EPA-600/2-84-024
                                                             January 1984 •  .

             A  GUIDE TO THE SELECTION OF MATERIALS
                FOR MONITORING WELL CONSTRUCTION
                    AND GROUND-WATER SAMPLING
                               by
                     Michael  J.  Barcelona
                          James  P.  Gibb
                         Robin A.  Miller

                   Illinois  State  Water Survey
          Department of  Energy  and Natural  Resources
                   Champaign*  Illinois   61820
                        EPA  No.  CR-809966
                       Project  Officers

                       / Marion  R.  Scalf
                       Leslie G.  McMillion

Robert S. Kerr Environmental Research  Laboratory,  Ada, Oklahoma
                              and
Environmental Monitoring Systems Laboratory,  Las Vegas, Nevada
             U.S. Environmental  Protection  Agency
                    REPRODUCED ti
                    NATIONAL TECHNICAL
                    INFORMATION SERVICE
                        US DIPHRTMlNi Of COMMfRCE
                          SPRIHCFitlC1, VA. 22161

-------

-------
     The information in this document has been funded wholly or in part
by the U.S. Environmental Protection Agency under assistance agreement
No, CR-8Q9966.  It has been subject to the Agency's peer and administra-
tive review, and it has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
                                     ii

-------

-------
                                     FOREWORD
     The  Environmental Protection Agency was established to coordinate
administration of the major Federal programs designed to protect the
quality of our environment.

     An important part of the Agency's effort involves the search for
information about environmental problems, management techniques and new
technologies through which optimum use of the Nation's air, land, and
water resources can be assured and the threat pollution poses to the
welfare of the American people can be minimized.

     EPA's Office of Research and Development conducts this search through
a nationwide network of research facilities.  As one of these facilities,
the Robert S. Kerr Environmental Research Laboratory is the Agency's center
of expertise for investigation of the soil and subsurface environment.  Per-
sonnel at the laboratory are responsible for management of research programs
to:  (a) determine the fate, transport and transformation rates of pollutants
in the soil, the unsaturated zone and the saturated zones of the subsurface
environment; (b) define the processes to be used in characterizing the soil
and subsurface environment as a receptor of pollutants; (c) develop techniques
for predicting the effect of pollutants on ground water, soil  and indigenous
organisms; and (d) define and demonstrate the applicability and limitations  '
of using natural processes, indigenous to the soil and subsurface environment,
for the protection of this resource.

     This report contributes to that knowledge which is essential in order
for EPA to establish and enforce pollution control standards which are
reasonable, cost effective, and provide adequate environmental  protection
for the American public.

    '•The subsurface environment of ground water presents many challenges to
effective monitoring efforts.   Physical, microbiological  and geochemical
forces interact over variable time frames as ground-water quality character-
istics evolve.   Monitoring data'.should be collected with minimum-disturbance
of the subsurface.   This goal  implies careful  attention to the  details of
sound drilling, well  construction, and sampling methodologies.

     The complex nature of ground-water monitoring further demands that cost-
effective choices of materials, target chemical  constituents,  and procedures
are made prior to implementation of the network design.   Attempts to cut
costs for the sake of short-term "savings" can result in substantial  added
expense.   This  may occur when  the goals of the monitoring effort are expanded
and information needs require  increased analytical detail.   There are few
merits to penny-wise, pound-foolish approaches to ground-water  monitoring.
                                             Hinton W.  Hall
                                                Director
                              Robert S.  Kerr Environmental  Research Laboratory

-------

-------
1  !

-------

-------
                                CONTENTS
Foreword ........................................ .......................................   iii
Figures .............. - ............................ . — .................................   vi
Tables ................................................. ' .................................  vii
Acknowledgements ....... . .......... . ........... ..... .................................  viii
Section 1 .  Introduction ............. . .............................................. ...    1
      Need for and Importance of Monitoring ....................................    2
Section 2 .  Objectives ............................... , ..................................    4
Section 3.  Ground-Water Monitoring Requirements ......... . .....................    5
      Resource Conservation and Recovery Act (RCRA) ................ . .......    5
      The Comprehensive Environmental Response, Compensation and Lia-
      bility Act (CERCLA) [[[    5
     , State Requirements for Ground-Water Monitoring ........................    6
Section 4.  Background .............................. . .................................    7
      Previous Studies ......................... . .......................... . ...........    7
        Monitoring in General ............... . ......................................    7
        Monitoring Well Design and Construction ................................    8
        Ground-Water Sampling [[[    9
        Materials Used in  Monitoring Efforts .....................................  10
      Current Research Efforts ............................... , — . ................  11
Section 5. -Monitoring Wells and Sampling Apparatus .............................  13
      Monitoring Well Components ..................... . . ........................ .  13
        Well Location ......... . ...... . ...................................... .........  14
        Well Diameter ............ . ..................................................  14
        Well Depth ............ , [[[  15
        Well Design and Construction Materials ..................................  15
        Drilling  Methods ............................... . ............ . . ..............  17
        Well Development .................... ... ........................ . ..... , .....  20
      Sampling Apparatus [[[  22
Section 6.  Sampling Strategy [[[ . .....  26
      Effects of Subsurface Conditions on Ground-Water Quality and
      Sampling ........................... .. ...........................................  26
      Effects of Hydrologic Conditions on Sampling Strategy ____ ..... ..........  27
Section 7.  Evaluation of Materials ....... . .......... . ............................ ....  31'
      Overview of Subsurface Conditions ..........................................  31
        Chemical Properties of Water and Their Effects on Various
        Materials ......... ' ................................ . ...........................  31
      Candidate Test Solutions for  Materials' Evaluation .......... . ....... . .....  33
      Preliminary Ranking  of Well  Construction/Sampling Materials ...........  34
      Evaluation of Selected Materials .............................................  37
        Teflon®  Well Casing .......... . ..............................................  38
        Stainless Steel Well Casing ........................... . .............. ........  38

-------
      Evaluation of Selected Materials (Continued)
        Casing Made from Other Ferrous Materials	,	  44
        Pumps Used  in Development	  45
        Grouts, Cements, Muds, and Drilling Fluids	<.,..  46
          Drilling aids	  46
          Seals, grouts and cements	  48
        Evaluation of Sample Collection Materials	  48
          Sampling devices	  49
          Tubing and transfer lines	  50
          Storage containers	,	  51
        Sources of Error in Monitoring Efforts .,	  52
          Comparison of analytical method performance with materials'
          related error	,	,	  53
Section 8. Cost Considerations	  58
      Description of an Example Monitoring Effort	  59
      Well Installation and Sampling Costs	,	,	  59
      Analytical Costs	-.	  61
      Project Cost Comparisons  for Selected Materials' Combinations  for
    ./ Networks of Varying Analytical Detail	  64
      Tailoring the Monitoring Well and Sampling Apparatus to the  Antici-
      pated Analytical Scheme	  65
Section 9, Conclusions	  66
Section 10, Recommendations	67
      General Recommendations	  67
      Specific Recommendations	  67
References	  69
                             LIST OF FIGURES

Figure                                                                      Page

5-1   Schematic diagrams of typical water supply (a) and monitoring (b) well
      installations	'  14
5-2   Schematic diagram of an air driven well development device, —,	   21
6-1   Percent of aquifer water versus time  for different transmissivities	   29
7-1   Sources of error  involved in ground-water monitoring programs  con-
      tributing to total  variance	   52
8-1   Capital costs for drilling and well construction of four point array	   61
                                       VI

-------
                              LIST OF TABLES

 Table                         •          ,                     ,  .            Page

 5-1    Description of Saturated  Zone-Sampling Devices for Use in Small-
       Diameter Boreholes and Monitoring Wells	:	   23
 7-1    Chemical Composition of Contaminated Ground Water Near Hazardous
       Waste Sites	•	'.	   33
 7-2   .Solution Composition for a Range of Ground-Water Conditions	   34
 7-3    Chemical Exposures Grouped  in General Solution Categories		   35
 7-4    Relative' Compatibility of Rigid Well Casing Material	   36
 7-5*   Relative Compatibility of Semi-Rigid or  Elastomeric Materials	'	   36
-7-6    Representatis'e Classes of Additives in Rigid PVC Materials  Used for
 1  ,    Pipe or Well Casing	   40
 7-7    Chemical "Parameters  Covered by NSF Standard; 14 for Finished Prod-.
 ,      ucts* and in Standard,Leach Tests...:'	   41
 7-8    Components of Drilling Fluids	.•	   47
 7-9    Frequency of Occurrence of Phthalate Esters in Waste water and Ground-
       Water Samples	'.	,	•	   50
 7-10   Analytical  Performance Data for Selected Water Quality
       Parameters.	>,	•.   54
 7-11   Analytical Performance  Data for Selected Inorganic Chemical
       Constituents	—;	'	   55
 7-12   Analytical Performance  Data for Selected Organic Chemical • j
       Constituents	v	   56
 8-1    Cost Estimates for Drilling, Well-Construction, and Sampling (1983
       Dollars)	:	'.	'	   60
 8-2    Description of Analytes and Costs for  Four Analytical Schemes for
       Ground-Water Samples	   62
 8-3    Analytical Cost Detail 	V	,	   63
 8-4    Comparison of Reanalysis Cost with Cost "Savings" on Materials	   63
'8-5   Total Project Costs  for Monitoring Programs ,	.•	   64
                                     vn

-------

-------
                             ACKNOWLEDGEMENTS

      This manual was prepared under cooperative agreement #CR 80996601  from
the U.S.  Environmental Protection Agency,  Office of Research  and Development,
under the direction of project managers Marion  R. Scalf of the EPA  R. S.  Kerr
Environmental Research Laboratory, Ada, Oklahoma, and Leslie G. McMillion of the
EPA Environmental Monitoring and Support Laboratory,  Las Vegas, Nevada.  ,
      The manual was written by Michael J. Barcelona, James P. Gibb, and Robin A.
Miller, all of the Water Survey Division  of the Illinois Department of Energy and
Natural Resources.
      A review panel  consisting  of agency  personnel,  ground-water  professionals,
representatives of industry and national  standards organizations participated in the
planning and review of the report. The authors wish to recognize the helpful guidance
and comments of the  following panel members:
         Gordon Bellen
         National Sanitation Foundation
         Olin Braids
         Geraghty & Miller, Inc.
         George Dixon
         USEPA Office of Solid Waste
         William Dunlap
         USEPA Robert S. Kerr Environmental Research Laboratory
         Robert Wilging and Robert Hinderer
         The B.F. Goodrich Company
      Particularly valuable inputs were received from many scientists and engineers
active in ground-water work and related fields. Among these contributors  the authors
would like to acknowledge the valuable comments, suggestions, and access to  unpub-
lished data from:
         Robert Brobst               City of Arttigo, WI
         Fred and James Doane        Industrial and Environmental Analysts
         Roy Evans                   USEPA/EMSL-Las Vegas, NV
         Thomas Imbriggiotta         U.S. Geological Survey
         Gregor Junk                 Iowa State University
         George Kish     '            U.S. Geological Survey
         Jerry Kotas                  USEPA Office of Drinking Water
         Jerry Leenheer               U.S. Geological Survey
         Paul Mills                   Meade Compuchem, Inc.
         Stan Mruk                  Plastic Pipe Institute
         Michael O'Hearn             Illinois State Water Survey
         Jack Sosebee                 Environmental Science and Engineering, Inc.
         Jerry Thornhill              USEPA/RSKERL-Ada, OK
         Bert Trussell                J.  M. Montgomery, Consulting Engrs.
         James Vennie                Department of Natural Resources, State  of Wisconsin
      The support of Pamela Lovett, Pamela Beavers, and a  host of other individuals too
numerous to mention is also greatly appreciated.
                                       vm

-------

-------
                                   SECTION 1
                                 INTRODUCTION
       Ground water provides the  base flow  of all  perennial streams and over 90
 percent of the world's freshwater resources. Ground water also is a source of one-half
 of the United  States' drinking water supplies (1). Yet until recently ground water has
 received only token scientific attention in terms of its quality and protection. Insufficient
 attention has been given to proper monitoring procedures to accurately determine the
 quality of ground water,                                                    -,
       With  the  beginning of the industrial  age,  a  variety of new_/chemicals were
 introduced into the hydrologic cycle by man's indiscriminate use and  waste disposal
 activities. "Since World  War II,  the explosive development of the  synthetic chemical
 industry added thousands of additional chemicals to the environment. Chemical Abstracts
 announced in  March 1972 that it had  registered  2,000,000 unique chemicals since
'January 1965" (2).  A large number of these organic and  inorganic chemicals enter
 both surface water and  ground  water not only through waste disposal but also as a
 result of normal use of these water resources.-
       The presence of many trace substances in ground water,  even  at very low
 concentrations, ma}' create short-term or long-term health hazards (3).  The inclusion
 of approximately 113 organic chemicals on the EPA priority pollutant list indicates the
 significance accorded these chemicals by U.S. health officials.
       The passage  of the Safe Drinking  Water  Act  (PL  93-523) in  1974  finally
 recognized ground water as a major source of drinking water and established standards
 of ground-water quality  protection. Later, passage of the Toxic Substance  Control Act
 (PL 94-469) and the Resource Conservation and  Recovery Act (PL 94-480) further
 recognized the significance of ground water  resources  and the  importance of their
 protection against the threats increasingly posed by human activities.
       Regulatory agencies are charged with regulating the disposal of waste to insure
 that the environment is  not adversely affected.  To accomplish this task,  these agencies
 must set design and operational  standards based on available technology  to minimize
 potential  pollution.  Disposal facilities then must  comply with these standards and
 monitor the effects  of their operation on  the  surrounding environment.  The use of
 wells or piezometers for collecting water samples  and water level data has been and
 probably  will continue to be the method for monitoring the effects of waste disposal
 facilities on ground  water.
       Considerable  research has been conducted to  develop analytical laboratory
 techniques to  detect the low levels  of the many constituents named in water  quality
 standards  or legislation related to ground-water  protection or waste management
 practice.  Monitoring well  construction, water sample  collection,  and  preservation
 techniques have been established by several different laboratories and  agencies in an
 attempt to  insure that water  samples  delivered  to the  laboratory are chemically

-------
representative of the  water contained in the ground. However,  there is considerable
controversy among laboratories, agency  policies, and researchers concerning proper
well construction and sampling techniques and appropriate procedures for preserving
the original chemical character of the samples. If monitoring wells and water samples
are to provide the performance yardstick for disposal facilities' design and operation,
the significance of various monitoring well construction and sampling procedures and
preservation techniques must be determined.


Need for and Importance of  Monitoring
      Ground-water monitoring is  essential to determine the quality of the nation's
ground-water resources and the effectiveness of ground-water pollution control regu-
lations.  Only through  the collection of accurate data  can  we ascertain the chemical
nature of our ground-water resources. Ground-water monitoring can be divided into
four types:  ambient monitoring, source monitoring, case preparation  monitoring, and
research  monitoring.
      Ambient ground-water quality monitoring  establishes  an understanding of char-
acteristic  regional water  quality variations  and changes  over time.  This  type  of
monitoring is normally accomplished through routine sampling of wells on an areal or
regional basis. The wells sampled are often public water supply, industrial or domestic
wells  as opposed  to specially constructed monitoring  wells.  The  sample collection
techniques also are often quite different  from those  for monitoring wells. Therefore,
data obtained from  ambient monitoring programs  may not  be comparable  to that
obtained by more rigorous well construction and sampling procedures. However, the
data are invaluable in detecting significant changes in water quality and protecting
public health.
      Source monitoring is the type normally conducted at a potential pollution source.
Source monitoring detects and quantifies the migration  of pollutants from potential
pollution  sources. Most regulatory  programs have monitoring guidelines that detail
minimum standards  for source monitoring. In an effort to minimize  cost to the
regulated community, a two-step approach to source monitoring has been used by most
states.
      The first step is a  minimum-requirement approach  based on the principle of
detecting leakage of pollutants from disposal or storage facilities. It is understood that
data collected  from a  detective  monitoring scheme  will  not define  the  extent  or
seriousness of a developing problem. Therefore, once the detective monitoring system
has sounded the alarm, most states require a more definitive monitoring program to
provide interpretive information concerning the extent and seriousness of the problem.
      The second step is interpretive monitoring which normally  requires expanded
information needs and the installation of additional wells. The purpose  of interpretive
monitoring  is to  define the limits and concentrations within the plume of pollutants
moving from the presumed source.  The addition of wells areally, and vertically nested
well sets often  is  required to adequately define plume  geometry  and concentration
gradients.
      Case-preparation or enforcement monitoring is undertaken by the regulatory com-
munity  to collect evidence,for prosecution of ground-water pollution cases. This type

-------
 of monitoring usually is developed in response to the language in the appropriate laws
 intended to be used in the prosecution of the case. The detailed information resulting
 from this type of monitoring generally is more specific with respect to concentrations
 and changes in  concentrations at the legal points of regulation. Further definition of
 plume geometry or concentrations normally  is not required for successful litigation.
       Research monitoring often results  in a level of data collection  far beyond that
 required for other types of monitoring  activities. Added  information is required to
 help understand-and document the  mechanisms controlling solute transport of pollu-
 tants. The collection of this type of monitoring data is  very rigoro.us and demanding
 although limited to specific research goals.
„ "     Selection of the types and levels of monitoring to be used in any situation should
 be made with care and judgment. The goals of a proposed monitoring program should
 be clearly stated and understood before consideration is given to what types of wells
 should be constructed, where they should be located, how deep they  should be, what
 materials-should be used, what chemical constituents should be analyzed, and  how
 samples  should  he collected. Successful and  cost-effective monitoring at  all  levels can
 be accomplished only after the 'goals or purposes of  the monitoring  program are
 thoroughly  understood.             .                          . -             '

-------
                                    SECTION 2
                                   OBJECTIVES
       This publication was produced as part of a cooperative agreement between the
 Illinois State Water Survey and the U.S. Environmental Protection Agency (EPA CR
 809966-01). The objectives of this project are as follows:
       1) To assess the state of our knowledge of the potential effects of well construction
    and pump and sampling materials on the integrity of ground-water samples, and
       .2) To develop  a guide  that will  facilitate the  selection of pumps and devices ,
    used to collect water samples, and materials used in the construction of monitoring
    wells.

       To achieve these objectives, the project was divided into two phases. Phase One
 entailed a thorough investigation and assessment of the existing literature. Phase Two
 involved laboratory and field studies to identify and assess efficient sampling techniques
 and materials for ground-water monitoring near waste disposal sites.
       This manual represents the culmination of the work in Phase One of the project.
 It aspires to fill in the gaps left by previous studies of ground-water monitoring. Subject
.areas  unexplored or treated superficially in the past are examined here in detail to
 present a more complete exposition of ground-water monitoring. These areas include
 the effects of materials on water samples, and the cost of materials vs' the'cost of the
 total monitoring project. Cost  is addressed from an important, but often overlooked,
 perspective: How the initial investment in materials suitable for a particular monitoring
 project can pay  dividends in terms of lower overall  project cost. Use of appropriate
 materials can obviate the need for the additional work involved in' designing and
 constructing new wells and in repeating sampling and analysis efforts when the use of
 inappropriate materials renders monitoring results unrepresentative of in-situ ground-
 water quality.          -/

-------
                                   SECTION 3
                  GROUNDWATER MONITORING REQUIREMENTS
       Requirements for ground-water monitoring exist at both the federal and the
 state levels. At the federal level, the Resource Conservation and Recovery Act (RCRA)
 and the Comprehensive Environmental  Response, Compensation, and Liability Act
 (CERCLA) provide for monitoring to regulate hazardous wastes  from generation to
 disposal to cleanup of substances accidentally released into the ground-water environ-
 ment.
 The Resource Conservation and Recovery Act (RCRA)
       Ground-water monitoring at applicable hazardous waste management facilities
 is required by regulations issued pursuant to  RCRA (Subpart F of 40 CFR  Parts 264
 and 265). While there are many intricate provisions in these rules, it is important to
 stress here that there are three basic monitoring objectives, and that there are two sets
 of requirements, one for existing facilities prior to permitting (Part 265) and one for
 permitted facilities (Part 264), The basic monitoring objectives are 1) to detect whether
 or not a facility is discharging hazardous wastes to the uppermost aquifer;  2) to
 determine whether concentrations of specific hazardous waste constituents are within
 prescribed limits; and 3) to measure the effectiveness of corrective actions.
       Prior to permitting, the rules are standardized. All facilities  follow the  same
 rules and the rules are self-implementing. That is, for existing (interim status) facilities
 the parameter selection is pre-set. This means that the federal rules provide no authority
 for the regulatory agency to require the use of specific materials or equipment for well
 construction  or sampling.  However, during  the  permitting  process, the regulatory
 agency will have the authority to evaluate an applicant's proposals and to make final
 specifications concerning well construction and sampling in the facility permit. There-
 fore, a major function of this document will'be to provide a resource  document to
 assist applicants and permit writers  in making these important determinations.
       The RCRA monitoring  regulations (Parts 264 and 265)  also require facility
 owners or operators to develop sampling and'analysis programs which  include moni-
 toring parameters, sampling schedules, sampling procedures, sample preservation and
 shipping procedures, and analytical procedures.
       For disposal facilities, the RCRA monitoring requirements are applicable through-
 out the post-closure care,period (Subpart G of 40 CFR Parts  264 and 265).


The Comprehensive Environmental Response, Compensation,
and Liability Act (CERCLA)
      While RCRA created one of the largest regulatory programs enacted by the
federal government to protect  ground  water, it could not address  the problem of

-------
abandoned waste sites  or  accidental spills of hazardous substances.  For  this reason,
Congress  enacted  CERCLA to deal with  the  cleanup problems created by those
uncontrolled hazardous waste sites where reponsibility cannot be assigned (e.g., aban-
doned sites) and accidents in the handling or transportation of hazardous substances.
The Act  provides for  the generation of a  trust (called Superfund) to  finance the
emergency cleanup of  accidentally  released hazardous substances and to assume the
financial liability of closed  permitted sites,
      At  present,  no specific systems of ground-water  monitoring are required by
CERCLA.  If hazardous contaminants from a waste facility or  accidental  spill are
suspected or known to be entering the ground water, the choice of monitoring equipment
and methodology to be used to determine the scope of the contamination is left to the
states. Monitoring systems may be patterned after those outlined in RCRA,  or states
may develop systems of their own.  Like  the  post-closure monitoring requirements of
RCRA, those of CERCLA are relatively vague and untested. As they are applied in
more cases, more specific recommendations, and perhaps regulations, for ground-water
monitoring may develop.
State Requirements for Ground-Water Monitoring
      Individual states may have their own requirements for ground-water monitoring.
While EPA allows states to take authority for imposing monitoring requirements under
RCRA and CERCLA, the actual requirements mandated by the various states may be
more stringent than those of these two acts. For this reason, it is important that those
involved in ground-water monitoring be  familiar with the requirements of the state in
which the monitoring is being performed. Such information  usually can  be  obtained
from the state's Environmental Protection Agency.

-------
                                   SECTION 4
                                 BACKGROUND
Previous Studies
      This section presents an overview of work previously published on the subject
of ground-water monitoring. These  works cover various aspects  of monitoring in
general, well design and construction, well  sampling, and materials'  considerations for
use in monitoring applications,

Monitoring in General
      The need for groundwater monitoring is evidenced in an article by Pettyjohn et
al. (4),  which presents an overview  of the  prevalence  and significance of organic
contamination in ground water. He explains where these organic compounds originate
and  includes observations  on their  chemistry, toxicity, mobility, sorption,  volatility,
dilution, biodegradation, and abiotic degradation.
      A number of authors and"researchers address  the federal  and state regulations
that  mandate monitoring.  The National Council of the  Paper Industry for Air and
Stream  Improvement (NCASI) (5) has  prepared a Guide to Groundwater Sampling
that describes the regulations developed under the Resource Conservation and Recovery
Act (RCRA), The Council highlights requirements for system design, sampling and
analysis, and monitoring preparation, evaluation and response. A USEPA publication
(6)  provides  guidance for owners and  operators of  hazardous-waste  land disposal
facilities in complying with the interim status requirements for  ground-water quality
monitoring in Subpart  F of 40 CFR  265. Clark and Sabel (7) report  on a survey
conducted to learn of states' requirements for monitoring wells, chemical analyses, and
data interpretation.
      The importance  of defining the  purpose of the monitoring program before
designing the monitoring system is addressed by Geraghty  and Miller, Inc. (8),  They
note that site-specific hydrogeologic data also play a part in system design and describe
elements of a comprehensive hydrogeologic investigation. On  monitoring in general,
they address  the design  and installation of monitoring well networks, including deter-
mining the number and location of wells; selecting well casing and screen materials;
and backfilling, sealing, developing and completing monitoring wells. They also discuss
sampling protocol.
      Grisak et al. (9) identify the technical  difficulties involved in monitoring ground-
water quality. Past ground-water monitoringf efforts are critiqued by Todd et al. (10).
A  manual prepared  for the  USEPA  by Perm et al. (11) presents  a comprehensive
discussion of the  many' aspects of ground-water monitoring.  Written  to address
monitoring at solid waste disposal facilities in particular,  the manual also covers general
monitoring principles,  including network  design; monitoring and  well technology;

-------
ground-water contamination by leachates; sample withdrawal, preservation and storage;
and analytical methods.
      Fenn et al.  (11)  discuss the methodology  of various monitoring  techniques,
including wells screened or opened over a single vertical section of aquifer, piezometers,
weil  clusters, single wells with multiple sample points, sampling during drilling, and
pore water extraction from  core samples. They list the advantages and disadvantages
of each of these techniques and advise how to implement them. Estimates of the costs
involved for the alternatives are supplied.
        »
      Tinlin (12) sho%vs, by example, site-specific  procedures for monitoring various
classes of ground-water pollution  sources. He covers brine disposal, plating  waste
contamination, landfill leachate, oxidation ponds, and multiple-source nitrate pollution.

Monitoring Well Design and Construction
      Lewis (13) stresses the importance of custom-designing ground-water monitoring
systems for specific monitoring objectives which are compatible with local hydrogeologic
and chemical conditions.
      Dunlap et al, (14) describe technology for construction of wells capable of providing
representative, uncomaminated samples of ground water. Scalf et al. (1) and Sisk (15)
detail various techniques for  constructing monitoring wells,  highlighting advantages
and disadvantages of each method listed.  Well  construction practices are elaborated
upon in a manual developed  for EPA  by the National  Water Well Association (16),
which sets forth standards prepared by NWWA for engineers and government personnel
whose expertise is not in the field of well construction.
      Selection of an appropriate well casing and screen for monitoring well applications
is addressed by Fenn et al. (11). Pettyjohn et al. (17) discuss well  casings and sampling
equipment made from glass, Teflon®, stainless steel, polypropylene, polyethylene, other
plastics and metals, and rubber, and suggest which materials are best suited to sampling
for organic contaminants. A manual that reviews the proper selection, installation and
utilization of thermoplastic water well casing has been published by the National  Water
Well Association and the Plastic Pipe Institute (18).
      Seanor and Brannaka (3) discuss the potential effects of well construction, including
well construction materials and drilling,  on the results of organic water quality analyses.
The Manual of Ground-Water Quality Sampling Procedures by Scalf et al. (1) describes the
principles of operation and the' advantages and disadvantages of the more common
types of drilling techniques suitable for construction of ground-water monitoring wells,
as does Fenn et al, (11)..
      Considerations for the selection of a  drilling method are outlined by Minning
(19),  who addresses common drilling and sampling procedures  for hazardous  waste
facilities. He also compares various aspects of different drilling methods, including cost,
effectiveness in various formations, depth capability, cross-contamination potential, and
quality  of samples. Luhdorff and Scalmanini (20) present a  methodology for  rating
several  drilling methods based on  their ability  to achieve typical tasks in a  ground-
water quality investigation. They illustrate the application of the rating system to two
ground-water contamination problems in California.

-------
 Ground-Water Sampling
      Recommended  methods of acquiring water data are presented by the USGS in
 a handbook published by its Office of Water Data Coordination (21). Scalf et al. (1)
 cover ground-water sampling comprehensively in a manual published by the National
 Water Well Association. The manual presents considerations in the selection of sampling
 procedures, including objectives of the sampling program, (characteristics and nature
 of pollutants, and  hydrogeology of the  area.  It also  describes  procedures currently
 utilized to sample ground water and highlights advantages and disadvantages of each.
 Sisk (15) also discusses sample parameter selection and sampling-considerations. He
 further describes various types of monitoring wells and presents examples that suggest
 appropriate sampling procedures for each.
      Gibb et al.  (22) provide  recommendations on  sampling  protocol and sample
 preparation, preservation, and storage, noting how these sampling steps can affect the
 chemical composition of the water samples. Grant (23) also provides information on
 sample collection and  preservation. He describes grab or discrete sampling and composite
 sampling and  explains the differences  between them. Wood (24) offers guidelines for
 collection and field analysis of ground-water samples for selected  unstable constituents.
 Schmidt (25) addresses sample collection, too, and speaks to the  question of optimum
 pumping rates and duration prior  to sampling, Scalf et al. (1) go  farther  and discuss
 sample records and chain-of-custody procedures. Others that detail the basics of sample
 withdrawal, preservation and storage include  Carriere and Canter (26) and Fenn et al.
 (11).
      General guidelines for  volume requirements, container preference, preservative
 measures,  and holding times when  sampling for specific parameters are provided in a
 useful table by Fenn  et al. (11).
      Suggestions as to types of sample containers and preservation techniques  to use
 when  analyzing for  various parameters can  be found  in many  of the  monitoring
 publications mentioned throughout this section. Regulatory agencies also may stipulate
 specific recommended practices. Several sources that deal with the effects of using
 particular  containers  in the presence of certain contaminants include King et al. (27),
 Struempler (28), Laterall  et al. (29), and Coyne and  Collins (30).
      Sampling of purgeable organic compounds performed as  part of the National
 Organic Monitoring Survey conducted by the USEPA is described by Brass et al. (31).
 A study by Dunlap et al.  (14) that presents methods for acquiring grab  samples of
 ground water suitable for total organic analysis, as well as other sampling information,
 aims to  provide a basic  capability for  sampling for organic pollutants  in shallow
 subsurface environments. Goerlitz and  Brown (32) recommend preservation techniques
 for organic substances in water.
      The potential for contamination posed  by equipment and  instrumentation used
.in ground-water monitoring is described by  Seanor and Brannaka (3). They discuss
 the  use of bailers and various pumps and describe how such sampling equipment can
 alter the chemical properties of ground water.
      Todd et al. (10) highlight the  need to check out the effects  of sampling materials
 on analytical results for the contaminants of interest. This type of research was performed
 by Gibb et al. (22), who  studied four types  of pumps to determine their effects on
 water samples.

-------
     Selection criteria for pumps are offered by Gass et al (33). A number of authors
detail various pumping and water collection devices, including McMillton and Keeley
(34), Buss and Bandt (35), and Tomson et al, (36).
     Everett et al. (37) provide detailed cost data for various monitoring methods and
techniques, They also supply a breakdown of capital and operational costs and suggest
sources of information for  updating the cost estimates provided. Fenn et al. (11) also
offer cost estimates for various monitoring techniques and well construction methods.
To aid. in the  selection of monitoring  methods, Tinlin (12) presents  hypothetical
illustrative examples for making decisions based,  in part, on cost comparisons among
monitoring alternatives.                                             i

Materials Used in Monitoring Efforts
     Scalf et al. (1) offer guidelines in the selection of well construction materials when
sampling for specific contaminants  in  Appendix  A of their manual.  An EPA report
prepared by Geraghty and Miller, Inc. (8) names materials best  suited to monitoring
efforts and notes exceptions to their general recommendations, Pettyjohn et al.  (17)
suggest materials of choice  when  sampling for organic contaminants.
     A number  of researchers and authors detail the limitations inherent in using
certain well construction materials. Scalf et  al. (1)  cite  the high cost of Teflon®-
constructed  wells, and a USGS study by Imbrigiotta and Martin (38)  draws attention
to PVC's  drawbacks in terms  of  its  lack of physical strength  and its  susceptibility to
damage. The same study points out that  metallic casing materials may be subject to
chemical degradation or dissolution.
     Some  previous studies provide in-depth analyses of these and  other problems
relating to  compatibility of well construction and  sampling materials  with various
contaminants. Miller (39) reports results of a laboratory study that aimed  to identify
chemicals that  may ieach from well casing material when  it is  exposed to selected
pollutants and  to quantify  adsorption of selected pollutants on  various  well-casing
materials. The  leaching of organic  and organotin compounds from PVC and CPVC
pipe was documented in a USEPA report by Boettner et al.  (40).
     Gass et al.  (33)  describes the various metals  used  in  water well  applications,
particularly  in well pumps, and the  types of corrosion that may afflict them. He also
lists well  and pump  construction materials in  order' of  preference  based on  their
resistance to certain types of corrosion. Seanor and Brannaka (3) point out the potential
sample contamination  posed by steel casing, which is exposed to oils and  solvents in
its production.  Another report by Fenn et al. (11) highlights  the advantages of PVC
casing over metals, and the contamination of  ground water by the adhesives and
primers associated with synthetic well casings'has been addressed by Sosebee et al. (41).
      The Uni-Bell Plastic  Pipe  Association (42) issued a report that describes the
results of research into the control of residual vinyl  chloride monomer (RVCM) in
PVC water  pipe. Dressman and McFarren (43) and  Sachs and Banzer (44) addressed
this same topic, as  did Mantel! and  others (45). Berens and Daniels (46)  discussed the
prediction of vinyl chloride monomer migration from rigid PVC pipe. Earlier, Daniels
and Proctor (47) reported  on  the extraction of RVCM from PVC bottles.
                                        10

-------
      Curran and  Tomson (48) describe the leaching of low levels of contaminants
from tubing made from various materials. Junk et al, (49) detail organic contamination
contributed to water samples by certain tubing materials. Study results on the adsorption
of specific contaminants on walls of containers made from certain materials are reported
by Massee et al, (50), Shendrikar et al. (51), Robertson (52,53), and  Eicholz et al. (54).
Current Research Efforts
      Monitoring methodology research presently is being conducted through three
principal EPA programs as well as through a number of U.S. Geological Survey projects,
private consultants, and foreign organizations. The EPA programs addressing sampling
methodology include the Ground Water  Research  Branch  of the Robert S. Kerr
Environmental  Research  Laboratory in Ada, Oklahoma; the Ground Water Branch of
the Environmental Monitoring and Support Laboratory in Las Vegas, Nevada; and the
Solid and Hazardous Waste  Research Section  of the  Municipal and Environmental
Research Laboratory in Cincinnati, Ohio.
      Research at RSKERL-Ada is directed principally at defining the processes which
control the movement of contaminants through the subsurface. Currently, two activities
sponsored by RSKERL  specifically  address sampling ground  water for  monitoring
purposes.  Through a cooperative agreement with the  National Center  for Ground
Water Research, the University of Oklahoma is investigating the possible sorption of
specific  trace organic pollutants and metals from ground water onto  the  surface of
monitoring well casing materials and the possible'leaching or release of such substances
from well casing materials into sampled water. Another cooperative agreement with
the  Illinois  State Water  Survey, partially  supported  by EMSL-Las Vegas, seeks to
evaluate and develop methods  for constructing, completing, and sampling ground-
water monitoring wells  to obtain representative physical, chemical, and biological
analyses. This manual is  a product of that research effort.
      Several projects in the RSKERL program have sampling components indirectly
related to ground-water  monitoring. These are concerned primarily with collecting
ground  water and uncontaminated subsurface material for various  research projects.
Cores of  subsurface materials are being  used  to measure  physical, chemical, and
microbial characteristics and to  relate these  to the ability of the subsurface to attenuate
pollutant movement.
      Stanford  University has  an extensive research  project designed  to evaluate
ground-water contamination risks resulting from hazardous waste disposal.  A portion
of this project involves extensive sampling of ground water at field installations by the
University of Waterloo.
      EMSL-Las Vegas is active in several  research areas directly related to ground-
water monitoring. In addition to co-sponsoring the Illinois effort with RSKERL-Ada,
EMSL-Las Vegas supports research into both drilling and non-drilling techniques for
ground-water monitoring as well  as into  methods for  vadose zone monitoring at
hazardous waste sites. Methods  under investigation that require drilling include down-
hole sensors for smal! diameter boreholes,  indicator parameters for hazardous constit-
uents, x-ray fluorescence and infrared spectrometry screening for hazardous wastes,
and  compound-specific  ground-water  monitoring using fiber optics technology  in
combination with laser fluorescence spectroscopy.

                                       11

-------
      Considerable effort at EMSL-Las Vegas is directed toward monitoring ground-
water quality using non-drilling or remote-sensing techniques, especially at  hazardous
waste sites. One project is investigating the feasibility of sampling by cone penetrometer
in lieu of conventional drilled wells. Other projects involve using geophysical methods
such as complex resistivity  for tracking Seachates from hazardous waste sites.
      Research at MERL-Cincinnati does  not specifically address ground-water mon-
itoring. The emphasis there is on providing technical support to RCRA and Superfund
activities of the Agency. Most research is related to remedial action activities at waste
disposal sites and/or the  evaluation of the effectiveness of these activities. Ground-,,
water monitoring  is necessarily a  part of  such  evaluations, but development of new
monitoring techniques is generally not a part of MERL's research efforts.
                                        12

-------
                                   SECTION 5
                MONITORING WELLS AND SAMPLING APPARATUS
      The success of a ground-water monitoring program depends, in part, on the
selection of the proper type of well.
      Production or traditional  wells often are used to obtain samples in  ambient
ground-water monitoring programs. Designed to yield large quantities of turbidity-free
water for potable or irrigation supplies, these wells generally tap the more permeable
portions of an aquifer. Chemical  data obtained from these wells depict  the quality of
water being delivered to the user community. Because water pumped from these wells
is  a  composite  of water from  different horizontal and vertical strata in the aquifer
systems, the presence of relatively  narrow or small plumes of polluted water can be
masked by dilution with water obtained from unaffected portions of the aquifer.
      Production or traditional wells should not be used for the more detailed source,
case-preparation, and research types of monitoring. Such detailed monitoring efforts
call for wells designed to determine the ground-water quality at a given location and
depth  within1 the geologic materials  being monitored.  All  available  geologic  and
hydrologic information for the site of interest should be reviewed prior  to selecting
preliminary locations and depths for monitoring wells. The potential paths of pollutant
movement from the  site should be  estimated and wells should be placed so that they
can  effectively  and quickly detect  releases.  Information gained during  the  drilling
process should be used to modify the monitoring plan to make it more effective.  -


Monitoring Well Components
      The principal reason for constructing monitoring wells is to collect ground-water
samples which,  upon analysis,  enable describing a contaminant plume and  tracking
movement of specific chemical or  biological constituents.  Obviously, the location of
monitoring wells spatially and vertically is important. Of equal importance is the design
and construction of monitoring wells to provide easily obtainable samples that will yield
reliable, meaningful information.  In general, monitoring well design and construction
follow production well design and construction techniques. However, emphasis is placed
on the  effect these practices may have on the  chemistry  of the water samples being
collected rather than on maximizing well efficiency.
      From this  emphasis, it  follows  that  an understanding of the chemistry of the
suspected pollutants and  the  geologic setting in  which  the  monitoring .wells are
constructed plays a major role in the drilling technique and materials used. There are
several  components to be  considered in the design of a monitoring well, including
location, diameter, depth,  casing, screen, sealing material, and well development. As
these components are  discussed in  detail,  it may 'be  helpful to refer to  Figure  5-1,
which portrays  two typical well installations: one for water supply and  the other for
ground-water quality monitoring.
                                       13

-------
                     Pump Discharge
                      Removable Cap
                      on Well Protector

J/S^//&S/s&//^
*
Schedule 80 Pipe _L
(4" dia. or greater) /
/
Backfill and/or '
Clay Slurry "*"/•
/
Sand or — *-;•
Gravel Pack ;
Well Screen — :
(length: entire
thickness of
water-bearing
material)

fill,
iJ
1^7 '
- L.
' /•
' <•
'. „,-,-,


^•n:;rri
- : 	



&//£z$?/£ys/fz>Js/ ^^ " ^^^//xS^v/i'oVi'V^v^ .
- Concrete or CemenK:;. ^ * f < . or^ntl
;_..'•'•••'• ' -' v i .''.- ;
" - e • '• '.' *• ' o .' '/ Schedule 80 —
s .' x ' TILL" ^ ' ' • , • pVC Pipe ^
Static '-'; .-.-•.-• :••-•; <2"dia.) ,
-Water Level '/. 0 '.'*"' o •','-.'.• '
•'-.'.... 'o ';-••.•.«• s' ^
0 •.''••« ' o /
'.».\,p/ l^../
vV :-.:-.- :-;::-;.'-: •••-••;•:•:: .-:-:. {
;";•"..• ^:---^v.-.V;\:-/-; ^^
... .0 •...•..,..-.•-....••_•;• 1
-..^SAND AND GRAVE L.-.-- — I— t
.*••'! ' (water-bearing) .". • . ' •' p\
'•'••' •'••' '•'•' •'.•••'• '•'-.''^^r'-''-
'•:'•'•'•.• ••'. e. •-•.'•:- v -.''-."• ':>-v-'.
.'''". • .' :. ' • • ' '"•,•'.' •'. ° .''•.' '' ' • o
'•';:. :-':-r^?>>Tr-. ::'..oV.

•
i
/
/
/ '\>
* t^
^
/
/
/
/
/
: —
•|

'///^VA-///-- .
Static
— Water Level
Cement or
•*- Bentonite
Seal
Sand or
"*" Gravel Pack
X
/C Well Screen
(2" dia.)

      a. Water Supply Wei
	SHALE BEDROCK	b.  Monitoring Well Piezometer
         Figure 5-1.  Schematic diagrams of typical water supply (al and monitoring tbl well installations
Well Location
     The location of a monitoring well should be selected on the basis of the purpose
of the program. This purpose may be to verify predictions of contaminant migration;
to detect contaminants in  drinking water supplies and thus protect public health;  to
activate a contingency plan such as a program for leachate collection; to protect the
operator; or  to reassure the public by demonstrating that their water quality is being
monitored.  Each  of these purposes  will require a somewhat specialized array  of
monitoring points and a somewhat different sampling program. The monitoring system
must be designed to suit the purpose(s) in mind.
     The  positioning  of  a  monitoring point in a contaminant  flow  path for  an
"interpretive" monitoring effort must be determined on the basis of reliable preliminary
data. To do this, the contaminant flow path must be clearly defined in three dimensions
during  the  "detective" phase.  Only then can  the  optimum location for proposed
monitoring wells be effectively determined.

Well Diameter
     A domestic water supply well is commonly 4 inches in diameter to accommodate
a submersible pump capable of delivering  5-10  gallons per  minute.  Municipal and
                                        14

-------
industrial supply wells have greater diameters to handle larger pumps and to increase
the available screened open area so the well can produce water more efficiently. Like
that of a water supply well, the diameter of a  monitoring well is determined by the"
size  of the  sampling  device  or pump. This strategy'works well in very permeable
formations,  but unlike  most  water supply wells, monitoring  wells are quite often
completed in very marginal water-producing zones. Pumping one or more well volumes'
of water  from a monitoring  well built in such zones may present  a serious problem if
the well  has a large diameter.  Large quantities of water must be disposed of, and it
may take several hours  or  even days  for enough water  to return to the  well for
successful collection of a representative sample. With the advent of several commercially
available small-diameter pumps (le'ss than 2 inches outside diameter) capable of lifting
water from  several  hundred  feet, it Is  rarely necessary-to construct monitoring wells
larger  than  2 inches in diameter. Additionally, the smaller the diameter, the less it will
cost for drilling and construction. A thorough discussion of advantages and disadvantages
of monitoring well diameter has recently been published (55).

Well Depth          "
     The depth of each monitoring well normally is determined by the geohydrologic
conditions at the site being monitored. Most "detective" monitoring wells are completed
in ' the first  relatively permeable  water-bearing zone  encountered,  since  potential
pollution sources requiring the installation of monitoring wells are frequently at ground
surface. Locating the  monitoring well in the  first relatively permeable zone therefore
yields an early warning of the migration of pollutants in most situations. However, care
must be taken to assure the .well is completed at a depth  sufficient  to allow for seasonal
water table fluctuations. Under confined or semi-confined (leaky) conditions the water
level will rise above  the top of the water-bearing zone. In this instance, the well should
be finished in the water-bearing zone and not above it.
             f                   "-
     If the water-bearing zone is  thick (greater  than  10 feet)  or contamination  is
known or suspected in deeper formations, multiple wells completed at different depths
should be used. For sampling at various depths, some engineers  have nested several
wells in a single borehole. This requires drilling a large diameter hole and exercising
special care  to ensure that the vertical  integrity of the sampling points is maintained.
It is important that monitoring wells be constructed so that they are depth-discrete,
that is, so that they sample from one specific formation or zone without interconnection
to others. To assure that this requirement is met, provisions for placing cement grout
above  and,  if necessary, below the  intake portion of the well,  must be made in the
design of the well.  ;                             -

Well Design and Construction Materials
     The type of material used for monitoring well casing can have a distinct effect
pn  the quality of the water  samples  collected. Galvanized casing will impart iron,
manganese,  zinc and  cadmium to  many 'waters. Steel  casing 'may impart  iron and
manganese to the water samples. PVC pipe  has been shown to release and adsorb trace
amounts of various organic constituents  to water after prolonged exposure.;PVC solvent
cements used' to attach sections of PVC  pipe also' have been shown  to release'significant
quantities of organic compounds. Teflon® and glass.are among the  most inert materials


                         ;   .           is'

-------
 that have been considered for monitoring well construction. Glass, however, is very
 difficult.and expensive to use under most field conditions. Stainless steel also has been
 found to  work satisfactorily. A detailed discussion; of materials is presented in later
 portions of the text.  ',                     •                     I
       In many situations,, it  may be possible to compromise accuracy or precision for
 initial cost. For example,  if the contaminants of-interest are already defined and they
 do not include substances  which  might bleed from,  sorb  or interfere  with  other
 analytical  methods, it might be reasonable to use wells cased with a  less expensive
 material. Wells constructed of less-than-optimum materials might be used for sampling
 if identically constructed wells are available in both uncontaminated and contaminated
 parts of the aquifer to provide ground-water samples for use as  "blanks." Obviously,
 such  "blanks" may not address adequately problems of adsorption on or leaching from
 the casing material induced by contaminants in the ground water. Careful consideration
 is required in each individual case,  and the analytical laboratory should be fully.aware
 of construction materials  used. .
       Care must be given to the preparation of the  casing and welt screen  materials
 prior to installation. As a minimum, both "should be washed with detergent and rinsed
 thoroughly with clean water. Care also: should be .taken  to ensure that these and other
 sampling  materials are protected from contamination by using some type of ground
 cover such as plastic sheeting for temporary storage in  the work area.
       AH wells should allow free entry of water. They also should produce clear, silt-
 free water; For drinking water supplies, the  reason •• is  obvious: sediment in the raw
 water can create  additional pumping  and  treatment costs and  lead  to the general
 ,unsuitability of the finished water.  With monitoring wells, sediment-laden water can
 greatly lengthen  filtering times and create  chemical interferences with the collected
• samples.    *                                           :
       Commercially manufactured well screens generally work best provided the proper
 slot,size is chosen. In formations where fine sand,  silt, and clay predominate, sawed qr
 torch cut slots  will not retain the material,  and the well may clog. It  may be helpful
 to have well screens of several slot-sizes on site so that the correct manufactured screen
 can be placed in the hole after the water-bearing materials have been inspected. Gravel-
 packing materials  compatible with the selected screen size will further help retain fine
 materials and also allow freer entry of water into the  well by creating a  zone  of higher
 permeability around the well. '
       Well screen  length  is an ^important consideration. When monitoring a potable-
 water-supply aquifer, the entire thickness of the  water-bearing formation should be
 screened. This provides a!n integrated  water sample similar to what would be found in
 the  drinking water supply. A monitoring  program  to describe  contaminant plume
 geometry requires sampling discrete intervals of the  water-bearing formation. In this
 situation,  screen lengths of no more than 5 feet (1.5 m)  should be used.  Thick aquifers
 would require completion of several wells at different depth intervals. In some situations,
 only  the  first  water-bearing  zone encountered  will require  monitoring. Here the
 "aquifer" may be only 6 inches to a few feet (0.2-2 m) thick, and the screen lengtfi
 should  be  limited to  I or 2 feet (
-------
such a case, the screen must be long enough to extend above the  water level in the
formations so  these lighter substances can enter the well.
      It is  critical  that the screened portion of each monitoring well access  ground
water from a specific depth interval. Vertical movement of water in the vicinity of the
intake and around the casing must be prevented to obtain samples representative of
that in the formation of interest. Specifically, rainwater can infiltrate backfiil materials
and dilute or  contaminate samples collected from  the screened portion of the well.
Vertical seepage of leachate or contaminated water from adjacent formations along the
well casing also may produce unrepresentative samples for the depth interval being
sampled. More serious indeed is the creation of a conduit in the annulus of a monitoring
well which could contribute to or  hasten the spread of contamination.
      Monitoring wells usually are sealed with neat cement grout, dried bentonite, or
bemonite slurry. While a neat cement grout often is recommended,  shrinkage and
cracking of the cement upon curing can create an improper seal. The .use of bentonite
traditionally has been  considered  more effective than neat cement grout. However,
recent studies have shown that some organic compounds migrate through bentonite
layers with little or no attenuation  (56). Therefore, expanding cement appears  to-offer
the greatest potential  for, the effective sealing of wells. There are .several commercial
formulations of this type of cement which generally  is more chemical resistant than
clays. The seal should extend from just above the well intake to  a level above the
highest known seasonal water level. Backfill cuttings then may be used to fill the annulus
in the unsaturated  zone provided that they  are free of contamination, and a mounded,
expanded cement collar can be placed around the well casing to divert surface drainage
from the "immediate area of the well casing. This collar should extend  well below the
frostline  and should be checked periodically for cracks or fissures,
      When sealing materials must be placed below the water table (or  where water
has risen in the borehole), it is recommended that they be  pumped  down the annulus
through  a tremie pipe  and filled from the bottom upward.

Drilling Methods
      The selection of a drilling method best suited for a particular installation is based
on the following factors, listed in  order of importance:
       1) Hydrologic information
          a. types of formations
          b. depth of drilling
          c. depth of desired  screen setting below water table
       2)  Types of pollutants  expected
       3)  Location of drilling site, e.g., dry lands or inside a lagoon
       4)  Design of monitoring well desired  ,
       5)  Availability of drilling equipment

       The principles of operation and the advantages and disadvantages of the more
 common types of drilling techniques suitable for constructing ground-water monitoring
 wells are discussed by Scalf (1). The following is a brief summary of material presented
 in that publication.
                                        17

-------
     Auger drilling frequently is accomplished by rotating a solid-stem, continuous-
flight auger into the soil.  As the augers are "screwed" into the soil, the cuttings are
brought .to the surface on the rotating flights. Auger bits are essentially of two types;
fish tail or drag bits for use in unconsolidated materials, or claw or finger bits for use
in consolidated or cemented soils.  Once the desired depth is reached,  the augers are
allowed to rotate to clean out the borehole.  The augers then are removed from the
borehole and  the well screen and  casing installed.'This method is best applied when
installing monitoring wells in shallow unconsolidated formations that will maintain an
open borehole long enough to permit emplacement of the well, piezometer, or sampling
device.
     Hollow-stem,  continuous-flight auger drilling differs from  the solid-stem augers
in that the stem is  hollow. Upon reaching  the desired depth, a  small-diameter casing
and  screen can be set inside the hollow stem. The augers then  are pulled  out as the
casing  is  held  in place. If the borehole stands open after withdrawing the augers, an
artificial gravel pack is placed opposite the screened portion of the well. If the material
collapses around the screened portion of the well,  those materials  are  ultimately
removed or altered in grain  size by  the development procedures employed after well
construction.
     Auger drilling rigs  generally  are mobile, fast, and inexpensive to operate in
unconsolidated formations. Because no drilling fluid is required, contamination problems
are minimized. However, these types  of rigs cannot be used in hard-rock drilling. Depth
limitations vary with the equipment  and types of soils, but usually are  a maximum of
150 feet.  Collection of formation  samples  is not expedited  by the use of solid-stem,
continuous-flight augers. However, conventional soil sampling techniques (split spoon
and Shelby tube sampling) can be accomplished effectively during the drilling procedure
through the hollow-stem, continuous-flight augers. Formation sampling has been covered
in detail  by Scalf et al. (1).
     Straight  rotary  drilling is accomplished by pumping a drilling mud down the
inside  of a rotating drill  pipe and allowing  it to  return to the surface through the
annulus. The drilling mud cools the drill bit, carries the cuttings to the surface, prevents
excess  fluid loss into the  formation and prevents the formation from caving. The
rotating drill pipe turns the bit, which cuts the formation, allowing the cuttings  to be
flushed out. The drilling fluid (or mud) may be clear water, water  mixed  with bentonite,
or water mixed with various synthetic or natural  drilling aids.  Rotary rigs generally
are available throughout the United  States. They are capable of  drilling in all types of
formations to  almost any depth desired for monitoring and are  fairly reliable in most •
formations. Casing  is  not required during the drilling and logging of formations.
     The  use  of drilling  fluids  or muds during construction can bias the  results of
samples-collected from wells. The introduction of clear water has a tendency to  allow
water to  migrate into permeable formations which  may be of interest. This water must
be effectively removed before the quality of "native ground water" can  be determined.
Drilling rnuds also present a problem during the well development and have the ability
to affect  the transport of certain pollutants as water moves toward the well. Organic-
based muds add significant quantities of foreign organic material into the aquifer system
(57,58,59). Studies concerning the  effects of these types of muds on the  analytical
results of  collected samples and  the biodegradability of the muds have  not  been
conclusive.

                                        18

-------
      Air rotary rigs operate in the same manner as the mud rotary rig except that air
is circulated down the drill pipe and returns with the cuttings up the annulus.  Some
rotary rigs are equipped to operate with either mud or air. Air rotary rigs are available
throughout much of the United States and are well suited for many drilling applkatons.
      Air rotary rigs operate best  in'hard rock formations.  Formation water is blown
out of the hole along with the  cuttings, so it is possible to determine when the first
water-bearing zone  is encountered. After  filtering of water blown from the  hole,
collection and field analyses may provide preliminary information regarding changes
in water quality for some parameters. Formation sampling ranges from  excellent  in
hard, dry formations to  nonexistent when circulation is lost in cavernous limestones
and other formations with cavities.
      Casing  is  required  to keep  the borehole  open  when  drilling in  soft, caving
formations below the water table. When more than one water-bearing zone is encoun-
tered and where the hydrostatic pressures are different, flow between zones will  occur
between the time when the drilling is done and the time when the hole can be properly
cased and one zone  grouted off. Commonly, no synthetic drilling aids are used  in air
rotary drilling. If the  air is filtered to capture compressor lubricants, contamination
can be  minimized relative to other methods. The use  of air rotary drilling in  badly
contaminated  subsurface  situations must be approached  carefully  to  minimize the
exposure of drilling  personnel to potentially hazardous  materials.
      A cable tool rig uses a heavy, solid-steel chisel-type drill bit suspended on a steel
cable which, when raised and dropped, chisels or pounds a hole through the  soils and
rock. When drilling  through the unsaturated zone, some water  must be added to the
hole.  Cuttings are suspended in  the water and periodically bailed. After enough  water
enters the borehole to  replace the water removed by bailing, no additional water  needs
to be added. When  soft,  caving formations are encountered, it is necessary to  drive
casing as the hole is advanced to prevent collapse of the hole. Often the drilling can
be advanced only a  few feet below the bottom of the casing. Because the drill  bit is
lowered through the casing, the hole created  by the bit is smaller than  the casing.
Therefore, the casing (with a sharp, hardened  casing shoe on  the bottom)  must.be
driven into the hole. The  shoe, in fact, cuts a slightly larger hole than does the drill
bit. This tight-fitting drive shoe cannot be  relied upon to form  a seal when overlying
water-bearing  zones  are encountered.
      Formation samples can be excellent when a skilled driller uses a sand-pump bailer.
Information regarding water-bearing zones is readily available during drilling. Relative
permeabilities and some water-quality data also can be obtained from different  zones
penetrated if a skilled operator is available. Cable tool rigs can operate satisfactorily in
all formations, but  they  are best  suited  for  large, caving, gravel-type formations or
formations with large cavities above the water table.
      Cable tool drilling is slow compared with rotary drilling. The necessity of driving
the casing along with drilling in  unconsolidated formations requires that the casing be
pulled back to expose  selected water-bearing zones. This process complicates the well
completion process and often increases cost. Relatively large diameter casing is  required
(minimum  4-inch casing) which increases the costs compared with rotary drilled wells
with plastic casing. The use of cable tool rigs for small-diameter (2-inch) wells is not
recommended.
                                        19

-------
     Reverse circulation drilling is accomplished by allowing water or mud to circulate
down the annulus and up the inside of the drill pipe (reverse flow direction from direct
rotary  mud drilling). This type of drilling  is used for construction of high-capacity
production wells and is not suited for small,  water-quality  sampling wells.  Custom
drilling techniques may  be necessary for specialized investigations (60).

Weil Development
     Development is a facet of monitoring-well construction that is often overlooked.
During the drilling process, fines are forced into the open borehole, forming a mud
cake that  reduces the hydraulic conductivity of the materials opposite the screened
portion of the well. To allow free entry of water into the monitoring well  and to
maximize  well yields (a particularly important factor for low-yield geologic materials),
this mud cake must be broken down and the fines removed from the well.
     Additionally, monitoring  wells  must  be  developed to provide  water free of
suspended  solids for sampling. When sampling for metal ions  and other inorganic
constituents, water samples must be filtered  and preserved at the well site at the time
of sample  collection. Improperly  developed monitoring wells will produce  samples
containing suspended sediments that will both bias the chemical analysis of the collected
samples and frequently cause clogging of the field filtering mechanisms (61).
     The development  procedures used for  monitoring wells are similar to those used
for production wells. The basic principles include creating alternately high and low
velocities  of water  flow in  the well to break  down  the mud  pack or loosen fines,
followed by pumping to remove the fines from the well and the immediate area outside
the well screen.
     Successful  development methods for relatively productive  wells include the use
of a surge block, bailing, and surging by pumping. A surge block is a  plunger device
that fits loosely inside the well casing. It is moved forcibly up and down, causing water
to surge in and out of the  well screen.  After  surging, the well must be pumped to
remove the fines carried into the well  screen and casing. The use of surge blocks for
monitoring  well development  has not  been  widely  attempted.  The potential for
damaging a relatively fragile, 2-inch-diameter PVC well with a tight-fitting surge block
has generally overruled the  potential  benefits that would be gained from  this type of
development.
     A bailer sufficiently  heavy to fall quickly through the water can be raised and
lowered rapidly through the screened  portion of the well. This  action will create the
same alternating surging action as the surge block. The use of bailers for developing
wells is more common than  the use of the surge block.
     Another method for developing  wells  in relatively productive geologic materials
is to surge either with a pump or by air.  When using a pump, the well is alternately
pumped and left idle to simulate  the  surging action desired to loosen the fines and
remove them from the  well. However, in most  applications, no outward movement of
water from the well is experienced, and bridging of fines moving toward the well limits
the effectiveness of this technique.
     When pumping with  air, the effectiveness of the procedure depends  on the
geometry  of the device injecting air into the  veil. Figure 5-2 illustrates a simple device
                                        20

-------
               Flattened nozzle with
                  1/8" opening
                                                 3/8" OD stainless or
                                                   ,copper pipe
                                                  Overall dimension
                                                   less than 1-'/z"
                       1/8" diameter hole (both sides)
                 Figure 5-2. Schematic diagram of an air driven well development device
that diverts air through the well screen to loosen the fines and forces air, water and
fines up the  well casing and out of the well. These devices are particularly effective
for developing monitoring wells in relatively productive geologic materials. Air devel-
opment  techniques also may cause the exposure  of drilling personnel  to hazardous
materials when badly contaminated ground water is present. Careful precautions must
be taken to minimize personnel exposure.
     Development procedures for monitoring wells in relatively unproductive geologic
materials is somewhat limited.  Due to the low  hydraulic conductivity of the materials,
surging of water in and out of the well  casing is  extremely difficult.  Also, when the
well is  pumped, the entry rate of water is inadequate to effectively remove fines from
the well bore and the gravel pack material outside the well screen.
     In this type of geologic setting, clean water can be circulated down the  well
casing,  out through the screen and 'gravel pack, and up  the open borehole prior to
placement  of the  sealant in  the  annulus.  Relatively high water  velocities can be
maintained and the mud cake  from the borehole wall can be broken down effectively
and removed. (Because of the  low hydraulic conductivity of geologic materials outside
the well, a negligible amount of water will penetrate the formation being monitored.)
Immediately following this procedure, the well sealant should be installed and the well
pumped.
     In summary, all monitoring wells should  be developed. The  additional time and
money spent for this important procedure will expedite sample filtration and result in
samples  more representative  of water contained in  the formation  being monitored.
The time saved in filtration alone will more than offset the cost of development.
                                        21

-------
Sampling Apparatus
      Many brands of pumps are available for use in small-diameter (2-inch) wells.
Table 5-1 Jists selected commercially available sampling  devices and  their principal
features. Devices in Table 5-1  can be divided into  five general  categories based on
their methods of operation: bailers, suction lift pumps, gas contact samplers, positive
displacement samplers, and syringe samplers.
      Bailing  is one of the simplest  and oldest methods for sampling small-diameter
monitoring wells. Bailers, which can be constructed  from a wide  variety of materials,
require  no power sources and are easy to transport.  They are economical and easy to
clean. One of the drawbacks of using a bailer is that it is time-consuming and sometimes
impractical when dealing with large quantities of water. Special care also must be taken
to keep the rope,  wire, or chain clean during bailing to prevent the introduction of
contaminants. The transfer of water from the bailer to the sample container may allow
aeration of the sample and outgassing of volatile chemical constituents. Bottom-draw
bailer designs  with dual check valves minimize these sources of bias.
      Suction lift sampling devices listed in Table 5-1 include peristaltic and Venturi-
type pumps. The peristaltic pumps are relatively portable and can operate over a wide
range of pumping rates.  However, their use is limited to situations where water levels
are less than  about 20 feet. The suction created by the peristaltic pumping action
changes pressure in the transfer lines and may result in degassing and loss of volatile
organic compounds from the sample.
      The Venturi pump operates in the same way as the domestic-size two-pipe jet
pump. Water is pumped  at the surface by a centrifugal pump. A portion of the water
is circulated clown a return pipe and through a  Venturi section where suction is created,
and water from the well is pulled into the Venturi section and up the discharge pipe.
The pressure drop created at the Venturi section  may cause degassing or volatilization
of organic compounds. The circulation of water back into  the  Venturi section and
priming vessel also makes it difficult to determine when a  particular  well volume is
being delivered to the surface.
      Gas-drive contact pumps  normally use nitrogen gas to force water  from a
sampling chamber up through a  discharge line. The surface-area-to-water ratio is small
to minimize the effects of the gas contact. Data  have not yet been obtained.to accurately
determine the effect of gas/water contact,
      Positive displacement pumps listed in Table 5-1 include the helical-rotor, piston,
and bladder-type pumps. All three can operate over a wide  range of pumping rates.
Though the pumps are relatively portable and easy to operate, they do require power
(electricity or compressed gas)  and are somewhat  difficult  to clean.  Available  data
suggest that these  pumps are  more versatile and  desirable  than  other devices for
sampling most chemical constituents.
      Syringe-type samplers are used to collect samples for analysis of volatile organic
compounds. After purging the well with another  pump, sampling devices are lowered
into the well and positive pressure is maintained to prohibit the  entry of water until
the desired  depth is reached. Pressure  is then released, and the  hydrostatic pressure
in the well  fills the syringe.  The sample is raised to the surface and  preserved for
analysis.
                                       22

-------
Table 5-1, Saturated-Zone Sampling Devices for Use in Small-Diameter Boreholes and Monitoring Wells


Manufacturer
BarCod Systems, Ine,



Cole-Farmer Inst.
Co.


ECO Pump Core,


Galiek Corp.


GeoEngmeer ing.
, Inc.

Industi ial and
IsO Environmental
03 Analysts. Inc.
(IEA!
IEA


Instrument Special-
ties Co. IISCO)

Keck Geophysical
Instruments,
Inc.
Leonard Mold and
Die Works, Inc.


Oil Recovery
Systems, Inc.


Q,E.D, Environmental
Systems, Inc.


Model name/
number
BarCad Sampler



Master Fie*
7570 Portable
Sampling Pump

SAMPLtfier


Bailer 219-4


GEO-MONITOR


Aquarius



Syringe Sampler


Model 2600
Well Sampler

SP-81 Submer-
sible Sampling
Pump
GeoFiller
Small Dia. Well
Pump 1#Q5QO!

Surface Sampler



Well Wrzard®
Monitoring
System (P 100)

Principle of
operation
dedicated; gas
df tve (poshive
displacement)

portable;
peristaltic
{suction}

portable; venturi


portable, grab
(positive dis-
placement}
dedicated; gas
drive (positive
displacement)
portable; bladder
(positive dis-
placement)

portable; grab
(positive dis-
placement!
portable; bladder
(positive dis-
placement!
portable; helical
rotor (positive
displacement)
portable; bladdei
(positive dis-
placement)

portable; grab
(positiVB dis-
placement)

dedicated; bladder
(positive dis-
placement)
Maximum outside
diameter /length
linchcsl
1.5/16



<1,0/NA




-------
                                                                             Table 5-1. (Concluded}
Manufacturer
Randolph Austin Co.
Robert Bennett Co.
Slope Indicator Co.
ISINCOl
Solinst Canada Ltd.
TJMCO MJg. Co.,
Inc.
TlMCO
Tole Devices Co.
Model name/
number
Model 500
Vari-Flow Pump
Model 180
Model 514124
Pneumatic
Water Sampler
5W Water Sampler
Std. Bailer
Air or Gas
Lift Sampler
Sampling Pump
Principle of
operation
portable; peri-
staltic (suction)
portable; piston
(positive dis-
placement)
portable ; gas
drive {positive
displacement!
portable; grab
(positive dis-
placement!
portable; grab
(positive dis-
placement!
portable, gas
drive (positive
displacement)
portable; bladder
(positive dis-
placement!
Maximum Oiitsido Construction Lift
diameter /length materials range
(Inches) fw/linvs & tubing) ill)

-------
      In addition to the effects the pumping mechanism may have on the sample, ihe
materials that contact water in the pump and delivery tubes also should be examined.
Chemical constituents can be sorbed (adsorbed or absorbed)and leached from improperly
selected materials in pumps. Considerations for materials selection is discussed later in
Section  7 of this  report.
      Dedicated  m-silit sampling devices also are available for saturated-zone sampling.
These devices are placed directly into  the geologic materials to be sampled and are
pumped by gas-drive contact. The literature does not address the suitability of these
devices  for organic  sampling.
      In-line  detection  and sampling  devices are  valuable  tools for ground-water
monitoring. During the  collection of samples, flow-through cells for  monitoring tem-
perature, pH, redox potential,  and conductivity should be used. These devices should
be constructed to allow for a constant flow of,water without the accumulation of gases.
The accumulation of gases on  the probes can cause erroneous values to be recorded.
Probes should be noncoiitaminating  or  as inert as possible.
      As with well  construction and drilling method precautions, all  sampling devices
should be carefully  cleaned prior to  use, A dilute hydrochloric acid rinse followed by-
successive  rinses  with deionized  water, acetone, and distilled water are routinely
recommended. In badly contaminated situations  a hot water detergent wash prior to
the above rinsing procedure may be necessary. Hexane rinses prior to  the final distilled
water rinse aid the remo%'al of sparingly soluble organic materials prior to sampling
for low-level organic pollutants.
                                        25

-------
                                   SECTION 6
                             SAMPLING STRATEGY
      The importance of proper sampling of monitoring wells cannot be overem-
phasized. Even when the wells are correctly located, constructed and developed, special
precautions must be taken to ensure that the sample collected is  representative of the
ground water at that location.  Care also is needed to ensure that  the sample is neither
altered nor contaminated by the sampling and handling procedures,
      To select proper sampling procedures for monitoring wells, several basic questions
must be reviewed.  First, the purpose  of  the  monitoring program and collection of
samples must be  reevaluated. Monitoring commonly is undertaken to comply with state
or federal regulations that set forth a  list of chemical constituents to be  monitored.
They also usually  stipulate the concentrations of various chemical constituents that
must be analyzed  and reported.  Sampling procedures for defining plume geometry
and  migration may  require different procedures  than  those  for routine regulatory
monitoring.  Sampling procedures for research projects may be particularly specialized
and  more demanding. In all  instances, it is essential to develop a  complete list of
chemical parameters to be measured and the sensitivity at which they will be measured.
Secondly, the physical limitations of the well and the well site must be considered. The
diameter of the well, depth to water,  length  and location of the well screen, and
accessibility of the well site  all bear on the practical application  of various  sampling
procedures.
      Prior to discussing the effects of well  flushing and pumping mechanisms  on
ground-water quality, it would be instructive to review briefly the factors controlling
ground-water quality. An understanding of these  factors enables one to  better plan
sampling efforts and minimize the  effects of reactions that may  occur due to  sample
handling.


Effects of Subsurface Conditions on Ground-Water Quality and Sampling
      The subsurface is a condensed,  geologic, and geochemical environment which
also serves as a habitat for microorganisms (62,63). It is difficult to describe the entire
range of subsurface conditions since local effects can drastically alter regional properties
of temperature, pressure,  oxidation-reduction potential, mineral solubility, solution
chemistry, and biological activity. The relevance of average conditions must be evaluated
on a case-by-case basis in order to account for the changes in ground-water chemistry
which  may occur during sample  collection and retrieval. Temperature and pressure
are major physical influences in this regard.
      Temperature fluctuations in  the upper  10m (33  ft) of the subsurface occur in
response to seasonal air temperature variations. In the United States, average ground-
water temperatures range from 3°C to 25°C (37°F to 77°F) to  a depth of 20 m (66
ft) (64). Local effects can be quite marked, however, since increases from 7°C to 16°C
                                       26

-------
(42°F to 60°F) above background have been observed in contaminated aquifers near
landfills (65), Pressure gradients in the subsurface may be approximated as those due
to the hydrostatic pressure of a column of water equivalent to the depth  of interest.
The strength of the pressure  gradient may vary, though it is generally about 0,1 atm
•  m"' (0,4-0.5 psi ' ft"1). Summarizing average physical conditions in the subsurface, one
can expect that from the screened portion of a  shallow monitoring well 10-20  m (33-
66 ft) deep to the surface, temperature and pressure differentials may range from 10-
30°C  (45-83°F)  and  from  2-3 atm,  respectively.  Regardless  of the choice of well
construction methods  or materials which contact the sample, the impact of temporal
and  spatial physical changes must be minimized to provide the most representative
ground-water sample possible.
       Chemical and biological indicators  of subsurface conditions vary  greatly  in
response to small changes in the above-mentioned  physical  factors. Considerations of
sample handling procedures and the compatibility of materials with the ground-water
system are subtle. In general, the subsurface environment is not in equilibrium with
ambient surface conditions.  This is particularly  true of the atmospheric gases oxygen
and  carbon dioxide, which exert control  over the  oxidation-reduction potential, pH,
buffer capacity, and chemical speciation of dissolved substances. Several reference works
on aquatic and ground-water chemistry provide  background information on the range
of potential conditions to which samples,  well  casing,  and  sampling devices may  be
subjected  (66,67,68,69).
       Within  an  aquifer, there are  six major processes which affect subsurface geo-
chemistry.  These  are:  1) compiex  formation;  2)  acid-base reactions;  3)  oxidation-
reduction processes; 4) precipitation-dissolution reactions; 5) sorptive interactions; and
6) microbiological processes.
       All of these processes can be affected by sample collection procedures- Shifts in
solution chemistry caused by rapid changes in temperature, pressure, or gas content
may result in nonrepresentative  samples.  This  is particularly  true of the carbonate
buffer system since ground water is normally oversaturated with carbon dioxide (CCX).
The system is sensitive to temperature  and pressure changes since gaseous, dissolved, •
and solid  chemical constituents participate in equilibrium reactions. For example, the
loss of dissolved CO2 on simultaneous temperature increase and sample depressurization
will cause an increase in the pH. The pH changes in turn affect mineral solubility, the
kinetics of iron and  manganese oxidation,  and hydrolysis reactions. Products of these
reactions in turn  may shift the chemical speciation  of nutrients and metals and either
stimulate or inhibit microbial activity.
       Many potential  changes can take place in a ground-water sample while being
collected or before it is preserved (22,70), The  previous discussion described general
subsurface conditions and examples  of solution chemical changes in inorganic constit-
uents if samples are improperly collected. The impacts on volatile organic compound
concentrations also can be quite marked. Samples collected from sites exhibiting unusual
chemical properties may respond differently.


Effects of Hydrologic Conditions on Sampling Strategy
       Hydrologic factors can exert controlling effects on the collection of representative
samples. The yield potential  of a  monitoring well  will  determine the length of  time

                                       27

-------
that a well must be pumped at a given rate to produce a sample representative of the
aquifer at that location, The transmissivity of the materials tapped by the well also is
important in determining  reasonable sampling  frequencies and  predicting rates of
ground-water movement. To determine these parameters, well or  slug tests  should be
conducted on the monitoring wells to be  sampled.

      Traditional analysis of well test data usually involves the use of equations derived
by Theis (71)  and Jacob (72). One of the basic assumptions made in deriving those
equations was  that all of the water pumped from a well during a  pumping or aquifer
test comes from the aquifer and that none comes from storage within the well. Since
this condition  is seldom  fulfilled in practice, particularly for the low-yield  situations
common!}' encountered for monitoring wells, these equations are somewhat inappro-
priate for describing the behavior of water levels during pumping for most monitoring
wells.

      Popadopulos and Cooper (73) presented an equation describing the  discharge
from a pumped well  which takes  into account  the volume of water removed from
casing storage.

      Drawdown values  calculated  with  the relation developed by Popadopulos and
Cooper differ significantly from those based on the Theis and Jacob equations. During
the early stage of a pumping test a relatively high percentage of discharge comes from
casing storage  and smaller drawdowns are observed than would be predicted by Theis
or Jacob methods. In  the  later stages of the test a  negligible quantity of water is
obtained from  casing storage  and all three drawdown calculation methods produce
equivalent results.  If  the effects of casing storage are not  taken into account, it is
possible to misinterpret the data and assume an erroneous transmissivity value based
on early drawdown data,

      Well  test data  For six monitoring wells studied in  Illinois (22) were analyzed
using the equations presented by Popadopulos and Cooper. At all of the sites studied,
the nonpumping water levels were significantly above the top of  the aquifers tapped,
suggesting artesian conditions. In this case, a storage coefficient of 0.0001 was used in
the analysis  of data. The drawdown values described by the Popadopulos and.Cooper
equation are  relatively  insensitive to changes  in  storage  coefficient.  The storage
coefficient values selected therefore should have little effect on the aquifer  properties
determined  at most sites.

       Using the  Popadopulos and Cooper equations, the  percent of aquifer water
pumped was calculated for a 2-inch-diameter well at a pumping rate of 500 mL/min
for a range of transmissivities and  time (see Figure 6-1). Relationships of this type can
be developed  for  any given set of hydrologic parameters' to predict the time at which
a high percentage of aquifer water would be obtained.

       Conducting slug tests on monitoring wells also provides  estimates  of aquifer
transmissivity. Analytical methods described  by Freeze and Cherry (68) are recom-
mended for the interpretation of slug test data. Once the transmissivity of aquifers has
been determined, the relationships developed by Popadopulos and Cooper  can again
be applied  to predict  the quantity  of water coming from the aquifer during various
phases of pumpage.
                                       28

-------
              100 -
            cc
            LLJ
            r-
            CC
            LLJ
            u_
            5
            o
            LLJ
            O
            cc
            UJ
            o.
                                                Q = 500 mL/mirt
                                           DIAMETER = 5.08cm
                                                          25     30
                                    TIME, minutes
                Figure 6-1.  Percent of aquifer water versus time for different transmtssivities

      Based on the  studies by Gibb and others (22), the following recommendations
are presented  for collection of ground-water samples:
      1) A brief 2 or 3  hour pumping test should be conducted on each monitoring
well to be sampled. Analysis of the pump test data and other hydrologic information
should be used to determine the frequency at which samples will be collected and the
rate and period of time each well should be pumped prior to collecting the sample. If
pumping tests  cannot be conducted, slug tests may be substituted to provide the needed
hydrologic information.
      2) A general  rule  'of thumb of pumping four to six well volumes will, in most
cases, produce samples representative of aquifer  water. For aquifers with  unusually
high transmissivities,  pumping for periods long enough to remove the "stagnant" water
column  may induce  migration of water from parts of the  aquifer remote from  the
monitoring well. The calculations  of percent aquifer  water with time provide a more
rational basis on which the length of pumping can be determined. Samples should be
collected in the minimum time required to produce water representative of the aquifer.
      3) A controlled sampling experiment [monitoring indicator parameters (pH, T,
fl'1, and Eh) or collecting samples  during  an extended period of pumping] should be
conducted to accurately determine the chemical quality of aquifer water and to verify
the response of the monitoring well to pumping, as predicted from the pump test data.
This is best accomplished with an in-line closed measurement cell (74). When the values
of the indicator parameters are observed to vary less than ± 10% over three consecutive
storage  volumes, the well may be presumed  to  have been  adequately flushed  for
representative sampling.  Once  the chemical  character and response of the monitoring
                                        29

-------
systems have been determined, chemical constituents for routine sampling can  be
selected.
      4) Based on the sensitivity  of the selected chemical parameters, a choice  of
pumps for routine'sampling can be made.
      5) The monitoring wells should be pumped at a constant rate for a period of
time that will result in delivery of at least 95 percent aquifer water. The rate and time
of pumping should be determined  on  the basis of the transmissivity of the aquifer, the
well  diameter, and the  results of the sampling experiment.
      6} Measurements of pH, Eh, temperature, and specific conductance also should
be made at the time of sample collection. Field determinations of alkalinity together
with a mineral analysis, the foregoing measurements and total dissolved solids permit
mass and charge balance calculations  to be made which are valuable analytical quality
control checks,
      The steps outlined above are designed for collecting representative samples for
inorganic analysis.  The same procedures likely  will produce representative samples for
nonvolatile organic analysis. The  use of a syringe  sampler (75) in conjunction with
pumping or a bottom-draw bailer is desirable for collecting samples for volatile organic
constituents.
      Special care  must be taken to  prevent cross-contamination when carrying sampling
apparatus from one well to another. The sampling devices must be cleaned thoroughly
to ensure that  contaminants from one well are not carried to the other. Cleaning
procedures should be tailored to the analytes of interest. The use of detergents, dilute
hydrochloric  acid,  hexane, and deionized rinse water often is necessary. In addition,
cleaning of sampling devices, all delivery tubes, and tether cables also must be performed
thoroughly. The effects of cross-contamination also can be minimized by sampling the
least contaminated wells first and progressing to the more contaminated ones. Dedicated
sampling devices for each well also  may be desirable in certain cases where the potential
for cross contamination is high.
      In the case  of monitoring wells that will not yield water at a rate adequate to
be effectively flushed,  different procedures must be followed, There are divergent
points of view on how flushing should be performed in these situations. The principal
difference in the arguments concerns the degree to which such wells should be evacuated.
One suggested  procedure includes the removal of  water  to the top of the screened
interval to prevent the exposure  of the gravel pack or formation  to atmospheric
conditions  (5).  Then  the sample  is   taken at a  rate which  would  not cause rapid
drawdown. On  the other hand, the wTells may be pumped dry and allowed to  recover.
The samples should be collected as  soon as a volume of water sufficient for the intended
analytical scfiteme re-enters the  well.  Exposure of water entering the well  for periods
longer than 2 or 3 hours may render samples unsuitable and unrepresentative of water
contained within  the aquifer system. Finally,  in  these cases, it may  be desirable  to
collect small volumes of water over a  period of time, each time pumping the well dry
and  allowing it to recover.  At  present there is  very little reliable data on which to
choose one sampling method over  another in "tight" formations.
                                        30

-------
                                   SECTION 7
                           EVALUATION OF MATERIALS
      The selection  of materials used for well construction and sample collection,
handling, and storage is a critical consideration in planning the well-conceived moni-
toring program. The materials of .choice should retain their structural integrity for the
duration of the monitoring program under actual subsurface conditions. They should
neither adsorb nor  leach chemical constituents which  would bias the collection of
representative samples. The material combinations also must be compatible with each
other-and the goals of the program. For example,  in a detective monitoring situation
where the presumed inorganic  contaminant source (e.g., brine or pickling liquor) is
held in a surface  impoundment, material selection should be inade so as to ensure the
reliability of analytical determinations of chemical constituents of the waste in  ground-
water. These parameters may include pH, specific conductance (fi"1), alkalinity, hardness,
total dissolved solids (TDS), chloride, and  trace metals.  In  this hypothetical situation,
the "background" ground  water is relatively high in ionic solids, and the goals of a
long-term (e.g., 30-year) monitoring program would be  most closely met by a properly
constructed network of non-metallic or corrosion-resistant monitoring  wells. .Sampling
gear and sample  handling precautions also should  be chosen carefully to ensure that
the samples are neither contaminated nor  biased by the effects of materials.

Overview of Subsurface Conditions
      The structural requirements for well casings to  withstand normal subsurface
pressures are met by most common piping  materials: steel, polyvinyl chloride, and iron
for depths up to ~30 m. In deeper monitoring situations, the use of corrosion-resistant
metallic casing of large diameter (>10 cm or 4 inches) may be required to  provide
necessary  structural integrity.  Local water  well construction practices  should  serve as
a guide.
      Metallic  corrosion  problems may be encountered  under either oxidizing  or
reducing conditions and are aggravated by high dissolved-solids content.  Other materials
(thermoplastics) may deteriorate under the influence of dissolved chemical substances
or direct  contact with wastes. Whether the well construction  or sampling materials
retain  their integrity  or not, there are   also  potential problems due to  microbial
attachment and growth  and the sorptive  capacity of the exposed materials for the
chemical species of interest. Representative sampling depends on the proper choice of
materials which can retain their integrity over the entire length of a well casing, from
the aerobic, unsaturated surface zone to the unique conditions in the  saturated zone.


Chemical Properties of Water and Their Effects on Various  Materials
     To achieve the goals of a  detective monitoring  program in a cost-effective  manner.
one must  carefully design and construct the sampling network after consideration of
                                        31

-------
the compatibility of casing and materials with the subsurface environment. If an initial
detective network is to be used during the interpretive phase, then the materials should
be compatible with  probable mixtures of ground water and chemical substances from
the source. Compatibility must be judged from  a structural and chemical standpoint.
Structural considerations are treated in detail in a National Water Well Association
manual (18). The main criterion for chemical compatibility should be that the long-
term interaction of the casing or sampling materials  with  the ground water will not
cause analytical bias in the interpretation of chemical  analysis of the water samples.
     For example,  assume that a long-term detective monitoring network  is to be
designed for an acidic metal-plating waste impoundment in a shallow  ground-water
system  where the background water is  low in pH and  high in  dissolved solids. The
contaminants of interest are Fe, Mn, Zn, Cr, Cd, and Pb.  Under these conditions,  it
would  be imprudent to construct the monitoring wells with black iron  or galvanized
steel, since inevitable corrosion of the casings would be expected to contribute several
of the  above metals to the water samples. One  may argue that  flushing the  well will
minimize this source of bias or that upgradient and downgradient wells will experience
the same degree of interference from corrosion.  This  argument may  hold  in some
cases; however, the  degree of interference  in specific wells will depend on the actual
conditions at each well site. High background levels of chemical substances contributed
by deteriorating well  casing might interfere with the analytical  determination  of the
parameters of interest, Very difficult interpretation problems may result if a downgra-
dient galvanized  well were at the margin  of a plume  of acidic waste, and  casing
corrosion contributed a high "background" of Fe, Mn,  Zn, and Cd to water samples
drawn  from both upgradient and downgradient  wells.         f
     In a long-term (e.g., 30-year) detective monitoring program, the judicious selection
of casing materials is of utmost importance. The corrosive  and leaching properties of
ground water over an extended period at a particular site are nearly impossible to
predict "a priori." Testing the materials of choice with admixtures of ground water
and  wastes from a  source may be a valuable step prior to the construction of the
monitoring well network. If organic  contaminants are of concern, it may be necessary
to investigate the leaching (or adsorption) of specific compounds at  the microgram-
per-liter level. In this  case, the impact of interfering  organic compounds  on the
separation and analysis of the compounds of interest (e.g., the priority pollutants) may
be quite serious. For example, the ubiquity of phthalate esters in flexible plastic (non-
polyolefinic) materials and in the environment has caused numerous analytical problems
in the  determination of many compound classes  in natural-water samples (76,77).
     The study of the effects of water or aqueous solutions on materials and vice versa
presents many obstacles to the investigator. For leaching effects alone, there are at least
six critical system variables which must be controlled or considered, including chemical
composition of the solution; temperature; rate of flow; composition of the material, its
age, pretreatment,  and surface  area exposed. For purposes of material  selection for
ground-water monitoring, static or flowing tests with solutions approximating the range
of solution composition expected should be sufficient. Protocols for conducting such
tests are detailed  in  several publications (78,79).  Additional references have been
included in  the section which follows along with the performance data  available for
selected materials.
                                        32

-------
 Candidate Test Solutions for Materials' Evaluation
       The composition of ground water or contaminated ground-water solutions varies
 quite  markedly, so there can be no  universal test solution applicable to the'specific
• requirements  of all ground-water monitoring programs.  Tap  water or  carbonate-1-
 buffered distilled water may be a suitable starting point if raw water from the site is
 unavailable.  This  is, in effect, what  is used in the national certification program  of
 piping materials for potable-water use conducted by the National Sanitation Foundation
 (80).  The chemical composition of  leachate and- contaminated  ground water From -
 investigations of sanitary or hazardous waste landfills provides some composition limits
 for additional test solutions. Municipal landfill leachate compositions are detailed in a
 number of publications  (11,81). The use of a standardized acetic acid solution has
.several distinct advantages over a synthesized leachate for testing purposes (81). >
     "  Leachate and ground water collected from the vicinity of industrial-waste handling
 sites  are difficult  to characterize due  to the  diversity  of  sites and wastes involved.
 However,  there are some .general indications from  the  recent literature  that two
 categories of chemical composition predominate.  Briefly, from a  survey of data from
 43 monitoring investigations of hazardous waste sites  in the United States, Shuckrow
 et  al. (82) identified  high organic/low  inorganic  and low organic/high inorganic
.solution compositions as accounting for'nearly 44 percent of the sites where ground-
 water samples were collected. High-inorganic samples contained inorganic constituents
 at five times the water-quality criteria  levels, while low-inorganic samples contained less
 than  the corresponding regulated levels. Similarly, high-organic samples contained
 greater than 400 micrograms " L"1 of individual hazardous organic constituents, while
 low-organic samples had less 'than 5 micrograms '• L"1 of such substances.
       Profiles of these two loose categories of contaminated ground-water composition
 are shown in Table 7-1.
       The high-organic group typically exhibited near-neutral pH  with TOG and COD
 levels  five  to ten times  the background values, TDS values were generally two to three
 times  background  levels. The four organic compound  classes noted below account for
 5-32 percent of the TOG. Specific compounds identified in these classes at concentrations
 exceeding 0.1  mg " L"1 include  phenols (pentachlorophenol,  phenol, nitrophehol);
       Table 7-1. Chemical Composition of Contaminated Ground Water Near Hazardous Waste Sites

                                High organic/law inorganic          ~     Low organic /high inorganic
                                        6-8           '                  .  3-6
pH
TOC
COD
TDS
  Phenols
  Organic Bases  •
  Aromatic Hydrocarbons
  Chlorinated Aliphatic
   .Hydrocarbons      '                   0,1-150
  All values in mg-L  . except for pH
    denotes insufficient data to1 present a range of values
  Source: Reference 82
,, > 10 • •
25-41,000
1,000-2,000 ' .-
0.5-3 . '' Zn
0.8-25 , ' Cd '
0.1-14 Cr.'
'. 1,000-13,000
1-100
1-8
1-200
                                                       As
10-10,000
                                         33

-------
organic bases (aniline, nitroaniline); aromatic hydrocarbons (benzene, toluene, xylene,
'substituted  benzenes); and chlorinated aliphatic  hydrocarbons  (dichloroethane, trich-
loroethane, trichloroethylene, tetrachloroethylene). The presence of these slightly water-
soluble compounds in ground water would be expected to affect sorptive or leaching
characteristics of the subsurface environment, particularly toward thermoplastic or
elastomeric materials.       ..                                  "•
       On the other  hand, the low-organic group showed mineral acidity and low pH
values. It also was associated  with TDS levels five to ten times ambient background.
Several metallic'elements (Zn,  Cd, Cr) normally.present in trace amounts were observed
at concentration's,exceeding background ground-water levels by factors of 10 to 1,000,
These low  pH.'high TDS solutions  would be  expected to be quite aggressive toward
.metallic casing materials after extended exposure periods.
       It is  clear that a  reasoned strategy for ground-water monitoring must consider
the effects of contaminated water on well construction materials. Unfortunately, there
is very little'published ^information on the performance of specific  materials in varied
hydrogeologic situations.  The 'monitoring strategy  must be  tailored to  fit  unique
situations, and strict guidelines'are currently unavailable.
       In the preceding discussion,  four categories of subsurface solution  conditions
were identified.  These  categories range from carbonate-buffered  water to leachate-
impacted ground water. The  categories are outlined in Table 7-2 with the principal
chemical species identified.
       The range of chemical exposures represented by these four categories  of solution
composition should  provide general test  cases for •consideration of compatible  well
construction materials.               '    •'

               Table 7-2. Solution Composition for a Range of Ground-Water Conditions
                         _                So/ids content.               Principal soluble
      Genera! category             pH          '  (mg-L"'l                   species present
  Buffered Weak Acid              5           100-200           NaVHCO, -, H*.  H, C03 
-------
               Table 7-3. Chemical Exposures Grouped in General Solution Categories
           weak acid
 Ammonium Carbonate
 Ammonium Chloride
 Calcium Carbonate
 Aqueous Carbon Dioxide
 Distilled Water
 Potassium Bicarbonate
 Sodium Acetate
 Sodium Benzoate
 Sodium Bicarbonate
 Sodium Carbonate
 Sodium Phosphate
 Sodium Suifide
 Sulfite Solutions
      Weak acid
Acetic Acid
Benzoic Acid
Citric Acid
Glycolic Acid
Fatty Acids
Formic Acid
Gallic Acid
Hydrogen Suifide (aq.
Lactic Acid
Oxalic Acid •
Tannic Acid
Tartaric Acid
   Mineral acid/
high dissolved sotids
Acid Mine Waters
Brine Acid
Hydrochloric Acid
Metallic Chlorides
Metallic Sulfates
Mixed Acids
Plating Solutions
Seawater
Sulfuric Acid
   Aqueous organic
     mixtures
Aniline
Beer
Benzene
Butyl Alcohol
Carbon Tetrachloride
Chlorobenzene
Cresol
Methyl Alcohol
Methylene Chloride
Naphtha
Phenol
Toluene
Trichioroethylene
Wine
Xylene
second-order effects  such as  adsorption,  absorption  or leaching  were  considered.
Compatibility ratings were as follows; (3 points) no significant deterioration, embrittle-
ment, or corrosion in general use; (2 points) the potential exists for deterioration,'etc.,
and  this  material is  recommended  only  after  testing; (1  point)  the  possibility  of
deterioration, etc., clearly  exists and is likely after extended use. Rigid materials used
for well casings or sampling gear, as well as semi-rigid or elastomeric materials for
tubing and other apparatus, were treated.in the same fashion. The compatibility of
each material was ranked  after the  total rating  in each category was converted  to a
percentage of the maximum possible  score.  The percent ratings in each category served
to rank the materials in order, and the sum of the ratings for each material in all  four
categories provides an overall ranking scheme.
      There are limitations in this  approach which stem from the lack of detailed
information  on testing conditions, the subjective  nature  or varied sources of judgment
of "significant" deterioration, and  the tenuous relationship  of exposures  to pure
chemicals  or dilute aqueous solutions to actual  subsurface conditions. However, this
preliminary ranking is a starting point for in-depth consideration of materials perform-
ance under  actual site conditions. Generally  recommended materials for monitoring
applications are noted'by Pettyjohn et al. (17).
      The following rigid well-casing materials were considered: PVC I  (unplasticized
polyvinyl  chloride), galvanized steel, carbon steel, Lo-Carbon Steel, Stainless Steels 304
and  316,  and Teflon®. Flexible  (or  semi-rigid)  materials commonly  used for pump
parts, sample transfer, lines, in-line  devices or storage  vessels were rated, including
flexible PVC (plasticized), polypropylene,  polyethylene (conventional), polyethylene
(linear),'polymethylmethacrylate (Lucite or Plexiglas), Viton®, silicone, and neoprene.
Glass was  not considered among  the rigid  materials due to the unavailability of well
screening and hazards associated with its use for casings. Rating results in  each category
and overall are shown in Tables 7-4 and  7-5 for the rigid and flexible or semi-rigid
materials,  respectively. Sources for  the  exposure ratings (83-91) are included in  the
references.
                                         35

-------
                  Table 7-4, Relative Compatibility of Rigid Well-Casing Material
                          PVC   Galvanized    Carbon  to-carbon    Stain/ess    Stainless
                           I       steel       steel     steel      steel 304    steel 316    Tefton®'
Buffered Weak Acid          100      56        51       59        97        100   '    TOO
Weak Acid                  98      59        43       47        96        100       100
Mineral Acid/High Solids      100      48        57       60'       ,80         82       100
Aqueous/Organic Mixtures    64      69        73    .   73        98        100       100
Percent Overall Rating        91      58        56       59        93         36       100
'Trademark of DuPont

                            Preliminary Ranking of Rigid Materials
                                   Teflon©
                                   Stainless Steel 316
                                   Stainless Stee! 304   • ,
                                   PVC I
                                   Lo-Carbon Steel
                                   Galvanized Steel
                                   Carbon Steel


               Table 7-5, Relative Compatibility of Semi-Rigid or Elastomeric Materials
                                      PB    PE
                         flexible   PP  conv,   linear  PMM   Vitort®"   Silicone  Neoprens   Teflon®'
 Buffered Weak Acid          97    97  100    97     90     92      87      85        100
 Weak Acid                  92    90   94    96     78     78      75      75        100
 Mineral Acid/High Solids     100   100  100   100     95    100      78      82        100
 Aqueous/Organic Mixtures     62    71   40    60     49   .  78      49      44        100
 Percent Overall Rating        88    90   84    88     78     87      72      72        WO
 'Trademark of DuPont

                     Preliminary Ranking of Semi-Rigid or Elastomeric Materials .
                                   Teflon®
                                   Polypropylene (PP)
                                   PVC flexible/PE linear
                                   Viton®
                                   PE conventional
                                   Plexiglas/Lucite (PMM)
                                   Silicone/Neoprene

        Both tables  clearly  show  that superior  performance  can be  expected of the
 polymeric materials under acidic or high-dissolved-solids conditions.  However, as the
 organic content of the solution increases, one  must be prepared for either direct attack
 on  the polymer  matrix or more subtle effects  due  to  solvent  absorption, adsorption
 and/or  leaching. The only  exception to  this  observation is Teflon®. Provided that
 sound construction practices are followed, Teflon®  can be  expected  to consistently
 outperform all other casing and sampling materials. The stainless  steels  predictably are
 the most chemical  resistant of the ferrous materials. Stainless steel performance may
 be sensitive to the chloride ion, which can cause pitting corrosion, especially over  long-
 term exposures under acidic conditions.  Given  the similarity  in price, workability, and
 performance,  the  remaining ferrous materials  provide  little advantage for casing/
                                            36

-------
screen construction. Indeed,  the manufacturers' recommendations  frequently were
contradictory (89,90,91), and  as a result, the rankings are essentially  equivalent.
      The semi-rigid or elastomeric materials ranking in Table 7-5 follows the general
chemical  resistance  expected  from  manufacturers'  recommendations. The  "true"
polymeric  materials  — Teflon®, polypropylene, polyethylene, Viton®, sil'icone, and
neoprene — may have an added advantage over polymer formulations (PVC)  for
rigorous applications since they generally contain  fewer  extenders,  stabilizers,  or
antioxidants which may cause interferences in analytical  determinations. One should
be aware of the fact that many flexible materials contain plasticizers, which are potentially
troublesome  contaminants, especially when industrial solvents are  encountered  in
ground-water systems. The polyolefin materials — polyethylene and polypropylene —
are exceptions  to this statement  since they generally do not contain plasticizers and
are more resistant to organic solvent attack than the  formulated plastics.
      The general rankings provided in the above tables should serve as a basis for
preliminary considerations  of the  suitability  of specific  materials for ground-water
monitoring applications. In the next section, the available literature on actual materials
performance in water-handling systems is reviewed for both rigid and flexible materials.
This more detailed information is provided in order to permk reasoned considerations
of performance in specific  applications.


Evaluation of Selected  Materials
      Since well  casing materials are rigid and nonporous, they present a very low
surface  area to water in the well bore relative to that of the adjacent  soil or aquifer
particles. There is an extensive body of literature dealing with  sorptive  interactions of
dissolved chemical species  in natural waters with solid surfaces. Most of these studies
describe the adsorption  of  trace metals or organic compounds (adsorbates) on mineral
particles (adsorbents). Surface area  (or particle size)  and the organic content of the
solid phase are cited almost universally as important variables in the adsorption  process.
Mineral phases such as quartz (92), aluminum  (93),  hydrous metal oxides (94), and
clays  (95), as well as natural  sediments (96),  have  been studied with surface areas
ranging from  5 to over 250  m2 "  g'1. These active  surfaces have been  observed  to
adsorb routinely  up  to  several hundred micrograms of adsorbate per square meter
surface area. The applicability of laboratory adsorption experiments to  the condensed
media of the subsurface is a matter of some controversy (97,98,99,100). However, a
simple qualitative comparison  of well casing versus  subsurface  solids should suffice  to
discount adsorptive interferences from materials selection considerations.
      As an example,  let  us  assume that we have  constructed a 50-foot (15.3 m)
monitoring well with a 2-inch  (~5  cm) diameter. The well  is screened  in the  lower 3
feet (1 m)  of a  33-foot (10 m) saturated thickness, and the  standing water level is  17
feet (~'5 m) from the land  surface. This hypothetical well  bore contains  approximately
21 liters of water  exposed to about 2 m2 of casing/screen surface, roughly a  10:1 ratio
of water to casing surface.  If the solids in the saturated zone were coarse sands with a
minimum surface area of 1 m* ' g"1, the casing water volume would have been exposed
to at least 104 irr  of solid surface! If we presume that the casing has  been in place long
enough to equilibrate with subsurface conditions, we may expect that the surface
                                        37

-------
activity of the coatings (microbial slimes,  organic films, or micro precipitates) on
neighboring particles and the casing material would be roughly equivalent. Thus, the
occurrence of adsorptive bias in our representative water sample, taken after purging
the stored volume, would be likely only if the well casing presented an extremely active
surface uncharacteristic of nonporous materials and if the rates of desorption/adsorption
were very fast relative to the duration of a sampling operation.
       Although reports on the rates of adsorption and desorption in aqueous solution
are sparse, most workers have employed at  least a 4-hour  equilibration  period in
laboratory studies. Reported half-times for maximum adsorption of metals and nutrients
are in  the range of 0.5-2 hours (27,29,54). These times are much greater  than the
time necessary to sample  most monitoring wells after the removal of stagnant water
from the well bore. On this basis, the potential bias effects due to adsorptive interactions
with well casings may be discounted. Such effects would be far more critical in  sample
transfer and  storage procedures prior to  separation or  analysis. A note of caution
should be added with regard to  the absorption  of  organic solvents by polymeric
materials. Past exposures of casing or tubing surfaces to high aqueous organic mixtures
may cause the migration of organic solvents into the polymer matrix. Normal cleaning
procedures may not be sufficient to remove this potential source of contamination in •
succeeding samples.  Also, freshly cleaned materials may represent active surfaces for
sorption or leaching effects,

Teflon® Well Casing
     Teflon® represents a nearly ideal well construction material. Inertness to chemical
attack, poor  sorptive properties, and low leach potential are clear advantages of rigid-
Teflon® PFA  for well screen and casing. However, these advantages are rather expensive
in comparison to other materials. Where situations allow, using  Teflon® casing and
screen in the saturated zone with  another suitable material as  the  upper casing may
be  a viable,  less-expensive alternative (17). The structural properties of Teflon® are
sufficient for the most exacting environments, and this factor gives it a clear advantage
over glass. Teflon® has not been reported  to contribute organic or inorganic contam-
inants  to aqueous solutions.

Stainless Steel Well Casing
     Stainless steel has been the material of choice for casing and screen when subsurface
conditions require a durable corrosion-resistant material. In the preliminary materials
ranking, 316 Stainless showed a slight edge over type 304, The principal compositional
difference between the two types is the  inclusion of 2-3 percent molybdenum  in type
316 (101). The molybdenum content gives 316 Stainless improved resistance to sulfur-
containing species as  well as sulfuric acid solutions. Resistance to oxidizing  acids  is
somewhat poorer than other chromium-nickel steels; however, reducing conditions are
more frequently encountered in well-casing applications.  The 316  Stainless  Steels are
less susceptible to the pitting or pin-hole corrosion caused by organic acids  or halide
solutions. They are the materials of choice in industries where excessive metal contam-
ination of process streams must be avoided (e.g., pharmaceuticals). Provided that surface
coating residues  from manufacture or storage are removed, stainless steel well casing,
screen, and  fittings can be  expected to function  nearly as  well  as Teflon® in  most
                                        38

-------
monitoring  applications. Chromium or  nickel contamination may result after long
exposure to very corrosive conditions; however, physical failure of the casing  would
probably accompany  or precede such an occurrence. Proper well flushing  prior to
sampling should be sufficient to minimize problems with these materials (102). Details
'of well-executed monitoring efforts in which stainless steel well'casing and screen have
been used successfully are provided by several recent publications (38,103,104).

PVC Welt Casing
     Polyvinyl chloride (PVC-Type 1) thermoplastic well casing is composed of a rigid
unplasticized polymer formulation with many desirable properties for monitoring well
construction. It is a rigid material with very good chemical resistance  except to low-
molecular-weight ketones, aldehydes, and chlorinated solvents. The preliminary ranking
in the previous section establishes PVC as a  close second to Teflon® and 316  Stainless
Steel with respect to resistance to acid solutions, and it may be expected to outperform
any of the ferrous materials in acidic environments of high ionic strength. There may
be potential problems when PVC is used in  contact with aqueous organic mixtures or
under  conditions which might encourage leaching of substances from  the  polymer
matrix. It should  be noted that manufacturers do not recommend the use of threaded
schedule 40 PVC  casing because of potential  mechanical failures. Schedule 80 threaded
PVC well casing is sufficiently durable for most well construction applications.

PVC product formulations and applicable standards
      There are  few  piping  materials that have received the scrutiny  to which PVC
products have been subjected.  Discussions  of durability,  health  effects  of leachable
components, and modes of fabrication have gone on since U.S. commercial production
began in the 1940's (105). A number of standard specifications exist  covering PVC
well casing (106,107,108,109), which  is tested by the National Sanitation Foundation
(NSF) using a protocol equivalent to that for all  plastic pipe used in potable  water
applications (80). In general, the chemical resistance of polymeric materials is improved
by the incorporation of fewer ingredients in the formulation (88), and unplasticized
PVC bears this out in tests of weatherability (110).
      There have been many concerns voiced about the  release of vinyl  chloride
monomer from PVC products. Process control technologies have significantly reduced
the total residual vinyl chloride monomer  (RVCM) levels in the resin and  finished
products. In fact, the NSF chemical,  taste,  and  odor testing protocol  (67) limits the
levels of RVCM to <10 ppm in NSF-listed products. This level of residual monomer
limits the potential leached concentrations  to 1-2  micrograms ' IA From  1977 to
1980, the RVCM failure rate in NSF tests of PVC formulations fell from 9 percent to
less than 1 percent (111). It would be expected that the leachable amounts of vinyl
chloride monomer decrease as the total  RVCM  levels in products are  reduced. This
has been borne out in several laboratory and field studies which demonstrate that vinyl
chloride monomer (VCM) does leach into potable water at the,low ppb level as a result
of prolonged solution  exposures. For example, an EPA field study of five potable water
supplies with plastic pipe manufactured  between 1964 and 1975 showed VCM levels
between 0.03 and" 1.4 microgram • L'1  (43). The higher levels  observed were from
samples taken from more recently  constructed installations.  Levels of VCM leached
                                       39

-------
into water from static or flowing tests have  been shown to be a function of exposed
surface area and actual levels of RVCM in the product (42,112,113). These laboratory
results on  samples that  meet NSF Standard 14 specifications (80)  show chat leached
VCM levels are in the <1 to 2 microgram •  L"1 range. Though these levels are below
those of toxicological concern (111) the potential exists for organic analytical interfer-
ences in monitoring situations where prolonged exposure.to aggressi%re aqueous organic
mixtures may occur.
       Rigid PVC  materials used for pipe and well casing with NSF-listingare essentially
free of plasticizers. Although plasticizers are unnecessary for rigid pipe applications, a
small number of sources of well casing may include plasticizers which are added along
with the thermal  stabilizer component. Plasticizer levels in such pipe samples would
not be expected to exceed  0,01%. These levels are far lower than  those in flexible
PVC formulations for tubing or sheet materials which can contain up to 30-50 percent
by weight  of plasticizer. Rigid PVC contains several  types of other additives at levels
approaching 5 percent by  weight,  which  may ,pose a source  of bias  or analytical
interferences  in ground-water monitoring  programs (113).  These additives include
pigments, antioxidants,  thermal stabilizers, and inorganic fillers  (114,115). Some rep-
resentative chemical classes of additives which have been used in rigid PVC manufacture
are contained in Table 7-6. There are clearly many possible combinations of substances
which  may be included in  PVC formulations. The potential for their  release into
aqueous solutions  or ground waters may be determined by  the individual formulation,
rigor of exposure, manufacturing techniques, and the chemical state of a particular
component in  the finished product. Here again,  NSF-listed  products for well casing
and  potable water applications are continually checked  and tested. Products that are
found to exceed the maximum contaminant levels set by the National Interim Primary
Dnnking  Water Standards in leach tests (Table 7-7)  do not qualify to carry the  NSF
logo. Their use should be avoided in monitoring  well construction since many manu-
facturers include compounding ingredients that are not permitted by the specifications.

               Table 7-6, Representative Classes of Additives in Rigid PVC Materials
                               Used for Pipe or Well Casing
                                (Concentration in wt, %}
Hear stabilizers (0.2-1,0%)                                         fillers (1-5%)
Dibutyltin diesters of lauric                                        CaC03
  and maleic acids                                               diatomaceous earth
Dibutyltin bis (lauryimercaptide)        .                            clays
Dibutyitin-(3-mercaptopropionate                                   Pigments
di-n-octyltin maleate
di-n-octyltin-S,S'-bis tsoctyl                   !                    TiO:
  mercaptoacetate                                              carbon black
di-n-octyltin-j3-mercaptopropfanate                                 iron and other metallic oxides
Various other alkyltin compounds                                   Lubricants (1-5%)
Various proprietary antimony compounds
                                                             stearic acid
                                                             calcium stearate
                                                             gtyeerol monostearate
                                                             montan wax
                                                             polyethylene wax
(Reference 11 51
                                         40

-------
                  Table 7-7. Chemical Parameters Covered by 1MSF Standard 14
                       for Finished Products* and in Standard Leach Tests
              Parameter

Antimony (SbS
Arsenic (As)
Barium (BaJ
Cadmium (Cd)
Chromium  {CO
Lead (Pb)
Mercury (Hg)
Phenolic substances
Residual vinyl chloride monomer* !RVCM)
Selenium (Se)
Tin (Sn)

t Mot covered under National Interim Primary Drinking Water Regulations,
* Total residua! after eorrplets dissolution of polymer matrix,

Tabulated vaiues are the maximum levels permissible in NSF-l!sied products
after standardized ieach lesrnc :n weakly acidic aqueous solution. [Carbonic
acid solution with "00 rng-L"1 hardness as CaCO|j with 0,5 mg- L"1 chlorine;
pH 5,0 to 0,2; and surface to solution ratio'of 6.5 cm1 -mL~l ]
Source: National Sanitation Foundation
Maximum contaminant level lmg-L~{ i
            0.05t
            0.05
            1.0
            0.01
            0.05
            0,05
            0.002
            0.05t
           10t
            0.011
            0.05
0.05T
Practical considerations and potential for analytical bias due to use of PVC well casing
       The use of NSF-lisied well casings provides us with a minimum standard material
that can be judged for suitability  in specific  monitoring situations. AD types of well
casings should be cleaned with detergent and rinsed with water prior to well construction
to remove processing  lubricants and release agents.  This is particularly true of PVC
well casing, which may be coated with natural or synthetic waxes, fatty acids, or fatty
acid esters.
       Threaded joints are the preferred means of connecting sections of PVC well
casing. In  this way, problems associated with  use of solvent primers and cements can
be avoided. Threaded joints on PVC  well casing (or pipe) can be provided  in  three
ways: solvent  cementing a molded thread adaptor to the end of the pipe, molded flush-
threaded joints built into each pipe section, and cutting tapered threads on  the pipe
with National Pipe Thread sized dies. The latter method is only recommended by the
industry for schedule 80 PVC well casing or pipe. Numerous studies have pointed out
analytical  interferences and  direct sources of bias caused by  the  migration of the
components  of solvent mixtures into  water samples  (79,41,116,117). The mixtures
frequently contain  two  or more  of the following  solvents:  methyl-ethylketone  (2-
butanone), methyl-butylkeiones, cyclohexanone, tetrahydrofuran, and dimethylformam-
ide.  Some of these substances may not be  among the analytes of interest; however,
their presence in a water  sample can cause severe problems in the determination  of
priority  pollutants (41).  In the experience of the Illinois State Water Survey,  even
minimal  solvent-cement application is sufficient to supply consistent levels of cement/
primer components above  100  ppb in actual ground-water samples despite proper well
development  and flushing prior to sampling. This problem may persist for months
after well  construction even  after repeated attempts to develop the wells. Prolonged
exposure to aggressive aqueous mixtures is probably  the single most important factor
                                          41

-------
contributing to leaching problems. This situation is decidedly different from the high-
volume flow-through testing conditions employed for PVC plumbing for potable water
transport applications,- There are various alternative joining procedures for PVC pipe
other than solvent cementing or threaded joints. These include: twist-lock or spline
unions which  employ O-ring seals.  For these procedures the integrity of the ring
material must  also be considered in evaluating appropriate materials for well casings.
      Other sources of potential bias from the use of PVC well casing may arise from
the additives present as compounding  ingredients, such as those in Table 7-6. These
substances are  added to color the pipe, protect it from oxidation or exposure to sunlight
and aid in the  maintenance of the integrity of the virgin resin in extrusion or molding.
For example,  thermal stabilizers such as the organotin compounds (e.g., dimethyltin-
bis-isooctylthioglycolate) are added to scavenge HC1 released when the resin is heated.
Liberated HC1 would otherwise attack the polymer matrix and degrade the product.
Ideally, minimal amounts of such substances should be added to insure product integrity.
In practice, however, it  is difficult to determine ideal additive concentrations and an
excess may be present in most  finished products.  Therefore, the initial compounding
ingredients, as weil as the products of their reaction  with species generated  during
processing  or   use, the  ingredients  themselves, and the  polymer matrix, must  be
considered among the species that-may leach or migrate from the finished product.
     , At present there are very few data available on the identity and concentrations
of chemical species  leachable  from PVC pipe  under  actual subsurface conditions.
Metallic and organometallic compounding ingredients have received the most attention
in the literature due mainly to  the ease of spectroscopic metal determinations relative
to that of specific organic compounds. Specific organic compound determinations are
difficult since both the mixtures of original ingredients and their reaction products are
present in a solid matrix which is more or less insoluble (114-118).
       Lead and cadmium compounds  are not permitted as compounding ingredients
in U.S.-manufactured, NSF-listed PVC well casing, There are a series of papers from
the United Kingdom which  demonstrate various aspects of the leaching process for
rigid  PVC materials  stabilized with lead stearate or tribasic lead_sulfate (78,119,120).
In this comprehensive study, R. F. Packham detailed the equilibration of lead-stabilized
rigid  PVC over a range  of pretreatment, aging, and leach-solution  conditions.  His
conclusions include the following points: 1) total Pb content was'not simply related to
the quantity extractable, although surface area was an important variable (he also noted
significant variations along a length  of pipe and the consequences of proportionately
longer heat exposure for larger-diameter pipe);  2) the bulk of the extractable  Pb
seemed to  be  present in  a  surface-rich  layer (~30 mg •  Pb • m'3),  which could be
removed by abrasion, ethanolic-NaOH  or oxidizing acids; 3) pre-exposure of rigid PVC
to organic acids, organic mixtures, and trichloroethylene actually increased the amounts
of Pb extracted by the standard bicarbonate solution;  4)  even after  removal of the
surface layer,  additional leachable Pb could be attributed to a slow diffusion process
of the order of 0.2 mg • Pb • week"1 • m's; and 5) there was some indication that lead
stearate stabilizers would support higher leachable lead quantities than inorganic tribasic
lead sulfate. This study significantly supplemented the earlier work of Niklasand Meyer
(121). The results of this work clearly  show some of the important variables involved
in the leaching of compounding ingredients from rigid  PVC well casing. Pretreatment
                                        42

-------
and  cleaning, aging,  and  the  effects  of multiple extractions  (static or  flowing test
conditions) are factors which should be considered in PVC well-casing applications.
Actual field  conditions  may  either decrease  or  increase  the long-term release  of
compounding ingredients (81)  through surface area effects, and the chemical species
of interest must be carefully considered,
      In more recent work, Boettner et al, (40) studied the release of alkykin species
from PVC pipe stabilized with dialkyltin-bis-isooctylthioglycolate compounds. Their
results clearly showed that levels of leachable organometallics were of the order of 10-
50 ppb on  initial exposure to chlorinated, buffered,  pH  5  carbonic- acid  solution.
Concentrations of leachable tin due to successive  exposures gradually decreased in a
biphasic manner for up to 3 weeks. The actual mobile tin species were not identified;
however, there was an indication that they were ionic in nature. The leaching of related
tin compounds from coatings is recognized  to be a complex process which may occur
from an active fraction  of the exposed area (122) and is certainly affected by the
chemistry of individual species (123). The leaching process is an active area of research
which will prove most helpful in the interpretation of chemical investigations of materials
effects in many environmental applications. Though the tin or antimony compounds
used as stabilizers are rarely of interest for monitoring,, their impact as analytical
interferences and/or sources of bias must be carefully considered,
      The bulk of the data available on PVC chemical resistance and leaching strongly
suggests that there are potential pitfalls  involved in the use of PVC well  casing  in
situations where trace chemical species are of interest. At this time, it is clear that PVC
exposed to aqueous organic mixtures  has the potential to act  as a source of foreign
organic or metallic compounds in  excess of what may occur in predominantly inorganic
solutions. Detailed monitoring efforts for organic compounds at the microgram • L'1
level may be significantly biased by the sole use of PVC well casing, particularly during
the initial study period. Thereafter, the slow diffusive release of PVC additives may  be
expected to continue for some time. Whether or not  these effects significantly bias
monitoring results will depend on specific conditions and the actual formulations used.
Caution is indicated from the available data on these processes. .
      The potential for solvent absorption, and the adsorptive uptake or release  of
organic compounds and metals by PVC pipe, has been discussed by several authors
(39). However, there are few data documenting this potential  under field conditions.
The  mechanism for solvent-cementing lies in the absorption of the solvent and partial
dissolution of the  PVC matrix  to produce a "solvent-weld." Therefore, the  exposure
of PVC to  low molecular ketone, aldehydes, acids, amides,  chlorinated alkenes  or
alkanes may cause the actual degradation of the polymer matrix and/or the release of
compounding ingredients which otherwise would remain in the solid. The concept of
discrete adsorption "sites" on PVC which may be expected to  equilibrate with stored
water in monitoring wells (124)  may be  applicable  to  certain  situations, though the
inevitable adsorption  of natural organic matter and microbial  coatings argue against
any specific  adsorptive interaction.  If circumstances do not  permit the use of a more
appropriate  material  (e.g.,  Teflon®, stainless steel) under  high-organic  or  unknown
conditions, then at least several paired wells should be constructed of a non-polymeric
material and PVC. This will allow at least an order-of-magnitude determination of the
potential bias due  to PVC  well casing.
                                        43

-------
      Further, manufactured casing and screen is preferable to off-the-shelf PVC pipe.
The practice of sawing slots in the pipe (e.g., home-made screens) should be avoided
since this procedure exposes fresh surfaces of the  material, increasing  the  risk of
releasing compounding ingredients or  reaction products. Also, it is very difficult to
properly slot casing materials by sawing operations.

                                                          i f
Casing Made from Other Ferrous Materials
      Ferrous  metal  well casing and screen materials, with the exception of stainless
steels, include carbon steel,  low carbon  or copper (0.2%) steels, and various steels with
a galvanized coating. In the  preliminary ranking contained in Table 7-4, these materials
ranked consistently poorer  than Teflon® and stainless steels. They do, however, show
an advantage  over rigid PVC in exposures to aqueous organic mixtures. The carbon
steels were formulated to improve resistance to atmospheric corrosion. To achieve this
increased resistance, it is necessary for  the material to  undergo alternate wetting  and
drying cycles. For  non-coated steels buried  in soils or in  the saturated zone,  the
difference between the corrosion resistance of either variety  is negligible (125). Both
carbon- and copper-steel well casings  may be expected  to  corrode, and corrosion
products may include oxides of Fe and Mn (and trace constituents), as well as various
metal sulfides. Under oxidizing conditions, the principal  products are  solid hydrous
oxides of these  metals, with a  large range of potential particle sizes.  The solids  may
accumulate in the well screen, at the bottom of the well, or on the casing surface.  The
potential also  exists for the  production  of stable colloidal oxide particles that can  pass
through  conventional membrane  filtration media  (126),  Reducing  conditions  will
generally provide higher levels of truly dissolved metallic corrosion products in  well
storage  waters (127,128). Galvanized steels are protected by a zinc coating applied by
hot dipping or electroplating processes, The corrosion resistance of a  galvanized steel
is generally improved over conventional steels; however, the products of initial corrosion
will  include iron, manganese,  and zinc (and trace  cadmium) species which may be
among the analytes of interest in a monitoring program.
      Corrosion  products from conventional or galvanized steels represent a potential
source of adsorptive interferences, The accumulation of the solid products has the
effect of increasing both the activity and  the exposed surface area  for adsorption,
reaction, and  desorption processes. Surface interactions can thereby cause significant
changes in dissolved metal or organic compound concentrations in water samples (102).
Flushing the stored  water from the well casing may not be sufficient to minimize this
source of bias because the effects of the disturbance of surface coatings or accumulated
products in the  bottom of the well would be difficult, if not  impossible,  to predict. In
comparison with glass,  plastic, and coated steel surfaces, galvanized metal presents a
rather active  surface for adsorption of orthophosphate (29). The age of the surface
and the total area  of exposure  have  been found to  be important variables  in the
adsorption process;  however, adsorption is not a  linear function  of galvanized-metal
surface  area.
      Field data  for  conventional and galvanized steels provide additional reasons for
caution in their use for well casings  or screens. The water well industry  routinely
chooses alternative nonconductive or corrosion-resistant materials in areas where normal
groundwater  conditions are known to  attack the  common steels.  In fact, regional or
                                        44

-------
local practices in the selection of water well  construction materials provide  valuable
preliminary-guidelines, for routine, monitoring efforts.

     The  preliminary ranking of ferrous materials  in  the  previous  section is  well
supported  by available monitoring data. Results from studies of the glaciated terrain
of Maine demonstrate that water samples from steel production  wells used for monitoring
consistently showed Fe and Mn levels 30-50 times higher than  those of samples from
adjacent PVC monitoring wells finished at the same depth (129).  In thess data, water
sample  composition {DO, T, and TDS)-was remarkably similar, with the exception of
iron and manganese levels. Again, though  iron and manganese may not be among the
parameters of interest in a  specific situation,  the  presence of their, metallic oxides or
soluble  complexes may interfere with the determination  of other metals. This source
of bias  may  be  particularly serious in  unfiltered  samples which cannot be preserved
properly in the  field,                    ,      :         :
     Galvanized steel well casing and screen  may seem  to be a less costly alternative
to the use of more appropriate materials (e.g., Teflon®, stainless'steel, PVC) in saturated,
high-dissolved-solids environments since it corrodes less rapidly than conventional steels
(130,131).  Under reducing  conditions at pH values between 5 and 7, the presence of
chloride, carbonate, and nitrate can encourage rapid aggressive attack of the material,
In some cases,  CO3= and  NO3~ may actually reverse the electrochemical  potential
between the  zinc oxide coating and the base metal, resulting in accelerated dissolution
of the iron pipe. Sulfur compounds, organic compounds, and dissolved copper concen-
trations  also are implicated in  the  rapid deterioration  of  galvanized  steels  under
saturated conditions. Monitoring data from wetland soils and bog environments disclose
that galvanized  imaterials are a liability/ in ground-water investigations. Instances of
casing  or screen dissolution under these conditions  resulting in  zinc  concentrations
approaching  10 ppm are common (38,132,133).  In  these cases, water samples "from
PVC and stainless  steel  monitoring  wells showed zinc concentrations  1-2 orders of
magnitude below those observed in samples from adjacent galvanized steel wells despite
'the fact  that comparable well  flushing techniques were  used. This is consistent with
the dissolution and migration of zinc from galvanized monitoring installations, providing
an additional source of bias that- careful flushing does not minimize.
     It appears, ttherefore,  that in corrosive  environments, galvanized steel  presents
little or-no advantage over .conventional steels, and more  appropriate materials should
be considered.  Some degree of compromise may be achieved by casing the  upper
unsa'turated zone with steel casing and then  using PVC, stainless, or  Teflon®  casing
and screen in the saturated  zone. However, the potential for abrasion during sampling
operations  or galvanic corrosion effects'should be evaluated prior to installation.


Pumps Used in Development
     The  large variety  of centrifugal, peristaltic,'impeller,  and  submersible  pump
designs  precludes an in-depth, discussion of their potential effects on the results of
ground-water monitoring efforts. According to the situation, one must carefully consider
the compatibility of the materials/ found in  high capacity  pumps  with subsurface
conditions. The methodology of monitoring  well development is probably far more
critical in this respect than the pumping mechanism or water-contacting materials. One
                                        45

-------
of the most serious problems, arises in the use of water of poor or indeterminate quality
for surging or jetting operations. This'water can migrate into the formation of interest
and alter local aquifer properties. Exhaustive pumping of a well after development to
determine hydraulic  conductivity from  time-vs-drawdown or recovery curves should
be sufficient to minimize this problem.
  ,,   Special precautions may be: called for in severely contaminated areas and where
flammable or toxic materials are present in ground water. In,these, cases,  a'plan  for
the management of water (as welTas cuttings) extracted during drilling or development
should be designed.  All'such operations should be carefully supervised  by the'chief
field engineer or  scientist.  The practice  of returning  contaminated water to the
formation is strongly discouraged since there is no tested procedure to account  for
changes which occur during surface storage or mixing.

Grouts, Cemenfs, Muds, and Drilling Fluids                                    \
   ,  'Various drilling aids, cements, and sealant formulations are  used to achieve two
,main goals:  to  maintain an open borehole  in rotary and cable tool operations in
unconsolidated  formations,  and to effect  a  seal between  the surface or overlying
formations and the casing/screened intervals so that runoff or other sources'of water
do not enter the well  bore. Problems involved  in obtaining representative water samples
stem mainly from the persistence of drilling aid components and the long-term  integrity
of grouts or cement seals.

Drilling aids  '                                                        *
       For the most part, water-based drilling fluids are used in freshwater applications
where the total-dissolved-solids content of ground water is below 10,000 mg •  L"1. The
fluids are introduced for several  purposes, including cooling and lubrication of the bit,
suspension and removal  of cuttings, stabilization of the borehole by building up a cake
on the sides of the hole, and minimization  of formation "damage  due to water loss or
penetration-of solids.  Freshwater muds are formulated mainly in  three types: 1)
bentonite, attapulgite or clay-based muds with pH  adjusted  to 9-9.5 with caustic; 2)
polymer-extended  clay  (organic)  muds; and 3)  inhibited  clay  muds  which utilize
lignosulfonates or lignin to counteract the effects of contaminants which would otherwise
destabilize the slurry and prevent effective cutting removal. The first two types are
used most frequently in  water-well drilling applications (134). Both of these main types
of mud formulations and a  spectrum of combined compositions  have been  used  for
the construction of monitoring wells. The basic ingredients in a drilling mud are shown
in Table 7-8, and the  main distinction L between  bentonite  and organic  muds is  the
addition of natural or  synthetic organic polymers  to  adjust consistency, viscosity, or
surface tension.
       For monitoring applications, there are distinct advantages to angering, air-rotary,
or clear-water rotary drilling techniques where conditions permit. The desired approach
is the least possible introduction of foreign materials into the  borehole. Compressor
lubricants for air-rotary rigs may rule out  this method for trace organic moniloring
work although''filters are available'to minimize such problems. In geologic situations
where water-based drilling fluids  are a necessity,,the predominantly inorganic clay muds
are  preferable  over those containing organic materials,  since  the ."introduction of
                                         46

-------
                      Table 7-8. Components of Well-Drilling Fluids
       Functional class                                    Examples
      Inert solids                          calcium and barium sulfates
      Inorganic salts             '          sodium chloride
      Active solids                         bentonite (~90% Na-monimorillonite clay),
                                        attapulgite, calcium carbonate
      Bacterrcides                         formaldehyde, hypochlorite
      Organic polymers                     partially hydrolyzed polyacrylamide, corn
                                        or bean starch, carboxymethylcellulose,
                                        copolymers of acrylamide end sodium
                                        acrylaie, lignosulfates/lignin
      (References 57, 134, 135, 136!

substrates for microbia! activity can seriously  impact the integrity of water samples
(57,58).  The decomposition  of the organic  components of drilling  muds may be
expected to  be a  function of their chemical -structure, microbial  populations, the
presence  of  nutrients,  and various physical  and chemical  factors  controlling the
distribution of organic substances in the subsurface. Studies to dale have been confined
primarily 10 oil or gas drilling operations in marine or subarctic regions where permafrost
impacts have been  investigated (137,138).
      Comprehensive freshwater drilling studies of the persistence of organic additives
are sparse; however, Brobst  and  Buszka (59) have  investigated the problem  using
chemical oxygen demand  (COD) as an indicator  of  the  organic mud.  In  this study,
monitoring wells were constructed by rotary methods using various clay and organic
muds. Since background COD levels were in the range of 2-10 mg • L' and the drilling
muds of interest were in the range of 1-10,000 mg • L"1,  COD was a reliable indicator
of persistent  analytical bias. Their  results show that, despite careful well development,
and flushing  prior  to sampling, COD levels from  mud-drilled -wells were consistently
3-10 times higher than the background levels  measured in adjacent wells.  Somewhat
longer persistence was noted for bentonite  muds as compared with organic muds, and
the effectiveness of supplements to  speed organic decomposition was not uniform. The
effects were found  to persist from  20 to  120 days after well completion.
      The potential consequences  of using drilling aids should be obvious.  Although
COD is a surrogate parameter for a number of reduced inorganic  and organic chemical
species, levels 3-10 times background can be expected  to seriously bias analytical results
in ground-water  monitoring programs. Filtration or well flushing prior to sampling
may not  be sufficient  to reduce mud-related bias. The stable colloidal-sized particles in
some muds will readily form emulsions in organic solvent extraction procedures. The
addition  of large amounts  of organic substrate for microorganisms into the subsurface
will have the effect of lowering  the oxidation-reduction  potential,  perhaps drastically
shifting both the chemical and biochemical processes in the vicinity of the  well  bore.
Since organic decomposition processes  will  depend largely on microbial activity,  there
may be  significant differences in the dominant pathways, rates, and products  from
installation to installation.  It is also most unlikely  that biocides, enzyme supplements,
or chemical oxidants  (hypochlorite) %vill totally  ameliorate the situation. An additional
word of  caution should be added against the return of mud-contaminated cuttings to
the borehole, since this will only increase  the amount of foreign material in the vicinity
of the sampling interval.
                                       47

-------
Seals, grouts, and cements
      Seals, grouts, and cements are the primary safeguards against the migration of
water from the  surface and  from overlying or adjacent formations into monitoring
wells. Surface seals also must be completed  with concern for the security at the well-
head by including, casing sheaths and locking caps.  Most seals between  the formation
of interest and regions above or below are made by the addition of clay materials or
cement. Bentonite clay can  swell from 10-15  times  in volume  after wetting with
deionized water. Variations in the composition of the contacting solution can severely
reduce the swelling of clay seals. Swelling volumes of 25-50 percent of  the maximum
values are not uncommon. The organic content of the solution in contact with the
clay can have a  dramatic effect on the integrity of the seal.  Organic compounds can
cause significant disruption of normal shrinking, swelling, or dehydration of the clay
lattice during alternate wetting and drying cycles (139). Alcohols, ketones, and other
polar organic solvents have a significant potential for these changes. On the microscopic
level, these phenomena can materially increase the  permeability of the clay seal. This
is presently an active area of  research which has wide application in well construction,
landfill liners,'and slurry or  grout cutoff walls (56,140). Macroscopic changes in the
permeability of clay or cement seals can occur due to solution channeling by aggressive
solvents, compaction or subsidence, and freezing and thawing processes at the surface.
      Chemical-resistant and  expanding cement formulations effectively minimize these
problems. Faulty seals or grouts can seriously bias the analytical results on water samples
from the formation of interest, particularly if water quality conditions vary or surface
soils are badly contaminated.  The impact of leaking seals may go far beyond the realm
of analytical interferences or  non-representative samples. A leaky well bore may act as
a conduit  to permit rapid contaminant  migration, which otherwise would  not have
occurred. This is one aspect of a ground-water monitoring  program which should not
be left  to an unsupervised drilling  crew or last-minute  substitutions  for preferred
materials.

Evaluation of Sample Collection Materials
     The  choices of sample collection devices, procedures, and  all  materials which
ultimately  contact water samples are probably the  most  critical  considerations in a
ground-water monitoring program.  The materials' related problems which may be
encountered  in well construction are secondary to those involved in sample  handling.
The careful  monitoring program planner must evaluate the potential of collection
mechanisms and all materials which  contact  the samples to  introduce interference or
bias  into the final analytical result. For example, a collection mechanism that creates
turbulent transfer of the  sample and the opportunity for  gas  exchange (e.g., air-lift
pumping  mechanisms) is clearly inappropriate in sampling  for  volatile organic com-
pounds  and pH- or redox-sensitive chemical species. The act of sampling alone would
alter solution composition, introducing bias'into subsequent analytical determinations
as well as  matrix effects which may not be totally accounted for by spiked field samples.
      Desirable attributes for sample collection materials include:
      1)  Durability, reliability, and ease  of repair
      2)  A minimum  number of moving parts or combinations of materials
                                        48

-------
       3)  Capability of being cleaned and sterilized effectively to prevent cross-contam-
          ination between sampling points
       4)  Capacity  for being checked for deterioration or malfunctions such as stuck
          check valves, clogging, and breakage
       5)  Verified low potential For introducing contamination, bias,  or interferences
          into the analytical results

       Each of these attributes plays an  important role in the overall performance of
monitoring efforts and bears directly on the successful retrieval of respresentative water
samples. The difficulties involved in the evaluation of materials for sample collection
apparatus stern mainly from the variety of combinations  of components in pumps (or
other samplers) and the properties  of polymeric and elastpmeric  materials for  tubing
or transfer lines,

Sampling devices
       Apart from  the actual mechanisms employed by sampling devices, the consid-
eration of materials is  of  prime .importance  in  the. choice of a suitable sampler,
Fortunately,  most types  of  devices  are  constructed in several  models which may  be
chosen for specific  monitoring situations. For example, bailers are presently fabricated
by commercial suppliers in  Teflon®,  stainless/Teflon®, stainless/PVC 1, and  totally
PVC  I. These materials satisfy the major critical materials' specifications. Problems
arise in materials selection with samplers employing non-rigid components. Even those
devices which incorporate a single pair  of 0-rings  may be limited in their application
by the material employed.
       The preliminary ranking of materials in Tables 7-4 and 7-5 serves-as a general
guide for materials selection for sampling devices.  Teflon® again incorporates most of
the characteristics  for an  ideal  material in sampling applications, However,  it is a
difficult material to machine and threaded components are  very  easily damaged. For
chemical resistance and  durability,  several materials other than stainless steel may  be
expected to perform satisfactorily in low-organic environments. These materials include
polypropylene, linear polyethylene, plasticized PVC, Viton®, and conventional polyeth-
ylene,  Viton® is a preferred material  for elastomeric parts,  since it may be expected
to give improved-chemical resistance over silicone  and neoprene.
       One may expect that the least desirable material in a  given sampler design will
eventually cause monitoring problems. For example, significant differences have been
observed  in the organic contamination  potential of impeller pumps as a function  of
impeller material (3). In this instance, consistently high PCB (polychlorinated biphenyl)
values  due to  cross-contamination  were observed in samples obtained  by  a  plastic
impeller pump relative to those obtained from a device with  a stainless steel  impeller.
Similarly,  solvent cements used in the construction of rigid-PVC  bailers can  result in
gross analytical errors for volatile organic compounds (41) in  samples collected shortly
after bailer fabrication.
       Flexible materials  in collection devices can be particularly problematic since they
owe their flexibility and resiliency to plasticizers as well  as a range of compounding
ingredients no less diverse than those contained in Table 7-6 for polyviny! chloride.
Here, a considerable overlap between materials selection for samplers and tubing  or
transfer lines arises. Careful consideration should be given to flexible components  of
                                         49

-------
diaphragm, bladder, or peristaltic pumps based on the discussion in the next section,
The intimate contact of water samples with tubing materials requires that absorption,
adsorption, and  leaching  potential be carefully considered for specific materials in
monitoring applications. True polymeric materials like the polyolefins may be expected
to cause fewer -problems than formulations or "sandwich" materials.

Tubing and transfer lines
      Tubing and transfer lines are available  in a variety of polymeric or elastomeric
materials.  Certain  applications (e.g.,  peristaltic or bladder pumps) demand  a  high-
resiliency material, and it  may be necessary to sacrifice chemical resistance to achieve
the desired structural performance. The bulk of common  tubing  materials,  with the
exception of Teflon®, contain a .wide range of additives. In addition  to the major classes
of additives in  Table 7-6, plasticizers,  lubricants, antistatic agents, tackifiers, and  other
ingredients may  be  present in flexible  synthetic materials  (115).  In  general,  true
polymers (e.g., polyolefins like polyethylene  and polypropylene) contain far  lower
amounts of such  ingredients. Formulations change frequently as manufacturers  strive
to keep production costs low,  so a  particular  material may show significant variation
from lot to lot. Piasticizers are frequently present at levels between 15 and 50 percent
of the total weight of flexible products. As a result of this fact and widespread plastics
usage, major plasticizers, such as phthalate esters, have been consistently identified in
environmental samples.
      There are numerous lab and  field  studies which detail the contamination of
water samples by contact with plastic tubing. Plasticized PVC is a particularly problematic
material in this respect, showing a  high potential for both the absorption/release of
organic compounds (3), and the leaching  of  compounding ingredients (49,141). By
these mechanisms, PVC tubing may contribute to cross-contamination between sampling
points as well as directly bias the determination of several classes of priority pollutants.
Several  studies have reported  data  that illustrate  such interferences, particularly due
to the presence of phthalates (142,143). Table 7-9 shows the frequency occurrence of
phthalate esters in industrial wastewaters (144) and ground-water samples. A comparison
of the frequency  levels at similar analytical sensitivities from USEPA monitoring in the
"Superfund" program (145) and those from the New York State study (143) relative
to that  in  industrial wastewaters is quite revealing.  Careful  sampling and analytical
quality assurance procedures drastically reduce the overall frequency of detection. This

                    Table 7-9. Frequency of Occurrence of Phthalate Esters
                          in Wastewater and Ground-Water Samples
                                                       N. Y. state            "Superfund"
                                   Industrial             public water            monitoring
        Phthalates                   wastewaters            supply wells             samples
bis-(2-ethy!hexyl) phthalate              42%                   98%                  0%
Dibutyl phthalate                      19        •            72                   4,8
Diethyl phthalate                       8                    35                    1.9
Butylbenzyl phthalate                    8                    26                  <1
Dioctyt phthalate                       6                    11                     1.1
Number of Samples                   2532-2998                56              '1150
                                  Savg. 2617)
Reference             ,              112                    39         '         113
                                         50

-------
 is supported by  the  fact that  the  "Superfund" sites  were of concern as potentially
 contaminated areas, while the New York State samples were from public water supply
 wells. It is quite  difficult to determine from the reported data whether analytical bias
 entered into these data sets from sampling or analytical procedures.  This is an area
 which deserves careful  attention in an overall ground-water monitoring effort,
       Polyethylene and polypropylene are clearly superior plastic materials for sampling
 applications in situations where Teflon® is not cost-effective. Teflon® is  the tubing
 material of choice in monitoring for low-level organic compounds in complex, chemically
 aggressive environments. Silicone rubber tubing for moving components of sampling
 devices represents a special case where alternate choices of material may not be feasible.
 The material is available in several grades which have widely varying compositions and
 additives.  Metallic  contamination from certain laboratory  grades of silicone rubber
 tubing can be quite serious at the ppb level. Fe and Zn concentrations 2-5  times those
 of control samples are  not uncommon even after short  contact times (146). Medical-
 grade silicone rubber tubing is relatively free of unreacted organic initiators (peroxides)
 or zinc and is a reasonable alternative to other tubing formulations. Together with the
 potential bias introduced by the suction mechanism of peristaltic pumps, the need for
 silicone rubber tubing makes it a poor choice of sampler  for detailed organic analytical
 schemes. The use of other elastomeric materials, such as natural rubber, latex, neoprene.
 or chloroprene, is not recommended for transfer lines or surfaces that contact ground-
 water samples.
      There is little information available on  the performance of flexible materials in
 ground-water applications. From the available  observations, Teflon®, polypropylene,
• and linear polyethylene may be expected to  outperform plasticized PVC, since they
 have superior chemical resistance over a range of environments and are less  likely to
 cause contamination or bias problems. Microbial transformations of plastics' additives
 introduces another dimension to the problem  posed by materials with high concentra-
 tions of additives.  There are a number of reports on the microbial  colonization of
 flexible PVC and the degradation of plasticizers from  the polymer matrix  (147).
      The closest evidence to actual  field testing of flexible materials comes from the
 testing of landfill liners  by exposure to municipal leachate (148) or actual waste streams
 (149). These observations were obtained on plasticized PVC and 11 other landfill liner
 materials. A major conclusion of the work confirms the  overall superior performance
 of plasticized PVC over most elastomers or chlorinated polyolefinic materials.

 Storage containers
      The choice of materials for sample storage containers for monitoring programs
 has been made by the USEPA after 15-20 years  of development by- environmental
 scientists all over the  world. The choice of container materials is dictated by the group
 of analytes of interest and the prescribed preservation techniques. Inorganic constituents
 are determined  in samples stored in high density linear polyethylene bottles, with the
 exception of ammonia, sulfide, or ferrous iron. These species demand oxygen-imperme-
 able  glass containers and short storage times prior  to analysis. Organic  chemical
 constituents, including  priority pollutant classes, TOC, and COD, are required to be
 stored in glass bottles with Teflon®-lined screw caps.
      Details on  specific preservation  procedures, sample volumes, and limits on sample
 storage times  are contained in several reference works (150,151). It is reasonable to

                                       51

-------
expect that preservation' procedures for organic surrogate parameters such  as BOD,
COD, and TOC are minimal steps to preserve samples for specific organic compounds
at-the ppb level.

Sources of Error in Monitoring Efforts
     In Figure  7-1, sources of error contributing to the overall variance (square of
the standard deviation from the mean) of ground-water monitoring program data are
schematically depicted. Site selection, sampling, and the analytical problems of meas-
urement,  reference samples and data handling are the principal  contributing factors.
The overall variance (S2) associated with each factor must be known and minimized in
order to permit reliable comparisons of related  data  sets, The relative magnitude of
analytical  error  and errors due to site selection or sampling will determine whether
systematic problems in network design, genuine trends, or significant concentration
differences can be observed in related samples.
     Errors may be classed as systematic or random. Simply  stated, random errors
enter into overall determinations by handling or human failures and contribute to the
lack of precision in a methodology.  Systematic errors  are the inherent sources of bias
or inaccuracy in an overall determination which may consistently prevent the reporting
of an accurate  result.  In  ground-water monitoring,  the overall determination  of a
chemical  constituent in a sample from a  particular well  includes flushing of standing
water to permit collection  of a representative sample; sample collection;  handling and
storage; and execution of the appropriate analytical, laboratory method.
     For  example, if a particular  chemical constituent can be determined analytically
with a precision  of ±20% relative  standard deviation, then errors arising from site
selection or sampling must be less than ± 20% to permit reliable evaluation of statistically
significant differences  in related samples.  If, on the other hand, systematic sampling
SITE
SELECTION
Sg = S VERT. + S HORIZ.


SAMPLING
S2
*!
Ss = S SYSTEMATIC +
S RANDOM + S NAT'L.


                                               MEASUREMENT
                                                 METHODS
                                                 REFERENCE
                                                  SAMPLES
  DATA
HANDLING
                   Thus the overall variance = S2 = S^ + S2 + S^ + S^ + S^

      Figure 7-1. Sources of error involved in ground-water monitoring programs contributing to total variance
                                         52

-------
 error (biased  accuracy) caused by the introduction  or  loss of this constituent From a
 particular sampling results in worsened overall precision, there is a danger of reporting
 a false positive or no trend when, in  fact, an unbiased series of samples would show
 the opposite to be true. This type of problem may go undetected and has been discussed
 by several authors (102,152,153, 154).
      To ensure quality and inter-comparability of ground-water monitoring data, one
 must seek to  eliminate the systematic sources of error (contamination, loss, or intro-
 duction of an  interfering constituent) and minimize the  random errors due to handling
 or  human  intervention. Systematic sampling errors could arise from  poor drilling
 techniques,  the use of persistent drilling fluids, improper well development, poor choice
 of well casing,  or inappropriate sample handling  materials for  the hydrochemical
 environment.  Sample collection mechanisms  and techniques also can significantly bias
 analytical results.
      It is obvious that the choice of materials  is a controllable source of systematic
 error. Whether or not materials' selection can adversely affect the quality of the analytical
 data will depend on the concentration range for the specific  chemical constituents of
 interest, the magnitude of analyte  loss, contamination  or interference contributed by
 the material, and the performance data for the  respective analytical methods.

 Comparison of analytical method performance with materials' related errror
      In order to best present the background information on various materials which
 will aid in making an appropriate selection, we  have chosen two degrees of analytical
 detail for hypothetical monitoring programs. The simplest program is patterned after
 minimum RCRA. compliance involving TOC (total organic carbon), TOX (total organic
 halogen), pH, and O'1 (specific'eonductance) determinations. The more detailed approach
 will  involve a complete priority pollutant scan on the  ground-water samples. The
 reported precision, accuracy,  and routine detection limits for the analytical determi-
 nations  will  be discussed.
      A minimal monitoring program may include  the  determination  of  selected
 indicator variables such as pH or conductance and organic surrogate parameters (TOC
 or TOX) to detect changes in the chemical quality of ground water. The introduction
 of acidic or  basic contaminants in excess of natural  buffer capacity would be indicated
 by pH changes, and these, as well as other inorganic substances, may cause a change
 in ionic composition which would  show up in the  solution conductivity. Increases in
 TOC and TOX may be expected to signal the introduction of organic compounds or
 halogenated organics.  The  significance of changes  in  these parameters  over  time is
judged by statistical comparison  of results from upgradient and downgradient wells.
 One may argue that if both wells were constructed identically, the effects of materials
 on the analytical results would be minimal. This may be true when the hydrochemical
 environments  at each well site are .identical. However,  it is precisely when there is a
 chemical difference between the subsurface zones as a result of a release that the need
 is greatest for reliable  signals  from high-quality  (unbiased) analytical data. Table 7-10
 contains performance data on the analytical  determinations of TOC, TOX, pH, and
 O"1. The table  includes precision data for repetitive determinations at various values of
 these parameters and corresponding accuracy information. Precision is reported as the
 relative standard deviation (r.s.d.) expressed as a percentage of the mean value. Accuracy
                                         53

-------
            Table 7-10. Analytical Performance Data for Selected Water Quality Parameters
    Parameter
    (method)
 TOC
 icombusticn-
  infraredj
 TOX
 (C-trap, pyrolysis,
  conductivity!
 pH
  (eiectrometnc)
 (conductivity
  bridge)
   Type of
determination'
                    Value
             107
             36,1
              0.09mg-C1-  •
                       org
IT'
              3,5 (pH units)
              7,1       j
              8.0
              7.7
             1 00 (micro Siemens)
             808
            1640
             536
 Precision
 (relative
 standard
deviation)
   %
  80
   7.8!

  14.0
  17.9
   6.0
          3.0
          2.8
          1.6
          1.3
          7.6
          8.2
          6.5
          1.1
                 Accuracy
                 fas bias}
+ 15.27
 +1.01

-11 4
-21.9
 +7.0 •
            -0,29
            +1.01
            -0.12

            -2.02
            -3.63
            -4.54
Practical
detection
  limit
tmg-L~lJ

 0.10
 0.01
          NA
                                                                       NA
Reference
   150


   155


   156



   150


   150
   150


   150
   = Interlaboratory Comparison; S = Single Laboratory Results
is  reported as bias or the difference of the analytically determined value  from the
"true"  or assigned value. Accuracy is the critical value to examine for the effect of
systematic errors.'  In an  actual monitoring program, one would be careful to note a
significant decrease in precision (increased r.s.d.) for repeated determinations on natural
samples, since the  "true" or assigned value is not known.  The goal, of course, is to
get representative samples.
      From Table 7-10 it is clear that the more  involved procedures for the surrogate
parameters (TOC  and TOX) involve  considerably larger relative standard deviations
and more variable  values of analytical bias than do the direct instrumental methods
for pH  and 0"'. To some extent "this is due to differences in laboratory conditions and
operator  consistency, and in the overall  efficiencies  of  the methods  for  standard
compounds. For high-quality data, sampling or  site  selection bias should be less than
the values for analytical bias. Therefore, one should weigh  the potential for materials'
effects  in  situations where  they may  contribute bias approaching that practically
attainable  in  the analytical method. For example, consider the use of plastic storage
bottles  which can contribute 2-3  rng • C • L"1 consistently  to TOC samples. At TOC
levels below 10-15 mg • C • L"1, this level of leached material would bias the determination
by more than 20 percent which would exceed the analytical bias. A grossly inaccurate
analytical result  would be observed, so this material is clearly inappropriate for the
application. A similar treatment would extend to the other parameters in Table 7-10.
      The overall  determinations of the surrogate parameters (TOC, TOX) or solution
properties such as pH and conductance are probably not very sensitive to materials'
effects,  since they measure classes of compounds or electrolyte properties. Gross errors
in this type of program would probably result from improper  construction or sampling
techniques, particularly inadequate well flushing prior to sampling,
                                          54

-------
       In a more detailed analytical data collection effort, the effects of materials may
be  much  more critical,  since  low levels of degradation or corrosion products and
leachables could directly bias the determination of specific chemical  species at concen-
tration levels well above detection or quantification limits. Sampling bias, apart from
material-related errors, remains a large gap in our understanding of overall inaccuracies
in the data. In this respect, each case must be carefully considered with regard to the
relative magnitudes of contributing errors for the .contaminants of interest.
       Tables 7-11 and 7-12 contain analytical method performance data for inorganic
and organic  chemical  constituents  which  might  be  included in  a  comprehensive
interpretive monitoring program. The performance data on these analytical methods
reflect the optimum condition of good  laboratory practice and high-quality reagent
water. Table 7-11 details performance data for inorganic chemical  parameters which
provide an indication of water quality and  the presence of inorganic contamination.
The precision and accuracy values are  expressed as  they were in  Table  7-10. -The
tabulated values indicate that routine analytical reproducibility is generally better than
20  percent and that  the mean  may be expected to be within about  10 percent of the
"true" value. Detection limits for the anionic constituents are in the  pprn  range, while
those for the metallic elements are in the low ppb range. The same overall accuracy
and precision ranges apply to the priority pollutant organic compounds in Table 7-12.
Modern analytical instrumentation enables the precise determination of many metallic
elements and organic compounds on a routine basis. Although accuracy varies somewhat
for compounds such as benzidine and the phthalate esters, a conscientious laboratory
staff can adjust raw analytical results for actual recoveries through the use  of suitable


        Table 7-11, Analytical Performance Data for Selected Inorganic Chemical  Constituents
                                                Precision
                                                 (relative              Practical
                                                standard    Accuracy   detection
    Parameter                           Type of      deviation}     fas bias!      limit
    {method!                        determination*   ,    %         %      (mg-L~'J     Reference
 CL~                                  I      '       9.1         2.16       1.0        150
 (titrimetry)                                          31        _Qgi       ^g
                                      S            2.9          -        -
 SO^                                  I           19,3        -8,26      1.0        150
 (turbidimetry)                                        5.9        -1.70       1.0
 Alkalinity             •                 |           15.9       +10.61       -         150
 (titrimetry}                                          4,5        -7.42
 N0"3                                  I           43.7        +8.30      0,1         150
 (colorimet'ry)                                      17.3        +2.82      0.1
 Major Cations:  Ca, Mg, Na, and K           S            4.4        ±2        0.01        150
 (AAS)t
 Trace Metals; Fe, Mn                     j          '27.8        *3        0.03        150
 (AAS)t
 Pollutant Metals: Cu, Cd, Pb, Ni, Zn         S            4,6        +6     '  0.01      155,157
 (AAS)  tt                              S            5.2        ±7        0.01
 *l = interlaborstory Comparison; S - Single Laboratory Results
 (AAS)t = Atomic Absorption Spectrometrv-Direci Aspiration/Flame
 (AAS)Tt - Aiom'ic Absorption Spectromatry-Graphite Furnace/Flameless
                                          55

-------
          Table 7-12. Analytical Performance Data* for Selected Organic Chemical Constituents
         Compounds

Vo tat iles
   benzene
   chlorobenzene
   1,1-dichtoroethane
   1,1,2-tnchlorcethane
   trichlcroethyiene
   tetracMoroethylene
   chioroform
   ethyl benzene
   toluene
   o-xylene
Bass-Neutrals
   anthracene
   banzidine
   benzo (A) pyrene
   bis (2-ethyihexyl) phthalate
   2-chloron3phthalene
   chrysene
 .  1,2-dich'orobenzene
   hexachiorobenzene
   nitrobenzene
   1,2,4 trichlorobenzene
   4-chlorophenyl phenyl ether
Acidics * *
   c-chlorophenol
   2-nitrophenol
   2,4-rjichlorophenol
   phenol
   pentacHlorophenol
   4-nitrophenol
   2,4,6-trichlorophenol
Pesticides**
   aldrin
   4,4' DDT
   dieldnn
   endrin
   heptachlor
   chlordane
                                              Precision
                                              frelative
                                              standard
                                             deviation)
12,0
13.1
19.1
12,8
14.0
13.0
17.5
12.5
12.6
 7.0

20.0
22.0
22.0
18.0
20.0
33.0
13.5
42.0
21.0
22.0
 5.0

22.0
20.0
20.0
19.0
31.0
20,0
26.0

 8,0
29.0
 2.5
11.0
11.0
 5.0
              Accuracy
               (as bias)
 29
 -6
  5
 -5
  6
-20
 10
 -7
 -4
 -5

 -7
-66
 11
 20
-14
-16
 -9
 50
 38
-26
  10
  5
  5
-39
  14
 -5
  10

-16
-23
 "™»J
 -6
-41
-20
           Practical
           detection
             limit
10
10
10
10
10
10
10
10
10
10

10
10
10*
10
10
10
10
10
10
10

25
25
25
25
25
25
25

10
10
10
10
10
10
                                                                                          Reference
143,156,157
143,156,167
143,156,157
143,156,157
 *The date represent average minimum levels of precision and accuracy obtainable by the referenced laboratories
  using 500 series Priority Pollutant Methods, Detection limits are those estimated [eveis routinely achieved by
  several laboratories for aqueous standards in each class. These s/alues will vary considerably depending on the
  analyst ano whether CC/MS, GC" /MS, or chromatographic methods with selective detectors are used.
 **lnterlaboratory values varied over an extremely wide range (greater than one order of magnitude), particularly
  for methods 624 and 625.
                                                   56

-------
standards carried through the procedure. The. laboratory staff cannot be expected to
account for sampling'or materials-related errors in the analytical .results.  '
      For the trace contaminants in Tables 7-11 and'7-12 determined at the 10 ppb
level, one may expect that the corrected analytical.results will be within 20 percent of
the actual value in the absence of bias or unknown interferences. It is the responsibility
of the monitoring program director to inform the lab of any known interferences as
well' as sampling-procedures and materials'  contacted which may bias the analysis of
ground-water samples.  For example, it  has been shown by several 'groups that solvent
cements  used  in  the construction of PVC  well casing  and bailers contain  chemical
constituents which  directly  interfere with  the  chromatography of several , priority
pollutants (41,79,' 116). Such'  interferences  may persist  for many months, even after
proper development of solvent-cemented PVC monitoring wells.  It may be possible to
assess analytical problems by saving a sample of the PVC primer or solvent cement for
complete analysis. However, the composition of these products has been shown to .vary
significantly from that  noted on  the label (41).
      There are many'Other sampling and materials-related problems which can bias
the analytical data from monitoring programs. Several decades of study on the effects
of such  artifacts have resulted in significant improvement  hr the quality of chemical
data.  This process has  taken place in fields as diverse as oceanography..(117,158). and
lunar geochemistry  (159), where the effects of "conventional"  sampling gear or an
inadvertant  fingerprint have  led to  substantial confusion in the scientific literature.
Ground-water -monitoring programs can benefit from  this body of experience, even
though the aims of  such efforts are  more practical.
                                        57

-------
                                    SECTION 8
                             COST CONSIDERATIONS
       Once the goals of a ground-water monitoring program have been identified, an
' estimate must be made of the costs that will be incurred in all the various tasks of data.
 collection.  Preliminary  screening  of existing data and hydrologic and  geophysical
 surveys are  valuable planning, tools which provide  estimates'of the type,  extent, and
 costs of data collection. Professional consultation  can supplement this planning by
 identifying potential  options for  altering  network design in, the event  of specific
 contingencies. As a rule, the degree of detailed information to  be collected is roughly
 proportional to the cost of the effort to obtain it. Both the minimum quantity and
 quality of data, as well  as  the  maximum, must  be  incorporated into the monitoring
 plan. Since  there  are 'many unknowns involved in ground-water investigations, the
 maximum cost/benefit condition may be achieved by designing the initial  plan so that
 minimum  information needs will be met and at least part of the network design will
 serve future  needs. Careful consideration of data reporting and standards for high-
 quality data are available to aid this effort (160).
       Water chemistry data collection entails the siting and construction of wells and
 sampling points followed by the collection, analysis and evaluation of the analytical
 data. The analytes of interest, necessary sensitivity, precision and accuracy at specified
 limits of quantitation must be identified in the,  planning stages. From this point, the
 particulars of materials' selection, sampling protocol development and  prospective
 interpretive' techniques can be planned 'to  insure that high-quality  information  is
 collected in  a cost-effective manner.
       The preceding chapter dealt  with the selection of appropriate material for a
 range of ground-water conditions. Also, sources of potential bias or inaccuracy which
 can result from the use of  inappropriate materials in 'contact with water samples were
 identified. The  choice of well casing (or sampling device) materials is a key element in
 planning a ground-wafer monitoring effort for two  reasons.  First, all of the water
 samples pass through  this point. Secondly, once the initial sampling points of a network
 are constructed, it will require  significant cost and effort to replace them should their
 emplacement or interactions with  the subsurface render them useless. An additional
 caution  in material selection is that bias or  inaccuracy  may be  detected too late  to
 salvage even the minimum  information required for program success. All of the other
 materials which contact the samples can be independently evaluated for potential bias
 and replaced if necessary  while the program is in motion. Fortunately, the case for
 choosing appropriate well casing materials  can be  made on a reasonable cost/benefit
 basis.
       The following discussion details the relative costs of essential elements of general
 ground-water monitoring efforts,  including well drilling and construction, sampling,
 and analysis. Relative costs are presented for four  suitable  combinations of well
 construction  materials over a  range of analytical  complexity.  It is presented as an
                                         58

-------
example of potential  planning  scenarios for a successful, cost-effective monitoring
program. A very detailed consideration of the costs involved in monitoring efforts has
been developed by Everett et al. (37).


Description of an  Example Monitoring Effort
      For purposes of discussion, assume that a network of one upgradient and 'three
downgradient, 2-inch,  50-foot wells is to be constructed at a waste disposal site. The
veils are. to be completed at 48-foot depths with 2 feet of screen and approximately
2 feet of casing above ground surface.  The screened  interval was chosen from the
results of a hydrogeologic and geophysical survey  Contract drilling/well construction
services will be used. Sampling is to be done by salaried personnel, and the purchase
of some equipment for sample collection and field determinations will be necessary.
Since very little chemical information is available on the wastes involved, the site, or
conditions, a range of  analytical schemes must be considered. The principal questions
in the mind  of the monitoring  network design staff are:  What are the background
chemical conditions present in ground  water upgradient from the site? How do we
distinguish background chemistry from chemical species which may be  contributed by
the  waste? How do we choose well construction materials so that we can obtain
representative samples of ground  water if both inorganic and organic contaminants
may be  found? The decision is made that the initial network design and installation of
sampling points will be operated for at least 5 years unless the results of the first year's
sampling indicate that  additional wells or an expansion  of the design is  necessary. The
cost-effective choke of well casing materials needs to be made  in order to insure the
quality of the analytical information and to build a degree of flexibility  into the 5-year
program.
                                                      i

Well Installation and Sampling Costs
      Consideration of the optimum requirements for reliable techniques and materials
for well  construction or sampling presented in the preceding sections limits the spectrum
of drilling techniques  and materials' selection.  Specific choices among the possible
combinations are likely to be very situation-dependent.  Therefore, the cost detail for
key elements of monitoring efforts will be somewhat general.
      Three types of  well casing/screen are to be considered; threaded rigid  PVC
(unplaslicized), 304 stainless steel (SS), and Teflon®. The costs involved in using paired
PVC/SS wells  are developed to enable  comparison with Teflon®.  Pairs of PVC/SS
wells appear to be a reasonable alternative when a range of inorganic and organic
contaminants are of interest under conditions where either PVC or SS may be attacked
or otherwise introduce bias  into subsequent analytical determinations.  These three
material combinations are recommended for well casing materials selection in planning
a monitoring effort.
      Table 8-1 contains a breakdown of approximate well construction and sampling
costs for the first year's operation of a monitoring effort. Drilling costs were estimated
in 1983 dollars for the mobilization and operation of a hollow-stem-auger drill rig and
for  the  installation of four  wells. The costs for  well  construction  materials  were
developed  for manufactured casings/screens currently available from commercial


                                       59

-------
              Table 8-1. Cost Estimates for Drilling, Well-Construction, and Sampling

                                    (In 1983 dollars)
Drilling
Mobilization (within 150 miles}
Hourly Fee
  (4 50-ft wells with 2" casing at 10 ffhf1
  plus formation sampling)
Well Construction Materials
Well casings and screen (2 ft)
Fittings, Well Protectors'**
Cement, Sand, Bentonite
           PVC
            700
            640
            460
Subtotal   S1800
             SS
            2000
             640
             460
            3100
            Subtotal

            PVC/SS
             7900*
              640
              460
             9000
Sampling***
Quarterly (10 man-days-yf1 ; at SlS.OOO-yr"1 )
Staff
Supplies
Equipment (pH and conductivity meter, electrodes,
  samplers and filtration apparatus!
First year costs for drilling, well construction
  materials and sampling
 PVC
$9850
   ss
11,150
Subtotal

PVC/SS
 17,050
            1st Year
            $  700
             3000

            $3700

            Teflon®
             8700
              640
              460
             9800
                                             $  600
                                               300
                                             3450

                                             $4350

                                             Teflon®
                                             17,850
  'includes cost of sxira drilling necessary for PVC/SS pairs
 "optional for SS wells
'"'field determinations of pH, conductance, and alkalinity are assumed to be included in sampling operations
suppliers. The added cost involved in drilling pairs of wells for the PVC/SS option
was  included in  the materials cost for purposes.of comparison. Sampling costs were
developed by including staff, supplies and equipment  outlays for the first year. This
procedure lumps together capital, personnel, and operating cost categories for purposes
of comparison. Details of financing arrangements and calculation of the actual cost of
a monitoring program  must  be made by the individuals involved  on a  case-specific
basis. It is clear from the data in the table that the first-year  well construction and
sampling costs for a PVC well network versus one of stainless steel differ  by less than
15 percent. For a program initially aimed at assessing background chemical conditions,
the added cost of installing stainless steel wells may be warranted to provide a degree
of flexibility in future operation should  organic contaminants  later be encountered.
The paired PVC/SS and Teflon® monitoring arrays are significantly more expensive
to construct and sample but are cost-effective choices in chemically aggressive subsurface
situations.
      A graphical comparison of the  capital costs for drilling and well construction as
a function of depth is shown in  Figure  8-1.  The figure projects a  linear increase in
costs correlated with the depth of monitoring interests  and demonstrates the similarity
in capital costs between the PVC, SS and the PVC/SS or Teflon®  options. An important
note here is  that  considerable cost savings may accrue  from casing the upper zones of
the well bore with a less expensive  material  and  using a more appropriate material
below static water levels.
                                          60

-------
                CO
                co
                O
                o
                0.
                <
                o
                               100
   200        300
DEPTH, feet
400
               Figurg 8-1, Capital costs for drilling and well construction of four point array
Analytical Costs
      The minimum degree of analytical detail necessary to adequately meet the needs
of ground-water monitoring programs varies significantly according to program goals.
In regulatory monitoring efforts, the prescribed parameters may be mandated and the
successful performance of the monitoring effort may be little more than an exercise.
There are pitfalls in  this minimal approach. For example, natural variations in ground-
water chemistry may not be evident from quarterly determinations of "contaminant-
indicator"  parameters  (e.g., RCRA analytical parameters, pH, specific conductance,
Total Organic  Carbon,  and Total Organic .Halogen).  If natural variability in the
background electrolyte chemistry is high, it may prove quite difficult to distinguish
natural versus contaminant-release fluctuations. A minimal interpretive analytical scheme
should include indicator parameters as well as a total-dissolved-mineral analysis in order
                                        61

-------
to provide a basis for future work as well as checks on sample handling and analytical
procedures.  This approach also enables close analytical quality control via mass and
charge balances for  the  evaluation of "representative sample" information, Further,
more detailed interpretive  monitoring schemes eventually may  be called for at the site
of interest. Well construction techniques  and materials should be chosen to permit
expansion of the analytical scheme, as well as monitoring network goals,  should the
need arise. After background conditions have been established in the early stages, the
goals can be extended and specific parameters of interest can be added to serve the
more  demanding needs  of detailed  interpretation. Well-conceived network planning,
design, and  operation will support  the detailed information  needs  which may  be
encountered in the future.
       In  Table 8-2, four sample analytical, schemes and corresponding derived param-
eters are listed. Average prices have been included from the price quotes of three high-
volume analytical laboratories based  on a minimum often samples (including replicate,
field blanks, and field standards) per  submission. One can see that as  the  number of
individual analytes is increased, the  cost  per sample increases  proportionately for
detailed inorganic  and organic analysis.
       After  the degree  of analytical  detail appropriate for the  program has been
chosen, a sampling  frequency must be determined.  Also,  the  number of replicate
samples must be decided upon in order to provide the necessary precision for evaluation
procedures. The sample monitoring program has been designed for quarterly sampling
frequency using the four  analytical schemes shown in Table  8-2.  At  least duplicate
                                         **                   *-
  Table 8-2, Description of Analytes and Costs for Four Analytical Schemes for Ground-Water Samples
                                                       Dissolved
                                                     (field-filtered'
                                       Tots/           snd preserved)          Cost per samplet
1. Minimal Detective '
  pH.rr1, roc, TOX                      *                                      ?s.oo

2, Minimal  interpretive                                                          175.00
  pH.rr1, TOC, TOX     =        '
  Alk, Cl~,  NOs", S04= ,  P04",  SiOz                           *
  Na+, K4, Ca++, Mg++,  Fe, Mn

3. Inorganic Interpretive                           '                               410.00
  pH, Q'1, TOC, TOX_
  Aik, CI",  NOs"; S04-,P04=, Si02, B
  Ma"1", K+, Ca++, Mg++,  Fe, Mn                                *
  Fe(ll), Zn, Cd, Cu, Pb, O, Ni               •  »
  Ag, Hg, As, Sb, Se, Be, Ti                                   *

4. Derailed Interpretive                                                         1400,00
  AD of the above (3) PLUS
  Volatiles     {Method 624)               *
  Base-Neutrals (Method 625)
  Acidics       (Method 625)
  Pesticides    (Method 808)               *
tCosts determined in 1983- U.S. dollars from the average of three high-volume
analytical services using EPA approved methodology.
A minimum of ten samples per submission is assumed.

                                         62

-------
samples of ground water from each well are  to  be collected, or  one  may elect to
concentrate more replicates  on the upgradient and one particular downgradient well.
       Five-year total costs are detailed in Table 8-3 for several combinations of analytical
schemes and sample  replication. The number of samples per year includes at least one
field blank and field  standards per sampling date at the stated level of replication. The
5-year totals  for analytical costs have been calculated at a 5 percent  annual increase in
the cost of these services.  In comparison with the  first-year drilling, well construction
and sampling costs presented in Table  8-1,  it is evident  that analytical costs  (even for
a minima! monitoring effort) make up a large share of total project costs.  Of particular
note is the fact that in a detailed interpretive network (3A, 3B, 4A, 4B), the difference
between choosing PVC over stainless steel well casing material (which is less likely to
introduce bias into the analytical data) results in a cost "savings" of $1300. A comparison
of reanalysis costs with apparent savings  by the choice  of a cheaper material  (PVC)
over  stainless steel are shown  in Table 8-4. One  can  readily  see  that  the  expected
"savings"  realized from  the use of an inappropriate,  cheaper material are actually
penalties.
      Ana/yiica/ scheme      '
1.  Minima! Detective
   (4 replicates of each sample)
2.  Minimal Interpretive-      '
   (Duplicates of each sample).
3.  A - inorganic Interpretive
   (4 replicates in 1st year;
   duplicates in years 2-5)
3.  B - Inorganic Interpretive
   (Duplicates of each sample)
4.  A -Detailed Interpretive
   (4 replicates in 1st year;
   duplicates in years 2-5)
4.  B- Detailed Interpretive   '
   (Duplicates of each sample)
  Table 8-3. Analytical Cost Detail*

            Samples' yr~l
            .     80

                 48

               80/48


                 48

               80/48


                 48
                                                                         Cost (Si
           1 Year

           4,800

           8,400
          32,800


          19,680

         112,000


          67,200
                                                                                  5 Years"

                                                                                   26,523

                                                                                   46,415
                                                                                  121,864


                                                                                  108,744

                                                                                  416,122


                                                                                  371,322
 "Quarterly sampling of 4 wells; unless otherwise specified, 4 replicates plus field blanks and
  field standards in duplicate for each analytical workup. Field determinations: pH, ft* and
  alkaniinitv included in sampling costs.
""Calculated at a 5% annual increase in cost,
              Table 8-4. Comparison of Reanalysis Cost with Cost "Savings" on Materials
         Analytical
          scheme

          1
          2
          3A
          3B
          4 A
          4B
 Single sample
reanalysis cost
    JS)

     450
    1050
    2460
    1640
    8400
    5600
Materials t
"ssvings"
   (S)

 1300
 1300
 1300
 1300
 1300
 1300
                                                                             Actual
                                                                            "savings"
                                                                              (S)

                                                                               850
                                                                               250
                                                                             -1160
                                                                              -340
                                                                             -7100
                                                                             -4300
         TDifference between PVC versus stainless steel well construction materials
                                            63

-------
 Project Cost Comparisons for Selected Materials' Combinations
 for Networks of Varying Analytical Detail
       The example  monitoring network has been developed as a general guide for
 planning network design. Though the emphasis has been on well casing materials, the
 methodology applies as well to the choice of materials for sampling gear or tubing. To
 put the costs of total project  implementation and well casing material in perspective,
 it is a relatively simple matter to repeat the process for a specific case to complete the
 cost/benefit analysis.
       In  Table  8-5,  the approximate  total  project  costs  for the sample analytical
 schemes and various well casing materials are presented. It is evident that the incremental
 increases for more idea! (and costly) materials is far less than the increases incurred in
 satisfying  more involved  analytical needs,  On the basis  of initial material costs, the
 most  expensive  materials options, paired  PVC/SS  and Teflon®,  make up a  small
 percentage of total project costs. One must  carefully weigh the value of the information
 needed versus  the  "insurance" value of making appropriate choices of materials which
 may contact the samples. For  long, term  five-year project operations, stainless  steel
 clearly is the material of choice for minimal analytical schemes.
       The selection of materials  for pumps and tubing are important to both the
 reliability and  the  cost/benefit  considerations of  ground-water monitoring  programs.
 They are somewhat less  critical than materials for well  construction since there are
 few significant additional  costs (e.g., drilling, mobilization, etc.) that must be included
 as well, except in  the  case of dedicated samplers. The cost comparison can be  made
 in a similar fashion to  that for well drilling and construction.

                        Table 8-5.  Total Project  Costs for Monitoring Programs
                           (S1.0K 1983 dollars  rounded to nearest S0.5K)
                            including the Percentage of Materials Costs
                                     1st Year                           5 Years t
        Mater/3/s
Analytical Scheme
1,    Minimal Detective
     % Increase over PVC
2.    Minimal Interpretive
     % Increase over PVC
3A.  inorganic Interpretive
     % increase over PVC
3B.  inorganic Interpretive
     % Increase over PVC
4A.  Detailed Interpretive
     % Increase over SS
4B.  Detailed Interpretive
     % Increase over SS
T5 year project costs  calculated using yearly totals, including: 1% annual increases per year for maintenance on well installations
 and 5% annual increases per year for supplies, sampling staff time and  analytical services. IPVC/SS installations require twice
 the maintenance effort and therefore exceed aM Teflon©costs after 5  years.)
MR _ Material is not  recommended for use with a detailed.organic analytical  scheme due to the likelihood of analytical bias
 and imprecision,
PVC
14.5
18.0
42.5
30.0
122,0
MR
77.0
NR
SS
16.0
10
19.5
8
44.0
3
31.0
3
123.0
78.0
PVC/SS
22.0
38
25.5
42
50.0
17
37.0
23
129,0
5
84.0
8
Teflon®
22.5
i5
26.0
44
50.5
18
37.5
25
130.0
6
85.0
9
PVC
42.0
62.0
137.5
124.0
431.5
NR
387.0
MR
SS
43.5
4
• 63.0
2
139.0
1
125.0
1
433,0
<1
388.0
<1
PVC/SS
51.5
23
73,0
18
147.0
7
134.0
8
441.0
2
396
2
Teflon©
50.5
20
70.0
13
146.0
6
132,5
7
440.0
2
395.0
2
                                           64'

-------
Tailoring the Monitoring Well and Sampling Apparatus
to the Anticipated Analytical Scheme
      The successful planning of a monitoring program demands the consideration of
several factors: the ultimate goal of the monitoring program; the hydrologic conditions
of the site to be monitored; the sources of pollution and their chemical nature; the
quality of the ground water to be protected; the levels at which selected indicator and
specific monitoring parameters are to be analyzed; and appropriate cost-effective choices
of materials, drilling or sampling techniques.
      Given these factors,  preliminary decisions can be made concerning the location,
depth, physical features, and materials selection for the monitoring wells. Appropriate
drilling  and  well development techniques also  must be selected. It is important to
maintain flexibility  in  this  preliminary planning stage because the initial monitoring
plan may require revision  as new information is  collected during well construction.
Preliminary planning must  be undertaken after consideration of the complexity of the
analytical scheme.
      The objective in siting a  monitoring well is to place the screened interval in the
predicted flow path of contaminant migration from  the monitored site.  Optimum
placement should allow for  early detection of contamination to prevent a crisis situation
caused by the undetected migration of pollutants. Well construction  materials should
be selected to minimize bias and interference with anticipated pollutants at the specified
levels of analysis. Careful attention should be given to possible expansion of the original
information needs of the monitoring effort.
      The selection of sampling apparatus also  should  be. based on the proposed
analytes of interest and planned (as well as achievable) levels of detection. Sampling
devices should be selected so that neither their component materials nor their pumping
operation will significantly  alter  solution chemistry.
      In the case of detective monitoring, consideration should be given lo the use of
the monitoring wells and sampling apparatus in a more rigorous interpretive monitoring
program. Upgrading wells and sampling devices in the early phases of monitoring may
save considerable time and expense  later.  Trying to save money by compromising on
materials quality or suitability in  detective monitoring may eventually increase program
cost by  causing reanalysis  or triggering  unwarranted interpretive monitoring.  The
choice of appropriate  robust materials and proven methods can provide significant
long-term benefits to a well-conceived  program.
      In summary,  each monitoring program must be approached on the basis of its
own information needs and conditions. There are no standard monitoring approaches
that prove satisfactory for all sites. The  multiple  goals of monitoring projects, the
heterogeneity of geologic materials, temporal and spatial variability  in chemistry and
hydrology, and the wide variety of chemicals to be monitored provide an  unlimited
number of unique conditions that call for different  approaches to monitoring. All of
these factors must be examined to determine the impact they may have on the reliability
of the analytical data as well as the long-term  costs and benefits involved in obtaining
this information.
                                       65

-------
                                  SECTION 9
                                CONCLUSIONS
      Ground-water monitoring is an important part of a reasoned resource manage-
ment and protection strategy. Monitoring programs include the modeling, planning,
analysis and interpretation of information on the subsurface environment of ground
water. These programs can  be  expensive as well  as time consuming. However, with
careful planning of field efforts, the  yield of high quality information will prove to be
invaluable to our understanding of the dynamics of ground-water systems,
      Critical considerations for the design of monitoring networks are the selection
of drilling and sampling techniques  in addition to the choice of materials which will
contact groundwater samples. With the knowledge of the principal chemical constituents
of interest, locai  hydrogeology, and an appreciation of subsurface geochemistry, appro-
priate selections of materials, drilling, and sampling techniques can be made. Whenever
possible, physical disturbance and the amount of foreign material introduced into the
subsurface should be minimized.
      The choices of drilling methods and materials for both well casing and sampling
apparatus are very  important decisions to be made in 'every type of ground-water
monitoring program. Details of network construction  can  introduce significant  bias
into monitoring data which frequently may be corrected only by repeating the process
of well  siting, installation, completion, and development. This can be quite costly in
time, effort, money, and loss of information. Undue expense is avoidable if planning
decisions are made cautiously with an eye to the future.
      Sampling  techniques similarly should be tailored to the information needs and
goals of the monitoring efforts.  If appropriate network design decisions are made, the
effect of sampling errors can be corrected before major 'decisions on expansion of the
monitoring goals are necessary.
      The expanding scientific literature  on subsurface phenomena and effective
ground-water monitoring techniques should be read and evaluated on a continuing
basis. This information alone will supplement published guidelines and suggestions for
application to specific monitoring efforts in the future.
                                       66

-------
                                  SECTION 10
                              RECOMMENDATIONS
General Recommendations
      The Agency should focus its aid and technical assistance to state ground-water
management and protection efforts so as  to encourage a degree of uniformity  and
consistency in monitoring efforts. A central element of federal policy should include a
mechanism for the retrieval and interpretation of monitoring information which includes
detailed site-specific data,on network design, construction, and operation to minimize
the continued use of unreliable  techniques or incompatible materials which result in
useless data collection. Of necessity, this must be conducted to  protect source confi-
dentiality in some instances, but this requirement should not prove to be an impediment
to effective technology transfer.
      Similarly, the claims of manufacturers or ground-water professionals concerning
the -integrity of materials, the reliability of specific designs or techniques,  and the
performance of sampling equipment should be very carefully evaluated on a case by
case basis. The concerted attention of design, technical, and analytical staff should be
trained on each aspect of a ground-water monitoring effort before the implementation
of the network.
Specific Recommendations
      Non-contaminating drilling methods are available to provide access holes to the
geologic formation of interest which need not be overly expensive or expose workers
to undue hazard even in badly contaminated situations. Well casings and screens should
be selected with the understanding that analytical requirements demand a* durable,
stable sampling point which can be  relied  upon to maintain the integrity of the in-situ
ground-water  condition. Well  casing and screen materials  in  decreasing  order of
preference for most  monitoring situations are:
         Teflon®, stainless steel, and rigid threaded PVC

      Commercially manufactured casing products are recommended over home-made
varieties. The choice of inappropriate, less expensive well casing/screen materials is
rarely cost-effective even in minimal" monitoring efforts with  limited analytical detail.
Future contingencies and the cost  of reanalysis can convert  apparent "savings" into
real  costs with a corresponding penalty in lost  time, effort, and  information. The
convenience of using PVC well casing or screens must be carefully  considered in  this
regard since the potential for biased analytical data clearly exists in  detailed analytical
schemes.
      The methods and materials involved in ground-water sampling are equally critical
to the collection of high  quality  monitoring 'information,  Overreliance  on simple,
traditional  collection mechanisms  (e.g.,  bailing) entails  a serious  potential  for  the
                                       67

-------
 continued collection of unreliable, poor data particularly where the analysis of volatile,
 pH-sensitive, or reduced chemical constituents is of interest. Integral sampling methods
 which minimize turbulence, atmospheric content, gas exchange, and depressurization
 are  preferable for these applications. Materials  which contact water samples during
 collection are no  less critical than storage vessels.  Recommended materials for pump
 parts, tubing, and associated apparatus in 'decreasing order of preference are;
          Teflon®,  stainless  steel, polypropylene,  polyethylene, linear polyethylene,
          Viton®, conventional polyethylene, PVC

       Though other  materials appear promising from  the  standpoint of structural
 integrity and chemical resistance, there are very few hard  data on which to recommend
 their use  in ground-water monitoring.
       Ground-water monitoring is more complex and challenging than the collection
 of reliable data-in natural surface waters.  The lessons of past monitoring efforts clearly
 demonstrate the need  for multi-disciplinary inputs to planning ground-water investi-
 gations. The input of both chemical professionals and laboratory personnel is essential
 to a successful program.
       The wise monitoring program director should  attempt  to consider carefully all
 existing information on local well drilling practices, hydrogeology, and the  potential
'impact of waste constituents on subsurface geochemistry  prior to implementation of a
 ground-water monitoring plan. In  this way, maximal benefits will accrue from the
 considerable outlay of funds,  time,  and effort involved in subsurface  monitoring
 activities. The most important result may  be-that in the future1 we will  be in  a far
 better position to effectively manage and protect our  ground-water supplies.
                                        68

-------
                                  REFERENCES
 l/'Scalf, M, R., J. F. McNabb, W. J, Dunlap, R. L, Cosby, and J. Fryberger, 1981. Manual
    of Ground-Water Quality Sampling Procedures. National Water Well Association, Wbr-
    thington, Ohio.
 2. Middleton, F, M.  1973. Organics  in Wraler Supply — The Problem, Proceedings, Fifteenth
 '   Water Quality Conference, University of Illinois, Champaign, Illinois.
 3. Seanor, A..M., and-L. K. Brannaka. 1981.  Influence of Sampling Techniques on Organic
    Water Quality  Analyses, Proceedings of the NationaL Conference on Management of
    Uncontrolled' Hazardous Waste  Sites, HMCR1,  9300  Columbia  Blvd.,  Silver Spring,
    Maryland, October.                  ^          .                      <\
 4, Pettyjohn, W. A., and A. W. Hounslow,  1982. Organic Compounds and  Ground-Water
    Pollution.  Proceedings,  Second Annual Symposium on Aquifer Restoration  and Ground
    Water Monitoring, National W;ater Well Association, Worthington, Ohio.
 5. National Council of the Paper Industry for Air and Stream Improvement:  1982. A Guide
    to Groundwater  Sampling. Technical  Bulletin 362, NCASI, 260'Madison Avenue, New
    York, New York.
 6. U.S.  Environmental  Protection Agency.  1982. -Ground-Water Monitoring Guidance for
    Owners and Operators  of Interim  Status Facilities.  EPA Report SW-963,  Office of Solid
    Waste and Emergency Response, Washington, D.C.
 7. Clark, T.  P., and G.i V. Sabel.  1980. Requirements  of State Regulatory Agencies for
    Monitoring Ground-Water  Quality at Waste Disposal Sites. Ground W'ater, 18, 2.
 8. Geraghty & Miller, Inc. 1978. Ground-Water Monitoring, EPA Contract 68-01-3703, Final
    Project Report.
 9. Grisak, G. E., R. E.Jackson, and J. F. Pickens. 1978.  Monitoring Groundwater Quality;
    The  Technical Difficulties.  Water Resources Bulletin, June.
10. Todd, D. K., R.  M.,Tinlin,  K. D. Schmidt, and L. G. Everett. 1976. Monitoring Ground-'
   ' Water Quality: Monitoring Metbodology.'"EPA-600/4-76-026, USEPA,.Las Vegas, Nevada.,
                     >
11. Fenn, D.,  E. Cocozza, J. Isbister,  O.  Braids, B. Yare, and  P. Roux.  1977. Procedures
    Manual  for Ground  Water Monitoring at  Solid  Wraste Disposal  Facilities, EPA/530/
    SW7611,  USEPA, Cincinnati, Ohio.
12. Tinliri, R. M., ed. 1976."Monitoring  Groundwater Quality:  Illustrative Examples.  EPA
    600/4-76-036,  USEPA,  Environmental  Monitoring and Support Laboratory, Office of
    Research and Development, Las  Vegas, Nevada,.
13% Lewis, R.  W. 1982. Custom Designing of Monitoring Wells for Specific  Pollutants and
    Hydrogeologic Conditions-. Second Annual Symposium on Aquifer Restoration and Ground
    Water Monitoring, National W-'ater Well Association, Worthington, Ohio.
14. Dunlap,  W. J., J.  F. McNabb, M.  R. Scalf, and R. L, Cosby. '1977. Sampling for Organic
    Chemicals and Microorganisms in the  Subsurface. Office of Research and Development,
    USEPA,  Robert S. Kerr Environmental Research  Laboratory, Ada, Oklahoma,^
15. Sisk,  S. W, 1981.  NEIC Manual for Groundwater/Subsurface Investigations at Hazardous
    Waste Sites. USEPA Office of Enforcement, National Enforcement Investigations-Center,
    Denver, Colorado.
                                         69

-------
 16. National  Water Well Association.  1976.  Manual of Water  Well Construction Practices.
  . '  EPA-570/9-75-001, USEPA, Office of Water Supply, Washington, D.C.
 17. Pettyjohn, W. A., W. J, Dunlap, R. L. Cosby, and J. W. Keeley. 1981. Sampling Ground
     Wafer for Organic  Contaminants.  Ground Water, 19, 2,-pp. 180-189,
 18. National  Water Well Association and the Plastics  Pipe Institute.  1980.  Manual on the
     Selection and Installation, of Thermoplastic Water Well Casing.
 •19. Minning,  R.  C.  1982, Monitoring Well Design and  Installation. Proceedings, ^ Second
     Annual Symposium on Aquifer Restoration and Ground  Water  Monitoring, National
     Water Well Association, Worthington, Ohio.
 20. Luhdorff, E. £., Jr., andj. C, Scalmanini.  1982. Selection of Drilling Method, Well Design
     and Sampling Equipment for  Wells to Monitor Organic Contamination. Proceedings,
     Second Annual- Symposium  on Aquifer  Restoration and  Ground  Water  Monitoring,
     National  Water Well Association, Worthington, Ohio.             .               <  •
 21.''U.S. Geological Survey. 1977. National Handbook of Recommended Methods for Water-
     Data Acquisition. USGS Office of  Water Data Coordination, Restpn, Virginia.
' 22. Gibb,  J.  P., R.  M.  Schuller, and R.  A. Griffin. 1981. Procedures for the Collection of
     Representative  Water Quality  Data  from Monitoring Wells.  Cooperative Groundwater
     Report 7, Illinois State Water  Survey and Illinois  State Geological Survey,  Champaign,
     Illinois.                    i
 23. Grant, D. M.  1982,  Right Place,  Time,'and Technique: Right Sample. WATER/Engi-
     neering and Management, June.               >t             "•              ,          •
 24. Wood, W. W.  1976. Guidelines  for  Collection and Field Analysis of Groundwater Samples
     for Selected Unstable Constituents.  U.S. Geological Survey Techniques for Water Resources
     Investigations, Book 1, Chapter D-2.
 25. Schmidt,  K.  D.  1982. How Representative 'are Water  Samples Collected  from .Weils?
     Proceedings,  Second  Annual Symposium on Aquifer Restoration .and Ground Water
     Monitoring, National Water Well Association, Worthington, Ohio.
 26. Carriere, G. D., and L. W. Canter.  1980. Planning for Ground Water Quality Monitoring.
     National  Center for Ground  Water Research, Norman, Oklahoma.
 27. King,, W. G., J, M.  Rodriguez,  and C. M. Wai. 1974. Losses of Trace Concentrations of
     Cadmium from Aqueous  Solution  During Storage in Glass Containers. Analytical Chem-
     istry, 46,  6, pp. 771-773.
 28. Struempler, A.  W.  1973.  Adsorption Characteristics of Silver,  Lead, Cadmium, Zinc and
     Nickel on Borosilicate Glass,  Polyetheylene and Polypropylene Container Surfaces. Ana-
     lytical Chemistry, 45, 13"!'
 29. Latterell, J. J.,  D. R. Timmons, R. F.; Holt,' and E. Sherstad. 1974. Sorption of Ortho-
     phosphate on the Surface of Water Sample Containers. Water  Resources Research, 10, 4,
     pp. 865-86,9.
 30. Coyne, R. V., and J.  A.  Collins.  1971.  Loss of Mercury from Water  During Storage.
     Environmental  Health Lab, McClellan AFB, California (NTIS AD-757 861).
 31.. Brass, H.J., M, A.  Feige,  T. Halloran, J. W. Mellow, D. Munch, and R. F. Thomas. 1977.
     The National Organic Monitoring  Survey: Samplings and Analyses  for Purgeable Organic
     Compounds. In Drinking  Water Quality Enhancement through Source Protection (R. B.
     Pojasek,  ed.), Ann  Arbor Science  Publishers, Ann  Arbor, Michigan.'
 32.. Goerlitz, D. F., and 'E. Brown. 1972. Methods for Analysis of Organic Substances in
    • Water. Chapter A3, Techniques of .Water-Resources Investigations, Book 5,- Laboratory
   n Analysis, .U.S. Geological  Survey.        x   '                 '             I
                                                               i
                                         •  70       .  "    •

-------
33. Gass, T, E., T. W, Bennett, J. D. Miller, and R. A. Miller. 1982. Manual of Well Maintenance
    and Rehabilitation Technology.  EPA report published by National Water Well Association,
    Worthington, Ohio.
34. McMillion, L. G.,  and J.  W.  Keeley.  1968, Sampling  Equipment  for Ground Water
    Investigations. Ground Water, 6, 2.
35. Buss, D. E, and  K. E. Bandt. 1981.  An All-Teflon® Bailer and an Air-Driven Pump for
    Evacuating Small-Diameter Ground-Water Wells. Ground Water, 19, 4.
36. Tomson, M. B.,  S.  Hutchins, J. M. King, and C. H. Ward. 1980.  A Nitrogen-Powered,
    Continuous Delivery, All-Glass-Teflon® Pumping System for Ground Water Sampling from
    Below 10  Meters. Ground  Water, 18, 5.
37, Everett, L.  G.,  K.  D. Schmidt,  R. M., Tinlin, and  D,  K.  Todd.  1976.  Monitoring
    Groundwater Quality: Methods and Costs. EPA-600/4-76-023, USEPA, Environmental
    Monitoring and Support Lab, Office of Research and Development, Las Vegas, Nevada.
38. Imbrigiotta, T. E., and A.  Martin, Jr. 1981.  Hydrologic and Chemical Evaluation of the
    Ground-Water Resources of Northwest Elkhard  County, Indiana. U.S. Geological Survey,
    Water Resources Investigation 81-53, Indianapolis,  Indiana,'140 p.
39. Miller, G.  D. 1982. Uptake and Release of Lead,  Chromium and Trace  Level Volatile
    Organics Exposed to Synthetic Well Casings.  Proceedings, Second Annual Symposium on
    Aquifer Restoration and Ground  Water Monitoring, National Water Well Association,
    Worthington, Ohio, pp. 236-245.
40. Boettner,  E.  A., G.  L.  Ball, Z.  Hollingsworth, and  R. Romulo. 1981.  Organic  and
    Organotin Compounds Leached from PVC and CPVC Pipe. EPA-600/1-81-062, USEPA,
    Washington, D.C.
41. Sosebee, J. B., P. C. Geisgler, D. L. Winegardner, and C.  R. Fisher. 1982. Contamination
    of Groundwater Samples with  PVC Adhesives and PVC  Primer from Monitor Wells.
    Environmental Science and Engineering, Inc., presented at ASTM Symposium on Haz-
    ardous Solid Waste, Orlando, Florida, January.
42. Uni-Bell Plastic Pipe  Association.  1981. Vinyl Chloride:  The Control of Residual Vinyl
    Chloride Monomer in PVC Water Pipe. Technical Report, Uni-Bell Plastic Pipe Association,
    2655 Villa Creek Drive, Suite 150, Dallas, Texas, 24 p.
43. Dressman, R, C., and E.  F, McFarren. 1978-. Determination of Vinyl  Chloride Migration
    from  Polyvinyl Chloride Pipe Into Water. Journal, American  Water  Works Association,
    January, pp. 29-30.
44. Sachs,  E A.,  and J. D. Banzer.  1974.  Vinyl  Choride — Release from PVC. Technical
    Papers,  Regional Technical Conference, New York  City meeting, October.
45. Mantell, G. J., J. T. Barr,,and  R.  K. S. Chan.  1975. Stripping VCM from PVC Resin.
    Chemical Engineering Progress, September,
46. Berens, A, R., and G. A. Daniels. 1976. Prediction of Vinyl Chloride Monomer Migration
    from Rigid PVC Pipe. Polymer Engineering and Science, August.
47. Daniels, G. A., and  D. E.  Proctor. 1975. VCM Extraction from PVC Bottles. Modern
    Packaging, April.
48. Curran: C. M., and M-. B. Tornson. 1982, Leaching of Trace Organics into Water from
    Five  Common Plastics. Personal Communication from authors to G.  D. Miller, National
    Center for Ground Water Research,  Norman, Oklahoma.
49. Junk, G. A.,  H.  J.  Svec, R. D. Vick, and M. Avery. 1974. Contamination of Water by
    Synthetic Polymer Tubes. Environmental Science and Technology, 8,  11, pp.  1100-1106.
                                          71

-------
50, Massee, R ,  F.  Maessen,  and J.  DeGeoij,  1981. Losses of Silver, Arsenic, Cadmium,
    Selenium and Zinc Traces from Distilled Water and Artificial Sea-Water by  Sorpdon on
    Various Container Surfaces, Analytica Chimica Acta,  127,
51. Shendrikar, A. D., V. Dharmarajan, H. Walker-Merrick, and P. W. West. 1976, Adsorption
    Characteristics of Traces of Barium, Beryllium, Cadmium, Manganese, Lead  and Zinc on
   ' Selected Surfaces. Analytica Chimica Acta,  84.
52. Robertson,  D. E. 1968a.  The.Adsorption  of Trace Elements in Sea Water on Various
    Container Surfaces. Analytica Chimica Acta, 42,
53. Robertson, D, E. 1968b. Rate of Contamination in Trace Element Analysis of Sea Water,
    Analytical Chemistry, 40,  7,
54. Eichholz, C. G., A. E. Nagel, and R, B. Hughes. 1965. Adsorption of Ions in Dilute
    Aqueous Solutions on Glass  and Plastic  Surface. Analytical Chemistry, 37,  7,  pp. 863-
55. Schalia, R., and P. L. Oberlander. 1983. Variation in the Diameter of Monitoring Wells.
    Water Well Journal,  May, pp. 56-57,                                 x>
56. Brown, K. W., J. Green, and j, C. Thomas, 1983, The Influence of Selected Organic
    Liquids on the Prerneability of Clay Liners. Presented at Ninth Annual Research Sym-
    posium, Land Disposal, Incineration, and Treatment of Hazardous Wastes, May 2-4, Ft.
    Mitchell, Kentucky,  USEPA-SHWRD/EPCS sponsored,
57. Richard, M, R. 1979. The Organic Drilling Fluid Controversy: Part 1. Water Well Journal,
    April, pp, 66-74.
58. Richard, M. R,  1979. The Organic Drilling Fluid Controversy:  Part  II, Water Well
    Journal, May, pp. 50-58,
59. Brobst, R, B,, and P. M. Buszka.  1983, Effects of Two Selected Drilling Fluids on Ground
    Water Sample Chemistry. (In preparation) Personal Communication.
60. Yare, B, S. 1975, The Use of a Specialized Drilling and Ground-Water Sampling Technique
    for Delineation of  Hexavalent  Chromium  Contamination  in an  Unconfined Aquifer,
    Southern  New Jersey Coastal Plain. Ground Water, 13, 2, pp. 151-154.
61. Strausberg, S. 1983. Turbidity Interferes with Accuracy in Heavy Metals Concentrations.
    Industrial Wastes, March/April, pp. 20-21.
62, Dunlap, W. J., and  J. F.  McNabb. 1973.  Subsurface Biological Activity in Relation to
    Ground Water Pollution. USEPA Report EPA-660/2-73-014, R. S. Kerr Environmental
    Research  Laboratory, Ada, Oklahoma, September, 60 p.
63. Allen,  M.  J. 1982; Microbiology of Potable Water and Ground Water. Journal, American
    Water Works Association, June, pp. 943-945.
64. Geraghty, J. J. 1967. Ground Water — A Neglected Resource. Journal, AWWA, 59, pp.
    820-828.
                                       i,
65. Kimmel, G. E, and O. C.  Braids.  1980. Leachate Plumes in Ground'Water from Babylon
    and  Islip'Landfills,  Long Island, New York. U.S. Geological Survey Professional Paper
    No.  1085, 38 p.
66. Matthess,  G. 1982. The Properties of Groundwater. Wiley-Interscience Publications, New
    York, New York, 406 p,
67. Drever, J. 1.  1982. The Geochemistry of Natural Waters. Prentice-Hall, Inc., Englewood
    Cliffs, New Jersey, 388 p.
68. Freeze, R- A., and |. A. Cherry.  1979. Groundwater. Prentice-Hall, Inc., Englewood
    Cliffs, New Jersey, 604 p.


                                          72

-------
69, Sturnm, W., and J, J. Morgan. 1981, Aquatic Chemistry, 2nd Edition, Wiley-Interscience,
    New York, 780 p.
70. Seanor, A.  M., and L. K. Brannaka.  1983. Efficient Sampling Techniques. Ground Water
    Age, April, pp. 41-46.
71. Theis, C. V. 1935. The Relation Between the Lowering of the Piezometric Surface and
    the Rate and Duration of Discharge of a Well Using Ground-Water Storage. Transactions,
    American Geophysics Union, 16.
72. Jacob, D. E. 1950. Flow of Ground  Water. Engineering Hydraulics, H. Rouse, ed., John
    Wiley and Sons, New York.
"73. Papadopulos, I. S., and H.  Cooper.  1967, Drawdown in a Well of Large Diameter. Water
    Resources Research, 3, 1.
74, Garvis, D, G., and  D. H. Stuermer,  1980. A Well-Head Instrument for Multi-Parameter
    Measurement During Well  Water Sampling. Water  Research, 14, pp. 1525-1527.
75. Gillhain, R. 1982. Syringe Devices for Ground-Water Sampling.  Ground Water Monitoring
    Review, Spring, pp. 36-39.
76. Singmaster, J. A.,  Ill, and  D. G. Crosby. 1976. Plasticizers as  Interferences in Pollutant
    Analyses. Bulletin,  Environmental Contamination Toxicology, 16, pp. 291-300.
77. Peterson, J. C., and D. H. Freeman. 1982. Phthalate  Ester Concentration Variations in
    Dated  Sediment Cores from the Chesapeake Bay. Environmental Science Technology, 16,
    8, pp.  464-469,
78. Packham,  R. F. 1971, The Leaching of Toxic Stabilizers from Unplasticized  PVC Water
    Pipe,  Part  I — A  critical study of laboratory test procedures. Water Treatment and
    Examination, 20, Part 2, pp.  108-124.
79. Boettner, E. A,, G. L. Ball, Z^Hollingsworth, and R. Aquino. 1982. Organic and Organotin
    Compounds Leached from  PVC and CPVC Pipe, EPA-HERL, Cincinnati,'Ohio, No, CR-
    806275,
80, National Sanitation  Foundation. 1980,  Plastic Piping Components and Related Materials
    (Standard Number 14). National Sanitation Foundation, 3475  Plymouth Road, P.O. Box
    1468, Ann Arbor, Michigan,  8  p. +  Appendices.
81,, Wilson, D.  C,, P. J. Young, B. C, Hudson, and G. Baldwin. 1982, Leaching of Cadmium
    from  Pigrnented Plastics in a Landfill Site. Environmental Science and Technology, 16,
    9, pp.  560-566. -
82. Shuckrow, A. J., A. P. Pajak, and C. J. Touhill. 1980.  Management of  Hazardous Waste
    Leachate.  No. CR-68-03-2766, Solid and Hazardous Waste Research Division, USEPA-
    MERL, Cincinnati, Ohio, NTIS PB-81-189359.
83. Robintech. Inc. PVC Technical Data Bulletin. Robimech, Inc., 2000 Two Tandy Center,
    Fort Worth, Texas, 76102.
84, Sarstedt, Inc. The Resistance of Polystyrene, Polyethylene, and Polypropylene to Chemical
    Compounds. Sarstedt, Inc., P.O. Box 4090, Princeton,  New Jersey, 08540.
85. Precision Chemical Pump Corporation. Chemical Resistance  Chart,  Bulletin 127-14,
    Precision Chemical Pump Corporation, 1396 Main Street, Waltham, Massachusetts, 02154.
86, Plasti-flo, Inc. Corrosion Resistance Chart: Industrial Chemicals. Plasti-flo, Inc., 2701 N.
    Pulaski Road, Chicago, Illinois, 60639.
87. Cole-Parmer Instrument Co, 1983. Masterflex Tubing Compatibility Tables and Plasticware
    Chemical Resistance Chan. Catalog 1983, Cole-Parmer  Instrument Co., 7425 N. Oak
    Park Avenue, Chicago, Illinois,  60648,


                                          73

-------
 88,  Plastic  Pipe Institute,  1973.  Thermoplastic Piping for the Transport of  Chemicals,
     Technical Report PPI-TR19-AUG1973, a Division of the Society of the Plastics Industry,
     Inc., 355 Lexington Avenue, New York, New York, 20 p.
 89.  Schweitzer,  P. A. 1969. Handbook of Corrosion Resistant Piping.  Industrial Press, Inc.,
     New York,  New York, 10016, 358 p.
 90.  Slunder, C.  J. and W. K. Boyd. 1971. Zinc: Its Corrosion Resistance. First Edition, Zinc
     Institute, Inc., 292 Madison Avenue, New York, New York, 10017.
 91.  Cheremisinoff, P, N., I. Fideli, and N, P. Cheremisinoff. 1973. Corrosion Resistance of
     Piping and  Construction Materials, Pollution Engineering, August, pp. 23-26.
 92.  Schindler, P. W., B. Furst,  R.  Dick, and R U. Wolf.  1976- Liquid  Properties of Surface
     Silanol Groups I. Surface Complex Formations with Fe'*, Cua+, Cd2*, and Pb2*. J. Colloid.
     Interf. Sci., 55, 469 p.
 93.  Kummert, R., and W. Stumm, 1980. The Surface Complexation of Organic Acids on
     Hydrous y-~A\2O3, J, Colloid,  Interf. Sci., 75, 2, pp.  373-385.
 94.  Vuceia.J., and J. J. Morgan. 1978. Chemical Modeling of Trace Metals in Fresh Waters:
     Role of Complexation and Adsorption. Environmental Science and Technology, 12, 12,
     pp.  1302-1309.
 95.  Davis, J; A, 1982. Adsorption of Natural Dissolved Organic Matter at the Oxide/Wate-
     Interface. Geochim. Cosmoichim  Acta, 46, pp. 2381-2393.
 96.  Lion, L. W., R. S.'Altmann, and J. O. Leckie. 1982. Trace-metal Adsorption Characteristics
     of Estuarine Paniculate Matter: Evaluation of Fe/Mn Oxide and Organic Surface Coatings.
     Environmental Science and Technology, 16,  10, pp. 660-666.
 97.  O'Connor, D., and J.  P. Connelly.  1980. The Effect of Concentration of Adsorbing Solids
     on the Partition Coefficient- Water Research, 14, pp. 1517-1523.
 98.  Schweich, D., and M. Sardin.  1981. Adsorption, Partition, Ion Exchange, and Chemical
     Reaction in Batch Reactors or Columns — A Review. J. Hydrol., 50, pp. 1-33.
 99,  Reardon, E. J.  1981. Kd's — Can They Be Used to Describe Reversible Ion Sorption
     Reactions in Contaminant Migration?  Ground Water, 19, 3, pp. 279-286.

100.  Coles, D. G.,  and L, D. Ramspott. 1982.  Migration of Ruthenium-106 in a Nevada Test
     Site Aquifer; Discrepancy Between Field and Laboratory Results. Science, 215, pp. 1235-
     1237.
101,  Allegheny Ludlum Steel Corporation.  1959. Stainless Steel Handbook; Allegheny Ludlum
     Steel Corporation, Pittsburgh, Pennsylvania, 120 p.
102.  Marsh, J. M.,  and J. W. Lloyd. 1980. Details of Hydrochemical  Variations  in Flowing
     Weils. Ground Water, 18, 4, pp. 366-373.
103.  Parks, W. S., D. D. Graham, and J.  F. Lowery. 1981. Chemical  Character  of Ground
     Water in the  Shallow Water-Table Aquifer at Selected Localities in  the Memphis Area,
     Tennessee.  USGS  Open-File Report 81-223, 29 p.
104.  Parks, W,  S,,  D.  D. Graham, and J. F.  Lowery.  1982. Installation and  Sampling of
     Observation Wells and Analyses of Water from the Shallow  Aquifer at Selected Waste-
     Disposal Sites in the Memphis Area, Tennessee. USGS Open-File Report 82-266, 32 p.
105.  Parish, C. A. 1969. Plastic Pipe and Water Quality.  Journal, AWWA, September,  pp. 480-
     482.
106.  American Society for Testing Materials. 1981. Standard Specifications for Thermoplastic
     Water Well Casing pipe and Couplings Made in Standard Dimension Ratios. ANSI/ ASTM
                                           74

-------
     F480-81, American Society  for Testing and Materials, 1916 Race Street,  Philadelphia,
     Pennsylvania, 19103, 18 p.                         ,       '   '
107. American Society for Testing Materials. 1982, Standard Specifications for IPS Rigid Poly
     (Vinyl' Chloride)' (PYC) Plastic Pipe, Schedules 40, 80, and 120, ASNI/ASTM D1785-
     82, American Society for  Testing and Materials, 1916 Race Street, Philadelphia, Penn-
     sylvania, 19103, 12 p.
108, U.S. Department  of Commerce-NBS. 1963. Polyvinyl Chloride (PVC) Plastic Pipe Com-
     mercial Standard  CS256-63. National Bureau of Standards, 15  p.
109. American Water Works Association.  1975. AWWA Standard for  Polyvinyl Chloride (PVC)
     Pressure Pipe, 4 in, through 12 in., for Water. AWWA, C900-75, American  Water Works
     Association, Journal, AWWA, December, pp. 694-704.
110. Plastics Pipe  Institute. 1973. Weatherability of Thermoplastic Piping, Technical Report
     PPI-TR18-March 1973, Plastics Pipe Institute, A  Division of the Society of the Plastics
     Industry, Inc., New York,  NY, 5 p.  •
111. McClelland, N. I.  1981. Monitoring for Toxicological Safety. Presented at the  International
     Conference on Underground Plastic Pipe, Pipeline Division of the American Society of
     Civil Engineers, New Orleans, LA, 1981,12 p.
112, Banzer, J. D.  1978. The  Migration of Vinyl Chloride Monomer from PVC Pipe into
     Water. In Proceedings of the National Technical Conference on Safety and Health with
     Plastics, 1978, Society of Plastics Engineers.
113. California Department of  Health and Human Services, 1980, Final Report of Potential
     Health Hazards Associated With the Use of Plastic Pipe in Potable Water Systems. Hazard
     Evaluation System and Information Services, Departments  of Health and Human Services
     and Industrial Relations, October, Includes appended results of HALTS study for State
     Toxic  Substances  Hazard Alert System  prepared by J.  M.  Montgomery  Consulting
     Engineers.
114. Crompton, T. R. 1971. Chemical Analysis of Additives in Plastic.  Pergamon Press, Oxford,
     England, 162 p.
115. Nass,  L. I. 1976. Encyclopedia of PVC, Volumes 1 and 2. Marcel Dekker, Inc., New York,
     Vol. 1 — 600 p.,  Vol. 2 — through p.  1249.
116. Reich, K, D.,  A. R, Trussell,  F. Y, Lien, L.Y.C. Leong, and R. R.  Trussell. 1981. Diffusion
     of Organics  from Solvent-Bonded Plastic Pipes : Used for Potable Water  Plumbing.
     Presented at American Water Works Association Conference, St. Louis, Missouri, June 7-
     II.
117. Wang, T. C,(  and J. L. Bricker, 1979. 2-Butanone and Tetrahydrofuran Contamination
     in the Water Supply. Bull.  Envir. Contam. Toxicol., 23, pp. 620-623.
118. Simpson, D., and B.  R.  Currell.  1971.  The Determination of Certain  Antioxidants,
     Ultraviolet  Absorbers, and Stabilizers in  Plastics Formulations  by  Thin Layer Chroma-
     tography. Analyst, 96, pp.  515-521,
119. Packham, R.  F, 1971. The Leaching of Toxic Stabilizers from Unplasticized PVC Water
     Pipe;  Part II. A Survey of Lead Levels  in a PVC Distribution System. Water Treatment
     and Examination, 20, 3, pp.  144-151.
120. Packham, R.  F. 1971. The Leaching of Toxic Stabilizers from Unplasticized PVC Water
     Pipe:  Part III. The Measurement of Extractable Lead in  PVC  Pipes. Water Treatment
     and Examination, 20, 3, pp.  152-164.
121. Niklas,  H., and W. Meyer.  1961. The Migration of  Lead from Pb-Stabilized Pipes.
     Kunstoffe, 51, pp. 2-6.


                                          75

-------
122. Monaghan, C, P., V, H. Kulkarni, and N, K, Good. 1978. Release Mechanisms of Organotin
     Toxicants from  Coating Surfaces: A  Leaching Model for Antifouling Coatings. Chapter
     22 (pp.  359-371) In Organometals and Organometalloids Occurrence and  Fate in the
     Environment, E E. Brinckman and J.  M. Beliama, editors, ACS Symposium Series 82,
     1978, 447 p.
123, Tobias, R. S. 1978.  The Chemistry of Organometallic Cations in Aqueous Media. Chapter
     9 (pp.  130-148) -In Organometals and Organometalloids  Occurrence  and  Fate  in the
     Environment, F. E. Brinckman and J.  M. Beliama, editors, ACS Symposium Series 82,
     1978, 447 p.
124, Miller, D., O, C. Braids, and W. Walker. 1977. The Prevalence of Subsurface Migr.ition
     of Hazardous Chemical  Substances  at  Selected Industrial Waste Land  Disposal Sites.
     Report of U.S. EPA Office  of Solid Waste,  November (NTIS-PB-275103), 165  p. +
     Appendices.
125. Swan, J, D. 1982.  Relative Corrosion  Resistance of Certain Steels in Soils and Waters.
     Materials Performance, June; also reprinted in Johnson Driller's Journal, 3rd-4th Quarter
     1982, pp,  6-9.
126. Boyle, E. A.,J. M. Edmond, and E. R. Sholkovitz. 1977. The Mechanism of Iron Removal
     in Estuaries.  Geochim. Cosmochim Acta, 41, pp. 1313-1334.
127, Singer, P. C., and W, Sturnm. 1970. The Solubility of Ferrous Iron in Carbonate-Bearing
     Waters. Journal, AWWA, 62,  pp. 198-202.
128. Jenne, E. A.  1968.  Controls on Mn, Fe, Co, Ni, Cu, and Zn Concentrations  in Soils and
     Water in Trace  Inorganics in  Water. ACS Advances in Chemistry Series No. 73, American
     Chemical Society, Washington, D.C.,  p. 337.
129. Blanchard, C. N., Jr. 1978,  Hydrologic and Chemical Analyses of the Old Town and
     Hampden, Maine, Well Fields with Regard to the High Iron and Manganese Concentration
     Problem. Unpublished M.S. Thesis, Department of Civil Engineering, Maine University,
     Orono (NTIS-PB-291741) August, 144 p.
130. Daesen.J. R, 1975. Mechanisms of Corrosion of the Coating. Pamphlet of the Galvanizing
     Institute, 28  p.
131. Hubbard, D. J., and C.  E.  A.  Shanahan. 1973. Corrosion of  Zinc and Steel in Dilute
     Aqueous Solutions. Br. Corros., J., 8, pp. 270-274.
132. Mechanich, D.  J.  1980. Tertiary Wastewater Treatment  Using a  Natural Peat  Bog.
     Unpublished M.S. Thesis, College of Natural Resources, University of Wisconsin, Stevens
     Point, May.
133, Imbrigiotta,  T,  E.  1983. Personal Communication.  USGS, Water  Resources  Division,
     Indianapolis, Indiana.
134. McGlothlin,  R.  E., and  H.  Krause.  1980. Water Based Drilling Fluids,  pp. 30-37. In
     Proceedings  of  the Symposium/Research on Environmental Fate and Effects of Drilling
     Fluids and Cuttings, USA, January 21-24, Lake Buena Vista, FL, Vol. 1 —  708 p., Vol.
     2 — 442 p.
135, N L Baroid.  1981. Product  Literature on EZ-MUD. N L  Baroid/N L Industries, Inc.,
     Houston, Texas, June.
136. I.M.C. Corporation. 1982. Product Literature on Fasthik® I and Fasthik® II. International
     Minerals and  Chemical Corporation,  Incore Division, Des Plaines, Illinois, February.
137. Proceedings  of  the Symposium/Research on Environmental Fate and Effects of Drilling
     Fluids and Cuttings, USA. Lake  Buena Vista, Florida, January 21-24, 1980, Vol.  1 —
     708 p., Vol. 2 — 442 p.


                                          76

-------
138. Strasher, M. T. 1980. Characterization of Organic Constituents in Waste Drilling Fluids,
     pp. 70-79. In Proceedings of the Symposium/Research on Environmental Fate and Effects
     on  Drilling Fluids and Cuttings, USA, Lake Buena Vista, Florida, January 21-24, 1980,
     Vol. 1 — 708  p., Vol. 2 — 442 p.
139, Mortland, M. M, 1970, Clay-Organic Complexes and Interactions. Advances in Agronomy,
     22, pp,  75-117.          '                         .                     -
140. Hunt, G., P, Spooner, V, Hodge, and P. Wagner. 1983, Collection of Information on the
     Compatibility of Grouts with Hazardous  Wastes. Presented at Ninth  Annual Research
     Symposium, Land Disposal,  Incineration, and Treatment of Hazardous Wastes, May 2-4,
     Ft.  Mitchell, Kentucky,  USEPA-SHWRD/EPCD sponsored.
141. Kreft, P., A. Trussell, J. Long, M, Kavanaugh, and R. Trussell. 1981. Notes and comments:
     leaching or organics from  a PVC-polyethylene-plexiglas pilot plant. Journal AWWA,
     October, pp. 558-560.
142. Tomson, M. B., J. Dauchy,  S. Hutchins, C. Curran, C. j. Cook, and C. H. Ward, 1981.
     Groundwater Contamination by Trace Level Organics from a Rapid  Infiltration Site.
     Water Resources, 15, pp.' 11094 116.
143. Schroeder,  R.  A., and D. S. Snavely. 1981. Survey of Selected Organic Compounds in
     Aquifers of New York  State Excluding  Long  Island. U.S. Geological  Survey  Water
     Resources Investigation 81-47, Albany, New York, 60 p.
144. Keith,  L. H.,  and  W. A, Telliard.  1979.  Priority Pollutants I —  a  perspective  view.
     Environmental Science and  Technology, 13, 4, pp. 416-423.

145. Evans,  R. 1982. Personal communication. USEPA-Environmental Monitoring and Sur-
     veillance Laboratory, Las Vegas, Nevada.
146. Segar, D. A., and G. A, Berberian. 1975, Trace Metal Contamination by Oceanographic
     Samplers. Chapter  3, pp, 16-21, in  Analytical Methods in  Oceanography, T.R.P. Gibb,
     Jr.,  ed., ACS Advances in Chemistry Series 147, American Chemical Society, Washington,
     B.C., 238 p.
147. Tirpak,  G. 1970. Microbial Degradation  of Plaslicized PVC.  Soc.  Plastics Engineers
     Journal, 26, pp. 26-30.
148. Haxo, H. E., jr., R. S. Haxo,  and T. F.  Kellogg.  1979.  Liner Materials- Exposed to
     Municipal Solid  Waste  Leachate.  Matrecon  Inc.  report prepared for USEPA-MERL,
     Cincinnati,  Ohio, 45268, EPA-600/2-79-038, July, 51 p.
149. Haxo, H. E., Jr., 1980.  Interaction of Selected Lining Materials with Various Hazardous
     Wastes-II. In Disposal of Hazardous Waste, proceedings of the Sixth Annual Research
     Symposium, March 17-20, Chicago, Illinois, EPA-600/9-80-010, pp. 160-180.
150. USEPA. 1979.  Methods for  Chemical Analysis of Water and Wastes. EPA-600/4-79-020,
     USEPA-EMSL, Cincinnati, Ohio 45279, March.
151. Hydrochemistry Team-German Chemists Association, 1981. Preservation of Water Sam-
     ples. Wrater Research, 15, pp. 233-241.
152. Summers, W. K. 1972.  Factors  Affecting  the Validity of Chemical Analyses  of Natural
     Water. Ground Water, 10, 2, pp. 12-17.
153. Eccles, L. A., and R, R.  Nicklen.  1978. Factors Influencing the Design of a Ground Water
     Quality  Monitoring  Network. American Water  Resources Association,  Water Resources
     Bulletin, Establishment of Water Quality Monitoring Programs, June, pp. 196-209,
154. Nelson, J. D., and R. C. Ward. 1981. Statistical Considerations and Sampling Techniques
     for  Ground Water Quality Monitoring. Ground Water, 19, 6, pp.  617-625.
                                          77

-------
 155. Experience of the State Water Survey Laboratories' participation in the USEPA Quality
     Assurance Program,
 156, Trussell, A.  R. 1983,  Precision  and Accuracy  Data on Trace Organic Compounds.
     Persona] communication,  Environmental Research  Laboratoryjam.es M.  Montgomery,
     Consulting Engineers, Inc.. Pasadena, California,  March.
 157, Mills, P. E.  1982. Precision and Accuracy Data on Trace Organic Compounds and Priority
     Pollutant Metals. Personal communication, Meade CompuChem, Inc., Research Triangle,
     North Carolina, August.
 158. Patterson, C. C, and D. M.  Settle. 1976. The Reduction of Order of Magnitude Errors
     in Lead Analyses of Biological Materials by  Evaluating and Controlling the  Extent and
     Sources of Industrial Lead Contamination Introduced during Sampling Collecting and
     Analysis. In Accuracy in Trace Analysis: Sampling,  Handling, and Analysis, P, LcFl.eur,
     editor, U.S. National Bureau of Standards Special Publication #422, pp. 321-351.
 159. Kvenvolden, K. A. 1975, Advances in the Geochemistry of Amino Acids. Annual Review
     of Earth and Planetary  Sciences, Vol. 3', pp,  183-212.
 160. American Chemical Society.  1980.  Guidelines for Data Acquisition  and  Data Quality
1     Evaluation  in Environmental Chemistry. American  Chemical Society Committee on  En-
     vironmental Improvement, 52, pp. 2242-2249.
                                           78

-------
                                   TECHNICAL REPORT DATA
                            (Please read Jaaructions on (he reverse before completing}
  REPORT NO.
  EPA-6QQ/2-84-Q24
            3, RECIPIENT'S ACCESSlON>NO,  _
                PBS 4   i U 7 7 9
   riTLE AMDSU8T1TLE
  "A GUIDE TO  THE SELECTION OF MATERIALS FOR MONITORING
   WELL CONSTRUCTION AND GROUND-WATER SAMPLING"
                                                           5, REPORT DATE
                                                               January  1984
            6, PERFORMING ORGANIZATION CODE
 7 AUTHORiSI
                                                           8. PERFORMING ORGANIZATION REPORT NO
   Michael J.  Barcelona, James P. Glbb, and
      Robin A. Miller
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
   Illinois State  Water Survey  •
   Department of Energy and Natural Resources
   Champaign, IL   61820
            10. PROGRAM ELEMENT NO,

                 CBPC1A
            11. CONTRACT/GRANT NO-
                 CR-809966
 12. SPONSORING AGENCY NAME AND ADDRESS
   Robert S. Kerr  Environmental  Research Lab.
   U.S. Environmental  Protection Agency
   Post Office Box  1198
   Ada, OK  74820
- Ada,  OK
13. TYPE OF REPORT AND PERIOD COVERED
  Interim  (05/82  - 05/83)
            14, SPONSORING AGENCY CODE
                 EPA/600/15
 16. SUPPLEMENTARY NOTES
 16. ABSTRACT
           The project was  initiated to supplement and  update existing guidance
      documents for effective, ground-water monitoring efforts.   The areas of
      primary concern were  the potential sources of errors  in chemical analyses
      of subsurface samples caused by well construction  and sampling materials,
      techniques or procedures.   A critical review of the literature was conducted
      on various aspects  of monitoring natural waters, materials'  performance
      data and unpublished  information on the success of various ground-water
      monitoring techniques.   The results of the literature review were collected
      and reviewed by a panel  of experts from government agencies, private hydro-
      logical consulting  firms,  the manufacturing industry  and  national standards
      organization.  The  publication consists of a thorough discussion of ground-
      water monitoring strategies, requirements and pitfalls.   It  concludes,with a
      detailed treatment  of the  costs and benefits of recommended  monitoring design
      criteria.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS  c,  COSATI Field/Group
   Hydrology
   Ground water
   Aquifers
   Chemistry and analysis of  pollutants
   Control techniques and equipment
Monitoring well  con-
 struction
                   68 D
 3, DISTRIBUTION STATEMENT
   Release unlimited.
                                              19. SECURITY CLASS (This Report)
                                               Unclassified
                                              20, SECURITY CLASS (This page)

                                               Unclassified
EPA Form 2220-1 (S-73!

-------

-------