&EPA
United States
Environmental Protection
Agency
                       Great Lakes
                       National Program Office
                       77 West Jackson Boulevard
                       Chicago, Illinois 60604
EPA905-R94-010
October 1994
Assessment and
Remediation
Of Contaminated Sediments
(ARCS) Program
BENCH-SCALE EVALUATION OF Rcc's
BASIC EXTRACTIVE SLUDGE TREATMENT
(B.E.S.T.)® PROCESS ON CONTAMINATED
SEDIMENTS FROM THE BUFFALO, SAGINAW,
AND GRAND CALUMET RIVERS
                      United States Areas of Concern

                      ARCS Priority Areas of Concern
                                 printed on recycled paper

-------
      /Bench-Scale Evaluation of RCC's Basic Extractive Sludge
      Treatment (B.E.S.T.®) Process on Contaminated Sediments
        from the Buffalo, Saginaw, and Grand Calumet Rivers
                           Prepared by

                           Clyde J. Dial
             Science Applications International Corporation
                          Cincinnati, Ohio
                              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           '

-------
                                       DISCLAIMER
The information in this document has been funded wholly or in part by the U.S. Environmental Protection
Agency  (EPA) under  Contract No. 68-C8-0062, Work Assignment No.  3-52,  to Science Applications
International Corporation (SAIC). It has been subjected to the Agency's peer and administrative review and
it has been approved for publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.

-------
                                  ACKNOWLEDGEMENTS
This report was prepared by the Engineering/Technology Work Group (ETWG) as part of the Assessment
and Remediation of Contaminated Sediments (ARCS) program.  Dr. Stephen Yaksich,  U.S. Army Corps
of Engineers (USAGE) Buffalo District, was chairman of the Engineering/Technology Work Group.

The ARCS Program was managed by the U.S. Environmental Protection Agency (USEPA), Great Lakes
National Program Office (GLNPO). Mr. David Cowgill and Dr. Marc Tuchman of GLNPO were the ARCS
program managers. Mr. Dennis Timberlake of the USEPA Risk Reduction Engineering Laboratory was the
technical project manager for this project. Mr. Stephen Garbaciak of USAGE Chicago District and GLNPO
was the project coordinator.

This report was drafted through Contract  No.  68-C8-0062, Work  Assignment No. 3-52, to  Science
Applications International Corporation (SAIC).  Clyde Dial  of SAIC was the principal author of the report,
with final editing and revisions made by Mr.  Garbaciak prior to publication.


This report should be cited as follows:

       U.S. Environmental Protection Agency. 1994. "Bench-Scale Evaluation of RCC's Basic Extractive
Sludge Treatment (B.E.S.T.®) Process on Contaminated Sediments from the Buffalo,  Saginaw and Grand
Calumet Rivers," EPA 905-R94-010, Great  Lakes National Program Office, Chicago, IL.

-------
                                         ABSTRACT
       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 (GLWQA) with Canada.  Under Section 118(c)(3) of the CWA, GLNPO is responsible
for undertaking a 5-year study and demonstration program for the remediation of contaminated sediments.
GLNPO has initiated an Assessment and Remediation of Contaminated Sediments (ARCS) Program to
carry out this responsibility. In order to develop a knowledge base from which informed decisions may be
made,  demonstrations of sediment treatment technologies are being conducted as  part of the ARCS
Program.  Bench-scale studies on the B.E.S.T.® Solvent Extraction Process, which is the subject of this
report,  took place at Resources Conservation Company (RCC) in Bellevue, WA on August 5 to 9, 1991.
The specific objectives for this effort were to determine process extraction efficiencies for polychlorinated
biphenyls (PCBs) and polynuclear aromatic hydrocarbons (PAHs); to conduct a mass  balance for solids,
water, oil, PCBs and PAHs; and to examine process effects on metals, oil and grease, and several other
parameters.

       The B.E.S.T.® Solvent Extraction Process was tested using sediment samples obtained from the
Buffalo River, Saginaw River, and Grand Calumet River. The concentration of the contaminants of concern
in the sediment were 0.3 to 22 mg/kg PCBs and 3 to 220 mg/kg PAHs.  The PCB and PAH concentrations
of 0.2 to 0.4 and 0.4 to 37 mg/kg, respectively, were found in the treated solids. This corresponds to PCB
and PAH removals of >95 to 99 percent  and  65 to 96 percent,  respectively.   Metals analyses were
performed on the treated solids and untreated  sediments. The data demonstrate that the treatment
process, as expected,  had little affect on metal removal from  the sediments.  The feed sediments and
treated solids were analyzed for percent moisture, oil and grease, total organic carbon (TOC), total volatile
solids,  and  pH.  Reductions in oil and grease concentrations (ranging from 80 to 99 percent) correspond
to sediment PCB and PAH removal. A mass balance was also carried out as part of this study for the
different constituents:  solids, oil, water, PCBs, and PAHs.

-------
                                   TABLE OF CONTENTS
Section

Disclaimer  	
Acknowledgements
Abstract	
Figures	
Tables	

1.0

2.0
                                                                             Page
3.0
Executive Summary

Introduction	
          2.1 Background	
          2.2 Sediment Descriptions
              2.2.1  Site Names and Locations for Each Sediment
              2.2.2  Sediment Acquisition and Homogenization . . .
          2.3 Sediment Characterization
          2.4 Technology Description . .
Treatability Study Approach
          3.1 Test Objectives and Rationale  	
          3.2 Experimental Design and Procedures
              3.2.1  Phase I
              3.2.2  Phase II
          3.3 Sampling and Analysis
              3.3.1  Sampling
              3.3.2  Analysis
 VI
vii

 1

 3

 4
 4

 4
 9

 9
10

12

12
14

14
15

18

19
19
                                             IV

-------
                             TABLE OF CONTENTS (Continued)


Section

4.0       Results and Discussion  	      21

          4.1 Summary of Phase I Results  	      21
          4.2 Summary of Phase II Results	      22

              4.2.1  Sediments/Treated Solids	      23
              4.2.2  Oil	      26
              4.2.3  Water 	      28
              4.2.4  Mass Balance  	      28

          4.3 Summary of Vendor Results	      33
          4.4 Quality Assurance/Quality Control	      34

Appendix A — B.E.S.T.® Bench-Scale Treatability Test Report 	      35
Appendix B — B.E.S.T.® Bench-Scale Treatability Test Plan	      64
Appendix C — Quality Assurance Project Plan	      72
Appendix D — B.E.S.T.® Treatability Study Analytical Matrix and Methods	     117
Appendix E — Battelle Data 	     123
Appendix F — Quality Assurance/Quality Control 	     143

-------
                                        FIGURES






Number



1      ARCS Priority Areas of Concern	       5



2      Buffalo River Sample Location  	       6



3      Saginaw River Sample Location 	       7



4      Grand Calumet River Sample Location  	       8



5      Flow Diagram of the B.E.S.T.® Process  	      11
                                            VI

-------
                                         TABLES


 Number                                                                              page

 1      Battelle and RCC Data - PCB Summary 	       1

 2      Battelle Data - Summary of Total PAHs	       1

 3      Mass Balance Summary	       2

 4      Battelle Data - Data Characterization of Feed Sediments 	       9

 5      Parameters for Analysis of ARCS Program Technologies	      14

 6      Sodium Hydroxide Addition	      16

 7      SAIC's Analysis Schedule for the Phase II Solvent Extraction of Buffalo River,
       Grand Calumet River, and Saginaw River Sediments	      20

 8      RCC Analyses	      21

 9      pH Adjustments	      22

 10     Battelle Data - Total PCBs	      23

 11     Battelle Data - Feed and Treated Solid PAH Concentrations	      24

 12     Battelle Data - Metals Concentration in the  Feed and Treated Solids	      25

 13     Battelle Data - Removal Efficiencies for Other Parameters 	      26

 14     Battelle Data - PAH Concentrations in the Treated Solids, Water and Oil	      27

 15     Battelle Data - PCB Concentrations in the Treated Solids, Water and Oil	      28

 16     Battelle Data - Solid Mass Balance	      29

 17     Battelle Data - Water Mass Balance  	      30

 18     Battelle  Data - Oil Mass Balance  	      30

 19     Battelle  Data - PCB Mass Balance 	      31

20     Battelle  Data - PAH Mass Balance 	      32

21      RCC Data - PCB Summary   	      33

22     RCC Data - Mass Balance Summary  	      33

                                           vii

-------
1.0    EXECUTIVE SUMMARY
       The B.E.S.T.® Solvent Extraction Process was tested using sediments obtained from the Buffalo
River, Saginaw River, and Grand Calumet River.  The contaminants of concern in the sediments for these
tests were PCBs and PAHs.  Samples of the feed material and the treated solids produced using the
B.E.S.T.® Solvent Extraction Process were analyzed by Battelle Marine Sciences Laboratory and RCC for
residual PCB contamination.  The data from these analyses are presented in Table 1.

                       Table 1. Battelle and RCC Data - PCB Summary
 Sample
       Feed
 (mg/kg, dry basis)
Battelle      RCC
  Treated Solids
 (mg/kg, dry basis)
Battelle       RCC
                                                                         Removal Efficiency
                                                                        Battelle
RCC
Buffalo River
Saginaw River
Grand Calumet River
0.32
21.9
15.0
0.60
21
22
<0.3
0.24
0.44
<0.03
0.18
0.23
>6
99
97
>95
99
99
       As these data obtained by RCC and Battelle demonstrate, PCB removal efficiencies for the Grand
Calumet River and Saginaw River sediments complement each other.  However, the removal efficiencies
determined by RCC and Battelle for the Buffalo River sediment are substantially  different.  This can be
attributed to the fact that the contaminant concentration in the raw Buffalo River sediment was close to the
analytical detection limit achievable by Battelle. The potential errors associated with these data undermine
the relevance of the removal efficiency obtained by Battelle for the Buffalo River sediment.

       Feed material and treated solids were also analyzed for residual PAH  concentrations. Table 2
outlines the analytical results obtained by Battelle.
                        Table 2. Battelle Data - Summary of Total PAHs
Sample
Buffalo River
Saginaw River
Grand Calumet River
Feed
(mg/kg dry basis)
9.90
2.70
230
Treated Solids Removal Efficiency
(mg/kg dry basis)
0.37
0.95
37.1
%
96
65
84

-------
       During the RCC analyses of the Buffalo River  and Saginaw River sediments, residual PAH
 concentrations of <0.2 mg/kg per compound were found  in the treated solids. Treated solids with PAH
 concentrations ranging from <1 to <3 mg/kg per compound were obtained for the Grand Calumet River
 sediment.  Because RCC was unable to report lower detection limits, comparisons between RCC and
 Battelle data are not conclusive.

       Metal analyses were performed on the treated solids and untreated sediments (see Table 11).  The
 Battelle data demonstrate that the treatment process, as expected, had little affect on metal removal from
 the sediments. The RCC data cannot be compared to the  Battelle data because these data were obtained
 using different analytical methods than those employed by  Battelle.  Because of the ashing of the sediment
 feed sample (potentially causing metals to be lost by volatilization) and because different methods were
 used to analyze the feed sediments and product solids, a reliable comparison of the RCC and Battelle data
 is not possible.

       The feed sediments and treated solids were analyzed for percent moisture,  oil and grease, Total
 Organic Carbon  (TOC), volatile solids, and  pH (see Table  12).   As the data  in Table 12 shows, the
 reductions in oil and grease concentrations (ranging from  80 to 99 percent) correspond to sediment PCB
 and PAH removal.

       A mass balance was also carried out as part of this  study. Table 3 summarizes the results obtained
 for the different constituents:  solids, oil, water, PCBs, and PAHs.

                     Table 3. Mass Balance Summary (percent recovered)
Solids
Sample
Buffalo River
Saginaw River
Grand Calumet
River
Battelle
98
99
92
RCC
97
98
86
Oil
Battelle
163
192
69
Water
RCC
112
137
97
Battelle
68
74
78
RCC
70
82
75
PCBs
Battelle
129
80
94
RCC
70
280
64
PAHs
Battelle
90
240
11
       Assuming that a full-scale application of this technology occurred and a volume of 500,000 tons of
sediment required treatment, RCC estimated that it would cost approximately $150 to $250/ton to treat the
material.  The cost is dependent on the quantity of material processed, the cleanup target and the settling

-------
characteristics of the waste. The waste would be treated at a rate of 200 to 300 tons per day using the
B.E.S.T.® Model  615 Unit operated on a  24-hour-per-day basis.  This  estimate includes mobiliza-
tion/demobilization costs  but does not account  for costs  associated with site excavation,  civil work,
applicable taxes, pre-screening needs, and overall site management and disposition of the product oil.

       Small vials of the residuals from the treatability test were retained and given to the EPA Technical
Project Manager for the GLNPO for "show" purposes. All quantities of the test products (water,  solids, and
oil residuals)  from each treatability test were sent to the analytical laboratory, Battelle Marine Sciences
Laboratory, for analysis.  Due to the quantities generated from the tests, none were retained and shipped
to EPA for possible further treatability studies.

2.0    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 (GLWQA) 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.  In response, GLNPO initiated the
Assessment and Remediation of Contaminated Sediments  (ARCS) Program.

       In order to develop a knowledge base from which informed decisions may be  made, bench- and
pilot-scale demonstrations of  sediment treatment technologies were conducted as  part of  the ARCS
Program. Information from remedial activities supervised by the U.S Army Corps of  Engineers  and the
Superfund program were also utilized. The Engineering/Technology (ET)  Work Group was charged with
overseeing the development and  application of the bench-  and pilot-scale tests.

       Science Applications International Corporation (SAIC) was contracted to provide technical support
to the ET Work Group. The effort consisted of conducting  bench-scale treatability studies on  designated
sediments to evaluate the removal of specific organic contaminants.  The bench-scale studies of the
B.E.S.T.® Solvent Extraction Process, which are the subject of this report, took place at  Resources
Conservation Company (RCC) in Bellevue, WA on August 5 to 9, 1991.  The specific objectives for this
effort were:   to  determine process  extraction  efficiencies  for polychlorinated  biphenyls (PCBs)  and

-------
 polynuclear aromatic hydrocarbons (PAHs); to conduct a mass balance for solids, water, oil, PCBs and
 PAHs; and to examine process effects on metals, oil and grease, and several other parameters.

 2.1    Background
       SAIC and its subcontractors have conducted seven bench-scale tests for the ARCS Program on
 four different sediments using four treatment technologies: B.E.S.T.® Solvent Extraction Process (RCC),
 Low Temperature Thermal Desorption Process  (ReTeC), Wet Air Oxidation (Zimpro Passavant),  and
 Anaerobic Thermal Process Technology (SoilTech). This report summarizes the approach used and results
 obtained during treatability testing of the B.E.S.T.® Solvent Extraction Process. The sediments used during
 this technology evaluation were obtained from the Buffalo River,  Grand Calumet River/Indiana Harbor
 Canal, and Saginaw River.

       The primary objective of this  portion  of the study was to determine the feasibility and cost-
 effectiveness of the B.E.S.T.® Solvent Extraction Process for treating and removing PCBs  and PAHs from
 the three sediments.  Based upon previous tests performed by RCC, it is their experience that the data
 obtained from the bench tests simulate full-scale operation. Thus, data generated by these tests may be
 used to estimate treatment costs for full-scale operation and to evaluate process feasibility.  The ability to
 evaluate process feasibility from these tests was also reported  by the  U.S. Environmental Protection
 Agency (EPA)  in their report entitled, " Evaluation of the B.E.S.T.® Solvent Extraction Sludge Treatment
 Technology - Twenty-Four Hour Test."

 2.2    Sediment Descriptions
       The sediments used for these tests are typical of sediments in the Great Lakes and their tributaries.
 They are representative of locations where future field demonstration projects may be conducted.  For the
 purpose of these tests, the primary contaminants in these sediments were PCBs and PAHs.

 2.2.1   Site Names and Locations for  Each Sediment
       GLNPO collected sediments for study from the  following areas around the Great Lakes: Saginaw
 River,  Michigan;  Sheboygan River, Wisconsin;  Grand Calumet  River/Indiana  Harbor Canal, Indiana;
 Ashtabula River,  Ohio; and Buffalo River,  New  York. SAIC was contracted to treat four of the sediments
 (from the Grand Calumet River/Indiana Harbor Canal, Buffalo River, Ashtabula River, and Saginaw River)
 using four different technologies. Samples from Grand Calumet River/Indiana Harbor Canal, Buffalo River,
 and Saginaw River were treated using the B.E.S.T.® Extraction Process.  A map is provided in Figure 1
which shows the ARCS Priority Areas  of Concern. Specifics of the sample location for the Buffalo River,
Saginaw  River and Grand Calumet River  are shown in Figures 2, 3, and 4, respectively.

-------
en
                                                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
  0  80 100  150 200
     KLOMETER9
US ENVIRONMENTAL PROTECTION AGENCY
(MEAT LAKES NATONAL PflOQRAU OFFICE
                           Rgure 1. ARCS Priority Areas of Concern

-------
    Ida
    Em
                                                                                 Buffalo River
O)
                                                                                          Sediment sample point
                                    Figure 2. Buffalo River Sample Location

-------
Saginaw River and Bay
        Sediment sample point
  "»»*
    '•ta»
    Figure 3. Saginaw River Sample Location




                   7

-------
                 Sediment sample point
          GRAND CALUMET RIVER
                                           M   IJ   2.0
                                            I     1
Figure 4. Grand Calumet River Sample Location

-------
2.2.2   Sediment Acquisition and Homogenization
       Prior to conducting the bench-scale treatability study using the B.E.S.T.® technology, the GLNPO
samples were homogenized  and  stored under refrigeration by the U.S. EPA Environmental Research
Laboratory in  Duluth, Minnesota.

       The homogenized sediments were sent to SAIC by the Duluth laboratory. Eighty ounces of each
sediment sample were then transferred by SAIC to RCC.  RCC used these samples to perform a series
of standard tests to determine if the waste samples were compatible with their process and to determine
optimum testing conditions and procedures for the treatability study (Phase I). The sediments used during
the treatability studies also originated from this stock and were forwarded to RCC by SAIC.

2.3    Sediment Characterization
       SAIC was responsible for the physical and chemical characterization of the raw sediment samples
used during the tests.  Table 4 provides characterization data of the sediments.  In order to limit inter-
laboratory variation, the different sediments and their residuals were analyzed by Battelle Marine Sciences
Laboratory in  Sequim, Washington.  Raw sediment analyses conducted by RCC are also included in this
report and can be found  in Appendix A. The raw sediment samples analyzed by RCC and Battelle were
collected simultaneously.
                  Table 4.  Battelle Data - Characterization of Feed Sediments
                              (mg/kg, dry basis, unless specified)

Total PCBs
Total PAHs
Moisture, % (as received)
Oil & Grease
TOC, % weight
Total Volatile Solid, %
pH, S.U. (as received)
Buffalo River
0.32
9.90
42.0
2420
1.98
4.03
7.29
Saginaw River
21.9
2.70
24.0
1350
0.83
2.09
7.30
Grand Calumet River
15.0
230
57.0
32200
17.03
14.2
7.35

-------
2.4    Technology Description
       The  B.E.S.T.® process is a patented  solvent extraction technology developed  by RCC.  This
process employs triethylamine (a solvent) to  extract contaminants from wastes.  Triethylamine is an
aliphatic amine produced by reacting ethyl alcohol and ammonia.  This solvent is distinguished from other
solvents because it is inversely miscible.  At temperatures below 65° F, triethylamine is completely miscible
with water, while at temperatures above 65° F, triethylamine and water are only partially miscible. Since
oil and water are similarly soluble in cold triethylamine, it can be utilized to treat wastes  containing both
contaminated oil and water.

       The  B.E.S.T.® process produces a single-phase extraction solution. If water and oil are present
in the feed material, a homogeneous mixture of triethylamine, water, and  oil is produced.  Any organic
contaminants contaminating the feed material, such as PCBs, PAHs, and volatile organic compounds
(VOCs) are trapped within the water  and  oil portion of the extraction solution. Since triethylamine achieves
intimate contact with the waste at nearly ambient temperatures and pressures,  emulsions (oil containing
the organic  contaminants) are not expected to occlude the solute. Thus the extraction efficiency of the
B.E.S.T.® process will not be compromised by feed mixtures with high water content.

       RCC utilizes triethylamine because it exhibits several characteristics that enhance its use in a
solvent extraction system. These characteristics, as reported by RCC, include:  1) a high vapor pressure
(therefore, the triethylamine can be recovered from the extract via simple steam stripping); 2) formation of
a low-boiling azeotrope with water (therefore, the solvent can be recovered from the treated solids by heat
with a low energy  input); 3)  triethylamine has an alkaline pH=10 (therefore,  some heavy metals are
converted to the hydroxide form, which  precipitate and exit the process  with the treated solids);  and  4)
triethylamine is only moderately toxic and readily biodegrades (data available in EPA document EPA-600/2-
82-001 a show that  a level  of 200 ppm triethylamine in water was degraded completely in 11  hours by
Aerobacter,  a common soil bacteria).

       A block diagram for the B.E.S.T.® Process is presented in Figure 5.  Since triethylamine is soluble
in water at temperatures below 65° F, the first extraction of the contaminated  material is  conducted  at
temperatures near 40° F. Therefore, the first extract solution will contain most of the water initially present
in the feed material. If the first stage extract contains sufficient water to allow a phase separation of the
triethylamine and water, the extract is heated to a temperature above the miscibility limit (130° F).  At this
temperature, the extract separates into two distinct phases; a triethylamine/oil phase and a water  phase.
The two phases are separated by gravity and decanted.
                                               10

-------
Solvent Recovery
          Recycled
          Solvent
  Was
1
         Extraction
         40°F(5°C)
 B.E.S.T.
                                          Solvent Recycle
                          Solvent / Oil
                                            Evaporator
             Separation
             130°F(55°C)
                           I
Solvent Recycle

 Water     I Water
 Stripper
                           Solvent / Water
                                           Solvent Recycle

                                                      Solids
                                 Solids
                                 Dryer
              Figure 5. Flow Diagram of the B.E.S.T.® Process (Source: RCC, Inc.)

-------
       At 130° F the solubility of oil (organic contaminants) in triethylamine increases. Since this enhances
the removal of oil from the contaminated solids, subsequent extractions are conducted at temperatures
above 130° F. Because these extracts contain mostly oil and very little water, they are combined with the
decanted triethylamine/oil  phase from the first extraction stage.  In the full-scale  unit, the  solvent is
recovered from the two phases by way of steam stripping and is recycled directly to the extraction vessels
for the solvent recovery portion of the process.  Residual triethylamine in the water and oil products is
usually low.

       Triethylamine is removed from the treated solids by indirect heating with steam.  A small amount
of steam may be added directly to the dryer vessel to provide the water required to form the low boiling
azeotrope.  Typically the residual triethylamine remaining with the treated solids biodegrades readily. Thus,
unless restricted by a contaminant not treated by the process, the dry treated solids may be used on site
as backfill.

       The B.E.S.T.® Process operates near ambient pressure  and temperature and at a mildly alkaline
pH.  Liquid temperatures vary from about 40 to 170° F and high pressures are not required.  A low-
pressure nitrogen blanket  creates a small positive pressure in tanks and vessels.  Since the process
operates in a closed loop  with one small vent for removal of non-condensing  gases, air emissions are
minimal.  RCC typically uses a water scrubber and activated carbon on this vent to minimize triethylamine
releases.

3.0    TREATABILITY STUDY APPROACH
3.1     Test Objectives and Rationale
       SAIC  was  contracted by the ARCS Program to test four technologies for removing organic
contaminants (PCBs and  PAHs) from  sediments typical of locations around  the Great Lakes.  This
treatability study was performed to determine the feasibility and cost-effectiveness of the B.E.S.T.® Solvent
Extraction Process for treating and removing PCBs and PAHs from three different sediments.  In order to
accomplish this, this bench-scale test had the following objectives:
       •  To record observations and data to predict full-scale  performance of the B.E.S.T.® process
       •  To take samples during the extraction tests and conduct analyses sufficient to allow  for
          calculation of mass balances for oil, water, solids and other compounds of interest
       •  To calculate the extraction efficiency of target compounds
       •  To obtain treated solids (300 g dry basis), water, and oil for independent analysis

                                              12

-------
       Based upon previous tests performed by RCC, it is their experience that the data obtained from
the bench test simulate full-scale operation.  Ultimately, this data may be used to estimate both the
feasibility and treatment costs associated with a full-scale application of the technology.  The ability to
evaluate process feasibility from these tests was also reported by EPA in their report entitled," Evaluation
of the B.E.S.T.® Solvent Extraction Sludge Treatment Technology - Twenty-Four Hour Test."

       A two-phase approach was used during this study. During Phase I, SAIC sent a sample of the
untreated sediments to RCC. These samples underwent a series of initial tests in order to determine the
optimum conditions to be used during the actual treatability tests (Phase II). During Phase II, wet sediment
from each of the three locations (Buffalo River, Grand Calumet River, and SJaginaw River) was sent to
RCC. Samples of raw (untreated) sediments and the various end products generated during the treatability
tests (Phase II) were obtained and analyzed by both SAIC and RCC.  The data generated by SAIC were
primarily used to determine treatment extraction efficiencies and mass balances. Vendor- or subcontractor-
generated data are reported and commented on when available.

       This study is only one part of a much larger program and is not intended to evaluate the treatment
of the sediments completely. In order to ensure that the data obtained from this study can be objectively
compared with data generated from the other studies performed in support of the ARCS Program, Battelle
Marine Sciences Laboratory was subcontracted to perform all analyses for the different treatability studies
performed by SAIC (seven treatability studies utilizing four technologies on four sediments).  The same set
of analyses listed in Table 5 was applied  during the characterization of each raw sediment  and end
products from the different treatability tests.  I n addition, representatives from SAIC observed how all Phase
II treatability tests were conducted.
                                               13

-------
                    Table 5.  Parameters for Analysis of ARCS Technologies
                                      Parameters
                          TOC/TIC                    Arsenic
                          Total Solids                 Barium
                          Volatile Solids               Cadmium
                          Oil & Grease                Chromium
                          Total Cyanide               Copper
                          Total Phosphorus            Iron (total)
                          PCBs (total & Aroclors)       Lead
                          PAHs (16)                  Manganese
                          pH                         Mercury
                          BOD                       Nickel
                          Total Suspended Solids      Selenium
                          Conductivity                 Silver
                                                     Zinc
3.2    Experimental Design and Procedures
3.2.1   Phase I
       Phase I was designed to allow RCC to explore a range of variables in order to set test parameters
which would optimize the performance of the B.E.S.T.® technology for the bench-scale tests (Phase II).
In order to accomplish this, samples of the different sediments were sent to RCC by SAIC prior to bench-
scale testing.  The amount of material sent was approximately 1 kg of each sediment, as specified by RCC.
       RCC  analyzed the  three raw  sediment samples to determine whether the sediments were
compatible with triethylamine.  During these compatibility tests, the feed samples were individually mixed
with cold triethylamine and then monitored to observe the amount of heat generated as well as any visual
signs that adverse reactions (such as an extremely exothermic chemical reaction) were occurring.  All
samples satisfied the compatibility criteria.

       Since triethylamine can be ionized at a low pH into unrecoverable triethylammonium salts, the pH
of the sample needed to be adjusted to approximately 11  during Phase II  in order to enable RCC to
efficiently recover the triethylamine from the separated phase fraction products. To determine the amount
of caustic needed to increase the pH of the raw feed to  the operating pH of the B.E.S.T.® process (pH =
11), RCC slurried  5 g portions of the feed samples with deionized water.  Incremental portions of the
caustic (50% sodium hydroxide) were added to bring the pH to 11.  The amount of caustic required to
adjust the pH  was recorded.
                                             14

-------
       After observing these simple mixing tests previously described, RCC applied its past experience
in selecting operating parameters for the technology to determine the number of extraction stages, solvent
to feed ratios, and the proposed temperature and mixing time for each stage for the bench-scale tests.

3.2.2   Phase II
       Phase II of the treatability program is referred to as the "B.E.S.T.® Bench-scale Treatability Test
Workup"  in  RCC's  B.E.S.T.® Bench-Scale Treatability Test  Plan.  This section outlines five  major
operations including: 1) Pre-treatment and first wash; 2) Second wash;  3) Third wash and solids drying;
4) Decantation; and 5) Distillation.  These procedures and associated equipment used are described in
Appendix B.

3.2.2.1 Procedures
Test Sample Preparation-
       The contaminated samples from the Buffalo River, Saginaw River, and Grand Calumet River sites
were gray-colored sediments with very little  debris present.  Each of the three samples contained free-
standing water. As  it was very difficult to homogenize the samples with the free-standing water present,
this water was decanted prior to conducting the bench-scale tests and proportionally recombined with the
portion used for the bench testing.  This was done by weighing the entire decanted sediment and the
portion used for the test.  The percentage of the portion used to the whole determined the percentage of
the decanted water  to recombine with the test sample.

       Bench-scale testing requires material greater than 1/4 inch  be  removed.   Full-scale processing
requires that the feeds be screened to remove only material greater than 1 inch in diameter. There was
no material greater than 1/4 inch in any of the three samples received. Therefore, the samples were not
screened.

       The following are summaries of  the five major operations.  With  slight variations, the  same
procedure was used to treat each of the sediments.

Pre-treatment and First Wash-
       During the treatability tests of the different sediments, each sample was placed in a 4-L resin kettle
immersed in a temperature-controlled  water bath set at 33° F. Each sample's pH was adjusted using
sodium hydroxide (NaOH) and 2.7  L of chilled triethylamine (2 percent  H2O).  The 2 percent water was
                                               15

-------
added to offset the amount of water remaining in the triethylamine after the extraction step.  The amount
of sodium hydroxide needed to adjust the pH of each sample to 11 was determined during Phase I  and
is listed in Table 6, as is the amount of sediment treated during each trial.

                             Table 6.  Sodium Hydroxide Addition
Sediment
Buffalo River
Saginaw River
Grand Calumet River
Sample Weight
(g)
700
700
1400
Caustic Added per
kg of Sediment
(ml 50% NaOH)
6.0
6.0
6.0
       While  immersed  in the cooling bath, the sample was mixed with the NaOH and the chilled
triethylamine using a air-driven prop mixer. Mixing occurred for approximately 20 minutes for the Saginaw
River and Buffalo River sediments with the pneumatic mixer in the chiller bath. Because of the high water
content of the  Grand Calumet River sediment, a larger sample size was needed to yield the quantity of
treated solids required. For this sediment, the first extraction was conducted in two steps, using 700 g of
feed for each step.  Each sample was mixed for 10 minutes.

       At the end  of the  first mixing  stage, the sample  was  allowed to separate  by  gravity.  The
particulates were then separated from the liquid by centrifuging the extract at 2,100 rpm for 10 minutes.
The solvent/oil/water/centrate were set aside for later decantation.  The solids from the centrifuge were
placed back into the resin kettle for additional wash stages.

Second Wash--
       Solids  recovered from the first extraction were mixed with 2.7 L of fresh triethylamine. Part of the
triethylamine was used to transfer solids from the centrifuge bottles into the mixing container.  For the
second extraction, the samples were heated  to 127 to 140° F. The mixture was  kept heated while mixing
was in progress.  Mixing was conducted with a pneumatic mixer for approximately 20 minutes. This sample
settled very  quickly.  The solvent/oil was  poured off and held  for later combination with the solvent/oil
portion from the decantation procedure.
                                              16

-------
Third Wash and Solids Drying--
       For the third wash, the same procedure that was used in the second wash was repeated.  Mixing
for this wash was for 30 minutes.

       The treated solids resulting from the third wash were then dried at 220° F in a forced-draft oven.
Occasional mixing in the oven to facilitate triethylamine volatilization was conducted.  This mixing was
accomplished by turning the sample in the oven with a clean spatula.  After the initial drying, a portion of
de-ionized water was added to wet the solids thoroughly. The solids were then redried in order to reduce
residual triethylamine concentrations further. To ensure that the triethylamine residual in the dried solids
was low, the solids were treated with caustic soda (applied with the de-ionized water) when the pH of these
solids was less than 10.  Sufficient caustic soda was added to raise the pH to approximately  10.5. The
required amount of caustic soda added  was determined on a small portion of the solids.

Decantation-
       The first stage extracts trap nearly all of the water present in the feed sample.  Because of this,
only the water from the first stage extracts is recovered.  There are two methods to decant water. The
decantation method is chosen depending on the water content of the feed.  The following methods were
employed to recover the water from the test series samples.

        Method 1: During the decantation of the Buffalo River and Grand Calumet River extracts, a 4-L
separatory funnel immersed in a temperature-controlled water tank was employed.  The tank was kept at
140° F by circulating water between the tank and a temperature-controlled water bath set at 140° F.

        Supernatant/centrate from the first wash (chilled to this point at 40° F) was heated to above 130° F
with continuous mixing on a hotplate and poured into the separatory funnel. Since above 130° F water is
no  longer miscible with the triethylamine,  the water settled to the bottom of the mixture. Forty minutes
quiescent residence time in the separatory funnel for the Buffalo River sample and 15 and 90 minutes for
the Grand Calumet River sample were required. The length of time required depended on how  long it took
for  near-separation of the layers.  A sample from the rag layer, which is an emulsion where any solids
present tend to collect and create a region where the triethylamine/oil/water  separation is not distinct, was
taken by RCC for later possible analyses.  Generally, the smaller the rag  layer is in comparison to the
triethylamine/oil and water phases,  the better the separation. The rag layers for all three sediments were
relatively small and within expected volumes.  SAIC did not collect  samples from this rag layer.
                                               17

-------
       Method?. Because of its low water content (<25%), the water present in the Saginaw River extract
was separated from the oil by evaporation instead of decantation.  When the first extract was evaporated,
the water in the triethylamine/oil/water mixture formed an azeotrope with the distilled triethylamine, leaving
the oil behind. After the triethylamine/water isotope had condensed, the water was decanted by heating
the triethylamine/water mixture above 130° F and pouring the mixture into the 4-L separately funnel. Since
no temperature control system was required, separation occurred immediately. This method produces a
much purer water stream and is preferable for low-water-content feeds where the extra energy cost to
evaporate the water is small.

Distillation-
       Water Layer: After recording the water's pH, which should have been >10, and its volume, the
water was stripped by steam at 110° C in a Buchi Rotovapor apparatus to ensure that the triethylamine left
the water. Periodically, the water volume and water pH were checked. When the water pH was <10, the
pH was adjusted to >12 and stripping continued until the pH of the water remained above 10. The pH was
periodically checked, and if found <10, the previous steps were repeated until the water pH remained above
10. At this point, distilling continued for 15 minutes longer before being terminated. The elevated pH is
necessary to  ensure that the majority of the triethylamine remains in the volatile molecular  form.

       Oil/Triethylamine Layer.  The bulk of triethylamine was removed from this layer by boiling the
triethylamine/oil mixture at 110° C in the Rotovapor (no steam necessary). The triethylamine condensed
as it evaporated and was collected separately. Normally the oil remaining in the flask would then be steam-
stripped of any residual triethylamine by adding a known quantity of water (typically 5 ml) to the hot oil in
the boiling flask of the  rotovapor and then  measuring the volume of distillate recovered.  When  all
triethylamine was removed, the amount of the distillate recovered would equal the amount of water added.
However, because of the low oil  content of the feed in this test, the amount of oil recovered was too small
to enable the  oil to be effectively stripped.  Triethylamine was allowed to remain in the Rotovapor, thereby
allowing the oil from the sample to remain in solution.  The homogeneous oil was able to be poured out
of the Rotovapor flask.   The final product oil/triethylamine  weight was recorded.  This  procedure is an
artifact of this test due to the small quantities of sediment.

3.3    Sampling and Analysis
       The Quality Assurance Project Plan is presented in Appendix C.
                                              18

-------
3.3.1   Sampling
       At the beginning of the Phase II treatability test, SAIC personnel observing Phase II packed and
shipped a sample of the untreated Buffalo River, Grand Calumet River and Saginaw River sediments to
SAIC's subcontract laboratory, Battelle, in accordance with written detailed instructions supplied to the SAIC
on-site representative. Each sample contained free-standing water which was decanted prior to conducting
feed analyses and was proportionally recombined prior to any analysis and bench testing. This was done
because it is very difficult to homogenize material when free-standing water is present. These samples
were obtained from separate unopened  containers of the sediments sent for Phase II.

       Although the samples would normally be screened to remove any material greater than 1/4 inch,
Phase I results indicated that no material of that size  or greater was present in the three samples.  Thus
the samples were not  screened.

       After the extractions were complete, samples of the final water, oil,  and solids residuals were
distributed to SAIC and RCC.  As specified in the Quality Assurance Project Plan (QAPP) a minimum of
300 g (dry basis) of solid material was required in order for Battelle to be able  to complete the necessary
analyses of  that  material.  Since the quantity of  oil and water was dependent on the  sediment  and the
technology employed, it was not possible to obtain  enough water and oil to perform the full scope of
analyses specified  in Table 7.

3.3.2  Analysis
       Two separate  sets of analyses were conducted by SAIC's subcontracted laboratory, Battelle, and
RCC on the  three raw sediments and the process products during Phase II. Battelle's data was used for
the results presented  in this report. RCC's data is discussed and commented upon, where possible, to
facilitate interpretation of the results of the treatability test.

3.3.2.1    Battelle Analyses
       Following the Phase II treatability test, Battelle conducted analyses on the three raw sediments and
the end products.  The number of analyses  conducted on these sediments and their residuals are listed
in Table 6. Descriptions of the analytical methods employed can be found in the QA Section of this report.

       Since the actual quantities of oil and water produced by the technology during the bench-scale
treatability tests were  not sufficient to perform all the analyses in Table 6, only PCB and PAH analyses
were performed on the water and oil.
                                               19

-------
                                        Table 7. SAIC's Analysis Schedule for the Phase II Solvent Extraction Evaluation of
                                                    Buffalo River, Grand Calumet River, and Saglnaw River Sediments
Parameters
Total Sol ids
(Moisture)
Volatile Solids
O&G
Metals
PCBs
PAHs
TOC
Total Cyanide
Total Phosphorus
PH
BOD
Total Suspended
Solids
Conductivity
QC Sample ()
and
Method Blank
(1)
YES
(1)
YES
(1)
YES
(0)
YES
(1)
YES**
(1)
YES**
(0)
YES
(0)
YES
(0)
YES
(0)
YES
NA
NA
NA
Untreated
Sediment
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S



MS



(1)
S
(1)
S
(1)
S
NA
NA
NA
'"



Tripli-
cate
(2)
S
(2)
S
(2)
S
(2)
S
(2)
S
(2)
S
NA
NA
•NA
NA



Treated
Solids
(3)
B,G,S
(3)
B.G.S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S
(3)
B,G,S

-

MS



(1)
S
(1)
S
(1)
S
NA
NA
NA




MSD




(1)
S
(1)
S
NA






Tripli-
cate
(2)
S
(2)
S
(2)
S
(2)
S



NA
NA
NA



Water
•" : • ^
NA*
NA
NA
(3)
B,G,S
(3)
B,G,S
NA
NA
NA
NA
NA
NA
NA
MS
f

'
NA
NA
NA
NA
NA
NA




MSD
"•

•>

NA
NA





%

Tripli-
cate
-
NA
NA
NA


NA
NA
NA
NA
NA
NA
NA
Oil




(3)
B,G,S
(3)
B,G,S







MS




(1)
S
(1)
S







Tripli-
cate
-



(2)
S
(2)
S







ro
o
        *  Not Analyzed
        ** A laboratory pure water spike is required for recovery determination
        (3) = Number of Analyses
         B = Buffalo River
         G = Grand Calumet River
         S = Saginaw Bay
MS = Matrix Spike
MSD = Matrix Spike Duplicate

-------
3.3.2.2    RCC Analyses
       RCC analyzed the sediment samples for moisture content, oil and grease, ash content, metals,
PAHs, PCBs, and particulate solids content.  RCC also conducted their own analyses on the products for
Phase II. Table 8 shows the analyses performed by RCC. Details on the analytical methods used by RCC
are presented in Appendix D. The following section is a summary of the RCC analyses conducted.

                                  Table 8. RCC Analyses
Matrix Sample
Raw Sediment
Treated Solids
Residual Oil
Residual Water
PAHs
yes
yes
no
no
PCBs
yes
yes
no
yes
Total
Metals
yes
yes
no
yes
Oil&
Grease
yes
yes
no
no
Solvent
no
yes
yes
no
Water
yes
no
no
no
Particulate
Content
yes
yes
no
no
TCLP
no
metals
no
no
PH
yes
yes
no
no
TCLP = Toxicity Characteristic Leaching Procedure


4.0    RESULTS AND DISCUSSION
4.1    Summary of Phase I Results
       RCC  analyzed the three raw sediment samples to  determine whether  the  sediments  were
compatible with triethylamine.  Since there were no visible signs indicating that adverse reactions were
occurring and the heat of the solution did  not exceed normal expectations, the B.E.S.T.® bench-scale
treatability tests proceeded.

       The amount of caustic (NaOH) needed to increase the pH of the raw sediments to the operating
pH of the B.E.S.T.® process (pH = 11) was  determined.  This information, as well as the  original pH of the
sample, is summarized in Table 9.
                                             21

-------
                                   Table 9.  pH Adjustments
Sediment
Buffalo River
Saginaw River
Grand Calumet River
Initial
PH
7.6
8.1
7.5
Caustic Added per
kg of Sediment
(ml 50% NaOH)
6.0
6.0
6.0
        A sieve analysis of the raw sediment was conducted to determine the screening and size reduction
requirements.   Since there was no material greater than 1/4-inch  in diameter, sample screening was
determined unnecessary.

4.2     Summary of Phase II  Results
        As stated previously, the concentrations of PAHs, PCBs, metals, total solids, volatile solids, and
oil and grease present in the  untreated sediments and treated solids are the critical measurements
associated with this study.  Oil and water residuals were analyzed to determine the fate of the contaminants
of concern from the process. Since insufficient water and oil was produced from the quantity of untreated
sediments used with the B.E.S.T.® process to perform all the analyses listed in Table 7, only PCB and PAH
analyses were performed on the water and oil residuals.   The following sections briefly address the
analytical results obtained for contaminant concentrations present in the raw sediments and the process
residuals (i.e.,  treated solids, water, and oil), as well as applicable extraction efficiencies.  The discussion
of Phase II results concludes with an analysis of the mass balance of the media and contaminants.  The
analytical data  received from Battelle can be found in Appendix E.

        Individual PAH compounds, PCB Aroclors, and  metals were quantitated during sample analyses.
In order to determine overall removal efficiencies for each class, it was necessary to sum these individual
results.  In  instances where all reported results were  less than the analytical  detection limits, total
concentrations are reported as less than the sum of the individual detection limits.  Where one or more
individual components are above detection limits, total concentrations are reported as the sum  of these
detected values.
                                              22

-------
4.2.1   Sediments/Treated Solids
4.2.1.1 PCBs
       Samples of the feed material  and the treated solids  produced using the B.E.S.T.®  Solvent
Extraction Process were analyzed for PCB contamination. The data from these analyses are presented
in Table 10.

                              Table 10. Battelle Data - Total PCBs
Sample
Buffalo River1
Saginaw River2
Grand Calumet River3
Feed
(mg/kg, dry basis)
0.32
21.9
15.0
Treated Solids
(mg/kg, dry basis)
<0.3
0.24
0.44
Removal Efficiency
%
>6
99
97
1 Identified primarily as Aroclor 1248
2 Identified primarily as Aroclor 1242
3 Identified primarily as Aroclor 1248

        As demonstrated by these data, PCB concentrations of 0.24 mg/kg and 0.44 mg/kg were found in
the treated solids generated from the Saginaw River and Grand Calumet River sediments, respectively.
This corresponds to PCB removal efficiencies of 99 and 97 percent.   At first glance, the Buffalo River
solids achieved a much lower removal efficiency (i.e., >6 percent).  This is attributed to the low  PCB
concentrations initially present in the untreated Buffalo River sediment and the high analytical detection
limits achieved.  As the concentration of a contaminant approaches analytical detection limits, the  error
associated with the analytical  readings obtained increases.  Thus, the relevance of the removal efficiency
achieved for the Buffalo River sediment is undermined by:  1) error associated with the measurement of
the contaminant concentrations near detection limits and  2) the high detection limits obtained  for the
samples.

4.2.1.2  PAHs
        Feed material and treated solids were also analyzed for PAHs.  As shown in Table 11, total PAH
concentrations of 0.37 mg/kg, 0.95 mg/kg, and 37.1  mg/kg were found in the treated solids produced by
treating the  Buffalo River, Saginaw River, and Grand Calumet River sediments. These values correspond
to removal efficiencies of 96,  65, and 84 percent, respectively.
                                               23

-------
     Table 11.  Battelle Data - Feed and Treated Solid PAH Concentrations (mg/kg, dry basis)
Buffalo River
Contaminant
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
ldeno(1 ,2,3-cd)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Total PAH
Feed
0.107
<0.080
<0.200
0.160
1.020
0.547
1.20
1.16
0.861
1.04
0.876
0.733
0.887
0.607
0.205
0.495
9.90
Treated '
0.037
<0.020
<0.030
<0.020
0.068
0.030
0.045
0.039
0.022
0.039
0.027
0.004
0.018
0.018
0.006
0.014
0.37
% Removal
64
NC
NC
>88
93
95
96
97
98
96
96
100
98
97
95
96
96
Feed
0.026
<0.020
<0.030
0.033
0.267
0.066
0.397
0.439
0.186
0.269
0.242
0.179
0.225
0.207
0.043
0.117
2.70
Saginaw River
Treated
0.024
<0.020
<0.020
<0.020
0.099
0.017
0.138
0.120
0.057
0.088
0.097
0.066
0.076
0.090
0.016
0.060
0.95
% Removal
08
NC
NC
>39
62
70
65
73
68
67
60
63
65
56
68
48
65
Grand Calumet River
Feed
4.40
2.32
4.40
4.62
15.2
5.63
32.0
32.0
18.3
24.4
19.2
13.4
20.6
14.7
5.22
13.8
230
Treated
2.25
0.121
0.726
1.08
5.64
1.47
3.11
3.55
3.13
3.99
1.89
1.32
3.22
1.36
1.93
2.35
37.1
% Removal
49
95
84
77
63
74
90
89
83
84
90
90
84
91
63
83
84
NC = Not Calculated

       Generally, the low removal efficiencies obtained for the PAHs in the Saginaw River sediment can
be attributed to the low concentration of PAHs initially present in the sediment and errors associated with
evaluating contaminant concentrations close to analytical detection limits.

       The removal efficiency of 84  percent for the total PAHs in the Grand Calumet River sediment
resulted in a final concentration of 37.1 mg/kg of PAHs in the treated solids. Additional extractions would
likely reduce PAH concentrations in the treated solids even further.
                                              24

-------
4.2.1.3 Total Metals
       The data in Table 12 highlight the removal achieved for the metal contaminants present in the
untreated feed and the treated solids. As demonstrated by the low or negative removal percentages, in
general, the B.E.S.T.® Solvent Extraction Process does not effectively remove metals.

Table 12.  Battelle Data - Metals Concentration in the Feed and Treated Solids (mg/kg, dry  basis)
Buffalo River
Contaminant
Arsenic
Barium
Cadmium
Chromium
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Zinc
Feed
12.7
413
2.10
109
70.2
42900
102
667
0.551
43.1
0.74
0.31
180
Treated
14.6
396
2.11
113
61.2
44200
102
684
0.627
42.1
0.87
0.24
190
%
Removal
-15
4
-0
-4
13
-3
0
-3
-14
2
-18
23
-6
Saqinaw River
Feed
2.21
322
4.14
107
58.8
7870
45.5
165
0.167
58.3
<0.3
0.84
140
Treated
2.85
319
4.26
118
64.1
8260
46.6
177
0.335
64.3
<0.3
0.82
169
%
Removal
-29
1
-3
-10
-9
-5
-2
-7
-99
-10
NC
2
-21
Grand Calumet River
Feed
22.8
317
8.56
2270
188
188000
582
3230
1.53
12.9
<0.3
4.84
2380
Treated
29.0
290
6.97
1710
223
82500
656
2540
1.46
<10
4.94
4.34
2810
%
Removal
-27
9
19
25
-19
56
-13
21
4
>22
NC
10
-18
 NC  =  Not Calculated


 4.2.1.4 Other Analyses
        The feed sediments and treated solids were analyzed for percent moisture, oil and grease, TOC,
 total volatile solids, and pH as shown in Table 13.  As shown by comparing data in Tables 10 and 11  with
 data in Table 13, reductions in oil and grease concentrations correspond to PCB and PAH removal.  This
 demonstrates that oil and grease analysis could possibly be used as a low-cost indicator for technology
 effectiveness for a given sediment.
                                              25

-------
              Table 13.  Battelle Data - Removal Efficiencies for Other Parameters
                               (mg/kg, dry basis, unless specified)
                        Buffalo River
Saginaw River
Grand Calumet River
Contaminant
Total PCBs
Total PAHs
Moisture, %
(as received)
Oil & Grease
TOC, % weight
Total Volatile
Solids, %
pH, S.U.
(as received)
Feed
0.32
9.90
42.0
2420
1.98
4.03
7.29
Treated
<0.3
0.37
3.72
238
1.21
3.91
10.30
Removal
>6
96

90
39
3

Feed
21.9
2.70
24.0
1350
0.83
2.09
7.30
Treated
0.24
0.95
0.16
265
0.58
1.73
10.73
Removal
99
65

80
30
17

Feed
15.0
230
57.0
32200
17.0
14.2
7.35
Treated
0.44
37.1
0.50
470
13.4
9.06
10.25
Removal
97
84

99
21
36

4.2.2  ON
       The concentrations of PAHs and PCBs in the oil extracted from the three sediments can be found
in Tables 14 and 15.  Final concentrations in the process solids and water have been  included as a
comparative measure of performance. Using values for percent oil determined in the samples received by
Battelle (i.e., 9.3 percent for Saginaw River extract,  6.3 percent for the Buffalo River extract, and 60.0
percent for the Grand Calumet River extract), these concentrations have been adjusted to account for the
triethylamine diluent found in the different oil samples received for analysis. The triethylamine was left in
these samples because of the low oil  content and small sample size used for these tests.

       Please note that the possibility for introducing error to these corrected oil concentrations does exist.
Analytically  determined values  are adjusted for  the amount  of  oil determined to be present in the
oil/triethylamine solution. When the PAH and PCB concentrations are adjusted, any error in this oil analysis
may be conveyed to the new concentrations. Samples with less oil (i.e., Saginaw River and Buffalo River)
are more likely to be affected.
                                              26

-------
        Table 14.  Battelle Data - PAH Concentrations in the Treated Solids, Water, and Oil
Buffalo River
Contaminant Solid Water
(ug/kg, dry) (ug/L)
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
lndeno(1 ,2,3-cd)pyrene
Dibenzo (a, h) anthracene
Benzo(g,h,i)perylene
Total PAH
37
<20
<30
<20
68
30
45
39
22
39
27
4
18
18
6
14
370
0.998
<0.4
<0.5
<0.5
<0.3
<0.4
<0.3
<0.3
<0.3
0.242
<0.3
<0.2
<0.3
<0.3
<0.3
0.184
1.41
Saginaw River
Oil' Solid Water
(ug/kg) (ug/kg, dry) (ug/L)
<9000
< 10000
<20000
24500
160000
126000
201000
183000
89400
1 20000
86900
68400
84900
67500
13300
39400
1260000
24
<20
<20
<20
99
17
138
120
57
88
97
66
76
90
16
60
950
1.44
<0.6
<0.7
<0.7
<0.5
<0.6
0.402
<0.5
<0.5
<0.4
<0.4
<0.3
<0.4
<0.4
<0.4
0.289
2.13
Grand Calumet
Oil" Solid Water
(ug/kg) (ug/kg, dry) (ug/L)
<9000
<20000
<20000
18000
170000
104000
280000
257000
1 1 6000
143000
125000
92400
1 1 4000
90500
31000
69500
1610000
2250
121
726
1080
5640
1470
3110
3550
3130
3990
1890
1320
3220
1360
1930
2350
37100
0.301
0.495
0.165
0.278
2.72
0.997
17.1
18.0
8.42
10.9
6.80
3.97
6.18
3.24
0.762
2.84
83.2
River
Oil'
(ug/kg)
<30000
48000
<60000
51800
213000
110000
636000
608000
349000
484000
432000
289000
433000
362000
75400
217000
4310000
* Corrected for actual volumes of oil present in oil/triethylamine samples analyzed.
  (9.3 percent for Saginaw River, 6.3 percent for Buffalo River, 60.0 percent for Grand Calumet River)
                                                    27

-------
       Table 15.  Battelle Data - PCB Concentrations in the Treated Solids, Water, and Oil

                    Buffalo River                 Saginaw River             Grand Calumet River
Contami-      Solids     Water    Oil3      Solids     Water     Oil3      Solids     Water     Oil8
nant	(ug/kg, dry)  (ug/L)   (ug/kg)  (ug/kg, dry)  (ug/L)    (ug/kg)  (ug/kg, dry)   (ug/L)   (ug/kg)

Total PCBs    <300      <0.6    62300     235      <0.6   5012000    440       4.8    268000
•  Corrected for actual volumes of oil present in oilAriethylamine samples analyzed.
  (9.3 percent for Saginaw River, 6.3 percent for Buffalo River, 60.0 percent for Grand Calumet River)
4.2.3  Water
       The concentrations of PAHs and PCBs in the water extracted from the three sediments can also
be found in Tables 14 and 15.  As the data demonstrate, individual PAH and PCB concentrations for the
Buffalo River and Saginaw River residual waters were mainly below the detection limits.  Please note that,
like the PAH and PCB concentrations associated with the treated solids and untreated sediments, the PAH
and PCB concentrations found in the Grand Calumet River residual waters were substantially higher than
the concentrations found in the Buffalo River and Saginaw  River residual waters.  Possibly  additional
extractions could reduce these concentrations to levels comparable to Buffalo River and Saginaw  River
concentrations.
4.2.4  Mass Balance
       For the B.E.S.T.® bench-scale treatability tests, good mass balance closures were obtained for the
solids, water, oil, PCBs, and PAHs.  The following sections address the different mass balances and
expand on those factors that influenced their closure. Tables are included in these sections which provide
the data used to calculate the  mass balance.

       During  the  mass  balance discussions, terms are introduced  which require  definition.  These
definitions are provided as follows:
       •  Input solids include the solids initially present in the sample plus those solids introduced by
          caustic addition.
       •  Output solids are the final product solids and the RCC samples taken during the tests.
       •  Input water includes the water initially present in the sample and the water contributed by the
          addition  of caustic.
       •  Output water consists  of the volume of product water obtained after the cold wash extraction.
                                               28

-------
4.2.4.1 Solids
       Closure of the solids was very good, ranging from 92 to 99 percent (Table 16).  Since the values
used to determine closure were simple weights rather than analytical results, the mass balance was not
compromised by errors associated with analytical methods.  Small quantities of solids deposited in the
vessels and containers used during the treatability study and found in the rag layer resulting from the cold
wash were not accounted for in the mass balance.

                          Table 16.  Battelle Data - Solid Mass Balance

                                     Buffalo River       Saginaw River    Grand Calumet River
 Input
   Total Feed, g                          900                899.7                1399.5
   H2O, %                                 41.96               23.98                57.04
   Total Feed Solids, g (dry)               522.4               682.6                 601.2
   Caustic (d=1.53)                           4.1                  4.1                  12.6
   Total Input, g (dry)                      526.5               686.7                 610.8
 Output
RCC Sample Cold Wash, g
RCC Sample 1st Hot Wash, g
RCC Sample 2nd Hot Wash, g
Final Dry Solids, g
Total Final Solids Output, g
(dry)
Recovery, %
11.2
16.3
29.2
461
517.7

98.3
30.6
23.4
21.3
606
681.3

99.2
6.9
7.1
12.3
485
511.1

91.9
4.2.4.2  Water
        Closure of the water mass balance ranged from 67.5 to 77.9 percent (Table 17).  Water lost to
evaporation and water remaining in the rag layer and in the solids after the product water was decanted
from the reactor vessel following the cold wash (i.e., beyond that directly associated with the triethylamine)
did not contribute to the output water recovered. Because the majority of the water was removed from the
raw sediments during the cold wash, closure calculations are based on data obtained during the cold wash
only.  The closed loop of the full-scale system would probably improve closure results.
                                              29

-------
                         Table 17. Battelle Data - Water Mass Balance

Total Feed, g
Water, %
Total Feed Water, g
Caustic Water, g
Total Input Water, g
Water Recovered, g
Net minus 2% TEA, g
Recovery, %
Buffalo River
900
41.96
377.6
4.1
381.7
262.8
257.5
67.5
Saginaw River
900
23.98
215.8
4.1
219.9
174.9
171.4
77.9
Grand Calumet River
1399.5
57.04
798.3
9.6
807.9
609.8
597.6
74.0
4.2.4.3 Oil
       The amount of oil initially present in the raw sediment and the amount of oil produced by the tests
were used to determine the oil mass balance.  The amount of oil present in either the treated solids or
water is known from past tests to be insignificant and, therefore, was not accounted for. In order to retrieve
the residual oil from the distillation flask, triethylamine was added to the oil so it could be poured.  Battelle
later determined the percentage of oil present in the resulting solutions (6.3 percent for the Buffalo River
solution, 9.2 percent for the Saginaw River solution, and 60.0 percent for the Grand Calumet River solution)
and compensated for these percentages when reporting final data.  The need to compensate for these oil
determinations possibly  introduced error into the  calculation of the oil mass balance.  The calculated
closures ranged from 69 to 192 percent as shown in Table 18.

                          Table 18.  Battelle Data - Oil Mass Balance

Feed Input, g
Oil and Grease, %
Input Oil, g
Final Oil & TEA, g
Oil, %
Final Oil, g
Recovery, %
Buffalo River
900
0.24
2.16
55.9
6.3
3.52
163
Saginaw River
899.7
0.14
1.22
24.4
9.2
2.34
192
Grand Calumet River
1399.5
3.22
45.1
51.6
60.0
31.0
68.7
                                              30

-------
4.2.4.4 PCBs
       The PCB mass balance was calculated using the amount of PCBs found in the feed, product oil
(with triethylamine), and product solids.  The contribution of the PCBs found in the  product water was
negligible and therefore not included in the mass balance calculations.

       The closures of the PCBs were good, ranging from 80 to 129 percent (Table 19). When calculating
these closures, it was assumed that the concentrations of the PCBs in the samples taken by RCC during
the treatability study were the same as those found in the final products. In reality, these concentrations
should be higher, since they were removed during earlier extraction stages. The need to compensate for
the excess triethylamine present in the oil extract solutions may have introduced errors to the determination
of the PCB concentrations found in the product oils. These errors could have subsequently been conveyed
to the PCB mass balance  closures.
                         Table 19. Battelle Data - PCB Mass Balance

Feed Input, g
Solids, %
Dry Solids, %
PCBs, ug/kg
PCBs Input, mg
Buffalo River
900
58.0
522.4
325
0.17
Saginaw River
900
76.0
684.2
21865
15.0
Grand Calumet River
1399.5
43.0
601.2
15003
9.02
 OUTPUT
 OH
  TEA & Oil wt., gm
  PCBs Cone., ug/L
  d of TEA & Oil, gm/L
  PCBs Output wt., mg
 Solids
  55.9
2702
 690
   0.22
     25.4
349,109
    752
     11.8
     51.6
117,156
    730
      8.28
Dry Solids wt., gm
PCBs in Solids, ug/kg
PCBs wt., mg
Total Output PCBs, mg
Recovery, %
522.7
ND
0.0
0.22
129
681.3
235
0.16
12.0
80
511.1
440
0.23
8.51
94
ND = Not Detected
                                             31

-------
4.2.4.5 PAHs
       The closures of the PAHs were calculated using the amount of PAHs found in the feed, product
oils (with triethylamine), and product solids.  The contribution of the PAHs found in the product water was
negligible and therefore not included in the mass balance calculations.

       The closures of the PAHs were good (Table 20), ranging from 90 to 111 percent for Buffalo River
and Grand Calumet River sediments.  Saginaw River sediments, however, realized a closure of 240
percent. This may be attributed to the low concentrations of PAHs initially present in the sediment and the
large errors associated with contaminant concentrations close to analytical detection limits.
                         Table 20.  Battelle Data - PAH Mass Balance
Buffalo River Saginaw
River
Feed Input, g
Solids, %
Dry Solids, %
PAHs, ug/kg
PAHs Input, mg
OUTPUT
OH
TEA & Oil wt., g
PAHs Cone., ug/L
d of TEA & Oil, g/L
PAHs Output wt., mg
Solids
Dry Solids wt., g
PAHs in Solids, ug/kg
PAHs wt., mg
900
58.0
522.4
9,900
5.17


55.9
54,864
690
4.44

522.7
370
0.19
900
76.0
684.2
2,700
1.85


25.4
112,070
752
3.79

681.3
950
0.65
Grand Calumet River
1399.5
43.0
601.2
230,000
138


51.6
1,886,715
730
134

511.1
37,100
19
       Total Output PAHs, mg
       Recovery, %
 4.63
90
  4.44

240
153

111
                                             32

-------
4.3    Summary of Vendor Results
       The analytical results and the extraction efficiencies and mass balances performed by RCC match
well with the analytical  results provided by  Battelle and the extraction efficiencies and mass balance
calculated by SAIC. The RCC data were used to make comparisons with the results obtained and to help
interpret data. Samples of the feed material and the treated solids produced using the B.E.S.T.® Solvent
Extraction Process were analyzed by RCC for residual PCB contamination.  The data from these analyses
are presented in Table 21.

                            Table 21. RCC Data - PCB Summary
Sample
(dry Basis)
Buffalo River
Saginaw River
Grand Calumet River
Feed PCB
Content, mg/kg
(dry basis)
0.60
21
22
Treated Solids
PCB Content, mg/kg
(dry basis)
<0.03
0.18
0.23
PCB Removal
Efficiency
(%)
>95
99
99
       Feed material and treated solids were also analyzed for residual PAH and for metals concentra-
tions. Residual PAH concentrations of <0.2 mg/kg per compound were found in the treated solids produced
by treating the Buffalo River and Saginaw River  sediments.  Treated solids with PAH concentrations
ranging from <1 to <3 mg/kg per compound were obtained for the Grand Calumet River sediments.  In tests
done by RCC, all of the treated solids passed the TCLP Toxicity Test for the leaching of metals.

       RCC performed a mass balance of solids, water, oil, and PCBs using their own data.  Table 22
summarizes the results of this mass balance.
                  Table 22. RCC Data - Mass Balance Summary (Recovery %)
Sample
Buffalo River
Saginaw River
Grand Calumet River
Solids
97
98
86
Oil
112
137
97
Water
70
82
75
PCBs
70
280
64
                                             33

-------
4.4    Quality Assurance/Quality Control
       The conclusions and the limitations of data obtained during the evaluation of RCC's B.E.S.T.®
Process are summarized in the following paragraphs.

       Upon review of all sample data and associated QC results, the data generated for the B.E.S.T.®
treatability study have been determined to be of acceptable quality.  In general, QC results for accuracy
and precision were good and can be used to support technology removal efficiency results.

       In some cases, the demonstration of removal efficiency for PAHs and PCBs may be limited if
relatively small amounts of these compounds are present in the untreated sediments. If minimal amounts
are present, then detection limits become a factor. Removal efficiency demonstration may be limited by
the sensitivity of the analytical methods.

       Refer to Appendix F for the complete analysis related to Quality Assurance/Quality Control.
                                             34

-------
                                 APPENDIX A

                      B JLS.T.® BENCH-SCALE TREATABILITY
                                  TEST REPORT

                          Great Lakes National Program Office
                    Buffalo River, Saginaw Bay and Indiana Harbor Sites
 I.   INTRODUCTION

 SUMMARY

 A bench-scale treatability test of the B.E.S.T. solvent extraction process was conducted on three
 pclychlorinated biphenyl (PCB) contaminated sediment samples. One sample was received from
 each of the sites, Buffalo River, Saginaw Bay and Indiana Harbor.  A summary of the bench-scale
 treatability test results follows:
                  BENCH SCALE TREATABILITY TESTS RESULTS
                          Feed PCB              Product Solids        PCB Removal
 Sample               Content, mg/kg         PCB Content, ma/kg      Efficiency, %
(dry basis)                 (dry basis)

Buffalo River                 0.60                < 0.03                > 95
Saginaw Bay                  21                     0.18                  99
Indiana Harbor               22                     023                  99
As can be seen from the data above, the PCB residuals of the treated solids (Product Solids) varied
from < 0.03 mgAg to 0.23 mgAg, yielding PCB removal efficiencies of > 95 to 99%.  Individual.
residual PAH concentrations in the treated solids were < 0.2 mg/kg for the Buffalo River and Saginaw
Bay samples and ranged from < 1 to < 3 mg/kg for the Indiana Harbor sample.

All of the treated solids readily passed the TCLP Toxicity Test for the leaching of metals.
                                        35

-------
 THE B JLS.T. SOLVENT EXTRACTION PROCESS

 The B.E.S.T. process is a patented solvent extraction technology using triethylamine as the solvent.
 Triethylamine is an aliphatic amine that is produced by reacting ethyl alcohol and ammonia.

 Triethylamine is an excellent solvent for treating hazardous wastes because it exhibits several
 characteristics that enhance its use in the solvent extraction system. These characteristics include:

      •    A high vapor pressure; therefore, the solvent can be easily recovered from the extract
          solution (oil, water, and solvent) via steam stripping.

      •    Formation of a low boiling temperature azeotrope with water, allowing the solvent to be
          recovered from the oil to very low residual levels (typically less than 100 ppm).

      •    A low heat of vaporization (1/7 of water), allowing solvent to be recovered from the
          treated solids with very low energy input.

      •    Triethylamine is alkaline  (pH=10); therefore, some heavy metals are converted to the
          hydroxide form, precipitate and exit the system with the treated solids.

     •    Triethylamine readily biodegrades.  Data available in EPA document EPA Data QRD
          USEPA  Washington. D.C.  20460. Feb. 1983 (reprint) Manual. Volume 1 600/2-82-001 a.
          shows that a level of 200 ppm triethylamine in water was degraded completely within 11
          hours by the common soil bacteria aerobacter.

A block diagram of the B.E.S.T. process is presented in  Figure 1. The first extraction of the
contaminated feed is conducted at low temperatures (about 40 degrees F).  At this temperature,
triethylamine is soluble with water. Therefore, the extract solution contains most of the water in the
feed sample. If the first extract solution contains sufficient water to allow a phase separation of the
solvent and water, the extract is heated to a temperature above the miscibility limit (130 degrees F).
At this temperature, the extract solution separates into two distinct phases, a solvent/oil phase and a
water phase. The two phases are separated by gravity and decanted.  The extract solution from the
subsequent stages is combined  with the decanted solvent/oil phase from the first extraction stage.
The solvent is recovered by steam stripping and evaporation.

Triethylamine is removed from the treated solids by indirect steam heating. A small amount of steam
may be  added directly to the dryer vessel to provide the water required to form the  low  boiling
temperature azeotrope. Residual solvent biodegrades readily,  allowing the treated solids to be used as
backfill at the site in some cases.

The  B.E.S.T. process operates near ambient pressure and temperature and at a alkaline  pH.
Temperatures of the liquid streams within the the unit vary from about 40 to 170 degrees F, and
elevated pressures are not required. This gives the B.E.S.T. process the advantage  that it can use
standard off-the-shelf processing equipment.
                                            36

-------
         RCC
CO
      Waate
                   B.E.S.T. PROCESS CONCEPT
                  Extraction
                  Recycled
                   Solvent
                                      Solvent Recovery
                                                 I
                                           Solvent (to recycle)

                             Solvent/OH	f
Extraction
              Subsequent
              Extractlone
                               First
                               Extraction
                               <3>40*F
   l
Separation
Steam
Stripping
                                                                           Oil
                                            Solvent/Water
               Solvent (to recycle)
                  f
                                          Steam
                                          Stripping
                                                                          Water
                    Solids
                                                           Solvent (to recycle)

                                                             t
                                          Solids
                                          Drying
                                                                         -Solids

-------
  AIR EMISSIONS AND ABATEMENT

  The B.E.S.T. process uses one vent to the atmosphere. The vent provides pressure equalization for
  the nitrogen blanketing system and a purge for noncondensible gases from process condensers. RCC
  uses a refrigerated condenser and an auxiliary water scrubber system to reduce solvent emissions from
  the vent

  During a performance test in February 1987 at the General Refining Superfund Site cleanup, a third
  party reported the following emissions from the B.E.S.T. process vent at a time when the auxiliary
  water scrubber was not in operation:
                                         Emission Rate. Ib/hr

                 Benzene                    0.00114
                 Mercury                  < 0.000000043
                 Toluene                     0.000614
                 Triethyiamine                0.0954
                 Xylene                      0.000884
 RCC expects air emissions from future operations to be similar to these results.  The use of the
 auxiliary water scrubber will lower the triethylamine release rate even further.  RCC now utilizes
 activated carbon filters on the single vent line to achieve zero emissions of triethylamine.
 EQUIPMENT DESCRIPTION

 RCC proposes using a B.E.S.T. Model 615 unit to treat the PCB-contaminated material at this site.
 The B.E.S.T. Model 615 unit has a design capacity of approximately 200 - 300 tons of feed per day.
 A flow schematic for the B.E.S.T. Model is presented in Figure 2.

 The B.E.S.T. Model 615 uses an extractor/dryer vessel to extract and dry the PCB-contaminated
 materials.  The extractor/dryer is a horizontal, steam-jacketed vessel that allows for solvent
 contacting, mixing, solids/solvent separation, solids drying, and solids conditioning in one vessel.
 The extractor/dryer vessel is an off-the-shelf assembly that has a long history of reliable performance
 in a wide range of process industry applications.

 Contaminated materials are excavated from the site and screened to one inch maximum dimension.
The screened material is then loaded into top-loading, bottom-discharge hoppers.  An overhead crane
 facilitates  the positioning and lowering of the loaded hopper onto the loading port  of the
extractor/dryer unit. The flow of material through the extractor/dryer system is shown in Figure 3.
Treated solids are discharged into hoppers and  transported to a holding area.

Figure 4 provides the standard Site Plan for RCC's B.E.S.T. Model 615.
                                           38

-------
                                               CONDENSER
               CONDENSER
             (DRYING CYCLE)
                         SOLVENT
                              OPTIONAL
                             CENTRIFUGE
     CLEAN
     SOLIDS
    PRODUCT
SOLVENT
                      SPENT
                      SOLVENT
                    TANK/SETTLER
  TANK
  n
r     i
           CHILLER
                     lANKh
                                    SOLVENT
                                    MAKEUP
                                 CLEAN
                                SOLVENT
                                  TANK
                                          SOLVENT
                                         EVAPORATOR
 SOLVENT
DECANTER
                                                                                PRODUCT
                               INOKCOHO
                                             PROCESS FLOW SCHEMATIC
                                                    r®
 B.E.S.TW MODEL  315/615

  SOILS  TREATMENT UNIT
                                                                     DWG NO.

                                                                     B-221
                                                                               »«»«
                                                                               1  OF
                                                                                    •"
                                                                             A
                                                                            A

-------
  f?CC
Httourttt
Conttrvitlon
Company
 B.E.S.T.® PROCESS STEPS
BATTERY LIMITS OPERATION
                                  SOUDS SETTUNG
                                AND SOLVENT DRAINING
WASHER/DRYER CHARQNO
                                                                 3

-------
                 Procession
Storane  Tanks   Equipment    Six Extractor/Dryers
                                                     RCSOURttS COM$tBV«nOM COUPtHt

-------
 BENCH-SCALE TREATABILITY TEST DATA CORRELATION TO FULL-SCALE
 PERFORMANCE

 In order to evaluate each potential application for the B.E.S.T. process. RCC has developed a low cost
 bench-scale treatability test protocol that provides data that closely simulates full-scale system
 performance.  The bench-scale treatability test data allows RCC to evaluate the feasibility of the
 process on a particular sample and to estimate treatment costs.

 The reliability of the bench-scale treatability tests to predict full-scale performance has been verified
 by the US EPA report Evaluation of the B.E.S.T. Solvent Extraction Sludge Treatment Technology -
 Twenty-Four Hour Test, by Enviresponse. Inc., under EPA Contract 68-03-3255.  A quote from this
 report evaluating the B.E.S.T. process states:
     "Resources Conservation Company has conducted many laboratory tests and developed
     correlations to which data from full-scale operations, such as the General Refining site,
     can be compared."

Figures 5 and 6 present data from two separate bench-scale treatability tests and full-scale operating
performance data at the General Refining. Inc.. Superrund site, as collected by an EPA contractor.
This data demonstrates a close correlation between bench-scale treatability test data and full-scale
operating data.

Bench-scale treatability testing provides valuable information about the use of the B.E.S.T. process at
full-scale including:

     •    The PCB removal efficiency from the sample.

     •    Solids separation requirements for full-scale operation.

     •    The separation efficiency of water from the water/solvent/oil solution by decapitation.

     •    General information on the partitioning of metals and organic compounds in the oil, water.
         and solids products.

     •    Full-scale operating parameters to develop treatment costs.
                                         42

-------
RCC
Resources
Conserirjffon
Company
                GENERAL REFINING SITE
PCB CONCENTRATIONS IN RAW SLUDGE & PRODUCT FRACTIONS
                       (ppm)
                          LAB SCALE TESTING (1986) FULL SCALE PROCESSING
                          TEST "A"       TEST "B"     FEB. 26-27,1987
 RAW SLUDGE (DRY BASIS) mg/kg  14.

 PRODUCT SOLIDS mg/kg          0.02

 PRODUCT WATER mg/L         <0.01

 "/..EXTRACTION EFFICIENCY      99.9
                                 12.

                                  0.14

                                 <0.01

                                 98.8
                                      13.5

                                      <0.13

                                      <0.005

                                     >99.0
ua
3
in

-------
COMPARISON OF BENCH SCALE TO FULL SCALE
Off**™"™'**" PHASE SEPARATION PERFORMANCE
• • ^* ^* Company FOR
GENERAL REFINING SITE SLUDGE
Raw Sludge
Oil % 36
Water % 56
Solids % 8
N/A Not Available
BS&W = 2.8%
Bench Scale
Separated Phase Fractions
Oil Water Solids
>97. .017 5.7
N/A <1.0
N/A >94.
Raw Sludge
27
66
7
Full Scale
Separated Phase
Oil Water
99. 0.0033
0.88 >99.
* 0.81
Fractions
Solids
0.81
<05
>98.
                                                                                                                                         •a
PUOAG4

-------
  BENCH-SCALE TREATABILJTY TEST DOCUMENTATION

  The documentation of the testing can be separated into three distinct categories. The following
  summarizes the procedures used for each step of the trcatability process:

      1.   When the samples were received in the laboratory, the shipment was checked for
          correctness of accompanying paper work, including Chain of Custody. The information
          was recorded both in a hardbound sample logbook and on a computer system that has been
          specifically designed by RCC for use in tracking samples. The samples were issued a
          discrete laboratory sample number, and a test request form was completed. The samples
          were kept in a refrigerator under controlled and documented temperature prior to any lab
          analysis or the trcatability study.  Chain of Custody records and other information received
          with the samples are kept as pan of the project file.

      2.   The bench-scale treatability testing was conducted in accordance with the test plan, and all
          records and observations taken during the simulation of the process were recorded in
          laboratory notebooks. The laboratory notebooks are the property of RCC, and each analyst
          and engineer has been issued a notebook. The notebooks are retained by RCC as
          permanent record of raw data collection.

     3.   Samples that were collected during the bench-scale test, including samples internal to the
          process, were submitted to the RCC analytical chemistry laboratory for further analysis.
          Each sample collected was issued a discrete laboratory number.  An analysis request form
          was completed. The samples were analyzed in accordance with RCC's Laboratory Quality
          Management Plan and  reviewed for correctness prior to issuance.  A file is maintained to
          permanently store  the accumulated test results from completion of the analytical testing.

The bench-scale treatability test plan is provided in Attachment 1. PCB chromatograms are given in
Attachment 2. PAH chromatograms are given in Attachment 3.
H.   BENCH-SCALE TREATABILITY TESTING

SAMPLE PREPARATION

The contaminated samples from the Buffalo River, Saginaw Bay and Indiana Harbor Sites arrived at
RCC's laboratory in July, 1991. The samples were labeled B-US-RCC, S-US-RCC and I-US-RCC,
respectively. All samples were gray-colored sediments with very little debris present.  Each of the
three samples contained free standing water. This water was decanted prior to conducting feed
analyses and was proportionally recombined prior to any analysis and bench testing. This effort was
taken since it was very difficult to homogenize the samples with the free standing water present

Bench-scale testing requires material greater than  1/4 inch be removed.  There was no material greater
than 1/4 inch in any of the three samples received.  Therefore, the samples were not screened.


                                        45

-------
FEED COMPOSITIONAL ANALYSIS

The feed was analyzed for percent oil, water, solids and metals per the following methods:

    •    The oil & grease content was determined as per Standard Methods for the Examination of
         Water and Wastewater. 16th Edition, Method 503D, with two exceptions: the extraction
         time was extended from 4 to 16 hours, and methylene chloride (MeQ2)was substituted for
         Freon based on RCC experience that MeCl2 is a better solvent for oils and greases.

    •    The water content was determined by weight loss at 70 degrees C.

    •    The paniculate solids content was determined by rinsing a known quantity of feed through
         a Whatman GF/C filter under vacuum with acetone followed by Med^ The ^idue was
         then dried and weighed. Since the paniculate solids content was by far the largest
         component of all of the feeds, the percent solids values below are by difference.

    •     The PCS concentration was determined per EPA Publication SW846 Test Methods for
         Evaluating Solid Waste. Method 8080. The sample extraction method was by Soxhlet
         extraction with 1:1 acetone:hexane for 16 hours.  The PCBs were quantitated as Arocior
         1242.

    •    The Polynuciear Aromatic Hydrocarbon (PAH) concentrations were determined per EPA
        Publication SW846, Method 8100 (1:1 Acetone:Hexane extraction solvent).

    •    The metals composition (except for Mercury) was determined by nitric acid digestion after
        ashing at 550 degrees C, followed by ICP analysis (EPA SW846, Method 6010).

    •    Mercury concentration was determined by the Cold Vapor Technique, Method 303F, of
        Standard Methods for the Examination of Water and Wastewater.

    •    Loss on ignition was determined by heating a sample from 105 to 550 degrees C which is a
        measure of the total organic content.
                                        46

-------
 The results of these analyses on a wet basis were as follows:
 Analyte
                               Feed Compositional Analysis
                                   (wet baste unless noted)
                         Sample Results
 Oil & Grease (by MeCl2),
 Water, %
 Solids. %
 Loss on Ignition, %

 PCBs, mg/kg, dry basis
        Buffalo River

             0.48
            41.
            59.
             4.8

             0.6
                                                    Saginaw Bay    Indiana Harbor
              0.36
             23.
             77.
             21
              2.4
             56.
             42.
             26.

             22
Individual feed PAH concentrations for the Buffalo River sample ranged from <0.5 to 6 mg/kg, for
the Saginaw Bay sample they were <0.4 to <3 mg/kg and for the Indiana Harbor sample they were <8
to <60 mg/kg.  These results are presented on page 17 and 18.

The heavy metals composition of each feed was as follows:
                            Feed Metals Composition, mg/kg
                                    (As received basis)
      Anaivte
      Antimony
      Arsenic
      Barium
      Cadmium
      Chromium
      Copper
      Lead
      Mercury
      Nickel
      Selenium
      Silver
      Sodium
      Zinc
                       Buffalo River
<30.
  34.
 <2.
  14.
  16.
  28.
   0.38
   9.7
<20.
 < 1.
 115.
  52.
Saginaw Bay

   < 15.
   <40.
     32.
    <3.
     62.
     37.
     37.
      0.10
     36.
   <25.
    < 1.
    150.
    120.
Indiana Harbor

    < 15.
    <35.
      76.
       6.4
     156.
      96.
     290.
       0.72
      29.
    <20.
     <0.7
     300.
   1,340.
                                         47

-------
 TRIETHYLAMINE COMPATIBILITY TEST

 Tricthylamine is a compound with a unique chemical structure. The geometry of the structure is
 tetrahedral, meaning that the nitrogen atom is at the center of a three-sided pyramid. The four points
 of the pyramid structure are occupied by three ethyl functional groups and one electron cloud. This
 structure gives triethylamine dual polarity characteristics. The ethyl groups are essentially nonpolar.
 the electron cloud is polar. Although triethylamine is a very stable solvent, there is a remote
 possibility that the electron pair can react with certain types of materials. In order to determine if this
 will occur with a sample, a compatibility test is performed.  This involves mixing of the sample with
 triethylamine and making observations as to the heat of solution and any other visual signs of
 reaction.

 When each feed sample was mixed with cold triethylamine, no visible sign of adverse reaction was
 observed, and the heat of solution was in a normal range. The triethylamine was observed to darken
 upon mixing, indicating that extraction of the organic compounds was occurring.

 Based on the favorable results of this preliminary test, it was decided that the B.E.S.T. bench-scale
 treatability test should proceed.
FEED pH ADJUSTMENT

Triethylamine can be ionized at low pH to triethyiammonium salts that cannot be removed from the
products. The alkaline nature of triethylamine will buffer the pH of the sample to a pH of around 9.
The solvent spent in the pH buffering will be lost. In order to efficiently recover the triethylamine
from the separated phase fraction products, the pH of the sample is adjusted to about 11 with causuc
soda.

A 5-gram portion of each feed sample was siurried  with deionized water. The pH of this mixture
indicated that caustic would need to be added to each sample. Incremental portions of causuc soda
(NaOH) were added to bring the pH to 11. The amount of causuc that was required to perform this
pH adjustment and the original sample pH is summarized below:
                              Sample pH and Caustic Dose

                                                         Caustic Dose
                                      pH            (mis 50% NaOH per kg)

         Buffalo River                 7.6                      6.0
         Saginaw Bay                 8.1                      6.0
       Indiana Harbor                7.5                      9.0
                                        48

-------
 SAMPLE EXTRACTION/PRODUCT SOLIDS

 A portion of the Buffalo River, Saginaw Bay and Indiana Harbor samples was prechilled by placing
 each in a 4-liter resin kettle, immersed in a temperature controlled water bath set at 0.5 degree C.
 Each sample pH was adjusted by adding caustic soda at the same time that chilled triethylamine was
 added. Mixing was performed by an air-driven prop mixer in the same 4-liter resin kettle immersed
 in the cooling bath.

 As expected, the solvent became colored for all three samples, indicating extraction of organic
 compounds was occurring. After mixing,  the solvent/oil/water liquid extract was separated from the
 solids by centrifugation. The liquid extract was temporarily set aside for testing as discussed later
 under DECANTATION OF WATER.

Two more extraction stages were performed on the solids, for a total of three extraction stages.  No
 additional caustic was added for the subsequent  extraction stages.  A sample of the Product Solids
 was collected for analysis as follows:
    •    The oil & grease content was determined as per Standard Methods for the Examination of
         Water and Wastewater. 16th Edition, Method 503D, with two exceptions: the extraction
         lime was extended from 4 to 16 hours, and methylene chloride (MeCty was substituted for
         Freon based on RCC experience that MeCl2 is a better solvent for oils and greases.

    •    Loss on ignition was determined by heating a sample to 550 degrees C for 3 hours.

    •    The PCB concentration was determined per EPA Publication SW846 Test Methods for
         Evaluating Solid Waste. Method 8080. The sample extraction method was by Soxhlet
         extraction with 1:1 Acetone:Hexane for 16 hours. The PCBs were quantitated as Aroclor
         1242.

    •    The metals composition was determined by aqua rcgia digestion, followed by ICP analysis
         (EPA SW846, Method 6010).

    •    The triethylamine content was determined by  shaker bath water extraction and packed
         column gas chromaiography with a flame ionization detector.

    •    The pH was determined by measuring the pH of a slurry of 5 grams of sample and 50 mis
         of deionized water. The slurry was tested by pH probe after mixing overnight.

    •    The Polynuclear Aromatic  Hydrocarbon (PAH) concentrations were  determined per EPA
         Publication SW846, Method 8100 (1:1 Acetone:Hexane extraction solvent).

    •    Mercury concentration was determined by the  Cold Vapor Technique, Method 303F, of
         Standard Methods for the Examination of Waste and Wastewater.
                                          49

-------
 PCB analytical results of all of the solid samples were as follows:



                             PCB Analysis Summary, me/kg
                                     (all data dry basis)


                               Buffalo River     Saginaw Bay      Indiana Harbor


Feed                               0.60             21                 22

Product Solids                      <0.03              0.18                0.23
Total polynuciear hydrocarbon (PAH) analytical results of the feeds and product solids from the three
samples are given on the following two pages. Due to a large number of interfering compounds eluting
at or near the retention times of the analytes of interest, it was not possible to report lower detection
limits.
                                         50

-------
                               PAH Summary for Buffalo River
                                            (nig/kg)

                                      Feed
                                      Product
                                       Solids
 Naphthalene
 2-Methylnaphthalene
 Acenaphthylene
 Acenaphthene
 Dibenzofuran
 Fluorene
 Phenanthrene
 Anthracene
 Fluoranthene
 Pyrene

 Benzo(a)anthracene
 Chrysene
 Benzo(b)fluoranthene
 B enzo(k)fluoranthene
 Benzo(a)pyrene
 Indeno(l,2,3-cd)pyrene
 Dibenzo(a,h)anthracene
 Benzo(gji,i)perylene
 <0.5
 <0.5
 <0.5
 <0.5
 < 1.
 <0.5
<0.5
<0.5
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2

                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                               <0.2
                              PAH Summary for Saginaw Bay
                                           (mg/kg)

                                      Feed
                                     Product
                                      Solids
Naphthalene
2-Methylnaphthalene
Acenaphthylene
Acenaphthene
Dibenzofuran
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene

Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Indeno( U,3-cd)pyrene
Dibenzo(a,h)anthracene
Benzo(gji,i)perylene
<0.4
<0.4
<0.4
<0.4
<0.4
<0.4
<0.8
< 1.
<2.
< 1.
<0.4
<0.4
<0.4
<0.4
<0.4
<0.4
51
<0.2
<0.2
<0.2
<0.2
<0.2
<0.2
<0.2
<0.2
<0.2
<0.2

<0.2
<0.2
<0.2
<0.2
<0.2
<0.2
<0.2
<0.2

-------
                            PAH Summary for Indiana Harbor
                                          (mg/kf)

                                    Feed
                                                                        Product
                                                                         Solids
 Naphthalene
 2-Methylnaphthalene
 Acenaphthylene
 Acenaphthene
 Dibenzoniran
 Fluorene
 Phenanthrcne
 Anthracene
 Fluoranthene
 Pyrene

 Benzo(a)anthracene
 Chrysene
 Benzo(b)fiuoranthene
 Benzo(k)fluoranthene
 Benzo(a)pyrene
 Indeno(1.2,3-cd)pyrene
Dibenzo(aji)anthracene
Benzo(gth,i)perylene
                                    < 6.
                                    < 6.
                                    <4.
                                    < 7.
                                   < 15.
                                   < 12.
                                   < 20.
                                   < 12.
                                   < 45.
                                   < 48.

                                   < 40.
                                   < 38.
                                   < 25.
                                   <28.
                                   < 33.
                                   < 13.
                                   < 10.
                                   < 13.
                                                                         < 3-
                                                                         < 3-
                                                                         < 1-
                                                                         < 1-
                                                                         < 3.
                                                                         < 1-
                                                                         < 3.
                                                                         < 1-
                                                                         < 2.
                                                                         < 2.

                                                                         < 2.
                                                                         < 3.
                                                                         < 2.
                                                                         < 1-
                                                                         < 1-
                                                                         < 1-
                                                                         < 1-
                                                                         < 1-
Additional product solids analyses, all on a dry basis since the product solids were dried in-process,
follows:

                                 Product Solids Analysis
                                  (three extraction stages)
Analyte
                                                Sample Results

                                Buffalo River       Saginaw Bav      Indiana Harbor
PCBs,mg/kg
Oil & Grease (by MeCl2), %
Triethylamine, mg/kg
Loss on Ignition, %
                                    <0.03
                                      0.22
                                     37.
                                      4.0
                                                         0.18
                                                         0.15
                                                        20.
                                                         2.1
                                                                           023
                                                                           0.52
                                                                          28.
                                                                          20.
                                         52

-------
  PCB removal efficiency is determined by comparing the amount of PCBs in the feed to the amount
  remaining in the environment after treatment. The fraction of PCBs remaining in the environment is
  calculated by dividing the PCB content of the product solids by the PCB content of the feed, on a dry
  basis. An example of the calculation, the Saginaw Bay sample, follows:
                                  Saginaw Bay Sample
                           PCB Removal Efficiency Calculation
     Fraction of PCBs remaining
     in environment
                     Product solids PCB Content (dry basis)
                     Feed PCB Content (dry basis)
                                             0.18 me/kg
                                             20.5 mg/kg
                                   = 0.00878
                  % Removal from      =      100 • (1 - fraction of PCBs remaining
                    environment                   in environment)

                                             100 • (1-0.00878)

                                             99.1 %

The reduction in the PCB content and the corresponding removal efficiency of PCBs from the
environment is summarized below for all of the samples:
                             Total PCB Removal Summary
Sample
  PCBs in
Feed, me/kg
  (dry basis)
PCBs in Product
  Solids, mg/kg
    (dry basis)
  Removal
Efficiency. %
Buffalo River
Saginaw Bay
Indiana Harbor
      0.60
     21
     22
     <0.03
       0.18
       0.23
   >95
     99
     99
                                         53

-------
 Total heavy metal analysis of the product solids was as follows:

                                     Product Solids
                              Total Metals Analysis, (me/kg)
 Analvte
                      Buffalo River      Saginaw Bav      Indiana Harbor

 Antimony                 < 20.              < 20.                 <20.
 Arsenic                   <5Q.              <50.                 <50.
 Barium                    73,                41                   igo.
 Cadmium                  < 3.               < 5.                  24.
 Chromium                  51.                93                  430.
 Copper                    62.                60.                 270.
 Lead                      110.                56.                 750.
 Mercury                     0.68                0.20                 1.6
 Nickel                     32.                53.                  88.
 Selenium                 < 30.              < 30.                < 30.
 Silver                     
-------
  TOXICITY CHARACTERISTIC LEACHING PROCEDURE ANALYSIS ON PRODUCT
  SOLIDS

  The product solids from each sample were extracted using the Toxicity Characteristic Leaching
  Procedure (TCLP) in accordance with Federal Register, March 29, 1990. Each TCLP leachate was
  analyzed for metals content The results from this analysis were as follows:
                                       Product Solids
                               TCLP Leachate Analysis, mg/I
 Analvte
Buffalo River
Saginaw Bav     Indiana Harbor
                Regulatory
                Level, mg/1
Arsenic
Barium
Cadmium
Chromium
Lead
Mercury
Selenium
Silver
   <0.5
    0.11
   <0.03
   <0.05
   <0.1
   < 0.002
   <0.3
   <0.01
    <0.5
     0.14
    <0.03
     0.15
    <0.1
    < 0.002
    <0.3
    <0.01
<0.5
  0.04
<0.03
  0.07
<0.1
  0.0024
<0.3
  0.05
  5
100
  1
  5
  5
  0.2
  1
  5
As can be seen from the above data, all product solids readily passed the TCLP test for metals.
                                         55

-------
 DECANTATION OF WATER

 The solvent recovered from each first extraction stage was separated into its aqueous and organic
 components.  Only the extract from the first extraction stage had a significant amount of water in
 solution, so only the water in the first stage extract is recovered. There are two methods available to
 recover the water in the first stage extract. Each method has  its advantages. The method chosen
 largely depends on the water content of the feed.

 In the first method of recovering the water, the triethylamine mixture is heated to  about
 140 degrees F. The water phase is separated from the triethylamine/oil phase by decamation.  At the
 elevated temperature, water is no longer miscible in triethylamine and settles to the bottom of the
 mixture. The  heated mixture is allowed to stand and separate for 30 minutes in a 4-liter separatory
 funnel. The separatory runnel is immersed into a clear tank into which water from  a temperature
 controlled water bath is pumped.  Excess water from the tank drains back into the water bath. The
 temperature controlled bath is set at 140 degrees F.  This low energy method is preferable for high
 water content feeds.

 Decantation (the first method) was used for both Buffalo  River and Indiana Harbor. The separation
 does not give only TEA/oil and water phases due to the presence of oil and solids in the mixture.  In
 between the two phases is a 'rag' layer or emulsion where any solids present tend to collect and  create
 a region where the TEA/Oil/Water separations is not distinct. The smaller this rag layer is in
 comparison to  the TEA/OH and water phases, the better the separation. For the Buffalo River sample
 the rag layer was only 2.6% of the entire TEA/Oil/Water mixture. For the Indiana Harbor sample the
 rag layer was  3.1%.  In  both cases, the rag layer is a very small fraction of the whole mixture,
 therefore, the separation of TEA/Oil from water was good.

 In the second  method of recovering water, the water from the feed is separated from the oil by
 evaporation as opposed  to decantation. When the triethylamine/oil/water first stage extract is
 evaporated, as described in section SOLVENT EVAPORATION/PRODUCT OIL, the water forms an
 azeotrope with the distilled triethylamine, leaving the oil behind. The water is then separated from
 the triethylamine of the condensed triethylamine/water azeotrope by decantation.  The
 triethylamine/water recovered from  the Solvent Evaporation/Product Oil  step is heated to
 140 degrees F, then poured into a 4-liter separatory funnel. Separation occurs immediately, so no
 temperature control system is required. This separation is highly effective because there is virtually
 no oil or solids  in the condensed triethylamine/water that would hinder the separation of triethylamine
 from water by  decantation. This method is preferable for low water content feeds where the extra
energy cost to evaporate the water is small.  This method  yields a much purer water stream, usually
avoiding the requirement for possible post-treatment of the water.

Evaporation (the second  method) was used for the Saginaw Bay sample because of its low  water
content (< 25%).  The aqueous phase separated from the organic phase, producing excellent results
for each.
                                         56

-------
 PRODUCT WATER

 Removal of residual triethyiamine from each decant water was accomplished by heating the water on
 a hot plate while maintaining an elevated pH. The elevated pH is necessary to ensure that the
 majority of the triethyiamine remains in the volatile molecular form. Triethylamine/waier azeotrope
 boils at about 170 degrees F. When the triethyiamine is removed, the water temperature increases to
 212 degrees F. Analysis of each stripped water, labeled Product Water, was conducted as follows:

     •    The triethyiamine content was determined by packed column gas chromatography with a
          flame ionization detector.

     •     There was insufficient sample for Total Petroleum Hydrocarbons, Oil & Grease, PCS or
          total metals analyses since the bulk of the water was sent to a third party lab.

Results of these analyses were as follows:


                              Product Water Analysis, mg/l


        Analvte                  Buffalo River     Saginaw Bav      Indiana Harbor

    Triethyiamine                       7.               JQ.                13.
        PH                            10.3               10.4               10.8
                                         57

-------
 SOLVENT EVAPORATION/PRODUCT OIL

 Recovery of product oil (the organic compounds in the feed) is normally accomplished in two steps.
 First, the  bulk of the triethylamine is recovered by distillation. This is done by boiling the
 triethylamine/oil mixture in a Bucni Rotovapor® apparatus. The oil remains in the boiling flask of
 the Rotovapor while the triethylamine is condensed as it evaporates and is collected separately.
 Second, any residual triethylamine is stripped from the oil by adding water to the hot oil in the boiling
 flask of the Rotovapor.  Water is added to form the low boiling triethylamine/water azeotrope. This
 second step was not used for these samples due to the  low oil content of the feeds. With a low oil
 content feed, the amount of oil recovered is so small that effective stripping is not possible. The oil
 from the sample was kept in a solution of triethylamine with only a fraction being completely dried of
 solvent In this way, the oil remains homogeneous and can be poured out of the rotovapor flask. This
 is vital for the integrity of the analyses on the oil, including PCBs from which  a PCB  mass balance is
 determined. The analysis of each product oil follows:

      •    The  metals composition was  determined by nitric acid digestion after ashing at
          550 degrees C, followed by ICP analysis (EPA SW846, Method 6010).

     •   The PCB  concentration was determined by dilution of the oil in hexane, followed by EPA
         SW846 (Test Methods for Evaluating Solid Waste). Method 3620. sulfuric acid and/or
         Florisil column cleanup. The prepared sample was then analyzed  by EPA SW846, Method
         8080.

The oil in a triethylamine diluent was analyzed and the results converted to  a pure oil (triethylamine
free) basis.  The results were as follows:
                             Product Oil Analysis, dry basis


Analvte               Buffalo River       Saginaw Bay        Indiana Harbor


PCBs. mg/kg                 4.0            1.600                 160

Metals, mg/kg

     Antimony            < 15.               < 20.                 < 4.
     Arsenic              < 40.               <50.                <  10.
     Barium                 3.                15.                   2.
     Cadmium               3.                 4.                 <0.6
     Chromium               5.                58.                  50.
     Copper               110.              3.800.                   12.
     Lead                   20.               290.                   6.
     Nickel                 20.              1.400.                   7.
     Selenium             < 25.               < 30.                 < 6.
     Silver                 < 0.8               39.                   0.2
     Sodium              400.            22,000.                  44.
    Zinc                   19.               100.                   5.
                                           58

-------
  m. MASS BALANCES

  The data gathered during the bench-scale treatability test provides the data required to calculate mass
  balances.  The mass balances have been segregated into four groups: solids, oil, water, and PCBs.
  SOLIDS MASS BALANCE

 The mass balance for solids is a comparison of the solids input during the test to the solids recovered
 after the test  The mass of solids input during the test includes the solids portion of the feed extracted
 and the solids portion of caustic soda added.  The solids portion of the feed extracted was calculated
 by multiplying the weight of feed extracted by the solids content as determined by analysis.  The
 solids portion of the caustic soda added was calculated by multiplying the weight of the 50 percent
 NaOH solution added by 0.50.

 The mass of the solids recovered from the test is equivalent to the sum of the product solids and
 samples taken for stage-by-stage assays. A summary of this data follows:

                                    Solids Mass Balance

Total Feed Extracted. Wei Basis
Solids Portion of Feed
Solids Portion of Caustic
Total Calculated Solids Input
Buffalo
900
530
+ 4
= 534
River
g
g
•1 g
g
Saginaw
900
690
+ 4.1
= 694
Bav
g
g
g
g
Indiana Harbor
1,400 g
584 g
+ 9.5 g
= 594 g
Weight of Product
Solids Recovered

Weight of Solids
Samples Recovered
    461    g
+    57  g
    666  g
+    75  g
    485   g
+    26   g
Total Solids Recovered
=   518  g
=   681   g
=   511   g
Recovery, %
    97
                                            59
     98
     86

-------
 OIL MASS BALANCE

 The oil mass balance was computed using the same method used in calculating the solids mass
 balance.  The oil & grease content of each feed was determined by extracting a sample of the feed
 with methylene chloride.  This oil &. grease content (by MeCl2) was multiplied by the weight of the
 feed input to determine  the amount of oil input. The mass of oil recovered from the test was
 equivalent to the product oil recovered. The residual oil in the product solids and product water was
 negligible when calculating an oil mass balance.

 The oil mass balances (based on methylene chloride) were as follows:

                                    Oil Mass Balance
 Buffalo River

 Saginaw Bay

 Indiana Harbor
Calculated
 Oil Input

  4.30 g

  3.24 g

 33.6  g
Equivalent Product
   Oil Recovered

      4.83  g

      4.44  g

     32.5   g
% Recovery

   112 %

   137 %

    97 %
Virtually all of the PCBs from the sample now reside in the product oil. For each sample, the weight
of PCB contaminated material was reduced from 900 grams (1400 grams for the Indiana Harbor
sample) to 3.4-30 grams corresponding to a 47-270 times reduction in mass.
WATER MASS BALANCE

The water mass balance was computed similarly to the method used for solids. The mass of water
input came from the water in the feed, plus the water introduced with the caustic. The water portion
of each feed was calculated by multiplying the weight of the feed by the water content as determined
by analysis. The water portion of the caustic input was calculated by multiplying the weight of the 50
percent NaOH solution by half.

The mass of water recovered was equivalent to the sum of the decant water, residual water in the
decant triethylamine/oil, the water contained in the rag layer, and residual water in the subsequent
extraction extracts. A summary of this data follows:
                                          60

-------
                                   Water Mass Balance
                                  Buffalo River      Saginaw Bay     Indiana Harbor
  Water Portion of Feed                    365 g           207  g            784   g

  Water Portion of Caustic            +     4.0 g       +     4.1 g        +    9.5 g
 Total calculated water input          =  369   g       =   211   g        =   793   g
 Water recovered from
 decant water                          258   g           172   g            598  g
 Total water recovered                =  258   g       =  172   g        =  598   g


 % Recovery                            70               82                75
The recovery of water was low. The temperature tends to increase above the triethylamine/water
miscibility limit when the treated solids arc centrifuged.  At these conditions, some water may have
exited the centrifuge with the solids. This water was lost when the solids were dried. In addition, a
portion of the water in the feed was left behind in the resin kettle after decantation of the first
extraction since it is not possible to decant all the solvent from the solids. This water was lost when
the solids were dried.  This portion of the water lost in the dryer is not accounted for in the water mass
balance. (In RCC's Pilot Unit, and Full-Scale Unit, all such water is recovered from the dryer.)


PCB MASS BALANCE

The PCB mass balance was computed similarly to the method used for oil. The mass  of PCBs input
was calculated by multiplying the weight of each feed by the PCB concentration as determined by
analysis. The PCBs recovered from the test reside in the product oil.  The PCBs in the product solids,
product water and recovered triethylamine were negligible when calculating a PCB mass balance.
The mass of PCBs recovered in the oil was calculated by multiplying the weight of oil recovered by
the PCB concentration as determined by analysis. The PCB mass balance for each sample was as
follows:
                                          61

-------
                                    PCB Balance
Calculated
PCBs Input
0.32 mg
14.2 mg
12.8 mg
Calculated
PCBs Recovered
0.22 mg
40.0 mg
8.3 mg
Total PCB
% Recovery
70 %
280 %
64 %
 Sample


 Buffalo River

 Saginaw Bay

 Indiana Harbor
 SUMMARY OF MASS BALANCE CALCULATIONS

 The following table summarizes the mass balance calculations for each of the constituents considered.
 The mass balances were based on the amount of the fraction recovered from the simulation divided by
 the calculated input amount to the simulation.
                             Mass Balance Summary, %


Sample

Buffalo River

Saginaw Bay

Indiana Harbor
Solids
97
98
86
on
112
137
97
Water
70
82
75
PCBs
70
280
64
                                        62

-------
IV.   CONCLUSIONS

The PCS-contaminated sediment samples finom tbe Buffalo River, Saginaw Bay and Indiana Harbor
Sites are suitable for treatment with the B.E.S.T. solvent extraction process. No problems were
observed  during testing of the samples.  Consequently, full-scale processing should be
straightforward.

      1.   The samples were chemically compatible with triethylamine.

      2.   The total PCB concentrations in the samples tested, 'Buffalo River', 'Saginaw Bay' and
          'Indiana Harbor' were 0.60,21 said 22 mg/kg, respectively.

      3.    After treatment, the PCB residual removal efficiencies were > 95% for the Buffalo River
          sample and 99% for the Saginaw Bay and Indiana Harbor samples.

      4.   The PAH residual concentration in the treated product solids were < 0.2 mg/kg for the
          Buffalo River and Saginaw Bay samples and ranged  from < 1 to < 3 mgAg for the
          Indiana Harbor sample.

     5.    All three treated solids readily passed the TCLP Toxicity  Test for leaching of metals.

     6.    Virtually all of the PCBs from the samples have been concentrated into the product oils.
          For each sample the weight of PCB contaminated material was reduced 47-270 times.
                                      *****
                                       63

-------
                           APPENDIX B
B.E.S.T.®  BENCH SCALE TREATABILITY TEST PLAN
            The original Plan contained proprietary information. RCC has removed that
            Information from this copy. For this reason, this copy contains blackened-
            out areas.
                         February, 1990
R«qit»f«d in ff» OS. P«»ne Otloi
                              64

-------
 1.0       TEST OBJECTIVES

 The test objectives are:

     o    To achieve good separation of each type of sample's phase components (oil, water,
          solids) with low thethylamine residuals in each.

     o    To record observations and data that will allow us to predict how a full-scale B.E.S.T.
          separation of the samples might proceed.

     o    Take samples during the extraction tests and conduct analysis sufficient to allow for
          calculation of mass balances for oil. water, solids and other compounds of interest

     o    To calculate the extraction efficiency of compounds of interest (i.e., PCB's if present)
          achieved during the bench-scale workups in order to determine the number of
          stages appropriate for full scale treatment of the site materials.

 Evaluation of attainment of these objectives will consist of analysis of the feed and products and
 observation of the bench-scale simulation in action.


2.0       TEST PLAN

The following tasks will be performed on the feed sample:

2.1       Characterize the Feed Sample

          o     Phase compositional analysis (oil. water, solids)
          o     Raw feed metals composition
          o     Analysis of other compounds of interest (i.e., PCB's if present)

2.2       Perform Preliminary Tests

          o     Sample pH adjustment characteristics
          o     TEA/Feed compatibility study

2.3       Conduct a B.E^.T. Bench-scale Treatabillty Test Wortcup

2.4       Feed Compositional Analysis

2.4.1      Analyze the feed for percent oil, water, and solids.

Determine the total oil per Standard Methods for the Examination of Water and Wastewater.
16th Edition,  Method 503D, with two exceptions; extend the extraction time from 4 to 16 hours
and substitute methylene chloride for Freon based on RCC experience that methylene chloride
is a superior solvent for oils and greases.

Determine the total water content by Karl Rsher titration. (The water content is generally not
determined by a simple oven test since most oils are relatively volatile.. The oven test is used if
the ratio of water to oil present in the sample is relatively large.)
                                         65

-------
Determine the paniculate solids content by rinsing a known quantity of feed through a Whatman
GF/C fitter under vacuum with acetone followed by methylene chloride, drying and weighing the
residue.

Normalize the oil + water + solids concentration to  100% if it is 90-110%.  Repeat the assays to
determine the source of the error(s) if the oil + water + solids concentration is less than 90% or
greater than 110%.
3.0       TEST PROCEDURE

3.1       Feed Characterization

In conjuction with the performance of a B.E.S.T. laboratory-simulation, a data set must to be
developed for the sample. The data required includes compositional (oil/water/solids), metals
analysis, as well as other compounds of interest.

     a.    Determine the total oil per Standard Methods for the Examination of Water and
          Wastewater. 16th Edition, Method 503D, with two exceptions; extend the extraction
          time from 4 to 16 hours and substitute methylene chloride for Freon based on RCC
          experience that methylene chloride is a superior solvent for oils and greases.

     b.    Determine the total water content by Karl Fisher titration. (The water content is not
          typically determined by a simple oven test since most oils are relatively volatile. An
          oven test will be used to cross-check the results of the Karl Fisher titration if the ratio
          of water to oil present in the sample is relatively large.)

     c.    Determine the paniculate solids content by rinsing a known quantity of feed through
          a Whatman GF/C filter under vacuum with acetone followed by methylene chloride,
          drying and weighing the residue.

     d.    Normalize the oil + water * solids concentration to 100% if it is 90-110%.  Repeat the
          assays to determine the source of the error(s) if the oil + water + solids concentration
          is less than 90% or greater than 110%.

3.1.1      Feed Sample Metals Composition

Dry approximately 10-20 gms  of the feed in the oven at 105°C. Use a ceramic crucible for this
task.  Dry for at least 6 hours.  Record the initial sample weight.

After drying at 105°C put the sample into the  muffle furnace and ash at 550°C for at least 3
hours. Once finished, let the sample cool to room temperature. Record the final sample (ash)
weight.

Digest the ash with nitric acid as a prelude to  analysis of metal concentrations by ICP.
                                       66

-------
3.1.2.     TCLP Extraction Metals (SVf 846 Method 1311)

Perform preliminary extraction on a representative portion of each sample to determine the
appropriate extraction fluid. After determining the correct extraction fluid, perform 18 hour
TCLP extraction on a portion of representative sample.

After filtering the leachate, perform metals analysis using SW846 Method 6010.

3.1.2      PCB Analysis

PCB analysis, if present, will be conducted in accordance with SW846 method 8080. Soxhlet
extraction will be used (method 3540). The Aroclor type found in each sample should be
recorded.

3.2 Preliminary Testing

TEA can be ionized at some pH conditions. In the ionic form TEA is non-volatile and will not be
recovered from the process product phase fractions.  To determine the proper pH  control
requirements for each sample, a pH adjustment test is conducted.

TEA has the potential to react with some rare types of samples. To determine if this will pose a
problem during the study a compatibility study will also be performed.

3.2.1      Adjustment Characteristics

Measure the sample pH using pH paper if the sample is mostly solid,  and pH probe if it is
mostly liquid.  If neither seem to be providing a good measurement (i.e., the pH paper color
cannot be read or the pH probe readings keep drifting), then add 100 mis of distilled water to
about 5 gms of sample, stir well, and remeasure pH.

Obtain 5-10 gms of sample and adjust pH with 5% NaOH solution.  If the pH could not be
measured as discussed above, be sure to add distilled water prior to the pH adjustment
measurements.

After the  sample pH has been adjusted to 11 or above, record the amount of caustic spent.
Cover the sample with parafilm and leave mixing overnight.  On the next day, check the pH
again and adjust with caustic soda if needed. Record all amounts of caustic spent.

3.2.2      TEA/Feed Compatibility Test

Mix about 5 gms of each type of sample with 50 mis  of cold TEA. Make observations about the
ability of the TEA to dissolve the sample.

Use a thermometer to measure the amount of temperature change when the sample is mixed
with TEA.

Note any effervescence that takes place such as the formation of hydrogen. Note any unusual
reactivity. If there is any effervescence or unusual reactivity, notify the Lab Director before
proceeding.
                                      67

-------
3.3  B.E.S.T. Bench-scale Treatablllty Test Workup

3.3.1      Pre-Treatment and First Wash

Adjust feed pH as determined previously.

Chill sample to below 40° F. Chill TEA to below 38°F.

Mix^^grams of the sample with^^tres of chilled TEA _   	
for ^Pninutes with the pneumatic mixer in the chiller bath (at < 40°F).
At the end of the first mixing stage, remove the particulates with the appropriate method
determined during the optimization testing.

Decant and retain supernatant/centrate/filtrate. Keep chilled at < 40°F until heating and
decantation can be performed.

Place the centrifuge cake or filter cake, if collected, back into the extraction beaker for
additional wash stages.

3.3.2     Second Wash

Mix recovered first extraction stage solids wfflf^iitres of fresh TEA. Fqrthe second and

TEA to transfer solids from the centrifuge botties (if used) into the mixing container. Keep the
mixture heated while mixing is in progress.

Mix forJFminutes with the pneumatic mixer.

Perform particle removal from extraction mixture again as performed for the first stage
extraction. If desired, collect a portion of the solids for later analysis.

3.3.3      Third Wash and Solids Drying

Repeat section 3.32 as a third wash.

Dry the solids at 220°F (105°C) in the forced draft oven. Mix occasionally to facilitate TEA
volatilization.
 After the initial drying, add a portion of de-ionized water adequate to thoroughly wet the solids,
 then redry in order to further reduce residual TEA concentrations. To insure that the TEA
 residual in the dried solids will be low. treat the solids with caustic soda (applied with the de-
 ionized water) if the pH of these solids is less than 10.  Add sufficient caustic soda to raise the
 pH to approximately 10.5. Determine the required amount of caustic soda on a small portion of
 the solids.

 NOTE:    Upon occasion, additional washes may be required to achieve required treatment
           standards.
                                         68

-------
 3.3.4      Decantatlon

 Use the large separator/ funnel that is in the temperature controlled bath.  Keep the water bath
 at 140°F.

 Heat supematant/centrate/filtrate from first wash (chilled to this point at 40°F) up to 140°F with
 continuous mixing on a hot plate. Pour into separatory funnel.

 Allow 30 minutes quiescent residence time in the separatory funnel. Decant only if the layers
 appear to have stopped separating. If separation is still proceeding, wait until a better
 separation is achieved, and then decant

 Record observations of the speed of separation and measure 'rag' volume. Rag layer should
 ultimately be centrifuged and the resulting TEA/oil and water layers should be added to the
 appropriate decanted fraction pnor to the stripping  operation.

 Record the weights of the TEA/oil, rag, and water fractions from the decantation.

 3.3.5      Distillations

 Water layer

 Record initial water pH  (should be >11).

Steam strip the water at 110°C (in the rotovap) until no TEA odor is detected in collected
distilling drops. Record  the initial water volume.

At this point, check water pH. If the pH is >10. distill for 15 minutes more and then terminate
distillation.

 If water pH is <10, adjust to >12 and continue distillation until no TEA is detected in the
collected distillate. Check the latter every 15 minutes, then recneck water pH.  If pH is still
below 10, then repeat this section (d.) until water pH is >10.

Oil/TEA laver

 Remove the bulk of TEA without steam at 110°C (in rotovap). Record initial and final oilfTEA
volume.

Steam strip the oil at 110°C. Perform this operation by adding a known quantity of water
(typically 5 mis) and then measuring the volume of distillate recovered. When all TEA is
removed, the recovery  of the distillate should be equal to the amount of water added.

Perform oil polishing by distilling without steam until the oil temperature reaches 120°C to
facilitate  excess water removal. Record the final product oil weight

Be sure to collect some of the final distillate for measurement of the extent of any volatile
organic carryover.

                                            69

-------
4.0       ANALYTICAL REQUIREMENTS

4.1       Feed Analysts

          o    Composition - % (wt) oil, water and solids by:

                     105°C total solid
                     Oil by Soxhlet extraction (by dichloromethane)
                     Solids by difference

          o    Total metals

                     Dryat105°C
                     Ash at 550°C
                     Nitric acid digest for ICP heavy metals determination

          o    Physical properties:

                     pH
                     Specific Gravity

          o    pH Adjustment:

                     Amount of caustic added to pH adjust the feed to pH of 11

          o    PCB's by Method 8080, if present

       	e	TCLP Mctola Dtfraetien Analysis

4.2       Product Solids Analysis

          o     Residual Oil and Grease by Soxhlet extraction

          o     Aqua regia digest for ICP total trace metals

          o     TCLP Extraction Metals Analysis

          o     PCB's. if present

          o     Residual TEA

4.3       Decant Water Analysis

          o    PCB's, if present and if sample size permits

           o    Oil & Grease by freon extraction (IR if volume is limited)

           o    Total metals

 4.4        Oil Analysis

           o    PCB's. if present


                                      70

-------
5.0       SAFETY CONSIDERATIONS

Because of the unknown reactivity of these samples, extreme care should be taken not to allow
a dangerous situation to develop. Results of the compatfciitty study should be reviewed with
the Lab Director prior to initiation of the full scale bench-scale extraction. The bulk of these
simulations will be done in the laboratory hood to decrease TEA emissions. Prior to testing, the
feed will be subjected to TEA-compatibility tests to verify that it does not react violently with
TEA.  Other safety precautions involving personnel conducting B.ES.T. laboratory work that will
be followed are described in RCC's Laboratory Safety Manual.  If you have any other safety
related concerns, contact the Lab Director.


NOTE:  Be sure to check that the ventilation system is working properly and wear
       appropriate protective equipment when handling these samples.
                                      71

-------
                     APPENDIX C
        QUALITY ASSURANCE PROJECT PLAN
                       FOR
    GLNPO - ASSESSMENT AND REMEDIATION OF
     CONTAMINATED SEDIMENT TECHNOLOGY
            DEMONSTRATION SUPPORT
                    Revision
                 February 15, 1991
                   Submitted to:

         U.S. Environmental Protection Agency
         Great Lakes National Program Office
                 230 S. Dearborn
               Chicago, Illinois 60604
                   Submitted by:

     Science Applications International Corporation
          635 West Seventh Street, Suite 403
              Cincinnati, Ohio  45203
EPA Contract No. 68-C8-0061, Work Assignment No. 2-18
          SAJC Project No. 1-832-03-207-50
                      72

-------
                                                           GLNPO - QAPjP
                                                           Section No.:    Q_
                                                           Revision No.:   2	
                                                           Date:        Feh 15. 1991
                                                           Page:        1 of 2
                                TABLE OF CONTENTS
 SECTION


 1.0    INTRODUCTION	

 2.0    PROJECT DESCRIPTION   	

 3.0    QUALITY ASSURANCE OBJECTIVES  . . .

 4.0    SAMPLE TRANSFER AND PREPARATION
            PROCEDURES 	

 5.0   ANALYTICAL PROCEDURES AND
            CALIBRATION  	

 6.0   DATA REDUCTION, VALIDATION AND
            REPORTING 	

 7.0   INTERNAL QUALITY CONTROL CHECKS

 8.0   PERFORMANCE SYSTEMS AUDITS  	

 9.0   CALCULATION OF DATA QUALITY
            IMPUCATORS 	

10.0   CORRECTIVE ACTION   	

11.0   QA/QC REPORTS TO MANAGEMENT  . . .

APPENDIX A - TECHNOLOGY SUMMARIES	
2

12

2
          REVISION    DATE
1

2
             1

             2

             2


             1

             1

             1

             1
1/9/91

2/15/91

2/15/91


1/9/91


2/15/91


1/9/91

2/15/91

2/15/91


1/9/91

1/9/91

1/9/91

1/9/91
                                          73

-------
                       QUALITY ASSURANCE PROJECT PLAN APPROVALS
   QA  Project Plan Title:    GLNPO   Assessment  and Remediation of Contaminated
                            Sediment  Technology Demonstration Suooort
  Prepared by:     Science  Applications  International  Corporation (SAIC)
        QA Project  Category:   II
                                          Revision Dace: January 9. 1990
SAIC's WorK Assignment Manager (print)
         Clyde J.  Dial
 SAlC's QA Manager (print)
 	Steve  Yaks:en	
 3U;PO '-eric Group Chair (print)
         Brian Schumacne"
 Af.CS  QA  Officer  (print)
        3ene Easterly
£?A. EMSL-LV, NRD  QA  Officer  (print,
        Salon Chr-'szensen
I7A Tecnnicai Project Manager (print
        Z'ave Cowo:" 1
AX.—>  .-rogras Manager .print;
                                                Signature
                                                Signature
                                                Signature
                                                Signature
                                                                               /Date
                                                                              Date
 Date
                                                                              Data
-ate
                                                                             -ate
                                           74

-------
 DISTRIBUTION LIST:
 Gene Easterly

 Brian Schumacher

 Tony Kizlauskas

 Thomas Wagner

 Clyde Dial

 Steve Garbaciak

 Dennis Timberlake

 Steve Yaksich

David Cowgill

Gary Baker

Vic Engleman
 U.S. EPA, EMSL (Las Vegas)

 LOCKHEED (Las Vegas)

 SAIC (Chicago)

 SAIC (Cincinnati)

 SAIC (Cincinnati)

 U.S. COE (Chicago)

 U.S. EPA, RREL (Cincinnati)

 U.S. COE (Buffalo)

 U.S. EPA, GLNPO (Chicago)

SAIC (Cincinnati)

SAIC (San Diego)
                                                     GLNPO - QAPjP
                                                     Section Nou   £
                                                     Revision No.:  2
                                                     Date:
                                                     Page:
                                               Feh IS 1991
                                               2of 2	
                                     75

-------
                                                          GLNPO - QAPjP
                                                          Section Nou   J_
                                                          Revision No.:  1
                                                          Date:        Jan. 9. 1991
                                                          Page:        1 of 2
  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 (GLWQA) with Canada.  Under Section
 118(c)(3)  of the  CWA, GLNPO  is  responsible  for  undertaking  a 5-year study and
 demonstration program for contaminated sediments.  Five areas are specified for priority
 consideration in locating and conducting demonstration projects:  Saginaw Bay, Michigan;
 Sheboygan  Harbor, Wisconsin; Grand  Calumet River, Indiana  (aka: Indiana Harbor);
 Ashtabula River, Ohio; and Buffalo River, New York.  In response, GLNPO has initiated
 an Assessment and Remediation of Contaminated Sediments (ARCS) Program.  The ARCS
 Program will be carried out through a management structure including a Management
 Advisory Committee consisting of public interest, Federal and State agency representatives,
 an Activities Integration Committee which is made up of the chairpersons of the technical
 work groups, and technical work groups.

      In  order to obtain the  broadest possible information base on  which to  make
 decisions, the ARCS Program will conduct bench-scale and pilot-scale demonstrations and
 utilize opportunities afforded by contaminated sediment  remedial activities by others, such
 as the Corps of Engineers and the Superfund program, to  evaluate the effectiveness of those
 activities.  These bench-scale and pilot-scale tests will be developed and conducted under
 the guidance of the Engineering/Technology (ET) Work Group for ARCS.

      SAJC has been contracted to supply technical support to the ET Work Group.  The
effort consists of conducting bench-scale treatability studies  on designated sediments  to
evaluate the removal of specific organic contaminants.
                                        76

-------
                                                       GLNPO - QAPjP
                                                       Section No.:   J_
                                                       Revision No.:  1_
                                                       Date:         Jan. 9. 199L
                                                       Page:         2 of 2
      Sediments have been obtained by GLNPO from various sites and represent the type
of material that would be obtained for onsite treatment. The primary contaminants of these
sediments are polychlorinated biphenyls (PCBs) and polynuclear aromatic hydrocarbons
(PAHs). Analyses to date show PCB concentrations are less than 50 ppm. These sediments
have been homogenized and packaged in smaller containers by EPA.
                                    77

-------
                                                          GLNPO - QAPjP
                                                          Section Nou   2-
                                                          Revision No.:  2-
                                                          Date:        Feb. 15. 1991
                                                          Page:        1 of 12
  2.0    PROJECT DESCRIPTION
  2.1    Background
        SAIC and its subcontractors will conduct seven (7) bench-scale (several liters) tests
  on wet contaminated sediments using four treatment technologies.

       The seven treatability tests (as currently planned) will utilize sediments from 4 sites
  (Saginaw  River, Buffalo River, Indiana Harbor Canal,  and Ashtabula  River).   Five
 sediments have been collected from these sites by GLNPO.   These samples have been
 homogenized by the U.S. EPA and  are being  stored under refrigeration  in 5 gallon
 containers by EPA in Duluth, MN.

       These five sediments are currently being analyzed in the U.S. EPA, Environmental
 Research Laboratory in Duluth. The Duluth Laboratory is analyzing the sediments for total
 organic carbon/total inorganic carbon  (TOC/TIC), particle size,  density of dry material,
 total sulfur, acid volatile sulfide, oil and  grease (O & G), total PCBs, PAHs (10), and metals
 including mercury.  Table 2-1  is a summary of the data received to date.

       A portion (small vial) of each residual of each treatability test may be retained and
 sent to the GLNPO office for "show" purposes.  If available, sub-regulated quantities of the
 solid and oil residuals from each test treatability study may also be retained and shipped to
 EPA for possible further treatment  studies.

       The following is a list of technologies and the proposed number of sediment samples
 to be  tested by each technology:
       a.     B.E.S.T.™  Extraction Process on three  samples (Buffalo River, Indiana
             Harbor, Saginaw TRP 6)
       b.     Low Temperature Stripping (RETEC) on one sample (Ashtabula River)
       c.     Wet Air Oxidation (Zimpro Passavant) on one sample (Indiana Harbor)
       d.     Low Temperature Stripping (Soil Tech) on two samples (Buffalo River and
             Indiana Harbor)
Summaries of these technologies are included in Appendix  A.

                                       78

-------
                                  TABLE 2-la.  Preliminary Analytical Results on ARCS Sediments
Description
Saginaw 221
Saginaw TRP6
Ashlabula River
/
Indiana Harbor
Buffalo River
Concentration (Mg/kgm)(a) Concentration (%)(*)
Total Total
PCB PAH
06 1.2
6.0 3.1
C C
0.2 96
0.4 5.6
c» Cd Ni Fe(%) Cr Zn Pb TOC OAG Moisture (b)
33 0.9 76 1.4 NO 240 30 1.4 O.I 40.3
81 4.7 110 0.9 200 200 47 1.2 0.3 3|.|
55 3.0 96 3.7 550 240 48 2.6 1.7 52.9
320 9.4 150 16 540 3300 780 2t 5.8 61.0
85 1.9 57 3.9 NO 200 94 2.0 0.5 41.5
(D
       (a) Concentration In ppm and dry weight basis unless otherwise Indicated.
       (b) As received basis.
                                      TABLE 2-lb.  Preliminary Particle Size Distribution (%)
Description
Buffalo River
Particle Size (a)
>50u 50-20 u 20-5 u 5-2 u 2-0.2 u 0.2-0.08 u <0.08u
198 12.1 29.0 11.8 24.3 2.4 0.6
Median
Diameter, u
9.3
       (a) u mlcarons

-------
                                                            GLNPO - QAPjP
                                                            Section No J   2.
                                                            Revision No.:  2
                                                            Date:         Feb. IS. 1991
                                                            Page:         3 of 12
  22.    Testing Program for Chemical Characterization
        SAIC shall be primarily responsible for the physical and chemical characterization
  of both  the sediment samples prior to testing and the residuals created during the tests.
  Analyses conducted by the vendors or subcontractors will not be depended on, but such data
  shall be  reported whenever available.

        Two different sets of chemical analyses will be conducted during the performance of
  the treatability tests: optimization test analyses and performance evaluation analysis.  The
  Phase I optimization test analyses will be conducted by the subcontractor or vendor during
  the series of initial technology tests. The Phase II performance evaluation analyses will be
  conducted by SAIC (or its analytical subcontractor) on the raw sediment sample prior to the
  treatability test run at optimum  conditions and  on  the  end products produced  by  that
 particular test.  These tests are described further in this section.

       In order to assure objectivity and consistency of data obtained from multiple vendors
 running different technology tests, SAIC shall conduct analyses as described in Table 2-2 for
 characterization of the sediments and the end products of the treatability tests at optimum
 conditions (Phase II).

       The analyses described for the solid fraction in Table 2-2 shall be performed by
 SAIC's analytical subcontractor once on a subsample taken from each sample  sent to each
 vendor or subcontractor for treatability tests (Phase II). This subsample will be taken at the
 same time that the sample for the  Phase  II treatablility study is taken by the vendor. This
 data will serve as the measure of the raw sediment quality for comparison to analyses of
 treated end products from each technology test that may be conducted on sediments from
 a particular area of concern.

       Each bench-scale technology test may actually involve the performance of multiple
laboratory simulations.  During the initial tests (Phase I), any analyses performed  by the
                                        80

-------
                                                           GLNPO - QAPjP
                                                           Section No.:   2.
                                                           Revision NOJ  2-
                                                           Date:        fph, 15 1991
                                                           Paige:        4 of 12
 vendor or subcontractor shall be reported, as  available.  For the tests run at optimum
 conditions (Phase II), SAIC shall conduct the full suite of analyses, as detailed in Table 2-2,
 on the end products if sufficient quantities are produced by the technology. Quotes solicited
 for each technology specified that a minimum 300 grams dry basis of treated solid had to
 be produced for SAICs analyses. Table 2-3 shows the apportionment of the 300 grams for
 the solid analyses.  The quantity of water is depended on the sediments and the individual
 technologies. To do all the analyses listed  in Table 2-2, and associated QC, approximately
 10 liters of water are required. Table 2-4 listed specified sample volumes for each  analysis,
 and gives a priority to each analysis.  It is possible that only the PCB and PAH analysis and
 associated QC will be performed on the water samples.  If any oil residue is  produced, it
 will be analyzed by dilution with  appropriate sample cleanup steps for PCBs and  PAHs.

       The data generated by SAICs analyses of the untreated sediment and the treated end
 products from the test at optimum conditions will be primarily relied upon to determine
 treatment efficiencies. Vendor- or subcontractor-generated data will not be relied upon but
 shall be reported when available.

 23    Required Permits

       Because of the small quantities of sediments required for the bench-scale treatabiliry
 tests, SAIC anticipates that no formal permits will be required to conduct these  tests. If this
 is  not the case  and permits  (such as TSCA, RD&D or RCRA permits) are required, the
subcontractor  will notify  SAIC and  the TPM  will  be notified to obtain approval  for
acquisition of the permit(s).

       All  unused  sediment samples requested by SAIC for the treatabiliry  test  and all
testing  residuals, except those requested  by  the  TPM  for "show" purposes and those
requested by the TPM for possible further testing, will be properly disposed of per federal
and state regulations.
                                        81

-------
                                                          GLNPO - QAPjP
                                                          Section No.:   2.
                                                          Revision No.:
                                                          Date:         Feb. 15. 1991
                                                          Page:         5 of 12

                                     TABLE 2-2

           Parameters and Detection Limits for Analysis of ARCS Technologies
Parameter
TOC/TIC
Total Solids4
Volatile Solids4
Oil & Grease4
Total Cyanide
Total Phosphorus
Arsenic4
Barium4
Cadmium4
Chromium4
Copper4
Iron (total)4
Lead4
Manganese4
Mercury4
Nickel4
Selenium4
Silver4
Zinc4
PCBs (total & Aroclors)4
PAHs (16)4-5
PH
BOD5
Total Suspended Solids4
Conductivity
so*
300
1000
1000
10
0.5
50
0.1
0.2
0.4
0.7
0.6
0.7
5
02
0.1
2
02
0.7
0.2
0.02
0.2
full range



Water
1000

1000
1000
10
10
1
2
4
7
6
7
50
2
0.01
20
1
7
2
0.07
2
full range
1000
1000
full range
Qift



















0.1
0.1




NOTES:
1      Detection limits for solids are ppm (mg/kgjgijLweigiit). The D.L.'s for metals should
       be obtainable by ICP except for As, Se, and Hg. If GFAA is used, the D.L-'s will be
       2 mg/kgm except Hg, Cd, and Ag which will be 0.1 mg/kgm.
-      Detection limits for water are ppb (ug/1). The D.LJs for metals should be obtainable
       by ICP except for As,  Se, Hg.  If GFAA is used D.L.'s will  be  1 ug/L except Hg
       which will be 0.01 ug/L.
3      Detection limits for oil are ppm (mg/1).
4       Parameters tentatively  identified for QC analyses.
5       Polynuclear aromatic hydrocarbons to be analyzed are the 16 compounds listed in
       Table 5-2.
                                        82

-------
                                                          GLNPO - QAPjP
                                                          Section No.:    2.
                                                          Revision No.:
                                                          Date:
                                                          Page:
2
Feh 15  1991
6 of 12	
                                       TABLE 2-3
                           Solid Sample Quantities for Analyses

Parameter
TOC/TIC
Total + Volatile Solids
Oil & Grease
Total Cyanide
Total Phosphorous
Metals (except Hg)
Hg
PCBs + PAHs
PH
Subtotals
Reserve
TOTAL
Initial
Sample (g)
15
5
20
10
5
5
1
30
20
111
—
—

OC (tf

10
40
—
_
15
3
90(60)3
—
158(128)
—
—

Total (g)
15
15
60
10
5
20
4
90
20
269(239)
31(61)
300

QC Approach
None1
Triplicate/Control
Triplicate/Control
None2
None2
MS/Triplicate
MS/Triplicate
(3)
None4



1  For sample set II that does not have such a limited quantity of solid, The QC described in
  footnote 3 will be implemented.

2  For sample set n, MS/triplicate QC will be implemented.

3  Quality control for untreated solids is Triplicate and spike and for treated solids matrix spike
  and matrix spike duplicate.

4  For sample set n, Triplicate/Control sample QC will be implemented. The control sample
  may be an EPA QC check sample, an NBS - SRM, a standard laboratory  reference solution,
  or other  certified reference material.
                                       83

-------
                                                        GLNPO - QAPjP
                                                        Section No.:    2	
                                                        Revision No.:   2	.	
                                                        Date:         Fftt?  15 199!
                                                        Page:         7 ^f 12
                                    TABLE 2-4

           Sample Volumes Required and Priority Ranking for Water Analyses

Parameter
TOC/TIC
Volatile Solids
Oil & Grease
Total Cyanide
Total Phosphorus
Arsenic
Barium
Cadmium
Chromium
Copper
Iron (total)
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Zinc
PCBs (total & Aroclors)
PAHs (16)
PH
BOD
Total Suspended Solids
Conductivity

Priority
1
5
6
1
7
4
2
2
2
2
2
2
2
3
">
4
->
9
1
1
7
7
5
7
Analysis
Volume, ml
25
d
1000
500
50
100
100
b
b
b
b
b
b
100
b
c
b
b
1,000
a
25
1,000
200
100
QC
Volume, ml
_
d
2000
—
—
300
300
b
b
b
b
b
b
300
b
c
b
b
2.000
a
—
—
400

QC
Approach
None (e)
Triplicate/Control
Triplicate/Control
None (f)
None (f)
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/Triplicate
MS/MSD
MS/MSD
None (f)
None (f)
Triplicate/Control
None (f)
Note:

a) same aliquot as PCBs
b) same aliquot as Barium
c) same aliquot as Arsenic
d) same aliquot as Total Suspended Solids
e) see footnote 2, Table 2-3
f)  see footnote 4, Table 2-3
                                       84

-------
                                                           GLNPO - QAPjP
                                                           Section NOJ   2_
                                                           Revision Nou  2-
                                                           Date:         Feb. 15. 1991
                                                           Page:         8 of 12
  2.4    Purpose of Phase I Experimental Design
        The purpose of the Phase I technology experimental design is for each subcontractor
  to establish a range of variables best suited for feasibly implementing their technology on
  a full-scale basis (Phase II). SAIC will send a quantity (specified by the vendor)  of each
  sediment to the vendor to accomplish this. All data generated by the vendor during Phase
  I will be supplied to SAIC for inclusion in the report for that technology. This information
 will include the operating conditions/parameters, the input/output data for the contaminants
 of interest to show the range of effectiveness associated with various operating conditions,
 and the quantities of the input material and the various residuals resulting from the test.
 The optimum set of conditions to be used for Phase II will be reported to SAIC  along with
 appropriate revisions to the Phase I experimental design to make it applicable to Phase II.

 2.5    Purpose of Phase n Treatahilitv Test
       SAIC will send another  container of sediment(s) to  the  vendor (quantity to be
 specified by the vendor). This container will not be opened until a representative of SAIC
 arrives for the scheduled treatability test(s).  Other observers from U.S. EPA, COE and/or
 the GLNPO may also be present during the Phase II treatability test(s).

      The new sample will be homogenized and a sample equivalent to a minimum of 300
gm of dry solids will  be set aside for characterization analyses (Table 2-2) by SAIC. SAIC
will observe the treatability tests and obtain samples of process residuals for analyses  (Table
2-2).   The bench-scale test(s)  must produce enough  solid residual  for all  vendor
requirements and a quantity equivalent to 300 gm of dry solids for SAIC analyses.  SAIC
can utilize up  to  10 liters of water for analysis  and 25  ml of the oil residual. The actual
quantities of water and oil that will be produced are dependent on the initial sediment and
the technology. All technologies except wet air oxidation are expected to produce an oil
residual. Also, if additional solid and/or oil residue is available,  EPA  may ask for these
materials to be sent to them for storage for possible future evaluation.
                                          85

-------
                                                           GLNPO - QAPjP
                                                           Section NOJ   2-
                                                           Revision No.:  2.
                                                           Dale:        Fffhi 15 1991
                                                           Page:        9 of 12
        All data generated by the vendor during Phase n is to be supplied  to SAIC for
 inclusion in  the report for that technology.  The vendor must stipulate in their work plan,
 prior to conducting the test(s), the process locations to be sampled, the frequency and the
 information  being obtained.

        All other residuals from both phases of the treatability study, including any untreated
 sediment, will be properly disposed of by the vendor.

       SAIC shall oversee the treatability test assessment(s) by vendors or subcontractors,
 including all  QA/QC aspects, monitoring and analysis. SAIC shall ensure compliance with
 the specific experimental design during the  tests conducted by vendors or  subcontractors.
 SAIC will make specific notes regarding the equipment being used, any pretreatment of the
 sediment(s), the operation of the equipment, and any post treatment of the residuals. SAIC
 personnel will pack the untreated sediment sample and the end product samples from the
 Phase II test for each technology in an appropriate  fashion for shipment from  the vendor
 or subcontractor to the laboratory SAIC is using for the analysis. Proper chain-of-custody
 procedures will be developed in the QAPjP and strictly followed by SAIC personnel.

      SAIC  plans  to  take photos of the equipment while at  the vendor's location for
 inclusion in the report.

      SAIC shall perform limited  interpretation of technology  test  results, specifically the
 development  of material  and  energy balances.  No test of air or fugitive emissions will be
 done. For material balances, estimates of the mass distribution of the analytes of interest
(Table 2-2) among the residuals will  be made.  The term energy balance is interpreted to
mean an estimation by the vendor of the energy input into the process at a pilot- or full-
scale.
                                        86

-------
                                                           GLNPO - QAPjP
                                                           Section NOJ   jj_
                                                           Revision No.:  2
                                                           Date:        Fah, is 1991
                                                           Page:        10 nf 12
        SAIC shall  collect any  information available  from the vendor or subcontractor
  concerning the actual or estimated costs of constructing and operating full-scale versions of
  the technology tested.

        The  purpose of this project is to  test five  technologies for removing organic
  contaminants (PCBs and PAHs) from sediments typical of locations around the Great Lakes.
  GLNPO is specifying the technologies and the sediment(s) to be treated by each technology.
  This study is only one pan of a  much larger program, and it is not necessarily intended to
  evaluate the complete treatment of these sediments. Other aspects or treatment options are
  being evaluated by a number of agencies, contractors, etc.

       Therefore, this study is based on the following assumptions:
       •     The percent removal of the PCBs and PAHs from the solid residual is the
             most important object of this study.
       •     The untreated sediments and solid residuals are the most important matrices.
       •     If water and oil residuals are generated by a technology, the existence of an
             appropriate treatment  or disposal option for these residuals  is assumed.
             PAHs  and PCBs will be determined in these residuals as  a cross  check of
             their fate in treating the solids.

       Based on the intents of this study, the critical measurements are PAHs, PCBs, metals,
total solids, volatile solids, and oil and grease in the untreated and treated solids.

2.6    Organization and Responsibilities
       A  project  organization  and   authority  chart  is  shown  in  Figure   2-1.   The
Environmental  Monitoring Systems Laboratory (EMSL) is cooperating with GLNPO and
SAIC on this evaluation.  Mr. Thomas Wagner is the SAIC Work Assignment Manager and
is responsible for the technical and budgeting aspects of this work assignment.  Mr. Clyde
Dial is QA Manager and is responsible for QA oversight on this work assignment.
                                       87

-------
                                            us i-PAi

                                             I'lUHliCI MANAIililt
II S P.PA QA MANAOnil




    
-------
                                                          GLNPO - QAPjP
                                                          Section NOJ   i.
                                                          Revision NoJ  2
                                                          Date:         Feb. 15. 1991
                                                          Page:         12 of 12
2.7    Schedule
      The Phase I experimental designs are scheduled for mid to late February 1990, and
the Phase El Treatability Tests are scheduled for March and April 1991.
                                       89

-------
                                                          GLNPO - QAPjP
                                                          Section NOJ    i.
                                                          Revision NOJ   2-
                                                          Date:         Feb. 15. 1991
                                                          Page:         1 of 2	
  3.0    QUALITY ASSURANCE OBJECTIVES
  3.1    Precision. Accuracy. Completeness, and Method Detection Limits
        Objectives for accuracy, precision, method detection limits, and completeness for the
  critical measurements of solids are listed in Table 3-1. Accuracy (as percent recovery) will
  be determined from matrix spike recovery for PAHs, PCBs and metals, and from laboratory
  control samples (certified reference material- CRM) for the remaining analyses.  Precision
  (as relative standard deviation) will be determined from the results of triplicate analyses for
  PAHs, PCBs, solids (total, volatile and/or suspended), oil and grease, and metals.  Matrix
 spike and matrix spike duplicate analyses will be used for treated solids for PCBs and PAHs.
 The completeness will be determined from the number of data meeting the criteria in Table
 3-1 divided by the number of samples that  undergo performance  evaluation analyses.

 3.2    Representativeness and Comparability
       Representativeness and Comparability are qualitative parameters.   The sediment
 samples have already been collected and have been reported to be representative of the
 areas to be remediated. The data obtained in this program will be comparable because all
 the methods are taken from  a standard EPA reference manual and all the analyses will be
 conducted at the same laboratory.  Reporting units for each analysis are specified in Section
 6 of this document and are consistent with standard reporting units in this  program.

 3.3    Method Detection Limits
       The target detection limits (TDLs) were specified by GLNPO (Table  2-2).  Based on
the analytical methods appropriate for the analyses and the amount of samples specified in
the methods, the detection limits listed in Table 3-1 should be achievable.   Generally the
instrument  detection limits are'defined as 3 times the standard deviation of 15 blanks or
standards with a concentration within a factor of 10 of the IDL.
                                       90

-------
                TABLE 3-1.  Quality Assurance Objectives Tor Critical Measurements
                              (Sediments and Treated Solids)
Parameter
Total Solids
Volatile Solids
Oil & Grease
Arsenic
Barium
Cadmium
Chromium
Copper
Iron (total)
lead
Manganese
Mercury
Nickel
Selenium
Silver
Zinc
PCBs (total
ft Aroclort (e)
PAIls (Table 5-2)
Method (a)
160.3
160.4
9071
3050/7060
3050/6010
3050/6010
3050/6010
3050/6010
3050/6010
3050/6010
3050/6010
7471
3050/6010
3050/7740
3050/6010
3050/6010
3540 or
3550/8080
3540 or 3550/
8270 or 8 100
Accuracy (b)
(as % recovery)
80-120
80-120
80-120
85-115
85-115
85-115
85-115
85-115
85-115
85-115
85-115
85-115
85-115
85-115
85-115
85-115
70-130
70-130
Precision (c)
%
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Method
Detection Limit (d)
(mg/kgm)
1000
1000
10
O.I
0.2
0.4
0.7
0.6
0.7
5
0.2
O.I
2
0.2
0.7
0.2
0.02
0.2
Completeness
%
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
00
90
90
(a) References are to "Methods for Chemical Analysis of Water and Wastes", EPA/600/4-79/020 or "Test Methods for
   Evaluating Solid Waste", SW-846, 3rd. Ed.
(b) Determined from MS or MS/MSD analyses for metals, PAHs, and PCBs; others determined from
   laboratory control samples.
(c) Determined as relative percent standard deviation of triplicate analyses, except PAIls and PCBs
   in treated solids where MS/MSD will be used.
(d) See Footnotes I and 2 of Table 2-2
(e) Detection limits based on extraction of 30 gram samples.
                                                                                                                                  p

-------
                                                           GLNPO - QAPjP
                                                           Section No_-   4_
                                                           Revision NoJ  1
                                                           Date:        Jan. 9. 199L
                                                           Page:        1 of 4
  4.0    SAMPLE TRANSFER AND PREPARATION PROCEDURES
        As described in Section 2, SAIC will receive a number of 5 gallon containers of
  previously homogenized sediments from the U. S. EPA in Duluth, Minnesota. The number
  of containers of each sediment is dependent on the final determination by GLNPO of which
  sediments will be tested by the various technologies.  Only if smaller portions of sediments
  are requested by the vendors will these containers be opened by SAIC.  If smaller portions
  are required, SAIC will resuspend the solids and water within an individual container by
  rolling, tumbling, and stirring of the contents.  The final stirring will  be  in the original
  containers using a metal stirrer as would be used to mix a 5 gallon container of paint. The
 metal stirrer is appropriate because metals are not the primary constituents of concern in
 these treatabiliry tests.

       The Chain of Custody Record shown  in Figure 4-1 will be completed for each cooler
 shipped to the subcontractor or vendor that will conduct the optimization and performance
 evaluation tests.   The samples  obtained from the vendor for analysis will  be labeled  as
 shown in Figure 4-2. The labels will document the sample  I.D., time and  date of collection,
 and the location from where the sample was taken. The amount/type of preservative that
 was added will also be recorded.

       SAIC personnel will pack and ship  the untreated  sediment  and the end product
 samples (residuals) from the optimum conditions  test for each technology. The amount of
 preservative will be recorded.  Samples will be labeled (see Figure  4-2) and shipped by
 overnight delivery service to the  laboratory  in coolers containing ice. If "blue ice" is used
 in the coolers,  samples will  be initially cooled with regular  ice prior to being packed in the
coolers with blue ice. The Chain  of Custody Record (Figure 4-1) will be completed for each
cooler shipped to the laboratory.
                                        92

-------
                                                          GLNPO - QAPjP
                                                          Section NOJ    ±
                                                          Revision No-   J
                                                          Date:
                                                          Page:
                                                                       Jan. 9  1991
                                                                       2 of 4
        Solid, sediment and oil samples require no preservative other than cooling to 4° C.
 The appropriate types of containers (solid and liquids), holding times, and preservatives for
 water samples are listed in Table 4-1.
            TABLE 4-1.  Sample Containers, Preservation and Holding Times
Parameter
TOC
Solids (Total,
Volatile &
Suspended
Oil and Grease
Total Cyanide
Container
P,G
P,G
G
P,G
Preservation of Water Samples
Cool 4° C, H2SO4 to pH < 2
Cool 4° C
Cool 4° C, H2SO4 to pH < 2
Cool 4° C, NaOH to pH > 12
Holding Time
28 days
7 days
28 days
14 davs
Total Phosphorous  P,G

                   P,G
Metals
(except Cr VI)

Cr(VI)
                   P,G
PAHs & PCBs
BOD5
PH
Conductivity
G teflon
lined cap
P,G
P.G
P,G
0.6g Ascorbic acid

Cool 4° C, H2SO4 to pH  < 2

HNO3 to pH < 2


Cool 4° C

Cool 4° C, store in dark


Cool 4° C



Cool 4° C
28 days

6 months except Hg
(Hg 28 days)

24 hours

Extract within 7 days
Analyze within 40 days

48 hours

Performed immediately

28 davs
                                       93

-------
GLNPO - QAPjP
gffpnn NOJ    ^
Revisioa No:
Date:
Page:
                                    hn, 9 1991 _
                                        *     -
|
w






•Q o
O
ft
£ .
S £
o
M
3
J
O
1
5
j
0
75
C


If
• e r
i ° <2
ctenct Appllci
tomvNonal Co
n Empfcir** Ownr
ly


Z
1 .
jj
zo c





B
g


*




i






1 n
' M
g a


>U-



















s
Phone Nunit

i!
UO2>-<_




















|
«
riONS. COMMENTS. 1
L INSTRUCTIONS 1
OBSERVA
SPECIA
.ZUC 01













!
1 t
s
I -

t
, i
Sampler (Slgna
l*««wyN» |

























































































































































































































































































































1



i

I
1
5

g




£•
1

j


£•

f
1
1





|
*
i














i
i
<
• j


s
f
2
I;
i
i\







\
1





i
!
I
B
?




:

.










I




B
i
S
8J
III

\
\

\l
*\
fl







J
I





i

1
j
|jj
i
;
\l
\\
|







r
1





i
1
fi
it Q
I
; *
M i
: !
f i
-5 «
] {


I




£
c

I
c


fc
1
1
i
1
u!
in



i
i
8
i
|







!
1





i








i










I




M
S
3
i
j
I
<_
;
h
U


* £
l!







j
I





i




I
1
|
s
[








i





!
l
c
o
1
s


f
j
!



n
I
i
•
u
i
T
I
\
\
a
i

1
i
!
i
R
ol/oni In
I
;
j
(ft
94

-------
                                                 GLNPO - QAPjP
                                                 Section No.:   4	—
                                                 Revision NOJ   I      —_
                                                 Date:         J«n 9 199L
                                                 Page:         4 of 4	
                 535 W. 7th Street. Suite 403, Cincinnati, OH 45203
Sample No.:
Sample Location/Date/Time:
Project Location/No.:
Analysis:
Collection Method:	Purge Volume:

Preservative:	
Comments:

                                            Collector's Initials
               Figure 4-2.  Example Sample Label
                              95

-------
                                                           GLNPO - QAFjP
                                                           Section NOJ   £.
                                                           Revision NOJ  2
                                                           Date:        Feb. 15. 199L
                                                           Page:        1 of 3
  5.0    ANALYTICAL PROCEDURES AND CALIBRATION
        Analytical procedures for all critical measurements are referenced in Table 3-1. The
  non-critical measurements are for  any residual water  and oil  remaining  after  the
  performance evaluation tests and some additional analyses on the solid samples.  The EPA
  procedures are specified in Table 5-1.

       The required calibration for all analyses are specified in the methods and will be
 followed. All instruments will be calibrated as specified in the methods prior to performing
 any analysis of the samples.  Internal QC checks, including initial calibration and continuing
 calibration checks, for the critical measurements are listed in Table 7-1.

       Table 5-2 contains the minimum list of the sixteen PAHs that must be determined
 by either analytical method.  Additional  compounds may be included, but none  of these
 sixteen may be deleted from the target list.

       The laboratory is responsible for  maintaining a preventive maintenance  program
consistent with manufacturers recommendations for  all  instruments required   for this
program.  In addition, they are responsible for having a sufficient supply of routine spare
pans necessary for  the  operation of the  analytical equipment  in order to complete the
analysis in a timely fashion.
                                       96

-------
                                                          GLNPO - QAPjP
                                                          Section No.:    £
                                                          Revision No.:   2.
                                                          Date:
                                                          Paige:
                                          Feb. 15. 1991
                                          2 of 3
                                     TABLE 5-1
             Analytical Methods for Critical and Non-critical Measurements
                                              Methods?
 Parameter
Solid
Water
Oil
TOC
Total Solids
Volatile Solids
Oil and Grease
Total Cyanide
Total Phosphorous
Arsenic
Mercury
Selenium
Other Metals
PCBs

PAHs

pH
BOD
Total Suspended Solids
Conductivity
9060
1603
160.4
9071
9010
3652
3050/7060
7471
3050/7740
3050/6010
3540 or
3550/8080
3540 or 3550/
8270 or 8100b
9045
NA
NA
NA
9060
NA
160.4
413.1
9010
3652
7060
7470
7740
3010/6010 (7760 Ag)
3510 or
3520/8080
3510 or 3520/
8270 or 8100b
9040
405.1
160.2
9050
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA

3580/8080

3580/8270
NA
NA
NA
NA
(a) References are to "Methods for Chemical Analysis of Water and Wastes", EPA/600/4-
79/020 or 'Test Methods for Evaluating Solid Waste",  SW-846, 3rd. Ed.

(b) Where options for methods are given,-Either is acceptable if the detection limits given
in Table 2-2 can be achieved.

NA - Not  analyzed
                                        97

-------
                                                     GLNPO - QAPjP
                                                     Section No.:   JL
                                                     Revision No.:  2
                                                     Date:        Feh 15. 1991_
                                                     Page:        3 of 3	
                                 TABLE 5-2

                                List of PAHs'
          Acenaphthene                  Chrysene
          Acenaphthylene                 Dibenzo(a,h)anthracene
          Anthracene                     Fluoranthene
          Benzo(a)anthracene             Fluorene
          Benzo(a)pyrene                 Inden(lA3-cd)pyrene
          Benzo(b)fluoranthene            Naphthalene
          Benzo(k)fluoranthene            Phenanthrene
          Benzo(ghi)perylene              Pyrene
PAH analyses must determine these 16 compounds at a minimum.
                                   98

-------
                                                          GLNPO - QAPjP
                                                          Section NOJ    £_
                                                          Revision NOJ   1
                                                          Date:         Jan. 9. 199L
                                                          Page:         1 of 1
 6.0   DATA REDUCTION, VALIDATION AND REPORTING
       Data will be reduced by the procedures specified in the methods and reported by the
 laboratory in the units also specified in the methods.  The work assignment manager or his
 designer will review the results and compare the QC results with those listed in Table 3-1.
 Any discrepancies will be discussed with the QA Manager.

       All data will be reviewed to ensure that  the correct codes and units have  been
 included.  All organic and inorganic data for solids will be reported as mg/kgm except TOC,
 oil & grease (O&G), moisture and iron that will be reported as percent and pH that will
 be reported in standard pH units. All metals  and organics in water samples will be reported
 as ug/1. TOC, solids (suspended and volatile), O&G, cyanide, phosphorus, and BOD will
 be reported as mg/1.  Conductivity will be reported as umhos/cm and pH as standard pH
 units.  After reduction, data will be placed in tables  or arrays and reviewed  again for
anomalous values. Any inconsistencies discovered will be resolved immediately, if possible,
by seeking clarification from the sample collection personnel responsible for data collection,
and/or the analytical laboratory.

      Data Tables in the report will be delivered in hard copy and on discs.  The discs will
be either in Lotus files or WordPerfect 5.1 files.
                                        99

-------
                                                       GLNPO - QAPjP
                                                       Section NOJ    2
                                                       Revision No_-   2.
                                                       Date:
                                                       Page:
                                      peb. 15. 199L
                                      1 of 7     __
7.0   INTERNAL QUALITY CONTROL CHECKS

      The internal QC checks appropriate for the measurement methods to be utilized for
this project are summarized in Table 7-1. These items are taken from the methods and the
QC program outlined in Section 3 of this QAPjP.
      For the GLNPO program, the following QC measures and limits are employed:
      on-going calibration
      checks
     method blanks
     matrix spikes
     replicates
 beginning, middle, and end of sample set for metals, pH,
 TOC/TIC, total cyanide, and total P
 mid-calibration range standard
 ± 10% limit unless otherwise stated
 ± 0.1 pH unit for pH
 ± 10 umhos/cm for conductivity at 25° C

 beginning, every 12, and end of sample set for PCBs and
 PAHs
 mid calibration range standard
 ± 10% limit

 one  per sample set for PCBs and PAHs
 < MDL limit unless otherwise stated
 beginning, middle and end  for metals, TOC/TIC, total
 P, total  cyanide, and pH
 beginning,  middle  and  end  for  conductivity  with
 acceptance limits of  < 1 umho/cm

 one per sample set
 1 to  1.5  times the estimated concentration of sample
 ±  15% limit for metals; ± 30% for PCBs and PAHs

triplicate analyses
RSD s 20%  unless otherwise stated
one per  sample set
 ± 0.1 pH unit for pH
± 2 umhos/cm for conductivity
                                  100

-------
                                                           GLNPO - QAPjP
                                                           Section No-    J
                                                           Revision No.:   2
                                                          Date:         Feb. IS 1991
                                                          Page:         2 of 7
        QC sample            -   - minimum of one per sample set
        (CRM)                -   ± 20% of known CRM
                              -   ± 0.1 pH unit for pH
                              -   ± 1 umhos/cm for conductivity
        surrogate spikes        -   added to each sample
        (PCBs and PAHs only) -   ± 30% recovery

 The surrogate for PCB analysis is tetrachlorometaxylene and the internal standard is 1,2,3-
 trichlorobenzene.
       Table 7-2 shows an analytical matrix that will be completed for each technology
 tested. For example, consider the case of a bench scale treatability test of (1 kilogram)
 Indiana harbor sediment by low temperature stripping. Based on the data presented in
 Table 2-la and assuming complete separation and recovery of oil, water, and solid, a 1
 kilogram sample of untreated sediment will produce 58 grams of oil, 610 ml of water, and
 332 grams of dry treated solids.  For the purpose of this program,  this sample set consists
 of 1 untreated solid, 1 treated solid, and the water and oil generated by the process. Table
 7-3 is a completed analytical matrix for this test. Table  7-3 is based on Tables 2-2 and 2-4
 and the QC approach described in this QA plan. The analysis of the water sample in this
 example is severely limited by the relatively small amount of sample obtained.

       Table 7-4 is a matrix summarizing the anticipated samples  to be analyzed for this
 project. The sets for each technology (see section 2.1) are:
       I      B.E.S.T.
       D     ReTec
       III    Wet  Air Oxidation
       IV    Soil Tech

       The Soil Tech process will process treated soils at two distinct points. Therefore,
four treated solids  are produced from the two untreated sediments.
                                      101

-------
                                            TABLE 7-1.  Internal QC Checks for Measurements
o
ro
Parameter
Solids
(Total &
Volatile
Oil & Grease
Metals
MelaU
PCBs (b)
PAHs
Method (a)
160.3
160.4
9071
6010
7000
series
8080
8270 or
8100
Initial
Calibration
Balance
(Yearly)
See Above
2 points
4 points
5 points
5 points
Calibration
Checks
Balance
Each Day
See Above
Every 10th
Sample
Every 10th
Sample
Every 10th
Sample
Every 12
Hours
Method
Blank
Yes
Yes
Yes
Ye«
Yes
Yes
MS/MSD
NA
NA
MS only
MS only
Yes (treated)
MS only (untreated)
Yes (treated)
MS only (untreated)
Triplicate
Sample
Analysis
Yes
Yes
Yes
Yes
NA (treated)
Yes (untreated)
NA (treated)
Yes (untreated)
QC
Sample
Yes
Yes
Yes
Yes
Yes
Yes
Surrogate
Spikes
NA
NA
NA
NA
Yes
Yes
          (a) References are to "Methods for Chemical Analysis of Water and Wastes*, EPA/600/4-79/020
            or "Test Methods for Evaluating Solid Waste*, SW-846, 3rd. Ed.
          (b) Second column confirmation of positive results is required.
          NA - Not Applicable
                                                                                                                                          R
                                                                                                                                           8
                                                                                                                                                 O

-------
                            TABLE 7-1.  Internal QC Checks for Measurements (continued)
Parameter
PH
Conductivity
Cyanide
Phosphorous
TOC/TIC
Method (a)
9045/9040
9050
9010
365.2
9060
Initial
Calibration
1 points
1 point
7 points
9 points
3 points
Calibration
Chech
Every 10th
Sample
Every 15th
Sample
Every 15th
Sample
Every 15th
Sample
Every 15th
Sample
Method
Blank
NA
NA
Yes
Yes
Yes
MS/MSD
NA
NA
NA
NA
NA
Triplicate
Sample
Analysis
NA
NA
NA
NA
NA
QC
Sample
Yes
Yes
Yes
Yes
Yes
Surrogate
Spikes
NA
NA
NA
NA
NA
(a) References are to "Methods for Chemical Analysis of Water and Wastes", EPA/600/4-79/020
  or "Test Methods for Evaluating Solid Waste". SW-846, 3rd. Ed.
NA - Not Applicable

S» -4

-------
                                                        TABI.K7-2.  Analytical Matrix
o
-P-
                            QC Sample
                               *nd
                            lelhod Bltnk
                                                                                                                                                SO!
                                                                                                                                                     '

-------
                                                       TABLE 7-3. Example
o
en
                         QC Simple


                           tnd


                        Method Blank
                                                                                                                               8

-------
             TABLE 7-4.  Analytical and QC Sample Matrix for GLNPO Trealability Studies (numbers of samples)
SAMPLE SET
SET I
Untreated S.
Treated S.
Water
Oil
SETW
Untreated S.
Treated S.
Water
Oil
SETH
Untreated S.
Treated S.
Water
Oil
SETHI
Untreated S.
Treated S.
Water
TOTALS
Solida
Water
Oil
roc/r/c
S<») QC(b)
3
3
2
4

1
I
1
1
I

16
-
1

3
2
3
-

5
3
TOTAL
SOLIDS
S QC

3


2
4


I
1

1
1

16


2


3
2


3
2

3
2

20

VOL
SOLIDS
S QC

3


2
4


1
1
1

I

16


2


3
2


3
2
3

3
2

20
3

OAO
S QC

3


2
4


1
1

1
1

16


2


3
2


3
2
3

3
2

20
3

TOTAL
YANIDB
S QC

3


2
4


1
1
1

1

16
1


-


-


3
3
3

-

6
3

TOTAL
PHOS
S QC

3


2
4


1
1
1

1

16
1


-


-


3
3
3

.

6
3

METALS
S QC

3


2
4


1
1
1

1
1

16
1


3


3
3


3
3
3

3
3

24


PCBt
S QC

3
3
3
2
4
2
2
I
1
1
I
1
1
1
16
6

2
1
3
3
2
I
3
3
2
2
3
3
2
2
20
9
PAH
S QC

3
3
3
2
4
2
2
I
1
1
I
1
1
1
16
6

2
1
3
3
2
1
3
3
2
2
a
3
2
2
20
9
PH
S QC

3
2
4
I
1
1
1
16
1
-
-
3
2
3
-
S
3
BOD
S QC
-
-
1
-
1
-
-
3
-
3
TSS
S QC
-
-
1
-
1
-
-
3
-
3
COND
S QC
-
-
1
-
1
-
-
3
-
3
(a) Number of original lamplea.
(b) Number of quality control (ampler A *3* represent* two additional repllcatei (triplicate determination) and • ipilce or control
  lample analysia resulting in an additional three QC analyses.  A *2* represents matrix spike/malrix spike duplicate analysis
  scheme resulting in an additional two QC inilysei.  A * I * indicates a blank spike or other control sample analysis resulting
  in one additional QC analysis.
(c) Treated and untreated solidf does not apply, and only one control sample per set will be analysed.

-------
                                                         GLNPO - QAPjP
                                                         Section NOJ   £_
                                                         Revision NOJ   JJ_
                                                         Dare:        Feb. 15. 1991
                                                         Page:        1 of 1	
 8.0    PERFORMANCE AND SYSTEM AUDITS
       The laboratory will perform internal reviews by the QA officer or a designee. These
 reviews should include, as a minimum, periodic checks on the analysts to assess whether they
 are aware  of and are  implementing the QA requirements specified in the ARCS QA
 program.

       The laboratory will be prepared to participate in a systems audit to be conducted by
 the SAIC QA Officer or his designee and/or ARCS QA Officer.

      The vendors of the various technologies have all been advised that a number of
representatives from SAIC, GLNPO, and other organizations will be present during
Phase II of the treatability studies.  Thus the ARCS QA officer can be present during
Phase II of any or all of the treatability studies.
                                      107

-------
                                                          GLNPO - QAPJP
                                                          Section NOJ    2_
                                                          Revision No-   1
                                                          Date:         Jan. 9. 1991
                                                          Page:         1 of 3
  9.0   CALCULATION OF DATA QUALITY INDICATORS

  9.1    Accuracy

        Accuracy for PAHs, PCB and metals will be determined as the percent recovery of

  matrix spike samples.  The percent recovery  is calculated according  to  the  following
  equation:


              % R  =  100% xC' ~£?	
                                Q

 where
       %R   =  percent recovery
       C,     =  measured concentration in spiked sample aliquot
       C0     =  measured concentration in unspiked sample aliquot
       C,     =  actual concentration for spike added


       Accuracy for the other critical measurements will be determined  from laboratory

 control samples according to the  equation:
             %R  =  100%  :2L
                             Q

where
       %R   =  percent recovery
       Cm    =  measured concentration of standard reference material
       C,     =  actual concentration for standard reference material
9.2   Precision

      Precision will be determined from the difference of percent recovery values of MS

and MSDs for PAHs and PCBs or triplicate laboratory analyses. The following equations

will be used for all parameters:
                                       108

-------
                                                   GLNPO - QAPjP
                                                   Section No.:   5	.
                                                   Revision NOJ  1
                                                   Date:       /an 9. 1991
                                                   Page:       2 of 3
 When 2 values are available:
                       [Q -  CJ x 100%
                          [Q  + CJ/2
where
      RPD  = Relative percent difference
      Cj    = The larger of two observed values
      C    = The smaller of the two observed values
When more than 2 values are available:
                          N
                          I
                        i = l
                                               N
2  _
               x  i
      N  i
                                   N -  1
where
      S   =  standard  deviation
      X.,   = individual measurement result
      N   =  number of measurements
      Relative standard deviation may  also be reported.
will  be calculated as  follows:

                             RSD = 100 5
                           If  so,  11
                                  109

-------
                                                         GLNPO - QAPjP
                                                         Section NOJ   JL
                                                         Revision No_-  1
                                                         Date:        Jan. 9. 199L
                                                         Page:        3 of 3
 where
       RSD  = relative standard deviation, expressed in percent
       .5   = standard deviation
       X   =  arithmetic mean of replicate measurement.
9.3    Completeness

       Completeness will be calculated as the percent of valid data points obtained from the

total number of samples obtained.

       % Completeness  =  VDP  x 100
                          TOP
where

      VDP  =  number of valid data points
      TOP  =  total number of samples obtained.
                                    110

-------
                                                             GLNPO - QAPJP
                                                             Section No.:    10
                                                             Revision No_-   1	
                                                             Date:         Jan. 9. 1991
                                                             Page:         1 of 2
  10.0   CORRECTIVE ACTION

        Corrective actions will be initiated whenever quality control limits (e.g., calibration
  acceptance criteria) or QA objectives (e.g., precision, as determined by analysis of duplicate
  matrix spike samples) for a particular type of critical measurement are not being met.
  Corrective actions may result from any of the following functions:
        •     Data Review
        •     Performance evaluation audits
        •     Technical systems audits
        •     Interlaboratory/interfield comparison studies

       AJ1 corrective action procedures  consist of six elements:
       •      Recognition that a Quality Problem exists
       •      Identification of the  cause of the problem
       •      Determination of the appropriate corrective action
       •      Implementation  of the corrective action
       •      Verification of the corrective action
       •      Documentation of the corrective action

       For these treatability studies after initial recognition of a data quality problem, the
data calculation will be checked first. If  an error is found, the data will be recalculated and
no further action will be taken.  If no calculation error is found, further investigation will
be conducted.  Depending on the  cause and the availability of the appropriate samples.
reanalysis or flagging of the original data will be utilized.
                                       111

-------
                                                        GLNPO - QAPjP
                                                        Section No_-   10
                                                        Revision No.:   1	
                                                        Date:        Jan. 9. 199L
                                                        Page:        2 of 2
      All corrective action initiations, resolutions, etc. will be implemented immediately and
will be reported in Sections One and Two (Difficulties Encountered and Corrective Actions
Taken, respectively) in the  existing monthly progress reporting mechanisms established
between SAIC, EPA-RREL, GLNPO, AND THE ARCS QA officer and in the QA section
of the final report. The QA Manager will determine if a correction action has resolved the
QC problem.
                                    112

-------
                                                           GLNPO - QAPjP
                                                           Section NOJ   11
                                                           Revision
                                                           Date:        Jan. 9. 1991
                                                           Page:        1 of 1
  11.0   QA/QC REPORTS TO MANAGEMENT
        This section describes the periodic reporting mechanism, reporting frequencies, and
  the final project report which will be used to keep project management personnel informed
  of sampling and analytical progress, critical measurement systems performance, identified
 problem conditions, corrective actions, and up-to-date results of QA/QC assessments.  As
 a minimum, the reports will include, when  applicable:

       •     Changes to the QA Project Plan, if any.
       •     Limitations or constraints on  the applicability of the data, if any.
       •     The  status of QA/QC programs, accomplishments and corrective actions.
       •     Assessment  of data quality in terms of precision,  accuracy, completeness,
             method detection limit, representativeness, and comparability.
       •     The final report shall include a separate QA section that summarizes the data
             quality indicators  that document the QA/QC activities that lend support to
             the credibility of the data and the validity of  the conclusions.

       For convenience, any QA/QC reporting will be incorporated into the already well-
established monthly progress reporting system between SAIC and EPA-RREL for all TESC
Work Assignments. In addition, copies of monthly reports will be sent to the ARCS QA
officer.  Any information pertaining to  the above-listed categories will be reported under
Sections  One through  Three (Difficulties Encountered, Corrective Actions Taken, and
Current Activities,  respectively)  in the monthly reports.
                                    113

-------
                            GLNPO - QAPjP
                            Section NOJ   Appendix A
                            Revision No.:  1	
                            Date:        Jan. 9. 1991
                            Page:        1 of 3
       APPENDIX A

TECHNOLOGY SUMMARIES
       114

-------
                                                            GLNPO - QAPjP
                                                            Sf-ftio" No-   Appendix A
                                                            Revision NOJ  1	
                                                            Dais:        Jan. 9. 1991
                                                            Page::        2 of 3	
  B.E.S.T.™ Process Description
  The B.E.S.T.™ process is a patented solvent extraction technology utilizing triethylamine
  as the solvent  Triethylamine is  an aliphatic amine that is produced by reacting ethyl
  alcohol and ammonia.  The key to success of the B.E.S.T.™ process is triethylamine's
  property of inverse miscibility. At temperatures  below 65°F, trietihylamine  is completely
  soluble with water.  Above this temperature, triethylamine and water are  only partially
 miscible.  The property of inverse miscibility can be utilized since cold triethylamine can
 simultaneously solvate oil  and water.

 The B.E.S.T.™ process produces a single phase extraction solution which is a homogeneous
 mixture of triethylamine and the water and oil (containing the organic contaminants, such
 as PCBs, PNAs, and VOCs) present in the feed material.  In cases where the extraction
 efficiencies of other solvent extraction systems are hindered by emulsions, which have the
 effect  of  partially  occluding the  solute  (oil  containing  the organic contaminants),
 triethylamine can achieve intimate contact at nearly ambient temperatures and pressures.
 This allows the B.E.S.T.™  process to handle feed mixtures with high water content without
 penalty in extraction efficiency.  This process  is  expected to yield solid, water, and oil
 residuals.

 Low Temperature Stripping
 Low-temperature stripping (LTS) is a means to physically separate volatile and semivolatile
 contaminants from soil, sediments, sludges, and filter cakes.  For wastes containing up to
 10% organics or less, LTS  can be used alone for site remediation.

 LTS is  applicable to organic wastes and  generally is not used for treating inorganics and
 metals.  The technology heats contaminated media to temperatures between 200-1000°F,
 driving  off water and volatile contaminants.  Offgases may be burned in an afterburner,
condensed to reduce the volume to be disposed, or captured by carbon adsorption beds.
For these treatability studies, only processes that capture the contaminants driven off will

                                        115

-------
                                                            GLNPO - QAPjP
                                                            $f»/t'f>f No-   /Appendix A
                                                            Revision No.:  1	
                                                           Date:         Jan. 9. 1991.
                                                           Page:         rlnf3
 be considered.  The process (for these treatability studies) is expected to yield solid, water,
 and oil residuals.

 Wet Air Oxidation
 Wet air oxidation is a process that accomplishes an aqueous phase oxidation of organic or
 inorganic substances at elevated temperatures and pressures. The usual temperature range
 varies from approximately 350 to 600°F (175 to 320°C). System pressures of 300 psig to well
 over 300 psig may be required. However, testing has been done at temperatures exceeding
 the critical point for water to limit the amount of evaporation of water, depending on the
 desired reaction temperature.  Compressed air or pure oxygen is the source of oxygen that
 serves as the oxidizing agent in the wet air oxidation process. This process is expected to
yield only solid and water residuals.
                                        116

-------
                                APPENDIX 0

                     B.E.S.T.®  TREATABIUTY STUDY

                          ANALYTICAL METHODS


                 Analytical Methods Used In B.E.S.T.® Workuos
  Feed Analysis

  Moisture:

  Oil & Grease:


  Ash Content:

  Metals Analysis:
 Polynudear Aromatic
      Hydrocarbons*:
                    Karl Fisher titration or Gravimetric @ 105°C, 16 hrs

                    Both EPA/SW846 Method 9071 and Methylene Chloride Soxhiet
                    gravimetric (16 hr. extraction)

                    Gravimetric @ 550°C for 16 hrs.

                    Digestion:   EPA SW846/3050 or, alternatively, ash digestion at
                               550°C,  followed by heating with nitric acid
                    Analysis: EPASW846/6010
                           EPA SW846/8100 (Methylene Chloride Extraction)

Benzene, Toluene, Xylene*:      EPA SW846/8020

pCB*:                        EPA SW846/8080, Method 3540 extraction (soxhlet
                             extraction with 1:1 acetonerhexane for 16 hours)
Product Water

Total Solids:

Total Dissolved Solids:

Total Organic Carbon:

Oil & Grease:


Triethylamme:

Metals:
                             Product Analysis



                        Standard methods 209A

                        Standard methods 209B

                        EPA 600/415.2

                        EPA 600/413.1 or 413.2 depending on level of sample and
                        quantity of sample available

                        See the attached RCC method

                        Digestion:   SW846/3005
                        Analysis:    SW846/6010
   • If present
                                   117

-------
         GC METHODS FOR TRIETHYLAMINE (TEA) ANALYSIS
             IN AQUEOUS SOLUTIONS, SOLIDS AND OILS
 I    Summary of Methods

     Triethyiamine (TEA) can be determined using Gas Chromatography with flame
     ionization detection. Aqueous solutions can be injected directly onto a packed
     column after pH adjustment and filtration. Solids are extracted into water without pH
     adjustment and then are analyzed using the same column and parameters as
     aqueous solutions. Oils are dissolved in a solvent (typically methylene chloride) and
     analyzed, using a megabore HP-1 Methyl Siiicone column.


II   TEA In Aqueous Solutions

    A.   Equipment and Operating Parameters

         1.    Gas Chromatograph: Hewlett Packard 5890A with 3392A Integrator

         2.    Column: 4% Carbowax-20M, 0.8% KOH, 60/80 Carbopack B

         3.    Injector Temp.: 200°C

         4.    Detector:  Flame  Ionization Detector (FID), set at 300°C

         5.    Oven Temperature and Time:
                   Initial Temp: 90°C,  Initial Time: 0 minutes
                   Final Temp: 170°C, Final Time: 30 minutes
                   Rate:  5°/min.

         6.    Column Flow:  ~30ml/min.

         7.    TEA Peak Retention Time: Approximately 8 minutes

   B.    Procedure

         1.    Standardization

             a)   Inject I microliter of 73 mg/1 TEA. This standard is prepared by
                  serial dilution from pure TEA (successive 1:100 dilutions from pure
                  TEA, which  is 730,000 mg/l TEA).

             b)   Use the peak area at approximately 8 min. retention time to
                  quantitate TEA.  Repetitive standards injections should agree to
                  within 10%.
                                  118

-------
 Product Oil

 Triethylamine:

 Viscosity:

 Water:

 Suspended Solids:

 Metals:
 See the attached RCC method

 Brookfield

 Karl Fisher titration

 Filtration/Gravimetric (Whatman GF/C)

 Product oil is diluted 1:10 with Xylene and filtered through
 GF/C filter, then analyzed with organometallic standards on
 ICP or, alternatively, ash digestion at 550°C, followed by
 heating with nitric acid
 Product Solids

 Residual Triethylamine:

 Oil & Grease:


 Metals:


TCLP:

Polynudear Aromatic
     Hydrocarbons':

Benzene, Toluene, Xylene*:

PCS*:
See the attached RCC method

Both SW846/9071 and Methylene Chloride Soxhlet
gravimetric (16 hr extractions)

Digestion:   1 gm sample refluxed with 15 mis Aqua Regia
Analysis:    SW846/6010

SW846/1311 followed by 3010 & 6010
   EPA SW846/8100 (Methylene Chloride Extraction)

   EPA SW846/8020

   EPA SW846/8080, Method 3540 extraction (soxhiet
   extraction with 1:1 acetone:hexane for 16 hours)
   * If present
                                    119

-------
          2.    Sample Preparation and Analysis
                a)    If necessary, dilute the sample in deionized. distilled water until the
                     TEA concentration is at or below 73 ppm and record the dilution
                     factor.
                b)    Inject 1 microliter sample.
                c)    Inject a standard at least once every 10 samples and at the end of
                     an analytical sequence.
          3.     Quantitation
                Quantitate the TEA using direct comparison of peak area
                 Peak Area of Standard   = (Peak Area of Sample)
                 	mg/l Standard            (mg/l of Sample)     x (dllutlon

III   TEA In Solids
    A.    Equipment and Operating Parameters
          The same equipment and parameters are used as in aqueous solutions.
          Additional equipment includes the following:
          1.    Shaker Bath (example: Forma Scientific model 2564).
          2.    50 ml Erienmeyer flasks.
          3.    Cover with Parafilm.
    B.    Procedure
          1.    Standardization: asinll.B.1.
          2.     Sample preparation and analysis.
               a.    Weigh out 3-5 g solids into an Erienmeyer flask. Record exact
                    weight.
               b.    Add 25 ml distilled water.
               c.    Cover with 1  layer of parafilm.
                                     120

-------
                d.    In the shaker bath, shake vigorously at ambient temperature for 1
                      hour.

                e.    Let the mixture stand quiescent allowing the solids to settle.

                f.     Continue as in II.B.2.

           3.    Quantitation

                Quantitation is the same as that for aqueous solutions, correcting for the
                extraction of the solids into the water.
                     Peak Area of Standard   _ Peak Area of Extract
                          TEA in Standard          jig/g TEA of Extract
               ng/gm TEA in Solids  = (jig/ml TEA in Extract) x (mis of Extraction Water)
                                                               (g of Solids)


IV  TEA In Oil

    A.    Equipment and Operating Parameters

          1.    Gas Chromatograph:  Hewlett Packard 5890 with 3392A Integrator

          2.    Column:  Megabore 15m x .53mm J&W BD1 Methyl Silicone

          3.    Injector Temp:  200°C

          4.    Detector: Flame lonization Detector (FID), set at 300°C

          5.    Oven Temperature and Time:
                    Initial Temp: 35°C. Initial Time:  12 minutes
                    Final Temp:  250°C, Final Time: 20 minutes
                    Rate: 25°C/min.

          6.    Column Flow: 2 ml/mm.

          7.    Makeup Gas: 20 ml/mm.

         8.    TEA Peak Retention Time: Approximately 7-8 min.
                                    121

-------
B.    Procedure

      1. Standardization

           a)    Inject 1 microliter 73 mg/l TEA (dissolved in GC grade methyiene
                chloride).  This standard is prepared by serial dilution from pure
                TEA (successive 1:100 dilutions into methyiene chloride from pure
                TEA, which is 730,000 mg/l TEA).

           b)    Use the peak area at approximately 7-8 minutes retention time to
                quantitate TEA. Repetitive standard injections should agree to
                within 10%.

     2.     Sample Preparation and Analysis

           a)    Dissolve the oil in methyiene chloride such that the TEA
                concentration is at or below 73 ppm.

           b)    Inject 1 microliter sample.

           c)    Inject a standard at least once every 10 samples.

     3.    Quantitation

          Quantitate the TEA using direct comparison of peak area.

               Peak Area of Standard     m    Peak Area of Sample
                  mg/l of Standard                mg/l of Sample
                                122

-------
ro
co
           SAIC-GLNPO (CF #361)
           CONVENTIONALS IN UNTREATED SEDIMENT
                                                                 B.E.S.T.
                                                                                                            REVISED
                                                                                                            2/18/92
% Total OH & Grease
MSLCode
MDL
Sponsor ID

% Moisture
0.01%
PH
NA
Volatile Solids
0.001%
(mg/kg)
0.1
TCC
% weight
0.007%
Total Cyanide
(mg/kg)
0.001
Total Phosphorus
(mg P/kg)
0.001
           361-5. Rep 1
           361-5, Rep 2
           361-5. Rep 3
           361-6
           361-7
           Method Blank
               S-US-RCC, Rep 1
               S-US-RCC. Rep 2
               S-US-RCC, Rep 3
               BUS-RCC
               I-US-RCC
STANDARD REFERENCE MATERIAL

MESS-1 SRM
In-house Concensus Value '

MATRIX SPIKE RESULTS

Amount Spiked
Sample
Sample + Spike
Amount Recovered
Percent Recovery

 REPLICATE ANALYSES
            361-5, Rep 1
            361-5, Rep 2
            361-5. Rep 3
                S-US-RCC, Rep 1
                S-US-RCC. Rep 2
                S-US-RCC. Rep 3
                     RSD%
23.98%
NA
NA
41.06%
57.04%
NA
7.30
NA
NA
7.29
7.35
NA
2.24%
2.03%
2.01%
4.03%
14.2%
0%
1485
1318
1256
2415
32165
0.1
0.83%
NA
NA
1.98%
17.03%
0.0117.
4.05
NA
NA
2.05
25.17
0.001 U
23.98%
NA
NA
41.96%
57.04%
NA
NA
NA
NA
NA
NA
NA
NA
23.98%
NA
NA
NA
7.30
NA
NA
7.29
7.35
NA
NA
NA
NA
NA
NA
NA
NA
7.30
NA
NA
NA
                                                                  NA
                                                                  NA
NA
NA
2.45
 2.3
NA
NA
23.98%
NA
NA
NA
7.30
NA
NA
NA
2.24%
2.03%
2.01%
6%
1485
1318
1256
9%
0.63%
NA
NA
NA
4.05
NA
NA
NA
            NA - Not analyzed
            U = Below detection limit
               - TOC value lor MESS determined based on past In-house analyses.  Not a statistical determination.
            NOTE:  All Conventional results are reported on a dry weight basis.
                                          742
                                           NA
                                           NA
                                          735
                                          2625
                                          0.089
NA
NA

NA
NA
NA
NA
NA

NA
NA
NA
NA
NA

NA
NA
NA
NA
NA
361-5

4305
1485
6383
4698
114%

NA
NA
NA
NA
NA
361-5

267.10
4.05
251.90
247.85
93%
361-6

2419
735
3088
2353
97%
§
m




                                           742
                                            NA
                                            NA
                                            NA

-------
SAIC-GLNPO (CF (361)
METALS IN UNTREATED SEDIMENT
(Concentrations

MSI Code
MX
361-5. Rep 1
361 5, Rep 2
381-5. Rep 3
361-6
361-7
In ug/g dry weight)

Sponsor ID

S-US-RCC. Rep 1
S US RCC. Rep 2
S-US HCC. Rep 3
BUS RCC
1 US RCC

Ag
»•
0007
0.87
0 85
0.80
0 31
4 84

As
MF
2 5
2.5 U
265
25 U
127
22.8

Ba
**
43
322
322
321
413
317

Cd
»A
0006
4.38
4.08
3.96
2.10
8.56

Cr
**
33
03
112
117
100
2270
B.E.S.T.

Cu
WF
5.6
55.8
67.3
63.4
70.2
188

%Fe
WT
0.26
0.780
0.765
0.616
4.200
18.770

HB
CVAA
0.0003
0.162
0.178
0.160
O.SS1
1.528

Mi
XV
66
162
ISO
173
667
3230

Nl
XRF
75
58. 1
55.2
61.5
43.1
12.0

Pb
XT
6 2
43.0
42.5
500
101.0
582 0
REVISED
2/21/02

Se
A*
022
0.22 U
0.22 U
0.22 U
0.74
0.22 U

Zn
Hf
It
125.7
132.6
162.1
180.3
2380
Method Blank
                                    0016
                                              NA
                                                        NA   0006
                                                                        NA
                                                                                NA
NA    0.00066
                                                                                                             NA
                                                                                                                     NA
                                     NA    0.22 U
NA
STANDARD REFERENCE MATERIAL
1646 SRM
certified
value
MATRIX SPIKE RESULTS
Amount Spiked
361-5*
361-5 + Spike
Amount Recovered
Percent Recovery
REPLICATE ANALYSES
361-5. Rep 1 S US RCC. Rep 1
361-5, Rep 2 S US RCC. Rep 2
381-5. Rep 3 S-US-RCC. Rep 3
RSD%
0.126
rC
rC

2
084
333
240
125%

087
085
080
4%
11.3
11 6
11 3

NS
NS
NS
NS
NS

1.86
265
2 12
18%
 J  - Values delected below MDL.
 U  -  Below detection  limit
 NA - Not analyzed/applicable
    - Mean  ol triplicated sample
 NS - Not spiked
 NOTE:  AH Metals results  are 'blank corrected "
406
NC
N3
NS
NS
NS
NS
NS
322
322
321
0%
0.42
0.36
±007
2
4 15
643
228
114%
4.30
408
3.06
5%
78
76
±3
N3
NS
NS
NS
NS
03
112
117
12%
106
18
±3
NS
NS
NS
NS
NS
55.8
57.3
63.4
7%
336
3.35
±01
NS
NS
NS
NS
NS
0.780
0.785
0.816
3%
0.066
0.063
±0012
1.072
0.166
2.207
2.041
103%
0.162
0.178
0.160
e%
339
376
±20
NS
NS
NS
N3
NS
162
150
173
4%
31.1
32
13
NS
NS
NS
NS
NS
68.1
55.2
61.5
B%
28.1
28 2
±18
NS
NS
NS
NS
NS
43.0
42.5
50.0
10%
0.75
NC
tc
2.72
022 U
3.10
3.10
114%
0.22 U
0.22 U
022U
NA
126.3
138
±6
NS
NS
NS
NS
NS
125.7
132.6
162.1
14%

-------
      SAIC-GLNPO (CF #361)

      CONVENTIONALS IN TREATED SEDIMENT
                                                   B.E.S.T.
                                                                                              REVISED
                                                                                              2/18/92
% Total Oil ft Grease
MSLCode
MDL
Sponsor ID

% Moisture
0.01%
PH
NA
Volatile Solids
0.001%
(mg/kg)
0.1
TOO
% weight
0.007%
Total Cyanide
(mg/kg)
0.001
Total Phosphorus
(mg P/kg)
0.001
O1
      361-8, Rep  1
      361-8, Rep  2
      361-8, Rep  3
      361-9
      361-10
      Method Blank
S-TS-RCC, Rep 1
S-TS-RCC, Rep 2
S-TS-RCC. Rep 3
B-TS-RCC
I TS-RCC
STANDARD REFERENCE MATERIAL

MESS-1 SRM
In-house Concensus Value *
0.16%
NA
NA
3.72%
0.50%
NA
10.73
NA
NA
10.30
10.25
NA
1.76%
1.70%
1.74%
3.91%
9.06%
NA
297
293
206
238
470
0.1
0.58%
NA
NA
1.21%
13.36%
0.011%
50.72
NA
NA
8.11
6B.5
0.001 U
                                         NA
                                         NA
                             NA
                             NA
NA
NA
NA
NA
2.45
 2.3
NA
NA
                                                        63
                                                        NA
                                                        NA
                                                        20
                                                       7214
                                                       0.089
NA
NA
       REPLICATE ANALYSES
       381-8. Rep 1
       361-8, Rep 2
       361-8, Rep 3
 S-TS-RCC, Rep 1
 S-TS-RCC, Rep 2
 S-TS-RCC, Rep 3
      RSD%
0.16%
NA
NA
NA
10.73
NA
NA
NA
1.76%
1.70%
1.74%
2%
297
293
206
19%
0.58%
NA
NA
NA
50.72
NA
NA
NA
       NA • Not analyzed
       U  - Below detection limit
          • TOC value for MESS determined based on past In-house analyses.
       *  - Mean for replicated sample.
       NOTE:   All Conventional results are reported on a dry weight basis.
                                                Not a statistical determination.
                                                        63
                                                        NA
                                                        NA
                                                        NA

-------
SAIC GLNPO (CF 1361)
METALS IN TREATED SEDIMENT
(Concentrations

MSLCode
MDL
361 -8. Rep 1
361 8. Rep 2
381 8. Rep 3
361 9
361-10
In ug/g dry weight)

Sponsor ID

S-TS-RCC. Rep 1
S TS-RCC. Rep 2
S-TS-RCC. Rep 3
BTSRCC
1- TS-RCC

AQ
AA
0007
1 03
076
067
024
434

As
»T
2.5
2.77
267
292
14 6
290

Ba
*r
43
325
321
310
396
290

Cd
AA
0.006
4.38
4 17
424
2.11
697

Cr
*t
33
130
113
110
113
1706
B.E.S.T.

Cu
XT
5.5
686
66.3
57.4
61 2
223

%Fe
*t
0.26
0.855
0627
0 795
4.42
825

HJ
CVAA
0.0003
0.505
0.290
0 209
0.627
1.458

Ml
HT
56
181
177
172
684
2540

Nl
wr
7.5
71.5
64 1
572
42 1
10 U

Pb
•Hf
6 2
533
450
41.4
101 5
6560
REVISED
2/5/92

Se
AA
0.22
0.22 U
022U
022U
0.87
4 94

Zn
we
78
194.0
185.1
147.0
189.7
2810.0
            Method Blank
                                                0020
                                                          NA
                                                                    NA   0.006
                                                                                    NA
                                                                                            NA
                                                                                                      NA    000013
                                                                                                                         NA
                                                                                                                                 NA
                                                                                                                                           NA
                                                                                                                                                  0.22 U
                                                                                                                                                              NA
to
en
            STANDARD REFERENCE MATERIAL
1646 SRM
ctrlllUd
v»lut
MATRIX SPIKE RESULTS
Amount Spiked
361-8*
361-8 * Spike
Amount Recovered
Percent Recovery
REPLICATE ANALYSES
361 -B. Rep 1 S-TS HOC. Rep 1
361-8. Rep 2 S-TS RCC, Rep 2
361-8. Rep 3 S-TS-RCC. Rep 3
RSO%
0 113
rC
N3

2
0 82
3 22
2 4
120%

1 03
0.76
0 67
23%
13 1
11 6
±1.3

NS
NS
NS
NS
NS

2 77
2 87
292
3V.
             J   . Values delected below MDL.
             U  -  Below detection limit
             NA - Not analyzed/applicable
                 .  Mean ol triplicated sample
             NS . Not spiked
             NOTE.  All Metals results aie 'blank corrected.'
411
N3
tc
NS
NS
MS
NS
NS
325
321
310
2%
042
036
±007
2
4.26
6.25
1 99
100%
4.36
4 17
4.24
3%
65
76
±3
NS
NS
NS
NS
NS
130
113
110
m.
22 3
IB
13
NS
NS
NS
NS
NS
68.6
66.3
57.4
B%
347
3.35
±0.1
NS
NS
NS
NS
NS
0.655
0.827
0.795
4%
0.066
0.063
±0012
1.987
0.335
2.282
1.947
98%
O.S05
0.290
0.209
46%
350
375
±20
NS
NS
NS
NS
NS
161
177
172
3%
38.1
32
±3
NS
NS
NS
NS
NS
71.5
64.1
57.2
11%
286
282
tt a
NS
NS
NS
NS
NS
533
450
41.4
13%
074
NO
NO
2 74
022U
2.99
2.99
109%
0.22 U
0.22 U
022U
NA
133.4
136
±6
NS
NS
NS
NS
NS
194.0
165 1
147.0
14%

-------
to
-vl
&/MU ULNh-U (Lit- »3B1)
PAH IN UNTREATED SEDIMENT
Low Molecular Weight PAHs (ng/g dry wl )

B.E.5.T.



REVISED
2/9/92
Naphthalene Acenaphthytene Acenapnttiene Fluorene Phananlhrene Anthraon*
MSL Code Sponsor ID
3615, Rep 1 R S US RCC, Rep 1
3615. Rep 2 R S US RCC. Rep 2
361-5. Rep 3 R S US RCC. Rep 3
361-6 BUS RCC
361-7 1 US RCC
Method Blank 3
Method Blank-R
STANDARD REFERENCE MATERIAL
SRM NIST1941
ccrlllUd v»lu»
MATRIX SPIKE RESULTS
Amount Spiked
361 5 i ~K -
361 5 • Spike R
Amount Recovered
Percent Recovery

26 B
24 B
27B
107B
4402 B
set
1 1

364
NC

933
26
406
380
41%

15
14 U
18
79 U
2322
58 U
1 1 U

54
NS

933
V6- <\%
577
RW 51T
66H loi'/»

19
20 U
22 U
113 U
4396
83 U
16 U

60U
NC

933
£9 
-------
to
CO
SAIC GLNPO (CF »361)
PAH IN UNTREATED SEDIMENT
High Molecular Welqhl PAHs (nq/q dry wl )


MSLCode Sponsor ID
381-5. Rep 1 R S US RCC. Rep 1
361-5. Rep 2 R S US RCC, Rep 2
361-5. Rep 3 R S US RCC, Rep 3
361-6 BUS RCC
361-7 1 US RCC
Method Blank 3
Method Blank- R
STANDARD REFERENCE MATERIAL
SRMNIST1041
cerlllled value
Amount Spiked __
381 5 • \ •»
361-5 + Spike R
Amount Recovered
Percent Recovery


Fhjotarv
Ihene
376 B
464 B
351 B
1197 B
32032 B
44
0

1114
1220
933
307
1075
678
73%


Pyrene

425 B
491 B
402 B
1157 B
32022 B
39
9

1034
1080
933
439
1077
637
68%


Benzo(a)-
anlhracene
184
210
163
861
18282
37U
5U

461
550
933
186
019
733
79%


Chrysene

285 B
299 B
242 B
1037
24399
36 U
S

703
NC
033
269
056
686
74%


Beruo(b)-
tuoranlhene
238 B
270 B
217B
878
19187
27U
6

766
760
033
242
1044
802
86%
B.E.S.T.

Benio (k)-
fluorantiene
180 B
200 B
158 B
733
13350
23 U
5

603
444
033
170
648
669
72%



Indeno

Dlbenzo
REVISED
2/9/02

Benio(a)- (l.2.3.c.d) (a.h)amhra- Benio(g.h.l)-
pyrene
224 B
250 B
200 B
887
20581
SOU
5

500
670
033
225
087
762
82%
pyrene
201 B
231 B
186 B
607
14653
27U
5

408
569
933
207
1095
888
9S%
cene
44
47
39
205
5224
35U
4 U

141
NC
033
43
012
660
03%

113B
120 B
100 B
405
13767
26 U
6

421
516
033
117
660
543
56%
         R  - Re-extracted sample results.
         •  - Mean ol klpllcaled sample
         B  - Analyle detected In Blank associated with sample.
         U  - Below detection  limits
         NC - Nol collided
         *  -  Value outside ol  Internal QC limits (40 120%)

-------
SAIC OLNPO (CF §361)
PAH IN UNTREATED SEDIMENT
B.E.S.T.
REVISED
2/5/03
1 Surrogate Recovery %

MSLCoda
361-5. Rep 1 R
361-5. Rep 2 R
361-5, Rep 3 R
361-6
361-7
Method Blank 3
Method Blank R

Sponsor ID
S US RCC. Rep 1
S US RCC. Rep 2
S US RCC. Rep 3
BUSRCC
1 US RCC


DB Naph-
thalene
37%*
34%*
37%'
25% *
27% '
26% '
51%
D10 Acenaph-
thalene
53%
47%
50%
45%
48%
26% *
62%
D12 Perylene

79%
63%
76%
64%
60%
90%
72%
        STANDARD REFERENCE MATERIAL

        SRMNIST1041
                                                              26%*
                47%
               74%
CD
        MATRIX SPIKE RESULTS

        Amount Spiked
        391-5  *
        361 5 4 Spike R
        Amount  Recovered
        Percent Recovery

        R  - Re extracted sample results.
        *  - Value outside ol  Internal QC  llmlls (40 120%)
        NA - Not appllcabie.
 NA
38%
40%
 NA
 NA
 NA
50%
54%
 NA
 NA
 NA
79%
88%
 NA
 NA

-------
CO
o
SAIC GLNPO (CF #361)
PAH IN TREATED SEDIMENT
Low Molecular Weight PAHs (ng/g dry wl )

B.E.S.T.
REVISED
2/18/92
Naphthalene Acenaphthytene Acenaphtiene
MSL Code Sponsor ID
361-8 H S-TS-RCC
381 9 R B-TS RCC
361-10 I-TS RCC
Method Blank-4
Method Blank R
STANDARD REFERENCE MATERIAL
SRMNIST1941
certified vilue
MATRIX SPIKE RESULTS
Amount Spiked
361 8
361-8 + Spike R
Amount Recovered
Percent Recovery
Amount Spiked
361-8 DUP
361-8 + Spike DUP R
Amount Recovered
Percent Recovery

24 B
37B
2253
38 U
11

364
tc

2222
24
723
699
31%'
1799
24
999
975
54%

14 U
15U
121
39 U
11 U

54
tc

2222
14 U
1004
1004
45%
1799
14 U
1262
1249
69%

19 U
21 U
726
56 U
16 U

60 U
NC

2222
19 U
933
933
42%
1799
19 U
1149
1130
63%
Fluorene Phenanlhrene Anthracene

16 U
18 U
1078
49 U
13 U

03
ND

2222
16 U
1165
1185
53%
1799
16 U
1281
1265
70%

99 B
68 B
5642
33 U
9

550
577

2222
99
2052
1954
88%
1799
99
1468
1369
76%

17
30
1474
36 U
9 U

165
202

2222
17
1658
1642
74%
1799
17
1370
1353
75%
                  R  - Re-extracted sample results.
                  U  - Below detection limits
                  NC - Not certified
                  •  -  Value outside ol Internal OC limits (40-120%)

-------
SAIC GLNPO (CF »361)
PAH IN TREATED SEDIMENT
Hlqh Molecular Weigh! PAHs (nq/g dry wl )

MSL Cod* Sponsor ID
361-8 R S-TS-RCC
381-0 R B-TS -RCC
361-10 I-TS-RCC
Method Blank-R
Method Blank R
STANDARD REFERENCE MATERIAL
SRM-NIST1941
certified value
MATRIX SPIKE RESULTS
Amount Spiked
361-8R
381 8 4 Spike R
Amount Recovered
Percent Recovery
Amount Spiked
361 8 R DUP
361-8 + Spike DUP R
Amount Recovered
Percent Recovery
B.E.S.T.
Fluoran-
thene
138 B
45 B
3105
25U
0

1114
1220

2222
138
2655
2717
122% '
1798
136
1698
1561
87%
Pyrene
REVISED
2/18/82
Indeno Dlbenio
Benzo(a)- Chryscne Benzofb)- Beruo(k)- B«nio(B>- (1,2.3.c.d) (a.h)anthra- B«nzo(g,hJ)-
anthracene
120B
38 B
3553
26 U
0

1034
1080

2222
120
2555
2434
110%
1789
120
1584
1473
62%
57
22
3126
26U
5 U

481
650

2222
57
2654
2506
117%
1788
57
1600
1632
81%
luoranlhena ftuorantiene
88 B
38 B
3882
24 U
5

703
rC

2222
88
2688
2502
113%
1788
88
1611
1623
85%
87B
27B
1887
16U
6

766
780

2222
87
2670
2773
125%'
1788
87
1734
1637
81%
88 B
4 B
1316
16 U
5

603
444

2222
66
2367
2301
104%
1788
66
1608
1443
80%
pyrene
76 B
18 B
3220
20 U
6

500
670

2222
76
2584
2508
113%
1788
78
1506
1430
79%
pyrene
00 B
18B
1360
18U
5

488
568

2222
80
2838
2848
128%'
1788
00
1703
1613
00%
cene
18
6
1830
24 U
4 U

141
rC

2222
16
2610
2584
117%
1788
16
1638
1622
80%
perytene
80 B
14 B
2354
18
6

421
616

2222
60
1786
1726
78%
1788
60
106S
1005
66%
R  -  Re-extracted sample results.
U  .  Below detection limits
NC - Not certified
*  - Value outside ol Internal OC  limits  (40-120%)

-------
          SAIC GLNPO (CF »361)

          PAH IN TREATED SEDIMENT
                                                             B.E.S.T.
                            REVISED
                            2/18/82
( Surrogate Recovery %
MSLCode
Sponsor ID
DB Naph-
thalene
01 0 Acenaph-
thalene
012 Perylene
          361-8 R
          361-0 R
          361-10
               S-TS-RCC
               B-TS-RCC
               I-TS-RCC
41%
31%*
25%'
46%
43%
S4%
67%
73%
85%
           Method Blank-R
           Method Blank-R

           STANDARD REFERENCE MATERIAL
                                                     21%*
                                                     61%
                19%'
                62%
                60%
                72%
           SRMNIST1041
                                                      28%*
                 47%
                74%
CO
to
MATRIX SPIKE RESULTS

Amount Spiked
36I-8R
361-8 + Spike R
Amount  Recovered
Percent Recovery

Amount Spiked
361 8 R DUP
361-8 + Spike DUP  R
Amount  Recovered
Percent Recovery

R  - Re extracted sample results.
•  - Values  outside of Internal QC limits (40-120%).
NA - Not applicable.
                                                                  NA
                                                                 41%
                                                                 31%'
                                                                  NA
                                                                  NA
                                                                   NA
                                                                 41%
                                                                 66%
                                                                   NA
                                                                   NA
                  NA
                 46%
                 40%
                  NA
                  NA
                  NA
                 46%
                 64%
                  NA
                  NA
                  NA
                 67%
                108%
                  NA
                  NA
                  NA
                 67%
                 87%
                  NA
                  NA

-------
             SAIC GLNPO (CF »361)

             PAH IN WATER
                                                                         B.E.S.T.
                                                                                                                              REVISED
                                                                                                                               2/3/92
              low Molecular Walghl PAHs (ng/U
MSL Code Sponsor ID
361-1
361-2
381 11
SWHRCC
B WHRCC
1 WR HCC
Naphthalene Acenaphlhytene Acenaphftene
1441 B
988 B
301 B
538 U
342 U
495
679 U
432 U
165
Fluorene Phenanlhrene Anthracene
641 U
408 U
278
421 U
268 U
2719
508 U
323 U
997
CO
00
Method Blank 2

MATRIX SPIKE RESULTS

Amounl Spiked
Blank 2
Blank-2 + Spike
Amounl  Recovered
Percent Recovery
                                                                1767
5000
1767
2552
 785
 16%'
                                                                                  963 U
                                                                                                  1216 U
              B - Analyle present In method blank associated with sample
              U -  Below detection limits
              * - Value outside ol Internal OC limits (40-120%).
                                                                                                                  1148 U
                                                                                                                                   754 U
                                                                                                                                                910U
5000
963 U
1072
1072
21% '
5000
1216 U
1143 U
1143U
0% '
5000
1148 U
1108
1108
22% '
5000
754 U
1504
1504
30%'
5000
910 U
1420
1420
28% '

-------
SAIC GLNPO (CF
PAH IN WATER
•361)



B.E.S.T.

REVISED
2/S/02

I Surrogate Recovery %
MSLCode
361-1
361-2
361-11
Sponsor ID
SWRRCC
BWRRCC
IWRRCC
08 Naph-
thalene
39%*
35%*
20%'
010 Acenaph-
thalene
44%
37%*
28%
012 Perylene
130%
111%
00%
CO
Method Blank-2

MATRIX SPIKE RESULTS

Amount Spiked
Blank-2
Blank-2 + Spike
Amount  Recovered
Percent Recovery
                                                                37%
 NA
37%'
20%'
 NA
 NA
                                                                                 38% '
 NA
38% '
20% '
 NA
 NA
                                                                                                 83%
  NA
 83%
109%
  NA
  NA
         *  - Value outside ol Internal QC limits (40-120%).
         NA  - Not applicable.

-------
 SAIC GLNPO (CF (361)

 PAH IN WATER

 High Molecular Weight PAHs (ng/l)
                                          B.E.S.T.
                                                                                          REVISED
MSLCode
361 1
361-2
361-11
Sponsor ID
S WR RCC
B-WR-RCC
I-WR-RCC
Fhrararv
Ihene
402
256 U
17056
Indeno Dlbenzo
Pyrene Beruo(a) Chrysen* Benzo(b)- Benzof»- Bento(i}- (1.2.3.c.d) (a.h)anlhra- Beruo(o,M-
anlhracene tuoranlhene Ruorantiene pyrene pyrane cent pcryton*
405 U 436 U 360 U 361 U 200 U 367 U 348 U 377 U 280
258 U 289 U 242 230 U 18SU 247U 221 U 240 U 184
17998 6418 10870 8708 3068 6181 3235 762 2841
 Method  Blank 2

 MATRIX SPIKE RESULTS

 Amount Spiked
 Blank 2
 Blank 2 + Split*
 Amount  Recovered
.^Percent Recovery
CO
                                           721 U
726 U
  B - Analyle present In method blank associated with sample
  U - Below detection limits
  * - Value outside ol Internal QC limits  (40-120%).
            613U
680 U
                                   648 U
                                               620 U
                                                          604 U
                                                                     623 U
                                                                                 676 U
                                                                                              S18U
6000
721 U
3816%
3816%
76%
6000
726 U
3830%
3830%
77%
6000
813U
6076
6076
140%'
6000
680 U
6752
5752
115%
6000
648 U
6620
8620
133%*
6000
620 U
6570
5570
112%
6000
604 U
4704
4704
eo%
6000
623 U
6686
5685
114%
6000
676 U
6011
6011
118%
6000
518
4854
4336
67%

-------
iAIC-GLNPO (CF 11361)
»AH IN OIL
bow Molecular Weight PAHs (no/ml)

: Sample
MSL Code Sponsor ID Density (g'ml)
361-3 S-OR-RCC 0
361-4 R B-OR-RCC 0
361-12. Rep 1 I-OR-RCC. Rep 1 0
361-12. Rep 2 I-OR-RCC. Rep 2 0
361-12. Rep 3 I-OR-RCC. Rep 3 0
Method Blank
OIL CONCENTRATIONS ON % OIL BASIS
Low Molecular Wekftl PAHs (oofcaoH)
MSL Code Sponsor ID
361-3 S-OR-RCC
361-4 R B^OHRCC
361-12. Rep 1 1 OR RCC. Rep 1
361-12. Rep 2 1 OR RCC, Rep 2
361-12. Rep 3 1 OH RCC. Rep 3
MATRIX SPIKE RESULTS
Amount Spiked
361-3
361-3 + Spike
Amount Recovered
Percent Recovery
7525
6003
7301
7301
.7301


%ON
0 25
6 20
5908
5998
59.06



Naphthalene
603 U
369 U
12127 DU
1 1 1 78 DU
11639DU
1774DU

Naphthalene
8664 U
8508 U
27694 DU
25526 DU
26351 DU

SOOOO
603 U
6260
5260
11%*
B.E.S.T.

Acenaphthylene Acenaphtiene
764 U
396 U
20391 D
22779 D
198590
1902U

Acenaphthylene
10977U
9121 U
46565 0
52019 D
45351 0

SOOOO
764 U
16600
16800
34%*
985 U
657 U
18291 DU
221840
18657 D
2675 DU

v*in6pnV)9n*i
13865U
12B32U
41770 DU
50660 0
42377 D

SOOOO
965 U
19300
19300
30%'


Fkiorene Phenanlhrene
12SO
1063
21142D
24980 D
21992 D
2241 DU

Fkiorene 1
17960
24483
482800
57045 D
50222 D

SOOOO
1250
32400
31150
62%
11800
6943
92741 0
97299 0
00220 D
1312 DU

•henanthrene
160540 D
160068
211786 D
222105 D
206020 D

SOOOO
11600
56700
44000
90%
REVISED
2/21/92
Anthncvnt
7210
6465
47165 D
61328 0
45668 D
1489 DU

Anthracene
103592 D
125930
107684 D
1172140
104289 D

SOOOO
7210
61800
54590
109%
R  -  Re-extracted sample results
D  - Samples diluted  1:10 and rerun
U  -  Below detection  limits.
• > Outside of Internal QC  limits (40 120%)

-------
CO
SAIC GLNPO (CF «361)
PAH IN OIL
Htoh Molecular Wolghl PAHs (ng/ml)
B.E.S.T.
Sample
Density
MSL Cod* Spomof ID (gyrrt)
361-3 S-OH-RCC 0
361-4 R &ORRCC 0
361-12. Rap 1 I-OR-HCC. Rep 1 0
361-12. Rap 2 1 OR RCC. Rep 2 0
361-12. Rep 3 1 OR RCC. Rep 3 0
Method Blank
7525
6903
.7301
7301
.7301

Fhioran-
Ihene
10500
8715
277651 D
287605 D
270480 D
808 DU
Pyrene
Benzo(a)-
anthracene
17900
7932
266075 D
274465 D
258488 0
895 DU
8070
3870
151300 D
160160 D
146080 D
846 DU
Chryten*
Benzo(b)-
Benzo (k)-
luoranthene duorantfiene
0060
5220
210701 D
220510 D
204500 0
80S DU
8710
3771
1 80630 D
1 00805 D
177767 D
646 DU
6430
2068
126840 D
120943 D
122228 D
520 DU
Benzo(*}-
pyrene
7020
3684
180062 D
108435 D
180074 D
688 DU
Indeno Dlbenro
(1.2,3.c.d) (a.h) anlhra-
pyrene
6300
2028
159145 D
165497 D
1507580
cen*
2160
670
32883 D
35296 D
30003 D
707 DU 604 DJ
REVISED
2/21/02
Benzo(g,h.l)-
perylene
4840
1708.00
06127 D
08204 D
00713 D
1430 O
OIL CONCENTRATIONS ON % OIL BASIS
Low Molecular WakjhlPAHs (ugAflolQ

MSL Cod*
Sponsor ID
361-3 S-OR-RCC
361-4 R B^ORRCC
381-12. Rep 1 I-OR RCC. Rep 1
361-12, Rep 2 I-OR-RCC. Rep 2
361-12, Rep 3 I-OR-RCC, Rep 3
MATRIX SPIKE RESULTS
Amount Spiked
361-3
361-3 * Spike
Amount Recovered
Percent Recovery


%OI
f*l
0.25
6.20
60.08
60.08
60.08








Fhioran-
Ihene
280172
200614
634051 0
656088 D
617606 D

50000
10500
83600
84100
128% '


Pyrene

267184
182760
607616 D
626776 D
500240 0

50000
17000
77500
59600
119%


Beiuofa)-
anthracene
116048
80370
346718 D
36S743D
336647 D

60000
6070
78600
70530
141%'


Ctuytene


B*iuo(b>-


Bemo(k>-
luoranthene luoranlieoe
143301
120480
4813680
603562 D
467228 D

60000
0080
86100
56120
112%
125144
86800
433065 D
4684850
405000 D

50000
8710
75800
67090
134%*
023BS
68386
280865 0
206741 D
270123 D

60000
6430
64300
67870
116%


B«UO(a)-
pyren*
113703
84870
433602 D
453151 D
4132770

50000
7020
72700
64780
130% '

Indeno
(l.2.3.e.d)
pyren*
00517
87465
363428 D
377033 D
3442760

60000
6300
77100
70800
142% '

Dlbenzo
(a.h) anthra-
cene
31034
13336
75002 0
80603 0
70571 D

60000
2160
81400
70240
158%*


Benzo(g,h.l)-
penrtene
60540
30367
210518 D
224467 D
207166 D

50000
4840
64700
50860
120%
           R  - Re-extracted •ample results.
           D  - Samples diluted 1.10 and re-run.
           U  -  Below  detection limits
           • . Outside ol Internal QC limits (40-120%)

-------
co
oo
SAIC-GLNPO (CF 1361)
PAH IN OIL

B.E.S.T.
REVISED
2/21/82
I Surrogate Recovery % I

MSLCode Sponsor 10
381-3 S-OR-RCC
301-4 R B-OR-RCC
381-12. Rep 1 I-OR RCC. Rep 1
361-12. Rep 2 1 OR RCC. Rep 2
361-12. Rep 3 I-OR-RCC. Rep 3
Method Blank
OIL CONCENTRATIONS ON % OIL BASIS
Low Molecular Wclgjil PAHs (uoAooN)

MSLCod* Sponsor ID
361 -3 S-OR-RCC
361-4 R B-OR-RCC
361-12. Rep 1 I-OR-RCC. Rep 1
361-12, Rep 2 I-OR-RCC, Rep 2
361-12. Rep 3 I-OR-RCC, Rep 3
MATRIX SPIKE RESULTS
Amount Spiked
361-3
361-3 + Spike
Amount Recovered
Percent Recovery
D8 Naph-
thalene
6% '
22%'
23% D*
30% 0
19% D*
60% D

DID Acenaph-
Ihalene
46%
39%'
81% D
63% 0
59% 0
118% D

1 Surrogate Recovery %
D8 Naph-
thalene
8%*
22%'
23% D*
30% D
10% 0*

NA
8% '
11% '
NA
NA
01 0 Acenaph-
thalene
46%
39%*
61% D
63% D
59% D

NA
46%
38%'
NA
NA
D12 Perylene

181%
85%
123% D'
108% D
108% D
73% D

1
01 2 Perylene

161%
05%
123% D*
106% D
108% D

NA
161%
130%
NA
NA
                     R  - Re-extracted  •ample results.
                     D  - Samples diluted 1:10 and re-run.
                     •   . Outside of Internal OC limits  (40 120%).
                     NA - Not applicable.

-------
CO
CD
         RE-PROCESSED RESULTS (1/92)
         PCBs IN UNTREATED SEDIMENT
         Concentrations In  ug/kg dry weight
     B.E.S.T
SAIC-GLNPO(CF#381)
2/18/92
MSLCode Sponsor ID
361-5. Rep 1 S-US-RCC, Rep 1
361-5, Rep 2 S-US-RCC. Rep 2
361-5. Rep 3 S-US-RCC. Rep 3
361-6 B-US-RCC
361-7 I-US-RCC
Blank-3
STANDARD REFERENCE MATERIAL
SRM-1 (HS-2)
certified value
MATRIX SPIKE RESULTS
Amount Spiked
361-5*
361-5 + Spike
Amount Recovered
Percent Recovery
Amount Spiked
Blank-3
Blank-3 + Spike
Amount Recovered
Percent Recovery
REPLICATE ANALYSES
361-5, Rep 1 S-US-RCC, Rep 1
361-5, Rep 2 S-US-RCC. Rep 2
361-5. Rep 3 S-US-RCC. Rep 3

Aroclor
1242
25372
16249
17666
200 U
200 U
200 U

40 U
N3

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS

25372
16249
17666
RSD% 25%
Aroclor Aroclor Aroclor
1248 1254 1260
200 U
200 U
200 U
325
11257
200 U

40 U
NC

N3
NS
NS
NS
NS
NS
NS
NS
NS
NS

200 U
200 U
200 U
0%
2591 E 100 U
1691 E 100U
2025 E 100 U
100 U 100U
3746 E 100 U
100U 100U

132 127
111 N3

609S NS
2102 NS
6129 NS
4027 NS
66V. NS
3333 NS
100 U NS
2498 NS
2498 NS
75% NS

Tetrachloro- Oclachloro-
m-Xylene naphthalene
60.9% 84.2%
58.0% 94.1%
59.6% 89.5%
63.3% 66.2%
61.7% 57.1%
256.1% ' 122.7% '

39.0% ' 42.4%
N3 N3

NA NA
58.8% 89.3%
61.0% 90.1%
NA NA
NA NA
NA NA
258.1% * 122.7% *
48.2% 55.8%
NA NA
NA NA

2591 100U 60.9% 84.2%
1691 100U 56.0% 94.1%
2025 100U 59.6% 89.5%
22% 0%
4% 6%
          U • Below detection limits.
          E • Values due to residuals from high Aroclors 1242 and 1248 levels.
          ' «  Value outside of Internal QC limits (40-120%).
          NC  -  Not certified.
          # -  Mean of 3 replicates.
          NS  = Not spiked.  NA =  Not applicable.

-------
RE-PROCESSED RESULTS (1/92)
PCBs IN TREATED SEDIMENT
Concentrations In uq/kq dry  weight
         B.E.S.T
    SAIC-GLNPO(CF#361)
2/18/92
                                     % Surrogate Recovery
MSLCode Sponsor ID
361-8 S-TS-RCC
361-9 B-TS-RCC
361-10 " I-TS-RCC
Blank-4
STANDARD REFERENCE MATERIAL
SRM-2 (HS-2)
certified value
MATRIX SPIKE RESULTS
Amount Spiked
361-8
361-8 -i- Spike
Amount Recovered
Percent Recovery
Amount Spiked
361-8
361 -8 + Spike DUP
Amount Recovered
Percent Recovery
U - Below detection limits.
J - Detected below detection limit.
Aroclor Aroclor Aroclor Aroclor lletrachloro- Octachloro-
1242 1248 1254 1260 | m-Xvlene naphthalene
205
100 U
100 U
200 U

200 U
N3

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS


100 U
100 U
440
200 U

200 U
N3

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS


30 J
50 U
50 U
100 U

100 U
111

2404
30 J
2060
2030
84%
2273
30 J
1169
1139
50%


SOU
SOU
SOU
100U

100U
ND

NS
NS
NS
NS
NS
NS
NS
NS
NS.
NS


35.2% *
34.5% '
89.7%
21 .0% *

44.8%
N3

MA
35.2% *
72.3%
NA
NA
NA
35.2% *
30.2% *
NA
NA


71.2%
90.0%
50.3%
71.5%

41 .7%
N3

NA
71.2%
89.2%
NA
NA
NA
71.2%
55.4%
NA
NA


  * . Value outside of Internal  QC limits  (40-120%).
  " . Early elutlng peaks present that do not match  Arolcor pattern.
  NC - Not certified.
  NS » Not spiked.
  NA = Not applicable.
Quantity estimated at -1000-2000 ppb, based on TCX response factor.

-------
RE-PROCESSED RESULTS (1/92)
PCBs IN WATER
Concentrations In  ug/L
     B.E.S.T
SA1C-GLNPO (CF #381)
2/18/92
MSLCode
361-1
361-2
361-11
Blank-1
Sponsor ID
S-WR-RCC
B-WR-RCC
I-WR-RCC

Aroclor
1242
0.2
0.2
0.2
0.2

U
U
U
U
Aroclor
1248
0.2
0.2
4.8
0.2

U
U

U
Aroclor
1254
0.1
0.1
0.1
0.1

U
U
U
U
Aroclor
1260
0.1
0.1
0.1
0.1

U
U
U
U
Tetrachloro-
m-Xylene
41.2%
35.2% *
43.5%
36.5% '
Octachloro-
naphlhalene
114.1%
98.0%
86.0%
98.0%
MATRIX SPIKE RESULTS

Amount Spiked
Blank-1
Blank-U Spike
Amount Recovered
Percent Recovery

U « Below detection limits.
* - Value outside of Internal QC limits (40-120%).
NC . Not certified.
 NS - Not spiked.
 NA - Not applicable.
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
50
0.1 U
42.6
42.6
85%
NS
NS
NS
NS
NS
NA
38.5% '
33.6% '
NA
NA
NA
98.0%
123.8%
NA
NA

-------
RE-PROCESSED RESULTS (1/92)
PCBs IN OIL
     B.E.S.T
8AIC-GLNPO (CF §361)
             2/26/92

% Surrogate Recovery

Sample
MSLCode Sponsor ID Density (g/ml)
361 -3 S OR RCC
361-4 B OR RCC
361-12, Rep 1 1 OR RCC. Rep 1
361-12. Rep 2 I-OR-RCC. Rep 2
361-12, Rep 3 I-OR-RCC, Rep 3
Blank-1 OH
0.7525
0.6903
0.7301
0.7301
0.7301

Aroclor
1242
318320
2000 U
2000 U
2000 U
2000 U
2000 U
Aroclor
1248
2000 U
2702
118616
120743
112109
2000 U
Aroclor
1254
30789
1000 U
1000 U
1000 U
1000 U
1000 U
Aroclor
1260
1000U
1000U
1000U
1000 U
1000 U
1000U
OIL mur«ITR»T10N9 ON % OIL BASIS
Concentrations In ug/Kq oil \
MSLCode Sponsor ID
361 -3 8 OR RCC
381-4 B OR RCC
361-12, Rep 1 I-OR-RCC, Rep 1
361 -1 2. Rep 2 1 OR RCC, Rep 2
381-12, Rep 3 I-OR-RCC, Rep 3
MATRIX SPIKE RESULTS
Amount Spiked
361-3
361-3* Spike
Amount Recovered
Percent Recovery
REPLICATE ANALYSES
381-12, Rep 1 I-OR-RCC. Rep 1
361-12. Rep 2 1 OR-RCC, Rep 2
381-12. Rep 3 IOH-RCC. Rep 3

%OII
(%)
0.25
8.20
50.08
50.08
50.08










RSD%
Aroclor
1242
4573563
46083 U
4587 U
4567 U
4587 U

NS
NS
NS
NS
NS

2000 U
2000 U
2000 U
NA
Aroclor
1248
28738 U
62258
270875
275732
256015

NS
NS
NS
NS
NS

118616
120743
112109
4%
Aroclor
1254
442371
23041 U
2284 U
2284 U
2284 U

60000
30780
84012
34123
68%

1000 U
1000 U
1000 U
NA
Aroclor
1260
Tetrachloro-
m-Xylene
65.0%
71.5%
03.7%
85.5%
85.2%
48.6%
Octachloro-
naphlhalene
00.5%
01.0%
96.0%
83.7%
90.4%
96.7%
% Surrogate Recovery
retrachloro-
m-Xylene
14368U 65.0%
23041 U 71.5%
2284 U 03.7%
2284 U 85.5%
2284 U 85.2%

NS
NS
NS
NS
NS

1000
1000
1000
NA

NA
65.0%
61.2%
NA
NA

U 93.7%
U 85.5%
U 65.2%
5%
Octachloro-
naphlhaUne
00.5%
01.0%
06.0%
83.7%
00.4%

NA
00.5%
00.4%
NA
NA

06.0%
83.7%
00.4%
7%
  U - Below detection limits.
  NC - Not certified.
  NS - Not spiked.
  NA - Not applicable.

-------
                                         APPENDIX F

                           QUALITY ASSURANCE/QUALITY CONTROL

        In order to obtain data of known quality to be used in evaluating the different technologies for the
 different sediments, a Quality Assurance Project Plan (QAPP) was prepared.  The QAPP specified the
 guidelines to be used to ensure that each measurement system was in control.  In order to show the
 effectiveness of the different technologies, the following measurements were identified in the QAPP as critical
 - PAHs, PCBs, metals, total solids, volatile solids, and oil and grease in the untreated and treated sediments.
 Other parameters analyzed in the sediments included pH, TOO, total cyanide, and total phosphorus. If water
 and oil residuals were generated by a technology, then PAHs and PCBs were determined  as a check on
 their fate in treating the sediments.  Each of these measurements and the associated quality control (QC)
 data will  be discussed in this section.
       Also included in this section are a discussion of the modifications and deviations from the QAPP
 and the results of a laboratory audit performed.  Any possible effects of findings on data quality will be
 presented.

 PROCEDURES USED FOR ASSESSING DATA QUALITY
       The  indicators used  to assess the quality of the data generated for this  project  are accuracy,
 precision, completeness, representativeness, and comparability. All indicators will be discussed generally
 in this  section; specific results for  accuracy, precision,  and completeness will be summarized in later
 sections.
 Accuracy
       Accuracy is the degree of agreement of a measured value  with the true  or expected value.
Accuracy for this project will be expressed as a percent recovery (%R).
       Accuracy was determined during this project using matrix spikes (MS) and/or standard reference
materials (SRMs).  Matrix  spikes are aliquots  of sample spiked  with a known concentration of target
analyte(s) used to  document the accuracy of  a method in a given sample  matrix.  For  matrix spikes,
recovery  is calculated as follows:
                                            c,-c0
                             %R    =      	         X      100
                                              Ct
                     where:  C,     =      measured concentration in spiked sample aliquot
                             C0     =      measured concentration in unspiked sample aliquot
                             Ct     =      actual concentration of spike  added
                                             143

-------
 An SRM  is a known matrix spiked with  representative target analytes used to document  laboratory
 performance.  For SRMs, recovery is calculated as follows:
                                              Cm
                              %R    =      _           X       100
                      where:  Cm     =      measured concentration of SRM
                              Ct     =      actual concentration of SRM
        In addition, for the organic analyses, surrogates were added to all samples and blanks to monitor
 extraction efficiencies.  Surrogates are compounds which are similar to  target analytes  in chemical
 composition and behavior.  Surrogate recoveries will be calculated as shown above for SRMs.
 Precision
        Precision  is the  agreement  among a set  of replicate measurements without assumption of
 knowledge of the  true value. When  the number of replicates is two,  precision is determined using the
 relative percent difference (RPD):
                                            (C, - C2) X 100
                              RPD    =      _
                                            (C, + C2)/2
                      where:  C-,      =      the larger of two observed values
                              C2      =      the smaller of two observed values
When the number of  replicates is three or greater, precision is determined using the relative standard
deviation (RSD):
                                             S
                              RSD    =      	   X      100
                                             X
                      where:  S      =      standard  deviation of replicates
                              X      =      mean of replicates
       Precision was determined during this project using triplicate analyses for those samples suspected
to be high in target analytes (i.e., untreated  sediments).  Matrix spike and matrix spike duplicate (MSD)
analyses were performed on those samples suspected to be low in target analytes (i.e., treated sediments).
A MSD is a second spiked sample aliquot with a known concentration of target analyte used to document
accuracy and precision in a given sample matrix.
                                              144

-------
Completeness
        Completeness is defined as the percentage of valid data points to the total number of data points
obtained.
                                           VDP
               % Completeness      =      	    X      100
                                           TOP
                      where:  VDP    =      number of valid data points
                             TOP    =      total data points obtained
        For this project, completeness was determined for each parameter for each technology evaluated.
                             (Add more)
Representativeness
        Representativeness refers to the degree  with which analytical results accurately and precisely
represent actual conditions present at locations chosen for sample collection.  Sediment samples were
collected prior to this demonstration and were reported to  be representative of the areas to be remediated.
Comparability
        Comparability expresses the extent with which one data set can be compared to another. As will
be discussed in more detail in the section MODIFICATIONS AND DEVIATIONS FROM THE QAPP, the data
generated are comparable within this  project and within other projects conducted for the ARCS Program.

ANALYTICAL QUALITY CONTROL
        The  following sections summarize  and discuss analytical procedures and  the results of the QC
indicators of accuracy and precision for each measurement parameter.
PAH Procedures
        Sediments and waters were extracted and analyzed using modified SW-846 procedures as described
in the section MODIFICATIONS AND  DEVIATIONS FROM THE  QAPP.  Oils were diluted 1:10 in hexane.
Three radiolabelled PAH surrogates were added to all samples and blanks prior to extraction. Daily mass
tuning was performed using decafluorotriphenylphosphine  (DFTPP) to meet the criteria specified in Method
8270. The instrument was calibrated at five levels for the sixteen polynuclear aromatic hydrocarbons (PAHs).
The RSD of the response factors for each PAH was required to be <25 percent.  Calibrations were verified
every 12 hours  for each PAH; criteria for % difference from the initial calibration was <25 percent. An
internal  standard, hexamethyl benzene, was added prior to cleanup and was used to correct PAH
                                             145

-------
 concentrations for loss during cleanup and extract matrix effects.  Quantification was performed using
 Selective Ion Monitoring (SIM).
 PAH QC Results and Discussion
         Surrogate recoveries for all PAH samples for the B.E.S.T. demonstration are summarized in Table
 QA-1.  If more than one of the  three surrogates fell outside the control limits used, corrective action
 (reanalysis) was necessary.  (Insufficient sample remained for reanalysis of water residuals).  Surrogate
 recoveries were generally low for samples and method blanks, indicating a possible analytical problem rather
 than matrix effects. An investigation indicated possible problems with the evaporator used to concentrate
 the extracts. This concentration step was not performed for the oil residuals and, as can be seen in Table
 QA-1, acceptable surrogate recoveries for the method blank were obtained.  In summary, low surrogate
 recoveries indicate that PAH target concentrations may be biased somewhat low. Since both the untreated
 and treated sediments were affected similarly, relative removal percentages should be valid.
        As required by the QAPP, triplicate analyses of the Saginaw River untreated sediment (S-US-RCC)
 were performed to assess precision.  These results are summarized in Table QA-2. A matrix spike was
 performed on this same sample to assess accuracy. These results are included in Table QA-2.  All RSD and
 spike recoveries fell within the control limits specified.  It should be  noted that these QC analyses were
 reanalyses;  the initial analysis yielded unacceptable surrogates and poor precision.  The effect of missed
 holding times on data quality will be discussed in the section HOLDING TIMES.
        As required by the  QAPP, a matrix spike  and  a matrix spike duplicate (MS/MD) analysis was
 performed for the treated Saginaw River sediment (S-TS-RCC). These results are presented in Table QA-3.
 While recoveries were generally acceptable, RPDs were consistently outside the  control limits specified in
 the QAPP. These RPDs,  however, are generally within acceptance criteria specified in Method 8270; data
 should be of sufficient quality to support project results.  It should be noted that  spike levels averaged up
 to one hundred times the target concentrations and that accuracy and precision data for this MS/MSD may
 not be indicative of the accuracy and precision obtainable at the target concentrations.
        Due to the minimal  amount of water generated by the B.E.S.T. process, no QC analyses were
 performed on this matrix.
       The QAPP specified that triplicate analyses and a matrix spike be performed on the Saginaw River
 oil residual (S-OR-RCC).  This sample was spiked but triplicate analyses were performed on the Indiana
 Harbor oil residual.  These results are summarized in Tables QA-4 and QA-5, respectively.
       One certified NIST standard reference material (SRM) was extracted and analyzed with the sediment
samples.  The recoveries for this standard are summarized in Table QA-6.
       Method blanks were  extracted and analyzed with each set of samples extracted.  Insignificant
                                              146

-------
                    TABLE QA-1.  PAH SURROGATE RECOVERIES, PERCENT
Sample
S-US-RCC, Rep. 1
S-US-RCC, Rep. 2
S-US-RCC, Rep. 3
B-US-RCC
I-US-RCC
Method Blank
Method Blank
S-TS-RCC
B-TS-RCC
I-TS-RCC
Method Blank
S-WR-RCC
B-WR-RCC
I-WR-RCC
Method Blank
S-OR-RCC
B-OR-RCC
I-OR-RCC, Rep. 1
I-OR-RCC, Rep. 2
I-OR-RCC, Rep. 3
Method Blank
d8-Napthalene
37 *
34 *
37 *
25 *
27 *
26 *
25 *
41
31 *
25 *
21 *
39 *
35 »
20 *
37 *
8 *
22 *
23 *
30 *
19 *
60
d 1 0-Acenapthalene
53
47
50
45
48
26 *
24 •
46
43
54
19 *
44
37 *
28 *
38 *
46
39 *
61
63
59
118
d12-Peryiene
79
83
76
64
60
90
90
67
73
85
50
130 *
111
90
83
161 *
95
123 *
106
108
73
Control Limits
40-120
I
I
I
I
I
I
40-120
I
I
I
40- 120
I
I
I
40-120
I
I
I
I
I
       *      Outside Control Limits


quantities of some PAHs were found in a few blanks; total concentrations are unaffected.  No corrections
for method blanks were performed.
                                          147

-------
TABLE QA-2.  PAH REPLICATE AND SPIKE RESULTS FOR S-US-RCC
Compound
Napthalene
Acenaphthylene
Acenaphthene
Ruorene
Phenanthrene
Anthracene
Ruoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
lndeno(1 ,2,3,c,d)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
NC
U
Replicate 1
dry ppb
26
15
19
33
249
64
376
425
184
265
238
180
224
201
44
113
Not Calculated
Undetected
Replicate 2
dry ppb
24
20U
20U
32
302
70
464
491
210
299
270
200
250
231
47
129


Replicate 3
dry ppb
27
18
30U
33
249
63
351
402
163
242
217
158
200
188
39
109


Mean
26
NC
NC
33
267
66
397
439
186
269
242
179
225
207
43
117


RSD
6.0
NC
NC
1.8
11
5.7
14
11
13
11
11
12
11
11
9.4
9.1


Precision %
Control Limits Recovery
20 41
| 62
I 57
I 64
I 64
| 63
I 73
| 68
I 79
I 74
| 86
I 72
| 82
| 95
| 93
I 58


Accuracy
Control Limits
40-120%
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I



-------
                       TABLE QA-3. PAH MS/MSD RESULTS FOR S-TS-RCC
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Ruorene
Phenanthrene
Anthracene
Ruoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo (b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
lndeno(1 ,2,3,c,d)pyrene
Dibenzo(a,h)anthracene
Benzo (g,h,i)perylene
MS Recovery
31 *
45
42
53
88
74
122 *
110
117
113
125 *
104
113
128 *
117
78
MSD Recovery
54
70
64
71
76
75
87
82
91
85
91
80
79
90
90
56
Accuracy Precision
RPD Control Limits Control Limits
32 * 40 - 120% 20
43 *
42 *
29 »
15
1.3
33 *
29 *
25 *
28 *
31 *
26 *
35 *
35 *
26 *
33 *






























               Outside Control Limits
PCBs
PCB Procedures
       Sediments and waters were extracted and analyzed using modified SW-846 procedures as described
in the section MODIFICATIONS AND DEVIATIONS FROM THE QAPP. Oil were diluted 1 no in hexane. Two
surrogates, tetrachloro-m-xylene and octachloronaphthalene, were added to all samples and blanks prior
to extraction. The gas chromatograph (GC) employed electron capture detection (ECD) and was calibrated
at three levels for each of four Arociors (1242, 1248, 1254, 1260). The RSD of the response factors for each
Aroclor was required to be <25 percent.  Calibrations were verified after every ten samples; criteria for %
difference from the initial calibration was <25 percent.  An internal standard, dibromooctafluorobiphenyl, was
added  prior to cleanup and was used to correct PCB concentrations for loss during cleanup and extract
matrix  effects.   Quantification  of Arociors was performed on two columns (DB-5, primary and 608,
confirmation) as a confirmation of their presence.
                                               149

-------
                         TABLE QA-4. PAH MS RESULTS FOR S-OR-RCC
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Ruorene
Phenanthrene
Anthracene
Ruoranthene
Pyrene
Benzo (a)anthracene
Chrysene
Benzo (b)fluoranthene
Benzo (k)fluoranthene
Benzo(a)pyrene
lndeno(1 ,2,3,c,d)pyrene
Dibenzo (a.h)anthracene
Benzo(g,h,i)perylene
MS Recovery % Control Limits
1 1 Not Specified
34
39
62
90
109
128
119
141
112
134
116
130
142
158
120















PCB QC Results and Discussion
        Surrogate recoveries for all PCB samples for the B.E.S.T. demonstration are summarized in Table
QA-7.  If both recoveries fell outside the control limits used, correction action (reanalysis) was necessary.
All samples were acceptable with respect to the surrogate criteria used.
        As required by the QAPP, triplicate analyses of the Saginaw River untreated sediment (S-US-RCC)
were performed to assess precision.  These results are summarized in Table QA-8.  A matrix spike using
Aroclor 1254 was performed on the same sample to assess accuracy. These results are included in Table
QA-8. RSDs were outside specified control limits but within precision specified in Method 8080; data should
be of sufficient quality to support project results.
        As required  by the QAPP, a matrix spike and a  matrix spike duplicate (MS/MSD)  analysis was
performed for the treated Saginaw Rive sediment (S-TS-RCC). These results are presented in Table QA-9.
The RPD was outside control limits; however, no Aroclor 1254 was found in the sample and the data is not
impacted.  Matrix spike recoveries were acceptable. Due to the minimal amount of water generated by the
B.E.S.T. process, no QC analyses were performed on this matrix.
        The QAPP specified that triplicate analyses and a  matrix spike be performed on the Saginaw River
                                             150

-------
 oil residual (S-OR-RCC).  This sample was spiked but triplicate analyses were performed on the Indiana
 Harbor oil residual. These results are summarized in Tables QA-10 and QA-11, respectively.
        One standard reference material (SRM)  certified by  the National  Research Council of Canada
 (NRCC) for Aroclor 1254 was extracted and analyzed twice with the sediment samples. Recoveries of 82.8%
 and 78.6% were obtained.  The average of 80.7% fell within the 80-120% criteria specified in the QAPP.
        Method blanks were extracted and analyzed with each set of samples extracted. No PCBs were
 found in any method blanks.

 METALS
 Metals Procedure
        Sediments were prepared for metals analysis by freeze-drying, blending, and grinding. Sediments
 for As, Cl, Hg, and Se were digested using nitric and hydrofluoric acids.  The digestates were analyzed for
 As, Cd, and Se by graphite furnace atomic absorption (GFAA) by SW-846 Method 7000 series using Zeeman
 Background correction. The digestates were analyzed for mercury by cold vapor AA (CVAA) using SW-846
 Method 7470.
        Sediments for As, Ba,  Cr, Cu, Fe, Mn, Ni, Pb, and  Zn were analyzed by energy-diffusive X-Ray
 fluorescence (XRF) following the method of Sanders (1987). The XRF analysis was performed on a 0.5g
 aliquot of dried, ground sediment pressed into a pellet with a diameter of 2 cm.
 Metals QC Results and Discussion
       Triplicate analyses of the Saginaw River untreated sediment (S-US-RCC) and treated sediment (S-TS-
 RCC) were performed to assess precision. Matrix spikes were analyzed for the same samples to assess
 accuracy. Results are summarized in Tables QA-12 and QA-13.  It should be noted that the sediments were
 not spiked for XRF analysis.
       Accuracy  and precision results for metals were acceptable with  only a few minor exceptions, as
 shown in Tables QA-12 and QA-13. These exceptions have little, if any, impact on data quality and project
 results.
       One NIST certified standard reference material (SRM) was digested and/or analyzed twice with the
 sediment samples for XRF, GFAA, and CVAA analyses. These results are presented in Table QA-14.
       Method blanks were digested and analyzed for the metals analyzed by GFAA and CVAA.  (Method
 blanks are not applicable to XRF analysis). If analyte was detected in the method  blank, blank correction
was performed. Minimal amounts of some metals were detected; data quality is not affected.
                                                151

-------
                                     TABLE QA-5. PAH REPLICATE"0 RESULTS FOR I-OR-RCC ng/ml
en
ro
Compound
Napthalene
Acenaphthylene
Acenaphthene
Ruorene
Phenanthrene
Anthracene
Ruoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
lndeno(1 ,2,3,c,d)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Replicate 1
ppb
30000 U
46600
50000 U
48300
212000
108000
634000
608000
346000
481000
433000
290000
434000
363000
75100
220000
(a) Replicate results represent values
NC
U
Not Calculated
Undetected
Replicate 2
ppb
30000 U
52000
50100
57000
222000
117000
657000
627000
366000
504000
456000
297000
453000
378000
80600
224000
after correction for percent oil concentration.


Replicate 3
ppb
30000 U
45400
42400
50200
206000
104000
618000
590000
336000
467000
406000
279000
413000
344000
70600
207000



RSD Control Limits
NC Not Specified
7.4
NC
8.9
3.8
6.1
3.1
3.0
4.4
3.8
5.8
3.1
4.6
4.7
6.7
4.1




-------
                      TABLE QA-6. PAH SRM RESULTS
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Ruorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo (a)anthracene
Chrysene
Benzo (b)f luoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
lndeno(1 ,2,3,c.d)pyrene
Dibenzo(a,h)anthracene
Benzo (g,h,i)pery1ene
Ftecovery, % Control Limits
NC 80 - 120%
NC
NC
NC
95
81
91
96
87
NC
98
136 *
75 *
88
NC
82















NC  = Not Certified
Outside Control Limits
            TABLE QA-7. PCB SURROGATE RECOVERIES, PERCENT
Sample
S-US-RCC, Rep. 1
S-US-RCC, Rep. 2
S-US-RCC, Rep. 3
B-US-RCC
I-US-RCC
Method Blank
S-TS-RCC
B-TS-RCC
I-TS-RCC
Method Blank
S-WR-RCC
B-WR-RCC
I-WR-RCC
Method Blank
S-OR-RCC
B-OR-RCC
I-OR-RCC, Rep. 1
I-OR-RCC, Rep. 2
I-OR-RCC, Rep. 3
Method Blank
Tetrachloro-m-xylene
53
56
51
63
62
25 *
35 *
34 *
89
21 *
41
35 *
44
36 *
65
71
92
86
84
48
Octachloronaphthalene Control Limits
85 40-120
94
92
70
61
130 *
76 40-
96
54
76
124 * 40 -
109
98
107
100 40-
103
107
97
102
105





120



120



120





      Outside Control Limits
                                  153

-------
                                    TABLE QA-8.  PCB REPLICATE AND SPIKE RESULTS FOR S-US-RCC
Replicate 1 Replicate 2 Replicate 3
Aroclor ppb dry ppb dry ppb dry Mean
1242/1248
1254
1260
Precision
RSD Control Limits
20
20
20
Accuracy
% Recovery Control Limits
„
40 - 120
—
01
-p.
                                           TABLE QA-9. PCB MS/MSD RESULTS FOR S-TS-RCC
              PCB
MS Recovery
MSD Recovery
PRO
Accuracy Control
    Limits
  Precision
Control Limits
           Aroclor 1254
                                                              40-120%
                                                                                      20

-------
                         TABLE QA-10. PCB MS RESULT FOR S-OR-RCC
PCB
Aroclor 1254
MS Recovery

Control Limits
Not Specified
 OIL AND GREASE

 Oil and Grease Procedures
        Sediment samples were extracted with freon using Soxhlet extraction according to SW-846 Method
 9071.  The extract was analyzed for oil and grease by infra-red (IR) as outlined in Method 418.1  (Methods
 for Chemical Analysis of Water and Wastes, 1983).
 Oil and Grease QC Results and Discussion
        Both the untreated and treated Saginaw River sediments (S-UC-RCC and S-TS-RCC) were analyzed
 for oil  and grease in triplicate. In addition, matrix spike was performed for S-TS-RCC. These results are
 summarized in Table QA-15.  All QC results fell within  specified control limits.

 TOTAL VOLATILE SOLIDS
 Total Volatile Solid Procedures
        Sediments were  analyzed for total volatile solids (TVS) following the procedures in Method 160.4.
 Methods for Chemical Analysis of Water and Waste, 1983) modified for sediments.  An aliquot of sediment
 was dried and then ignited at 550°C.  The loss of weight on ignition was then determined.
 Total Volatile Solid QC Results and Discussion
        Both the untreated and treated Saginaw River sediments (S-US-RCC and S-TS-RCC) were analyzed
 for TVS in triplicate. Results are summarized in Table QA-16.  Both RSDs fell within specified control limits.

 AUDIT FINDINGS
        An audit of the Battelle-Marine Sciences Laboratory was conducted on September 25 and 26,1991.
 Participants included  EPA, GLNPO, and SAIC personnel. The path of a sample from receipt to reporting
was observed  specifically for samples from these bench-scale treatability tests.  Two  concerns  were
 identified in the  organic laboratory: 1) the preparation, storage, record-keeping, and replacement of
                                              155

-------
                            TABLE QA-11. PCB REPLICATE RESULTS FOR I-OR-RCC
Aroclor
1242/1248
1254
1260
Replicate 1 , ppb



Replicate 2, ppb



Replicate 3, ppb



Mean



RSD



Control Limits
20
20
20
                     TABLE QA-12.  METALS REPLICATE AND SPIKE RESULTS FOR S-US-RCC
Metal
Ag
As
Ba
Cd
Cr
Cu
Fe(1)
Hg
Mn
Ni
Pb
Se
Zn
Method
GFAA
XRF
XRF
GFAA
XRF
XRF
XRF
CVAA
XRF
XRF
XRF
GFAA
XRF
Replicate 1,
ppm dry
0.87
1.86
322
4.39
93
55.8
0.780
0.162
162
58.1
43.0
0.3 U
126
Replicate 2,
ppm dry
0.85
2.65
322
4.08
112
57.3
0.765
0.178
159
55.2
42.5
0.3 U
133
Replicate 3,
ppm dry
0.80
2.12
321
3.96
117
63.4
0.816
0.160
173
61.5
50.9
0.3 U
162
Mean













RSD













Precision
Control
Limits













%
Recovery













Accuracy
Control
Limits













*     Outside Control Limits        NS



(1)    Results in Percent for Fe      U
Not Spiked    NC



Undetected
Not Calculated

-------
                                  TABLE QA-13.  METALS REPLICATE AND SPIKE RESULTS FOR S-TS-RCC
Metal
Ag
As
Ba
Cd
Cr
Cu
Fe(1)
Hg
Mn
Ni
Pb
Se
Zn
Method
GFAA
XRF
XRF
GFAA
XRF
XRF
XRF
CVAA
XRF
XRF
XRF
GFAA
XRF
Replicate 1,
ppm dry
1.03
2.77
325
4.38
130
68.6
0.855
0.505
181
71.5
53.3
0.3 U
194
Replicate 2,
ppm dry
0.76
2.87
321
4.17
113
66.3
0.827
0.290
177
64.1
45.0
0.3 U
165
Replicate 3,
ppm dry
0.67
2.92
310
4.24
110
57.4
0.795
0.209
172
57.2
41.4
0.3 U
147
Mean
0.82
2.85
319
4.26
118
64.1
0.826
0.335
177
64.3
46.6
0.3 U
169
RSD
23
2.7
2.4
2.5
9.2
9.2
3.6
46*
2.6
11
13
NC
14
Precision
Control
Limits
20
I
I
I
I
I
I
I
I
I
I
I
I
%
Recovery
120*
NS
NS
100
NS
NS
NS
98
NS
NS
NS
101
NS
Accuracy
Control
Limits
85-115
—
—
85-115
—
—
—
85-115
—
—
—
85-115
™"~
en
            *      Outside Control Limits



            (1)    Results in Percent for Fe



            NS    =      Not Spiked



            U     =      Undetected



            NC    =      Not Calculated

-------
                       TABLE QA-14.  METALS SRM RESULTS, % RECOVERY
Metal
Ag
As
Ba
Cd
Cr
Cu
Fe
Hg
Mn
Ni
Pb
Se
Zn
SRM-1
NC
113
NC
117
85.5
124*
104
105
93.3
113
101
NC
96.7
SRM-2
NC
97.4
NC
117
103
110
100
105
90.4
97.2
99.6
NC
93.0
Control Limits
80-120%












        *       Outside control limits.
        NC     =      not certified.


standards is not well-documented; and 2) the nonstandard procedures used to extract, clean up and analyze
samples needs to be documented with reported data.
        During the audit, the use of nonstandard procedures was discussed. It was concluded that data
comparability within this project and within the ARCS program  should not be an issue, as the Battelle
laboratory has performed all analyses to date.  However, comparability to data generated outside the ARCS
program is not possible.
MODIFICATIONS AND DEVIATIONS FROM THE QAPP

        Laboratory activities significantly deviated from the approved QAPP in two areas-analytical
procedures and quality assurance (QA) objectives. Specific deviations and their effect on data quality are
discussed in this section.
                                              158

-------
en
CO
                     TABLE QA-15. OIL AND GREASE REPLICATES AND SPIKE RESULTS FOR S-US-RCC AND S-TS-RCC
Sample
S-US-RCC
S-TS-RCC
Replicate 1,
ppm dry
1480
297
Replicate 2,
ppm dry
1320
293
Replicate 3,
ppm dry
1260
206
Mean
1350
265
RSD
8.4
19
Precision
Control Limits
20
20
% Recovery
114
NS
Accuracy
Control Limits
80-120%
           NS
Not Spiked
                                  TABLE QA-16. TVS REPLICATES FOR S-US-RCC AND S-TS-RCC
Sample
S-US-RCC
S-TS-RCC
Replicate 1, % dry
2.24
1.76
Replicate 2, % dry
2.03
1.70
Replicate 3, % dry
2,01
1.74
Mean
2.09
1.73
RSD
6.1
1.8
Control Limits
20
20

-------
 ANALYTICAL PROCEDURES
        The Assessment and Remediation of Contaminated Sediments (ARCS) Program was initiated by the
 Great Lakes National Program Office (GLNPO) to conduct bench-scale and pilot-scale demonstrations for
 contaminated sediments. To date, all laboratory analyses performed in support of the ARCS Program have
 been done at the Battelle-Marine Sciences Laboratory (MSL) in Sequim, Washington. Standard procedures
 used by Battelle-MSL often do not follow those procedures identified in SW-846 and the QAPP.  While these
 nonstandard procedures yield results of acceptable quality, comparability with analyses performed outside
 the ARCS Program is not possible.
 PAH Analysis
 •       Samples were co-extracted with PCB samples using a modified SW-846 extraction procedure which
        entailed rolling  of the sample in methylene chloride and an additional clean-up step using high
        pressure liquid  chromatography (HPLC).  An internal standard, hexamethyl benzene, was added
        prior to this clean-up step to monitor losses through the HPLC.  Final results were corrected for the
        recovery of this internal standard.  A second internal standard, d12-phenanthrene, was added prior
        to analysis; however, no corrections were made based on its recovery. Neither of these internal
        standards are specified in Method 8270.
 •       SW-846 Method 8270 was modified to  quantify the samples using Selective Ion Monitoring (SIM)
        Gas Chromatography/Mass Spectrometry (GC/MS). This modification results in improved detection
        limits.
 •       Three radiolabelled PAH compounds were used as surrogates rather than those recommended in
        Method 8270., Recoveries of these compounds should better represent the recoveries of target
        PAHs.
 PCB Analysis
 •       Samples were extracted  using the modified extraction procedures  as described  for the PAH
        analysis.  An internal standard, dibromooctafluorobiphenyl, was added prior to the HPLC clean-up
        to monitor losses. Final results were corrected for the recovery of this standard. A second internal
        standard, 1,2,3-trichlorobenzene (required by QAPP)  was added prior to analysis;  however, no
        corrections were made based on its recovery.
•       Quantification of PCBs was not done on a total basis as required by SW-846 Method 8080 but by
        quantifying four peaks for each Aroclor and averaging these results.                   
-------
 •      The  surrogate required by the QAPP, tetrachloro-m-xylene,  was used.  A second  surrogate,
        octochloronaphthalene, was also added to monitor extraction efficiency.
 Metals Analysis
 •      Nine of the 13 metals analyzed for sediment samples were measured by energy-diffusive X-Ray
        fluorescence (XRF) - As, Ba, Cr, Cu, Fe, Mn, Ni, Pb, and Zn. This procedure yields a total metals
        concentration instead of the recoverable metals determined by SW-846 methods.
 •      Sediments for Ag, Cd, Hg, and Se were subjected to an acid digestion using nitric and hydrofluoric
        acids. This digestion again yields total rather than recoverable metals.
 Oil and Grease
 •      Oil and grease extracts for sediments were analyzed using infrared (IR) detection rather than the
        gravimetric procedures specified in the QAPP. This should have no effect on data quality.
 QUALITY ASSURANCE OBJECTIVES
        Many of the QA objectives and internal QC checks criteria specified in the QAPP (particularly for
 organic analyses) are not routinely achievable by standard or nonstandard methods.  To avoid  excessive
 reanalyses (both costly and time-consuming), acceptance criteria established internally by Battelle were used
for this project.  These internal limits are adequate for use in determining whether or not project results are
valid.
 PAH Analysis
 •       Both  surrogate and matrix spike objectives for PAHs were specified in  the QAPP to be 70-130%.
        For surrogates, Battelle actually used internal limits of 40-120%, with one  of the three surrogates out
        of limits being acceptable.  If more than one surrogate did not fall within 40-120%, reanalysis was
        required.  For matrix spikes, internal limits of 40-120% were also used; no reanalyses however, were
        performed based on exceedences of these limits.
•       Limits for continuing calibration checks were specified as  ±10% in the QAPP; limits of ± 25% were
        used.
PCS Analysis
•       Both surrogate and matrix spike objectives for PCBs were specified in the QAPP to be  70-130%.
        For surrogates, Battelle actually used internal limits of 40-120%. If both surrogates exceeded these
        limits, re-extraction was performed.  For matrix spikes, internal limits of 40-120% were also used;
        no reanalyses, however, were performed based on exceedences of these limits.
                                               161

-------
       Limits for continuing calibration checks were specified as ±10% in the QAPP; limrts of ±25% were
       used.
Metals Analysis
       Samples analyzed by XRF cannot be spiked.  Therefore, no measure of sample accuracy was
       obtained for those metals previously identified as being analyzed by XRF. An SRM was analyzed,
       providing a means to measure method accuracy for eight of the nine metals determined by XRF (all
       but Ba).
                                                                                     e.qarep.bgr

                                            162

-------
  Data  Verification Report For Assessment and
Remediation of Contaminated Sediment Program
                      Report Number 8
                  (SAIC, Bench-Scale Tests)
                             By

           M. J. Miah, M. T. Dillon, and N. F. D. O'Leary

      Lockheed Environmental Systems and Technologies Company
                    980 Kelly Johnson Drive
                    Las Vegas, Nevada 89119
                         Version 1.0
                   Work Assignment Manager
                     Brian A. Schumacher
              Exposure Assessment Research Division
           Environmental Monitoring Systems Laboratory
                    Las Vegas, Nevada 89193
           Environmental Monitoring Systems Laboratory
               Office of Research and Development
              U. S. Environmental Protection Agency
                    Las Vegas, Nevada 89193

-------
                               ABSTRACT

       Data  submitted  by the Science  Applications International Corporation
 (SAIC) of Cincinnati, Ohio, have been  verified for compliance of the QA/QC
 requirements of  the  Assessment and Remediation  of Contaminated Sediment
 (ARCS) program.  This data set includes results  from bench-scale technology
 demonstration  tests  on  wet contaminated  sediments  using  four  treatment
 technologies, namely, B.E.S.T.  (extraction process), RETEC (low temperature
 stripping), ZIMPRO (wet  air  oxidation), and  Soil Tech  (low temperature
 stripping). The primary contaminants in these sediments were polychlorinated
 biphenyls (PCBs) and polynuclear aromatic hydrocarbons (PAHs). In addition,
 metal contents and conventional (% moisture,  pH,  % total volatile solids, oil and
 grease, total organic carbon  (TOC), total cyanide, and total phosphorus) in these
 sediments were also considered for this project. The objective of the bench-scale
 technology  demonstration  study  was  to  evaluate  four different  treatment
 techniques for removing different organic contaminants from sediments.   Both
 treated and untreated sediment samples  were analyzed to determine  treatment
 efficiencies.

       A total of seven sediment samples from four different areas of concerns
 (Buffalo River,  Ashtabula River, Indiana Harbor,  and  Saginaw River)  were
 analyzed under the bench-scale technology demonstration project. The samples
 from these areas of concern  (AOCs) were collected by the Great Lakes National
 Program Office (GLNPO)  in Chicago,  IL, and  sample homogenization was
 performed by the U. S. EPA in Duluth,  MN.  SAIC was primarily responsible
 for the characterization of the sediment samples  prior to testing and for the
 residues created during the test.  The solid fraction analyses were performed by
 SAIC's analytical subcontractor Battelle-Marine Sciences Laboratory of Sequim,
 Washington, and Analytical  Resources Incorporated of Seattle, Washington.

       The submitted data sets represent analyses of untreated sediments, as well
 as solid, water, and  oil residues obtained by using different treatments. The
 verified data set is divided into several parameter groups by sampled media. The
 data verifications are presented in parameter groups that include:  metals, PCBs,
 conventionals, and PAHs.

       The results of the verified data are presented as  a combination  of an
 evaluation  (or  rating) number and any  appropriate data flags  that may be
 applicable. The templates used to assess  each  individual analyte are attached in
case the data user needs the verified data of a single parameter instead of a
parameter group.

-------
                                  INTRODUCTION
       The bench-scale technology  demonstration  project was  undertaken to evaluate  the
efficiencies of four techniques used for the removal of specific contaminants from wet sediments
collected from designated Great Lakes areas of concern.  Four  different sediment treatment
techniques, namely, B.E.S.T (Basic Extraction Sludge Technology), RETEC, ZIMPRO, and Soil
Tech were considered for evaluation.  B.E.S.T. is a solvent extraction process, RETEC and Soil
Tech are low temperature stripping techniques, and ZIMPRO is a wet air oxidation technique.
Wet sediments were collected by the Great Lakes National Program Office (GLNPO) from four
Great Lakes sites, namely, the Buffalo River in New York, the Saginaw River/Bay (referred to
as Saginaw River throughout  the following discussions)  in Michigan, the Grand Calumet
River/Indiana Harbor (referred to as Indiana Harbor throughout the following discussions) in
Indiana, and the Ashtabula River in Ohio. The four techniques were used to treat the sediment
samples from these four sites.   The sediment samples represent the sediment  that would be
obtained for on-site treatment.

       The B.E.S.T. process is a patented solvent extraction technology that uses the inverse
miscibility of triethylamine as a solvent.  At 65° F, triethylamine is completely soluble in water
and above this temperature, triethylamine and water are partially miscible.   This property of
inverse miscibility is used since cold triethylamine can simultaneously solvate oil and water.
RETEC and the Soil Tech (low temperature stripping) are techniques to separate volatile and
semivolatile contaminants from  soils, sediments, sludges and filter cakes. The low temperature
stripping (LTS) technology heats contaminated media to temperatures between 100 -2Qtf F,
evaporating off water and volatile organic contaminants. The resultant gas may be burned in
an afterburner and condensed to a reduced volume for disposal or can be  captured by carbon
absorption beds.   For these treatability studies, only the processes that capture  the driven off
contaminants were considered.  The ZIMPRO (wet air oxidation) process  accomplishes an
aqueous phase oxidation of organic  and inorganic  compounds at elevated  temperatures and
pressures.  The temperature range for this process is between 350 to 600" F (175 to 320* Q.
System pressure of 300 psi to well over 300 psi may be required.  In this process, air or pure
oxygen is used as an oxidizing agent.

       Samples for the technology demonstration projects were obtained by GLNPO (Chicago,
Illinois) and were analyzed by Battelle-Marine Sciences Laboratory (Battclle-MSL, Sequim, WA)
and  by Analytical Resources  Incorporated  (Seattle, WA).  To  evaluate the bench-scale
technologies, the  sample analyses  were divided  into four  parts:  (1) raw untreated  sediment
samples, (2) treated sediments, (3) water residues, and (4) oil residues. The amount of residues
available for the  analyses depended upon  the corresponding sediment samples  and on  the
individual technology used to treat those sediment samples.

       The analyses of sediment and residue parameters for these projects were divided into four
different categories: (1)  metals, including Ag, As, Ba, Cd, Cr, Cu, Fe, Hg, Mn, Ni, Pb, Se,
and Zn; (2) polychlorinated biphenyls (PCBs); (3) polynuclear aromatic hydrocarbons (PAHs);

-------
and (4) conventionals, including percent moisture, pH, percent total volatile, oil and grease, total
organic carbon  (TOC),  total cyanide, and  total phosphorus.   Analyses  of metals and
conventionals were performed on treated and untreated sediment samples only for B.E.S.T.,
ZIMPRO, and Soil Tech, while for the RETEC process, analyses of metals and conventionals
were performed on treated and untreated sediment samples as well as water residue samples.

       No oil residues were produced by the ZIMPRO technique (wet air oxidation treatment
technique), while in  the other three techniques, oil residues were analyzed after appropriate
sample cleanup steps for PCBs and PAHs.
       QUALITY ASSURANCE AND QUALITY CONTROL REQUIREMENTS
       The objective behind all quality assurance and quality control (QA/QC) requirements is
to ensure that all data satisfy predetermined data quality objectives.   These requirements are
dependent on the data collection process itself. Under the bench-scale technology demonstration
project, QA/QC requirements were established for:

    1.  Detection limits,
    2.  Precision,
    3.  Accuracy,
    4.  Blank analyses,
    5.  Surrogate and matrix spike analyses, and
    6.  Calibration
             a) initial
             b) ongoing.

       Four parameter groups analyzed in the sediment and water residue phases were of interest
in the bench-scale technology demonstration project. These groups included:  (a) metals, (b)
PCBs,  (c) PAHs, and (d) conventionals.  The conventionals included:  percent moisture, pH,
percent total volatile, oil and grease, TOC, total cyanide, and total phosphorus.  In addition,
total solids,  total suspended solids, and conductivity were included in the conventionals group
for RETEC conventional analyses. The analyses for metals and conventionals were performed
for solids only, except for RETEC, where metals and conventionals were analyzed in solid and
water residue phases. Parameter groups analyzed in the oil residue phase are PCBs and PAHs.
The objective of these analyses was to characterize samples both before and after each treatment
was applied.

       The detection limits for metals, PCBs, PAHs, and  conventionals (where appropriate)
were defined as, three times the standard deviation for  15 replicate analyses of a sample with
an analyte concentration within a factor of 10 above the expected or required limit of detection.
Individual parameter detection limits are presented in the approved quality assurance project plan
for SAIC on file at the Great  Lakes National Program Office in Chicago, IL.

-------
       Precision requirements were based on analytical triplicate analyses for all parameters of
sediment samples and treated residues, at the rate of 1  per 20 samples.   The results of the
triplicate analyses provided the precision for the analytical laboratory.  An acceptable limit was
the coefficient of variation less than or equal to 20 percent.  The precision requirement was
established for all variable types in this project.   For treated sediments, the relative percent
difference (RPD) between the matrix spike and matrix spike duplicate was  used as a measure
of precision with an acceptance limit of less than 20% .

       Accuracy was defined as the difference between the expected value of the experimental
observation and its "true" value.  Accuracy in this project was required to be assessed for each
variable type using analysis of certified reference materials, where available, at the rate of 1 per
20 samples. Acceptable results must agree within 20 percent of the certified range.  Since no
PCBs and PAHs were expected to be detected in the treated sediment, matrix spikes and matrix
spike duplicate analyses were required during the analyses of treated sediment for the organic
parameters.  Matrix  spike analyses were used as a  measure of accuracy for treated sediment
analyses, with an acceptance limit of ±30% from the  known value.

       Matrix spikes were required to be  used at a rate of 1 per 20 samples and to be within
plus or minus 15 percent of the spiking value for metals  and 70 to 130 percent of the spiking
value for organics (PCBs and PAHs).

       Surrogate spike analyses were only required for each sample in organic analyses. The
acceptable limits for the surrogate  recovery  was  between 70 and  130 percent of the known
concentration.

       The observed values  should have  been less than  the method detection limit for each
parameter for method blanks (run at the beginning,  middle, and end of each analytical run).

       The ongoing calibration checks were required at the beginning, middle, and end of a set
of sample analyses for all variable types.  The maximum acceptable difference was ±10% of
the known concentration value in the mid-calibration range. Initial calibration acceptance limits,
for metals, was the _> 0.97 coefficient of determination for the calibration curve, while a %RSD
of the response  factors of less than or equal to 25% was required for organics.

                            RESULTS AND DISCUSSION
       The ARCS QA program was formally adopted for use when SAIC received final approval
from the  GLNPO on May  31,  1991.   An evaluation scale, based  upon the QA program
developed for the ARCS program, was developed to evaluate the success of the data collection
process in meeting the QA/QC requirements of the ARCS program. The following section
discusses how to interpret the data verification results.

-------
The Verification Process and Evaluation Scale

       For verification purposes, the data set from each technology was divided into 4 different
sample media as follows:

                           1.  Untreated sediment,
                           2.  Treated sediment,
                           3.  Water residue, and
                           4.  Oil residue.

       The verification process included QA/QC compliance checking for accuracy, precision,
matrix spike analysis,  surrogate spike analysis,  blank  analysis,  detection limits, initial and
ongoing calibration checks, and holding times as well as checks on calculational correctness and
validity on a per parameter/analyte basis. Compliance checks were performed to ensure that the
QA/QC measurements and samples:  (a) met their specified acceptance limits; (b) had reported
results that were supported by the raw data; and (c) were analyzed following good laboratory
practices, where checking was possible. Upon completion of the verification process, a final
rating was assigned for each of the individual categories.  The final ratings are presented as a
combination of a number value and a flag list.

       The numerical value for the rating of a given parameter was assigned based upon the
successful completion of each required QA/QC  sample or measurement.  The QA/QC samples
were broken down into four different sample groups, namely, accuracy,  precision, blanks, and
spike recoveries.   A fifth category was included for  QA/QC measurements to address the
successful completion of instrument calibrations  (both initial and ongoing) and the determination
of method detection limits.  If the laboratory  successfully met the acceptance criteria of 50
percent or more of the parameters in a given QA/QC sample group, then the laboratory received
the full value for that category.  For example, if 50 percent or more of the reagent blanks for
the metals in sediment analyses had measured  values below the method detection limit, then
three points were  awarded  for that category, assuming reagent blanks were the only blank
samples analyzed by the laboratory.  The individual point values for each QA/QC sample type
or measurement and the minimum acceptance levels for each category are presented in Appendix
B.  The final numerical rating presented for each parameter category is the summation of the
point values from each of the five categories.

       Along with each numerical rating, a list of appropriate flags has been attached to the final
rating value (Appendix C). The flag indicates where discrepancies  exist between the laboratory
data and the acceptance limits of the required QA program.  Different flags are presented for
each category of QA sample (accuracy, precision, blanks, and spike recoveries) and for the
QA/QC measurements (instrument calibration and detection limit determination). The flags have
a letter and subscript configuration, such as A,.  The letter of the  flag represents the category
of the discrepancy while the subscript designates the form of the discrepancy.  For example, the
A flags indicate discrepancies  in the use of accuracy  checking  samples, such  as reference
materials or standards.  A flag with a subscript of 1  indicates that the laboratory failed to meet

-------
the acceptance criteria. Using the example of the A, flag, this flag would then indicate a failure
of the laboratory to meet the QA/QC requirements for the use of reference materials in their
appraisal of accuracy.  A flag with the subscript 0 indicates that no information was received
(or no standards were available in the case of accuracy) from the analytical laboratory, and
therefore, no points could be allotted towards the final calculated rating value for that particular
category.   It should be noted that the 0 flag does not necessarily indicate that the analytical
laboratory did not perform the QA/QC analyses, only that no information was received from the
laboratory.

       The subscript 9 flag indicates that the sample  category or QA/QC  measurement is not
applicable to that particular parameter or parameter group (Appendix C).  For example, an S,
flag indicates that a matrix spike for that given parameter or analyte is not applicable,  such as
was the case for percent moisture.  Where subscript 9 flags occur, an adjustment to the passing
and maximum scores (to be discussed) for a parameter group was made and will be reported in
the appropriate  tables.

       A complete presentation of the QA/QC rating factors (point values by sample type) and
the various data flags  and their subscripts are presented in Appendices B and C,  respectively.
A more complete discussion of the rating scale can be found in the report submitted to the
RA/M  workgroup by Schumacher and  Conkling entitled,  "User's Guide to the  Quality
Assurance/Quality Control Evaluation Scale of Historical Data Sets."

       Individual parameter flags are presented in the templates found in  Appendix D. The
objective of the presentation of the individual  flag templates is to help the data user  make a
determination regarding the useability of the data set for any given purpose and to provide the
data user with a means to assess any individual parameter that may be of specific interest

The Interpretation and Use of the Final Verified Data Rating Values

       The data verification scale was developed to allow for the proper rating of the verified
data  and  the  subsequent  interpretation  and  evaluation  of the ratings.   Two  different
interpretations can be made using the ratings provided in this report, namely, the actual or "true"
rating and  the potential rating.  The  first interpretation is based upon the formal ARCS QA
program, while the second interpretation scale is based upon the  "full potential"  value of the
submitted data set. In the following sections, each interpretation of the results will be discussed.

Data Interpretation Based  upon the Formal ARCS OA Program

       For each of the four parameter categories,  the data were initially verified for QA/QC
compliance following the requirements specified in the  signed QAPP submitted by SAIC and the
ARCS QAMP on file at the GLNPO in Chicago, Illinois.

       Table 1  provides the verified data ratings for each variable class for the four different
technologies studied based on the current  ARCS QA program.  The ratings of these variable

-------
 classes are presented to provide the data user with  a means for comparing the ARCS QA
 program-based verified results with other data sets, using the same or similar parameters, that
 were generated prior to and after the initiation of the  formal ARCS  QA program.

       Table 2 provides the data user with the full compliance and acceptable scores presented
 for each parameter group based upon the current ARCS QA program. The full compliance score
 represents the numerical rating value if all required QA/QC samples and measurements were
 performed by the analytical laboratory and successfully met all the QA/QC requirements of the
 ARCS QA program. An acceptable score is lower than the full compliance score and accounts
 for laboratory error that can be reasonably expected  during an analysis of multiple samples.
 Any final rating value less than the acceptable score indicates that problems were identified in
 the data that could adversely effect the quality of the data.  The acceptable score was set at 60
 percent of the full  compliance  score.  To determine  the percentage of QA/QC samples and
 measurements successfully analyzed for a given parameter versus the number analyzed following
 the complete ARCS QA protocols, divide the numerical rating received by the full compliance
 score.  An acceptable data set, in this case, has a rating of 60 percent or greater.

       In some cases, all  the QA/QC requirements may not be applicable (e.g., matrix spikes
 for percent solids are not applicable). If this is the case, a  flag with the subscript 9 was used,
 and the full compliance and acceptable scores were adjusted by lowering the score on appropriate
 number of points for nonrequired sample type, as identified in Appendix B. An example of this
 situation is % moisture,  as indicated in  Table 1, the subscript  9  flag has  been applied to
 accuracy, blank, detection  limit, and  spike samples.  Therefore,  the full  compliance and
 acceptable scores (Table 2) are only based upon the possible points for the successful completion
 of the remaining QA/QC samples that have cumulative points value of 8 (Appendix B).
Data Interpretation Based upon the" Potential' Value of the Data Set

       A second interpretation scale has been presented to allow the data user to establish the
"full potential" value of the submitted data set.   The numerical value and associated flags
presented in the first interpretation can be considered as an absolute rating  for that data set or
parameter.  These ratings were based upon all the data submitted to Environmental Monitoring
Systems Laboratory - Las Vegas (EMSL-LV) and to Lockheed for review by the analytical
laboratory. If one or more parameter or parameter groups qualifying flags had the subscript of
5, 6, 9, or 0 (Appendix C), the required information was not available or not applicable at the
time of sample analysis, and consequently was not included during the data verification and
review process. The equivalent point value(s) for each individual sample type may be added to
the reported point sum to give the data user the full potential value of the data set.  This process
assumes that if the "missing" QA/QC samples or measurements were performed,  the results
would fall within the ARCS QA program specified acceptance limits.  For example, if the point
value (including qualifying flags) for the metals was  6-Bo  Q, D0 So, then the data  user could
potentially add 14 points to the score since the blank analyses, spike information, detection limit,

-------
and calibration (initial and ongoing) information was not available for verification. The resulting
data would then have a rating of 20.
      TABLE 1. Verified Data Ratings Based on the Current ARCS QA Program
Untreated
Sediments
Metals
% Moisture
pH
%TVS
Oil and grease
TOC
Total cyanide
Total
phosphorus
PCBs
PAHs
B.E.S.T.
12-CoD0
0-A,B,CoD,P0S,
0-A,B,CoD,P0S,
6-A,CoD,S,
15-A, C,
12-C6P0S9
14-AoPo
14-AoP,
17-B,D0
17-D0 S,
ZBVfPRO
12-CoD0
3-A,B,CoD,S,
0-A,B,CoD,PoS,
3-A,BoCoD,S,
6-A, B, C« D, S,
12-C6P0S,
14-AoP0
14-Ao P0
14-A, B2 D0
11-8,005,5,
Soil Tech
12-Q D0
0-A,B,QD,P0S,
0-A,B,C0D,P0S,
6-A, Co 0,5,
6-A, B, C. D, S0
12-C6 P0 5,
H-AoP0S0
14-Ao P0
14-A, B, DO
17-D0 5,
RETEC
12-QDo
3-A,B,CaD,S,
3-A, B, Co D, 5,
6-A, C, D, S,
9-A, De C, So
9-C6D0P,S,
8-AoD0P,So
11-AoD.So
H-A.BjDoS,
20-D0
Treated
Sediments
Metals
% Moisture
pH
%TVS
Oil and grease
TOC
Total cyanide
Total
phosphorus
PCBs
PAHs
12-CoD,
0-A9B9C0D,P0S9
0-A,B,CoD,PoS,
6-A, 0, 0,5,
15-A,C«
12-C6P0S,
14-AoPo
14-AoPo
14-BjDoP,
14-D0P,S,
12-CoDo
0-A,B,CoD,P0S,
3-A,B,CoD,S,
3-A,BoCoD,S,
6-A, B, C6 D, 5,
12-C, P0 5,
14-Ao P0
14-Ao P0
H-A.BjDoP,
17-D0S,
12-C, D0
3-A, B, Co D, 5,
0-A,B,CoD,PoS,
6-A,CoD,S,
9-A, B, C. D,
12-C4P0S,
14-Ao P0
14-Ao P0
14-B,D0P,
14-DoP.S,
12-C.D.
3-A,B,C,D,S,
3-A,B,C.D,S,
6-A,C,D,S,
6-A,C,D,P,So
12-C.D.S,
11-AoD.Po
14-Ao D.
14-A, B, D0
20-D,

-------
TABLE 1. Verified Data Rating Based on the Current ARCS Program
(Continued)
Water
residue
Metals
% Moisture
PH
Total
Suspended
Solids
%TVS
Total Solids
Oil and grease
TOC
Total cyanide
Total
phosphorus
Conductivity
PCBs
PAHs
**
**
»*
**
**

**
**
**
**
**
14-B2 D0 P0
11-AoDoP.S,
«*
**
««
**
**

**
**
*«
**
**
14-Bj D0 P0
17-D0 S,
**
**
«»
**
**

**
*«
**
**
**
S-A^DoPoS,
s.
17-D0 P0
20
»»
3-A,B,Q,D,S9
e-A.CoD.S,
6- A, Co D9 S,
6-A,CoD,S,
12-A, C, D0
9-AoC4D0S,
14-A,D0
14-AoD0
9-AoC.D,S,
5-AoBjDoPoS,
s.
ll-A.DjP.S,
Oil residue
PCBs
PAHs
H-A.BjDoS,
H-AoBjD0S,
*
*
17-B, Do
14-Bj Do Sj
11-B2D0P0S3
17-BjDo
*   No oil residue was produced by this treatment
**  Analyses were not conducted for this treatment

-------
TABLE 2.  Full Compliance and Acceptable Scores Based on the Current ARCS QA Program
Variable Class
Metals in Treated Sediment
Metals in Untreated Sediment
% Moisture
pH
%TVS
Oil and grease
TOC
Total cyanide
Total phosphorus
Conductivity
Suspended Solids
Total Solids
PAHs
PCBs
Full Compliance
20
20
8
8
9
17
17
20
20
14
9
9
23
23
Acceptable
12
12
5
5
6
11
11
12
12
9
6
6
14
14
       Table 3 presents the verified data ratings for each variable class in the four technologies
based on their full potential value.  All data qualifying flags with the subscripts 5, 6, 9, or 0
have been removed. The appropriate point values for each of the 5, 6, or 0 flags (Appendices
B and C) were added to the final rating scores for  each parameter or parameter group.  In
contrast, the removal of the subscript 9 flags resulted in an adjustment to the full compliance and
acceptable scores, and noj in an addition to the calculated point scores since these analyses were
not applicable to the methodologies used by the laboratory (Table 2).

-------
                                                                      10
TABLE 3. Verified Data Ratings Based on the Full Potential of the Data set
Untreated
Sediments
Metals
% Moisture
pH
%TVS
Oil and grease
TOC
Total cyanide
Total phosphorus
PCBs
PAHs
B.E.S
-------
                                                                                      11
TABLE 3. Verified Data Ratings Based on the Full Potential of the Data set
(continued)
Water
residue
Metals
% Moisture
pH
%TVS
Oil and grease
TOC
Total cyanide
Total phosphorus
Conductivity
Suspended Solids
Total Solids
PCBs
PAHs
**
**
**
**
*«
**
**
**
**
**
«*
20-Bj
17-P.S,
**
**
**
**
*«
**
**
««
»*
**
**
20-Bj
20-83
**
**
*«
**
«*
**
**
**
«*
«*
**
14-A.BjS,
23
20
8
8
6
17
17
20
20
14
6
6
20-Bj
14-A, P, S,
Oil residue
PCBs
PAHs
14-A, Bj S,
17-8,5,
*
*
20-B,
17-8,5,
20-82
20-Bj
       * No oil residue was produced by this treatment
       **  Analyses were not conducted for this treatment

       To evaluate  the data using the values presented in Table 3, the final ratings should be
compared to the full compliance and  acceptable scores presented in Table 2.  The data user
should bear in mind that these values are only the potential values of the data set and assumes
that  the "missing"  QA/QC data could have been or were performed  successfully by the
laboratory. Any value falling below the acceptable value presented in Table 2 clearly indicates
that major  QA/QC violations were identified and the data should be used with  a great deal of
caution by  the data  user.

-------
                                                                                   12
Data Verification Results for Bench-scale Technology Demonstration Project

B.E.S.T.

       The B.E.S.T. technology was evaluated by analyzing sediment samples and their treated
residues (treated sediments, water residues, and oil residues) for metals,  conventionals, PCBs
and PAHs. PCB and PAH analyses were performed for sediments, water, and oil residues.  The
metals and conventional analyses were performed for the sediment samples only.

       In the majority of the cases studied, the accuracy objective was satisfactory for the metal
analyses  in treated and  untreated sediments.   Of  the thirteen  metals  analyzed, accuracy
information was not available for Ba, Se, and Ag. In both treated and untreated sediments, ten
of the thirteen metal analyses (As, Cd, Cr, Cu, Fe, Pb, Mn,  Hg, Ni, Pb,  and Zn) satisfied
ARCS specified QA/QC requirements  for accuracy.  Four of the thirteen metal analyses (Cd,
Hg, Se, and Ag) satisfied QA/QC requirements  for blank  analyses, while the remaining nine
metals (As, Ba, Cr, Cu, Fe, Mn, Ni, Pb, and Zn) were analyzed by XRF  techniques. In all of
the XRF analyses, results from blank  sample analyses were not applicable.  Both initial and
ongoing calibration for Cd,  Hg, Se, and Ag analyses  met the ARCS QA/QC specifications for
both treated and untreated sediments, while for the remaining nine metals (As, Ba, Cr, Cu, Fe,
Mn, Ni, Pb, and Zn) calibration information was not available. Detection limits information for
metal analyses in treated and untreated sediments were not  available for verification except for
Cd, Hg, Se, and Ag where detection limits were satisfactory.  The precision information for the
metal analyses in  treated sediment  was not available for Se, but  was  satisfactory  for the
remaining  elements, with the exception of Hg,  where precision  information did not satisfy
QA/QC requirements.  The precision information for the metal analyses in untreated sediment
was not available for Se,  but was satisfactory  for the remaining twelve metal  (Ag, As, Ba, Cd,
Cr, Cu, Fe, Hg, Mn, Ni, Pb, and Zn) analyses. The matrix spike information for both treated
and untreated sediment analyses were satisfactory for Cd, Hg, and Se,  were  unsatisfactory for
Ag, while the remaining nine metals (As, Ba, Cr, Cu, Fe, Mn,  Ni, Pb,  and Zn) were analyzed
by XRF techniques.  In all of the XRF analyses, results from  matrix spike analyses were not
applicable.

       Of  the  seven conventional analyses,  the accuracy information  in  both  treated  and
untreated sediments was satisfactory for TOC and was not available for  total cyanide, and total
phosphorus. In the remaining four conventional analyses, accuracy was  not applicable.  In both
sediments, five of the seven conventionals (%TVS, oil and grease, TOC, total  cyanide, and total
phosphorus) satisfied QA/QC requirements for blank analyses,  and the blank information was
not applicable for moisture, pH, and TVS. Both initial and ongoing calibration information was
satisfactory for all conventional analyses in both treated and untreated sediments except for
moisture and pH where calibration information was not available and for TOC and oil and grease
where ongoing calibration information was not available. Detection limits were satisfactory for
four (oil and  grease,  TOC, total cyanide, and total phosphorus) of the seven conventional

-------
                                                                                     13

analyses in treated and untreated sediments, and were not applicable for moisture, pH, and TVS.
The precision  information was satisfactory for two  (%TVS, oil and grease) of the seven
conventional analyses in treated and  untreated sediments.   No precision  information was
available for the remaining five conventional analyses in treated or untreated sediments.  The
matrix spike information for both treated and untreated sediment analyses were  satisfactory for
oil and grease, total cyanide, and total phosphorus, while for the remaining four conventional
analyses the matrix spike information was not applicable.

       In treated sediments, untreated sediments, and water residues, the accuracy objective for
PCBs was satisfactory for Aroclor 1254 analyses only and could be used to represent the whole
PCB group.  No accuracy information was available for the remaining three Aroclor analyses.
In oil residues, accuracy information was not satisfactory for PCB analyses.  In  both sediments
and in both residues, PCB analyses did not satisfy ARCS  specified QA/QC requirements for
blank  analyses indicating  potential  contamination at  the laboratory.   Initial and  ongoing
calibration was satisfactory for all PCB analyses in both treated and untreated sediments as well
as in water and oil residues.  Detection limit information were not available for PCB  analyses
in treated and untreated sediments and  for water and oil residues.  In  the untreated sediments,
the  precision information was satisfactory  for  Aroclors  1242  and  1254,  and no precision
information was available for Aroclors  1248 and 1260.  In  the treated  sediments, the precision
information was not satisfactory for Aroclor 1254, and no precision information was available
for  Aroclors 1242, 1248, and 1260.  In water residues, no precision information was available
for  any of the Aroclors.  In oil residues, the precision information was satisfactory for Aroclor
1248,  and no precision information was available for Aroclors  1242, 1254, and 1260.  The
matrix spike for Aroclor 1254 was satisfactory for both sediment and water residue analyses and
could be used  to represent the whole  PCB  group.  The matrix spike for Aroclor 1254 was
unsatisfactory for the analyses of  oil residue.  In both sediment or residue analyses, no matrix
spike information was  available  for Aroclors  1242, 1248, and  1260.   The surrogate spike
recoveries were satisfactory for PCB analyses in both sediments and residues.

       In eight of sixteen PAH  analyses of treated  and  untreated sediments, the accuracy
objective was satisfactory.  No accuracy information was available for six PAHs (naphthalene,
acenaphthylene, acenaphthene, fluorene, chrysene, and dibenzo(a,h)anthracene) analyses in both
treated and untreated sediments.   The accuracy objective was  not satisfactory for benzo(k)
fiuoranthene and benzo(a)pyrene in treated or untreated sediments. No accuracy information was
available for any  of the PAH analyses in  water and  oil residues.  In treated and untreated
sediments, and in water residues, PAH  analyses satisfied ARCS specified QA/QC requirements
for blank analyses. In all cases of oil residues, the blank analyses exceeded the MDL indicating
potential contamination at the laboratory. Initial and ongoing calibration limits for PAH analyses
met the ARCS QA/QC specifications for both treated and untreated sediments and water and oil
residue analyses.  Detection limit information was not available for PAH analyses in treated and
untreated sediments, nor for water and oil residues.  In untreated sediments and oil residues, die
precision information was satisfactory for all PAH analyses, except for acenaphthene in untreated
sediment, and  naphthalene in oil  residues where no precision information was available.  In
treated sediments, the precision information was satisfactory for fluorene, phenanthrene, and

-------
                                                                                    14

anthracene but was unsatisfactory  for the remaining PAH analyses.  In water residues, no
precision information was available for PAH analyses except for benzo(g,h,i)pyrene where
precision was unsatisfactory.  The matrix spike information was satisfactory for twelve of sixteen
PAH analyses in treated sediment and for eight of the sixteen analyses in untreated sediment and
in water and  oil residues.   Surrogate recoveries were  not satisfactory for PAHs in either
sediment and residue analyses.
ZIMPRO

       The  ZIMPRO technology  was evaluated by analyzing  sediment  samples,  treated
sediments,  and water residues for metals, conventionals, PCBs, and PAHs. PCB and PAH
analyses  were performed for both sediment and water residues.  The metals and conventional
analyses  were performed for the  both sediment samples only.

       In the majority of the cases studied, the accuracy objective was satisfactory for the metal
analyses  in treated and untreated sediments.   Of  the thirteen  metals analyzed,  accuracy
information was not available for Ba, Se, and Ag. In both treated and untreated sediments, ten
of the thirteen metal analyses (As, Cd, Cr, Cu, Fe, Pb, Mn, Hg, Ni, and Zn) satisfied ARCS
specified QA/QC requirements for accuracy.  Four of the thirteen metal analyses (Cd, Hg, Se,
and Ag) satisfied QA/QC requirements for blank analyses, while the remaining nine metals (As,
Ba, Cr, Cu, Fe, Mn,  Ni, Pb, and Zn) were analyzed by XRF techniques.  In all of the XRF
analyses, blank sample analyses are not applicable.  Both initial and ongoing calibration for Cd,
Hg, Se, and Ag analyses met the ARCS QA/QC specifications for both treated  and untreated
sediments while for the remaining nine metals (As,  Ba, Cr, Cu,  Fe, Mn,  Ni,  Pb, and  Zn),
calibration  information was not available.  Detection limit information  for metal analyses in
treated and untreated sediments was not available for verification except for Cd, Hg, Se, and
Ag where the detection limits were satisfactory.  The precision for the metal analyses in treated
sediment was  not satisfactory for As, but was satisfactory for the remaining elements.  The
precision information for the  metal analyses in  untreated .sediment was satisfactory  for all
elements. The matrix spike information for both treated and untreated sediment analyses were
satisfactory for four (Cd, Hg,  Se, and Ag) of the thirteen elements while the remaining nine
metals (As, Ba, Cr, Cu,  Fe, Mn,  Ni, Pb, and Zn) were analyzed by XRF techniques.  In all of
the XRF analyses, results from matrix spike analyses were not applicable.

       Of the  seven  conventional analyses, the accuracy  information in  both treated and
untreated sediments was satisfactory for TOC and was not available for total cyanide,  and total
phosphorus. In the remaining four conventional analyses, accuracy was not applicable. In both
sediments, three of the seven conventionals (TOC, total cyanide, and total phosphorus) satisfied
QA/QC requirements for blank analyses.  The blank information was unsatisfactory for oil and
grease, was not available for %TVS, and the blank information was not applicable for moisture
and pH.  Both initial and ongoing calibration information was satisfactory for all conventional
analyses  in both treated  and untreated sediments except for % moisture, pH, and TVS where
calibration  information was not available, and for TOC and oil and  grease, where ongoing

-------
                                                                                     15

calibration information was not available.  Detection limits were satisfactory for three (TOC,
total cyanide, and total phosphorus) of the seven conventional analyses in treated and untreated
sediments.  Detection  limits were unsatisfactory for oil and grease analyses  in treated and
untreated  sediments and were not applicable for % moisture, pH, and %TVS.  The precision
information was satisfactory for pH, %TVS, and oil and grease analyses in treated, and for
%moisture, %TVS, and oil and grease analyses in untreated sediment. No precision information
was  available  for  % moisture,  TOC, total cyanide,  and total phosphorus analyses in treated
sediment and for pH, TOC, total cyanide, and total phosphorus analyses in untreated sediments.
The  matrix spike information for both treated and untreated sediment analyses were satisfactory
for total cyanide and total  phosphorus, were unsatisfactory for oil and grease while for the
remaining four conventional analyses the matrix spike information was not applicable.

       The accuracy objective was unsatisfactory for the PCB analyses in treated and untreated
sediments for  Aroclor  1254.  No  accuracy information was available for the remaining  three
Aroclor analyses in treated and untreated sediments.  In water residue, the accuracy objective
for PCBs was satisfactory for Aroclor 1254  analyses only and could be used to represent the
whole PCB group.  No accuracy  information was available for the remaining three Aroclor
analyses in water residues.  In water residues and in both treated and untreated sediments, the
blank  analyses exceeded  the  detection limits  specified in  the QAPP indicating potential
contamination at the laboratory.  Initial and  ongoing calibration was satisfactory for all PCB
analyses in both treated and untreated sediments as well as in water  residues.  Detection limits
information were not available for PCB analyses in treated and untreated sediments, nor in the
water  residues.   In untreated sediment analyses, most PCB observations were below the
instrument detection limits, therefore it was not possible to calculate  meaningful  precision
information for PCB Aroclors, with the exception of Aroclor 1248 analyses, where precision
information satisfied QA/QC requirements.   No precision information was available for PCB
analyses in treated sediments,  except  for Aroclor 1254 in treated sediment where it did not
satisfy QA/QC requirements. In the water residue, no PCB precision information was available.
The matrix spike for Aroclor 1254 was satisfactory for both sediments,  and the water residue
analyses and could be used to represent the whole PCB group.  The matrix spike information
for sediments  and water residue analyses for  Aroclor 1242, 1248, and 1260 were not available
for verification.  The surrogate recoveries were satisfactory for PCB analyses in sediment and
residue analyses.

       In ten  of the sixteen PAH analyses  in treated sediment  and nine of the sixteen  PAH
analyses  in untreated  sediments,  the accuracy objective  was satisfactory.    No  accuracy
information was available for six PAHs (naphthalene, acenaphthylene, acenaphthene, fluorene,
chrysene, and dibenzo(a,h)anthracene) analyses in treated and untreated sediment. The accuracy
objective was not satisfactory for  benzo(k)fluoranthene in  untreated  sediment.   Accuracy
information in water residue was unsatisfactory for naphthalene, acenaphthylene, acenaphthene,
phenanthrene, and benzo(a)pyrene. Accuracy was satisfactory for the rest of the PAH analyses
in water  residues.  In treated sediments  and water residues, PAH analyses satisfied ARCS
specified QA/QC requirements for blank analyses. In all cases of untreated sediment analyses,
the blank analyses exceeded the detection limit  specified in the QAPP.  Calibration limits for

-------
                                                                                     16

 PAH analyses met the ARCS QA/QC specifications for both treated and untreated sediments,
 and also for water residue analyses.  Detection limits information were not available for PAH
 analyses  in treated  and  untreated sediments, nor for the water  residues.   The precision
 information was satisfactory  for  PAH analyses  in both  sediments except  for naphthalene,
 acenaphthylene, acenaphthene, fluorene, and benzo(a)pyrene analyses  in treated sediment and
 for  naphthalene, acenaphthene, phenanthrene, and benzo(a)pyrene in water residue,  where
 precision was unsatisfactory.  The matrix spike information was satisfactory  for fifteen of the
 sixteen PAH analyses in treated sediment, for five of the sixteen analyses in untreated sediment
 and for eleven  of the sixteen analyses in  water residues.   Surrogate recoveries were not
 satisfactory for PAHs in  the sediment and residue analyses.
 SOIL TECH

       The Soil Tech technology was evaluated by analyzing sediment samples and their treated
 residues (treated sediments, water residues, and oil residues) for metals, conventionals, PCBs,
 and PAHs. PCB and PAH analyses were performed for sediment and residues. The metals and
 conventional analyses were performed for the sediment samples only.

       In the majority of the cases studied, the accuracy objective was satisfactory for the metal
 analyses  in  treated  and  untreated sediments. Of the  thirteen  metals  analyzed,  accuracy
 information was not available for Ba, Se, and Ag.  Four of the thirteen metal analyses (Cd, Hg,
 Se, and Ag) satisfied QA/QC requirements for blank analyses, while the remaining nine metals
 (As,  Ba, Cr, Cu, Fe, Mn, Ni, Pb, and Zn) were analyzed by XRF techniques.  In all of the
 XRF analyses, blank sample analyses are not applicable.  Both initial and ongoing calibration
 for Cd, Hg,  Se, and Ag analyses met the ARCS QA/QC specifications for  both treated and
 untreated sediments while for the remaining nine metals (As, Ba, Cr, Cu, Fe,  Mn, Ni, Pb, and
 Zn), calibration information was not available.  Detection limits information for metal analyses
 in treated and untreated sediments were not available for verification except for Cd, Hg, Se, and
 Ag where detection limits were satisfactory.  The precision information for the metal analyses
 in treated sediment was  not available for Se and Hg  but was  satisfactory for the remaining
 elements with the exception of Cr, where precision information  did not satisfy the QA/QC
 requirements.  The precision  information for  the metal analyses in  untreated sediment was
 satisfactory for all metal analyses. The matrix spike information were satisfactory for four (Cd,
 Hg, Se, and Ag) of the thirteen elements for treated sediments and two (Cd, Hg) of the thirteen
 elements for untreated sediments.  The matrix spike information were unsatisfactory for Se and
 Ag analyses in untreated sediments.  The remaining nine metals (As, Ba, Cr, Cu, Fe, Mn, Ni,
 Pb, and Zn) were analyzed by XRF techniques.  In all of the XRF analyses, results from matrix
 spike analyses were not applicable.

       Of the  seven conventional analyses, the  accuracy  information in both  treated and
 untreated sediments was satisfactory for TOC and was not available for total cyanide, and total
phosphorus. In the remaining four conventional analyses, accuracy was not applicable.  In both
 sediments, four of the seven conventionals (%TVS, TOC, total cyanide, and total phosphorus)

-------
                                                                                     17

satisfied QA/QC requirements for blank analyses, and the blank information was not applicable
for moisture and pH, while blank analyses was not satisfactory for oil and grease.  Both initial
and ongoing calibration information was satisfactory for all conventional analyses in both treated
and untreated sediments, except for % moisture, pH, and  %TVS where calibration information
was not available.   Ongoing calibration information was not available for TOC and oil and
grease.  Detection limits were satisfactory for three (TOC, total cyanide, and total phosphorus)
of the seven conventional analyses in treated and untreated sediments.  Detection limits were
unsatisfactory for oil and grease and were not applicable for %moisture, pH, and  %TVS. The
precision information was satisfactory  for % moisture, %TVS,  and oil and grease in treated
sediments.  The precision information was satisfactory for %TVS, and oil and grease in treated
sediments.  No precision information was available for the remaining conventional analyses in
treated or untreated sediments.  The matrix spike information were satisfactory for oil and
grease, total phosphorus, and total cyanide in treated sediment analyses and for total phosphorus
in untreated sediment analyses.  The matrix spike information were not available for oil and
grease  and total cyanide in  untreated sediment analyses.   While  for  the  remaining four
conventional analyses, the matrix spike information was  not applicable.

       The accuracy objective was satisfactory for the PCB analyses in treated sediments and
in oil  residue analyses for Aroclor 1254 only and could  be used to represent the whole PCB
group. The accuracy objective was unsatisfactory for the PCB analyses  in untreated sediments
and in water residue analyses for Aroclor 1254.  No accuracy information was available for the
remaining three Aroclor analyses in sediment or residue analyses. In both residues and in both
treated and untreated sediments, the blank analyses exceeded the detection limits specified in the
QAPP, except for Aroclor 1260 in oil residue. Initial and ongoing calibration was satisfactory
for all PCB analyses in both treated and untreated sediments,  as well as in both water and oil
residues. Detection limit information was not available for PCB analyses in both sediments and
residues.  In untreated sediment analyses,  most PCB observations were below the instrument
detection limits, therefore, it was not possible to calculate meaningful precision information for
PCB Aroclors,  with the  exception of Aroclor  1248 analyses, where  precision information
satisfied QA/QC requirements.  No precision information was available for PCB analyses in
treated sediment, except for Aroclor 1254, where it did not satisfy QA/QC requirements. No
precision information  was available  for PCB analyses in oil and water residues, except for
Aroclor 1248 in oil residue, where precision was satisfactory.  The matrix spike for Aroclor
1254 was satisfactory for both sediments  and the oil residue analyses and could be used to
represent the whole PCB group.  The matrix spike for Aroclor 1254 was unsatisfactory for the
water residue analyses, and the matrix spike information for both sediment and residue analyses
for Aroclor 1242,1248, and 1260 were not available for verification. The surrogate recoveries
were satisfactory for PCB analyses in sediment and residue analyses, except for water residue
where surrogate information was not available.

        In eight of sixteen PAH analyses in treated  and untreated sediments, the accuracy
objective was satisfactory. No accuracy information was available for six PAHs (naphthalene,
acenaphthylene, acenaphthene, fluorene, chrysene, and dibenzo(a,h)anthracene) analyses in both
treated and untreated  sediments.  The accuracy objective was not satisfactory for benzo(k)

-------
                                                                                     18

fluoranthene in  treated  or untreated  sediments nor  for benzo(g,h,i)perylene  in  untreated
sediment.   Accuracy  information was  satisfactory  for the PAH analyses  in water  and oil
residues.  In treated and untreated sediments and water residues, PAH analyses satisfied ARCS
specified  QA/QC requirements for blank analyses.   In all cases of oil residues,  the blank
analyses exceeded  the MDL.  Calibration limits for PAH analyses  met the ARCS QA/QC
specifications for both treated and untreated sediments as well as water and oil residue analyses.
Detection limit information  was not available  for  PAH analyses in treated and  untreated
sediments nor for water and oil residues. In untreated sediment and oil residues,  the precision
information was satisfactory for all PAH analyses, except for acenaphthene and acenaphthene
in untreated sediment, and naphthalene in oil residues, where no precision information  was
available.  In treated sediments,  the precision information was satisfactory for  naphthalene,
acenaphthylene acenaphthene, fluorene,  phenanthrene, and anthracene, and was unsatisfactory
for the remaining PAH analyses.  In water residues, no precision information was available for
any of the PAH analyses. The matrix spike information was satisfactory for twelve of sixteen
PAH analyses in treated sediment, and for thirteen of the sixteen analyses in untreated sediment
and ten of the sixteen analyses in water and all analyses in oil residues.  Surrogate recoveries
were unsatisfactory for PAHs in either sediment and  oil residue analyses but were satisfactory
in water residue.
RETEC

       The RETEC technology was evaluated by analyzing sediment samples and their treated
residues (water residues and oil residues) for metals, conventional, PCBs and PAHs.  PCB and
PAH analyses were performed for sediment and residues. The metals and conventional analyses
were performed for both sediment samples and water residues.

       In a majority of the cases studied, the accuracy objective was satisfactory for the metal
analyses in treated and untreated sediments.  Of thirteen metals analyzed, accuracy information
was not available for  Ba, Se, and Ag.   In both  treated and untreated sediments,  ten of the
thirteen metal analyses (As, Cd, Cr, Cu, Fe, Pb, Mn, Ni, Hg, and Zn) satisfied ARCS specified
QA/QC requirements  for accuracy.   The  accuracy objective was  satisfactory for  all metal
analyses in water, except for Se, where accuracy did not satisfy QA/QC requirements.  Four of
the thirteen metal analyses (Cd,  Hg, Se, and Ag) satisfied  QA/QC  requirements for blank
analyses.  The remaining nine metal analyses (As, Ba, Cr, Cu, Fe, Pb, Mn, Ni, and Zn) were
analyzed  by  XRF techniques.  In all of the XRF analyses, blank sample analyses are not
applicable. In water residue, blank analyses were satisfactory for all metals except for Fe, Mn,
and Se, where blank analyses exceeded the detection limits specified in the QAPP, and for Ba,
where no information  regarding  blank analyses  was  available.   Both  initial and ongoing
calibration met the ARCS QA/QC specifications for Cd, Hg,  Se, and Ag  for both treated and
untreated  sediments, and for all metals in water residue analyses.  While in both treated and
untreated  sediments the remaining nine metals (As, Ba, Cr,  Cu,  Fe,  Pb, Mn, Ni, and Zn),
calibration information were not available.  Detection limits information for metal analyses in
treated and untreated sediments were not available for verification, except  for Cd, Hg, Se, and

-------
                                                                                    19

Ag,  where  detection limits were satisfactory.  Detection  limits for metal analyses in water
residue were satisfactory, except for Mn,  Se,  and Zn, where  detection limits  exceeded  the
QA/QC requirements. The precision information for the metal analyses in treated  and untreated
sediments, and in water residue was satisfactory for all elements, except for Hg in treated
sediment, and Se and Hg in water residue analyses, where precision information did not satisfy
QA/QC  requirements.    The matrix spike information for  treated  sediment analyses  were
satisfactory  for Cd, Hg,  and Ag, and was not satisfactory for Se.  The matrix spike information
for untreated sediment analyses were satisfactory for Cd and Hg, and was not satisfactory  for
Se and Ag.  The remaining nine metals (As, Ba, Cr, Cu,  Fe, Pb, Mn, Ni, and Zn) were
analyzed by XRF techniques for treated and untreated sediment.  In all of the XRF analyses,
matrix spike analyses are not applicable.  The matrix spike information  for water residue
analyses was satisfactory for all metals except for Ag where  matrix spike information did not
satisfy QA/QC requirement.

       Of the seven conventional analyses in both treated  and untreated sediments, accuracy
information was  satisfactory for TOC, and was not available for  total  cyanide, or total
phosphorus.  In the remaining four conventional analyses accuracy was  not applicable. Of ten
conventional analyses in water residue, accuracy information  was not available for TOC, total
cyanide, total phosphorus, and conductivity. In the remaining  seven  conventional analyses
accuracy was not applicable.  In both treated and untreated sediments and in water residue
analyses,  %TVS, oil and grease, TOC, total cyanide, and total  phosphorus satisfied QA/QC
requirements for blanks.  Also, the blank information was satisfactory for total solids and total
suspended solids in water residue analyses.  The blank information was not applicable for the
remaining conventional  analyses in  sediment and  water residue analyses.   Both  initial and
ongoing calibration information was satisfactory for all conventional analyses in both sediment
and  water residue, except for % moisture (in sediment), pH,  and TVS, TSS,  TS where
calibration information was not available, and for TOC and oil and grease, where ongoing
calibration information was not available.  Detection limit information was not available in both
treated and untreated sediments and in water residue for oil and grease, TOC, total cyanide, and
total phosphorus, and was not applicable for the remaining conventional analyses.   In treated
sediment, the precision  information  was not satisfactory for  oil  and grease and  no precision
information was available for total cyanide.  In untreated sediment, the precision information
was not satisfactory for total cyanide, and no precision information was available for TOC. The
precision information was satisfactory for the remaining five conventional analyses in treated and
untreated  sediments.  In water residue, the precision information was; satisfactory  for all the
conventional*, except for moisture, where no precision information was  available. The matrix
spike information was not available for oil and grease, and was satisfactory for total cyanide and
total phosphorus in treated sediment analyses.  The matrix spike information was  not available
for oil and grease, total cyanide, and total phosphorus in untreated sediment analyses.  The
matrix spike information was satisfactory for oil and grease, total cyanide, and total phosphorus
in water residue analyses. The matrix spike information for the remaining conventional analyses
was not applicable for sediment and water residue analyses.

-------
                                                                                     20

       The accuracy objective was unsatisfactory for the PCB analyses in treated sediments,
untreated sediments, and oil residue for Aroclor 1254 and could be used to represent the whole
PCB group.  No accuracy information was available for the remaining three Aroclor analyses
in treated  and untreated sediments. No accuracy information was available for PCB analyses
in water residues.   In both sediments and residues,  the blank analyses exceeded the detection
limits specified in  the QAPP.  Both initial and ongoing calibration for PCB analyses met the
ARCS QA/QC specifications for both treated and untreated sediments, as well as for water and
oil residues.   Detection  limit information was not  available for  PCB in  either  sediments or
residue analyses.  The precision information  for the PCB analyses  in treated and untreated
sediment was satisfactory for Aroclor 1254.  In all  remaining analyses, precision information
was not available.  The matrix spike was satisfactory for Aroclor 1254 in treated  sediment and
in oil residue analyses, and could be used to represent the whole PCB  group. The matrix spike
information was not available for the  remaining Aroclors in treated sediment and oil residues.
The matrix spike information was not available for PCB analyses in untreated sediment and in
water residues.  The surrogate recoveries were satisfactory for PCB  analyses in sediment and
residue analyses.

       In ten of the sixteen PAH analyses in treated sediments and in  seven of the sixteen PAH
analyses in  untreated  sediments, the accuracy  objective was satisfactory.   No accuracy
information was available for six PAHs (naphthalene, acenaphthylene acenaphthene, fluorene,
chrysene,  dibenzo(a,h)anthracene) analyses in  treated and untreated  sediment. The accuracy
objective  was not  satisfactory  for benzo(k)fluoranthene, benzo(a)pyrene, and  benzo(g,h,i)
perylene in untreated sediment.  Accuracy information was satisfactory  for fourteen of the
sixteen PAH analytes in oil residue.  Accuracy information was unsatisfactory for PAH analyses
in   water   residue,   except   for   benzo(k)fluoranthene,   indeno(l,2,3,c,d)pyrene,
dibenzo(a,h)anthracene.  The blank analyses for the PAHs in treated and untreated sediment was
satisfactory in all cases except for acenaphthylene, acenaphthene, fluorene,  phenanthrene,  and
anthracene. In water residues, all PAH analyses satisfied ARCS specified QA/QC requirements
for blank analyses.  In all oil residues, the blank analyses exceeded the detection limit specified
in the QAPP. Both initial and ongoing calibration information for PAH analyses met the ARCS
QA/QC specifications for both treated and untreated sediments,  and also for water and oil
residue analyses.  Detection limit information was  not available  for PAH analyses in  either
sediments or  residues. The precision information was satisfactory  for PAH analyses in treated
sediments,  except  for   benzo(k)fluoranthene, where  precision  did not satisfy  QA/QC
requirements.  The precision information  was  satisfactory for PAH analyses  in  untreated
sediments except for acenaphthylene and acenaphthene,  where precision information was not
available, and for benzo(k)fluoranthene, where precision did  not satisfy QA/QC requirements.
The precision information  was satisfactory  for  PAH  analyses  in  oil residue, except  for
benzo(k)fluoranthene, where precision information did not satisfy QA/QC requirements.  In
water residue, precision was unsatisfactory for  PAH  analyses except for benzo(k)fluoranthene,
indeno(l,2,3,c,d)pyrene,  and dibenzo(a,h)anthracene,  where precision was satisfactory.  The
matrix spike  information was satisfactory  for ten  of the sixteen PAH analytes  in treated
sediment, for fourteen of the analytes  in untreated sediment,  for thirteen of the analytes in oil
residues, and  for three of the analytes in water residues. Surrogate recoveries were satisfactory

-------
                                                                                  21

for PAHs in both treated and untreated sediments as well as for oil and water residue analyses.



Summary

       Based on the compliance with the ARCS QA/QC requirements, SAIC was capable of
supplying acceptable results for metals, conventionals, PCBs, and PAHs. The results received
for all four technologies satisfied ARCS QA/QC requirements.

       An examination of results of the bench scale technology demonstration data set indicates,
that SAIC could have successfully provided acceptable data for all parameters. The data user
should be aware that some QA/QC discrepancies were identified, as indicated by subscript 1 and
2 flags in Table 3.

-------
         APPENDICES A and D




       are not included in this report.




Copies are available from GLNPO upon request.

-------
      APPENDIX B
QA/QC Sample Rating Factors

-------
CATEGORY
RATING FACTORS
        CATEGORY
SCORE ACCEPTABILITY LEVEL
Accuracy
Precision
Certified Reference Material
Analytical Replicate
   3


   3
Acceptable = 3


Acceptable = 3
Spike Recovery


Blanks

Miscellaneous
Matrix Spike                    =  3
Surrogate Spike (organics)         =  3

Blanks                         =  3

Instrument Calibration (initial)      =  3
Instrument Calibration (on going)   =  2
Instrument Detection Limit        =  3
            Acceptable = 3
             (organics)  = 6

            Acceptable = 3
                                                             Acceptable = 3

-------
   APPENDIX C
Data Verification Flags

-------
 A =  Accuracy Problem

 AO = no standard available/no information available
 A, = accuracy limit for the reference materials exceeded
 A, = accuracy is not applicable
B = Blank Problem

BQ = no information available
B; = reagent blank value exceeded MDL
B, = blanks are not applicable
C = Calibration Problem

CQ = no information available
C| = initial calibration problem
Cj = on-going calibration problem
Q = no information on initial calibration
C6 = no information on on-going calibration
C, = on-going calibration is not applicable
D = Detection Limit Problem

D0 = no information available
D, = detection limit exceeded
D9 = detection limit is not applicable

-------
H = Holding Times Exceeded
P = Precision Problem

P0 = no information available
P, = precision limit for analytical replicate exceeded the QA/QC
     requirements
P3 = MSD exceeded the QA/QC requirement
P9 = precision is not applicable
S = Spike Recovery Problem

S0 = no information available on spike
S, = limit of matrix spike recovery exceeded
S2 = limit of surrogate spike recovery exceeded
Si = no information available on matrix spike recovery
S6 = no information available on surrogate spike recovery
S, = spike recovery not applicable

-------