United States
               Environmental Protection
               Agency
             Sediment Issue
Determination  of Rates and Extent of
Dechlorination in PCB-Contaminated
Sediments During  Monitored Natural
Recovery
  Challenges

  Case Study
    Lake Hartwell
    Superfund Site
Purpose

The National Risk Management Research Lab-
oratory (NRMRL) of the U.S. Environmental
Protection Agency (U.S. EPA) is developing
effective, inexpensive remediation strategies
for contaminated sediments. This program
theme includes the study and development of
field monitoring tools to evaluate Monitored
Natural Recovery (MNR) processes in aquatic
sediments (1). Previous researchers have dem-
onstrated the potential for PCB dechlorination
in sediments using other evaluation methods,
such as chiral chemistry (2) and laboratory mi-
crocosms (3).  In contrast, the approach taken
Lake Hartwell
                                    by NRMRL involved characterization of dechlorination in sediments
                                    using PCB congener fingerprinting and polytopic vector analysis
                                    (PVA).
NRMRL conducted studies to evaluate the long-term recovery of
PCB-contaminated sediments via reductive dechlorination, including
the magnitude, extent, and rates of ortho and meta plus para dechlo-
rination reactions with sediment depth and time (4, 5). The informa-
tion learned is summarized in this Sediment Issue, which is intended to
be used as a reference for site managers and U.S. EPA decision makers
who may be considering MNR as a contaminated sediments manage-
ment strategy. The data summarized in this Sediment Issue were gener-
ated during field studies conducted by NRMRL in cooperation with
U.S. EPA Region 4 at the Sangamo-Weston/Twelvemile Creek/Lake
Hartwell Superfund Site in Pickens County, SC.

Introduction

In aquatic environments affected by contaminated sediments, risk
management strategies focus on either removing the contaminated
material or interrupting exposure pathways by which contaminants
might pose an ecological or human health risk over time. This is
generally achieved by dredging, capping, or MNR. MNR relies on
naturally occuring processes to reduce risk to humans and/or ecologi-
cal receptors, including physical, biological, and chemical mechanisms

-------
that act together to contain, destroy, or otherwise
decrease the bioavailability or toxicity of contaminants in
sediment (1).

In the case of PCB-contaminated sediments, where buried
sediments are anaerobic and ample organic carbon is pres-
ent, PCBs may undergo metabolic transformation through
a process known as reductive dechlorination (6-8).  This
metabolic transformation is mediated by bacteria, result-
ing in a hydrogen substitution for a chlorine on the PCB
molecule. Generally, higher-chlorinated PCBs are more
susceptible to transformation via reductive dechlorination,
resulting in the accumulation of lower-chlorinated PCBs (9).

Reductive dechlorination preferentially removes chlorines
from the meta and para positions of PCBs (6), shown in
Figure 1, which has been shown to lead to the conservation
of biphenyl rings and ortho chlorines.  Dechlorination of
meta and para chlorines can result in relative detoxifica-
tion through elimination of coplanar-like congeners and
aryl receptor-mediated toxicity (10, 11) and through the
transformation of generally more toxic, higher-chlorinated
congeners to generally less toxic,  though more mobile,
lower-chlorinated congeners (12).

Dechlorination patterns can be evaluated  by measuring
historical dechlorination patterns in vertical sediment
core profiles and by multivariate receptor  modeling, such
as PVA. PVA, a  sophisticated statistical procedure, is a
self-training receptor/mixing model that separates complex
mixtures into several contributing patterns and determines
their contributions to each sample (13).  PVA has been
used increasingly for the characterization of PCB sources
and to determine the transformation of PCBs  in complex
environmental settings (7-9, 14-16).

Methods
Identification and Characterization of Weathered and
Dechlorinated PCB Congener Patterns.  Detailed PCB
analysis of 107 congeners (PCB molecules) was used to
measure historical dechlorination patterns in vertical sedi-
ment core profiles. Assuming the PCB source remains
unchanged,  congener shifts (patterns showing an accumu-
Figwe 1. Meta (3,3',5,5') and Para (4,4') Positions of PCB Molecule.
lation of lower-chlorinated congeners and a corresponding
decrease in higher-chlorinated congeners) are characterized
by comparing surface sediment congener distributions with
the distributions in buried sediments.

Congener-specific PCB data are evaluated using PVA. The
use of a multivariate, variance-based technique, such as
PVA, offers broad insight into mixing systems at a very
high level of detail (sample-level resolution) using only data
concerning the response variables of a system (13).  PVA
provides estimated compositions of contributing PCB fin-
gerprints (end-members) directly from the analysis of the
ambient data. The end-member (EM) refers to the original
Aroclor formulation that the PVA analysis has determined
and points to as being the original material that caused the
contamination. In other words, PVA analysis  is applied
to a PCB data set that is comprised of an environmen-
tally altered PCB  congener composition (e.g.,  weathered,
degraded), and the result of the analysis is an identification
of the original Aroclor(s) that the contamination came
from some time ago. PVA also provides estimates of the
relative contribution of each EM in each sample. Only
after the EM  patterns are resolved are they compared to
literature-reported source and alteration patterns (e.g., Aro-
clor compositions, PCB dechlorination patterns, and other
weathering patterns).  Using an exploratory approach with
minimal a priori assumptions helps ensure  that unknown
or unforeseen PCB source or alteration mechanisms are
not overlooked. Use of PVA, in conjunction with other
forensic analysis approaches, can be important for  under-
standing long-term fate and formulating efficient remedia-
tion strategies.

Determination of Dechlorination Rates.  Sediment age
dating, using lead-210 (210Pb) and cesium-137 (137Cs),
and linear relationships between meta plus para chlorine
concentrations and sediment age are used to determine
dechlorination rates (4).  These rates are compared with
literature-reported rates and to rates calculated using this
equation by Zwiernik et al. (17):

MDR= 1.16Ce + 6.37  (r2 = 0.960)

where MDR  (nmol Cl'/g sediment/week) is the maximum
dechlorination rate and Ce (ug/L) is the aqueous phase
PCB concentration.  The aqueous-phase PCB  concentra-
tions were estimated  based on equilibrium-partitioning  co-
efficients,  fraction of organic carbon, and total PCB
(t-PCB) concentrations.  The magnitude and extent of
PCB dechlorination is measured by plotting ortho chlo-
rines and meta plus para chlorines per biphenyl molecule
(i.e., mole chlorine/mole PCB) with sediment depth.
Linear regression analysis is used to determine the rate of
meta plus para removals over time (as will be shown in
later examples in this Sediment Issue).

-------
Challenges

Challenges associated with using PCB dechlorination and
the above methods to assess the extent of natural recovery
include:

• Direct Access to Reference Data Sets — Direct compari-
  son of the congener patterns observed in sediments under
  study to known or historic dechlorination patterns can
  greatly facilitate interpretation; however, raw historic data
  often are unavailable.

• Assuming Relativeness of Surface Sediment Distributions
  to the PCB Source — It is assumed that surface sediments
  are characteristic of the PCB source; however, if historical
  data from surface sediments are unavailable, this assump-
  tion may be impossible to prove.

• Consistency of Dechlorination — Dechlorination may
  be very extensive in most of the cores, but it may not be
  consistent from core to core or at various depth intervals
  within a single  core.
• Dechlorination Rates — Rates of dechlorination are de-
 rived from the laboratory and may be higher than
 dechlorination rates measured directly in the field.

•Analytical Limitations — Chemical analysis may result
 in the coelution of certain PCBs, making it impossible
 to calculate the numbers of meta and para chlorines per
 biphenyl molecule separately. In order to address
 coeluting congeners in this study, the meta and para
 chlorines were summed  so that the magnitude and
 extent of PCB dechlorination were measured using
 ortho chlorines and meta plus para chlorines per
 biphenyl molecule (i.e.,  moles chlorine/mole
 PCB). However, as subsequently noted later
 in this Sediment Issue, three factors were
 considered before summing the meta and para
 chlorines to calculate dechlorination rates: 1) the
 uniformity of source material in the sediments,
 2) the conservation of ortho chlorines with sediment
 depth and time, and 3) the linear relationship between
 the meta and para chlorines and time.

Case Study

Lake Hartwell Superfund Site
The methods described were conducted as part of an
evaluation of the recovery of PCB-contaminated lake
sediments at the Sangamo-Weston/Twelvemile Creek/
Lake Hartwell Superfund site (Pickens County, SC),
shown in Figure 2. The Sangamo-Weston plant manu-
factured capacitors from 1955 to 1978  using dielectric
fluids containing Aroclors 1016, 1242, and 1254.
   Though the Sangamo-Weston plant no longer exists, waste
   disposal practices dating back to earlier operations led to
   PCB-contaminated sediments in Lake Hartwell, down-
   stream of the Sangamo-Weston plant (18).

   A total of 280 samples from 18 sediment cores and surface
   sediment samples were collected during two annual events
   conducted in 2000 and 2001. Sediment sample locations
   are shown in Figure 2 as alpha-numeric transects. Sediment
   samples were analyzed for 107 PCB congeners, particle size
   distribution, total organic carbon (TOC), moisture con-
   tent, and radio isotopes (210Pb and 137Cs).

   1.  PCB Dechlorination: End-Member Characterization.
   PCB congener fingerprinting and PVA were used for iden-
   tification and characterization of weathered and dechlori-
   nated PCB congener patterns.

   Congener shifts were characterized  by comparing surface
   sediment congener distributions with the distributions in
                               Sangamo-Weston
                                       Plai
                                        Sediment
                                     Impoundments
                                    10 Transect
                                     Locations
                                     Lake Hartwell
Figure 2. Location Map of Lake Hartwell Transects. Lake Hartwell
sediment cores were collected at 10 transect locations. Surface sediment
samples were collected in Twelvemile Creek and in Town Creek above and
below the former Sangamo-Weston plant. Reprinted with permission from
(5) . Copyright 2005, American Chemical Society.

-------
 buried sediments and then matched with literature-report-
 ed dechlorination patterns from other sites.  Examples of
 congener distributions (mole percent) are shown in Figure
 3  for surface and buried samples from the Transect L core
 collected in 2000.  Congeners are plotted as they appear
 chromatographically, from left to right and generally from
 low to high molecular weight (from lower-chlorinated to
 higher-chlorinated congeners).  Figure 3-c illustrates the
 net congener shifts from high to low chlorinated con-
 geners due primarily to dechlorination, although other
 weathering mechanisms such as biotransformation, sorp-
 tion, and benthic mixing also may have been involved.
 Negative concentrations in Figure 3-c represent a net
 concentration loss, while positive concentrations repre-
 sent a net concentration gain.  Comparison of the surface
 and buried congener distributions in the Transect L core
 reveals a congener shift from higher- to lower-chlorinated
 congeners with sediment depth and age.
PVA results were used to determine PCB source and
dechlorination EM patterns and their distributions in each
of the cores.  Analysis of PVA goodness-of-fit diagnostics
indicated a four end-member model.

Figure 4 shows the congener profile of the first end
member (EM-1) as an example. The similarities between
congener profiles were evaluated by calculating the cosine
theta (cos 0) values (19). The cos 0 value represents the co-
sine of the angle between two multivariate vectors, in this
case the two matrices formed by the congeners used for
each respective sample.  A cos 0 value of 0 would indicate
two completely dissimilar, orthogonal vectors, and a cos
0 value of 1.0 would indicate two identical vectors. The
cos 0 values were calculated for various Aroclor mixtures
compared to EM-1, including mixtures of Aroclors 1016,
1242,  1248, and 1254.  Comparison of EM-1 with the
50/50 Aroclor 1248/1254 mixture shown in Figure 4-b
b.U
.0
4.0
.0
.0
1.0 -




*\

T"'


:'|




I


n






L


;?:
























T

ll



1^








rr:









77






i:

|
||





-.













|










Pi
1
iLiiJiii


_






gr







F

ill




i


_








0
r-n
1


-5 (




:n

..


n




ir


:;:

iterval (a)



o ..„ „
Congener Distributi
_k _
> to ^ o> n o K
b b b b b c
b U.U "
0
5 6.0
4.0
2.0 -|?
-2.0 -
-4.0
c n

IN
-:-
:i:
!|
1 illl ^L..

j|! *
5.4
1
i
-n i
mm \\
m

0.
3%
^

35-40 cm interval (b)
|li|_pii[_|is|| p^ pi [ii|iii|ii|ii| n-R PI— n r-n-r-

7-
Difference of (0-5 cm interval) and (35-40 cm interval) (c)
:
! jij
drim~p""i"iTr H !rf^
1 u l!;r
1
                                            ^rOOOOOiKoooSuoooc_>F^C)C)C>c>oSCI:lcacc:lcDIrlcIlcI3crlcnQDL?>D:ICDSCDIIlcI3IIlcIlcIlcD CQ^HCDCDCD^:
                                            SS£IKImffffSffiKSlEIgEsa:a:a.g^oooooooooS;ooSooooooooSooo^
                                            ^L       S^L                             "Ss          "5LS
Figure 3. Congener Distributions Showing Dechlorination Characteristics for Lake Hartwell Sediments. Panel (a) shows Transect L (year
2000) surface sediment (0-5 cm) congener distribution; panel (b) shows the congener distribution in the 35-40 cm interval of the year 2000
Transect L core; panel (c) shows the net dechlorination in the 35-40 cm interval by subtracting the congener distribution in panel b from
panel a. Reprinted with permission from (4). Copyright 2005, American Chemical Society.

-------
resulted in the highest cos 0 value of 0.93- Comparison of
EM-1 with the 50/50 Aroclor 1242/1254 mixture shown
in Figure 4-c resulted in a cos 0 value of 0.83- Both
values are close to 1.0 and suggest strong  similarities be-
tween these two Aroclor mixtures and EM-1. Congener
profiles for EM-2 through EM-4 are provided by Magar et
d. (5).

2. PCB Dechlorination: Rates and Extent. An evalu-
ation was conducted to determine the extent of reductive
dechlorination by the preferential loss of meta and para
chlorines, the conservation of ortho chlorines with sedi-
ment depth and age, and the historical transformation of
higher-chlorinated PCB congeners (congeners with four or
more chlorines) to mono-, di-, and trichlorobiphenyl con-
geners with sediment depth and time.  The magnitude and
extent of PCB dechlorination was measured by plotting
ortho chlorines and meta plus para chlorines per biphenyl
molecule (i.e., moles chlorine/mole PCB) with sediment
                                                    depth, as shown in Figure 5- The loss of meta and para
                                                    chlorines and the  conservation of ortho chlorines with
                                                    sediment depth suggests that the PCBs in the sediments
                                                    underwent reductive dechlorination after burial.

                                                    Dechlorination rates were determined on a sample-by-
                                                    sample basis for the Lake Hartwell site by plotting the
                                                    numbers of meta plus para chlorines per biphenyl molecule
                                                    (moles chlorine/mole PCB) with sediment age.  Linear
                                                    regression was used to determine the rate of meta plus para
                                                    chlorine removals over time. Results of the linear regres-
                                                    sion analyses and  corresponding meta plus para dechlori-
                                                    nation rates measured for 2000 and 2001  cores  are shown
                                                    in Table 1. Three factors made it possible to calculate
                                                    meta plus para dechlorination rates in this study: 1)  the
                                                    source material was relatively uniform throughout the Lake
                                                    Hartwell surface sediments, 2)  ortho chlorines were  well
                                                    conserved with sediment depth and time (the number of
                                                    ortho chlorines did not change with depth since the input
                                                    composition did not change with time), and 3)  the linear
    10.0
     8.0
     6.0
#
 5
 +•»
 .c
 «
 0>
 0)
 O)
 o
u
 w
_rc
 O
 8.0 '

 6.0

 4.0

 2.0 H
 o.o
10.0

 8.0
                                                                                        Generated by  (a)
                                                                                        the PVA model
                                        jaHi
I
JO
niliiL   Hii
                                                                   lii
                                            50/50 Arochlor (b)
                                            1248/1254 mix
JLJL._
     0.0
                                                                                            50/50 Arochlor (c)
                                                                                            1242/1254 mix
                                                                                          *- 
-------
               1    1.5     2
               mol Chtorine.'mol PCB
  1      1.5
molCniorine.'molPCB
Figure 5. Numbets of Ottho (•) and Meta Plus Pata (A) Chlorines pel- Biphenyl Mol-
ecule (Moles Chlorine/Mole PCB) with Sediment Depth for Sediment Cotes Collected in
2001. Figures include (a) Cotes T-OA, (b) T-OB, (c) T-OC, and (d) T-LC. Adapted from
(4). Copyright 2005, American Chemical Society.
Table 1.  Measuted Dechlotination Rates Using Moles of Meta Plus
Pata Chlorines Ovet Measuted Time Intervals fot Buried Sediments.
relationship observed between the num-
ber of meta plus para chlorines and time.

For comparison to rates calculated using
the equation by Zwiernik et al. (17), the
average dechlorination rates for the 2000
and 2001 transect cores (i.e., 0.0403 and
0.0266 mole ClVmole PCB/yr, respec-
tively, from Table 1) correspond to 2.9 x
10'6 mole ClVg PCB/week and 2.2 x 10'6
mole Cl'/g PCB/week, assuming 261 g/
mole for Aroclor 1242 (the MDR equa-
tion was calculated using experiments
with Aroclor 1242) and 52 weeks per
year. Thus, average dechlorination rates
assuming a sediment PCB concentration
of 60 mg/kg (60,000 nanograms/gram)
equates to 0.174 and 0.132 nmole ClYg
sediment/week for the 2000 and 2001
rates, respectively.

An inverse of the dechlorination rate
indicates the time required for chlorine
removal.  Thus, for the average dechlo-
rination rate for the 2001 transect cores,
the amount of time for chlorine removal
would be about 33 years.

Conclusions

At PCB-contaminated sediment sites,
dechlorination may be an important
component of the natural recovery of the
sediments when anaerobic conditions
Cote

T-O
T-N
T-L
T-I
T-T6
Avetage

T-OC
T-LA
T-LB
T-IA
T-IB
Avetage
Depth
Range (cm)

0-
0-
0-
0-
0-


20-
15-
0-
0-
0-


40
40
35
60
20


100
-35
60
20
15

Date Range"
(yeat)

1981 -
1958-
1979-
1980-
1960-


1952-
1962-
1987-
1993-
1996-


1999
1997
1999
1999
1998


2000
1984
2001
2000
2000

No. of
Samples
2000 Transect Cores1
8
8
7
11
4

2001 Transect Coref
11
4
7
4
3

Petcent Dectease
(mole Cl /mole PCB/yt)
(95% Confidence Limits)

6.03 (6.01,
3.21 (3.20,
5.76 (5.73,
6.25 (6.22,
2.81 (2.80,
4.03 (4.02,

1.52(1.50,
6.33 (6.29,
8.33 (8.31,
14.1 (14.1,
16.4(16.1,
2.66 (2.65,

6.05)
3.23)
5.79)
6.27)
2.82)
4.04)

1.53)
6.37)
8.35)
14.1)
16.6)
2.68)
Cottelation
Coefficient

0.92
0.85
0.89
0.88
0.97
0.79

0.68
0.94
0.97
0.99
0.85
0.52
 a Dates were determined using 210Pb and 137Cs age dating techniques.

 b The cores for the year 2000 were collected at the centerline of the 10 transects shown in Figure 2, i.e., one core to the depth indicated per
 transect.

 c The cores for the year 2001 were collected laterally along (or across) three of the 10 transects shown in Figure 2. Three cores were collected to
  depth across Transect O (designated A, B, and C), three across Transect L (again designated A, B, and C), and two across Transect I (designated
  AandB).

-------
exist and the source(s) of PCBs in sediment remain
unchanged.  Under anaerobic conditions, the primary
metabolic pathway for PCBs is reductive dechlorina-
tion in which  chlorine removal and substitution with
hydrogen by bacteria result in a reduced organic com-
pound with  fewer chlorine molecules (6-8).

Reductive dechlorination of PCBs preferentially
removes chlorines from the meta and para positions
(6), which has been shown to lead to the conservation
of biphenyl  rings and ortho chlorines in laboratory
dechlorinating enrichment cultures (14, 15).  Gener-
ally, higher-chlorinated biphenyls are preferentially
dechlorinated over lower-chlorinated congeners resul-
ting in the accumulation of mono-, di-, and trichlo-
robiphenyls  (9).  As such, reductive dechlorination
can be identified by the historical transformation of
higher-chlorinated PCB congeners with sediment
depth and time, and through the preferential  loss of
meta and para chlorines and the conservation of ortho
chlorines.  Determination of dechlorination rates is
useful for predicting the number of years for chlorine
loss to occur (i.e., for predicting when positive  impacts
of dechlorination could be observed).

PCB congener fingerprinting and multivariate receptor
modeling such as PVA can be used as exploratory data
analysis tools to characterize PCB sources and altera-
tion patterns.  PCB sources and alteration patterns
are determined by comparing PVA EM patterns with
known source patterns (i.e., Aroclors or Aroclor mix-
tures) and literature-reported alteration patterns.  PVA
also is used to characterize the  vertical and lateral dis-
tributions of PCB source and dechlorination patterns.

Utilization of PCB dechlorination to assess  the extent
of natural recovery has proven to be a valuable tool;
however, challenges associated with this tool may be
encountered.  Some of these challenges have been
summarized herein to enable site managers to account
for them in  their research design and interpretation of
results.

References

(1)   United States Environmental Protection Agency. 2005-  Contaminat-
     ed Sediment Remediation Guidance for Hazardous Waste Sites, OSWER
     9355.0-85, EPA540/R05/012. December, http://www.epa.gov/
     Superfund/resouces/sediment/pdfs/guidance/pdf.
(2)   Wong, C.S., A.W. Garrison, and W.T. Foreman. 2001. Enantiomeric
     Composition  of Chiral Polychlorinated Biphenyl Atropisomers  in
     Aquatic Bed Sediment. Environ. Sci. TechnoL, 35 (1): 33-39.
(3)   Pakdeesusuk,  U., WJ. Jones, CM. Lee, A.W. Garrison, W.L. O'Niell,
     D.L. Freedman, J.T. Coates, and C.S. Wong. 2003- Changes in
    Enantiomeric Fractions during Microbial Reductive Dechlorination of
    PCB132, PCB149, and Aroclor 1254 in Lake Hartwell Sediment Micro-
    cosms. Environ. Sci.  TectmoL, 37 (6): 1100-1107.

(4)  Magar, V.S, R.C. Brenner, G.W.Johnson, andJ.F. Quensen, III. 2005-
    Long-Term Recovery of PCB-Contaminated Sediments at the Lake
    Hartwell Superfund Site:  PCB Dechlorination.  2. Rates and Extent. Envi-
    ron. Sci. Technol.,39 (10): 3548-3554.

(5)  Magar, V.S., G.W.Johnson, R.C. Brenner, J.F. Quensen, III, E.A. Foote, G.
    Durell, J.A. Ickes, and  C. Peven-McCarthy.  2005- Long-Term Recovery of
    PCB-Contaminated Sediments at the Lake Hartwell Superfund Site: PCB
    Dechlorination.  1. End-Member Characterization. Environ. Sci. TechnoL,
    35»(10): 3538-3547.

(6)  Bedard, D.L. and J.F. Quensen, III.  1995-  Microbial Reductive Dechlo-
    rination of Polychlorinated Biphenyls. In: Microbial Transformation and
    Degradation of Toxic Organic Chemicals., pp.  127-216, L.Y. Young and C.
    Cerniglia (Eds.). Wiley-Liss Division, John Wiley & Sons, Inc.; New York,
    NY.

(7)  Mohn, WW, and J.M. Tiedje. 1992. Microbial Reductive Dechlorina-
    tion. Microbiol. Rev., 6: 482-507.

(8)  Magar, V.S. 2003-  PCB  Treatment Alternatives and Research Directions.
    / Environ. Eng., 129 (11): 961-965.

(9)  Quensen, J.F., III, and J.M. Tiedje. 1997- Evaluation of PCB Dechlorina-
    tion in Sediments.  In:  Methods in Biotechnology., Bioremediation Protocols,
    pp. 257-273, D. Sheehan (Ed.).  Humana Press, Inc.;  Totowa, NJ.

(10) McFarland, V andj. Clarke. 1989-  Environmental Occurrence, Abun-
    dance, and Potential Toxicity of Polychlorinated Biphenyl
    Congeners: Considerations for a Congener-Specific Analysis. Environ.
    Health Perspect., 81:  225-239.

(11) Quensen, J.F., M.A. Mousa, S.A. Boyd, J.T. Sanderson, K.I. Froese, and
    J.P. Giesy.  1998.  Reduction of Ah Receptor Mediated Activity of PCB
    Mixtures Due to Anaerobic Microbial Dechlorination.  Environ. Toxicol.
    Chem.,17(5}: 806-813.

(12) Agency for Toxic Substances and Disease Registry (ATSDR). 2000. Toxico-
    logical Profile for Polychlorinated Biphenyls (PCBs). Division of Toxicology,
    Toxicology Information Branch; Atlanta, GA.

(13) Barabaas, N., P. Adriaens, and P. Goovaerts. 2004. Modified Polytopic
    Vector Analysis to Identify and Quantify a Dioxin Dechlorination Signature
    in Sediments. 1. Theory.  Environ. Sci. TechnoL.,38 (6): 1813-1820.

(14) Tiedje, J.M., J.F. Quensen, III, J. Chee-Sanford, J.P. Schimel, and S.A.
    Boyd. 1993- Microbial Reductive Dechlorination of PCBs. Biodegrada-
    tion, 4\ 231-240.

(15) Quensen, J.F., III, J.M. Tiedje, and S.A. Boyd.  1990. Dechlorination of
    Four Commercial Polychlorinated Biphenyl Mixtures (Aroclors) by Micro-
    organisms from Sediments.  Appl. Environ. Microbiol., 56: 2360-2369.

(16) van Dort, H.M. and D.L. Bedard. 1991. Reductive Ortho and Meta
    Dechlorination of a  Polychlorinated Biphenyl Congener by Anaerobic
    Microorganisms. Appl. Environ. Microbiol.., 57: 1576-1578.

(17) Zwiernik, M.J., J.F.  Quensen, III, and S.A. Boyd. 1999. Residual
    Petroleum in Sediments Reduces the Bioavailability and Rate of Reductive
    Dechlorination of Aroclor 1242. Environ. Sci. Technol.,33 (20): 3574-3578.

(18) United States Environmental Protection Agency. 1994. Superfund Record
    of Decision: Sangamo-Weston/Twelvemile Creek/Lake Hartwell Site, Pickens,
    GA: Operable Unit 2. EPA/ROD/R04-94/178.

(19) Davis, J.C. 1986.  Statistics and Data Analysis in Geology. John Wiley &
    Sons, Inc.; New York, NY.

-------
United States
Environmental Protection
Agency

Office of Research and Development
National Risk Mangement
  Research Laboratory
Cincinnati, OH 45268

Official Business
Penalty for Private Use
$300

EPA/600/S-08/012
August 2008
www.epa.gov
PRESORTED STANDARD
 POSTAGE & FEES PAID
         EPA
   PERMIT No. G-35
                                           Recycled/Recyclable
                                           Printed with vegetable-based ink on
                                           paper that contains a minumum of
                                           50% post-consumer fiber content
                                           processed chlorine free

-------