EPA/600/A-96/093

 Design and Interpretation of Microcosm Studies for Chlorinated Compounds
                           Barbara H. Wilson and John T.  Wilson
                       National Risk Management Research Laboratory,
                    U.S.  Environmental Protection Agency,  Ada, Oklahoma
                                                                      ^
                                       Darryl Luce
                                         Region 1
                     U.S. Environmental Protection Agency, Boston, MA
 Introduction
 'There are three lines of evidence used to support natural attenuation as a remedy for chlorinated
 solvent contamination in ground water.  They are 1) documented loss of contaminant at field
 scale, 2)  geochemical analytical data, and 3) direct microbiological evidence.  The first line of
 evidence (documented loss) involves using statistically significant historical trends in contaminant
 concentration in conjunction with aquifer hydrogeological parameters,such as seepage velocity
 and dilution to show that a reduction in the total mass of contaminants is occurring at the site.
 The second line of evidence (geochemical data) involves the use of chemical analytical data in
 mass balance calculations to show that decreases in contaminant concentrations can be directly
 correlated to increases in metabolic by-product concentrations.  This evidence can be used to
 show that concentrations of electron donors or  acceptors in ground  water are  sufficient to
 facilitate degradation of the dissolved contaminants (i.e., there is sufficient capacity).  Solute fate
 and transport models can be used to aid the mass balance calculations, and to collate information
 on degradation.

•Microcosm  studies are often used  to  provide a  third  line  of evidence.   The potential  for
 biodegradation of the contaminants  of interest can be  confirmed by the  use  of microcosms,
 through comparison of removals  in the  living treatments  with removals in the  controls.
 Microcosm studies also permit an absolute mass balance determination based  on biodegradation
 of the contaminants of interest.  Further, the appearance of daughter products in the microcosms
 can be used to confirm biodegradation of the parent compound.
                                            1

-------
 When to Use Microcosms,

 There are two fundamentally different applications of microcosms.  They are frequently used in
 a qualitative way  to  illustrate the  important  processes  that control the  fate  of organic
 contaminants.    They  are  also used  to estimate  rate  constants  for bjotransformation  of
 contaminants that can  be  used in a  site-specific  transport  and  fate  model  of a plume  of
 contaminated ground water.  This paper only discusses microcosms for  the second application.

 Microcosms should be used when there is no other way to obtain a  rate constant for attenuation
 of contaminants, in particular, when it is  impossible to estimate  the rate of attenuation from data
 from monitoring wells in the plume of concern.  There are  situations where it is  impossible to
 compare  concentrations  in monitoring wells  along a  flow  path due to legal or physical
 impediments.  In many landscapes, the direction of ground-water flow (and water table elevations
 in monitoring wells) can vary over short periods of time due  to tidal influences or changes in
 barometric pressure.  The direction of ground-water flow may also be affected by changes in the
 stage of a nearby river or pumping wells in the vicinity.  These changes in ground-water flow
 direction  do not  allow  simple  snap-shot comparisons of concentrations in monitoring wells
 because of uncertainties  in identifying the flow path.  Rate  constants from microcosms can be
 used with average flow conditions to estimate attenuation at some point of discharge or point of
 compliance.

 Application  of Microcosms.

 The primary objective of microcosm studies is to obtain rate constants applicable to  average flow
 conditions.  These average condition can be determined by continuous monitoring of water table
 elevations in the aquifer being evaluated.  The product  of the  microcosm  study and the
continuous monitoring of water table elevations will be a yearly or seasonal estimate  of the extent
of attenuation along  average flow paths.  Removals seen  at  field scale can be  attributed to
 biological activity.  If removals in the microcosms duplicate removal  at field scale, the rate
constant can  be used for risk assessment purposes.
                                            2

-------
 Selecting Material for Study

 Prior to choosing material for microcosm studies, the location of major conduits of ground-water
 flow should be identified and the geochemical regions along the flow path should  be determined.
 The important geochemical regions for natural attenuation of chlorinated aliphatic hydrocarbons
 are  regions  that are actively  methanogenic; regions that exhibit sulfate reduction and  iron
 reduction  concomitantly; and  regions  that  exhibit iron reduction alone.   The  pattern  of
 biodegradation of chlorinated solvents varies  in different regions.  Vinyl  chloride  tends  to
 accumulate during reductive dechlorination of trichloroethylene (TCE) or tetrachloroethylene
 (PCE) in methanogenic  regions (1,2); it does  not  accumulate to  the  same extent  in  regions
 exhibiting iron  reduction and sulfate reduction (3). In regions showing iron reduction alone, vinyl
 chloride is consumed but dechlorination of PCE, TCE, or dichloroethylene (DCE)  may not occur
 (4).  Core material from each geochemical region in major  flow paths represented by the plume
 must be acquired, and the  hydraulic conductivity of each depth at which core material is acquired
 must be measured.  If possible, the microcosms should be constructed with the  most transmissive
 material in the flow path.

 Several characteristics of ground water from the same interval used to collect the core material
 should be  determined.    These characteristics include  temperature, redox potential, pH,  and
 concentrations of oxygen, sulfate, sulfide, nitrate,  ferrous iron, chloride, methane, ethane, ethene,
 total organic carbon, and alkalinity. The concentrations of compounds of regulatory concern and
 any breakdown products for each site  must be determined.  The ground  water should be analyzed
 for methane to determine if methanogenic conditions exist and for ethane  and ethene as daughter
 products. A comparison of the ground-water chemistry from the interval where  the cores were
acquired to that in neighboring monitoring wells will demonstrate if  the collected cores are
representative of that section of the contaminant plume.

 Reductive dechlorination of chlorinated solvents requires an electron donor to allow the process
to proceed.  The  electron donor could be  soil  organic matter, low molecular weight organic
compounds (lactate, acetate, methanol, glucose, etc.), H2,  or a co-contaminant such as landfill
                                            3

-------
 leachate or petroleum compounds (5,6,7).  In many instances, the actual electron donor(s) may
 not be identified.

 Several characteristics of the core material should  also be evaluated.  The initial concentration
 of the contaminated material (^g/kg) should be identified prior to construction-of the microcosms.
 Also, it is necessary to know if the contamination is present as a nonaqueous phase liquid
 (NAPL) or in solution. A total petroleum hydrocarbon  (TPH) analysis wijl determine if any
 hydrocarbon-based oily materials are present.  The water-filled porosity is a parameter generally
 used to extrapolate rates to the field.  It can be calculated by comparing wet and dry weights of
 the aquifer material.

 To insure sample integrity and stability during acquisition,  it is important to quickly transfer the
 aquifer material into a jar,  exclude  air  by adding ground  water,  and seal the jar  without
 headspace. The material should be cooled during transportation to the laboratory.  Incubate the
 core  material at the ambient ground-water temperature in the dark before the construction of
 microcosms.

 At least one microcosm study per geochemical region should be completed. If the plume is over
 one  kilometer in  length, several microcosm studies per  geochemical region may need to be
 constructed.

 Geochemical Characterization of the Site

 The  geochemistry of the subsurface affects  behavior of  organic  and inorganic contaminants,
 inorganic minerals, and microbial populations.  Major geochemical parameters that characterize
the  subsurface encompasses (1) pH; (2) redox potential,  Eh;  (3) alkalinity;  (4) physical and
chemical  characterization of the solids;  (5)  temperature;  (6)  dissolved constituents, including
electron  acceptors; and (7) microbial processes.   The most important of these in  relation  to
biological  processes  are redox potential,  alkalinity, concentration  of electron acceptor, and
chemical nature of the solids.

-------
 Alkalinity  Field indications of biologically active portions of a plume  may  be identified  by
 increased alkalinity, compared to background wells, from carbon dioxide due to biodegradation
 of the pollutants. Increases in both alkalinity  and pH have been measured in portions of an
 aquifer contaminated by gasoline undergoing active utilization of the gasoline  components (8).
 Alkalinity can be one of the parameters used when identifying where to collect biologically active
 core material.

 pH  Bacteria generally prefer a neutral or slightly alkaline pH level with an optimum pH range
 for most microorganisms between 6.0 and 8.0; however, many microorganisms can tolerate a pH
 range  of 5.0 to 9.0.  Most ground  waters in uncontaminated aquifers are within these ranges.
 Natural pH values may be as low as  4.0 or 5.0 in aquifers with active oxidation of sulfides, and
 pH values as high as 9.0 may be found in carbonate-buffered  systems (9),   However, pH values
 as low as 3.0 have been measured for ground waters contaminated with municipal waste leachates
 which often contain  elevated concentrations of organic acids (10).  In ground  waters contaminated
 with sludges from cement manufacturing, pH values as high as 11.0 have been measured (9).

 Redox The oxidation/reduction (redox) potential (Eh) of ground water is a measure of electron
 activity that indicates the relative ability of a solution to accept or transfer electrons. Most redox
 reactions in the subsurface are microbially catalyzed during metabolism of native organic matter
 or contaminants.  The only elements that are predominant participants  in aquatic  redox processes
 are carbon, nitrogen, oxygen, sulfur, iron, and manganese (11).  The principal oxidizing agents
 in ground water are oxygen, nitrate, sulfate, manganese  (IV), and iron  (III).  Biological reactions
 in the subsurface both influence and are affected by the redox potential and the available electron
 acceptors. The redox potential changes with the predominant electron acceptor,  with reducing
conditions increasing  through  the sequence  oxygen, nitrate, iron, sulfate, and  carbonate.  The
redox  potential decreases in  each  sequence, with methanogenic (carbonate  as the  electron
acceptor) conditions  being most reducing.   The interpretation of redox potentials in ground waters
is difficult (12).  The potential  obtained in ground waters is a mixed potential  that reflects the
potential  of many  reactions and cannot be used for quantitative  interpretation (11).   The
                                             5

-------
 approximate location of the contaminant plume can be identified in the field by measurement of
 the redox potential of the ground water.

 To overcome the limitations imposed by traditional redox measurements, recent work has focused
 on the measurement of molecular hydrogen to accurately describe the predominant in situ redox
 reactions (13, 14, 15).  The evidence suggests that concentrations of H2 in ground water can be
 correlated with specific  microbial processes, and these concentrations can be used to identify
 zones of methanogenesis, sulfate reduction, and iron reduction in the subsurface (3).

 Electron acceptors  Measurement of the available electron acceptors is critical in identifying the
 predominant microbial  and  geochemical processes occurring  in  situ at the time of sample
 collection.  Nitrate and sulfate are found naturally in most ground waters  and will subsequently
 be used as electron acceptors once oxygen is consumed. Oxidized forms of iron and manganese
 can be used as electron acceptors  before sulfate reduction commences.   Iron and manganese
 minerals solubilize coincidently with sulfate reduction, and their reduced forms scavenge oxygen
 to the extent that  strict  anaerobes  (some sulfate reducers and  all methanogens)  can  develop.
 Sulfate is found in  many depositional environments, and sulfate reduction may be very  common
 in many contaminated ground waters.  In environments where sulfate is depleted, carbonate
 becomes the electron acceptor with methane gas produced as  an end product.

 Temperature  The temperature at all monitoring wells should be measured to determine when the
 pumped water has stabilized and  is ready for collection.  Below approximately  30  feet, the
temperature in the subsurface is fairly consistent on an annual basis.  Microcosms should be
 stored at the average in  situ temperature.  Biological growth can  occur  over a wide  range of
temperatures, although most microorganisms are active primarily between  10° and 35°C (50° to
95°F).

Chloride Reductive dechlorination results in the accumulation of inorganic chloride.  In aquifers
with a  low background  of inorganic  chloride,  the  concentration of inorganic chloride should
 increase as the chlorinated solvents are degraded. The sum of the inorganic chloride plus the
                                           6

-------
 contaminant being degraded should remain relatively consistent along the ground-water flow path.

 The following tables (Table 1,  Table 2) list the geochemical  parameters, contaminants, and
 daughter products that should be measured during site characterization for natural attenuation.
 The tables include  the analyses that  should be performed, the optimum  range  for natural
 attenuation of chlorinated solvents, and the interpretation of the value in relation to biological
 processes.

 Microcosm Construction

 During construction of the microcosms,  it is best if all manipulations take place in an anaerobic
 glovebox.  These gloveboxes exclude oxygen and provide an environment where the integrity of
 the core material may be maintained, since many strict anaerobic bacteria are sensitive to oxygen.
 Stringent aseptic precautions not necessary for microcosm construction.  It is more important to
 maintain anaerobic  conditions of the aquifer material  and solutions added to the  microcosm
 bottles.

 The microcosms should have approximately the same ratio of solids to water as the in situ aquifer
 material, with a minimum or negligible headspace.  Most bacteria in the subsurface are  attached
 to the  aquifer solids.  If a microcosm has an excess of water, and the contaminant is primarily
 in the  dissolved phase, the bacteria must consume or transform  a great deal more contaminant
 to produce the same  relative change in the contaminant  concentration.  As a result, the kinetics
 of removal at field scale will be underestimated  in the microcosms.

 A minimum of three replicate microcosms  for both living  and control treatments should be
 constructed for  each sampling event.   Microcosms  sacrificed  at each sampling interval are
preferable to microcosms that  are repetitively sampled.  The compounds of regulatory interest
 should be added at concentrations representative of the  higher concentrations found in  the
geochemical region of the plume being evaluated.  The compounds  should be added  as  a
concentrated aqueous solution.  If an aqueous solution is not feasible, dioxane or acetonitrile may
                                            7

-------
 be used as solvents.  Avoid carriers that can be metabolized anaerobically, particularly alcohols.
 If possible, use ground water from the site to prepare dosing solutions and to restore water lost
 from the  core barrel during  sample collection.

 For long term microcosm studies, autoclaving is the preferred method for sterilization. Nothing
 available to sterilize core samples works perfectly.  Mercuric chloride is excellent for short term
 studies (weeks or months).  However, mercuric chloride complexes to clays, and control may
 be lost as it is sorbed over time.   Sodium azide is effective in repressing  metabolism of bacteria
 that have cytochromes, but is not effective on strict anaerobes.

 The microcosms should be incubated in the  dark at the ambient temperature of the~aquifer.  It
 is preferable that the microcosms be incubated inverted in an anaerobic glovebox.  Anaerobic jars
 are also available that maintain an oxygen-free environment for the microcosms.  Dry redox
 indicator strips can be placed in  the jars to assure that anoxic conditions are maintained. If no
 anaerobic storage is available, the inverted microcosms can be immersed in approximately two
 inches of water  during incubation.  Teflon-lined butyl  rubber septa are excellent for excluding
 oxygen and should be used if the  microcosms must be stored outside an anaerobic  environment.

The studies should last from  one year to eighteen  months.  The  residence time of a plume may
be several years to tens of years at field scale.  Rates of transformation that are slow in terms of
laboratory experimentation may have a considerable environmental significance.  A microcosm
study  lasting only a  few weeks to  months may not have the resolution to detect slow changes that
are of environmental significance.  Additionally, microcosm studies often distinguish a  pattern
of sequential biodegradation of the contaminants of interest and their daughter products.

Microcosm Interpretation

As a practical matter, batch microcosms with an optimal solids/water ratio, that are sampled every
two months in triplicate, for up to eighteen  months, can resolve biodegradation from  abiotic
losses with a detection limit of 0.001  to 0.0005 per  day.  Rates determined from replicated batch
                                             8

-------
 microcosms are found to more accurately duplicate field rates of natural attenuation than column
 studies.  Many plumes show significant attenuation of contamination at field calibrated rates that
 are  slower than  the detection limit  of  microcosms  constructed  with  that aquifer material.
 Although rate constants  for modeling purposes are more appropriately acquired from field-scale
 studies, it is reassuring when the rates in  the field and the rates in the laboratory agree.

 The rates measured in the microcosm study may be faster than the estimated field rate. This may
 not be due to an error in the laboratory study, particularly if estimation of the field-scale rate of
 attenuation  did not account for regions  of preferential flow in  the  aquifer.  The regions of
 preferential flow may be determined by use of a down-hole flow rneter or by use of a geoprobe
 method for  determining hydraulic conductivity in one to two feet sections of the aquifer.

 Statistical comparisons can  determine if removals of  contaminants  of  concern  in the living
 treatments are significantly different from zero or significantly different from any sorption that
 is occurring.   Comparisons are made  on  the first-order rate of removal,  that is,  the slope of a
 linear regression  of the natural  logarithm  of the  concentration  remaining  against time  of
 incubation for both the living and control microcosm. These slopes (removal rates) are compared
 to determine if they are  different, and if so, extent of difference that can be detected at a given
 level of confidence

 The Tibbetts Road Case Study

The Tibbetts Road  Superfund  Site in Barrington,  N.H.. a former private home, was used to store
drums of various chemicals  from 1944 to 1984,  The primary ground-water contaminants in the
overburden and bedrock  aquifers were benzene and TCE with respective concentrations of 7,800
jig/L and 1,100 ng/L.  High concentrations of arsenic, chromium, nickel, and lead were also
found.

Material collected at the  site was used to construct a microcosm study evaluating the  removal of
benzene, toluene, and TCE.  This material was acquired from the most contaminated source at
                                            9

-------
 the site, the waste pile near the origin of Segment A (Figure 4). Microcosms were incubated for
 nine months. The aquifer material was added to 20-mL headspace vials, dosed with one mL of
 spiking  solution,  capped  with  a Teflon-lined,  gray  butyl  rubber  septa, and sealed with  an
 aluminum crimp cap.  Controls were prepared by autoclaving the material used to construct the
 microcosms overnight.  Initial concentrations for benzene,  toluene, and TCE were, respectively,
 380 [ig/L, 450 [ig/L, and 330 [ig/L.  The microcosms were  thoroughly mixed by vortexing, then
 stored inverted in the dark at the ambient temperature of  10°C.

 The results (Figures 1, 2, and 3; Table 3) show that significant biodegradation of both petroleum
 aromatic hydrocarbons and the chlorinated solvent had occurred.   Significant removal in the
 control microcosms also occurred for all compounds. The data exhibited more variability in the
 living microcosms than in the control treatment, which is a pattern that has been  observed in
 other microcosm studies.  The removals observed in the controls are probably due to sorption;
 however, this study exhibited more sorption than typically seen.

 The rate  constants determined from the microcosm study for  the three compounds are shown in
 Table 4.  The appropriate rate constant to be used in a model or a risk assessment would be the
 first-order removal in the living treatment minus the first-order removal in the control, in other
 words the removal  that is in excess of the removal in the  controls.

The first-order  removal  in the  living  and control microcosms was estimated  as the linear
regression of the natural  logarithm of  concentration remaining  in each microcosm in each
treatment against time  of incubation.  Student's t distribution with n-2 degrees of freedom was
used to estimate the 95% confidence interval.  The standard  error of the difference of the rates
of removal  in living and control microcosms was estimated as the square root of the sum of the
squares of the standard errors of the living and control microcosms, with  n-4 degrees of freedom
(16).

Table 5  presents the concentrations of  organic compounds and  their  metabolic  products  in
monitoring wells used to define  line segments  in the aquifer for estimation of field-scale rate
                                           10

-------
 constants.  Wells  in this aquifer showed little accumulation  of trans-DCE; 1,1-DCE; vinyl
 chloride; or ethene, although removals of TCE and  cis-DCE were extensive.  This can  be
 explained by the observation (4) that iron-reducing bacteria can rapidly oxidize vinyl chloride to
 carbon dioxide. Filterable iron accumulated in ground water in this  aquifer.
                                                                      ^
 The extent of attenuation from well to well listed in Table 5, and the travel time between wells
 in a segment (Figure 4) were used to calculate first-order rate constants for each segment (Table
 6). Travel time between monitoring wells was calculated from site-specific estimates of hydraulic
 conductivity and from the hydraulic gradient.  In the area sampled  for the microcosm study, the
 estimated Darcy flow was 2.0 feet per year.  With an estimated porosity in this particular glacial
 till of 0.1, this corresponds to a plume velocity of 20 feet per year.

 SUMMARY

 Table 7 compares the first-order rate constants estimated  from the microcosm studies to  the rate
 constants estimated at field scale.  The agreement between the independent estimates of rate is
 good;  indicating that the rates can appropriately be used  in a risk  assessment.   The rates  of
 biodegradation documented in the microcosm study could easily account for the disappearance
 of trichloroethylene, benzene,  and toluene observed at field scale.  The rates estimated from the
 microcosm study are several-fold higher than the rates estimated at  field scale.  This may reflect
 an underestimation  of the true rate in the field.  The estimates of plume velocity  assumed that
 the aquifer was homogeneous.  No attempt was made in this study  to correct the estimate  of
plume  velocity for the influence of preferential flow paths.  Preferential flow paths with a higher
hydraulic conductivity than average would result in a faster velocity of the plume, thus a lower
residence time and  faster rate of removal at field scale.

References
 1.  Weaver, J.W., J.T. Wilson, D.H. Kampbell.  1995. EPA Project Summary.  EPA/600/SV-
    95/001.  U.S. EPA.  Washington, D.C.
 2.  Wilson, J.T., D. Kampbell, J.  Weaver, B. Wilson, T. Imbrigiotta, and T.  Ehlke.  1995.

                                           11

-------
     Symposium on  Bioremediation of Hazardous Wastes: Research, Development, and Field
     Evaluations.  U.S. EPA.  Rye Brook, N.Y.

 3.  Chapelle, F.H.   1996.  Identifying redox conditions that favor the natural attenuation of
     chlorinated ethenes in contaminated ground-water systems. Proceedings of the Symposium
     on Natural Attenuation of Chlorinated Organics in Ground Water.  September 11-13, 1996.
     Dallas, TX.

 4.  Bradley, P.M. and  F.H. Chapelle.  1996. Anaerobic mineralization of vinyl  chloride in
     Fe(III)-reducing aquifer sediments.  Environmental Science and Technology. In Press.

 5.  Bouwer,  E.J.    1994.   Bioremediation  of chlorinated solvents using alternate electron
     acceptors.  In: Handbook of Bioremediation. Lewis  Publishers.  Boca Raton, FL.

 6.  Sewell, G.W. and  S.A. Gibson.   1991.   Stimulation of the reductive dechlorination  of
     tetrachloroethylene  in anaerobic aquifer microcosms  by the addition of toluene.
     Environmental Science and Technology.  25(5):982-984.

 7.  Klecka, G.M., J.T. Wilson, E. Lutz, N. Klier, R. West, J. Davis, J. Weaver, D. Kampbell,
     and  B. Wilson.  1996.  Intrinsic remediation  of chlorinated  solvents in ground  water.
     Proceedings of the  IBC/CELTIC  Conference on Intrinsic Bioremediation.  March  18-19,
     1996.  London, U.K.

 8.  Cozzarelli, I.M., J.S. Herman, and M.J. Baedecker. 1995.  Fate of microbial metabolites  of
     hydrocarbons  in a  coastal  plain aquifer: the role of electron  acceptors.   Environmental
     Science and Technology. 29(2):458-469.

 9.  Chapelle, F.H.  Ground-water Microbiology and Geochemistry. John Wiley & Sons, Inc.
    New York, New York.

10. Baedecker, M.J.,  and W.  Back, 1979. Hydrogeological processes and chemical reactions at
    a landfill.  Ground Water.  17(5):429-437.

11. Stumm, W., and J.J.  Morgan.  1970.  Aquatic Chemistry.  Wiley Intcrscience.  New York,
    New York.

12. Snoeyink, V.L. and D. Jenkins.  1980.  Water Chemistry.  John Wiley & Sons.  New York,
    NY.

13. Chapelle,  F.H.,  P.B. McMahon, N.M. Dubrovsky, R.F.  Fugii,  E.T. Oaksford, and D.A.
    Vroblesky.  1995.  Deducing the  distribution of terminal electron-accepting processes  in
    hydrologically diverse groundwater systems.  Water Resources Research.  31:359-371.

14. Lovley, D.R., F.H. Chapelle, and J.C. Woodward.  1994. Use of dissolved H2 concentrations

                                          12

-------
    to determine distribution of microbially catalyzed redox reactions in anoxic groundwater.
    Environmental Science and Technology.  28:1255-1210.

15.  Lovley, D.R.  and S.  Goodwin.  1988.  Hydrogen concentrations  as an indicator of the
    predominant terminal electron-accepting reactions in aquatic sediments.  Geochemica et
    Cosmochimica Acta.  52:2993-3003.
                                                                   *s
16.  Glantz, S.A.  1992.  Primer of Biostatistics.  McGraw-Hill, Inc.  New York, NY.
                                          13

-------
Table 1. Geochemical Parameters
Analysis
Redox Potential
Sulfate
Nitrate
Oxygen
Oxygen
Iron II
Sulfide
Hydrogen
Hydrogen
pH
Range
<50 millivolt
against Ag/AgCl
<20 mg/liter
<1 mg/liter
<0.5 mg/liter
>1 mg/liter
>1 mg/liter
>1 mg/liter
>1 nMolar
<1 nMolar
5 < pH< 9
Interpretation
Reductive pathway possible
Competes at higher concentrations with
pathway
Competes at higher concentrations with
pathway
Tolerated, toxic to reductive pathway at
concentrations
Vinyl chloride oxidized
Reductive pathway possible
Reductive pathway possible


reductive
reductive
higher



Reductive pathway possible, vinyl chloride may
accumulate
Vinyl chloride oxidized
Tolerated range


                                     14

-------
Table 2.  Contaminants and Daughter Products
 Analysis
                  Interpretation
 PCE
 TCE
 1,1,1 -Trichloroethane
 cis-DCE
 trans-DCE
 Vinyl Chloride
 Ethene
 Ethane
 Methane
 Chloride
 Carbon Dioxide
 Alkalinity
Material spilled
Material spilled or daughter product of perchlqroethylene
Material spilled
Daughter product of triehloroethylene
Daughter product of triehloroethylene
Daughter product of dichloroelhylenes
Daughter product of vinyl chloride
Daughter product of ethene
Ultimate reductive daughter product
Daughter product of organic chlorine
Ultimate oxidative daughter product
Results from interaction of carbon dioxide with aquifer
minerals
                                          15

-------
Table 3. Concentrations of TCE, Benzene, and Toluene in the Tibbetts Road Microcosms
Compound
TCE


Mean ± Standard
Deviation
Benzene


Mean ± Standard
Deviation
Toluene


Mean ± Standard
Deviation
Time Zero
Microcosms
328
261
309
299 ± 34.5
366
280
340
329 ±44.1
443
342
411
399 ± 51.6
Time Zero
Controls
337
394
367
366 ± 28.5
396
462
433
430 ± 33.1
460
557
502
506 ± 48.6
Week 23
Microcosms
1
12.5
8.46
7.32 ± 5.83
201
276
22.8
167 ± 130
228
304
19.9
184± 147
Week 23
Controls
180
116
99.9
132 ±42.4
236
180
152
189 ± 42.8
254
185
157
199 ± 49.9
Week 42
Microcosms
2
-2
2
2.0 ± 0.0
11.1
20.5
11.6
14.4 ± 5.29
2
2.5
16.6
7.03 ± 8.29
Week 42
Controls
36.3
54.5
42.3
44.4 ± 9.27
146
105
139
130 ±21.9
136
92
115
114 ± 22.0
                                      16

-------
Table 4,  First-order Rate Constants for Removal TCE, Benzene, and  Toluene  in the
Tibbetts Road Microcosms.
Parameter

TCE
95% Confidence Interval
Minimum Rate Significant at 95% Confidence
Benzene
95% Confidence Interval
Minimum Rate Significant at 95% Confidence
Toluene
95% Confidence Interval
Minimum Rate Significant at 95% Confidence
Living
Microcosms
Autoclaved
Controls
Removal
Above
Controls
First-order Rate of Removal (per year)
6.31
±2.50

3.87
± 1.96

5.49
±2,87

2.62
± 0.50

1.51
± 0.44

1.86
±0.45

3.69
±2.31
1.38
2.36
• ± 1.83
0.53
3.63
±2.64
0.99
                                      17

-------
Table 5. Concentration of Contaminants and Metabolic By-products in Monitoring Wells
along Segments in the Plume used to Estimate Field-scale Rate Constants.
Parameter
Monitoring well


TCE
dj-DCE
trans-DCE
1,1-DCE
Vinyl Chloride
Ethene
Benzene
Toluene
o-Xylene
m-Xylene
p-Xylene
Ethyl benzene
Methane
Iron
Segment A
SOS
Up
Gradient

200
740
0.41
0.99
<1
<4
510
10000
1400
2500
1400
1300
353

79S
Down
Gradient
Segment B
70S
Up
Gradient
52S
Down
Gradient
Segment C
70S
tJp
Gradient

......... 	 . — —-^ug/mcrj— — • — —---—-•-— — — — .
13.7
10.9
<1
<1
<1
<4
2.5
<1
8.4
<1
22
0,7
77

710
220
0.8
<1
<1
7
493
3850
240
360
1100
760
8

67
270
0.3
1.6
<1
<4
420
900
71
59
320
310
3

710
220
0.8
<1
<1
7
493
3850
240
360
1100
760
8

53S
Down
Gradient

3.1
2.9
<1
, <1
<1
<4
<1
<1
<\
<1
<1
<1
<2
27000
                                       18

-------
Table 6.  First-order Rate Constants in Segments of the Tibbetts Road Plume.
Flow Path Segments in Length and Time of Ground-water Travel

Compound
TCE
cis-DCE
Benzene
Toluene
o-Xylene
/n-Xylene
p-Xylene
Ethylbenzene
Segment A
130 feet = 6.5 years
Segment B
80 feet = 4.0 years
Segment C
200 feet .= 1 0 years
First-order Rate Constants in Segments ( per year)
0.41
0.65
0.82
>1.42
0.79
>1.20
0.64
1.16
0.59
produced
0.04
0.36
0.30
0.45
0.31
0.22
0.54
0.43
>0.62
>0.83
>0.55
>0.59
>0.70
>0.66
                                       19

-------
Table 7. Comparison of First-order Rate Constants in a Microcosm Study, and in the Field at the
Tibbetts Road NPL Site.
Parameter


Trichloroethylene
Benzene
Toluene
Microcosms Corrected for
Controls
Average
Rate
Minimum Rate
Significant at 95%
Confidence
Field Scale
Segment A
*s
Segment B


3.69
2.36
3.63
1.38
0.53
0.99
0.41
0.82
>1.42
0.59
0.04
0.36
Segment C

0.54
>0.62
>0.83
                                          20

-------
Figure Titles









Figure 1.  TCE Concentrations in the Tibbetts Road Microcosm Study.




Figure 2.  Benzene Concentrations in the Tibbetts Road Microcosm Study.^




Figure 3.  Toluene Concentrations in the Tibbetts Road Microcosm Study.




Figure 4. Location of Waste Piles and Flow Path Segments at the Tibbetts Road Superfund Site.
                                          21

-------
                    DISCLAIMER
The U.S. Environmental Protection Agency through it's
Office of Research and Development partially funded and
collaborated in the research described here.  It has been
subjected to the Agency's peer review and has been approved for
publication in an EPA document.
                                                            /
                                                           L

-------
1000
 100
                    10      15     20      25      30      35     40      45
                                                                              o TCE Microcosm
                                                                              • TCE Control
   Figure 1. TCE Concentrations in the Tibbetts Road Microcosm Study.

-------
   1000
    100
O)
3
                      10      15      20     25      30

                                   Time  (Weeks)
35
40
                   a Benzene Microcosm

                   • Benzene Control
45
           Figure 2.  Benzene Concentrations in the Tibbetts Road Microcosm Study.

-------
                   1000
1  \
                    100
                O)
                3
                        0
10     15      20     25      30

             Time  (Weeks)
35
                                                        a Toluene Microcosm

                                                        • Toluene Control
40
45
                           Figure 3. Toluene Concentrations in the Tibbetts Road Microcosm Study.

-------
Waste Pile

Ground Water
Flow Segment

-------
                                        TECHNICAL REPORT DATA
                                (Please read Imtrucitons en the reverse before cemplciingf
 1. REPORT NO.
       EPA/600/A-96/093
2.
 4..TIT,LI
   DESIGN AND INTERPRETATION OF MICROCOSM  STUDIES  FOR
   CHLORINATED COMPOUNDS
                                  5. REPORT DATE
                                  5  PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
1)BARBARA H.  WILSON AND JOHN T. WILSON,  (2) DARRYL LUCE
                                                                   3. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
DU.S.  EPA, NRMRL, SPRD         (2)REGION  1
  P.O.  BOX  1198                   U.S. EPA
  ADA,  OKLAHOMA 74820             BOSTON, MASSACHUSETTS
                                                                    IQ. PROGRAM ELEMENT NO,
                                  1 I. CONTRACT/GRANT NO.

                                    IN-HOUSE RPJW9
 12. SPONSORING AGENCY NAME ANO ADDRESS
  U.S.  EPA, NRMRL, SPRD
  P.O.  BOX 1198
  ADA,  OKLAHOMA 74820
                                                                    13. TYPE Of REPORT ANO PERIOD COVERED
                                                                      BOOK CHAPTER	
                                  14. SPONSORING AGENCY CODE
                                                                      EPA/600A
15. SUPPLEMENTARY NQT6S
16. ABSTRACT
       Table 7 compares the first-order rate constants estimated from the microcosm studies to the rate
       constants estimated at field scale.  The agreement between the independent estimates of rate is
       good; indicating that the rates can appropriately be used  in a risk assessment.  The  rates of
       biodegradation documented in the microcosm study could easily account for the disappearance
       of trichloroethylene, benzene, and  toluene observed at field scale. The rates estimated from the
       microcosm study are several-fold higher than the rates estimated at field scale.  This may reflect
       an underestimation of the true rate in the field.  The estimates of plume velocity assumed that
       the aquifer was homogeneous. No attempt was made in this study to correct the estimate of
       plume velocity for the influence of preferential flow paths.  Preferential flow paths with a higher
       hydraulic conductivity than average would result in a faster velocity of the plume, thus a lower
       residence time and faster rate of removal at  field scale.
                                    KEY WORDS ANO DOCUMENT ANALYSIS
a. DESCRIPTORS

is. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
s
b. IDENTIFIERS/OPEN ENDED TERMS

19. SECURITY CLASS iTIns Krpnrti
UNCLASSIFIED
20. SECURITY CLASS fTliit page:
UNCLASSIFIED
C COS ATI FieiU, CfOup

21 NO. OP "AGES
25
22. <"*iCS
    Form 2220—1 (R«-». 4—77)    PREVIOUS EDITION is oasoi_£Te

-------