-80
<5-10
<20


NA


<7.5
<7.5
<7.5

<15
<15
<15

<90
<90
<90


Unli-
mited

      Experimental Suction Samplers
Cellu-
lose
Acetate
Hollow
Fiber
Samplers
Membrane
Filter
Samplers
Vacuum
Plate
Samplers


Cellulose
Acetate

Noncell-
ulose
Polymer
Cellulose
Acetate
PTFE
Alundum
Ceramic
Fritted
Glass
Stainless
Steel
<2.8


<2.8


<2.8

PTFE
7
1.2-3.0
4-5.5
7

>100


>100


>100

5-10
20
>100
50
20

HL


HB


HL

HB
HL
HL
HL
HL

<60-80


<60-80


<60-80

NA
<20
<60-80
50
<20

<7.5


<7.5


<7.5

<7.5
<7.5
<7.5
<7.5
<7.5

NA = Not Available
HB = Hydrophobic
HL = Hydrophilic
                     13

-------
pore-liquid to equilibrate  following acid flushing.  Sintered

stainless steel samplers used in virus studies are chlorinated

and rinsed with a  10%  solution of sodium thiosulfate to

neutralize free chlorine  (30).  Other pre-installation procedures

(e.g. pressure testing) are described by Everett and McMillion

(5) and Timco  (31).



Suction sampler installation and sampling procedures are

described by U.S.  Environmental Protection Agency  (19),

Soilmoisture (22), Timco  (31), Linden (32), Rhoades and Oster

(33) Klute (34), Brose et al. (35), Morrison  (2), Cole et al.

(36), Wengel and Griffen  (37), Brown et al. (38), and Chow (39).



Vacuum Lysimeters1



Vacuum lysimeters  generally consist of a porous cup mounted on

the end of a tube, similar  to a tensiometer (see Figure 2).  A

stopper is inserted into the upper end of the body tube and

fastened in the same manner as the porous cup or, in the case of

rubber stoppers, inserted tightly (19).



A variety of materials have been used for the porous segment

including nylon mesh (40),  fritted glass (41), sintered glass

(42), Alundum®, stainless steel (43), polytetrafluorethylene
     *A device used to measure the flux of water within a soil
monolith, usually undisturbed; or to collect percolating water
for analyses.

                               -14-

-------
STOPPER —•;
                              HAND
                              SUCTION
                              PUMP
              SUCTION LINE
                   WATER
                  SAMPLE
VACUUM
     SAMPLE
     FLASK
                   POROUS
                   CUP
    Fgura 2. Vacuum Lysimeter (Courtesy, Soilmoistura Equipment Corp.. 1981).


                     -15-

-------
 (PTFE)  (31), and ceramics  (22).  The sampler body tube has been
made with PVC, ABS, acrylic, stainless steel (44) and PTFE (31).
The stopper is typically made of rubber  (19), neoprene, or PTFE.
The outlet lines are commonly polycarbonate, PTFE, rubber,
polyethylene, polypropylene, Tygon®, nylon, stainless steel, and
historically, copper.  Fittings and valves are available in
brass,  stainless steel, PVC, and PTFE.

Vacuum  lysimeters transfer samples directly to the surface via a
suction line.  Because the maximum suction lift of water is about
7.5 m,  these samplers cannot be operated below this depth.  In
practice, suction lifts of even 7.5 m may be difficult to attain.

PRESSURE-VACUUM LYSIMETERS

These samplers, depicted in Figure 3, were developed by Parizek
and Lane (45) for sampling pollutants moving in the vadose zone
beyond the reach of vacuum lysimeters.  Again, the porous segment
is usually a porous cup at the bottom of a body tube.-  Two lines
are forced through a two-hole stopper sealed into the upper end
of the body tube.  The discharge line extends to the base of the
sampler and the pressure-vacuum line terminates a short distance
below the stopper.  At the surface, the discharge line connects
to a sample bottle and the pressure-vacuum line connects to a
pressure-vacuum pump.  The sampler and its components are
commonly made out of the same materials used for vacuum

                               -16-

-------
                                    PINCH
                                   CLAMPS
czrr
1
g;=
i /
L^~\
/ «, ^
PRESSURE-VACUUM <=* *"
HAND PUMP <=>
v

BODY— ^— -
TUBE0 «

o
cC>
0 *
«^_ BACKFILL
Q *> MATERIAL
c?
c>
<»

Rgure 3.  Pressure-Vacuum Lysimeter (Courtesy. Soilmoisture Equipment Corp.. 1989).


                         -17-

-------
lysimeters.  Pressure-vacuum lysimeters first collect pore-liquid
in the  body tube  by  application  of vacuum through the pressure-
vacuum  line.  The sample  is  then retrieved by pressurizing the
sampler through the  same  line; this pushes the sample up to the
surface through the  discharge line  (see Figure 3).

Because samples are  retrieved under pressure, these samplers can
be used below 7.5 m.   However, when positive pressure is applied
for sample retrieval,  some of the sample may be forced back out
of the  cup.  At depths of over about  15 m, the volume of sample
lost in this manner  may be significant.  In addition, pressures
required to bring the  sample to  the surface from depths greater
than 15 m may be  high  enough to  damage the cup or to reduce its
hydraulic contact with the soil  (46,  47).  Rapid pressurization
causes  similar problems.  Morrison and Tsai (48) developed a tube
lysimeter with the porous section located midway up the body tube
instead of at the bottom  (see Figure  4).  This design mitigates
the problem of sample  being  forced back through the cup.
However, it does  not prevent problems with porous segment damage
due to  over pressurization or rapid pressurization.  The sleeve
lysimeter (which  is  not presently available commercially) was a
modification of this design  for  use with a monitoring well (2).
Another modification is the  casing lysimeter which consists of
several tube lysimeters threaded into one unit (see Figure 5).
This arrangement  allows precise  spacing between units (31).
                               -18-

-------
   PRESSURE-
 VACUUM LINE
n
 PVC BODY
     TUBE
THREADED.
 COUPLING
i==g
u
»





i
\


i

i
i
\
!

Y
*
\
t
J
«
^
THREADED
COUPLING
DISCHARGE
LINE



^ 	 POROUS
SECTION
n\tr* \mv?\ i
                               POINT
  Figure 4. Tube Prassura-Vacuum Lysimetar (Morrison and Tsai, 1981).
                 -19-

-------
                                     FLUSH
                                  THREADED
                                    CASING
         DISCHARGE
           LINES
                              PRESSURE-
                                VACUUM
                                   LINES
   GROUND
  SURFACE
BENTONITE
     SEAL
    SILICA
     PACK
BENTONITE
     SEAL
    SILICA
    PACK
                    LYSIMETER
                                    BLANK
                                     TUBE
                                   POROUS
                                   SECTION
BLANK j
 TUBE^
                                           Vi
                                   FLUSH
                               THREADED
                                  CASING
               Rgure 5 Casing Lysimeler (Courtesy. Timco Mfg., Inc., 1989)
                         -20-

-------
Nightingale et al.  (49) described a design which allows incoming
samples to flow into a chamber not in contact with the basal,
porous ceramic cup  (see Figure 6).  The ceramic cup is wedged
into the body tube without adhesives or threading.  The sampler
was used to sample the vadose zone, the capillary fringe and the
fluctuating water table in a recharge area.  Knighton and
Streblow (50) reported a sampler with the porous cup mounted on
the top of a chamber (see Figure 7).  These designs also allow
pressurization for sample retrieval without significant liquid
loss.  However, because the porous cups are exposed to pressure,
possible damage due to over pressurization or rapid
pressurization remains a problem.

HIGH-PRESSURE VACUUM LYSIMETERS
High-pressure vacuum lysimeters operate in the same manner as
pressure-vacuum lysimeters.  However, they include one-way check
valves and a transfer vessel or chamber between the sampler and
the surface (see Figure 8).  These accessories prevent sample
loss through the porous section during pressurization, and
possible cup damage due to over pressurization.  The samplers are
manufactured generally using the same materials as vacuum
lysimeters (22, 31).
                               -21-

-------
  PRESSURE-
    VACUUM
        LINE
SAMPLE
DISCHARGE
LINE
                    =	VALVE
             no
           ;•.>.-:••••>


          *•••' "•"•»:•;•>•*•'••••*
                          SAMPLE
                          RESERVOIR
                          STANDPIPE
                        POROUS
                        CERAMIC
                        CUP
Figure 6. Modified pressure-vacuum lysimeter (Nightingale, et al, 1985).
                 -22-

-------
   PRESSURE-
     VACUUM
         LINE
 RUBBER
STOPPER
SAMPLE
DISCHARGE
LINE
          POROUS
    «	  CERAMIC
          CUP
                                       TUBE
                                       BODY
 Figure 7. Knighton and Streblow-type vacuum tysimeter (Knighton and Streblow, 1981).
                   -23-

-------
    SOLID PLUG
         CHECK
         VALVE
     DISCHARGE
           LINE
         CHECK
         VALVE

         SOLID
          PLUG
PRESSURE-
VACUUM
LINE

PRESSURE
SEAL
                              TRANSFER
                              CHAMBER
"O"RING
PRESSURE
SEAL
                              POROUS
                              CUP
Rgura 8 High Pressure-Vacuum Lysimeter (Courtesy. Soilmoisture Equipment Corp., 1989)
                     -24-

-------
FILTER TIP SAMPLERS

Filter tip samplers consist of two components:  a permanently
installed filter tip, and a mechanically-retrievable glass sample
vial  (see Figure 9).  The filter tip includes a pointed end to
help with installation, a porous section, a nozzle, and a septum.
The tip is threaded onto riser pipes which terminate at the
surface.  The sample vial includes a second septum.  When in use,
the vial is seated in an adaptor which includes a disposable
hypodermic needle to penetrate both the septa, allowing sample to
flow  from the porous segment into the vial.

The body of the filter tip is constructed from a variety of
materials, including thermoplastic, stainless steel, or brass.
The attached porous section is available in high density
polyethylene, porous ceramic, or sintered stainless steel.  The
septum is made of natural rubber, nitrile rubber, or fluororubber
(51, 52).

A sample is collected from a filter tip sampler by lowering an
evacuated sample vial down the access tube to the porous tip.
The vial is coupled with the porous tip via the hypodermic needle
and sample flows through the porous section into the vial.  Once
full, the vial is mechanically retrieved (see Figure 9).
                               -25-

-------
FILTER /
TIP \
X
^^_
Wm
1

f.ff
	
-^
+ 	 RISER
PIPE
SAMPLE
VIAL
— SEPTUM
DOUBLE ENDED
HYPODERMIC
NEEDLE
— SEPTUM
POROUS
* — SECTION
Figure 9.  Filter tip sampler (BAT Envitech Inc., 1988).
                 -26-

-------
EXPERIMENTAL SUCTION SAMPLERS  (UNSATURATED/SATURATED SAMPLING)

Experimental samplers, described in the literature, are usually
limited to research applications because of their fragility.  For
the most part, these samplers  are not commercially available.
However, most of these samplers may be easily fabricated.
Experimental suction samplers  operate on the same principles as
vacuum lysimeters, and are also limited to depths of less than
7.5 m (see Table 2).

CELLULOSE-ACETATE, HOLLOW-FIBER SAMPLERS

These samplers consist of a bundle of cellulose-acetate hollow
fibers (see Figure 10).  The bundle of flexible fibers is pinched
shut at one end and attached to a suction line at the other end.
The suction line leads to the  surface and attaches to a sample
bottle and source of suction in the same manner as a vacuum
lysimeter.  Levin and Jackson  (53) described similar fibers made
from a noncellulosic polymer solution.

MEMBRANE FILTER SAMPLERS

Membrane filter samplers were  described by Morrison (2), U.S.
Environmental Protection Agency (19) and Stevenson (54).
Figure 11 shows that a sampler consists of a membrane filter of
polycarbonate, cellulose acetate, cellulose nitrate or PTFE

                               -27-

-------
                    SUCTION
                    LINE
                    CELLULOSE-
                    ACETATE HOLLOW-
                    FIBER BUNDLE
                    SEALED
                    END
Figure 10. Cellulose-acetate hollow-fiber sampler.
           -28-

-------
mounted in a  "swinnex"  Type  filter  holder  (54, 55,  56).  The
filter rests  on  a glass fiber  prefilter.   The prefilter rests on
a glass fiber "wick" which in  turn  sits on a glass  fiber
collector.  The  collector is in  hydraulic  contact with the soil,
extending the sampling  area  of the  small diameter filter  (see
Figure 11).   A suction  line  leads from the filter holder to the
surface.  At  the surface, the  suction line is attached to a
sample bottle and suction source in a manner similar to vacuum
lysimeters.

BARREL LYSIMETER

There are two limitations with suction samplers.  First, they may
not sample from  macropores (unless  the macropores are directly
intercepted).  Second,  their results cannot be used in
quantitative  mass balance studies.  Horby  et al. (57) described
an installation  which could be used to surmount these problems
(see Figure 4).  A barrel-sized  casing (e.g., 57 cm outside
diameter by 85.7 cm high) is placed in a support device and
gently pushed into the  soil with a  backhoe.  As the casing is
pushed, soil  is  excavated around it to help with insertion.  The
process results  in an encased  monolith of  undisturbed soil.  The
monolith is then rotated and lifted, pressure-vacuum lysimeters
are placed in its base, and the  bottom is  sealed.   Subsequently
the assembly  is  placed  back into the ground at the  monitoring
site (see Figure 12).   All fluid draining  through the monolith is

                               -29-

-------
  LINE
HOUSING
                          BACKFILLED
                              SOIL
                                         BARREL
                                         LYSIMETER
SAMPLE
 LINES
         UNDISTURBED
                 SOIL
                         SUCTION
                        SAMPLERS
         Figure 12.  Barrel lysimeter (after Hornby et ad, 1986).
                   -31-

-------
                        MEMBRANE
                           FILTER
                              CUT
                                        (a)
                                     FILTER/
                                     SUPPORT/BASE

                                     SEALING
                                     SHOULDER
                                CAP
SEALING CLAMP

 SAMPLE
 BOTTLE
SUCTION
    LINE
BACKFILL-
                   VACUUM
                   INDICATOR
                                                 (b)
                                             SUCTION LINE
                   SUCTION
                   SOURCE
POLYETHYLENE
SHEETS
      MEMBRANE
        FILTER
       SAMPLER
                  GLASS
                   FIBER
                  "WICK"

              GLASS FIBER
              COLLECTOR
                                                FILTER/
                                                SUPPORT/BASE
                                      "SWINNEX"
                                      FILTER
                                      HOLDER
                                       MEMBRANE
                                       FILTER

                                       GLASS FIBER
                                       PREFILTER


            Figure 11.  Membrane filter sampler (Stevenson, 1978):
                   a) Preparation of filter sampler,
                   b) Installation of fitter sampler.
                               -30-

-------
     VACUUM
  HAND PUMP
RECEIVING
BOTTLE
        UNDISTURBED
        SOIL
   DISCHARGE
      TUBING
BACKFILL
 TRENCH
    WALL
     VACUUM
      PLATE
                                    COLLECTED
                                    SAMPLE
         Figure 13. Vacuum plate sampler installation (after Cole, 1958).
                       -33-

-------
collected by the  samplers.   Inasmuch as the boundaries of the
system are sealed, the  flux  of  liquid through the system requires
maintaining a vertical  downward hydraulic gradient by applying
continual suction to the samplers.

VACUUM PLATE SAMPLERS

A vacuum plate sampler  consists of a flat porous disk fitted with
a nonporous backing attached to a suction line which leads to the
surface (see Figure 13).  Plates are available in diameters
ranging from 4.3  to 25.4 cm  and custom designs are easily
arranged (2, 22).  Plates are available in Alundum®, porous
stainless steel  (43), ceramic (1.2 to 3.0|A.m maximum pore size)
or fritted glass  (4 to  5.5 HJTI maximum pore size) (59, 22, 60, 61,
36, 63, 64).  The nonpermeable  backing can be of fiberglass
resin, glass, plastic,  or butyl rubber.

OPERATIONAL CONSTRAINTS OF SUCTION LYSIMETERS

The inherent heterogeneities of unsaturated pore-liquid movement
and chemistry limit the degree  to which samples from suction
samplers can be considered representative.  This is because the
small cross-sectional areas  of  suction samplers may not
adequately integrate for spatial variabilities in liquid movement
rates and chemistries (27, 65,  66, 67).  Biggar and Nielsen (67)
suggested that results  of chemical analyses from suction sampler

                               -32-

-------
TABLE 3a. POROUS MATERIAL INTERACTIONS*

Alb
Alkali-
nity
Ca
C
C03
HC03
Cd
Cl
Cr
Cu
Fe
H
K
Absorbs
Species






C(ll)

C(19)
C(ll)
C(ll)

0(5,6,15)
Oesorbs
Species
cC(2)d

0(1,2,18)
CAP (18)
A(14)
FG(22)
C(2)
C(2)


C(3)
PTFE ( 3 )
A(3)
C(3)
PTFE ( 3 )
PTFE ( 3 )
A(3)

C(18)°
A(14)
Screens
Species













No
Signif .
Inter-
aciton
C(16)
SF(ll)
C(3,6,10,
11,25)
PTFE ( 3 )
A(3)
FG(18,22)
CAF(IO)



C(3)
PTFE ( 3 )
A(3)
C(ll,25)
SF(ll)

A(3)
C(3,25)
SF(ll)
C(l,25)
CAF(18)
FG(18,22)
No
Inter-
action
-






PTFE (13)


PTFE (13)


35

-------
samples are good  for qualitative but not quantitative
comparisons, unless the variabilities of the parameters involved
are established.  Law  (3) came to similar conclusions, stating
that results from suction sampling could not be used for
quantitative mass-balance studies.

Chemical interactions  between porous segments and the liquids
which pass through them affect the validity of pore-liquid
samples collected with suction samplers (68).  Potential
interactions can  include sorption, desorption, cation exchange,
precipitation, and screening  (69).  These interactions can also
occur with all other parts of the samplers which liquids contact.
However, the much higher surface area within the pores of porous
segments makes them the most critical element chemically.
Table 3 presents  the results of a literature review for porous
section/pore-liquid interactions.  An attempt has been made to
document the pertinent features of the listed studies.  However,
the reader should refer to the original papers to determine if
experimental techniques are applicable to the situation of
interest.  The absence of entries for a constituent relative to a
material does not infer absence of interactions.

EXPERIMENTAL ABSORPTION SAMPLERS (UNSATURATED/SATURATED SAMPLING)

Absorbent samplers depend on the ability of a material to absorb
pore-liquids (2).  Samples are collected by placing the sampler

                               -34-

-------
TABL]
	
Pb
Si02
Si
S04
Sr
Zn
High
Molec .
Wt.
Cmpds .
4-
nitro-
phenol
Chlori-
nated
Hydro-
carbons
Diethyl
Phthal-
ate
Naph-
thalene
Acenaph-
thene
2 3a. PORO
Absorbs
Species







PTFE(23)
PTFE(23,
24)

PTFE(23)
PTFE(23)
US MATERIAL
1
Desorbs
Species

C(2)


C(ll)
C(ll)






INTERACTIONS * ( Continued )
• 	
Screens
Species






0(17,21)
CAF(IO)





No
Signif .
Inter-
aciton


0(4)
PTFE(4)
0(11)





PTFE(23)


No
Inter-
action
PTFE(--13)




PTFE(13)






NOTES ON Table 3a:
a:   Comparisons of materials based  on  this table should be
     made cautiously. Differing experimental techniques should
     be  considered as  a source  of differing  conclusions.
     Undocumented  factors often  include  material age  and
     sampling history.
b:   Valence states are often not reported in studies.
                           37

-------
TABLE 3a. POROUS MATERIAL INTERACTIONS* (continued)

Mg
Mn
Na
NH4
N
NO2
N03
NO3-N
(N02+
N03 ) -N
P
P04
PO4-P
Absorbs
Species
C(6)
C(ll)
C(6)
C(4,12)





0(1,5,8,
15,18)
C(4,5,7)

Desorbs
Species
C(2,3,ll,
18)
A(3,14)
CAP (18)
A(3)
C(12,18)
A(14)
CAF(18)
FG(18,22)

FG(22)







Screens
Species






CAF(IO)
C(10)
CAF(IO)




No
Signif .
Inter-
aciton
C(10,25)
PTFE ( 3 )
CAP (10)
FG(18,22)
C(3)
PTFE ( 3 )
A(14)
C(l,ll,
25)
PTFE ( 4 )

C(4,5)
PTFE ( 4 )
C(4,8)
PTFE ( 4 )

C(5)
CAF(18)
FG(18)
PTFE ( 4 )
CAF(IO)
C(10)
:AF(10)
No
Inter-
action
PTFE (13)
PTFE (13)
PTFE (13)









36

-------
TABLE 3b.  REFERENCES AND NOTES ON EXPERIMENTAL TECHNIQUES (a)
Reference
Number in
Table
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Citation
Number
in Text
108
24
106
109
110
27
28
69
33
53
29
111
2
23
112
68
3
113
11
114
34
56
115
116
105
Porous
Section
was
washed
X
X
X
X

X
X
X

X

X
X
X


X




X


X
Results
are a
Function
of Several
Factors




X
X







X

X

X
X


X
X
X
X
(a) Absence of Information on expenmentaj
the techniques were not specified in 1
Dilute
Solutions
were
Tested













X








X
X

Tests
performed
on non-
porous
materials






















X
X

L techniques means that
:he citation.
                                39

-------
NOTES ON TABLE 3a (Continued)

c:   Abbreviations:
     1.   C = porous ceramic
     2.   PTFE = porous  polytetrafluoroethylene
     3.   A = porous Alundum
     4.   CAP = cellulose acetate fibers
     5.   FG = fritted glass or glass fibers
     6.   SF = silica flour

d:   Numbers in parenthesis refer to references in Table 3b

e:   Example: Reference 18 in Table  3b  (i.e., citation number
     113  in  text)  found  that  there  is  no  significant
     interaction of cellulose acetate fibers with potassium in
     solution.  The  porous section was washed prior to testing
     and  results  were  found to  be a  function of  several
     factors.
                           38

-------
CERAMIC ROD SAMPLERS

These samplers consist of solid, tapered ceramic rods.  Prior to
installation, the rods are boiled in distilled water, dried, and
weighed.  The rods are simply installed by driving them into the
soil.  After a period of time, the rods are withdrawn, weighed,
and again boiled in distilled water.  The water is then analyzed
(71).

PROBLEMS WITH EXPERIMENTAL ABSORPTION SAMPLERS

As with other samplers, there are problems with chemical
absorption, desorption, precipitation, cation exchange and
screening of various pore-liquid components as a function of the
sampler materials. Tadros and McGarity (70) discussed these
concerns in relation to sponge samplers.  Shimshi (71) provided a
good discussion of the limitations of sampling for nitrate with
ceramic rod samplers.

FREE DRAINAGE SAMPLERS (SATURATED SAMPLING)

A free drainage sampler consists of some sort of collection
chamber which is placed in the soil.  Pore-liquid in excess of
field capacity is free to drain through soil (usually through
macropores) under the influence of gravity.  Hence, these
samplers collect liquid from those portions of the vadose zone

                               -41-

-------
in contact with soil.  Liquid is allowed to absorb into the
sampler material over time.  The sampler is then removed, and
liquid is extracted for analyses.  The simplicity of these
samplers have made them attractive to some investigators.

Physically, absorbent methods are limited to soils approaching
saturation.  Sampling requires removing the device and bringing
it to the surface.  Because of this requirement, repeat sampling
at the same location is difficult.  Although the sampler may be
placed back at its original location, identical hydraulic contact
with the soil cannot be guaranteed.

SPONGE SAMPLERS

This sampler includes a cellulose-nylon sponge seated in a
galvanized iron trough (70).  Samples are collected by pressing
the dry sponge against a soil surface with a series of lever
hinges.  The sponge is left in place until a sufficient volume of
pore-liquid has been collected for analyses.  Theoretically,
there is no maximum sampling depth for sponge samplers.  However,
because access trenches are required for operation, installations
are restricted to shallow depths dictated by excavation equipment
and safety considerations.
                               -40-

-------
          Pan Lysimeter
          Glass Block Lysimeters
          Caisson Lysimeters

          Wicking Soil Pore-Liquid Samplers
          Trough Lysimeters
          Vacuum Trough Lysimeters
          Sand-Filled Funnel Samplers

PAN LYSIMETERS

A pan lysimeter generally consists of a galvanized, metal pan of
varying dimensions  (see Figure  14).  A copper tube is soldered to
a raised edge of the pan.  Plastic or Tygon tubing connects the
copper tube to a collection vessel.  Any liquid that accumulates
on the pan drains through the tubing into the vessel (19, 45).

GLASS BLOCK LYSIMETERS

Barbee and Brown (72) developed free drainage samplers made from
hollow-glass bricks (see Figure 15).  These glass bricks, which
are produced as ornamental masonry, have dimensions-of 30 by 30
by 10 cm and have a capacity of 5.5L.  To build a sampler, nine
holes, 0.47 cm in diameter, are drilled along the perimeter of
one of the square surfaces of a brick.  Nylon tubing is inserted
into one of the holes to allow  for sample removal.  The

                               -43-

-------
which are intermittently saturated due to events such as
rainfall, flooding, or irrigation.  This gravity drainage creates
a slightly positive pressure at the soil-sampler interface
causing fluid to drip into the sampler.  Some free drainage
samplers apply a small suction in order to break the initial
surface tension at the soil-sampler interface.  Samples are
retrieved either by accessing the samplers at depth or by drawing
samples to the surface through a suction line.

Suction samplers can also be used to sample free drainage flow.
However, the small area of those samplers compared to the spacing
of macropores limits their usefulness for this application.  In
addition, suction must be applied to suction samplers to collect
samples, even under saturated conditions.  Free drainage samplers
are passive collectors which automatically collect the
percolating liquids.

Free drainage samplers are classified differently by various
authors, depending on the installation methods.  Many free
drainage samplers are installed in the side walls of trenches and
are referred to as trench lysimeters.  However, free drainage
samplers are also installed in the walls of vertical caissons.
The principle behind each of the samplers is essentially the
same.  However, the materials and construction differ.  Free
drainage samplers include the following:
                               -42-

-------
                                         SUCTION
Figure 15.  Glass block lysimeter (after EPA, 1986).
               -45-

-------
      12 IN 	

      SIDE VIEW
            COPPER
            TUBING
                              GALVANIZED
                              16 GAGE
                              METAL PAN
         15 IN
       PLAN VIEW
Figure 14.  Example of a pan lysimeter (EPA, 1986).
          .44-

-------
                 WATER
                                CORRUGATED
                                 STEEL PIPE
                                  CAISSON
NATIVE
  SOIL
      PVC
     HALF
  SCREEN
             COLLECTOR
                 PIPE
                               SAMPLE
                              COLLECTOR
      Figure 16.  Example of a caisson lysimeter (Schmidt and Clements, 1978).
                     -47-

-------
collecting surface is  fitted with a fiber glass sheet to improve
contact with the  soil.  Pore-liquid collection is enhanced by a
raised lip along  the edge  of the surface.

Level blocks are  critical  for retrieving the bulk of the sample.
However, the inside glass  surface is uneven and has low spots
("dead spots") where residual sample collects between sampling
cycles.  This leads to cross-contamination of samples.

CAISSON LYSIMETERS

A caisson lysimeter consists of collector pipes, radiating from a
vertical chamber  (2).  A design used by Schmidt and Clements (73)
consists of a nearly horizontal, half-screened PVC casing (see
Figure 16).  Schneider et  al. (74) designed a similar system
consisting of the following components:  (1) a stainless steel
tube extending diagonally  upward through the caisson wall into
the native soil,  (2) a screened plate assembly within the tube to
retain the soil,  (3) a purging system used to redevelop the
sampler when it becomes clogged, (4) an airtight cap that
prevents exchange between  the air in the caisson and the soil
air.

NICKING SOIL PORE-LIQUID SAMPLERS

Hornby et al. (58) described a wicking sampler (see Figure 17),

                               -46-

-------
which combines the attributes of free drainage samplers and
pressure-vacuum lysimeters.  The sampler collects both free
drainage liquid and liquid held at tensions to about 4 cb.  A
standing "Hurculon" fibrous column acts as a wick to exert a
tension on the soil pores in contact with a geotextile fiber
which serves as a plate covering a 30.5 by 30.5 by 1.3 cm pan.
The terminus of the fibrous column is sealed into the cap of a
tubular chamber.  This chamber also contains an inlet pressure-
vacuum line and a sample collection tube.  Materials for the
sample collection tube depend on the constituents being sampled.
Glass and PTFE were recommended materials when sampling for
organics (58).

TROUGH LYSIMETERS

Trough lysimeters, also known as Abermayer lysimeters, rely on a
trough or pail to collect pore-liquid.  A fiberglass screen is
suspended inside the trough to maintain a firm contact with the
edges of the sampler and the soil.  The screen is lined with
glass wool and covered with soil until the soil is even with the
top of the trough (75).

Morrison (2) reported a trough lysimeter in which two parallel
metal rods are inside the trough, in contact with the bottom side
of the screen, and bent toward the collection tube (see
Figure 18).  Liquid that enters the trough migrates along these

                               -49-

-------
                              GEOTEXTILE FIBER
     HURCULON   !
        FIBERS
PLEXIGLASS PLATE
      DISCHARGE LINE
                      I!
                      IS
                      U
                      u
                      \\
                       \\
                       I:
                       ii
                       u
                       I !
            TYGON HOSE
            CONTAINING
         HURCULON FIBERS
PRESSURE-
 VACUUM
   LINE
           Figure 17.  WIcking type soil pore-liquid sampler (Hornby et al, 1986).
                         -48-

-------
rods towards the collection tube in response to capillary forces.
A modification of this design consists of a metal trough with a
length of perforated PVC pipe mounted inside.  The trough is
filled with graded gravel so that coarse material is immediately
adjacent to the PVC pipe and fine sand is at the edges and the
top of the trough.  The pipe is capped at one end while the other
end is connected to a sample container via a drainage tube (2).

VACUUM TROUGH LYSIMETERS

Montgomery et al. (76) described a vacuum trough lysimeter
consisting of a metal trough equipped with two independent
strings of ceramic pipe, each 13 mm in diameter.  The primary
purpose of this design is to sample free drainage.  However, the
device also apparently allows extraction of samples under applied
suctions of up to 50 cb.  The ceramic pipes act as a vacuum
system, and samples are extracted through a suction line.

SAND-FILLED FUNNEL SAMPLERS

Brown and Associates (77) discussed a sand-filled funnel for
collecting freely draining liquid (see Figure 19).  The funnel is
filled with clean sand and inserted into the sidewall of a
trench.  The funnel is connected through tubing to a collection
bottle.  Application of suction to a separate collection tube
pulls the sample to ground surface.

                               -51-

-------
        FIBERGLASS
          SCREEN
           METAL RODS

SAMPLE LINE
         Figure 18. Trough lysimeter (after Jordan, 1968).
                 -50-

-------
PERCHED GROUND-WATER SAMPLING  (SATURATED SAMPLING)

Perched water occurs where varying permeability layers in the
vadose zone retard downward movement of liquid.  Over time,
liquid collects above lower permeability layers and moisture
content may increase to saturated (19, 9).  Once soil becomes
saturated, wells and other devices normally installed below the
water table can be used to collect samples.

Sampling perched liquid is attractive because the perching layer
collects liquid over a large area.  Such integrated samples are
more representative of areal conditions than suction samples
(78).  This also allows the sampler to potentially detect
contaminants which may not be moving downward immediately
adjacent to the sampler.  In addition, larger sample volumes can
be collected than those which can be obtained by suction
samplers.  Everett et al. (7, 9) discussed the incorporation of
perched ground-water sampling into monitoring programs.

Perched water systems can be difficult to find and delineate.
Surface and borehole geophysical methods (e.g. neutron logging)
and video logging of existing wells are often used.  Also,
perched systems tend to be ephemeral.  Therefore, suction
samplers are sometimes required as backups.  As with all
samplers, potential chemical interactions between sampler
materials and the constituents of interest should be considered.

                               -53-

-------
   LINE
 HOUSING
          SUCTION
            LINE
EXCAVATED TRENCH
  WHICH MAY BE
    BACKFILLED
UNDISTURBED
         SOIL
NATIVE
SOIL
BACKFILL
                         SAND
                         FILLED
                         FUNNEL
                                      COLLECTED
                                      SAMPLE
         Figure 19. Sand filled funnel sampler installation (EPA, 1986).
                      -52-

-------
VALVED
 PORT
     \
       PACKER
  SHORT
SCREENED
 INTERVAL
                 IBENTONITE
                 'SEAL
                PACK
          OPEN
          PORT
SEAL
SAMPLER
                           VALVED
                            PORT
                                                           SHORT
                                                         SCREENED
                                                         INTERVAL
       Figure 20.  Examples of point sampling systems (after Patton and Smith, 1988).
                             -55-

-------
Because these samplers are usually installed for other purposes/
incompatibility of materials with monitoring objectives is often
a problem.  Everett et al. (9), Dunlay (79), and U.S.
Environmental Protection Agency (80) discussed this topic.

Following are some of the methods for sampling perched
groundwater:

          Point samplers
          Wells
          Cascading water samplers
          Drainage samplers

POINT SAMPLERS

Point samplers are open ended pipes or tubes/ such as piezometers
or wells with short-screened intervals/ installed for the purpose
of collecting samples from a discrete location in saturated
material (see Figure 20).  Samples are collected by bringing
liquid which flows freely into the device to the surface by one
of a variety of methods.  Figure 20 presents various point
sampler configurations which have been used (2/ 9, 19, 86/ 87/
88, 89, 90, 91, 92, 93, 94, 95, 96).
                               -54-

-------
                    LOCKABLE
                PROTECTIVE COVER
   WELL CAP
     CASING

    SURFACE
 CASING SEAL
        ZONE
      OF LOW
PERMEABILITY
      WELL
    SCREEN
    DENSE PHASE
   SAMPLING CUP
     CONCRETE
     SURFACE
     SEAL
      SURFACE
      CASING

-     POTENTIO-
      METRIC
      SURFACE"A"

  BACKFILL

      POTENTIO-
      METRIC
      SURFACE"B"

      SEAL
                                       FILTER
         Figure 22.  A monitoring well installed to sample from the lower of two
               ground-water zones (after Riggs and Hatheway, 1988).
                        -57-

-------
                  LOCKABLE
              PROTECTIVE COVER
WELL CAP
  CASING
BACKFILL
     SEAL
    WELL
  SCREEN
 DENSE PHASE
SAMPLING CUP
                                       CONCRETE
                                       SURFACE
                                       SEAL
                                      POTENTIO-
                                      METRIC
                                      SURFACE
                                      FILTER
    Figure 21. A monitoring well with the uppermost ground-water level Intersecting
            the slotted well screen (after Riggs and Hatheway, 1988).
                    -56-

-------
                    LOCKABLE
                PROTECTIVE COVER
   WELL CAP
   BACKFILL


CASING SEAL
       ROCK
CONCRETE
SURFACE
SEAL
                                         OVERBURDEN
                                         CASING
 SOIL
                                      OPEN BOREHOLE
   Figure 23. An open-hole ground-water monitoring well in rock (after Riggs and Hatheway, 1988).
                         -59-

-------
CASCADING WATER SAMPLERS

Cascading water occurs when a well is screened throughout a
perched layer and the underlying water table  (see Figure 24) or
when water leaks through casing joints at the perched layer.
Because the water table is lower than the perched layer, water
flows into the well in the portion open to the perched layer/ and
cascades downward to the water table.  This situation is common
in some areas where the practice has been to install water wells
with large screened intervals (44).  Samples are collected by
capturing liquid flowing into the well from the perched layer
before it cascades down to the water table (78).  Alternatively,
water samples pumped from a well that has been shut down for a
period of time represent ground water that has been influenced by
cascading water (see Figure 24).

DRAINAGE SAMPLERS

Shallow perched systems may spread contamination, cause problems
with structures, or interfere with agriculture.  Drainage systems
are installed to alleviate these problems.  These systems cause
gravity flow of perched ground water to a ditch or sump from
which it is pumped out.  This outflow can be sampled.  Typical
drainage systems include tile lines, half perforated pipes,
                               -58-

-------
synthetic sheeting, or even layers of gravel and sand.  Depending
on the design of the system, it may be possible to sample
outflows which drain different areas.

Schilfgaarde (96) contains numerous papers on the design and
construction of drainage systems.  Donnan and Schwab (97)
described sampling from agricultural drainage systems.  Gilliam
et al. (99), Gambrell et al. (100), Eccles and Gruenberg (101)
and Gilliam et al. (102) described sampling from tile drainage
lines.  Gilliam et al. (99) and Jacobs and Gilliam (103)
described sampling from drainage ditches.  Wilson and Small (104)
described a lateral drain sampler installed beneath a new
sanitary landfill.
                               -61-

-------
                          PRODUCTION
                             WELL
                               I
                                PERFORATED
                                INTERVAL
VADOSE
  ZONE
PERCHED f
    ZONE!
PERCHING
   LAYER
        AQUIFER
      AQUICLUDE
                                                PERCHED
                                                WATER
                                                LEVEL

                                  CASCADING
                                  WATER
                                     _WATER
                                        TABLE

                                     AQUIFER ZONE
                                     INFLUENCED
                                     BY CASCADING
                                     WATER
     Figure 24. Conceptualized cross section of a well showing cascading water
            from perched zone (after Wilson and Schmidt, T979).
                            -60-

-------
tracer chemical should also be conservative.  That is, in the
case of an organic compound, the tracer chemical should not be
reactive and it should not interact with the environment, for
example, by adsorption on to the materials of the media being
monitored.

As might be expected even when the final indicator parameters for
RCRA ground-water monitoring were selected (pH, specific
conductivity, total organic carbon, and total organic halogens),
it became apparent quickly that many uncertainties associated
with these indicators were still present when they were compared
with the properties for an ideal indicator parameter.  As a
result, one of the conclusions reached in the RCRA ground-water
monitoring program was the need to develop guidance for selecting
and using indicator parameters by specifying conditions such as
type of operations or hydrologic situations under which valid
correlations could be expected.  Unfortunately, ground-water
regimes being examined would not be expected to be readily
comparable with the varieties of conditions which actually exist
in the vadose zone.  In addition, site-specific differences
between aquifers and vadose zone sediments in different regions
of the country could in some cases invalidate the use of the same
indicator parameters.

In the vadose zone, therefore, it appears more reasonable to
select a series of indicators that would be consistently valid

                               -63-

-------
   SELECTING INDICATOR PARAMETERS FOR PORE-LIQUID SAMPLING OF
             PETROLEUM HYDROCARBON CONTAMINATED SITES

One of several original concepts used by the Environmental
Protection Agency  (EPA) in  its  ground-water monitoring program
developed under the Resource Conservation and Recovery Act (RCRA)
was the selection  of indicator  parameters and tracer chemicals
for the detection/monitoring phase.  The purpose of these
measurements was to provide an  initial indication of whether
migration of hazardous constituents in the ground-water regime
had occurred.  This concept can also be applied to contaminant
migration monitoring in the vadose zone.

Among the key factors to be considered in selecting any indicator
parameter is to find a compound that is naturally absent from the
overall system to  be monitored, and yet is related to the
compounds that are in the source being monitored.  The indicator
parameter(s) should be selected from the broad range of compounds
stored in the UST  since they will behave in a similar manner in
the environment.   Also, analytical sensitivity, precision,
accuracy and costs of the monitoring procedures should be
reasonable.  If a  tracer that is exotic by definition and absent
from the system being monitored cannot be found, then the next
best choice might  be an indicator parameter which provides a
measure of relative change  rather than indicating only the
presence or absence of this particular compound.  An effective

                               -62-

-------
C8)  (cracking) range and produce a high percentage of branched
(reforming) chains, both of which increase the octane rating.

Petroleum products make up a family of related organic products.
Gasoline contains a mixture of low-boiling hydrocarbons typically
made up of branched-chain alkanes, alkenes, and cycloalkenea
(aromatic hydrocarbons).  In general, the chain lengths are
between C4 to C8.

Aviation jet fuels are generally composed of light petroleum
distillates consisting of mixtures of gasoline and kerosene with
kerosene containing carbon chains from about C9 to C14, and
having a somewhat higher octane rating than automotive gasoline
since it contains a large fraction of aromatic hydrocarbons.
Diesel fuels are generally composed of carbon chains ranging rom
Cll to C20 and contain very small quantities of aromatic
hydrocarbons.

Another category of petroleum products consists of lubricating
oils.  They contain higher molecular weight hydrocarbons with
carbon chains ranging from C20 to C28 and contain very low
concentrations of aromatic hydrocarbons.  Waste oils may contain
some additional products generated by thermal decomposition of
the lubricating oil during its use.
                               -65-

-------
for similar situations within the realm of vadose zone monitoring
applications, and which would also be compatible with and be
representative of the various types of materials which represent
typical contaminant sources.  The family of indicator parameters
for this purpose will likely be considerably larger than for
ground-water monitoring programs.  By understanding the various
environments which can exist and the various petroleum
hydrocarbons which are stored and represent contaminant sources,
an indicator parameter system can be selected to be somewhat more
dependent on site-specific conditions than the indicator
parameters typically used for ground-water monitoring programs.

CHEMISTRY OF PETROLEUM PRODUCTS

Petroleum products are complex substances made up of many
chemical compounds generally produced from petroleum crude oil.
Petroleum crude oil is subjected to fractional distillation which
separates a number of individual hydrocarbon products based on
their boiling point.  In addition to distillation, the organics
are also subjected to catalytic cracking and catalytic reforming.
Cracking refers to the breaking of long carbon chains to form
shorter carbon chains with lower molecular weights.  Reforming
refers to transforming straight-chain hydrocarbons into branched-
chain hydrocarbons.  The purpose is to produce a hydrocarbon
fraction (gasoline) which is in the four and eight carbon (C4 to
                               -64-

-------
the chemicals that have been used by the major petroleum
companies to specifically identify their products.  Since they
vary from manufacturer to manufacturer, they are not likely to be
candidates for indicator parameters.  However, if a site has
multiple potential sources of leaks, they might be useful in
identifying which source is the contributor.

Physical and chemical information of petroleum hydrocarbon
compounds typically found in gasoline are summarized in Table 4.
This information includes water solubility of hydrocarbons,
boiling points of individual compounds, and vapor pressures.  The
list of gasoline components was selected from the California
Leaking Underground Fuel Tank (LUFT) Field Manual.

Each of these individual compounds in this complex mixture of
chemicals have their own chemical and physical properties which
will determine how they behave in the particular subsurface
environment into which they are introduced.  Since most of the
components are of lower molecular weight, they normally have high
vapor pressures, and, therefore, will be present in the vapor
phase.  In addition, these lower molecular weight compounds all
have some solubility in water ranging from several thousands
parts per million to a few parts per million.  As a result, in
pore liquids, these organics will be present as two distinct
phases in the environment.  The water soluble portion of the
organic is known as the water phase.  The water insoluble,

                               -67-

-------
Prior to the introduction of the higher octane gasolines
(unleaded and super unleaded gasolines), compounds like
tetramethyl lead and tetraethyl lead had been added to the lower
octane gasolines to reduce knocking and increase or boost the
octane ratings.  Other  organic compounds, such as ethylene
dibromide (EDB) or ethylene dichloride  (EDC), have also been
added to leaded gas primarily because of their ability to remove
lead from the engine.   With the gradual phasing out of leaded
gasoline, and, as a consequence, phasing out the use of organic
lead compounds and EDB  and EDC as additives, large amounts of
aromatic hydrocarbons such as benzene and toluene are being added
to boost gasoline octane ratings for increased engine performance
and to prevent knocking.

In recent years, methanol and ethanol have also been added to
gasoline to make products referred to as "gasahol."  The alcohols
are also used primarily to boost the octane, and, as a secondary
benefit, reduce the effect of water in the gasoline.  Another
gasoline additive that  has been used recently to boost octane
ratings is methyl tertiary butyl ether.

Another category of compounds is made up of gasoline additives.
Some of these additives, while they represent a very small
fraction in gasoline, are chemical compounds with markedly
different properties from the branched chain and aromatic
hydrocarbons which contain only carbon and hydrogen.  These are

                               -66-

-------
immiscible portion is known as the non-aqueous phase.  Since non-
halogenated hydrocarbons have densities less than water, they
will float in the water phase.  If the hydrocarbons are
halogenated, i.e, contain bromine and chlorine, like EDB and_EDC,
their densities are greater than water, and any immiscible  -
portions of these compounds will sink to the bottom of the water
phase.  This is further complicated by the fact that most organic
compounds are subject to biodegradation, and some of the
degradation products from microbial processes can also be present
in either phase.

FACTORS AND CONSIDERATIONS THAT AFFECT THE SELECTION OF INDICATOR
PARAMETERS

As a first step in selecting indicator parameters, it is
necessary to define, in general, the products to be monitored.
Petroleum hydrocarbon compounds can be divided into several major
categories that can be further defined by products stored in
tanks.  It is also helpful to further divide some of the major
products stored in tanks.  It is also helpful to further divide
some of the major products into subcategories.  The major product
categories can be defined as complex multi-component systems,
simple multi-component systems, and single component systems.
                               -69-

-------
TABLE 4. SELECTED PHYSICAL AND CHEMICAL PROPERTIES FOR MAJOR
COMPONENTS FOUND IN A TYPICAL GASOLINE MIXTURE

Benzene
Toluene
Ethylbenzene
m-Xylene
o-Xylene
p-Xylene
n-Butane
n-Pentane
n-Hexane
Ethylene
dibromide (EDB)
Ethylene
dichloride (EDC)
Tetraethyl Lead
Tetramethyl Lead
'Based on compoun
9At 20«C
C0ecomposes in wal
Note: Unless otherv
Sources: 1) Montog
Chelsea,
2) Verech
Nostrand
Reinhot
3}DeVitt,
I
Solubility
(ppm)
1791
550
152
146
175
198
61 a
35
138
250
8650
N/A°
N/Ae
Boiling Point
CO
80.1
110.6
136.2
139.1
144.4
138.3
-.5
36.1
69
131.6
83.7
110
110
Vapor
Pressure
(mm Hg)
95.19
22.0
10.0
8.3
6.6
8.76
1823
430"
1208
11s
8%
.15"
22.5*
Henry's Law
Constant
(atm-m'/mole)
5.48x10*
6.74x10*
6. 44. 1O*
6.3 x Iff3
5.1 x10-a
7x1ff3
.947
1.23
1.68
N/A
9.8X10"
N/A
N/A
Specific
Density*
.8765
.8669
.8670
.8642
.8610
.8610
.6012
.6262
.6603
2.701
1.2351
1.659
1.99
d at 20«C and water at 4«C
er, ultimately to Pb"
rise noted, physical properties listed at 25'C
omery, J.H, and Welkom, LM. 1 990. Groundwater Chemicals Desk Reference. Lewis Publishers.
Ml.
ueren. K. 1983. Handbook of Environmental Data on Organic Chemicals, 2nd Edition. Van
d Co., New York, NY.
D.A. et al. 1987. Soil Gas Sensinq for Detection and Maopinq of Volatile Organics. National Well
Water Association, Dublin, OH.
68

-------
suited as an indicator parameter.  Since the simplest examples
                                                            x
are those tanks containing single component products stored as
raw materials or finished products, the choice of an indicator
parameter for this case will also be the simplest.

The two subcategories which exist in a single component system
are organic and inorganic liquids.  The inorganic chemicals are
most likely strong acids or strong bases, or highly concentrated
solutions of inorganic salts which are water soluble.  The
organic chemicals are single chemical products such as benzene,
acetone, or chlorobenzene.  In these cases, the logical indicator
parameter to identify for single component systems is the
particular organic or inorganic chemical which is stored in a
potential contaminant source.  Inorganic products can be
monitored using generic indicator parameters such as electrical
conductivity and/or pH, while the organic products can be
monitored using a sensor for specifically detecting the presence
of that organic compound.

Storage tanks containing simple multi-component systems
consisting of simple mixtures of organic chemicals are also
relatively simple to monitor.  Indicator parameters can most
likely be selected from the organic component present in the
tanks that has the highest concentration provided its physical
and chemical properties lend themselves to measurement in the
field.  In general, simple multi-component systems contain less

                               -71-

-------
The largest single product category under multi-component systems
would be petroleum products.  The category of petroleum products
can be further divided into subcategories, ie., automotive fuel,
aviation fuel, diesel fuel, and other petroleum products such as
lubricating oils, waste oils, etc.  As examples of simple multi-
component systems, two subcategories would be mixtures of
compatible chemicals stored for recycling and mixtures of
compatible hazardous wastes.  Examples of single component
systems would be tanks storing either a raw material or finished
product.  These substances, as well as any other substance in a
single component system can be further separated into two general
subcategories - organic materials and inorganic materials.
Indicator parameters that will eventually be selected as
representative of a particular petroleum hydrocarbon compound
will most likely be chosen from each one of these various
subcategories wherever possible.  While the three generic
divisions and the major categories within them are convenient for
describing petroleum hydrocarbon compounds, the subcategories
represent that particular population which contains the
characteristic element(s) that can be utilized as the indicator.

The next step in selecting indicator parameters is to determine
the chemical components which are present in each one of these
subcategories.  Their chemical and physical properties are then
used to assist in selecting those chemical(s) which are best
                               -70-

-------
be affected by the subsurface depending on their chemical and
their physical properties which differ from component to
component.  Without a thorough understanding of these chemical
and physical properties and how these properties govern the
behavior in the vadose zone, it is difficult to make any
determination of what component or group of components should be
selected as an indicator parameter(s).

Finally, the selection is made more difficult by the fact, as
noted earlier, that the vadose zone is also a multicomponent
system.  It is composed of three separate phases:  solids,
liquids and gases.  Each phase may also consist of several
different substances.  For example, the solid phase, soils, is
typically a variable mixture of coarse to fine-textured
particles.

The analytical method selected to measure an indicator parameter
is an important consideration.  Some products may lend themselves
in particular to simple analytical measurements such as pH or
conductivity which can be made in the field, while other products
would require more sophisticated laboratory analytical
techniques.

Any attempt to deal with early detection monitoring of leaks in
the vadose zone requires an understanding of the nature of the
product which has been released and its behavior in the

                               -73-

-------
than four to  five  individual chemicals and all of the materials
are typically chemically compatible.  For example, one typical
storage unit  might contain only hydrocarbons with similar boiling
points and vapor pressures being stored prior to recycling, while
another typical unit could contain chlorinated hydrocarbons to be
incinerated.  Since the organic compounds within each tank would
have to be compatible  for safety requirements, it is quite likely
they would have similar boiling points and vapor pressures.
Therefore, in most instances, it is relatively straightforward to
select an indicator parameter for tanks containing simple multi-
component systems.

Of all three  divisions, the complex multi-component system is the
most difficult one to  select a chemical from that would be a
suitable indicator parameter since, by definition, there are many
individual chemical substances present in varying amounts.  This
division is unique since, for all practical purposes, the only
substances that make up the division are petroleum products.  As
discussed earlier, there are several different petroleum
products, e.g, jet fuel, gasoline, kerosene, and other examples.
The differences are small, however, such as lower or higher
molecular weight fractions or percentages of aromatic
hydrocarbons.  All are basically the same complex multi-component
mixtures of hydrocarbons.  The presence of all of these chemical
compounds greatly  complicates the behavior of the mixture when
they are released  into the environment as they will interact and

                               -72-

-------
In any monitoring system, false positives or false negatives can
occur.  A false positive will lead to the conclusion that a
release has occurred when no release, in fact, has occurred.  A
false negative will lead to the conclusion that no release has
occurred when, in fact, a release has occurred.  Therefore, an
ideal indicator parameter system would prevent the occurrence of
false negatives and false positives.  For example, if two or more
indicator parameters are identified to be present in an
established ratio in the source mixture, then the detection of
only one of these components would not conclusively indicate that
a new leak is occurring.  Further, detecting one of these
components could suggest that: 1) an "old" or "aged" leak which
has been subjected to biodegradation has occurred or, 2) a new
leak has occurred and the sampled liquid may not contain both
indicator parameters due to contaminant partitioning at the
sampler porous membrane or in the soil prior to reaching the pore
liquid sampler.

All of this information would seem to indicate that although
indicator parameters can be selected to monitor and identify
contaminant releases, it is unlikely at this time that any single
indicator parameter can be selected for all systems that must be
monitored.  In fact, it would appear that, specific indicator
parameters are typically used for specific types of materials
stored in the underground environment.  The goal, therefore, is
to define those indicator parameters which can be used

                               -75-

-------
subsurface environment.  Consequently/ a knowledge of (1) the
physical and chemical properties of products which are potential
contaminants/  (2) the mechanisms responsible for contaminant
releases/ and  (3) knowledge of the environment into which the
chemicals are  leaking/ is necessary.  Only with this
understanding  can any comprehensive program provide early warning
for detecting  leakage prior to its introduction into the ground-
water regime.

The fact that  the vadose zone is made up of a series of totally
different and  unique environments varying substantially from site
to site/ which are almost never homogeneous/ is likely the most
difficult problem to address.  The vadose zone is a complex
environment consisting of components of both saturated and
unsaturated zones.  The capillary fringe zone is the next deeper
zone normally  encountered and is underlain by the aquifer or the
saturated zone.  The migration of hazardous constituents from
leaking petroleum hydrocarbon contaminant sources is even more
complex since  the mechanism for migration is different within
each of these  zones.  Therefore/ the fundamental understanding of
the migration  of hazardous constituents must include an awareness
of all of the  hydrogeologic complications which can exist in each
zone and must  consider as a consequence the multi-dimensional
interactions which can occur.
                               -74-

-------
affecting migration of liquids from a leaking tank and their
affects on monitoring methodologies, parameters relating to each
process controlling transport and fate, and monitoring
methodologies showing the interaction of the processes with
monitoring technology, are summarized in several tables.  For the
purposes of this report, it is useful to modify and expand these
tables to include the concept of indicator parameters.  A brief
discussion of each indicator parameter's physical and chemical
properties has been included to assist in understanding how they
will control migration of liquids.

In selecting indicator parameters for monitoring the release of
petroleum hydrocarbon liquids, it is necessary to understand the
processes controlling subsurface migration or transport of these
liquids and to recognize that there will most likely be three
components present affecting transport or migration: (1) the
stored product, (2) the product vapor, and (3) the water.  The
processes which govern product and vapor mobility and migration
are, in general, very poorly understood.

Two distinct aspects of the migration process can typically be
identified:  transport processes and fate processes.  Transport
processes are those responsible for the actual movement or
migration of the constituents through the soils in the
subsurface, while fate processes are those responsible for
transformation or retardation of the constituents which of course

                               -77-

-------
unequivocally.  Furthermore, the goal should be to provide
families of specific  indicator parameters which are likely to
indicate that any potential release source with its associated
system is leaking so  that  this system can be used until another
more general approach becomes apparent.  This approach will then
accommodate both those single component and simple multi-
component systems as  well  as the complex, multi-component
products, which represent  the single largest percent of petroleum
hydrocarbons.

As has been pointed out previously, the vadose zone is extremely
complex and any attempt to select indicator parameters without
making some simplifying assumptions would not be successful.  If
some generalities are used to simplify the conditions in the
underground environment, then indicator parameters can be
selected to monitor for leaks until a better understanding of
transport and fate processes are provided by continued research
efforts.  In addition, by  collecting information from indicator
parameters used for monitoring, a database can be acquired which
would lead to areas where  future research efforts could best be
concentrated.

FATE AND TRANSPORT CHARACTERISTICS OF INDICATOR PARAMETERS

In a recent report entitled "Processes Affecting Subsurface
Transport of Leaking  Underground Tank Fluids" (117), processes

                               -76-

-------
Each of the fixed parameters of a liquid, which are its chemical
and physical properties, are discussed separately in the report.
Solubility is a key parameter which influences persistence,
transport, and the ultimate fate of the leaking liquid in the
natural underground environment.  Adsorption, dispersion,
diffusion, volatilization, degradation, toxicity, and other "
similar properties are functions of the solubility of the liquid
and its concentration in water.  Density of the liquid influences
migration.  For example, those liquids with densities less than
that of water "float" on the water layer, while those which are
denser than water sink to the bottom of the water layer.  The
viscosity of a liquid affects its flow velocity or its ability to
transport, migrate within the subsurface environment.

The surface tension of the liquid is responsible for the
capillary effects and the wetting of the soil particle surfaces
in the underground, and the dielectric constant, which is a
measure of the dipole moment of the liquid, influences its
ability to adsorb on the soil particles.  A liquid with a high
dielectric constant has a high dipole moment and would adsorb
strongly on an ionic solid (soil grains with charged particles on
their surface, such as clay).  In a mixture of water and an
organic compound, water adsorbs preferentially since it has the
highest dielectric constant.  Vapor pressure and the boiling
point of a fluid have an affect on what portions of the fluid
will be in the vapor phase of the subsurface environment within

                               -79-

-------
has an effect on migration.  By understanding these aspects of
the migration process, suitable indicator parameters for
monitoring tank leakage can be selected.  Transport can occur by
both liquid and vapor migration; soil surface and interface
properties between the various phases in the unsaturated zone and
biological degradation can affect the fate or distribution in the
environment.

Major attenuation mechanisms for organic chemicals are
biodegradation, adsorption, and volatilization, all of which will
be determined by the physical/chemical properties of the fluid.
Although biodegradation is important in the attenuation of
organics, it is specific to organics rather than all chemicals in
general and not generally applicable to every chemical.
Excluding mercury, volatilization is also unique to organic
chemicals and is governed by their fixed properties such as vapor
pressure, boiling point, and Henry's law constant, but also by
the variable properties of the environment such as the air-filled
porosity, pore size and shape, and temperature.

Adsorption is determined by the fixed properties of the liquid,
its molecular structure, dielectric constant, solubility, and
viscosity as well as the variable properties of the environment
such as the soil organic content, the volumetric water content,
and surface area of soil particles.  Adsorption occurs to some
extent for all chemicals.

                               -78-

-------
Another major attenuation mechanism for organics is
volatilization.  Although detecting the presence of a leak
through monitoring for vapor release is very cost effective,
there are problems associated with this approach because of the
many complications which can occur as a result of the subsurface
environment.  To simplify these complexities, indicator    -
parameters can be identified by considering only the
physical/chemical properties of the liquids, and ignoring the
interactions which can occur in the subsurface.  Using this
approach, vapor monitoring can be considered as an adjunct for
monitoring the leak.  For example, vapor monitoring can be
considered an early warning system that would indicate the
possibility of a problem in the subsurface.  Such an approach
should be supplemented by appropriate monitoring for pore
liquids. This approach could prove useful particularly in those
instances where it is not feasible either technically or
economically to monitor continuously for liquids, but where an
early alert would be desirable.

SELECTION OF INDICATOR PARAMETERS

As discussed in earlier sections, selection of indicator
parameters for single component systems and even simple multi-
component systems is straightforward.  The complex multi-
component system consists essentially of petroleum products in
the system for which it is more difficult to select indicator

                               -81-

-------
the vadose  zone.  Vapor phase  hydrocarbons can exhibit enhanced
mobility in vadose  zone pore spaces not occupied by liquid,
potentially migrating extended distances to redissolve into pore
water at some remote location.

Adsorption  and volatilization  can be discussed in terms of how
the physical/chemical properties affect interaction of the
liquids in  the underground environment with the variable
characteristics of  the unsaturated zone and to what extent site
environmental factors have on  their behavior.  Adsorption is
applicable  only to  liquids which are organic substances.  It is
responsible for the attenuation mechanisms which remove organics
through adsorption.  The moisture content and organic content of
the soil, the soil  surface area, and the pore/water chemistry all
influence adsorption processes and determine the transport and
fate of the liquid.

The physical/chemical properties of liquids in the subsurface
environment cannot  be altered  except through biodegradation.  In
addition, variable  conditions  in the subsurface environment are
responsible for surface chemical affects that influence transport
and fate. It follows that chemical constituents under
consideration as indicator parameters should be selected based on
the properties of the subsurface environment; consequently, it is
more suitable to select several sets of environmental indicator
parameters  for different sets  of conditions.

                               -80-

-------
Because so many previous studies have included three other alkyl
benzenes - toluene, ethyl benzene, and total xylenes, they will
also be included in this list of candidates.  Other criteria such
as the chemical/physical properties of petroleum products, rather
than concentrations as the sole criteria, can be used to screen
indicator parameters from the list of potential candidates.

Brief descriptions of these  10 organic compounds, presented
below, provide a summary of  chemical and physical properties
which will assist in the selection of the most appropriate
indicator parameter(s).

Benzene is a clear, colorless, highly flammable liquid.  It is
the simplest of the aromatic hydrocarbons and has the highest
water solubility of all of the potential indicator parameters.

Toluene or methyl benzene, an aromatic hydrocarbon, is a
flammable liquid very slightly soluble in water and has been used
as a gasoline additive for blending in the preparation of
unleaded gasolines.

Ethyl benzene, also an aromatic hydrocarbon, is a derivative of
benzene with one of the hydrogens on the ring substituted with an
ethyl group.  It is a colorless, flammable liquid that is
essentially insoluble in water.  It is a product of the cracking
and reforming catalysis from crude petroleum oil.

                               -83-

-------
parameters.  However, by using the assumptions made in previous
sections, some reasonable choices can now be made.

There are thousands of individual organic chemical compounds in
the many commercial petroleum products. Consideration of only
those hydrocarbons appreciably ubiquitous at potential sources of
petroleum hydrocarbon release reduces the  number to under 15 -
three straight-chain alkanes, six branched-chain alkanes, and six
alkyl benzenes.  One of the elements, lead (Pb), is present at a
concentration significantly higher than any of the others, and
two non-petroleum organics, ethylene dibromide and ethylene
dichloride, account for the majority of the additives although
they are only present in a few parts per million (approximately
100 ppm to 300 ppm).

The 15 petroleum organics can be further reduced in number by
considering only the five elements that represent 60 percent of
those present in appreciable quantities.  Including the element
lead and the additives EDB and EDC, eight potential choices that
can be used as indicators for the presence of a gasoline leak can
be identified as follows:
     2    straight-chain alkanes
     2    branched-chain alkanes
     1    alkyl benzene
     1    element
     2    additives

                               -82-

-------
area soils  (117) and may not be seen in pore-water samples above
the detection limit.

Ethvlene dichloride (EDO or 1, 2-dichloroethane is similar to
EDB with chlorine substituted for bromine.  It is a liquid with a
density greater than water and very high water solubility.  Both
of these compounds are used as scavengers for lead in engines
that use leaded gasoline.
Tetraethvl lead is an organo-metallic compound that is used as a
gasoline additive to prevent knocking in automotive engines.  It
is used only in leaded gasolines.  It is essentially insoluble in
water, but is soluble in benzene, gasoline and other similar
organic compounds.

Each of these chemical and physical properties will be important
in determining the chemical's behavior in the subsurface
environment.  For example, benzene has a relatively high vapor
pressure, a specific gravity less than water, a viscosity
slightly less than water, and a high water solubility.  These
properties can now be used to predict how this compound will be
distributed in the subsurface - the solid phase, the liquid
phase, and the vapor phase.  It will dissolve in water to the
extent of its equilibrium solubility and will then form a
separate phase which is immiscible in water but will float on the

                               -85-

-------
water table since it has a specific gravity less than water.  Its



high vapor pressure will allow  a portion of it to vaporize and



enter the vapor phase and its relatively low viscosity will allow



it to migrate readily through the soil.







One need in particular is to differentiate between  "old" leaks



and "new" leaks.  In many instances for existing installations,



past operations will have released some amount of product prior



to when the requirements for monitoring were recognized.  When a



monitoring system is installed, background values of organic



compounds may be high enough that if the tank now begins to leak,



it will not be recognized because of the high background.  In



particular, this will be the case when the parameters that are



being used to monitor leaks are the more volatile aromatic



hydrocarbons commonly used; such as benzene, toluene and xylene,



and if attention is given exclusively to the individual



concentrations of each organic  compound.







Comparison of the relative ratios of leaked organic liquid



constituents to the relative ratios of the constituents found in



automotive gasoline may provide a possible means to differentiate



between an "old" leak and a "new" leak.  Over time, the various



reactions and interactions within the soil environment which have



been discussed earlier (differences in adsorption, chemical and



biological degradation, volatilization, vapor transport and



dissolution, and partitioning in soil pore-liquid between the





                               -86-

-------
various organic components) will all begin to change the relative
ratios of the leaked gasoline in the soil environment from that
in the original leak source.  Thus, over time, the "old" or aged
leaked gasoline organic component ratios will be different from
the "new" leaked gasoline organic component ratios.  While
background concentrations of the leaked residual "old" gasoline
components will alter the concentrations of the "new" leaked
gasoline components to some extent, the ratios should still be
close.  By obtaining data from existing old leaking tank
installations and comparing these ratios to the new product
ratios, the validity of this method of differentiating an "old"
leak from a "new" leak can be established.  The effectiveness of
this approach is dependent upon the recognition that changes in
gasoline product blends and source crudes occur with time.

                            CONCLUSION

A review of the cited literature reveals that it is difficult,
and perhaps impossible, to obtain pore-liquid samples which are
not altered by the sampling process (this concept is roughly
analogous to Heisenberg's Uncertainty Principle for quantum
mechanics).  Investigators should choose sampling devices and
methods which provide the least altered samples.  However, cost
considerations will dictate a point at which increased sample
representativeness is not practical.  At this point, it is the
investigator's responsibility to document the types of alteration

                               -87-

-------
caused by the sampling process.  The research necessary to
quantify these alterations has increased in recent years as
vadose zone monitoring concepts have matured.

Early warning monitoring of contaminant movement in the vadose
zone requires knowledge of potential release sources, an
understanding of the chemical and physical properties of the
potential contaminant, and familiarity with its behavior in the
subsurface environment.  The vadose zone is a widely varied and
typically nonhomogeneous environment which includes regimes of
both unsaturated and saturated flow.  No one single indicator
parameter can currently be identified for use in all systems that
must be monitored.  Specific indicator parameters must be
identified for use in monitoring each specific potential
contaminant material and application.  Ideal properties of an
indicator parameter are dictated by the factors identified in
this report.

Liquid alterations caused by the sampling process are important.
Probably, however, alterations of non-dilute solutions are
generally less significant than the inherent, spatial
variabilities of pore-liquid chemistries.  Such variabilities are
caused by preferential flow and physical heterogeneities (105,
106, 107).  As a result, even an unaltered pore-liquid sample
should only be viewed as representing temporally the sample
location and not representing spatially any other point.  This

                               -88-

-------
limitation does not detract, however, from the value of vadose
zone monitoring systems in a comprehensive ground-water
monitoring and protection program.  The primary purpose of
installing such systems is early warning and detection of
pollutants moving from a release source.
                               -89-

-------
                        LITERATURE CITED
 1. Wilson, L.G., Monitoring in the Vadose Zone;  A Review of
    Technical Elements and Methods, U.S. Environmental
    Protection Agency, EPA-600/7-80-134, 1980.

 2. Morrison, R.D., Ground Water Monitoring Technology, Timco
    MFG., Inc., Prairie DU Sac, Wisconsin, 1983.

 3. Law Engineering Testing Company, Lysimeter Evaluation Study,
    American Petroleum Institute, May 1982, 103 pp.

 4. Brown, K.W., Efficiency of Soil Core and Soil Pore-Liquid
    Sampling Systems, U.S. EPA/600/52-86/083, Virginia,
    February, 1987.

 5. Everett, L.G.; McMillion, L.G., Ground Water Monitoring
    Rev.. 1985, vol 5, pp 51-60.

 6. Everett, L.G., Ground Water Monitoring Rev., 1981, vol 1, pp
    44-51.

 7. Everett, L.G.; Wilson, L.G.; and McMillion, L.G., Ground
    Water. 1982, vol 20, pp 312-324.

 8. Wilson, L.G., Ground Water Monitoring Rev., 1982, 2., 31-42.

 9. Everett, L.G.; Hoylman, E.W.; Wilson, L.G.; McMillion, L.G.,
    Ground Water Monitoring Rev., 1984, vol 4, pp 26-32.

10. Thomas, G.W., and Phillips, R.E., J. Environ. Qual., 1979,
    vol 8, pp 149-152.

11. Anderson, L.D., Ground Water, 1986, vol 24, pp 761-769.

12. Severson, R.C.; Grigal, D.F., Water Resources Bull., 1976,
    vol 12, pp 1161-1169.

13. Shuford, J.W.; Fritton, D.D.; and Baker, D.E., J. Environ
    Qual. 1977, vol 6, pp 736-739.

14. Tyler, D.D.; Thomas, G.W., Jour Environ Qual, 1977, vol 6,
    pp 63-66.

15. Wilson, L.G. Thirteenth Biennial Conference on Ground Water,
    September 1981, p 38.

16. Wilson, L.G., Groundvater Monitoring Rev, 1983, vol 3, pp
    155-165.
                              -90-

-------
                  LITERATURE CITED (Continued)
17. Robbins, G.A.; Gemmell, M.M., Ground Water Monitoring Rev,
    1985, vol 5, pp 75-80.

18. Merry, W.M.; Palmer, C.M., Proceedings of the NWWA
    Conference on Characterization and Monitoring of the Vadose
    Zone, Nat Water Well Assoc,  1985, p 107.

19. Permit Guidance Manual on Unsaturated Zone Monitoring for
    Hazardous Waste Land Treatment Units, U.S. Environmental
    Monitoring Systems Laboratory, Office of Solid Waste and
    Emergency Response, EPA/530-SW-86-040, 1986a, 111 pp.

20. Ball, J.; Coley, D.M., Proceedings of the Sixth National
    Symposium and Exposition on Aquifer Restoration and
    Groundwater Monitoring, NWWA/EPA, 1986, p 52.

21. Hillel, D., Fundamental of Soil Physics, Academic Press,
    Inc., Orlando, Florida, 1980, p 413.

22. Soilmoisture Equipment Corporation, Sales Division, Catalog
    of Products, Santa Barbara, California, 1988.

23. Neary, A.J.; Tomassini, F., Can J. Soil Sci, 1985, vol 65,
    pp 169-177.

24. Wolff, R.G., Am J. of Sci. 1967, vol 265, pp 106-117.

25. Wood, W.W., Water Resources Res. 1973, 9, pp 486-488.

26. Wilson, L.G., in Ground Water and Vadose Zone Monitoring;
    Nielsen, D.M.; Johnson, A.I., Eds; Am Soc Testing Materials,
    Philadelphia, PA, 1990, pp 7-24.

27. Debyle, N.V.; Hennes, R.W.; Hart, G.E., Soil Sci, 1988, vol
    146, pp 30-36.

28. Bottcher, A.B.; Miller, L.W.; Campbell, K.L., Soil Sci,
    1984, vol 137, pp 239-244.

29. Peters, C.A.; Healy, R.W., Ground Water Monitoring Rev,
    1988, vol 8, pp 96-101.

30. Timco Manufacturing, Inc., Timco Lvsimeters, Sales Division,
    Prairie DU Sac, Wisconsin, 1988.

31. Linden, D.R., U.S. Pep. Agric. Technical Bull.. 1977, 1562.
                              -91-

-------
                  LITERATURE CITED (Continued)
32. Rhoades, J.D.; Oster, J.D., In Methods of Soil Analysis.
    Agronomy; Klute, A. Ed., Am Soc Agron, Soil Sci Soc Am,
    Madison, Wisconsin, 1986, Vol. 9, pp 985-1006.

33. Methods of Soil Analyses, Klute, A. Ed., Agronomy, American
    Society of Agronomy, Soil Sci Soc Am, Madison, Wisconsin,
    1986, Vol. 9.

34. Brose, R.J.; Shatz, R.W.; Regan, T.M., Proceedings of the
    Sixth National Symposium and Exposition on Aquifer
    Restoration and Ground Water Monitoring, Nat Water Well
    Assoc, 1986, pp 88-95.

35. Cole, D.; Gessell, S.; Held, E., Soil Sci Soc Am Proceed,
    1968, vol 25, pp 321-325.

36. Wengel, R.W.; Griffin, G.F., Soil Sci Soc Am J, 1971, vol
    35, pp 661-664.

37. Brown, K.W., Thomas, J.D., and Aurelius, M.W., Soil Sci Soc
    Am J. 1985, vol 49, pp 1067-1069.

38. Chow, T.L., Soil Sci Soc Am J. 1977, vol 41, pp 19-22.

39. Quin, B.F.; Forsythe, L.J., New Zealand J. Sci, 1976, vol
    19, pp 145-148.

40. Long, F.L., Soil Sci Soc Am J, 1978, vol 42, pp 834-835.

41. Starr, M.R., Soil Sci. 1985, vol 140, pp 453-461.

42. Mott Metallurgical Corporation, Sales Division, Catalog of
    Products, Farmington, Conn., 1988.

43. Smith, C.N.; Carsel, R.F., 1986, Soil Sci Soc Amer J, vol
    50, pp 263-265.

44. Smith, S.A.; Small, G.S.; Phillips, T.S.; Clester, M., Water
    Quality in the Salt River Pronect. A Preliminary Report,
    Salt River Project Water Resource Operations, Ground Water
    Planning Division, Phoenix, Arizona, 1982.

45. Parizek, R.R.; Lane, B.E., J. Hvdrol. 1970, vol 11, pp 1-21.

46. Trainor, D.P., M.S. Thesis/Independent Report, The
    University of Wisconsin, Madison, Wisconsin, 1983.
                              -92-

-------
                  LITERATURE CITED (Continued)
47. Young, M, Proceedings of Monitoring Hazardous Waste Sites,
    Geotechnical Engineering Division/ Am Soc Civil Engin,
    Detroit, Mich., 1985.

48. Morrison, R.D.; Tsai, T.C., Modified Vacuum-Pressure  ...
    Lysimeter for Vadose Zone Sampling, Calscience Research
    Inc., Huntington Beach, California, 1981.

49. Nightingale, H.I.; Harrison, D.; Salo, J.E., Ground Water
    Monitoring Rev/ 1985, vol 5, pp 43-50.

50. Knighton, M.D.; Streblow, D.E., Soil Sci Soc Am J, 1981, vol
    45, pp 158-159.

51. BAT Envitech, Inc., Sales Division, Catalog of Products, BAT
    Envitech Inc.,  Long Beach, California, 1988.

52. Haldorsen, S.; Petsonk, A.M.; Tortensson, B.A., Proceedings
    of the NWWA Conference on Characterization and Monitoring of
    the Vadose Zone, Nat Water Well Assoc, 1985, p 158.

53. Levin, M.J.; Jackson, D.R., Soil Sci Soc Am J, 1977, vol 41,
    535-536.

54. Stevenson, C.D., Environ Sci Technol. 1978, vol 12, pp 329-
    331.

55. Wagemann, R.; Graham, B., Water Res, 1974, vol 8, pp 407-
    412.

56. Sales Division, Catalog of Products, Cole-Panner Instrument
    Company, 1988.

57. Hornby, W.J.; Zabick, J.D.; Crawley, W., Ground Water
    Monitoring Rev, 1986, vol 6, pp 61-66.

58. Sales Division, Catalog of Products, Corning Glass Works,
    New York, 1988.

59. Duke, H.; Kruse, E.; Hutchinson, G., USDA Agricultural
    Research Service,  ARS, 1970, 41-16.

60. Tanner, C.B.; Bourget, S.J.; Holmes, W.E., Soil Sci Soc Am
    Proceed. 1954, vol 18, pp 222-223.

61. Cole, D.W., Soil Sci. 1958, vol 85, pp 293-223.
                              -93-

-------
                   LITERATURE CITED (Continued)
 91. Riggs, C.O., Proceedings of the Workshop on Resource
     Conservation Recovery Act Ground Water Monitoring
     Enforcement: Use of the Technical Enforcement Guidance
     Document and Compliance Order Guide/ Am Soc of Testing
     Materials, 1987.

 92. Hackett, G., Ground Water Monitoring Rev, 1987, vol 1, pp
     51-62.

 93. Riggs, C.O.; Hatheway, A.W., Proceedings of the ASTM
     Conference on Ground Water Technol, Am Soc of Testing
     Materials, 1986.

 94. Hackett, G., Ground Water Monitoring Rev, 1987, vol 7, pp
     51-62.

 95. Taylor, T.W.; Serafini, M.C., Ground Water Monitoring Rev,
     1988, vol 8, pp 145-152.

 96. Drainage for Agriculture. Schilfgaarde, J.V., Ed., Agronomy
     Series, American Society of Agronomy, Madison, Wisconsin,
     1974, Number 17.

 97. Donnan, W.W.; Schwab, G.O.  In Drainage for Agriculture,
     Schilfgaarde, J.V., Ed., Agronomy Series, American Society
     of Agronomy, Madison, Wisconsin, 1974, Number 17, pp 93-114.

 98. Gilliam, J.W., Daniels, R.B.; Lutz, J.F., J. Environ Qual,
     1974, Number 17, pp 93-114.

 99. Gambrell, R.P; Gilliam, J.W.; Weed, S.B., J. Environ Qual,
     1975, vol 4, pp 311-316.

100. Eccles, L.A.; Gruenberg, P.A., Proceedings: Establishment of
     Water Quality Monitoring Programs, Am Water Resources Assoc,
     1978, p 319.

101. Gilliam, J.W.; Skaggs, R.W.; Weed, S.B., J. Environ Qual,
     1979, vol 8, pp 137-142.

102. Jacobs, T.C.; Gilliam, J.W., J. Environ Qual, 1985, vol 14,
     472-478.

103. Wilson, L.G.; G.G. Small, Hydraulic Engineering and the
     Environment, Proceedings 21st Annual Hydraulics Specialty
     Conference, Am Soc Civil Engin, 1973, 427.
                               -96-

-------
                  LITERATURE CITED (Continued)
78. Dunlap, W.J.f Some Concepts Pertaining to Investigative
    Methodology for Subsurface Process Research, U.S.
    Environmental Protection Agency, 1977, p 167.

79. U.S. Environmental Protection Agency, RCRA Groundwater
    Monitoring Technical Enforcement Guidance Document, Office
    of Waste Programs Enforcement, Office of Solid Waste and
    Emergency Response, OSWER-9950.1, 1986b, p 208.

80. Reeve, R.C.; Doering, E.J., Soil Sci. 1965, vol 99, pp 339-
    344.

81. Pickens, J.F.; Cherry, J.A.; Coupland, R.M.; Grisak, G.E.,
    Merritt, W.F.; and Risto, G.A., Ground Water Monitoring Rev,
    1986, vol 6, pp 322-327.

82. Patton, F.D.; Smith, H.B., Groundwater Contamination Field
    Methods. Am Soc of Testing Materials, STP963, Philadelphia,
    PA, 1988.

83. Pickens, J.F.; Grisak, G.E., Ground Water. 1979, vol 17, pp
    393-397.

84. Campbell, M.; Lehr, J., Water Well Technol. McGraw-Hill Book
    Co., New York, New York, 1973, p  681.

85. Scalf, M.R.; McNabb, J.F.; Dunlap, W.J.; Cosby, R.L.;
    Fryberger, J., Manual of Ground Water Sampling Procedures,
    Nat Water Well Assoc, Ohio, 1981, p 93.

86. Minning, R.C., Proceedings of the Second National Symposium
    on Aquifer Restoration and Ground Water Monitoring, Nat
    Water Well Assoc, Columbus, Ohio, 1982, p 194.

87. Richter, H.R.; Collentine, M.G., Proceedings of the Third
    National Symposium on Aquifer Restoration and Ground Water
    Monitoring. Nat Water Well Assoc, Columbus, Ohio, 1983, p
    223.

88. Gass, T.E., Water Well J. 1984, vol 38, pp 30-31.

89. Driscoll, F.G., Groundwater and Wells, Johnson Division, St.
    Paul, Minnesota, 1986, p 1089.

90. Keely, J.F.; Boateng K., Ground Water, 1987, vol 25, pp 3-4.
                              -95-

-------
This  form  is  required  for
all  research  products.         ORD CLEARANCE FORM
         Report
                               2. Series
                                                        3. Lab/Office Draft No.
                                                         TIP* 89-0889A/EMSL-LV#0889A
                                                                                         It. Copyright Permission     j
                                                                                         	 Yes (Attached)	 No_X N/AJ
   5A. Original Document Title:

      Criteria for Selection of Equipment and Indicator Parameters for Direct  Pore-Liquid Samplers

   58. Final Document Title, if changed:

      Criteria for Selecting Monitoring Devices and Indicator Parameters for Direct Pore-Liquid Sampling of Petroleum
      Hydrocarbon Contaminated Sites
  6. Author(s),  Affiliation, and Address (identify
     EPA authors with  Lab/Office)
  L. G. Everett, S.  J.  Cullen, R. G. Fessler; University of
    California at Santa Barbara, 93101
  D. W. Dorrance, ENSR, 3000 Richmond Avenue, Houston, TX  77098
  L. G. Wilson,  University of Arizona, Tuscon, AZ  85721
  10.  DU/Obj./PPA/Project/Deliverable Output No.

       D109/I15/01/0889A
11.  Technical Information (Program)  Manager

  	     jP  -  &•  ^—£~&rr-~*-^^)
  Signature   Faye E. Cromar,  OPS
                                                Date/   '
    Signature of sender (if other than TI(P)M)   Date
  12.  Signature/Date^

                  U.
   Laurence A. Eccles, AMU
   Joseph J. DLugosz, Chief, AMU
  Eugene P. Meier, Director,  AMD
                                                                  7. Project Officer/FTS Telephone

                                                                    Laurence A. Eccles/545-2385
                                                                 8. Contract/IAG/Assistance Agreement No.

                                                                    CR813350
                                                                 9. Product (check one)

                                                                 	 Peer Reviewed Journal Article (complete block  13)

                                                                 XX Published Reports: Project Report/Summary/
                                                                    (magnetic tapes/floppy diskettes). Method,Research
                                                                    Report, User's Manual. Handbook, Criteria
                                                                    Document, Health Assessment Document, Technology
                                                                    Transfer Report, Proceedings(Conferences,Symposia,
                                                                    workshops)

                                                                 	Symposium Papers .and Book Chapters

                                                                 	 Internal Report (distribution restricted
                                                                    to EPA

                                                                 	 Miscellaneous (newsletter, research brief,
                                                                    trade paper) non-peer reviewed journal article,
                                                                    abstracts, oral  presentations (complete block  13)

                                                                 _ Unpublished Report
                                                                 13. Bibliographic Citation (include Month/Year)
                                                                  	 Accepted     	 Published
                                                                   14. Distribution (use block  16,  if necessary)
  ISA. This Publication
   X_ does not have policy implications for EPA
   	 has policy implications  for EPA
        (memo attached) *
16. Comments
TU-'s document has been reviewed.  It does not conta(l/( policy implicat
                                                                     7
                                                                   tions.
                                          Eugene P. Meier,  Director, AMD
  EPA-362(Cin)(Rev.3/87)  *  If policy implications, attach memo.

-------
                                     TECHNICAL MANUSCRIPT REVIEW FORM
Title/ Draft No.
Criteria for Selecting Monitoring
Devices and Indicator Parameters for
Direct Pore-Liquid Sampling of Petroleum
Hydrocarbon Contaminated Sites
Date Review Requested Date Review R equired
11/18/90 11/21/90
Type of Publication/Audience
Application User's Guide
Review Coordinator (e.g.. PO, TIM. Supervisor)
Authoits)
Everett, Cullen, Fessler, Dorrance, Wilson
Project Officer/Organization/Address '
Mr. Larry Eccles
USEPA-EMSL
944 East Harmon Ave.
Las Vegas, NV 89119
Reviawer/Organization/Address
Mr.' Bill Ullojn
Vadose Research, Inc. , ,.
351 21sL — alreel N~TT '"^.-''^ ('.' '^''j'-J i-.> '•••,-- f '•
Canton, Ohio 44701. .s
(,216) 454-6706
       You are asked to review and comment on the attached manuscript. Feel free to make notations on the manuscript as well as in
       comments section below, particularly regarding your recommendations for revisions. If you are unable to review the
       manuscript by the required data above, please return it now. Your suggestions for alternate or additional reviewers wilt be
       welcomed.
SUMMARY RATING
Please rate the manuscript as follows: Satisfactory Unsatisfactory

lalityof data and validity of analytical ,
woundness of conclusions ..
Editorial quality .^___ ^


-?>.*-.•:- /><•//.• >cm" . J
RECOMMENDATIONS
tJ2rf) Acceptable as is
Q (2) Acceptable after minor revision
D (3) Acceptable after major revision
G (4) Not acceptable
If you have checked either 3 or 4,
please specifically state reason(s) in
the comments space below.
Jj,JL* &(t&^ K-K * J
Reviewer's Signature Date .
Commsnts: (Use extra $fie«ts if Headed.)

-------
TECHNICAL MANUSCRIPT
Title/ Draft No. Auth
Criteria for Selecting Monitoring _
Direct Pore-Liquid Sampling of Petroleum JB<
Hydrocarbon Contaminated Sites Mr
US
94
1 a
jDate Review Requested Date Review R equirad
11/19/90 11/21/90 Revie
Type of Publication/Audience Mi
Application User's Guide D:i
Review Coordinator (e.g., PO. TIM, Supervisor) !
1 «
P«
REVIEW FORM
or(s)
ett, Cullen, Fessler, Dorrance, Wilson
R Off leer/Organization/ Address '
. Larry Eccles
EPA-EMSL
4 East Harmon Ave.
s Vegas, NV 89119
wer/Organization/Address
-. 'Ed W. Hoylman
.rector of Hydrology
;rzog & Associates
118 Redwood Way, Ste. 203
staluma, CA 94954 (707) 792-5622
You are aaked to review and comment on the attached manuscript. Feei free to make notations on the manuscript as well as in
comments section below, particularly regarding your recommendations for revisions. If you are unable to review the
manuscript by the required date above, please return it now. Your suggestions for alternate or additional reviewers will be
welcomed.
SUMMARY RATING
Please rate the manuscript as follows: Satisfactory Unsatisfai
COPtflHT 3"P9 ^^ .
QrgprvtatfO" '"** P'990ntiltijtf -
^w-t*z^cc<-^^
^fyU^O'ls Aj^^c^f^ C^M^W^.-V^ &SU-
) .
15/82)
s RECOMMENDATIONS
«ory ^a»1 ' Acceptable as is
Q (2) Acceptable after minor revision
LJ (3) Accepiaoie after major revision
~~^~ O (4) Not acceptable
If you have checked either 3 or 4,
^^MMM ntaafio tnprifirflllv QfatA rpn^nn/4t ifi
the comments space below.

f . rf/ fl
/^y^x^y^W^^ llftfl?C
1 Reviewer'^ rfignaty/e'*^ Da(e /
/ /
/Ct^^y jj/lu^^i^^\_ &
/T /

-------
 OI'•>*>'«/•,
o" 	 '•!
             UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                         OFFICE OF RESEARCH AND DEVELOPMENT
                ENVIRONMENTAL MONITORING SYSTEMS LABORATORY-LAS VEGAS
                                  PO. BOX 93478
                            LAS VEGAS. NEVADA 891 93-3478
                            (7O2/798-21OO- FTS 545-21OO)
                                 DEC 2 01990
SUBJECT:
Transmittal of Deliverable Report (89-0889A), "Criteria for
Selecting  Monitoring Devices and Indicator Parameters for Direct
Pore-Liquid Sampl>rfg. of Petroleum Hydrocarbon Contaminated Sites"
(USER'S GUIDE) . '   '
FROM:
TO:
Robert N. Snelli
Acting Director/^
             <
-------
      7.   Provides approach  to  indicator selection and data  interpretation.

      8.   Discusses  factors  relevant  to sampling for petroleum products.

      9.   Explains importance of  fate and  transport considerations  in
           selecting  indicators.

      10.   Guides development of QA/QC plans.

      The concept of  monitoring  pore-liquids  is  the most certain approach  to
 identifying  the migration of contaminants.   Representative pore-liquid samples
 provide direct evidence of the  presence or absence of contaminants, whereas
 other available techniques do not; e.g., soil gas sampling and neutron
 logging.   The research supporting the preparation of this document  focused on
 evaluating the performance and  potential utility of available pore-liquid
 sampling equipment.  The underlying research also indicates  that  there are
 significant  limitations with available pore-liquid sampling  equipment.  Ve
 identified many of the limitations several years ago and have been  working on
 the development of new equipment  to improve  our pore-liquid  sampling
 capabilities.  Presently, pore-liquid sampling needs to be used in  conjunction
 with  other techniques, such as  soil gas sampling and neutron logging, for an
 effective  monitoring network; i.e., pore-liquid sampling alone is not intended
 to provide all of the information needed to  determine the probable  migration
 of all contaminants.  The objective in preparing this guide  is to assist
 investigators in determining where pore-liquid sampling fits into their
 monitoring networks.

      This  output meets its intent  of  contributing to the acceleration and
 acceptance of vadose zone monitoring  techniques.  The EPA requires  vadose
 zone  monitoring for  land treatment and has determined that it should be
 included in  monitoring strategies  for other  types of regulated facilities.  In
 California,  vadose zone monitoring already is required for most types of
 operating  and closed hazardous and solid waste  facilities.   California
 regulations  make reference to official EPA guidance, EPA/530SV86040, "Permit
 Guidance Manual on Unsaturated Zone Monitoring  for Hazardous Waste  Land
 Treatment  Units," and rely heavily on other  EPA publications for  implementing
 their vadose zone monitoring strategy.  In addition, this document  will
 provide an essential reference  for a  new official OSV guidance document for
 vadose zone  monitoring.

 Attachment

 cc (w/attachraent):
 Neal  Durant, OSW (OS-321)
Tom Baugh, OMMSQA (RD-680)
 Margaret Hawkins, OMMSQA (RD-680)

 cc (w/o attachment)
Rick  Linthurst,  OMMSQA (RD-680)
Don Clay,   OSWER (OS-100)
Vern Myers, OSW (OS-321)

-------