&EPA
       United States
       Environmental Protection
       Agency
          Great Lakes National Program Office
          77, West Jackson Boulevard
          Chicago, Illinois 60604
EPA905-R96-012«X
   July 1996
Assessment and
Remediation
of Contaminated Sediments
(ARCS) Program

BENCH-SCALE  EVALUATION OF
BIOREMEDIATION FOR THE TREAT-
MENT OF SEDIMENTS FROM THE
ASHTABULA, BUFFALO, SAGINAW,
AND SHEBOYGAN RIVERS
                           ® United States Areas of Concern

                           • ARCS Priority Areas of Concern

-------
Bench-Scale  Evaluation of Bioremediation for the Treatment  of
 Sediments from the Ashtabula,  Buffalo, Saginaw  and  Sheboygan
                             Rivers
                          Final Report
                          Prepared by
                         W. Jack Jones
                        Rochelle Araujo
                         John E.  Rogers
                 Ecosystems Research Division
             National Exposure Research Laboratory
        United States Environmental  Protection Agency
                       Athens,  Georgia
                            For the

 Assessment and Remediation of Contaminated Sediments  (ARCS)
                            Program
              Great Lakes National Program Office
             U.S.  Environmental Protection Agency
                       Chicago, Illinois.
                           U.S. Environmental Protection Agency
                           Region 5, Library (PL-12J)
                           77 West Jackson Boulevard, 12th Floor
                           Chicago, IL  60604-3590

-------
  Bench-Scale Evaluation of Bioremediation for the Treatment of
  Sediments  from the Ashtabula. Buffalo,.  Saginaw  and Sh
                             Rivers
                            ABSTRACT
     The potential of microbial degradation for the reduction of
organic contaminants was examined in sediments from four Areas of
Concern in the Great Lakes Region: the Ashtabula, Buffalo,
Saginaw and Sheboygan Rivers.  Only indigenous organisms were
tested in the bioremediation study; organisms from other sites or
genetically engineered bacteria were not employed.  The ability
of anaerobic bacteria to remove chloride ions (reductive
dechlorination) from polychlorinated biphenyls (PCBs)  was tested
at three of the four sites.  Separate tests were undertaken to
assess the ability of aerobic microorganisms to degrade poly-
nuclear aromatic hydrocarbons  (PAHs) in sediments from the
Buffalo River.
     Bench-scale studies were performed to determine the
dechlorinating potential of natural bacteria for
biotransformation of historically contaminated (PCB) sediments
and to assess enhancement of dechlorinating activity via nutrient
addition. PCB-contaminated sediments from the Ashtabula River, ^
the Saginaw River, and the Sheboygan River were tested for their
ability to dechlorinate "historical" PCBs and highly chlorinated
PCB congeners  (added to samples) under anaerobic conditions. PCB-
contaminated Ashtabula River sediments were spiked with
2,3,3',4,4'-pentachlorobiphenyl  (Penta-CB), 2,3,3',4,4',5-
hexachlorobiphenyl  (Hexa-CB),  and 2,2',3,4,5,6,6'-
heptachlorobiphenyl  (Hepta-CB).  Dechlorination of the added
Penta-CB, Hexa-CB, and Hepta-CB congeners was observed after
respective lag periods of 5, 3, and 4 months.  Significant
dechlorination occurred after  6 months when the sediment was
spiked with combinations of  either two or all three congeners.
The average number of chlorines per biphenyl decreased from 5.9
to 3.1 in Ashtabula River sediments amended with a combination of
all three congeners. PCB-contaminated sediments  from the  Saginaw
River were amended with Aroclor 1260  in the absence and presence
of inorganic nutrients  [revised anaerobic mineral medium  (RAMM)].
Sediments amended with Aroclor 1260 plus modified RAMM exhibited
an increase in the mole percent of mono-, di-, and tri-
chlorobiphenyl  (CB) homologs and  a decrease in the mole percent
of homologs with higher numbers of chlorine atoms per biphenyl
when compared to unamended controls.  In experiments amended with
Aroclor  1260 only, an increase was observed in the mole percent
of only  the mono- and di-CB  homologs/ the mole percent of the
remaining homolog groups decreased. PCB contaminated Sheboygan
River sediments were amended with  5,  10 and 20 mg/L of
2,2',3,3',4,5,6,6'-octachlorobiphenyl  (Octa-CB).  The percentages
of Octa-CB remaining in the  samples after anaerobic incubation
for 8 months ranged  from 10  to 35  percent.  A decrease in the
higher chlorinated  (5 to 8 chlorines  per biphenyl) homologs,

-------
 including the  added Octa-CB,  and a  corresponding increase  in the
 lower  chlorinated (1 to  3  chlorines per  biphenyl)  homologs,  was
 observed.  The  major products  of reductive  dechlorination of  Octa-
 CB  amended sediments were  di-CB congeners.
     Sediments from Buffalo River were found  to  be contaminated
 with PAHs at concentrations ranging from 0.37 mg/kg for fluorene
 and benz(b)fluoranthene  to greater  than  10  mg/kg for fluoranthene
 and pyrene.  The  indigenous microorganisms  from  Buffalo River
 were capable of quickly  degrading two- and  three-ring PAHs and
 slowly degrading  four-ring compounds, but showed little activity
 on  compounds with greater  than  four rings.  Each of the two- and
 three-ring compounds supported  growth of sediment  bacteria in
 mineral media;  the compounds  with four rings  and higher did  not
 support growth.
     An enrichment culture was  developed by exposing Buffalo
 River  sediments to each  compound in a mixture of 16 PAHs and
 combining the  cultures adapted  to each of the compounds to form
 an  enrichment  mixture with a  wide range  of degradative activity.
 Oxygen did not  limit degradation of PAHs by the  enrichment
 culture in sediment  slurry experiments.  Additions  of phosphate,
 either alone or with nitrogen,  enhanced  the rate of degradation
 as  measured by  the mineralization of pyrene;  additions of
 nitrogen  as ammonium nitrate  decreased the rate  slightly.  The
 addition  of organic  matter extracted from sediment  increased the
 initial rate of pyrene mineralization by supporting faster growth
 of  the degrader populations.  Trace  metals were  not  inhibitory to
 degradation of  PAHs  in aerated  systems.  The  presence of
 additional PAHs decreased  the rate  of mineralization  of pyrene;
 this inhibition could be reduced by  treating  the sediment
 slurries with hydrogen peroxide.
     The enrichment  culture, when added  back  to  non-sterile
 sediments  (bioaugmentation), did not enhance  degradation at
 inoculation densities below 106  cells/ml. Above that density,
 mineralization  of  pyrene was  first-order, indicating that growth
 of  the bacteria was  not needed to initiate degradative activity.
     Sediment had  both an  inhibitory and a stimulatory influence
 on  the rate of  degradation of pyrene.  The stimulatory effect
 could  be explained by the utilization of sediment organic carbon
 as  a growth substrate by the bacteria.    However, at sediment
 slurry concentrations above 5%,   the rate and  extent of
 degradation decreased with increasing sediment slurry
 concentrations due to sorption of the compound.
     Triton X-100, at concentrations both above  and below the
 critical micelle concentration  (CMC), inhibited  the
mineralization of pyrene when added during the growth phase  of
the organisms.   When the surfactant was  added after
mineralization of the available phase of the PAH,
 "solubilization" of the remaining PAH by the  surfactant resulted
in  slight increases in mineralization.

-------
                        Table of Contents


Section

Abstract  	 1

List of Figures   	v

List of Tables	V11:L

List of Abbreviations/Acronyms  	  ix

Acknowledgements  	 x

1.0  Introduction 	

     1.1  Background	;-
     1.2  Sediment Descriptions 	 3
          1.2.1  Site Names and Locations	3
          1.2.2  Sediment Acquisition and Homogenization   ... 4
     1.3  Background on Bioremediation Technology  	  10
          1.3.1  PCB Biodegradation	1:L
          1.3.2  PAH Biodegradation	12

2.0  Treatability Study Approach   	  16

     2.1  Study Objectives and Rationale   	  16
          2.1.1  PCB Biodegradation	16
          2.1.2  PAH Biodegradation	16
     2.2  Experimental Design  	  17
          2.2.1  PCB Biodegradation	17
          2.2.2  PAH Biodegradation	17
     2.3  Experimental Methods   	  19
          2.3.1  PCB Biodegradation	19
          2.3.2  PAH Biodegradation	20

3.0  Results  and Discussion	24

     3.1  Summary of PCB  Biodegradation  Results  	  24
          3.1.1  Bench Scale Experiments with Sheboygan
                 River Sediments	24
          3.1.2  Bench Scale Experiments with Ashtabula
                 River Sediments	28
          3.1.3  Bench Scale Experiments with Saginaw
                 River Sediments	30
     3.2  Summary of PAH  Biodegradation  Results  	  35
          3.2.1  Assessment  of Potential for Biodegradation
                 of PAHs	35

          3.2.2  Processes Affecting the Rate of Aerobic
                 Biodegradation of PAHs	38

                                iii

-------
                  Table of Contents (continued)
Section                                                      Page
     3.3  Discussion and Interpretation 	  47
          3.3.1  PCB Biodegradation	47
          3.3.2  Aerobic PAH Biodegradation 	  49
4.0  Conclusions	54
5.0  References	55
6.0  Appendix	59
                               IV

-------
                        List of Figures
Figure                        Title


   1     ARCS priority areas of concern.                      5

   2     Map of Ashtabula River area of concern.              6

   3     Map of Buffalo River area of concern.                7

   4     Map of Saginaw River area of concern.                8

   5     Map of Sheboygan River area of concern.              9

   6     The chemical structures of PAHs.                     14

   7     Metabolism of PAHs.                                  15

   8     The design of biometer flasks used for
        radiorespirometry.                                   23

   9     Profile of amended and historical PCB
        biotransformation in  (a) unamended control
        (sterile and nonsterile) sediments and  (b) 20
        mg/L 2,2',3,3',4,5,6,6'-octachlorobiphenyl  (Octa-
        CB) amended sediments.                               25

   9     Profile of amended and historical PCB
        biotransformation in  (c) 5 and  (d) 10 mg/L Octa-
        CB amended sediments.                                26

   10    Evidence for the reductive dechlorination of PCBs
        as indicated by the reduction  in the average
        number of chlorines per biphenyl in unamended
        (live and sterile controls) and in Octa-CB
        amended  (5-20 mg/L) sediments  (Sheboygan River).     27

   11    Reductive dehalogenation of amended PCB congeners
        [either 2,3,3',4,4'-pentachlorobiphenyl (Penta-
        CB), 2,3,3',4,4',5-hexachlorobiphenyl  (Hexa-CB),
        or 2,2',3,4,5,6,6'-heptachlorobiphenyl  (Hepta-
        CB)] added  (a) individually or  (b) as  a mixture
        in Ashtabula River sediment slurries.                29

   12    Profile of biotransformation of historical  and
        amended PCBs  in  (a) live control experiments and
        in  (b) 2,3,3',4,4'-pentachlorobiphenyl  (Penta-CB)
        plus 2,3,3',4,4',5-hexachlorobiphenyl  (Hexa-CB)-
        amended sediment slurries.                           31

   13    Profile of reductive  dehalogenation  of historical
        and amended PCBs in Ashtabula  River  sediment
        slurries as measured  by the reduction  in  the
        average number of  chlorines per biphenyl.            32
                                v

-------
                 List of Figures  (continued)


14    Reductive dehalogenation of PCBs in (a) sterile
      control and (b) unamended (live) Saginaw River
      sediments after anaerobic incubation for 25
      weeks.                                              33

15    Reductive dehalogenation of PCBs in Saginaw River
      sediments after anaerobic incubation for 25 weeks
      and amended with (a)  Aroclor 1260 and  (b)  Aroclor
      1260 plus RAMM.                                     34

16    Effect of aeration on the mineralization of
      phenanthrene by the indigenous bacteria in 10%
      sediment slurries prepared with Buffalo River
      sediments.                                          39

17    Effect of nutrient additions on the aerobic
      mineralization of phenanthrene by the indigenous
      bacteria in 10% sediment slurries prepared with
      Buffalo River sediments.                            41

18    Effect of nutrient additions on the
      mineralization of pyrene when Buffalo River
      enrichment culture was added to 10% sediment
      slurry.                                             41

19    Effect of addition of sediment organic matter on
      the aerobic mineralization of pyrene by a Buffalo
      River enrichment culture in MSB containing added
      organic carbon.                                     42

20    Effect of inoculum size on the aerobic
      mineralization of pyrene when Buffalo River
      enrichment culture was added to 10% sediment        42
      slurry.

21    Effect of sediment concentration on the aerobic
      mineralization of pyrene in sediment slurries
      when Buffalo River enrichment culture was added
      to 0% 5%,  10%,  25%,  and 50% sediment slurries.
      Values represent the mean of triplicates.            43

22    Effect of initial addition of Triton X-100 (CMC =
      130 mg/L)  on the aerobic mineralization of pyrene
      when Buffalo River enrichment culture was added
      to 10% sediment slurry.                             43

23    Effect of addition of Triton X-100 (0.5 CMC =
      62.5 mg/L)  after the aerobic mineralization of
      available pyrene in 10% (A),  25% (B)  and 50% (C)
      sediment slurries to which Buffalo River
      enrichment culture was added.                       44
                             VI

-------
                 List of Figures (continued)


24    Effect of additions of heavy  metals (X  = 25 ug/mL
      Pb,  25 ug/mL Cu and 2.5 ug/mL basal
      concentration)  on the aerobic mineralization of
      pyrene when Buffalo River enrichment culture was
      added to 10% sediment slurry.                       46

25    Effect of additions of secondary PAHs on the
      aerobic mineralization of pyrene when Buffalo
      River enrichment culture was  added to 10%
      sediment slurry, with and without pretreatment
      with hydrogen peroxide.                             46
                             VI1

-------
                  List of Tables

                       Title                         page


Recovery of deuterated PAHs extracted from
Buffalo River sediment slurries.                     22

Rate of dechlorination of amended
octachlorobiphenyl  (Octa-CB) in Sheboygan River
sediment.                                            28

Concentrations of PAHs in homogenized bulk
Buffalo River sediment.                              35

Indicators of aerobic growth of indigenous
microorganisms from Buffalo River sediments on
PAHs as sole substrate and with sediment organic
matter as co-substrate.                              36

Aerobic biodegradation by indigenous Buffalo
River microorganisms of PAHs (20 mg/kg)  added to
sediment slurries.                                   37

Comparison of rate of aerobic PAH degradation by
an enrichment culture added to Buffalo River
sediment slurries to that by indigenous              38
organisms.

The effect of aeration method on the rate of
degradation of PAHs in Buffalo River sediment
slurries.                                             39

Trace metal concentrations in Buffalo River
sediments.                                            45
                     VI11

-------
             List  of Abbreviations/Acronyms
Ace         Acenaphthene
Acy         Acenaphthylene
Ant         Anthracene
ARCS        Assessment and Remediation of Contaminated
            Sediments
BaA         Benz(a)anthracene
BaP         Benzo(a)pyrene
BgP         Benzo(ghi)perylene
Bip         Biphenyl
BRS         Buffalo River sediment
CB          Chlorinated biphenyl
Chy         Chrysene
CWA         Clean Water Act
DbA         Dibenz(a,h)anthracene
ET          Engineering/Technology Work Group
Fir         Fluoranthene
Flu         Fluorene
GLNPO       Great Lakes National Program Office
Hepta-CB    2,2',3,4,5,6,6'-heptachlorobipheny1
Hexa-CB     2,3,3',4,4',5-hexachlorobipheny1
HMW         Higher molecular weight
LMW         Lower molecular weight
MSB         Minimal salts broth
Octa-CB     2,2*3,3',4,5,6,6'-octachlorobiphenyl
PAH         Polycyclic aromatic hydrocarbons
PCB         Polychlorinated biphenyl
Penta-CB    2,3,3',4,4'-pentachlorobiphenyl
Phe         Phenanthrene
Pyr         Pyrene
RAMM       Revised anaerobic  mineral  medium
                            IX

-------
                         ACKNOWLEDGEMENTS


     We gratefully acknowledge the efforts of all those who
assisted in conducting the research that is the basis of this
report.  Recognition is extended to Dr. Daniel A. Wubah
(University of Georgia, Athens, GA and Biology Department, Towson
State University, Towson, MD) for his contribution to the
research conducted on the dechlorination of PCBs.  The excellent
technical assistance of Ms. Rebecca Adams for PCB analyses and
Ms. Tere Simmons for performing PCB extractions is also greatly
appreciated. The authors also thank Blasland and Bouck Engineers,
P.C., the Detroit District of the U. S. Army Corps of Engineers,
and NRRI, University of Minnesota-Duluth for collection of PCB
and PAH contaminated sediments. Recognition is also given to
Marirosa Molina  (Environmental Research Laboratory, U.S.
Environmental Protection Agency, Athens,  Georgia)  and Dr. Brian
Nummer (University of Georgia, Athens,  Georgia and Department of
Biology,  Tennessee Technical University,  Cooksville,  Tennessee).
                                x

-------
1.0  INTRODUCTION

     The Great Lakes National Program Office (GLNPO) leads
efforts to carry out the provisions of Section 118 of the Clean
Water Act (CWA) and to fulfill U.S. obligations under the Great
Lakes Water Quality Agreement with Canada.  Under Section
118(c)(3) of the CWA, GLNPO was responsible for undertaking a 5-
year study and demonstration program for the remediation of
contaminated sediments.  Five areas were specified for priority
consideration in locating and conducting demonstration projects:
Saginaw River and Bay, Michigan; Sheboygan River, Wisconsin;
Grand Calumet River/Indiana Harbor Canal, Indiana; Ashtabula
River, Ohio; and Buffalo River, New York.  GLNPO created the
Assessment and Remediation of Contaminated Sediments  (ARCS)
Program to implement these demonstration projects.  The ARCS
Program is managed by an Advisory Committee comprised of federal,
state, and public interest representatives, and an Activities
Integration Committee made up of the chairpersons of various
technical work groups.

     To make informed decisions concerning the design of
demonstration projects, the Engineering/Technology  (ET) Work
Group within the ARCS Program was charged to evaluate and test
available removal and remedial technologies for contaminated
sediments, to  select promising new technologies for further
testing, to demonstrate alternatives at priority consideration
areas, and to  estimate contaminant losses during remediation.
The ET Work Group was to select technologies that were available,
implementable, and economically feasible.  As part  of this
effort, the effectiveness of selected developmental sediment
remediation technologies was to be tested at the bench scale.

      The Environmental Research Laboratory, Athens, Georgia
 (AERL),  as a member  of the ET Workgroup, provided technical
assistance in  the areas of bioremediation.  AERL conducted bench-
scale  studies  to examine the degradation of polychlorinated
biphenyls  (PCBs) in  Saginaw River  and Bay, Sheboygan  River, and
Ashtabula River  sediments and the degradation of polycyclic
aromatic hydrocarbons  (PAHs) in Buffalo River sediments.  As  part
of this  effort a workshop entitled "Biological Remediation  of
Contaminated Sediments, with  Special Emphasis on the Great
Lakes" was held in Manitowoc, Wisconsin  in June  of  1990  (Jafvert
and Rogers, 1991).   The research results presented  in this  report
were  designed  to address key areas of research needs  identified
at this  workshop.


1.1   BACKGROUND

      In  June 1990, a workshop,  "Biological Remediation of
Contaminated Sediments, with Special Emphasis on the  Great
Lakes,"  was held to  identify research that would advance the  use

-------
 of bioremediation to cleanup contaminated sediments.   Two general
 research areas were identified as information gaps:  (1)  The
 specific processes and mechanisms controlling observed
 degradation rates and patterns,  and (2)  issues associated with
 extrapolation of bench-scale studies  to  pilot- or  full-scale
 field studies.

      Clearly,  a significant  amount of information  about  the
 biological  transformations of pollutants already is known from
 process  research.   Much of this  research is  at the
 phenomenological level.   The results  of  these studies  have helped
 identify empirically,  or allude  to mechanistically, the
 interactions  among microorganisms,  pollutants,  and sedimentary
 and aqueous media in which they  exist.   These interactions can be
 complex,  even for simple systems,  such as  the transformation of  a
 single compound by a pure microbial culture  in homogeneous
 solution.   In this simple system,  characterization of  the
 degradation process requires an  understanding of nutrient and
 growth requirements,  the kinetics  of  transformation reactions,
 degradation pathways,  pollutant  concentration dependencies,
 effects  of  alternative substrates  and electron acceptors,
 temperature dependencies, the  effects  of metabolic inhibitors,
 and in some cases,  the effects of  varying  carbon sources.   The
 additional  complexity  associated with  investigating the  same
 microbial degradation  process  in natural or  manipulated  sediments
 is  obvious.  Additional  consideration  must be given to organic
 and inorganic  inhibitor  availability,  combined inhibitory
 effects, pollutant  bioavailability  and the kinetics of this
 availability,  and  amended microorganism  competition or
 cooperation with the  indigenous  bacteria.

     Although a complete  understanding of how these processes
 interact at specific sites would result  in the most direct
 approaches  to treatability,   a comprehensive  understanding may  not
 always be necessary.   In  many cases, biological treatment
 efficiency  may be  significantly  enhanced above background levels
 by  regulating a few limiting factors.  These  factors can  be
 identified  at the bench-scale through  simple  process studies.  In
 many cases, differences  in these controlling  factors are  reasons
 for the site- or sediment-specific nature of  biological
 treatability successes or failures.  Clearly, while much  is
 known, a better definition of the chemical, physical and
 biological processes or  factors  controlling  observed
 transformation rates and pathways in natural  and manipulated
 sediments will enhance the frequency and degree of bioremediation
 successes.

     AERL has concentrated its efforts on determining the
processes that limit the degradation of PCBs  and PAHs in
 sediments.  These bench-scale studies were conducted using actual
 sediment from the Ashtabula  River, Sheboygan River and Saginaw
River for PCBs, and the Buffalo River for PAHs.

-------
1.2  SEDIMENT DESCRIPTIONS

     The sediments used in the AERL bench-scale studies were
collected from four of the five ARCS priority consideration Areas
of Concern (AOCs). The sediments were located in areas in which
serious impairment of beneficial uses of water or biota are known
to exist. The sediments were collected from locations where
future pilot and/or field demonstration projects may be
conducted. The primary contaminants in the sediments were PAHs
and PCBs. The map provided in Figure 1 shows the locations of the
ARCS Priority AOCs.

1.2.1  Site Names and Locations

     Sediments were collected by GLNPO from the Saginaw River,
MI, the Sheboygan River, WI, the Grand Calumet River/Indiana
Harbor, IN, the Ashtabula River, OH, and the Buffalo River, NY
AERL conducted bench-scale studies with sediment samples from the
Ashtabula River, Sheboygan River and Saginaw River in the PCB
studies and from the Buffalo River in the PAH studies.  Specifics
of the sample locations for the Ashtabula River, Buffalo River,
Saginaw River, and Sheboygan River are shown in Figures 2, 3, 4
and 5, respectively.

     The Ashtabula River  is located in northeast Ohio and is
primarily  fed by  a rural  and agricultural drainage basin covering
an area  of approximately  350 square kilometers  (Figure 2). The
Fields Brook is a major tributary  of the Ashtabula River; the
large and  diverse concentration of chemical plants located on the
Fields Brook was  a primary  source  of chemical pollutants to the
Ashtabula  River.  Numerous chlorinated organic compounds,
including  PCBs, hexachlorobenzene, and chlorinated aliphatics, as
well as  heavy metals,  are known pollutants  in this area  (Sanders,
1991).   The Fields Brook  site  was  placed on the National
Priorities List in 1983.

     The Buffalo  River is located  in western New  York state
 (Figure  3).  Heavy industrial  activity,  including chemical
synthesis  and power  generation, has  resulted  in the  contamination
of Buffalo River  sediment with PAHs, PCBs,  chlordane,  heavy  met-
als  and cyanide  (Lee et  al. ,  1991).

      The Saginaw  Bay watershed (Figure  4)  consists  of 22,550
square  kilometers and serves  as the  source  of drinking water for
over 300,000 people  (Goudy,  1991).  Industry  located along the
River  is quite  diversified;  the major  industries  include
automobile manufacturing, metal fabrication,  chemical production,
and  food processing. A 25-30  cm depth  section of  a  sediment core
collected in the  Bay had a  total  PCB concentration  of 574 ug/ml,
and  a  recent study showed that the major PCB contaminants are a
mixture of Aroclor 1242,  1254 and 1260  (Brandon et  al.,  1991).

-------
      The  Sheboygan  River  and Harbor  site  (Figure  5)  is  located at
 the western  shore of  Lake Michigan.   The  site  includes
 approximately  22 kilometers of the Sheboygan River  and  the  38
 hectare harbor (Eleder, 1991) .  Contaminants of concern at  this
 site  include PCBs,  dioxins, PAHs  and heavy metals.   Analysis of
 the sediments  from  this site indicates that the original PCB
 mixtures  used  in this area were Aroclors  1242, 1248  and 1254
 (Skogerbee et  al.,  1991).
1.2.2
Sediment Acquisition and Homogenization
     Sediment samples were collected with a mechanical dredge by
the Army Corps of Engineers from a site in the upper Buffalo
River that had previously been identified as a "hot spot" of PAH
contamination.  The samples were identified as station 47,
corresponding to site number 747 on the map insert. Samples
collected by the Corps were shipped to the U.S. EPA's
Environmental Research Laboratory at Duluth, Minnesota, where
they were screened, homogenized, and then shipped to AERL.

     Samples were stored and shipped on ice.  Upon receipt of
samples at AERL,  the sediments (in 5 gal. plastic buckets) were
mixed thoroughly,  screened for debris (2 mm sieve) and divided
into 0.5- and 1-liter subsamples.  All subsamples were
individually labelled and identified in a logbook. Samples were
stored at 4°C  until  use.

     Ashtabula,  Saginaw,  and Sheboygan River sediments were
collected as grab samples that were immediately placed in glass
containers (filled to capacity)  and sealed with Teflon-lined
caps.   The sediments were characterized primarily as mixtures of
silt and sand contaminated with PCBs and low concentrations of
oils and greases.  Samples were shipped on ice and were stored at
4 C  until use.

-------
                                       ARCS' PRIORITY
                                    AREAS OF CONCERN
              ARCS AREAS OF CONCERN
                 1. SHEBOYGAN RIVER
        2. GRAND CALUMET RIVER / INDIANA HARBOR
                3. SAGINAW RIVER/BAY
                 4. ASHTABULA RIVER
                  5. BUFFALO RIVER
Assessment and Remediation of Contaminated Sediments
•  M  MO 1*0
I  I  I  I
                                                    M CMVMJNUf MIAl mOTICTKM MXNCV
                                                    CMC AT UWM NATDMM. rMOOMAM OFFCt
 Figure 1. ARCS priority areas of concern.

-------
                                     Ashtabula River
                   The sediment sample was a composite comprised of
                   subsamples taken throughout the Ashtabula River system
Figure  2.  Map  of Ashtabula  River area of concern.

-------
                                                              Buffalo River
                                                                      Sediment sample point
                                                                                     JbftJB
Figure 3. Map  of  Buffalo River area  of  concern.

-------
                Saginaw River and Bay
                       Sediment sample point
Figure 4. Map of Saginaw River area of concern,

-------
              /
             -N
                     1 Miles
Sheboygan
  Faib
                                                        I«ceiit4 Am    22

                                                        ComimimiT Citim    •

                                                        to*

                                                        tract toM boinxUtj
       Figure  5. Map  of Sheboygan  River area of concern.

-------
 1.3  BACKGROUND ON BI©REMEDIATION TECHNOLOGY

     A major challenge in today's world is to maximize the
 benefits that accrue from the industrial, agricultural and
 domestic use of xenobiotic chemicals while minimizing any adverse
 effects on natural ecosystems.  Some of the means to achieve this
 goal are waste minimization, proper management of waste disposal,
 and clean up of sites already contaminated.  Contaminated sites
 can include surface and subsurface soil, groundwater, and
 freshwater and marine sediments with their associated water.
 Often the cleanup or remediation of such sites may employ
 techniques from a number of disciplines, including biology,
 chemistry, geology, hydrology, and engineering.

     For decades, biological treatment has been utilized in
 composting and in treatment of municipal and industrial
 wastewater.  Within the past decade, interest has grown in the
 use of microorganisms for treatment of hazardous wastes that
 contaminate a variety of environments.  Biological treatment,
 either alone or combined with other technologies, compares
 favorably with physical and chemical treatments on the bases of
 cost and completeness of chemical transformation.  The cost of
 bioremediation is usually less than the cost of other methods of
 waste disposal.

     The use of indigenous microorganisms to clean waste sites
 has the potential to make bioremediation a very cost-effective
 technology.  This is especially true for cases in which only
 minor adjustments to the environment,  such as addition of
 nutrients, are required to achieve cleanup.  Biological treatment
 also offers the possibility of complete transformation of organic
 waste.  Physical treatment of wastes,  including extraction,
 adsorption, solidification,  encapsulation,  and filtration, only
 separates or immobilizes but does not destroy waste.  Thermal
 processes, such as incineration,  either completely destroy waste
 or reduce its volume,  but may produce toxic ash and volatile
 compounds.  Both chemical and biological treatment of wastes may
 result in hazardous and stable components.   However, biological
 treatment offers the most flexible technology option to destroy
 hazardous waste components to acceptable environmental levels. In
 contrast to biological treatment,  alternate technologies may
 require further treatment of wastes or extended storage of
 treated waste.

     Bioremediation in aquatic environments often requires
 treatment of sediments,  pore water,  and surface waters.  Sediments
 can serve as long-term sources of pollutants.   Contaminated water
 or sediment may be biologically remediated in situ or in
bioreactors.   Although in situ treatment may be less expensive
than bioreactor treatment,  it often requires longer time periods
 (months to years)  to achieve clean-up standards and may be
extremely limited by physical constraints at the site.   Treatment

                               10

-------
in bioreactors may be complete within a shorter time (on the
order of weeks) because optimal conditions (i.e., pH,
temperature, nutrient concentration^, etc.)  for biodegradation
can be more easily maintained.  However, physical removal of the
contaminated material to the bioreactor is required and much
higher operating expenses and maintenance costs are an additional
consideration as are the risks associated with the removal and
transportation of the contaminant.

1.3.1  PCB Biodearadation

     For PCB-contaminated sediments, reductive dechlorination
reactions (as mediated by anaerobic bacteria) preferentially
transform the more highly chlorinated PCB congeners to less-
chlorinated derivatives that are more amenable to aerobic
degradation. In this instance, the anaerobic and subsequent
aerobic processes are complementary and result in a reduction of
toxic  (higher chlorinated, coplanar) PCB congeners, and possibly
even the complete biological destruction of PCBs through
subsequent aerobic oxidation  (Harkness et al., 1993).  An
important prerequisite for using this method at a remediation
site is to assess the ability of indigenous microorganisms to
transform the pollutants at the site and to develop techniques to
enhance their activities.

     PCB transformation in anaerobic environments such as the
sediments of lakes and rivers was inferred in the mid 1980^s from
the studies of Brown and coworkers  (1987a,b). These investigators
noted  an alteration in the expected profile of PCB congeners in
contaminated sediments from the Hudson River compared to the
congener profile of the original contaminating Aroclor. The
alterations were characterized by a reduction in the
concentration of the more highly chlorinated PCB congeners and an
increase in the concentration of the more lightly chlorinated
congeners. Further, the observed shifts in the congener profiles
were congener specific, with  selective or preferential removal of
meta-  and para-chlorines from the more highly chlorinated
congeners.
PCB congener showing the numbering
system and chlorine positions
                                11
                                                (para)
                                                 4
                                                       (meta)

-------
 The transformed  sediments were thus enriched with  ortho-
 substituted,  lightly  chlorinated PCB  congeners.  It was therefore
 proposed that dechlorination of the more highly  chlorinated PCB
 congeners was catalyzed by anaerobic  microorganisms residing  in
 the contaminated sediment.  The biologically mediated reductive
 dechlorination of PCBs from contaminated sediments was
 subsequently  demonstrated in several  laboratory  investigations
 (Quensen et al.,  1988,1990; Bedard and Van Dort, 1992).   In some
 studies, the  microbial inoculum was obtained by  "washing" PCB-
 contaminated  sediments with anaerobic medium and collecting the
 supernatant  (Quensen  et al., 1990; Nies and Vogel, 1990).  Other
 investigators demonstrated reductive  dechlorination of PCBs in
 actively methanogenic sediment slurries spiked with PCB mixtures
 (Aroclors) or specific PCB congeners  (Van Dort and Bedard, 1991;
 Wubah and Rogers, 1992).  In each study, evidence for reductive
 dechlorination was demonstrated by observing changes in the PCB
 congener profile  with time.

     To date, there have been only a  limited number of studies
 that attempt  to understand the factors that affect the reductive
 dechlorination of PCBs in historically contaminated sediments.
 Abramowicz et al. (1990)  reported that addition  of inorganic
 nutrients enhanced reductive dechlorination of endogenous PCBs in
 laboratory incubations of Hudson River sediments. In a recent
 study using methanogenic sediment slurries contaminated with
 Aroclor 1260, Bedard  and Van Dort (1992) reported that
 dehalogenation of amended brominated  biphenyl congeners
 significantly stimulated the reductive dechlorination of
 endogenous (historical) PCBs.  In an earlier study,  Bedard and
 coworkers (1990)   reported that amendment of sediment from Woods
 Pond (Housatonic  River, Massachusetts) with a high concentration
 (350 uM) of either 2,3',4',5-chlorobiphenyl (CB)  or 2,3,4,5,6-CB
 stimulated reductive  dechlorination of "historical" PCBs;
 transformation of congeners with para chlorines was especially
 evident.

     One of the many  objectives of this study was to determine
 the reductive dechlorination potential of PCB-contaminated
 sediments from the Saginaw,  Sheboygan, and Ashtabula Rivers and
 to examine the effects of addition of PCB congeners and mineral
 nutrients on  the  reductive dechlorination of endogenous
 (historical)  PCBs by the resident microbial populations.

 1.3.2  PAH Biodegradation

     PAHs are of environmental significance because of their
 resistance to biodegradation,  their chronic toxic effects on
humans and other species,  and their widespread occurrence at
 contaminated waste sites.   Low solubilities and high partition
 coefficients  dominate the behavior of PAHs in biological  and
environmental systems.  The high partition coefficients of PAHs
determine their  high relative  concentrations in aquatic animals

                               12

-------
and sediments as compared to the water column. With log Kows
(octanol/water partition coefficients) in the range of 2.5 - 6.5,
PAHs may be present in sediments at a hundred to nearly a million
times their aqueous concentrations.  Sediments, then, serve as a
long-term source of PAHs to the water column and to biota.

     The persistence of PAHs in sediments is a function of their
chemical structures (Figure 6). Low molecular weight  (LMW) PAHs,
such as naphthalene, are both more soluble and more volatile than
are the higher molecular weight (HMW) PAHs, and are lost from
sediments more rapidly through volatilization, dissolution and
biotransformation. High molecular weight PAHs are relatively
insoluble and resistant to microbial degradation, and may persist
indefinitely  (Cerniglia and Heitkamp, 1989) .

     Given the tendency of PAHs to adsorb to sediments, there has
been a great deal of interest in the potential for PAH
degradation under anoxic conditions.  Anaerobic degradation of
monoaromatic compounds by anaerobic respiration, fermentation and
photometabolism has been demonstrated (Young, 1984).  Naphthalene
and acenaphthene have been degraded to nondetectable  levels under
denitrifying conditions where water is the oxygen source for
oxidation of the ring  (Mihelcic and Luthy, 1988) .  Such activity
is of limited utility in bioremediation; no anaerobic degradation
of higher molecular weight PAHs has been reported to  date.

     Under aerobic conditions, molecular oxygen is the source of
oxygen for the degradation of PAHs.  Degradation may  follow
either a dioxygenase or a monooxygenase pathway  (Figure 7). In
bacteria, dioxygenases incorporate both atoms of oxygen, leading
to the formation of dihydrodiols, with subsequent ring fission
and assimilation, whereas fungi hydroxylate PAHs as  a prelude to
detoxification  (Dagley, 1981).

     Aerobic bioremediation also has been tested as  a cleanup
strategy for materials, primarily soils, from contaminated
environmental sites where soils have been highly contaminated
with PAHs from creosote and other wood-preserving compound
mixtures. At  such sites, high  levels  of PAHs are found in
conjunction with other organics such  as pentachlorophenol and
organic solvents, with the total organic pollutants  often
comprising 5 to 10% by weight.

     Both slurry reactors and  in situ remediation have been
attempted with PAH-contaminated soils. Mueller  et al.  (1991)
reported that indigenous organisms successfully  degraded phenols
and LMW-PAHs  in creosote-contaminated soil slurries  within 30
days.  The addition of nutrient amendments did  little to  enhance
the effectiveness of degradation of HMW-PAHs.  In soils that
contained 4-, 5- and 6- ring PAHs  in  concentrations  ranging from
10.4 mg/kg  (benzo (b) fluorene)  to 61.7 mg/kg
 (benzo (b/ k) f luoranthene) , the  addition of nutrients  to slurry

                                13

-------
                     WfWHALENE
                       (NAP)
                       aUORENE
                        (FIR)
                             CHRiSENE
                               (CHY)
       AKTHRACCNE
         (ANT)
       ACtNAPIHENE

          (AH)
       oo
         PHENANTtfCNE
           (WE)
                                                                     ACENAPIHYLENE
               BENZ(o)ANTHRACENE
                              PIRENE
                               (Prt?)
                aUOWWTHENE
                   (FIT)
                  fen;o(g,h,i)perylene
Benz(o)pyrene
Oiben2(o,h)anthrocene
Figure  6.  The  chemical  structures  of PAHs.
                                             14

-------
  Polycyclic
  Aromatic
Hydrocarbon
                        Arene Oxide
                                                      OH
                                    Epoxide
                                   Hydrolase
                                                      OH
                                                      OH
                                             trans-Dihydrodiol
         H

          OH
NAD*
   i
        r-OH  Dehydrogenase
        H             V»     .  R
                       OH
cis-Dihydrodiol
                                                  Catechol
                                                              (>-Glucoside
   Figure 7. Metabolism of PAHs  (from Cerniglia  and  Heitkamp 1989).

-------
reactors resulted in the reduction of the HMW-PAHS by 48 percent
after 12 weeks of incubation.  The same group of compounds was
reduced by only 35 percent in reactors operated without nutrient
supplement. The in situ degradation rate for PAHs and phenols in
soils was much slower; 8 to 12 weeks were required for 50 percent
removal of organic contaminants, with little reactivity of HMW-
PAHs.

     Few studies have specifically addressed the issue of
bioremediation in sediments.  Most field applications of
bioremediation have attempted to decontaminate soil, and usually
soil with relatively high concentrations of PAHs.  Soil sites
have usually been contaminated directly by disposal of wastes, by
spills, or by application of sewage sludges  (Wild et al.r  1991),
whereas sediments are typically contaminated as a result of
sorption of the compounds via an aqueous exposure.  Although the
latter may result in concentrations that pose environmental
threats, they are seldom as high as those resulting from direct
dumping.

     Sediments present several challenges to bioremediation:
relatively low concentrations of PAHs, predominantly anoxic
environments, an overlying water column that both limits physical
access and disperses any soluble additions that may be added
 (e.g. nutrients or oxidants), and high concentrations of organic
matter that both sorb the PAHs and support competitive
consumption of added nutrients and oxidants.

2.0  TREATABILITY STUDY APPROACH

2.1  STUDY OBJECTIVES AND RATIONALE

2.1.1  PCB Biodegradation

     The objectives of this  study were to determine the  extent  of
past (in situ) dechlorination activity, to assess the potential
for  further biotransformation of PCBs, and to examine the  effects
of addition of specific PCB  congeners and mineral nutrients  on
the  reductive dechlorination of amended and  historical PCBs  by
the  resident  sediment microbial populations  in  sediment  samples
from the Ashtabula, Saginaw  and Sheboygan Rivers. The primary
goal was to identify  factors that might influence the rate and
extent  of  biotransformation  of  the historical PCBs  in the
contaminated  sediments.

2.1.2   PAH Biodegradation

     The objectives of this  study were to identify  the
environmental parameters that constrained the biodegradation of
PAHs in Buffalo River sediment  and to  assess the potential for
their  manipulation, either  in situ or  ex situ,  to enhance  the
rate of remediation.   The  study considered the  mechanisms

                                16

-------
documented in the literature, as derived from both laboratory and
field experiments, in light of the specific characteristics of
the contaminated sediments themselves.  Rates of degradation in
sediment were compared and contrasted with those reported for
other media and processes that enhanced the rate and extent of
degradation were identified and investigated.


2.2  EXPERIMENTAL DESIGN

2.2.1  PCS Biodegradation

     Enhancement of PCB biotransformation in sediments has been
reported to occur in laboratory investigations via addition of a
limiting nutrient, such as an inorganic mineral or other specific
organic compounds, or via addition of specific PCB and/or
polybrominated biphenyl  (PBB) congeners. Experiments were
therefore conducted to examine the effects of inorganic nutrient
and PCB congener addition on PCB biotransformation in the various
contaminated river sediments.

     Slurries prepared with Sheboygan River sediments were
amended with 2,2'3,3',4,5,6,6'-octachlorobiphenyl  (Octa-CB) to
yield a final Octa-CB concentration of 5 to 20 mg/L wet sediment
slurry  (approximately 0.2 g dry weight sediment/ml wet slurry).
Sterile controls were prepared both with and without addition of
Octa-CB. Live controls  (non-sterilized) without addition of Octa-
CB were also prepared.

     Slurries prepared with Ashtabula River sediments were
amended with 2,3,3',4,4'-pentachlorobiphenyl  (Penta-CB),
2,3,3',4,4',5-hexachlorobiphenyl  (Hexa-CB), and 2,2',3,4,5,6,6'-
heptachlorobiphenyl  (Hepta-CB) either alone  (individual
congeners, 10 mg/L)  or in all congener combinations  (total of 10
mg/L amended congeners). Sterile controls were also prepared as
described previously.

     Slurries prepared with Saginaw River sediments were amended
with Aroclor 1260  (final concentration of 50 mg/L) both in the
absence and presence of revised anaerobic mineral  medium  (RAMM).
RAMM  (see Appendix) was prepared according to the  procedures of
Shelton and Tiedje  (1984) and was added to the sediment slurries
as a concentrated, anoxic solution. Sterile controls were
prepared as previously described.

2.2.2  PAH Biodegradation

     Enhancement of  bioremediation of PAHs has been reported in
field and reactor  studies. Although enhancement can be associated
with manipulations such as nutrient additions, many variables
affect the rate  of PAH transformation.  Thus, an understanding of
the processes that impact the rate of degradation  is essential

                                17

-------
for predicting the feasibility and effectiveness of
bioremediation of these compounds on other than a case-by-case
basis.

      In light of the objectives of this project, experiments were
designed to: 1) develop an enrichment culture of PAH-degrading
bacteria derived from Buffalo River sediment, 2) characterize the
activity of the enrichment culture and contrast it with the
activity of indigenous organisms, 3) determine whether the
enrichment culture maintained and expressed its activity when
reintroduced into sediments, and 4) investigate the processes
that  affected the rate of transformation of PAHs by the
enrichment culture in sediments.

      The factors that determine the rate of degradation of PAH
compounds in the environment can be considered to be intrinsic,
that  is, controlled by properties of the chemical itself;
extrinsic, controlled by the environment; or combinations of the
two.  Thus, oxygen or nutrient concentrations and bacterial
densities are extrinsic limitations, toxicity to bacteria is an
intrinsic limitation, and sorption of low-solubility PAHs to
sediment organic matter is a combination of the two. Extrinsic
factors are more readily modified, and were investigated first as
possible limitations on the rates of biodegradation.
Environmental processes considered in these investigations focus
on the extrinsic factors that may limit the rate and extent of
biodegradation of PAHs and the interactions between bacteria and
the sediment environment that may determine that activity of
degrader organisms on the substrates of interest.

     The processes that influence the rate and extent of PAH
degradation in sediments were investigated by comparing the
kinetics of degradation under controlled conditions.  The
potential for degradation of PAHs and the range of compounds
susceptible to transformation were assessed by amending sediments
with mixtures of PAHs.  The concentrations of PAHs in Buffalo
River sediments as sampled were low. Thus, PAHs were added to
these sediments in order to elevate concentrations to levels that
supported the analysis of differences in rates of degradation in
response to experimental treatments. Moreover, given the relative
unavailability of hydrophobic materials that have been associated
with organic sediments for long periods,  the amendment of
sediments with PAHs permits the optimization of the environmental
conditions, other than sorption, that control the rate of
degradation. The added mixture consisted of up to 16 compounds at
individual concentrations of approximately 20 mg/kg sediment,
with representative compounds from each size class of PAHs,  based
on number of rings.   The structure of the compounds added is
shown in Figure 6.   The transformation of the PAHs was assessed
by extracting the entire sediment slurry from among sets of
replicate treatments at a specified time, and analyzing the
extract by gas chromatography.  The sampling time periods

                                18

-------
depended on the concentrations of the PAHs and the relative
activity of the sediments or inoculum.

     The processes that influence the rates of degradation were
investigated by measuring the kinetics of degradation of a model
compound.  Pyrene was selected as the model on the basis of its
relative resistance to degradation in unmodified sediments.
Pyrene degrades slowly in sediments, and few organisms that
utilize pyrene as a growth substrate have been identified.
Pyrene is also available in radiolabelled form, which facilitates
frequent data collection over short time periods needed for
kinetic studies. Thus the trapping of radiolabelled C02  yielded
data amenable to kinetics interpretation; specifically,  the time
series data provided by radiorespirometry permit accurate rate
comparisons as well as indications of bacterial acclimation times
and growth.  The enrichment culture developed from Buffalo River
sediment was able to mineralize pyrene and was used as the
inoculum in the mineralization studies.
2.3  EXPERIMENTAL METHODS

2.3.1  PCB Biodegradation

     Anoxic slurries of the PCB-contaminated sediments were
prepared in an anaerobic glove box containing an atmosphere of 5
percent hydrogen and 95 percent nitrogen.  For all sediment
samples, one volume of wet sediment was combined with two volumes
of anoxic  (N2  sparged)  distilled water and the  mixture stirred
for approximately 5 minutes.  Approximately 50 milliliters of the
sediment slurry were added to 120-ml amber serum bottles and
amendments were added within 1 hour.  Serum bottles were crimp-
sealed with a Teflon-lined septum and aluminum cap and vigorously
mixed to distribute the amended PCB congener evenly. Live control
(no sterilization) sediment slurries were prepared without PCB
congener amendment. Sterile controls were prepared for all
experiments by autoclaving sediment slurries at 121°C for 30
minutes on 3 consecutive days, followed by PCB congener
amendment.  Experiments were performed in triplicate.  All
vessels were incubated in the dark at room temperature  (25°C)
inside the anaerobic glove box.

     At specified time intervals  (usually 1-2 months), the
sediment slurries were thoroughly mixed. Slurry samples
(approximately 5 ml) were withdrawn, transferred to screw-cap
tubes, and sealed with a Teflon-lined cap.  Five milliliters of
acetone and 100 pi of surfactant  [10%  (v/v) Triton X-100] were
added to each sample and the sediment slurry was shaken
thoroughly.  Surfactant was added to enhance the recovery of PCBs
from the contaminated sediments. Samples were then placed in an
ice bath, sonicated for 10 minutes at 30-35W using a titanium
microtip probe  (Ultrasonic Power Corp., Freeport, Illinois), and

                                19

-------
centrifuged at 3500 x g for 10 minutes.  The supernatant was
transferred to a 250-ml separatory funnel and 10 ml of hexane-
acetone  (9:1, v/v) were added.  The separatory funnel was
vigorously shaken and the aqueous layer was discarded.  The
remaining organic phase was sequentially washed with 5 ml each of
2%  (w/v) Nad, concentrated H2S04, and  2%  (w/v) NaCl. A detailed
description of the PCB extraction procedure is presented in the
Appendix.

     Congener-specific PCB analysis was performed using an HP-
5890 gas chromatograph (Hewlett-Packard Co., Palo Alto,
California) equipped with a DB-1  (J & W Scientific, Folsom,
California) capillary column  [30m by 0.25mm (inside diameter);
0.25pm film thickness] and an electron capture detector.
Identification and quantitation of PCB congeners were performed
by comparing the areas and retention times of individual peaks
with those in a standard mixture of Aroclors 1232, 1248 and 1262
in the ratio 2.5:1.8:1.8 (Mullin, 1985).  Congener identities
were confirmed by comparing the order of elution with that
published by Mullin et al.  (1984).  A three point calibration
curve was used to quantitate added PCB congeners as well as
historical PCBs.   Data were acquired and analyzed using Maxima
820 software  (Dynamics Solutions, Ventura, California) on an NEC
PowerMate 386SX computer connected to the gas chromatograph
through a system interface module (Waters Corp.,  Milford,
Massachusetts).  A detailed QA/QC plan for the congener-specific
PCB analysis is presented in the appendix.


2.3.2  PAH Biodegradation

     An enrichment culture that could degrade the widest range of
PAHs was developed by enriching the sediment bacterial population
for individual PAH degraders,  and then combining the individual
cultures to form a mixed enrichment culture. The individual
enrichments were developed by incubating 10% slurries (w/v) of
Buffalo River sediment in minimal medium with each of the PAHs
shown in Figure 6, except for the 6- ring compounds. Each of the
2- and 3- ring PAHs were added to the sediment slurry at 50 mg/L;
the 4- and 5- ring PAHs were added at 20 mg/L. The compounds were
added at concentrations substantially higher than environmental
concentrations in order to maximize the population of degrader
organisms.  The 6-ring PAHs were not added to the enrichment
media because these compounds do not support microbial growth and
would likely be toxic to bacteria.  The sediment slurries were
incubated on a shaker (150 rpm)  at 30 C for  3  weeks. After  3
weeks, a subsample of the sediment slurry was transferred to
minimal medium (Appendix)  containing the same PAH compound on
which the organism had been enriched.  Each of the PAH-degrading
strains was evaluated for activity by plating the enrichment on
agar plates containing the PAH to be tested.


                                20

-------
     The activities of the indigenous organisms and the
enrichment culture were assessed by introducing each source of
bacteria to 10% sediment slurries to which had been added a
mixture of all of PAHs shown in Figure 6, each at a nominal
concentration of 20 mg/kg sediment  (dry weight of sediment).   The
amount of PAH remaining was determined throughout the period of 2
to 6 weeks following dosage.

     PAHs were added to sediments as a primary mixture made by
combining 2 mg each of acenaphthene, acenaphthylene, fluorene,
phenanthrene, anthracene, fluoranthene, pyrene, benzo(a)pyrene,
dibenz(a,h)anthracene, and benzo(g,h,i)perylene in 10 ml CH2C12.
The spiking mixture was stored at -4°C.   Sediment slurries were
spiked by adding 50 pi of the PAH mixture to wet sediment  (as
received, equivalent to 20 g dry weight) in a 250-ml flask.  The
solvent was allowed to evaporate overnight at 4°C prior to
addition of the experimental medium  (e.g., minimal salts broth)
to form the slurry.

     Sampling intervals depended on the concentration of PAHs
added to the sediments and the activity of the bacterial
inoculum.  Estimates of the appropriate intervals were based on
outcomes of prior experiments.  At the time intervals indicated
in the figures for each experiment  (2-6 weeks), the entire
contents of replicate flasks were extracted. The sediment
slurries, which had been amended with PAHs, were centrifuged to
remove water, treated with methanol or sodium sulfate to remove
residual water, and extracted twice with methylene chloride on a
wrist-action shaker overnight.  Detailed procedures for
extraction and GC analysis are presented in the appendix.

     The recovery of PAHs extracted from sediments was calculated
from the additions of a mixture of  six deuterated PAHs,  (Table
1).  Deuterated compounds are those that differ from the non-
deuterated compound by the replacement of a number of hydrogen
atoms with the heavier form of hydrogen, or deuterium. By virtue
of the greater molecular weight, the deuterated form elutes from
the gas chromatograph slightly later than the standard form.  The
identification of separate peaks for the extracted compound and
the recovery spike permits the calculation of recoveries  from a
single analysis, rather than the two analyses required by the
method of standard additions.  Recoveries were calculated to
assess the effectiveness and the consistency of the analytical
procedures used; the PAH data presented herein have not been
corrected for recovery. The procedure  for analysis of PAHs by gas
chromatography also is presented in the appendix.
                                21

-------
 Table  1. Recovery  of deuterated PAHs extracted  from Buffalo River
 sediment slurries.
     Compound            % Recovery1           S.D.
d8- naphthalene
dl 0 - acenaphthene
dl 0 -phenant hr ene
d!2-chrysene
d!2-perylene
32.85
46.15
60.38
70.27
76.77
4.3
6.5
4.9
5.3
6.3
 Recoveries calculated as mean and standard deviation (S.D.)  of
13 samples.

     In experiments in which the PAH concentration was determined
by extraction and GC, each treatment was run in duplicate or
triplicate.  At each sampling, at least one treatment replicate
was sampled in triplicate, in order to compare the variance due
to sampling to that due to differential performance of replicates
within treatments.  All data are represented as averages of
replicate or triplicate determinations.

     Radiolabelled PAHs can be readily quantified with a high
degree of precision and were used for several types of
experiments in this project in a technique known as
radiorespirometry. No extraction is required to measure the
radiolabelled compounds, either as the parent compound in the
sediment slurry or as carbon dioxide produced by respiration of
the parent. This permits frequent, precise sampling, which is
necessary for comparing the rates of transformation of PAHs in
response to treatments.  Moreover, although disappearance of a
parent PAH can indicate the occurrence of a transformation, the
detection of radiolabelled carbon dioxide from labelled PAHs can
indicate the full mineralization of the parent and hence the
potential for supporting the growth of bacteria capable of their
degradation.

     Mineralization experiments were performed in biometer flasks
or Erlenmeyer flasks modified to function as biometer flasks
(Figure 8).  The sediment slurry or liquid sample was incubated
in the main chamber of the flask and CO2 was  trapped in  0.1 N
NaOH in a 25-ml vial suspended in the flask by wire from a rubber
stopper.  A 3-ml syringe with a needle pierced through the
stopper was connected to tubing and was used to replace the NaOH.
An aliquot of NaOH (I ml) was combined with 5 ml scintillation
cocktail (ICN Ecolite)  and radioactivity was determined on a
Beckman L56000LL scintillation counter (Fullerton,  California)
after chemiluminescence had subsided.  Samples were counted for 5
minutes or to 2% counting precision.   The counts were corrected

                               22

-------
by the external standards method.  Data are reported as the
cumulative number of counts trapped as radiolabelled carbon
dioxide (14C02) ,  expressed as the percentage of counts added to
the experimental flask as labelled PAH.  Thus, the data represent
the fraction of the PAH transformed and respired by the bacteria.
The percent 14CO2 seldom approaches 100% because a fraction of the
original PAH is incorporated into biomass or metabolic products.
Thus, a characteristic yield of 14C02 results from the utilization
of any given substrate by a given bacterium under similar growth
conditions.  All data are the mean of triplicate measurements
unless otherwise specified.
Figure 8. The design of biometer flasks used for radio-
respirometry.
     To determine whether specific PAHs were mineralized by  the
enriched culture, approximately 0.025 uCi of the  labeled compound
and 2 ppm unlabeled substrate were added to sediment  slurries  and
14CO2 evolution was monitored over time. The compounds tested
were: phenanthrene, fluoranthene, pyrene and benzo(a)pyrene,
which were obtained in radiolabeled form  ([9-14C]phenanthrene
 (Sigma Chemicals),  [3-  C]fluoranthene  (Chemsyn Labs,  Lexana,
Kansas),  [4,5, 9,10-14C]pyrene  (Chemsyn  Labs)  or [7,10-
14C]benzo(a)pyrene  (Sigma  Chemicals)).   Sediments  were inoculated
with the enrichment culture.
                                23

-------
3.0  RESULTS AND DISCUSSION

3.1  SUMMARY OF PCB BI©DEGRADATION RESULTS

3.1.1  Bench-scale experiments with Sheboygan River sediments.

       The most prominent PCB homologs detected in the
contaminated Sheboygan River sediments were trichlorobiphenyls
and tetrachlorobiphenyls  (Figures 9 a-d, week 1); the total PCB
concentration was 65.6 + 5.3 mg/L. In unamended control and
autoclaved control sediments (Figure 9a), no appreciable change
in the profiles of historical PCBs was evident after 30 weeks of
anaerobic incubation at 25°C.   The average number of chlorines
per biphenyl (3.9) in the live, unamended control sample remained
essentially constant over the course of the 30-week incubation
period (Figure 10).  However, in all experiments amended with
Octa-CB,  there was a decrease in the concentration of hepta-,
hexa-, penta-,  tetra- and tri-CB congeners and an increase in the
concentration of di- and mono-CBs. The mole percentage of mono-
CBs was less than 1% at the onset of the experiment (week 1);
after anaerobic incubation for 30 weeks, this homolog group
accounted for approximately 8% of the total PCB congeners in
sediments amended with 20 mg/L Octa-CB  (Figure 9b).  The major
products of reductive dechlorination were di-CB congeners; this
homolog group increased from 2.5 to 40 mole percent after 30
weeks of incubation.  The most prominent di-CB peak detected in
Octa-CB amended sediments consisted of two ortho-substituted
congeners (2,2'- and 2,6-CB).

     Two additional homolog groups, tri- and tetra-CBs, initially
accounted for approximately 80% of the total PCB homologs in the
contaminated sediments (week 1, Figures 9 a-d).   After anaerobic
incubation for 30 weeks,  these two CB homolog groups were reduced
to less than 50% of the total mole percent of homologs in the
Octa-CB amended sediments.   The mole percentage of penta-, hexa-
and hepta-CB homologs were reduced from 8, 5 and 2% to 4, 3 and
1%, respectively.

     In addition to the changes observed in the PCB homolog
distribution after anaerobic incubation for 30 weeks,  there also
were changes in the average number of chlorine atoms per biphenyl
in the various treatments (Figure 10).  In the unamended and
sterile Sheboygan River sediments, the average number of chlorine
atoms per biphenyl remained constant throughout the experiment,
indicating little or no detectable dechlorination activity.
However,  in the Octa-CB amended sediments, the average number of
chlorines per biphenyl (total of historical plus amended PCBs)
decreased from 4.2 to 2.8 (± 0.1 ), 2.5 (± 0.3)  and 2.2  (±0.3)
in experiments amended with 5,  10 and 20 mg/L Octa-CB,
respectively.
                                24

-------
                           Sheboygan River Sediment (Control)
   60.00
   40.00
    20.00
    0.00
         MONO    Dl
TRI   TETRA  PENTA   HEXA   HEPTA   OCTA   NONA   DECA
             HOMOLOG
                    Sheboygan River Sediment + 20 mg/L Octachlorobiphenyl
                                                                     B
   60.00
   40.00
   20.00
    0.00
         MONO
                Dl
                      TRI
                           TETRA
                                  PENTA   HEXA
                                    HOMOLOG
                                              HEPTA
                              OCTA
                                    NONA
                                          DECA
Figure 9. Profile of  amended  and historical PCB biotransformation
in  (a)  unamended control (sterile and  nonsterile)  sediments and
(b)  20 mg/L  2,2',3,3',4,5,6,6'-Octachlorobiphenyl  (Octa-CB)
amended sediments. Experiments were conducted  under anaerobic
conditions using PCB-contaminated sediment from the Sheboygan
River.
                                   25

-------
                     Sheboygan River Sediment + 5 mg/L Octachlorobiphenyl
        60.00
        40.00
        20.00
        0.00
             MONO   Dl    TRI   TETRA  PENTA  HEXA  HEPTA  OCTA  NONA  DECA
                                   HOMOLOG GROUPS
                     Sheboygan River Sediment +10 mg/L Octachlorobiphenyl
        60.00
        40.00
        20.00
        0.00
             MONO   Dl    TRI   TETRA  PENTA  HEXA  HEPTA  OCTA  NONA  DECA
                                   HOMOLOG GROUPS
Figure 9.  Profile of  amended and  historical PCB biotransformation
in  (c) 5  and  (d)  10 mg/L Octa-CB  amended sediments.  Experiments
were conducted  under  anaerobic conditions using PCB-contaminated
sediment  from the Sheboygan  River.
                                     26

-------
   
   co
   i—
   
-------
Table 2. Rate of dechlorination of amended octachlorobiphenyl
(Octa-CB) in Sheboygan River sediment.
     Experiment                    Rate (mean ± S.D)

     Autoclaved control                      ND

     Unamended control                       ND

     Sediment 4- 5 mg/L Octa-CB          0.22 ± 0.01

     Sediment + 10 mg/L Octa-CB         0.35 ± 0.01

     Sediment + 20 mg/L Octa-CB         0.43 ± 0.01
     Rate measured as ug atoms of chlorine removed per gram
     sediment per month

     ND = no dechlorination
3.1.2  Bench-scale experiments with Ashtabula River sediments.

     Unamended Ashtabula River sediments and sediments to which a
single chlorobiphenyl congener had been added exhibited no
appreciable dechlorination activity during the first 3 months of
anaerobic incubation (Figure lla).  Lag periods of 3, 4 and 5
months were observed before the onset of dechlorination of the
amended Hexa-CB, Hepta-CB and Penta-CB congeners, respectively.
After 6 months incubation, approximately 75% of the amended Hexa-
CB and Hepta-CB were transformed while the same amount of the
Penta-CB was transformed after 7 months incubation.  No further
dechlorination was observed after incubation for an additional 2
months.  Approximately  20 to 30% of the original concentration
of all amended PCB congeners remained at the end of the
incubation period.
                                28

-------
     O)
     co

     CD
    DC

    CD
    O
    OL
100

 80

 60

 40

 20

  0
-•— penta-CB
-D—hexa-CB
 x- hepta-CB
                      8          16

                           Time (weeks)
                                    24
                                    32
        100
      o>
     •5  80
      CO
      E
      V
      cc
      OQ
      o
      O_
 60

 40

 20

  0
     penta-CB
     hexa-CB
     hepta-CB
                                                        B
                       8          16
                                     24
                                      32
                            Time (weeks)
Figure 11. Reductive dehalogenation of amended PCB congeners
[either 2,3,3',4,4'-pentachlorobiphenyl (Penta-CB),
2,3,3',4,4',5-hexachlorobiphenyl (Hexa-CB),  or 2,2',3,4,5,6,6'-
heptachlorobiphenyl (Hepta-CB)]  added (a)  individually or (b)  as
a mixture in Ashtabula River sediment slurries. Results are
expressed as the percent of the amended PCB congener
concentration.
                               29

-------
      In  experiments  amended with two  congeners,  a  lag period  of  6
months was  observed  before the  onset  of  dechlorination  of  all
amended  congeners  except  for  sediments amended with  a combination
of Penta-CB and Hexa-CB.  In this case, significant dechlorination
was detected after incubation for  5 months.  The Hexa-CB congener
was transformed more rapidly  in sediments  amended with  either
Penta-CB or Hepta-CB compared to sediments amended with .Hexa-CB
alone.   In  sediment  slurries  amended  with  a mixture  of  all three
congeners,  a lag period of 6  months was  observed before
appreciable dechlorination occurred  (Figure lib).  In this case,
all three congeners  were  transformed  at  a  similar rate.
Approximately 20 to  30% of the  original  concentration of all
amended  congeners  remained after 8 months  incubation.

     Addition of the Penta-CB,  Hexa-CB,  and Hepta-CB congeners,
singly or as mixtures, to Ashtabula River  sediment slurries
resulted in enhanced reductive  dechlorination of historical PCB
congeners in a manner similar to that observed for Sheboygan
River sediment amended with Octa-CB.  The  results presented in
Figure 12 are representative  of the PCB  homolog  distributions for
Ashtabula River sediment  slurries after  32 weeks incubation
amended  with either  the Penta-, Hexa-, or  Hepta-CB congener or
combinations  thereof. Figure  12a depicts the results for the  live
control  sediment.  Figure  12b  depicts  results for sediment amended
with Penta-CB plus Hexa-CB.

     The  average number of chlorine atoms per biphenyl  (total of
historical  plus amended PCBs)  decreased  in all experiments
amended  with  PCB congeners, but not in sterile and live
(unamended)   control  experiments (Figure  13a).  A representative
plot of  the  average  number of chlorines per biphenyl for
experiments  amended  with  all  three congeners is presented in
Figure 13a;   the average number  of chlorines/biphenyl lost for
experiments  amended  with  individually added congeners is
presented in  Figure  13b.  No  appreciable changes were observed in
the congener  profile  or the total PCB concentration  of the
historical  PCBs for  both  sterile and live controls. However,   the
molar increase in  lower chlorinated congeners was consistent  with
the molar decrease in higher  chlorinated congeners (historical
PCBs)  for all experiments amended with specific PCB congeners.

3-!-3  Bench-scale experiments with Saginaw River sediments.

     Unamended Saginaw River sediment slurries and sterile
controls  (Figures  14a-b)  exhibited little or no appreciable
changes  in the profile of PCB congeners after incubation at 25°C
for 25 weeks. In experiments amended with Aroclor 1260 plus RAMM,
a significant increase in the mole percentage of mono-,  di- and
tri-CB homologs and  a decrease in the mole percentage of homologs
with higher  numbers of chlorine atoms per biphenyl was evident
(Figure  15b). Significant reductions in the mole percentages   of
tetra-,  hexa-, hepta-, octa-,  and nona-CB were also evident.

                                30

-------
Concomitant with the reduction in the mole  percentage of the
higher chlorinated homolog groups, the mole percentage of the
mono-, di-, and tri-CB homolog groups increased from 1, 7.5, and
18% to 4.5, 15,  and 20%, respectively. These data clearly
indicate that  active, PCB-dechlorinating  populations were
present. In experiments amended with Aroclor 1260 alone, an
increase was evident in the mole percentage of mono- and di-CB
homologs only  (Figure 15a). The mole percentage of the tri-CB
through the nona-CB homolog groups decreased significantly. The
mono- and  di-CB congeners, which accounted  for 1% and 7.5% of the
total PCBs at  week number 1, increased to 8% and 28% after
anaerobic  incubation for 25 weeks.
                Ashtabula River Sediment (Control)
      30

      20

      10

       0
         MONO
Dl    TRI   TETRA  PENTA  HEXA  HEPTA  OCTA  NONA   DECA

              HOMOLOG GROUPS
            Ashtabula River Sediment •(• Pentachlorobiphenyl + Hexachlorobiphenyl
                                                             B
      30

      20

      10
         MONO   Dl   TRI   TETRA PENTA  HEXA  HEPTA  OCTA  NONA  DECA

                             HOMOLOG GROUPS
 Figure 12.  Profile of biotransformation of historical  and amended
 PCBs  in (a)  live control experiments and in (b) 2,3,3',4,4'-
 pentachlorobiphenyl  (Penta-CB)  plus 2,3,3',4,4',5-
 hexachlorobiphenyl (Hexa-CB)-amended sediment slurries.
 Experiments were conducted  under anaerobic conditions  using PCB-
 contaminated sediment from  the  Ashtabula River. Sterile control
 samples behaved identically to  the live control sample.
                                 31

-------
      G
      o>
      (0
      k.
      
-------
                          Saginaw River Sediment (Autoclaved Control)
       50.00
       40.00
     5 30.00
     o 20.00
       10.00
        0.00
             MONO   Dl    TRI   TETRA  PENTA   HEXA  HEPTA   OCTA  NONA   DECA
                                    HOMOLOG GROUPS
                                                                       B
                            Saginaw River Sediment (Live control)
       50.00
       40.00
       30.00
     3 20.00
       10.00
        0.00
             MONO   Dl
TRI   TETRA  PENTA  HEXA  HEPTA  OCTA  NONA   DECA
          HOMOLOG GROUPS
Figure  14. Reductive dehalogenation  of PCBs in (a)  sterile
control  and  (b) unamended  (live)  Saginaw  River sediments after
anaerobic incubation for 25  weeks.
                                      33

-------
                           Saginaw River Sediment + Arodor 1260
       30.00 T
       20.00
     £
     a.
       10.00
        0.00
            MONO    Dl    TRI   TETRA  PENTA  HEXA   HEPTA  OCTA  NONA  DECA

                                   HOMOLOG GROUPS
                       Saginaw River Sediment + RAMM + Aroclor 1260
                                                                       B
       30.00 r
       20.00
     o_
     UJ
       10.00
        0.00
             MONO   Dt     TRI   TETRA  PENTA  HEXA  HEPTA  OCTA   NONA   DECA
                                   HOMOLOG GROUPS
Figure 15.  Reductive  dehalogenation of  PCBs  in Saginaw  River
sediments  after  anaerobic  incubation  for 25  weeks  and amended
with (a) Aroclor 1260 and  (b)  Aroclor  1260 plus RAMM.
                                     34

-------
3.2  SUMMARY OF PAH BIODEGRADATION RESULTS

3.2.1  Assessment of potential for biodegradation of PAHS

3.2.1.1  PAH concentrations in homogenized sediment material

     The concentrations of PAHs in Buffalo River sediment are
given in Table 3.  The compounds present in highest
concentrations are pyrene and fluoranthene, both four-ring PAHs.
Other studies that have analyzed samples from several locations
in the Buffalo River have noted similar patterns of PAH
concentrations  (USAGE Buffalo District, 1993).   Thus, because it
is present at high concentrations, and for the reasons described
earlier, pyrene was selected as a model PAH for subsequent rate
experiments.


Table 3. Concentrations of PAHs in homogenized bulk Buffalo River
sediment.


Compound                      mg/kg  (dry weight)1      S.D.
2
Fluorene
Phenanthrene
Fluor anthene
Pyrene
Benz (a) anthracene
Chrysene
Benz (Jb) f luoranthene
Benz (k) f luoranthene
Benz (a) pyrene
Indeno (1,2, 3-d) pyrene
Dibenz (a) anthracene
Benz (g, h, i) perylene
0.37
5.83
12.10
11.66
2.55
5.26
0.37
0.08
1.28
0.53
ND
0.72
0.08
1.33
2.44
2.18
0.65
0.64
0.08
0.11
0.13
0.37

0.35
  Mean of triplicate determinations.
2 Standard deviation of triplicate determinations.


3.2.1.2  Aerobic degradation of PAHs by both indigenous bacteria
from Buffalo River sediments and enrichment cultures derived from
Buffalo River sediments

     Organisms from the Buffalo River are capable of aerobic
growth on biphenyl, naphthalene, acenaphthene, acenaphthylene,
fluorene, phenanthrene, and anthracene, as demonstrated by colony
formation and turbid growth in media derived from sediment
slurries  (10% w/v) amended with 10 mg/L of individual PAHs  (Table
4).  The fluorene and anthracene degraders produced very little
turbidity and no colonies on agar, which suggests very low

                                35

-------
activity  on those compounds. Aerobic growth  on phenanthrene
produced  an orange transformation product  in the medium, which
subsequently turned yellow, suggesting that  an oxidized product
may be a  metabolic intermediate, which is  further  oxidized.   One
microbial species may carry out the sequential metabolic
transformations, or the  intermediate compound may  be used  as  a
substrate by other organisms.  No bacterial  growth was detected
with fluoranthene, pyrene, chrysene, or benz(a)anthracene  as  the
sole substrate, or with  sediment organic extract as a co-
substrate.
Table 4.  Indicators of aerobic growth of indigenous
microorganisms from Buffalo River sediments on PAHs as  sole
substrate and with sediment organic matter as co-substrate.
Compound
MSB/PAH1
                                   MSB/PAH/
                                   Sediment
                                   Extract2
Isolated
Colonies on
PAH + MS Agar
Biphenyl
Naphthalene
Acenaphthene
Acenaphthylene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene
Benz(a)anthracene-
                 w
                 ++/orange
                 w
                  ++++/yellow
                  ++
 Growth in broth was determined visually (+++)   >( ++)   > (+)  >
 (w)  >  (-)  = turbid growth to no growth.  Growth on agar was
determined by observing colony formation.
2Sediment extract was prepared by adding 50g Buffalo River
sediment and 150ml water.  The mix was autoclaved, then shaken
for 24 hours on a wrist-action shaker.  Sediment was centrifuged
and the supernatant retained as sediment extract.

     When Buffalo River sediment slurries  (20% w/v) were amended
with a mixture of 16 PAHs (20 mg/kg each),  all of the compounds
for which degraders were detected (Table 4) were measurably
degraded after 6 weeks (Table 5).  Likewise, those PAHs for which
no degraders were detected were not degraded.
                                36

-------
Table 5. Aerobic biodegradation by indigenous Buffalo River
microorganisms of PAHs  (20 mg/kg) added to sediment slurries,


                         Concentration(mg/kg dry weight)

Compound                   Start                6 weeks



Acenaphthene                13.2                   1.1

Acenaphthylene               8.9                    ND

Fluorene                    16.3                   3.2

Phenanthrene                19.4                   4.1

Anthracene                  21.2                   4.6
xNo degradation of fluoranthene,  pyrene,  benzo(a)anthracene,
chrysene, benzo(a)pyrene, dibenz(a,h)anthracene, and
benzo(gfh,i)perylene occurred within the 6-week period.
2ND = not detected


     Table  6 shows that the enrichment culture degraded a wider
range of PAHs at a faster rate than did the unenriched sediment.
After 2 weeks of incubation, the degradation  of 4- ring PAHs by
the enrichment culture was more than double that of the sediment,
although neither inoculum was capable of transforming the 5- and
6- ring compounds during that period.
                                37

-------
Table 6. Comparison of rate of aerobic PAH degradation by an
enrichment culture added to Buffalo River sediment slurries to
that by indigenous organisms.
PAH Class                                 Concentration
                                         (mg/kg dry weight)

                              Start               14 days
                                        Enrichment     Indigenous
2-,
4-
5-,
3-
ring
6-
ring
ring
80.
58.
46.
9
9
3
8.
21.
43.
1
7
4
(58%)
( 0%)
13
44
42
.6
.7
.9
(77%)
(23%)
( 0%)
Values in parentheses are percentages of parent PAH compounds
lost.
3.2.2  Processes affecting the rate of aerobic biodegradation of
PAHs.

     To determine whether oxygen was limiting in flask
experiments incubated on a bench-top shaker, the rates of
degradation of PAHs using three methods of aeration were
compared: diffusion (stationary flasks),  agitation, and air
infusion  (sparging).   Sediment slurries,  either shaken or left
stationary, were monitored for phenanthrene mineralization and
for PAH degradation.   Mineralization was rapid and leveled off
quickly.  Thus measurements of mineralization were terminated
after 14 days; the rate was determined by the slope of the
initial increase.  The extent of PAH degradation was measured by
extracting the slurries after 28 days of incubation. The shaken
sediment slurry mineralized phenanthrene more rapidly  (7.68%/day)
and to a greater extent than did the stationary sediment slurry
(3.17%/day), as shown in Figure 16 and Table 7. However, the
difference in PAH degradation after 4 weeks between aerated
(shaken, sparged) and non-aerated  (stationary)  was small.  Air
sparging, using airstones to mix and aerate sediments, did not
increase the extent of degradation after 28 days.  The 2- and 3-
ring PAHs were readily degraded in all three treatments  (55%, 42%
and 44% respectively;  Table 8).   The 4-ring PAHs were also
degraded  (47%, 37% and 22%, respectively).   Although slightly
more loss of the 5- and 6-ring PAHs occurred after 28-day
incubations than after the 14-day incubation, they were
relatively resistant to degradation.

                                38

-------
Table 7. The effect of aeration method on the rate of degradation
of PAHs in Buffalo River sediment slurries.
PAH Class
  Concentration
(mg/kg dry  weight)
                 Start
                              Shaking
       28  days

       Stationary
Sparged
2-, 3- ring
4- ring
5-, 6- ring
29.9 14.1 (55%)1 18.0 (42%) 17.3 (44%)
42.0 22.5 (47%) 26.5 (37%) 33.1 (22%)
37.0 27.4 (16%) 30.1 ( 8%) 27.6 (15%)
Values in parentheses indicate percentage loss of parent PAH
compound.
                    STATIONARY

                    AERATED
Figure 16. Effect of aeration on the mineralization of
phenanthrene by the indigenous bacteria  in  10%  sediment slurries
prepared with Buffalo River  sediments. Values represent the mean
of triplicates.

                                39

-------
     The addition of nitrogen (1000 mg/L (NH4)2N03)and phosphorus
(1000 mg/L K2HPO4 and  200 mg/L KH2P04) , or phosphorus  alone,  in-
creased mineralization of phenanthrene.   However, addition of
nitrogen alone decreased the mineralization of phenanthrene
(Figure 17).   Sediment slurries without any inorganic nutrient
additions mineralized phenanthrene at nearly the same rate
(7.7%/day) as with nitrogen and phosphorus added (9.9%/day),
following an increased lag period.  Pyrene,  likewise, was
mineralized at comparable rates in media with or without
inorganic nutrient addition (Figure 18).

       Although pyrene can be mineralized by the enrichment
culture, the addition of sediment extract increased the initial
rate of degradation at all additions tested (Figure 19).  The
addition of sediment extract also transformed the shape of the
mineralization curve from logarithmic to first order, suggesting
that the presence of the sediment organic matter either shortened
an induction period or greatly increased the rate of growth of
the degrader bacteria.

     To test whether the increased rate of degradation of pyrene
in the presence of sediment organic extract was due to the growth
of a larger active degrader population,  different numbers of
bacteria were inoculated into sediment containing pyrene (Figure
20).  An inoculum of 106 produced rapid mineralization of pyrene,
and the first-order shape of the curve resembled that of the
previous experiment.  Smaller inocula (1 - 104)  resulted in much
slower mineralization.  The shape of the slower curves was nearly
linear, suggesting that growth either did not occur, or occurred
very slowly.   This observed threshold may result from failure of
the enrichment culture to successfully compete with indigenous
organisms when introduced at low inoculation densities.

     Both the rate and extent of mineralization of pyrene by the
Buffalo River enrichment culture decreased as the concentration
of sediment in the slurry increased (Figure 21) due to sorption
of pyrene on the sediment.  Triton X-100, a neutral surfactant,
was added to pyrene sorbed to sterile sediment.  In all cases,
that is, above and below the critical micelle concentration
(CMC), the presence of the surfactant reduced the rate of
mineralization of the compound by the enrichment culture (Figure
22).  The extent of inhibition was inversely related to the
amount of surfactant added.

     When the pyrene was sorbed to the sediment surfaces for a
period of time prior to the addition of the surfactant, and the
bacteria had utilized all of the available form of the compound
(as indicated by a leveling off of the mineralization curve), the
addition of surfactants did not consistently increase the amount
of PAH degraded  (Figure 23) .
                                40

-------
               Minimal (alts 
-------
Figure 19. Effect of addition of sediment organic matter on the
aerobic mineralization of pyrene by a Buffalo River enrichment
culture in MSB containing added organic carbon.  The cultures
were prepared with MSB and 100%, 10%, and 1% by volume of the
supernatant solution of a 10% sediment slurry prepared with MSB.
Values represent the mean of triplicates.
                           10
15
20
25
                               DAYS
Figure 20. Effect of inoculum size on the aerobic mineralization
of pyrene when Buffalo River enrichment culture was added to 10%
sediment slurry. Values represent the mean of triplicates.

                                42

-------
                              DAYS
Figure 21. Effect of sediment concentration on the aerobic
mineralization of pyrene in sediment slurries when Buffalo River
enrichment culture was added to 0%, 5%, 10%, 25%, and 50%
sediment slurries.  Values represent the mean of triplicates.
Figure 22. Effect of initial addition of Triton X-100 (CMC - 130
»g/L) on the aerobic mineralization of pyrene when Buffalo River
enrichment culture was added to 10% sediment slurry.  Triton X-
100 concentrations were 10%, 50%, 100%, and 300% of CMC.  Values
represent the mean of triplicates.

                                43

-------
     M
     60
    40
     20
     80
    '60
     40
     20
     60
     60
     40
     20
Figure 23. Effect of addition of Triton X-100  (0.5  CMC =62.5
mg/L) after the aerobic mineralization of  available pyrene in 10%
(A), 25%  (B) and 50% (C) sediment slurries to  which Buffalo River
enrichment culture was added. Surfactant was added  at day 15,
after mineralization has leveled off. Values represent the mean
of triplicates.
                                44

-------
The addition of Triton X-100 at day 15 produced an increase in
the subsequent rate of mineralization in sediment slurries of
either 25 or 50%, although no clear relationship between the
amount of surfactant added and the extent of degradation increase
was observed.  The linearity of the mineralization curve
following the addition of the surfactant suggests that a first-
order process, in this case, solubilization of the compound, is
controlling the rate of mineralization.

     Figure 24 shows that even very high concentrations of heavy
metals in the oxidized state do not affect the rate of
mineralization of pyrene by sediment bacteria.  The metals were
added as the most soluble complex under aerobic conditions, at
concentrations well above the concentrations measured in Buffalo
River sediments  (Table 8).


Table 8. Trace metal concentrations in Buffalo River sediments.
(from Dial, 1994)
Trace Metal                        mg/kg (dry weight)
     CU                             33
     Cd                              1.9
     Ni                             57
     Fe                              3.9
     Cr                            110
     Zn                            200
     Pb                             94
     Figure 25 shows that the degradation of pyrene by the
enrichment culture is slower in the presence of several
combinations of 2-, 3-, and 4- ring PAHs.  The enrichment culture
is able to mineralize pyrene; therefore, it can derive energy
from that compound.  The addition of 2- and 3- ring PAHs did not
increase the rate of growth, and hence the rate of degradation,
for the culture, but rather slowed the rate, suggesting that the
smaller compounds may be preferred as substrates by the bacteria.
Additions of 4-ring PAHs repressed pyrene mineralization to a
greater extent than did the additions of the smaller compounds.
The inhibition caused by additions of both classes of PAHs was
partially ameliorated by the pretreatment of the sediment slurry
with hydrogen peroxide  (7.5 ml per flask).
                                45

-------
                                 DAYS
Figure 24. Effect of additions  of  heavy metals (X = 25 ug/ml Pb,
25 ug/ml Cu and 2.5 ug/ml basal concentration) on the aerobic
mineralization of pyrene when Buffalo River enrichment culture
was added to 10% sediment slurry.   Metal concentrations were 0,
IX, 10X, and 100X. Values represent the mean of triplicates.
               2-,3- ring

               4-ring

               2-. 3- ring + H202

               4-ring + H202

               Control
Figure 25. Effect of additions  of  secondary PAHs on the aerobic
mineralization of pyrene when Buffalo River enrichment culture
was added to 10% sediment  slurry,  with and without pretreatment
with hydrogen peroxide. Values  represent the mean of triplicates.
                                46

-------
3.3  DISCUSSION AND INTERPRETATION

3.3.1  PCB Biodegradation

     The Sheboygan River sediments used in this study were
originally contaminated with hydraulic fluids containing Aroclors
1248 and 1254 (Sonzogni and David, 1991) ;  the most prevalent PCB
congeners in a mixture of Aroclors 1248 and 1254 are tetra-,
penta- and hexachlorobiphenyls.   Our initial analyses of the PCB
congener profile of Sheboygan River sediment used in this study
revealed a higher concentration of lower chlorinated congeners
and a lower concentration of highly chlorinated congeners than
would be found in the original contaminating material. This
indicated that the PCB contaminants had been transformed since
their discharge.  Further, based on the distribution and weight
percentages of indigenous PCB congeners in contaminated Sheboygan
River sediments, Sonzogni and David (1991) suggested that
microbial transformation of PCBs occurred in situ.

     Unamended sediment slurries prepared from Sheboygan River
sediments exhibited no appreciable changes in PCB homolog
distribution after 30 weeks of anaerobic incubation. These
results suggest that rates of reductive dechlorination of the
historical PCB congeners are extremely slow and it is likely that
some factor  (nutritional, physiological,  physicochemical) limits
PCB degradation in natural sediments.  In contrast to the
unamended Sheboygan River sediments, addition of Octa-CB to these
sediments resulted in the accumulation of ortho-substituted
congeners, indicating that dechlorination had occurred at the
meta and para positions.  This dechlorination pattern has also
been described in laboratory incubations of actively
dechlorinating, PCB-contaminated Hudson River sediments  (Brown et
al.r 1987a,b; Ye et al., 1992).  In our experiments, the pattern
of dechlorination of historical PCBs in sediments with added
Octa-CB was similar; the concentration of PCB homolog groups with
three or more chlorine atoms decreased with time and the major
products were mono- and di-CBs.  These results illustrate the
preferential dechlorination of the more highly chlorinated PCB
congeners by the indigenous anaerobic microorganisms. The rate of
dechlorination of Octa-CB was estimated in experiments with 5 to
20 mg/L of added Octa-CB, and the results indicated that
dechlorination was directly proportional to the concentration of
Octa-CB added.

     In a previous investigation using Sheboygan River sediments,
reductive dechlorination of PCBs was not observed in experiments
in which bacteria eluted from contaminated sediments were used as
the inoculum  (Sonzogni and David, 1991).  Results from our study
indicate that naturally occurring microorganisms mediate the
reductive dechlorination of both amended and historical PCBs in
Sheboygan River sediments incubated anaerobically under
laboratory conditions. It has been suggested that different

                                47

-------
physiological groups of dechlorinating bacteria are present in
PCB-contaminated sediments (Quensen et al.,  1990; Ye et al.,
1992).  The use of 'eluted' microorganisms as inocula in
dechlorination experiments may exclude some microbial populations
that are attached to sediment matrices and that may be important
for transformation of complex organic chemicals. This hypothesis
can only be tested through additional studies with PCB-
contaminated sediments in which reductive dechlorination has been
demonstrated.

     Addition of PCB congeners to Ashtabula River sediments also
resulted in enhanced dechlorination of historical PCBs as was
observed for Sheboygan River sediments. The congeners added in
this study, Penta-CB, Hexa-CB and Hepta-CB,  contained one, one,
and four ortho-chlorines per biphenyl, respectively; the onset of
dechlorination of these congeners, either alone or in
combinations, occurred prior to dechlorination of the historical
PCBs. Although all amended congeners were dechlorinated in
Ashtabula River sediment slurries, it was expected that Hepta-CB
would be more resistant to anaerobic transformation than the
other amended congeners since reductive dechlorination of meta
and para chlorines are preferred. This was not the observed
result. The onset of dechlorination of the amended congeners
occurred in the order: Hexa-CB, followed by Hepta-CB and Penta-
CB.  A possible explanation for the longer lag period before
onset of dechlorination of Penta-CB is the lack of a significant
population of Penta-CB dechlorinating microorganisms in the
sediment.  In experiments with mixtures of the amended congeners,
the results are less clear. Because /neta-dechlorination of the
Hexa-CB congener on the less chlorinated ring produces the
identical Penta-CB congener used in our studies, the calculated
rates of dechlorination may be biased  (an underestimate for
Penta-CB dechlorination) and account for the apparent slower rate
of dechlorination of Penta-CB in the presence of Hexa-CB.  The
slower degradation rate of Hepta-CB may be attributed to the
presence of four ortho-chlorine atoms, which likely makes this
congener more resistant to anaerobic biotransformation.

     The results from the present study demonstrate the
dechlorination capacity of PCB-contaminated Sheboygan River and
Ashtabula River sediments and support the theory of Sonzogni and
David  (1991)  that a threshold PCB concentration exists for
reductive dechlorination of chlorobiphenyls.  No appreciable
dechlorination of historical PCBs was observed in unamended
sediment slurries. Three explanations are proposed for the
stimulation of reductive dechlorination of historical PCBs in
sediments by addition of specific PCB congeners: (a) the
bioavailability of PCBs was enhanced, thus providing an available
electron acceptor for oxidation reactions;  (b) the growth of
indigenous PCB dechlorinating microorganisms was stimulated;  (c)
amended PCB congeners induced dechlorinating activity of
indigenous microbial populations. It has recently been

                                48

-------
demonstrated that the reductive dechlorination of 3-
chlorobenzoate by the anaerobic isolate Desulfomonile tiedjei
provides energy for growth (Dolfing, 1990; Mohn and Tiedje,
1990).  If microorganisms that reductively dechlorinate PCBs
derive energy from the process, then this would provide a
selective advantage for their growth in PCB-contaminated
environments. Further experiments are necessary to examine the
questions of PCB bioavailability, growth of dechlorinating
populations in historically contaminated sediments, and to
determine the threshold concentration for PCB biotransformation.

3.3.2  Aerobic PAH Biodegradation

     The present-day concentrations of PAHs in Buffalo River
sediments indicate that some transformation of PAHs has probably
occurred.  Ideally, residual concentrations should be compared to
a previously measured concentration, or absent that, a weight-
averaged mass loading, to determine whether losses of compound
have occurred.  Historical data for Buffalo River are sparse and
variable spatially.  Nonetheless, patterns of compound relative
abundance indicate biological activity.  The lower molecular
weight and more soluble compounds, such as naphthalene, may be
lost from sediment by either biological or physical processes,
but the loss of the 3- and 4-ring compounds relative to the 5-
and 6- ring compounds is an indication of biological activity.

     PAHs may persist in sediments due to limitations of
bacterial metabolic activities or because the compounds are
present in a form that is not available to the bacteria.
Bacterial activity may be limited by inherent me tabolic
capability, by inadequate supplies of nutrients and/or electron
acceptors, or by inhibitory interactions with other organisms
(e.g. competition, predation) or other compounds  (e.g. toxic
metals).

     The native microorganisms of the Buffalo River demonstrate
the capability to degrade PAHs in sediment slurries amended with
PAHs, yet those compounds persist in river sediments.  One
possible explanation for the restricted activity  of indigenous
bacteria in sediments is the limitation of growth of the degrader
population by other factors, such as competition  with other
organisms for substrates or nutrients.   Alternatively,
indigenous bacteria may not express the metabolic capability to
degrade some of the PAH compounds in the presence of preferred
substrates.

     PAHs of less than five rings were degraded by an enrichment
culture derived from Buffalo River sediments.  These bacteria
utilized a greater range of compounds than did the indigenous
organisms, and were able to grow using some PAHs  as energy-
yielding substrates.  Rates of transformation of  PAHs by the
enrichment culture were faster than the rates for indigenous

                                49

-------
bacteria.  Since the enrichment culture was derived from the
Buffalo River sediment, this suggests that the genotypes for PAH
degradation are present in the sediment, yet some environmental
conditions prevent their natural growth and/or expression.

     It has been well established that PAH biodegradation is far
more efficient under aerobic conditions.  However, except for the
surface layer, sediments are primarily anaerobic environments.
Moreover, efficient aeration is difficult to attain under in situ
field conditions, where sediment oxygen demand often exceeds
mixing capabilities.  Without appropriate physical constraints,
the high rates of mixing needed to achieve aeration under field
conditions result in unacceptable levels of sediment
resuspension.  Under such conditions, alternate oxygen delivery
systems, such as liquid oxidants, may be more efficacious.
Mechanical aeration is more feasible for engineered systems.

     Experiments indicated that degradation was slightly faster
in aerated slurries than in non-agitated slurries.  Dilute
slurries were tested in these experiments; the differences
between aerated and non-aerated systems is expected to be more
pronounced  in the more concentrated slurries characteristic of
field conditions.  On the other hand, aeration of slurries by
sparging did not increase the rates of degradation; rather, a
slight decrease was observed.  This suggests that not only is
there a maximal aeration rate corresponding to other rate
limitations on the compound transformations, but that higher
rates of agitation of the sediment slurries may be detrimental to
microbial activity by shearing bacteria from sediment surfaces.

     Mineralization of PAHs indicates complete conversion of the
parent compound to C02  and  H20, yielding energy for growth of the
bacteria.  There have been few reports of the mineralization of
PAHs with four or more aromatic rings; most transformations of
larger PAHs reported in the literature have been supported by co-
metabolism of other substrates from such sources as sediment
organic matter.  The second substrate may be required by the
organisms as a source of energy,  either because the metabolism of
the PAH does not yield energy or because the compound is so
insoluble that too little is available to the bacteria to support
growth and/or activity.

     In the former case, cometabolism may result in incomplete
transformation of the parent compound.  Cometabolism of PAHs in
the environment may result in the accumulation of toxic by-
products.   Thus, the enhancement of mineralization of these
compounds by the introduction of enriched degrader bacteria may
reduce the residual toxicity associated with their partial
transformations.  The higher rates of mineralization of pyrene
resulting from the additions of sediment-derived organic matter
suggest that additional substrates may be utilized by the
enrichment culture to enhance growth.  However, the concurrent

                               50

-------
mineralization of the pyrene indicates that such utilization of
additional substrates does not preclude PAH degradation, but
rather enhances it to a slight extent.

     PAH compounds may affect the metabolism of other PAH
compounds in a number of ways, including inhibition (or toxicity)
or cometabolism.  Most of the large-ring PAHs are mutagenic
and/or carcinogenic; they are toxic to bacteria as a result of
their intercalation in cell membranes.  Such toxicity may
contribute to their persistence in the environment.  Conversely,
given the importance of cometabolism in the degradation of larger
PAHs, it is possible that the smaller PAHs may support
cometabolism of the larger PAHs by providing energy to the
degrader bacteria.  In these investigations, the additions of
additional PAHs, either 2-, 3- or 4- rings, inhibited the
mineralization of pyrene by the enrichment culture.

     Additions of nitrogen did not significantly enhance the rate
of mineralization of PAHs, which suggests that nitrogen is not
growth-limiting in Buffalo River sediments.  Phosphate additions,
either alone or with nitrogen (as MSB) increased the rate and
extent of degradation.  In the absence of additional phosphorus,
20% mineralization is not achieved in 5 days; with phosphorus,
that level of mineralization is achieved in 3 days, suggesting
that phosphorus additions either decrease the acclimation period
for degradative activity or increase the rate of growth of the
bacteria.

     When degradation results from the activity of the enrichment
culture alone, that is, in sterile sediment, degradation is
proportional to bacterial numbers, but when the enrichment
culture is added to sediment with a viable indigenous bacterial
population, a threshold population of the enriched bacteria must
be achieved to effect degradation.  At population densities below
106,  the mineralization of pyrene was slow and nearly linear,
even if incubation times were long enough to allow growth of the
smaller populations.  The inability of the enrichment to grow may
have resulted from competition with indigenous sediment organisms
or predation by sediment protozoa.  Ramadan et al. (1990) have
reported a similar occurrence; the introduction of a p-
nitrophenol degrading pseudomonad into nonsterile lake water was
unsuccessful at population densities below 4 x 104.   Given the
much higher densities of indigenous microflora in sediment, it  is
reasonable that the threshold density for successful introduction
is higher in sediment than in lake water.  Thus, bioaugmentation
of degrader bacteria into sediments may only be successful if
large numbers of active bacteria can be introduced and
maintained.

     Several other processes could be manipulated to increase the
rate of degradation, but the system response was not
straightforward.  The concentration of sediment in the  slurry was

                                51

-------
critical in determining the rate of the reaction.  An optimal
concentration resulted from the competing influences of sediment
as a source of nutrients and organic co-substrates to the
degrader bacteria and as a sorbent that decreased the
availability of the PAHs for degradation.  In the Buffalo River
sediments used in this study, the optimal slurry concentration
was approximately 5%; this number will likely be different for
different sediments.

     Sediments are frequently contaminated with high
concentrations of heavy metals.  In river bottom sediments, due
to the prevailing anoxic conditions, metals are predominantly in
their reduced state.  Most heavy metals are non-toxic in the
reduced form, but may be toxic in the oxidized state.
Biodegradation of PAHs requires oxygen and their transformation
has been shown to proceed most rapidly under well-aerated
conditions.  Thus, those conditions that best favor
biodegradation of PAHs will also convert sediment metals to their
most potentially toxic states.  When soluble forms of metals were
added at concentrations that exceeded the ambient concentrations
in Buffalo River sediments, no consistent inhibition of
degradation occurred. This suggests that the bacteria may have a
mechanism for resistance to trace metals in sediments, or that
the metals may be made unavailable, and hence nontoxic, to the
bacteria by another mechanism.

     The rate of degradation of the larger PAHs may be limited by
their availability to the bacteria.  Their low solubilities
dictate that very low concentrations will exist in the aqueous
phase; unless the bacteria can enhance the rate of
solubilization, the available concentrations of compound may be
too low to support microbial growth.

     The addition of a nontoxic surfactant slightly increased the
availability of sediment-bound PAHs if the surfactant was added
after the bulk of the available compound had been degraded.  When
the surfactant was added prior to initiation of mineralization of
pyrene, degradation was inhibited; it is likely that the
surfactant itself was utilized by the bacteria as a growth
substrate, precluding adaptation of the bacteria to metabolism of
the PAH.  The observed increase in turbidity supports this
explanation.  Tsomides et al.  (1995) tested the effects of
several nonionic surfactants on the degradation of PAHs sorbed to
sediments; all of the surfactants tested, except for Triton X-
100, were toxic to the bacteria and inhibited degradation of
phenanthrene.

     Although surfactants can increase the apparent solubility
and degradability of PAHs from solid crystalline form  (Volkering
et al., 1995), there is little evidence for enhancement of
degradation of PAHs sorbed to sediments.   Additions of Triton X-
100, at concentrations well above the CMC, increased the apparent

                                52

-------
solubility of phenanthrene but did not significantly increase its
mineralization.  The compound within the surfactant micelle
structure is apparently not available for biodegradation
(Volkering et al., 1995; Roch and Alexander, 1995). On the other
hand, an increase in biodegradation by the addition of
surfactants at sub-CMC concentrations has been reported
(Aronstein et al., 1991); the mechanism for this result has not
been elucidated.

     PAHs appear to be less available for degradation after long
contact with soils or sediments.  Hatzinger and Alexander (1995)
report that, when phenanthrene was incubated with sterile soils
for varying lengths of time prior to inoculation with a bacteria
capable of degrading the compound, longer incubation times
resulted in less degradation.  The extent of degradation
corresponded to the amount of phenanthrene that could be
extracted with organic solvents or by exposure to earthworms, and
this "available" fraction decreased with soil resident time.
This phenomenon is known as aging, and is more pronounced in
soils with higher organic content. The addition of nutrients
slightly increased the rate and extent of degradation in both
aged and unaged soils, but did not reverse the effects of aging.
The authors concluded that aging may be due to slow diffusion of
nonionic organic chemicals within the solid-phase organic matter
in soils, or to entrapment and retarded diffusion within the
small pores in soils aggregates.  Sonication increased the
initial rates of degradation in aged and unaged soils, but did
not appreciably increase the extent of degradation.  This strong
association between an organic contaminant and soil is likely to
limit the effectiveness of in situ bioremediation. Given the high
organic content of Buffalo River sediments, it is very likely
that aging of PAHs has rendered them resistant to degradation via
a similar mechanism.

     Most of the processes found to increase the rate of
degradation of PAHs could be utilized either in situ or in an ex
situ engineered facility.  Nutrients, oxidants, and competent
bacteria could be added to sediments in situ with injection pumps
or with the construction of semi-enclosed weirs.  Increasing the
availability of the more recalcitrant PAHs, on the other hand,
requires the formation of less concentrated sediment slurries and
enhancement of PAH availability, either by disruption of sediment
particulate matter  (e.g. sonication) or the addition of
surfactants.  Such treatments can best be carried out in ex situ
contained facilities.
                                53

-------
4.0  CONCLUSIONS
     Several conclusions can be made from these studies.   They
are:

•    For the three PCB contaminated sediments examined, intrinsic
     rates of PCB degradation were not observed during the course
     of the investigation under the experimental approaches used
     herein.  No evidence for further PCB dechlorination was
     detected in these historically contaminated sediments
     despite evidence that PCB dechlorinating microorganisms were
     present.

•    Reductive dechlorination of the historical PCBs were
     stimulated by the addition to the sediments of single or
     multiple PCB congeners containing five or more chlorines.
     Both the added PCB congener(s) as well as congeners already
     present in the sediments were dechlorinated primarily to
     mono- and di-chlorinated PCBs.

•    Under aerobic conditions, microorganisms native to Buffalo
     River sediments mineralized PAHs with less than five rings
     and partially degraded some PAHs with five and six rings.
     The activity of native bacteria can be increased by
     enrichment on specific substrates and reintroduction to
     contaminated sediment (bioaugmentation).

•    Degradation of PAHs was not stimulated by the addition of
     surfactants at levels above or below the CMC.  Surfactants
     generally inhibited PAH degradation in this study.

•    Microbial degradation of PAHs was not limited by nitrogen
     and was not inhibited by potentially toxic metals also found
     in these sediments.  Inhibition of degradation by the large
     ring PAHs may be partially alleviated by treatment with
     peroxide.

•    Bioaugmentation of PAH-degrading bacteria into PAH-
     contaminated sediments may only be successful if large
     numbers of active microorganisms can be introduced and
     maintained in an active state.

     An evaluation of these results with respect to effective
bioremediation of the contaminated sediments suggests to the
authors that in situ treatment of these contaminated sediments is
not practical.  Removal of the sediments to a treatment facility
or alternative treatment location would be required.  PCBs could
be treated by a combination of anaerobic pretreatment, to convert
the highly chlorinated PCB congeners to congeners containing one
or two chlorines, followed by aerobic degradation of the less
chlorinated PCB congeners.  Treatment of PAH contaminated

                                54

-------
sediments could best be accomplished and controlled by aerobic,
above-ground treatment systems.  Bioaugmentation with competent
PAH degrading bacteria is an alternative treatment option for PAH
remediation but may be costly due to the large amount of
microbial inocula required for effective and timely treatment.
Complete degradation of high MW PAHs will require treatment to
enhance their availability to bacteria.

5.0  REFERENCES

Abramowicz, D. A., M. J. Brennan, and H. M. van Dort. 1990.
Anaerobic and aerobic biodegradation of endogenous PCBs, p. 55-
69. In Research and Development Program for the Destruction of
PCBsr Ninth Progress Report. General Electric Co. Corporate
Research and Development, Schenectady, N.Y.

Aronstein, B. A., Y. M. Calvillo and M. Alexander. 1991. Effects
of surfactants at low concentrations on the desorption and
biodegradation of sorbed aromatic compounds in soil. Environ.
Sci. Technol. 25:1728-1731.

Bedard, D. L., S. C. Bunnell, and H. M. van Dort. 1990. Anaerobic
dechlorination of endogenous PCBs in Woods Pond sediment, p.  43-
54. In Research and Development Program for the Destruction of
PCBsr Ninth Progress Report. General Electric Co. Corporate
Research and Development, Schenectady, N.Y.

Bedard, D. L. and H. M. Van Dort. 1992.  Brominated biphenyls can
stimulate reductive dechlorination of  endogenous Aroclor 1260 in
methanogenic sediment slurries. Abstr. 92nd Gen  Meet. Amer.  Soc.
Microbiol., p 339.

Brandon, D. L., C. R. Lee, J. W. Simmers, H. E. Tatem and J.  G.
Skorgerbee. 1991. "Information Summary, Area of Concern: Saginaw
River and Saginaw Bay." Miscellaneous  Paper EL-91-7, US Army
Engineer Waterways Experiment Station, Vicksburg, MS.

Brown, J. F., Jr., D. L. Bedard, M. J. Brennan, J. C. Carnahan,
H. Feng, and R. E. Wagner. 1987a. Polychlorinated biphenyl
dechlorination in aquatic sediments. Science 236:709-712.

Brown, J. F., R. E. Wagner, H. Feng, D. L. Bedard, M. J. Brennan,
J. C. Carnahan, and R.  J. May. 1987b.  Environmental
dechlorination of PCBs. Environ. Toxicol. Chem. 6:579-593.

Cerniglia, C. E., and M. A. Heitkamp.  1989.  Microbial
degradation of polycyclic aromatic hydrocarbons in the aquatic
environment.  In Metabolism of polycyclic aromatic hydrocarbons  in
the  aquatic environment. U. Varanasi,  ed., CRC Press, Boca Raton
FL,  pp.  41-68.
                                55

-------
Dagley, S. 1981. New perspectives in aromatic catabolism. In
Microbial Degradation of Xenobiotics and Recalcitrant Compounds,
Leisenger, T., R. Hutter, A. M. Cook, and J. Nuesch, eds.,
Academic Press pp. 181.

David, M. M. 1990. PCB Congener Distribution in Sheboygan River
Sediment, Fish and Water. M.S. thesis. University of Wisconsin,
Madison, WI.

Dial, C. J. 1994. Bench-scale evaluation of RCC's Basic
Extractive Sludge Treatment (B.E.S.T.) Process on contaminated
sediments from the Buffalo, Grand Calumet and Saginaw Rivers.
U.S. Environmental Protection Agency, Cincinnati, OH. Publication
No. EPA-905-R94-010.

Dolfing, J. 1990. Reductive dechlorination of 3-chlorobenzoate is
coupled to ATP production and growth in an anaerobic bacterium,
strain DBC-1.  Arch. Microbiol. 153:264-266.

Eleder, B. L. 1991. Sheboygan River and Harbor, Sheboygan,
Wisconsin. In Biological Remediation of Contaminated Sediments,
with Special Emphasis on the Great Lakes. Jafvert, C. T. and J.
E. Rogers (eds.). U.S. Environmental Protection Agency, Athens,
GA. Publication No. EPA/600/9-91/001.

Goudy, G. 1991. Saginaw River/Bay Area of Concern.  In Biological
Remediation of Contaminated Sediments, with Special Emphasis on
the Great Lakes.  Jafvert, C.  T. and J. E. Rogers (eds). US
Environmental Protection Agency, Athens, GA. Publication No.
EPA/600/9-91/001

Harkness, M. R., J. B. McDermott, D. A. Abramowicz,  J. J. Salvo,
W. P. Flanagan, M. L. Stephens, F. J. Mondello, R. J. May, J. H.
Lobos, K. M. Carroll, M. J. Brennan, A. A. Bracco, K. M. Fish, G.
L. Warner, P. R. Wilson, D. K. Dietrich, D. T. Lin,  C. B. Morgan,
and W. L. Gately. 1993. In situ stimulation of aerobic PCB
biodegradation in Hudson River sediments. Science 259:503-507.

Hatzinger, P. B., and M. Alexander.  1995.  Effect of aging of
chemicals in soil on their biodegradability and extractability.
Environ. Sci. Technol. 29:537-545.

Jafvert, C. T., and J. E. Rogers. 1991. Biological Remediation of
Contaminated Sediments, With Special Emphasis on the Great Lakes.
Report of a Workshop. US Environmental Protection Agency
Publication No. EPA/600/9-91/001.

Lee, C. R., D. L. Brandon, J.  W. Simmers, H. E. Tatem, and J. G.
Skogerboe. 1991.  Information Summary, Area of Concern: Buffalo
River, New York.  Miscellaneous Paper EL-91-9, Department of the
Army, Waterways Experiment Station, Corps of Engineers, Vicksburg
MS.

                                56

-------
Mihelcic, J. R. and R. G. Luthy. 1988. Microbial degradation of
acenaphthene and naphthalene under denitrifying conditions in
soil-water systems. Appl. Environ. Microbiol. 54:188-1198.

Mohn, W. W. and J. M. Tiedje. 1990. Strain DBC-1 conserves energy
for growth from reductive dechlorination coupled to formate
oxidation. Arch. Microbiol. 153:267-271.

Mueller, J. G., S. E. Lantz, B. O. Blattman, and P. E. Chapman.
1991. Bench scale evaluation of alternative biological treatment
processes for the remediation of pentachlorophenol- and creosote-
contaminated materials:  solid-phase bioremediation. Environ. Sci.
Technol. 25:1045-1055.

Mullin, M.D. 1985. Report from PCB Workshop. US EPA Larger Lakes
Research. Groslsle, MI.

Nies, L. and T. M. Vogel. 1990.  Effects of organic substrates on
dechlorination of Aroclor 1242 in anaerobic sediments.  Appl.
Environ. Microbiol. 56:  2612-2617.

Quensen, J. F., Ill,  J.  M. Tiedje, and  S. A. Boyd. 1988.
Reductive dechlorination of polychlorinated biphenyls by
anaerobic microorganisms from sediments. Science 242:752-754.

Quensen, J. F., S. A. Boyd and J. M.  Tiedje.  1990.
Dechlorination of four commercial polychlorinated  biphenyl
mixtures  (Aroclors) by anaerobic microorganisms from sediments.
Appl. Environ. Microbiol. 56: 2360-2369.

Ramadan, M. A., O. M. El-Tayeb, and M.  Alexander.  1990.   Inoculum
size as  a factor  limiting the success of inoculation for
biodegradation. Appl. Environ. Microbiol. 56: 1392-1396.

Sanders, P. 1991. Fields Brook Superfund Site/Ashtabula River
Area of  Concern.  In Biological Remediation  of Contaminated
Sedimentsf  with Special  Emphasis on the Great Lakes. Jafvert,  C.
T.  and J. E. Rogers  (eds). US Environmental Protection Agency,
Athens,  GA. Publication  No.  EPA/600/9-9I/001.

Shelton, D. R. and J. M. Tiedje. 1984.  General method for
determining anaerobic biodegradation  potential. Appl. Environ.
Microbiol.  47:850-857.

Skogerbee,  J.  G., C.  R.  Lee, D. L. Brandon, J. W.  Simmers and  H.
E.  Tatem.  1991. Information  summary,  Area of Concern: Sheboygan
River, Wisconsin. Miscellaneous Paper EL-91-6.  US Army Engineer
Waterways Experiment  Station, Vicksburg, MS.
                                57

-------
Sonzogni, W. C. and M. M. David.  1991. PCB dechlorination in the
Sheboygan River. In Biological Remediation of Contaminated
Sediments, with Special Emphasis on the Great Lakes. Jafvert, C.
T. and J. E. Rogers (eds.). U.S. Environmental Protection Agency,
Athens, GA. Publication No. EPA/600/9-91/001.

Tsomides, H.J., J.B. Hughes, J.M. Thomas, and C.H. Ward.  1995.
Effect of surfactant addition on phenanthrene biodegradation in
sediments.  Environ. Toxicol. Chem. 14:953-959.

USAGE Buffalo District. 1993. Pilot-scale Demonstration of
Thermal Desorption for the Treatment of Buffalo River Sediments.
U. S. Environmental Protection Agency Great Lakes National
Program Office, Chicago, IL.  Publication No. EPA/905/R93/005.

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.

Volkering, F., A.M. Breure, J.G. van Andel, and W.H. Rukens.
1995.  Influence of nonionic surfactants on bioavailability and
biodegradation of polycyclic aromatic hydrocarbons.  Appl.
Environ. Microbiol. 61:1699-1705.

Wild, S. R., J. P. Obbard, C. I. Munn, M. L. Berrow, and K. C.
Jones. 1991. The long-term persistence of polynuclear aromatic
hydrocarbons (PAHs) in an agricultural soil amended with metal-
contaminated sewage sludges.  Sci. Total Environ. 101:235-253.

Wubah, D. A. and J. E. Rogers. 1992. Reductive dechlorination of
three highly chlorinated polychlorinated biphenyl (PCB) congeners
by sediments from a PCB-contaminated Great Lakes site. Amer. Soc.
of Microbiology Annual Meeting.  New Orleans, LA. Abstr. #Q31.

Ye, D., J. F. Quensen, J. M. Tiedje, and S. A. Boyd.  1992.
Anaerobic dechlorination of polychlorobiphenyls  (Aroclor 1242) by
pasteurized and ethanol-treated microorganisms from sediments.
Appl. Environ. Microbiol. 58: 1110-1114.

Young, L. Y. 1984. Anaerobic degradation of aromatic compounds.
In Microbial Degradation of Organic Compounds, D. T. Gibson (ed).
Marcel Dekker Inc., New York, New York, pp.487-523.
                                58

-------
6.0
 APPENDIX
                        QUALITY ASSURANCE PROJECT PLAN

                     BIOLOGICAL REMEDIATION OF PCB AND PAH
                            CONTAMINATED SEDIMENTS

                     U.S. ENVIRONMENTAL PROTECTION AGENCY

                         ECOSYSTEMS RESEARCH DIVISION

                     NATIONAL EXPOSURE RESEARCH LABORATORY

                                ATHENS,  GEORGIA
Approved:
W. J. Jont
Ecosyster
	LA
/sterns^
           UU-.
           -Principal Investigator
          Urch Division,  Athens GA
             -Principal/Investigator
                             , Athens  GA

                             ^LJ/f  .
R.
  Josystems
           Ss
    '.." Iftge'rs,  Projeict Manager
Ecosystems Researehr Division,  Athens  GA
                                       59

-------
                               TABLE OF CONTENTS









INTRODUCTION  	  61




PROJECT DESCRIPTION  	  61




PROJECT ORGANIZATION AND RESPONSIBILITY	  61




QUALITY ASSURANCE OBJECTIVES 	  62




SAMPLING PROCEDURES  	  62




SAMPLE CUSTODY	  63




INSTRUMENT CALIBRATION PROCEDURES AND FREQUENCY	  63




ANALYTICAL PROCEDURES  	  65




DATA REDUCTION, VALIDATION,  AND REPORTING 	  71




INTERNAL QUALITY CONTROL CHECKS 	  72




PERFORMANCE AND SYSTEM AUDITS 	  72




PREVENTIVE MAINTENANCE 	  72




ROUTINE DATA ASSESSMENT PROCEDURES	  73




CORRECTIVE ACTION  	  74




REFERENCES 	  75




TABLES	  76
                                       60

-------
                                 INTRODUCTION

      The Biological Remediation of PAH and PCB Contaminated Sediments project
is a research activity of the Biology Branch of the EPA Environmental Research
Laboratory in Athens,  Ga.  The Biology Branch has a long history in research
on the degradation and remediation of hazardous organic wastes under anaerobic
and aerobic conditions.  The Athens laboratory researchers are working with
the Engineering/Technology Workgroup of the Assessment and Remediation of
Contaminated Sediments (ARCS) program and the Bureau of Mines to investigate
bioremediation technologies as remediation alternatives for contaminated
sediments.

      The ARCS program is authorized as part of the Water Quality Act of 1987
and is administered by the Great Lakes National Program Office of the EPA.
The two major objectives of the ARCS program are to assess the extent of
sediment pollution in designated areas of concern and to identify and
demonstrate options for the removal and/or treatment of the contaminated
sediments.  The ARCS program is to be completed in the years 1988-1992.

      In 1991, Congress directed the Department of the Interior, Bureau of
Mines (BOM) to investigate quality of bottom sediments in the North Branch of
the Chicago River, and the potential for remediation of contaminated
sediments.  As part of this investigation, the BOM coordinated its research
with other agencies and groups, including the U.S. Army Corps of Engineers,
the Metropolitan Water Reclamation District of Greater Chicago, the Northeast
Illinois Planning Commission, Friends of the River and the U.S. EPA.  The
current EPA research is supported by an interagency agreement with the BOM.

      The Athens ERL effort will be predominantly laboratory research, with
pilot and field studies to be designed based on outcomes of laboratory
studies.


                             PROJECT DESCRIPTION

      During a workshop held July 1990 (sponsored by the ARCS program and
Environment Canada), an initial review was made of bioremediation methods and
options for their use in reducing contamination of Great Lakes sediments.  The
workgroup determined that the biological remediation technologies for these
classes of organic contaminants warranted examination by the ARCS Program
because the technologies were in rapidly advancing stages of development.
Major contaminants of interest included polychlorinated biphenyls (PCBs) and
polyaromatic hydrocarbons  (PAHs).  PCB analytes will include total PCBs as
well as individual PCB congeners.  PAH analytes will include, but not be
restricted to, benz [a] anthracene-, benzo [b] fluoranthene, benzo [k] fluoranthene,
benzo[a]pyrene and chrysene.

      PAHs and metals are the contaminants of concern in the North Branch of
the Chicago River. In complementary research, the BOM will investigate
bioremediation as an option  for the treatment of metals in sediments, and the
EPA will concentrate on the  environmental factors that influence the rate of
PAH degradation, with an emphasis on those factors that can be managed to
enhance the rate and range of biotransformation in engineered systems.


                    PROJECT ORGANIZATION  AND RESPONSIBILITY

      The Athens ERL is one  of twelve research laboratories operated by the
USEPA Office of Research and Development.  It is headed by a Director, with a

                                      61

-------
Program Operations Director and a Research Operations Director.   The
individual research groups report through Branch Chiefs to the Research
Director.

      The Biological Remediation of PAH Contaminated Sediments project is
under the direction of Dr. John E. Rogers, Project Manager.  Day-to-day
technical direction is coordinated by Dr. Rochelle Araujo and Dr. W. Jack
Jones, Co-Principal Investigators.  Independent quality assurance oversight
activities will be performed by the Biology Branch QA Officer, Mr. John Pope,
and the AERL QA Officer, Dr. Robert R. Swank.   Project staff includes a
microbiologist and a post-doctoral researcher.


                         QUALITY ASSURANCE OBJECTIVES

      The purpose of this quality assurance project plan is to describe the
procedures to be taken to assure that the measurement data produced are of
known quality based upon their intended use.

Completeness - The objective for completeness  in this research effort is 90
percent.

Representativeness - After the samples have been mechanically mixed to visible
homogeneity, representative subsamples from each sample lot will be extracted
and analyzed for comparison upon sample characterization.  Buffalo River and
the North Branch of the Chicago River sediment are representative of PAH
contaminated sediments. These two areas represent possible field demonstration
sites. The Saginaw River/Bay sediment is representative of a PCB contaminated
sediment.

Comparability - For this project, comparability for each test will be assured
through use of the same analytical procedures  for all similar samples.  The
analytical data should be reported in the same units for each test and for all
samples collected from a site.

Precision and Accuracy -  Known standards, QC  samples,  replicates, blanks,
surrogate spike and treated matrix spike samples will be used to assess
precision, accuracy and detection limit of all analyses.

Detection Limits - The detection limit for the analysis of TOC is 300 mg/kg,
PAHs 0.2 mg/kg, and PCBs  (Aroclor) 1.0 mg/kg.   Detection limit determination
for the ARCS program is done by the analysis of 15 low level standards or
blanks with a concentration within a factor of 10 times the instrument
detection limit.  The limit is then set as three times the standard deviation
of the multiple measurements.  This determination will be done prior to the
analysis of the experimental samples.


                              SAMPLING PROCEDURES

     Initial procurement of the sediment samples is the responsibility of
GLNPO for Buffalo River and PCB contaminated sediments and the Bureau of Mines
for Chicago River.  When the samples are received at Athens ERL, splitting
will be accomplished by mechanically mixing the received material until visual
homogeneity is observed, then the sample will  be completely split into equal
subsamples.  These wet sample splits and original study site samples will be
stored in the dark at 4° ± 2° C.  Weekly temperature measurements will be
recorded in a bound notebook.  Wet-sample splits will be consumed through the
course of the project, so holding times may be as long as one year.  Periodic

                                      62

-------
storage checks to ensure the lack of breakdown of organics will be done on the
subsamples when they are removed from the storage area and prior to testing.
The initial baseline analysis of each subsample should be within + 30% of the
original analysis.
                                SAMPLE CUSTODY

     Upon receipt of the samples at the Athens ERL,  they will be logged into a
sample logbook.  The following information will be documented in the logbook:

         *   Date and time of sample receipt
         *   Project number
         *   Field sample number
         *   Laboratory sample number (assigned during log-in procedure)
         *   Sample matrix
         *   Sample parameters
         *   Storage location
         *   Log-in person's initials

     As noted, the sample will be split into quantities suitable for
characterization and testing, and each split will be numbered and labeled.
When a researcher takes one or more of these splits for processing studies,
its number, date, name of the researcher, and purpose will be entered in the
log.  As further subsamples are generated, they will be assigned numbers
consistent with the original split.  The principal investigators will serve as
sample custodian for the purpose of receiving samples and verifying sample
records.


                INSTRUMENT CALIBRATION PROCEDURES  AND  FREQUENCY

     Research equipment, capable of calibration, will be calibrated once daily
during each period in which the device is used.  The calibration standards
will be from Ultra Scientific and will include the constituents of concern for
the project.  Required frequency and acceptance criteria for the ARCS program
are listed in the Internal Quality Control section.

     The following instruments will be utilized in this project in accordance
with the working standards supplied by each operations manual:

  1)   Hewlett Packard 5890 Series II Gas Chromatograph
      Manual #5890-90110

  2)   Hewlett Packard 3396A Integrator
      Manual #03396-60920

  3)   Hewlett Packard 7673 Automatic Sampler
      Manual #07673-60995

  4)   Orion 611 Ph Meter
      Manual #502700-058

  5)   Mettler PM400 and PM4000 Balances
      Manual #ME-702395

  6)   Dohrman DC-85A Carbon Analyzer
      Manual #915-239

                                      63

-------
 7)   Hewlett Packard 3396 Series II Integrator
      Manual #03396-90105

 8)   Hewlett Packard 5970 Mass Selective Detector
      Manual #05970-90049

 9)   Hewlett Packard 59970C Chemstation
      Manual #59970C-90034

10)   Beckman LS6000LL Liquid Scintillation Counter with external standards
      source

      Below is an overview of the calibration procedures for the analytical
instruments that will be used for this project.  The concentrations of the
calibration standards for each method will be determined by the detection
limit and linear response of the range.
Instrument

GC (PAHs}
Hewlett Packard 5890
with a 30-meter DB-5
and flame ionization
detector (FID)

GC/MSD (PAHs)
Hewlett Packard 5970
GC (PCBs)
Hewlett Packard 5890
with a 30 meter DB-1
column (0.25 urn i.d.)
and electron capture detector
(ECD)

Analytical balance
HPLC (PAHs)
pH meter
TOC
LSC
          Procedure

Meet criteria described above in a 3
point initial calibration with 20, 50,
and 100 mg/L standards.
Meet MSD tuning criteria followed by
chromatographic acceptance criteria (same as
above)  with the 3-point initial calibration
with 20, 50, and 100 mg/L standards.

Meet chromatographic acceptance criteria
(such as degradation, peak shape,
sensitivity, signal to noise ratio, and
retention time stability) in a 3-
point initial calibration with 6.1, 30.5,
and 61 mg/L standards.

Prior calibration check with class S weights
in the gram and milligram range.  Other checks
as appropriate in expected weighing
range.

Meet chromatographic acceptance criteria  (same
as for the GC/MS) with a multipoint initial
calibration at applicable concentrations.

Three point calibration at Ph 4, 7, and 10.
Calibration check every 10 samples.

Daily single-point calibration in triplicate.
Check standard every 20 samples.

Daily checks for acceptable background and
sample quench.
                                      64

-------
                            ANALYTICAL  PROCEDURES

     The research group will perform characterization and separation
procedures on the samples.  PAHs will be analyzed using methods adapted from
EPA SW 8461.   Quantitative analysis  for  PAHs  will be  preformed by GC/FID and
GC/HPLC. PCBs will be analyzed using a modification of methods from EPA SW
8461.   Quantitative  analysis for PCBs  will  be performed by GC/ECD and
qualitative analysis will be by GC/MSD.  Measurements of pH will be in
accordance with methods 9040 and 9045 from EPA SW 8461;  measurements  of TOC
will adhere to method 9060 also from EPA SW 8461.

     Initial efforts will be directed toward ascertaining the appropriate
methods for monitoring the losses of PAHs and PCBs from contaminated sediment
incubated under aerobic and anaerobic conditions,  respectively.  Aerobic
conditions will be maintained by mixing the reaction vessel on a temperature
controlled reciprocal shaker  (150 rpm) .  Anaerobic conditions will be
maintained by conducting  experiments in an anaerobic chamber containing a gas
mixture of 95% N2 and 5%  H2. Sterile controls will be prepared by autoclaving
samples for 30 min on three consecutive days.

Extraction and Clean-up of Sediment Slurry Samples for Analysis of PAHs.

Extraction

1.   Vortex sediment slurry for 1 minute.  Immediately  remove 5 ml samples  to
     50ml centrifuge tubes. Triplicate  samples are collected  for PAH analysis;
     another  set of triplicates is collected  for determination of sediment  dry
     weights. Record amount of  sediment in sample  (grams  dry weight/5  ml).
2.   Centrifuge  sample at 8000  rpm for  20 minutes in refrigerated  (4° C)
     Sorvall  Superspeed  (RC-2)  centrifuge. Pour off  aqueous supernate  and
     extract  with 5 ml CH2C12.   Combine  organic phase with subsequent sediment
     extract.
3.   Add  15ul of surrogate  mixture containing 4mg/L  each  of d!4-l,4-
     dichlorobenzene, dlO-acenapthene,  dS-naphthalene,  d!2-perylene, dlO-
     phenanthrene,  and d!2-chrysene to  sediment.  Refrigerate  sample at 4°C
     2h-overnight to allow CH2C12  to evaporate and  for  PAHs to adsorb to the
     sediment.   A surrogate spike of  15ul  (60pg of each surrogate compound)
     yields a final analyte concentration  of  10 mg/L after the sample  has  been
     split as described.
4.   Dry  the  sediment by  adding lOg sodium sulfite  (15g of granular  form).
     Mix  immediately to a sandy texture. Water was removed from  sediment
     samples  of  greater than  10 g by  preliminary extraction with methanol
      (3X).  The  methanol  was  subsequently  pooled with  methylene  chloride
     extracts and back-extracted with a 3% saline  solution.
5.   For  sediments  dried  with sodium  sulfite, extract  first with 15ml  of  a 2:1
     CH2C12/  acetone mixture on a wrist-action shaker (Burrell Model  75) at
     full-speed  setting  for 18h.  Centrifuge  sample  (4°C,  8000 rpm, 20min.).
     Pour off supernate  to a  pennyhead  centrifuge  tube.   Cap  and store at 4°C.
6.   Repeat  extraction with 10ml  CH2C12  to  sediment and shake  18h.
7.   Centrifuge  sample  (4°C,  8000 rpm, 20 min.).  Combine supernatants  in a
     graduated cylinder.   Adjust  volume of methylene chloride to 21  ml.  Put
     sample  into a  storage tube.
                                       65

-------
Clean-up

1.   Remove one third (7ml)  of each sample to a separatory funnel.  Add 10ml
     IN NaOH.  Shake well and vent.
2.   Remove organic layer to pennyhead-stoppered centrifuge tube.
3.   Exchange methylene chloride for hexane by adding 7 ml of hexane and
     evaporating methylene chloride under a stream of dry nitrogen.   Repeat
     with a second aliquot (5ml) of hexane.  Evaporate down to 5ml +/-.
4.   Pass sample in hexane through silica columns.  Preparation of silica
     columns will be as follows: Heat silica gel at 110°C for 24h to activate.
     Place silanized glass wool on bottom of 10 ml  syringe fitted with a
     Teflon luerlock valve.   Saturate silica gel with hexane and add to column
     to reach the 10 cc mark.  Add ±1 g of sodium sulfate to top.  Seal with
     aluminum foil and parafilm and refrigerate until use.)   Wash with 20ml
     hexane.
5.   Elute sample with CH2C12 (2 x 10 ml aliquots).   Collect in test tubes and
     pool aliquots into a pennyhead-stoppered centrifuge tube.
6.   Evaporate methylene chloride under a stream of dry nitrogen to 1.0-1.5ml.
                                      66

-------
GC Analysis

1.   Prepare 1 hexane blank and 3 or 4 standards  (1, 2.5, 5ppm,  or  lOppm) .
     Prepare standards with 1ml 2X concentration  standard mixture.  Add lOppm
     surrogate mixture to each standard.
2.   Add 200ul dibutyl chlorendate internal standard  (lOmg/ml)  to each  sample
     and to standards.  Pipet internal standard to all samples  in one group
     with the same pipet to increase accuracy.
3.   Accurately top sample volumes to 2.0ml with  CH2C12.
4.   PAHs were quantitated by gas chromatography  using an HP  5890 Series II
     (Hewlett Packard) equipped with a DBS capillary column  (30 m X 0.32 mm,  C
     & W Scientific) and an FID detector.  Helium (30 psi) was  used as  the
     carrier gas, while air (40 psi) and hydrogen (30 psi) supplied the FID
     detector.  Injector and detector temperatures were  300°C.  The
     temperature program was 40°C for 3 min.,  10°C/ min to  125°C, 4°C/ min. to
     310°C,  and 310°C was held for 3 min.   The injection volume was lul.
5.   PAH concentrations were determined with  a Chemstation software package
     from standard curves obtained from analysis  of the  standard mixture of
     PAHs.
Media
Bushnell-Haas Minimal salts medium  (MSB)

K2HP04  	  1000 mg
 (NH4) 2NO3 	 1000 mg
CaCl,-2H,O
20 mg
KH2PO«  	  1000 mg
MgSO<-7H2O	  200 mg
Fed, 	    5 mg
Stock solutions

K2HPO4  	  10  g
 (NH4)2N03 	 10 g
CaCl2-2H2O  	   0.2  g
                             KH2PO«  	 10 g
                             MgSO4-7H2C  	  2 g
                             ddH,0
                                                1 L
Mix  and  aliquot  into  100  ml  volumes,  then autoclave.   Stocks are 10X.
FeCl3 	  0.5 g in 100 ml ddh2O, sterile filtered.  Use 1 ml per 1L.


Sediment Extract MSB  Medium

      Sediment  extract was prepared by combining 50g BRS and 150ml H20.   The
mix  was  autoclaved, then  shaken for 24h on wrist-action shaker.  Sediment was
pelleted by centrifugation and the supernate was retained as sediment extract.
MSB  stock solutions  (10X)  were added to achieve a final MSB concentration of
IX.
 Extraction of PAHs

      Two methods were used to extract PAHs from sediments, depending on the
 concentration of PAHs and the amount of sediment to be extracted.  Laboratory
 experiments in which PAHs were added to sediment slurries resulted in higher
 PAH concentrations than encountered in the field.  Thus, smaller volumes of
 sediment needed to be extracted to yield measurable quantities of the
 compounds.  Such sediment slurries were extracted by drying with sodium
 sulfate and extracting with methylene chloride, as described below.
                                       67

-------
Sediments that had not been spiked with additional PAHs contained lower
concentrations of the chemicals; hence larger volumes of sediment were
required.  Such large quantities were not amenable to drying with sodium
sulfate.  In such cases, the methylene chloride extraction was preceded with
methanol extractions until all water was removed.
     Thus, water was initially removed from experimental sediment slurries by
centrifugation, followed by drying with sodium sulfate  (2:1 by weight).
Unless otherwise specified, all centrifugation was carried out in a Sorvall
RC-3 at 4000 x g at 10°C for 10 min.   The  sediment slurries were  subsequently
extracted with 15 ml of a mix of CH2C12:acetone  (2:1) on a wrist action shaker
(Burrell, Pittsburgh, PA) for 18-24 h.  The sediment was pelleted by
centrifugation and the supernate was retained.  Extraction was repeated with
10 ml of CH2C12.   All extraction supernatants were pooled, back-extracted
with I N NaOH, and the total volume was adjusted to 21 ml with CH2C12 prior to
clean-up.
     When PAHs were to be determined in samples that had not been spiked with
additional compound, sediment quantities of 10 g or greater were required to
yield sufficient analyte.  In such cases,  the sediment was extracted three or
four times with 10-ml quantities of methanol after the bulk water had been
removed by centrifugation.  The methanol was retained and combined with the
subsequent methylene chloride extracts (two 10-ml extractions on the wrist-
action shaker, as above).  The pooled extract was back-extracted with 3%
saline to remove the methanol, but retain any compound that may have been
removed with that solvent.
     A deuterated PAH mixture  (J&W Scientific, Folsom,  CA) was added (to yield
10 ppm in the final analyte) prior to extraction to evaluate extraction and
clean-up efficiencies.  Typical recoveries of deuterated PAHs were naphthalene
45%, acenaphthalene 60%, phenanthrene 80%, and perylene 95%.

Clean-up of PAH samples.

     After either of the described extractions, the methylene chloride solvent
was reduced in volume and a solvent exchange into hexane was performed.  The
sample,  in hexane, was applied to an activated silica clean-up column  (63-200
mesh) to remove sediment-derived organics, and the PAHs eluted in methylene
chloride.   Dibutyl chlorendate (J&W Scientific) was added as an internal
standard at a final concentration of 20 ppm and the sample volume was adjusted
to 2.0 ml.

PAH additions to sediment slurries.
     PAHs were added to sediments as a primary mixture made by combining 2 mg
each of acenaphthene, acenaphthylene, fluorene, phenanthrene, anthracene,
fluoranthene, pyrene, benzo(a)pyrene, dibenz(a,h)anthracene, and
benzo(g,h,i)perylene (Sigma Chemicals, St. Louis, MO) in 10 ml CH2C12.  The
spiking mixture was stored at -4°C.   Sediment slurries were spiked by adding
50 pi of the PAH mixture to wet sediment  (equivalent to 20 g dry weight) in a
250 ml flask.  The solvent was allowed to evaporate overnight at 4°C prior to
addition of the experimental medium  (e.g., Bushnell Haas minimal salts broth).

Mineralization of radiolabelled PAHs.

     Radiolabelled PAHs can be easily quantified with a high degree of
precision and were used for several types of experiments in this project.
Although disappearance of a parent PAH can indicate the occurrence of a
transformation, the detection of radiolabelled carbon dioxide from labelled
PAHs can indicate the full mineralization of the parent and hence the
potential for supporting the growth of bacteria capable of their degradation.


                                      68

-------
Moreover, the use of radiolabelled compounds, detected either as disappearance
of the parent or production of carbon dioxide, permits frequent, precise
sampling, which is necessary for comparing the rates of transformation of PAHs
in response to treatments.
     To determine whether specific PAHs were mineralized by the enriched
culture, approximately 0.025 uCi of the labelled compound and 2 ppm unlabeled
substrate were added to sediment slurries and 14CO2 evolution was monitored
over time. The compounds tested were: phenanthrene, fluoranthene, pyrene and
benzo(a)pyrene, which were obtained in radiolabelled form  ([9-14C] phenanthrene
 (Sigma Chemicals),  [3-14C] fluoranthene  (Chemsyn Labs, Lexana, KS),  [4,5,9,10-
14C]pyrene  (Chemsyn  Labs) or  [7,10-14C]benzo (a)pyrene  (Sigma  Chemicals)).
Sediments were inoculated with the enrichment culture.

Enrichment of PAH degraders from Buffalo River Sediments.

     An  enrichment  culture of microorganisms  from  Buffalo  River sediment
 (BRS) were developed by exposing the sediment to a mixture  of PAHs for a
period of three weeks.  BRS was combined with mineral salts broth  (MSB) and a
mixture  of PAHs  (20 mg/kg each of the  sixteen PAHs of concern)  in 250-ml
 flasks,  which were  incubated shaking  (200 rpm) at 28oc  for  3 weeks.   The
 culture  was transferred  <1%) every 3-4 weeks.  After 3 months,  in order to
 determine the activity of the enrichment on  each of the PAHs as well  as the
mixture  of PAHs,  and to determine the  dependence of the culture on nutrients
 derived  from sediment, the enrichment  was inoculated into:  (1)  BRS +  MSB +
 each  PAH individually,  (2) BRS + MSB + PAHs,  (3) MSB +  PAHs, and (4)  MSB +
 each  PAH individually.  In addition to testing the activity of  each individual
 enrichment, a uniform inoculant with optimal  activity was  developed by
 combining all of  the cultures.  The resulting enriched  culture  was maintained
 by  sequential transfer in BRS + MSB +  PAHs every 3-4 weeks.  It is this
 enriched culture  that is the experimental inoculum unless  otherwise specified.


 Quality control  and statistical analyses.

      The procedures used  to  assure quality control and  validity of statistical
 analyses are described. Deuterated PAHs were added to sediment  samples  to
 monitor extraction  efficiency. The extraction efficiency varied with  the
 molecular weight  of the  PAH.
      In experiments in which PAH  concentration was determined by extraction
 and GC,  each treatment was  run in duplicate  or triplicate.  At  each  sampling
 at  least one treatment  replicate  was  sampled in  triplicate,  in  order  to
 compare the variance due  to  sampling  to  that due  to  differential performance
 of  replicates  within treatments.  All  data are represented as  averages  of
 replicate or triplicate  determinations.

 Extraction and Cleanup  of Sediment Samples  for PCS Analysis.
 Sediment preparation.

      Transfer  entire  sample  jar  contents  to  a pre-weighed,  labeled glass  pie
 plate.  Remove  any visible large  debris (stones),  and reweigh the pan.  Place
 pan in a hood  to air dry for 3 days.   Reweigh. Scrape contents  of the pan into
 a mortar and  grind to  a consistent grain  with a  pestle.  Sieve  the sediment
 through three  consecutive sieves  and into a  catch pan:  first #4 - 1.40 mm,
 second #18 -  1.00 mm,  and lastly #45 - 0.355 mm.  Transfer  the  sieved sediment
 into a clean,  labeled  sample jar.  Cap and store in  the refrigerator at 4 +
 2°C.
                                       69

-------
CALCULATIONS

     % dry weight:

     % dry weight =             g of dry sample   x  100
                                g of sample
PCS extraction of dried sediment.
     Transfer 5.0 g of air-dried sediment into a 15.0 ml test tube and cap
with a teflon-lined screw cap. Add 0.25 ml of 1% (v/v) Triton-X 100, 0.100 ml
of the 100 ppm matrix spike solution  (when applicable), and 10.0 ml acetone to
the sample.  Vortex, and place on the wrist-action shaker for 24 hours.
Visually check to be certain that the sample is well mixed. Centrifuge the
sediment/acetone mixture in the test tube at 2200-2400 rpm for 10 minutes at
room temperature. Transfer all the supernatant solution into a 125 ml
separatory funnel with a 5.0 ml graduated pipet. Record the volume recovered.
Place test tube with sediment plug in a 105°C oven  overnight;  record weight
and subtract test tube weight for dry sediment weight. Add 10.0 ml of the
extraction solvent and 5.0 ml of 2%  (w/v) sodium chloride solution and shake
for 1 min. Allow the sample to settle before draining the lower layer. Add 5.0
ml of concentrated sulfuric acid and shake for 1 min, and allow the sample to
settle.  If there is still color in the non-aqueous layer, reshake and then
drain. Add 5.0 ml of 2% (w/v)  sodium chloride and shake for 1 min.  Allow the
lower layer to separate and drain. Transfer the non-aqueous layer with a 5.0
ml pipet to a labeled 15 ml test with a Teflon-lined screw cap.  Record volume
and dilute to exactly 10.0 ml with hexane.

Sample cleanup.

     Prepare florisil spice tubes (Analtech #11-96) with the addition of
approximately 2 g florisil.  Transfer spice tubes to anaerobic glove box and
add approximately 10 g cleaned (reduced) copper filings to each.  Add
approximately 2 g florisil to the top of the tube,  and remove from the glove
box. Pre-wet spice tubes with 2 rinses of 5 ml hexane. Transfer 5 ml of
extract onto the spice tube column and vacuum drain slowly.  Collect the
extract.  Rinse 3 more times with  5 ml hexane, collecting all rinses. Measure
volume of collected extract,  record,  and dilute to 20 ml with hexane in a
labeled glass scintillation vial. Transfer 4.5 ml of extract into a labeled
shorty vial. Add 0.5 ml internal standard solution (10 ppm octachloro-
naphthalene;OCN). Cap with Teflon-lined cap and shake to mix. Transfer 1-2 ml
to an amber (labeled)  autosampler vial.   Cap with a Teflon-lined crimp seal.
Seal and store remaining extract in the shorty vial in the freezer (-20C).
Sample remaining in the scintillation vial may be disposed as hazardous waste.
All solvents used should be pesticide grade or better.


Solutions for PCB extraction.
Extraction solvent (5 mg/L of congener #151). Make up 1 liter of 9:1
hexane/acetone by measuring 900 ml hexane and 100 ml acetone into a 1 liter
amber volumetric flask.  Completely rinse the contents of a 5 mg vial of BZ
#151 (Ultra Scientific #RPC-051) into the volumetric flask using the 9:1
solvent.  Mix thoroughly, cap, seal,  and store in the refrigerator.

Matrix spike solution  (100 mg/L of congener #158) .  Completely rinse the
contents of a 5 mg vial of BZ #158 (Ultra Scientific #RPC-109)  with hexane
into a 50 ml amber volumetric flask.   Fill the volumetric flask to volume with
hexane.  Mix, cap,  seal,  and store in the freezer.

                                      70

-------
2%  (w/v)  sodium chloride solution.  20 gm of high purity sodium chloride
(Aldrich #22,351-4)  is added to 1000 ml of distilled water.   Mix,  cap,  and
store in the refrigerator.

1% (  w/v)  Triton X-100 solution.  1 ml of Triton X-100 {Aldrich #28,210-3) is
diluted to 100 ml with distilled water in a 100 ml volumetric flask.  Mix,
cap,  and store in the refrigerator.

10% sulfuric acid solution. 10 ml of concentrated sulfuric acid (Baxter
Diagnostics) is diluted to 100 ml with distilled water in a volumetric flask.
Mix,  cap,  and store in the hood.

Internal standard solution (10 mg/L OCN). 20 mg of octachloronapthelene  (OCN)
(Ultra Scientific #RCN-012) is rinsed with hexane into a 20 ml amber
volumetric flask.  Fill the volumetric flask with hexane (this is now the 1000
ppm stock solution).  Transfer 5 ml of the 1000 ppm solution to a 500 ml amber
volumetric flask, and dilute to the mark with hexane.  Mix,  cap, seal,  and
store in the freezer.

Copper preparation. Transfer 500 g of copper metal - fine granular
(Mallinckrodt/Baxter #4649-500) into a clean sample jar.  Cover copper with
10% sulfuric acid, shake vigorously, and let sit a minimum of 24 hours.  Pour
off sulfuric acid, rinse 3 times with 200 ml distilled water, followed by 3
rinses with 200 ml acetone. Blow down remaining acetone with argon  or
nitrogen.  Transfer copper to the anaerobic glove box immediately.

Florisil preparation. Transfer 100 gm Florisil 100-200 mesh
(Mallinckrodt/Baxter #6876) into a glass container.  Loosely cap and store in
120°  oven.

Sediment preparation. Transfer entire sample jar contents into  a labeled, pre-
weighed glass pan.  Remove any visible  large stones.   Reweigh.  Allow
sediment to air dry for  3  days.  Scrape contents of the pan into a  large
mortar and  grind  to a consistent grain.  Sieve the ground sediment  through
three consecutive sieves;  first #14  - 1.40 mm, second #18 - 1.00 mm, and
lastly #45  - 355  urn.  Pour final sieved sediment into a clean,  labeled  sample
jar.  Cap and store in the refrigerator. All solvents should be pesticide
grade or better.


                   DATA REDUCTION,  VALIDATION,  AND REPORTING

Data Recording  -  All  raw data  will be put  into laboratory notebooks, as
mandated  in AERL  LOP  5400-32.

Data Reduction  -  Wherever possible,  the initial  data  reduction  will be
computerized.   Where  data reduction  is  not computerized,  calculations will be
performed in permanently bound laboratory  notebooks with  carbon copy pages.

Data Validation - The individual  analysts  will verify the completion of the
appropriate data  records to  verify  the  completeness  and correctness of  data
acquisition and reduction.   The  principal  investigator  will periodically
review  computer and manual data  reduction  results  and will inspect  laboratory
notebooks and/or  data sheets  to  verify data reduction correctness and
completeness  and to ensure adherence to the specified analytical method
protocols.  Calibration  and  QC data  will  be examined by the  individual
analysts  to verify that  all  instrument  systems  are in statistical control,  and
that QC objectives for precision,  accuracy,  completeness, and method detection
limit  are being met.

                                       71

-------
Reporting - Results will be given in standard units,  as specified by the
analytical methods. Reporting units will be given in mg/L for all liquid
samples and in mg/kg for soils and other solid matrices.   Before final
reporting of results,  the QC Monitor will review data for adherence to QC
objectives.  All results will be available in hard-copy and computer readable
format to the ARCS or BOM Project Managers.  All reports and documentation
required, including chromatograms and mass spectra,  calibration records, and
QC results, will be clearly labeled with the laboratory sample number and
associated field sample number.


                        INTERNAL  QUALITY  CONTROL  CHECKS

     At the discretion of the researcher, duplicate tests will be conducted to
ensure the accuracy and reproducibility of the analyses.   Typically, these
replicate tests will be conducted on random samples.   These repeat test
analyses will be a double check on analytical work to help ensure that
consistent results are obtained throughout the entire research program and
should fall within 20%.  The analytical internal quality control procedures
listed in Table A will be taken to ensure the stated criteria for precision
and accuracy (Table B).

     QC control charts will be kept on file for blanks,  QC samples, and mid-
range calibration standards.

     Triplicate sample analysis giving acceptable precision is ± 20% RSD.

     Samples in a group of 20 or less is called a set.  Data from a sample are
acceptable when spike recoveries fall in the range of 70 to 130, except for
highly volatile compounds.  An analytical batch is a group of samples run with
the same reagents on the same instrument under the same conditions.  The
analytical batch is acceptable when the QC chart shows the analysis to be in
control and that the QC measures have been acceptable.

     Conservation of mass will be used to validate mass measurement data.


                         PERFORMANCE AND  SYSTEM AUDITS

Internal Performance Audits - An audit will initiated by the principal
investigators using reference standards to assess adequacy of pollutant
concentration measurements.

Internal Systems Audit - An audit will be performed by the project manager at
the initiation of the project covering sampling splitting procedures, work
plans, selection of methods, and data recording.

External Performance Audits - Initiated by the principal investigators  at the
midpoint of the project using reference samples from an external source  (where
available) to assess the performance of pollutant analyses.

External Systems Audit - May be performed by the ARCS or BOM QA officer.


                            PREVENTIVE MAINTENANCE

     All the required  maintenance is performed in-house by the analyst
according  to the procedure described in the instrument operations manual
except for that which  can only be performed by a representative of  the

                                      72

-------
instrumentation company.  Instrument maintenance logbooks will be kept with
each instrument and will be updated by the operator whenever either routine or
non-routine maintenance procedures are performed.


                      ROUTINE DATA ASSESSMENT PROCEDURES

Surrogate Spike Recovery - The percent recovery for each surrogate standard
will be calculated and tabulated, and control limits established.  A surrogate
standard is a compound that when spiked into an environmental sample, behaves
in the same manner as the analyte or analytes of interest.  The surrogate
standard is added prior to extraction and is used to determine chemical
recoveries from environmental samples.  Surrogate standards for PAHs in
sediment are deuterated PAH mixtures.  Samples to be analyzed for PCBs will be
spiked with an appropriate surrogate  (dibutylchlorendate or a specific PCB
congener such as congener #151, which is not a major constituent of the
analyzed sample). All samples will be spiked with a surrogate standard unless
a specific exception is made in the method itself.  If a surrogate standard
falls beyond the control limit, the data will be regarded as unreliable,
corrective action will be taken, and the analysis will be repeated.
     Percent recovery for the surrogate spike samples will be calculated as
follows:

                        %P = 100% x  (Qd/Qa)

where:

     %P = percent recovery;
     Qd = quantity determined by analysis; and
     Qa = quantity added to the sample.

     A tabulation of percent recoveries will be maintained for each surrogate
spike.  -The tabulation will include the analysis date and the percent
recovery.

Precision and Accuracy - Precision will be defined in terms of relative
percent difference of the duplicate analysis, where applicable.  Precision
will be calculated using the following equation  for relative percent
difference:

                                RPD =      (C, - C,j^     x  100
                                           [(GJ+ C2)/2]


where:

     RPD = relative percent difference;
     G!  = the larger of the two values;  and
     C2  = the smaller of the two values.

If calculated from three or more replicates, the relative standard deviation
(RSD) will be used rather than RPD.  The following equation is applicable for
RSD:

                          RSD =  (s/y) x 100%

where:

     RSD = relative standard deviation;

                                      73

-------
     s = standard deviation;
     y = mean of replicate analyses.


The calculation for accuracy in matrix spikes is as follows:

                          %R = 100% x [ (S - U)/Csa]

where:

     %R = percent recovery;
     S = measured concentration in spiked aliquot;
     U = measured concentration in unspiked aliquot;
     Csa = actual concentration of spike added.


The calculation for accuracy in standard reference materials  (SRM) is as
follows:

                           %R = 100% x (Cm/Csra)

where:

     %R = percent recovery/-
     Co, = measured  concentration  of  SRM;
     C3rm =  actual  concentration of SRM.


Completeness - Completeness will be defined as followed for all measurements:

                           %C = 100% x (V/n)
where:

     %C = percent completeness;
     V = number of measurements judged valid;
     n = total number of measurements.


                               CORRECTIVE ACTION

     Corrective action will be implemented whenever a system is not in
compliance.  Corrective action schemes are based on data acceptability limits,
identification of defects, summarizing defects, and tracing of defects to
their source.  We will implement measures to correct identified defects,
maintain documentation of the results of the corrective process, and monitor
the process until each defect is eliminated.  In the case of laboratory non-
compliance issues,  corrective actions are likely to be immediate in nature and
most often will be implemented by the analyst.  In these cases, corrective
actions usually involve recalculation, reanalysis, or repetition of a sample
run.  Corrective actions may also be started as a result of performance and
system audits.  These actions may require altering analytical procedures,
sampling,  or using a different batch of sample containers.

     Corrective actions fall into two categories: 1) handling of analytical or
equipment malfunctions; and 2) handling of nonconformance or noncompliance
with the QA requirements that have been set forth.  All corrective measures
taken will be included in the Laboratory Record Book.
                                      74

-------
     Corrective action required to conform to the specifications will be
recorded by the QC monitor,  and reported to the Principal Investigator within
five days.  Corrective actions will be documented.

     Failure of any sample to meet any of the above stated limits for
precision and accuracy or failure to fall within limits of detection stated in
Section — QUALITY ASSURANCE OBJECTIVES -- will be considered criteria for re-
analysis of the sample or sample set.  This may include recalculation,
repetition of a sample run,  or reconsideration of the appropriateness of the
method used.  In the case of a change in methodology, all samples of the type
in question will be reanalyzed.  All corrective action taken will be noted in
the Laboratory Record Book.
REFERENCES

1.  Test Methods for Evaluating Solid Wastes (SW 846), Third Edition  (USEPA
     1986).

2.  USEPA Athens Environmental Research Laboratory Operations Manual, December
     1982.
                                       75

-------
Table A.  Summary of Internal Quality Control Procedures

Procedure                Frequency         Requirement

Calibration Standard     3  per set
Method Blanks
Matrix Spikes1
    contaminated
                         1 per set
1 per set
    uncontaminated
1 per set
    treated  duplicate

Replicates
    triplicate  analysis  1 per set
   duplicate  test

QC Sample

Surrogate Spikes1




Control Charts
as applicable

see spikes

all samples
                  Mid-range standard shot at
                  beginning,  middle,  and end of sample
                  set for pH and TOC.

                  Mid-range standard shot at
                  beginning,  every 12,  and end of set
                  for PAHs and PCBs.

                  ± 20% of the known value

                  < detection limit (3  per set for TOC
                  at beginning,  middle,  and end of
                  set)
1.0 to 1.5 times the estimated
concentration of sample

± 30% of the spike for PAHs and PCBs
± 30% of the spike for TOC

Spike within a factor of 10 of the
instrument detection limits

± 30% of the spike for PAHs and PCBs
± 30% of the spike for TOC

20% RPD
triplicate analysis of one sample

20% RSD .

20% RPD



applicable to PAHs and PCBs

30% RSD
± 30% of the spike
                        postrun analysis  ± 3 standard deviations
:The use  of  deuterated  PAHs makes matrix and surrogate spikes equivalent.
                                      76

-------
Table B.  Data Quality Objectives for Precision and Accuracy

   Measurement          Analysis
                        Method(s)a
                        8100 & 8310
Parameter

PAHs
standard
sample replicates
test duplicates
surrogate spike
     PCBs
     standard           8080 & 8270
     sample replicates     "
     test duplicates       "
     surrogate spike       "

     PH
     standard           9040 & 9045
     sample replicates     "
     TOC
     standard
     matrix spike
     sample replicates
                   9060
     Surrogate spikes  (PAH)
     Matrix  spikes  (PCBs)
     untreated sample   8080,8270,8100,8310
     treated sample            "
                                           Precision
                                                20% RSD
                                                20% RPD
                                           20% RSD
                                           20% RPD
                                           0.1 pH unit
                                           0.1 pH unit
                                           20% RSD
                                           20% RPD
Accuracy


± 20%


± 30%


± 20%


± 30%


± 0.1 pH unit
± 20%
± 30%
                                                             +  30%
                                                             ±  30%
 'Test Methods  for  Evaluation  Solid Wastes   (SW  846), Third  Edition  (USEPA 1986)
                                       77

-------
Table C.     Constituents of Revised Anaerobic Mineral Medium  (RAMM)

Component         per  unit volume  (liter)

KH2PO4              0.27    g
K2HPO,              0.35    g

NH,C1              0.53    g
CaCl2-2H2O         75       mg
MgCl2-6H2O         100      mg
FeCl2-4H2O         20       mg

MnCl2-4H2O         0.5      mg
H3BO3               0.05    mg
ZnCl2              0.05    mg
CuCl2              0.03    mg
NaMo04-2H2O        0.01    mg
CoCl2-6H2O         0.5      mg
NiCl2-6H2O         0.05    mg
Na2SeO3             0.05    mg

NaHC03             1.2      g    (optional*)
Na2S-9H2O          0.5      g    (optional*)


*not added in the present  study
                                         78     * U.S. GOVERNMENT PRINTING OFFICE: 1996 - 748-159

-------