EPA/540/2-89/038
     SUPERFUNDTREATABILITY
            CLEARINGHOUSE
               Document Reference:
Research Triangle Institute. "PCB Sediment Decontamination Process-Selection for
  Test and Evaluation." Approximately 175 pp. Prepared for U.S. EPA, HWERL.
                    September 1987.
              EPA LIBRARY NUMBER:

           Superfund Treatability Clearinghouse - FBZZ-1
                                U.S. Environmental Protection
                                Reg'oo5,Librar^ (p,.],,,
                                77 West Jackson EC ;
                                Chicago, JL 606C'.*-' '

-------
                 SUPERFUND TREATABILITT CLEARINGHOUSE ABSTRACT
Treatment Process:

Media:

Document Reference:
Document Type:

Contact:
Site Name:

Location of Test:
Physical/Chemical - Dechlorination

Soil/Generic

Research Triangle Institute.  "PCB Sediment
Decontamination Process-Selection for Test and
Evaluation," and slide presentation on "Effective
Treatment Technologies for the Chemical Destruction
of PCB."  Approximately 200 pp.  Prepared for U.S.
EPA, HWERL.  May 1987.

EPA ORD Report

Dr. Clark Allen
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC  27709
919-541-5826

Guam (Non-NPL)

Research Triangle Park, NC
BACKGROUND;  This document is a report describing the assessment of seven
alternative treatment processes that show potential for decontaminating
polychlorinated biphenyl (PCB)-contaminated sediments.  The processes are
KPEG, MODAR Supercritical Water Oxidation, Bio-Clean, Ultrasonics/UV, CFS
Extraction, B.E.S.T., and Low Energy Extraction.  Each process was evalu-
ated using five criteria:  the probability of cleaning sediments to 2 ppm
or less; the availability of a test system; the test and evaluation effort
required; the time required for future availability of a commercial treat-
ment process; and the probable cost of treatment using the process.  The
evaluation of the criteria for each process was carried out by engineering
analysis of available data and site visits to developers' facilities.  This
report deals with the KPEG process for the destruction of PCBs.
OPERATIONAL DiFOBMATIOH;  The KPEG process was demonstrated in the treat-
ment of contaminated soil on Guam by way of the Galson Terraclean-Cl
process.  This destroys PCBs by nucleophilic substitution.  Potassium
hydroxid* is reacted with polyethylene glycol (PEG) to form an alkoxide.
The alkoxide reacts to produce an ether and potassium chloride.
    Addition of an RO-group enhances the solubility of the molecule and
makes it less toxic.  The reaction may continue until several chlorine
atoms are removed from the PCB molecule.  The reagent consists of a mixture
of PEG, potassium hydroxide, and dimethyl sulfoxide (DMSO).
    Contaminated soil or sediment is fed to the reactor from 55-gallon
drums.  An equal volume of reagent is added to the soil in the reactor.
The reagent is blended with the soil using a stainless steel bladed mixer.
    During operation of the system, contaminated reagent is mixed with
make-up reagent in the reagent storage tank and recirculated into the
reaction vessel containing contaminated soil  The reaction vessel is heated
3/89-39                                                 Document Number:  FBZZ-1

     NOTE:  Quality assurance of data may not be appropriate for all uses.

-------
 (150°C)  and the soil and reagent are kept mixed until the reaction is com-
 plete.   Volatilized material from the bulk storage tank and the reaction
 vessel are vented through a charcoal adsorption unit.  Vater vapor is
 condensed and used as wash water.   The reagent  is  decanted, weighed,  and
 stored for reuse.  The soil is washed twice with water to remove excess
 reagent,  and the wash water is held for analysis and  possible treatment
 with activated carbon.
     The  treated soil is held for analysis.   If  PCB concentration is greater
 than 2 ppm,  the soil is retreated.   QA/QC procedures  are not discussed.
 PERFORMANCE;  It was found that all of the processes  assessed have merit.
 In  selecting the most promising ones,  a ranking system was used based on
 the five criteria mentioned in the  background section.   The processes were
 ranked comparatively as to the desirability for thorough testing and  evalu-
 ation.   The KPEG process was ranked 5th with a  score  of 0.58,  within  a
 range of scores from 0.49 to 0.62.   Laboratory-scale  KPEG treatments  were
 applied  and there was a reduction of PCB levels to 17.5 ppm by treating the
 soil 5 hours at 1150 to 120°C.   Residual PCBs were qualitatively identified
 as  penta- and hexa-chloro biphenyl.   These congeners  had been reduced 75
 percent  and 60 percent,  respectively,  by the treatment.   Galson reported
 reduction from 1800 to 2.3 ppm by treatment at  150°C  for 2 hours.

 CONTAMINANTS:

 Analytical  data is provided in the  treatability study report.   The
 breakdown of the contaminants by treatability group is:

 Treatability Group             CAS  Number        Contaminants

 W02-Dioxins/Furans/PCBs         1336-36-3        Total  PCBs
                                11096-82-5        PCB-1260
3/89-39                                                 Document Number:  FBZZ-1

     NOTE:  Quality assurance of data may not be appropriate for all uses.

-------
                SUPERFUND TREATABILITT CLEARINGHOUSE ABSTRACT


 Treatment Process:       Thermal Treatment  -  Critical Water  Oxidation

 Media:                   Soil/Generic

 Document Reference:      Research Triangle  Institute.   "PCB  Sediment
                         Decontamination  Process-Selection for Test and
                         Evaluation,"  and slide  presentation on  "Effective
                         Treatment Technologies  for  the Chemical Destruction
                         of  PCB."  Approximately 200 pp.  Prepared for U.S.
                         EPA,  HWERL.   May 1987.

 Document Type:           EPA ORD Report

 Contact:                Dr. Clark Allen
                         Research Triangle  Institute
                         P.O.  Box 12194
                         Research Triangle  Park,  NC   27709
                         919-541-5826

 Site Name:               Guam  (Non-NPL)

 Location of Test:        Research Triangle  Park,  NC

 BACKGROUND;  This document  is a report describing the assessment of seven
 alternative treatment  processes that  show  potential for decontaminating
 polychlorinated  biphenyl (PCB)-contaminated  sediments.  The processes are
 KPEG, MODAR Supercritical Water Oxidation, Bio-Clean, Ultrasonics/UV, CFS
 Extraction, B.E.S.T.,  and Low Energy  Extraction.  Each process was
 evaluated using  five criteria:   the probability  of  cleaning sediments to 2
 ppm or  less; the availability of a test  system;  the test and evaluation
 effort  required; the time required for future availability  of a commercial
 treatment process; and the  probable cost of  treatment using the process.
 The evaluation of the  criteria  for each  process  was carried out by
 engineering analysis of  available data and site  visits to developers'
 facilities.  This report deals  with the  evaluation  of a critical water
 oxidation process to destroy PCBs.
 OPERATIONAL INFORMATION;  The MODAR Supercritical Water Oxidation process
 utilizes water above critical conditions (374°C  and 22.1 MPa) to increase
 the solubility of organic materials and  oxygen  to effect a  rapid oxidation,
 destroying organic contaminants.   The PCBs are  found  in a slurry or sludge
 type material.  The report attempts to evaluate  systems available from C.F.
 System and Enseco.  However, the  source  of the  bench-scale  study is not
 given, neither are sampling procedures,  QA/QC procedures, or conclusions.
 PERFORMANCE:  It was found  that  all of the processes  assessed have merit.
 In selecting the most  promising ones, a  ranking  system was  used based on
 the five criteria mentioned in  the background section.  The processes were
 ranked comparatively as  to  the  desirability  for  thorough testing and
 evaluation.  The MODAR supercritical water system was ranked 6th with a
 score of 0.57,  within  scores which ranged  from 0.49  to 0.62.  The
destruction efficiency for PCB  is  given  in Table 1.
3/89-18                                            Document Number:  FBZZ-2

   NOTE:  Quality assurance of data nay not be appropriate for all uses.

-------
CONTAMINANTS:

Analytical data  is provided  in  the  treatability study report.  The
breakdown of  the contaminants by  treatability group is:

Treatability  Group              CAS  Number        Contaminants

W02-Dioxins/Furans/PCBs         1336-36-3         Total PCBs
                                  TABLE 1

                 WASTE DESTRUCTION EFFICIENCY MODAR/CECOS
                     DEMONSTRATION ORGANIC WASTE TEST
Contaminant
PCS
Feed rate
(g/min)
9.1xlO~2
Liquid
effluent
rate (g/min)
<3.1xlO~7
Gaseous
effluent
rate (g/min)
<4.4xlO~6
Destruction
efficiency
%
>99.9995
Note:   This is a partial listing of data.  Refer to the document for more
        information.
3/89-18                                            Document Number:  FBZZ-2

   NOTE:  Quality assurance of data may not be appropriate for all uses.

-------
RESEARCH  TRIANGLE  INSTITUTE


                                                          September 1987
                          PCB Sediment Decontamination
                                    Processes—Selection for
                                           Test and  Evaluation
                                                                    by

                                                         Ben H. Carpenter
                                                    Research Triangle Institute
                                              Research Triangle Park, NC 27709
                                                    Contract No.: 68-02-3992
                                                 RTI Project No.: 471U-3065-65
                                 Project Officers: Donald L. Wilson, T. David Ferguson
                                 Hazardous Waste Environmental Research Laboratory
                                                      Cincinnati, OH 45268
                                  Hazardous Waste Engineering Research Laboratory
                                            Office of Research and Developmen*
                                            U.S. Environmental Protection Agp
                                                      Cincinnati. OH

POST OFFICE BOX 12194  RESEARCH TRIANGLE PARK.  NORTHCAR1'

-------
                                   NOTICE
    The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency under Contract No. 68-02-3992 to
the Research Triangle Institute.  It has-been subjected to the Agency's peer
and administrative review, and it has<*been approved for publication as an EPA
document.  Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.

-------
                                  ABSTRACT

    Eight alternative treatments for PCB-contaminated sediments have been
assessed as candidates for immediate thorough test and evaluation.  The proc-
esses are:  Basic Extraction Sludge Treatment (B.E.S.T), UV/Ozone or Hydrogen/
Ultrasonics Technology, Bio-Clean Naturally-Adapted Microbe, Potassium Poly-
ethylene Glycolate (KPEG), Low Energy Extraction. MODAR Supercritical Water
Oxidation, Critical Fluid Systems (CFS) Propane Extraction, and Battelle In
Situ Vitrification.
    The processes were evaluated using five criteria:  the probability of
cleaning sediments to 2 ppm or less; the availability of a test system; the
test and evaluation effort required; the time required for future availability
of a commercial treatment process; and the probable cost of treatment using
the process.  These criteria were addressed by .engineering analysis of avail-
able data and site visits to developers' facilities.
    The processes were ranked comparatively as to the overall desirability of
thorough test and evaluation using all five criteria collectively.  Two rating
methods were applied:  a multiplicative model using a Desirability Function
and a linear model, d-SSYS, using weighted utility functions.  Both methods
converted the process characteristics to ratings on a scale from 0 to 1 (worst
to best).  The Desirability approach normalized the characteristic using the
difference between acceptable and borderline values; d-SSYS normalized the
characteristic using the difference between the maximum and minimum values.
In calculating the overall score, the factors were weighted equally in the
Desirability Function.  Probable cost of treatment and test and evaluation
effort were assigned weights 4 to 5 times those of the other three character-
istics in the d-SSYS ranking.  These independent approaches gave final overall
desirability scores as follows:
                                     iv

-------
                       LIST OF ABBREVIATIONS AND SYMBOLS
 ABBREVIATIONS

 ATCC
 EPA
 d-SSYS
 HWERL
 uses
 ONSO
 PEG
 PCP
 CF Systems
 B.E.S.T.
 RCC
 TEA
 CST
American Type Culture Collection
Environmental Protection Agency
Computer Model for Rating Alternatives
Hazardous Waste Engineering Research Laboratory
Unified Soil Classification System
Dimethyl sulfoxide
Polyethylene glycol
Pentachlorophenol
Critical Fluids Systems
Basic Extraction Sludge Treatment
Resources Conservation Company
Triethylamine
Critical Solution Temperature
SYMBOLS

Btu
.3
gal
kJ
kWH
Ib
MPa
mt
TRUs
British thermal unit
cubic meter
gallon
kilojoule
kilowatt hour
pound
mega pascal
metric ton
transuranics

-------
                              UNIT CONVERSIONS






1 atmosphere * 0.098066 MPa



1 Btu - 1.05587 kJ



1 Btu/lb - 2.326 kilojoule per kilogram



1 Btu/min - 0.023575 hp




1 cal (20*C) - 4.1819 joule



1 centistoke » 1 centipoise/density of liquid



1 cu ft - 7.4805 gal



1 cu ft - .028317 m3



1 foot-lb - 3.7662 x 10"7 kilowatt-hours




1 (ft-lb)/min - 3.0303 x 10'5 hp



              - 2.2597 x 10~5 kilowatts




1 gallon - 3.7854 liters




1 gallon - 0.0037854 m3




1 horsepower (hp) « 0.7457 kilowatt



1 joule - 2.77778 x 10'7 ktffl



1 kg eal.  - 3.9685 Btu



1 kJ - 0.94709 Btu



1 kilowatt-hr - 3.6 x 106 joules - 3.6 x 103 kJ



1 liter - 0.035316 cu ft



1 Ib - 0.45359 kg




1 lb/in2 - 0.0068948 NPa




1 lb/in2 * 0.068046 atmospheres




R = 8047.2 (kgf x m/*K)
                                       vii

-------
                                  CONTENTS

SECTION                                                                 PAGE
NOTICE 	      i i
FORWORD	     ill
ABSTRACT 	      iv
LIST OP ABBREVIATIONS AND SYMBOLS 	      vi
UNIT CONVERSIONS 	     vi i
LIST OP TABLES 	.	     xii
LIST OP FIGURES 	      xv
ACKNOWLEDGEMENTS	    xvii
SECTION 1  INTRODUCTION 	       1
   1.1  BACKGROUND 	       1
   1.2  PURPOSE 	       1
   1.3  APPROACH 	       2
SECTION 2  CONCLUSIONS 	       4
SECTION 3  TECHNICAL ASSESSMENT 	       6
   3.1  RANKING CRITERIA	       9
        3.1.1  Standard for Acceptable Cleanup .of PCBs 	       9
        3.1.2  Probable Cost of Treatment After Performance
               is Proven 	      10
        3.1.3  T and E Effort Required 	      12
        3.1.4  Availability of a System for Test 	      17
        3.1.5  Likely Future Commercial Availability of the
               Process 	      17
   3.2  EFFECT OP DIFFERENCE IN SEDIMENTS ON THE TREATMENT
        PROCESS 	      17
   3.3  BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS.
        RESOURCES CONSERVATION COMPANY 	      26
        3.3.1  Availability of System to Test 	      26
        3.3.2  Process Description 	      26
        3.3.3  Information fro* Prior Studies 	      30
        3.3.4  Field Tests 	      32
        3.3.5  Additional Data Needs  	      41
        3.3.6  Probable Cost of Treatment After Demonstration 	      44
        3.3.7  Environmental Characteristics 	      47
        3.3.8  Health and Safety Characteristics 	      47
        3.3.9  When Process Can Be Made Available 	      47
                                    Vlli

-------
                              CONTENTS (cont.)

SECTION                                                                 PAGE
   3.7  LOW ENERGY EXTRACTION PROCESS BY APPLIED SCIENCE DEPARTMENT
        OF NEW YORK UNIVERSITY 	       88
        3.7.1  Availability of System for Test	       88
        3.7.2  Process Description 	       89
        3.7.3  Information fro* Prior Studies  	       91
        3.7.4  Additional Data Needs 	       93
        3.7.5  Probable Cost of Treatment After Demonstration 	      100
        3.7.6  Environmental Characteristics 	      102
        3.7.7  Health Characteristics 	      104
        3.7.8  When Process Can Be Made Available 	      104
   3.8  SUPERCRITICAL WATER OXIDATION PROCESS.  MODAR.  INC	      104
        3.8.1  Availability of System for Test  	      104
        3.8.2  Process Description 	      105
        3.8.3  Information from Prior Studies -.	      107
        3.8.4  Pilot and Field Tests 	      107
        3.8.5  Additional Data Needs 	      113
        3.8.6  Probable Cost of Treatment After Demonstration 	      115
        3.8.7  Environmental Characteristics 	      119
        3.8.8  Health Characteristics 	      120
        3.8.9  When Process Can Be Available 	      120
   3.9  PROPANE EXTRACTION PROCESS. CRITICAL FLUID SYSTEMS (CFS)  ...      120
        3.9.1  Availability of System to Test  	      120
        3.9.2  Process Description 	      121
        3.9.3  Information from Prior Studies  	      123
        3.9.4  Field Tests 	      124
        3.9.5  Additional Data Needs 	      124
        3.9.6  Probable Cost of Treatment After Demonstration 	      126
        3.9.7  Environmental Characteristics 	      128
        3.9.8  Health and Safety Characteristics 	      128
        3.9.9  When Process Can Be Made Available 	      128

-------
                               LIST OF TABLES


TABLE                                                                   PAGE

   1   INITIAL SCREENING OP TREATMENT PROCESSES 	        7

   2   HUDSON RIVER SEDIMENT CHARACTERISTICS 	       13

   3   HUDSON RIVER SEDIMENTS GRAB SAMPLES PROPERTIES 	       14

   4   HUDSON RIVER SEDIMENTS, METAL CONCENTRATIONS AND PCB
       CONTENT 	'.	       15

   5   UNIFIED SOIL CLASSIFICATION SYSTEM 	       25

   6   BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS PC8-
       CONTAMINATED SOIL COMPOSITION ANALYSIS 	       30

   7   BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS PCB-
       CONTAMINATED SOIL MULTIPLE EXTRACTION WITH TEA 	       32

   8   BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS FIELD
       TESTS,  SUMMARY OF TYPES OF ANALYSES OF SITE SAMPLES 	       35

   9   FEED STOCK COMPOSITIONS,  TESTS OP BASIC EXTRACTION SLUDGE
       TREATMENT PROCESS  	       37

  10   BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS
       COMPONENTS TESTING,  ANALYSIS OF SOLIDS PRODUCTS 	       38

  11   BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS COMPONENTS
       TESTING.  ANALYSIS OF TREATED PROCESS WATER EFFLUENT 	       39

  12   BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS COMPONENTS
       TESTING.  ANALYSIS OF OIL PRODUCT 	       40

  13   SAMPLES/ANALYSES FOR BASIC EXTRACTION SLUDGE TREATMENT
       (B.E.S.T.) PROCESS.  PRELIMINARY TESTS 	       45

  14   SAMPLES/ANALYSES FOR BASIC EXTRACTION SLUDGE TREATMENT
       (B.E.S.T.) PROCESS,  PILOT TESTS 	       46

  15   COST ESTIMATE FOR BASIC EXTRACTION SLUDGE TREATMENT
       (B.E.S.T.) PROCESS 	       48

  16   SAMPLES/ANALYSES FOR UV/OZONE OR HYDROGEN/ULTRASONICS
       TECHNOLOGY 	       56

  17   CAPITAL COST ESTIMATE FOR UV/OZONE/ULTRASONICS TREATMENT 	       58
                                     Xll

-------
                           LIST OF TABLES (cont.)


TABLE                                                                   PAGE

  36   COST ESTIMATES, CF SYSTEMS ORGANIC EXTRACTION SYSTEM 	     129

  37   SAMPLES/ANALYSES BATTELLE IN SITU VITRIFICATION PROCESS,
       DEMONSTRATION TESTS 	     135

  38   WHOLE BODY RADIATION DOSES FROM ROUTINE OPERATIONS 	     142

  39   PUBLIC DOSE COMMITMENTS FROM ROUTINE OPERATIONS 	     142

  40   OCCUPATIONAL DOSES FROM ACCIDENTAL RELEASES 	     143

  41   PUBLIC DOSE COMMITMENTS FROM POSTULATED ABNORMAL
       OCCURRENCES 	     143

  42   ACCEPTABLE AND BORDERLINE VALUES FOR PROCESS
       CHARACTERISTICS 	     145

  43   OVERALL DESIRABILITY OF IMMEDIATE T AND E OF THE EIGHT
       CANDIDATE PROCESSES 	•	     147

  44   SCALED RATINGS OF EIGHT TREATMENT PROCESSES  	     150

  45   DETERMINISTIC SCORES FOR TREATMENT PROCESSES 	     152
                                    xiv

-------
                              LIST OF  FIGURES
IGURE                                                                  PAGE

  1     Basic  Extraction Sludge  Treatment  (B.E.S.T.)  process.
       components  test  unit  	       27

  2     Basic  Extraction Sludge  Treatment  (B.E.S.T.)  process.
       large-scale skid-mounted unit  	       28

  3     Basic  Extraction Sludge  Treatment  (B.E.S.T.)  process  flow
       diagram,  showing optional  countercurrent extractor  	       29

  4     Site map  for Basic  Extraction  Sludge Treatment  (B.E.S.T.)
       process cleanup  of  PCB-contaminated sludges  	       33

  5     UV/Ozone/Ultrasonics  process flow  diagram  	       50

  6     Schematic of Bio-Clean Naturally-Adapted Microbe process as
       applied to  sediments  	       62

  7     Typical pentachlorophenol  decay rates vs.  time  	       64

  8     Typical PCP decay curve,  concentration of  pentachloro-
       phenol vs.  time  	."	       65

  9     Schematic of KPEG with ONSO process 	       76

10     Schematic of scaled-up KPEG with DMSO process 	       85

11     Preliminary material  balance and flow diagram 	       90

12     Solvent leaching of PCBs  from  Waukegan Harbor sludge  	       92

13     Horizontal  belt  filter 	       95

14     GREERCO continuous counter-current contactor  	       96

15     Schematic of the NOOAR Supercritical Water Oxidation
       process 	      106

16     Schematic of NOOAR Supercritical Water Oxidation process
       as applied  to liquid wastes 	      108

17     MOOAR Supercritical Water  Oxidation process demonstration
       plot plan 	      109

18     Schematic showing major  components of the  NODAR pilot
       unit 	      ill
                                    xv

-------
                                 ACKNOWLEDGMENTS

     The contributions, guidance, and advice provided by the EPA Technical
Project Managers. Donald L. Wilson. T. David Ferguson, and by Charles J.
Rogers, Chief of Chemical and Biological Staff, are gratefully acknowledged.
Dr. Albert J. Klee, Chief of Chemical Biological Detoxification Branch.
provided valuable contributions concerning process rating Methodology.
Dr. John Glaser provided helpful input to the information on biological
treatment processes.  At RTI, Robert Truesdale and Paula Hoffman contributed
to the discussion of the effects of soil and sediment types on the
performance of treatment processes.  Dr. Clark Allen provided input and
assisted in the format of the section on process rating.  Dr. J. J. Spivey
provided constructive review of the report.
     The developers of the process described herein contributed performance
data, process flow sheets, and cost information without which an assessment
would not have been possible.  Their contributions are also gratefully
acknowledged and are referenced throughout the report.
                                  xvii

-------
                                 SECTION 1
                                INTRODUCTION

1.1  BACKGROUND
    The PCB-contaminated sediment problems in New Bedford, Massachusetts (EPA
Region I) in New York State (EPA Region II) and in Haukegan, Illinois (EPA
Region V) are reported to be the worst in the nation in terms of PCB
concentration and the total quantity of PCBs present.  In addition, there are
numerous industrial lagoons with large quantities of PCB contaminated
sediments.  The dredging of the sediments for decontaminating harbors, rivers.
and lagoons is unacceptable without effective disposal/treatment methods for
PCB contaminated sediments.
    EPA Regional Offices are being asked to comment on the technical and
economic feasibility of chemical/biological processes for clean-up of these
sediments and sludges.  The Regional Offices do not have adequate data to
recommend any of a number of processes proposed or being tested/evaluated for
the decontamination of sediments containing PCBs.
    In the first phase of evaluation, the Research Triangle Institute identi-
fied eight candidate treatment processes which showed potential as alterna-
tives to chemical waste landfill and to incineration.  Seven of these required
further test and development.  Some had been tested using soils, but none had
been tested specifically on PCB-contaminated sediments.  The sediments of
concern differ from soils in several properties that influence the performance
of unit operations involved in the treatment processes.  The details of this
study are presented in the published project report entitled "PCB Sediment
Decontamination:  Technical/Economic Assessment of Selected Alternative
Treatments" (NTIS Number PB87-133 112/AS).
1.2  PURPOSE
    The purpose of this study is to establish suitable factors for further
assessment of the candidate processes that have been identified, to review
these processes against these factors and identify additional data needs, and
to provide a basis for the selection of three processes for a defensible.

-------
    The probable cost of treatment after demonstration and approval of cue
process was determined from vendor-supplied information or our own estimates.
Cost elements given include:  capital, energy, labor, maintenance, process
quality control and testing.  The capital cost is recovered over the 2.5 years
of projected operation.  Labor and profit/contingency are estimated at uniform
rates for all processes for purposes of comparative evaluation.
    Environmental and health characteristics assessed include all process feed
and waste streams, reagents, and operating hazards.
    The lapsed time required to demonstrate process performance, then to
design, construct-and check out a full-scale process is projected, based on
needs for additional data and requirements of the developers.
    The projection is based on the conduct of treatment system tests at condi-
tions determined in laboratory tests, with sampling and analysis of process
feed and exit streams.  Additional testing for engineering design data has
been added as prescribed by the developer.  The design and construction time
for a full-scale unit, estimated by the developer, is shorter for those proc-
esses with unit operations that have previously been scaled to the size
necessary for cleanup of 152.000 m3 of contaminated sediment per year.  Where
the need exists to establish a basis for size of unit not known to be in
existence and demonstrated, the needed time is increased, up to six months.
    The processes are compared and rated using the results of the assessment.
Based on composite ratings, three processes showing the highest rating are
recommended for immediate test and evaluation (Section 4).

-------
                                  SECTION 2
                                 CONCLUSIONS

    Eight emerging treatment processes for decontamination of PCB-contaminated
sediments have been evaluated as candidates for thorough test and evaluation
(T and E) using a test system judged of sufficient size by the developer to
provide performance, cost, and scaleup data for a large commercial plant.  The
processes assessed include:  Basic Extraction Sludge Treatment (B.E.S.T); Bio-
Clean Naturally-Adapted Microbe; Critical Fluid Systems Propane Extraction:
Potassium Polyethylene Glycolate, Galson; Low Energy Extraction, New York
University; MODAR Supercritical Water Oxidation; UV/Oxone or Hydrogen/Ultra-
sonics Technology; and Battelle In Situ Vitrification.
    The processes were evaluated using as criteria:
    •  The probability of cleaning sediments to <2 ppm PCBs:
    •  The probable cost of treatment;
    • .The relative level of Test and Evaluation effort to be supported by
       EPA;
    •  The availability of a processing system to test; and
    •  The likely future commercial availability of the process.
    While all the processes except perhaps In Situ Vitrification merit further
development for treatment of sediments, comparative simultaneous evaluation of
their ratings on a scale of 0 to 1 gave the following results:

                                                   Relative Desirability of
                                                 Thorough Test and Evaluation
                                                 Desirability        d-SSYS
               Process                           	score            score
Basic Extraction Sludge Treatmer .                    0.623           0.8127
  Resources Conservation Company
UV/Ozone or Hydrogen/Ultrasonics Treatment.          0.621           0.8010
  Ozonic Technology, Inc.
Naturally-Adapted Microbes Process,                  0.617           0.7533
  Bio-Clean. Inc.

-------
                                 SECTION 3
                            TECHNICAL ASSESSMENT

    This section describes the development of criteria for ranking the proc-
esses, discusses the characteristics of soils and sediments as they relate to
the application of treatment processes, and describes each process assessment.
    The Phase 1 study screened 64 process technologies and selected eleven for
assessment.  The eleven processes assessed were:  KPEG, O.H.M. Methanol
Extraction, Advanced Electric Reactor, EPRI (Acurex) Solvent Nash, Bio-Clean.
Vitrification, LARC, MODAR Supercritical Water Oxidation, Soilex Solvent
Extraction, Sybron Bi-Chem 1006 PB, and Composting.  The assessment showed the
first eight of these to have potential for reduction of PCB concentrations to
the desired background levels (1 to 5 ppm) or less, with minimal environmental
impacts and low to moderate cost.  All of the eight except the Advanced
Electric Reactor require further development and testing.
    The Soilex Solvent Extraction, Sybron Bl-Chem 1006 PB, and Composting
processes ranked lowest in overall desirability and were dropped from further
consideration.
    The seven candidate processes that required further development and test-
ing (KPEG.  O.H.M.  Methanol Extraction, EPRI (Acurex) Solvent Wash, Bio-Clean.
Vitrification, LARC, and MODAR Supercritical Water Oxidation) were screened at
the start of the Phase 2 study for availability of a continuing developer and
a treatment system for use in test and evaluation of the process.  The results
of this screening are given in Table 1.  Three processes were eliminated from
further consideration.  The Solvent Wash process is not available for assess-
ment because its sponsor,  the Electric Power Research Institute, is seeking a
firm to undertake the further needed development of the process before it is
ready for further consideration.  The developer of the OHM Extraction process
has chosen not to invest in this process.  The developer of the LARC process
has not identified sufficient markets and the process is not available from
then.
    Meanwhile, four technologies not assessed in the Phase 1 study have become
available:   the Basic Extraction Sludge Treatment (B.E.S.T.) process: the

-------
      Process
      I.ARC
Basic Extraction Sludge
Treatment
      (K  Systcub  Propane
      bxtract ion
00
      Ultrasonlcs/UV Technology
      Low Energy Extraction
      Process
                                                 TABLE 1.   (Continued)
                                                 Contact
Continuing
Developer
                                                                                            Test Systea(s)
                                                                                              Avai lab It:
George Anspach
Atlantic Research Corporation
5390 Cherokee Avenue
Alexandria. VA  22312

Nark Tose
Resources Conservation Co.
3101 N.E. Northup Hay
Bel levlew. HA  98004
(206) 826-2376

Thoaas J. Cody, Jr.
CP Systems Corporation
25 Acorn Park
Cambridge NA  02140
(617) 492-1631

Edward A. Pedzy
Ozonlc Technology. Inc.
90 Herbert Avenue
P. 0. Box 320
Closter. NJ  07624
(201) 767-1225

Walter Brenner/Barry Rugg
New York University
Dept. of Applied Science
26-36 Stuyvvsant Street
New York. NY  10003
(212) 598-2471
                                                                                   Yes
                        Yes
    Yes
                                                                                                 Yes
    Yes
                                                                                                 Yes
    Yes
                                                                                               Planned
      'This  process was identified  as  tin: Acurcx process  in  the  Phase  1 study.

-------
    These levels (10 ppra, 25 ppm, and 50 ppm) are to be attained by removing
all contaminated soil exceeding these levels.  The removed soil is subject to
disposal regulations:  cleanup to <2 ppm.  For this reason, permits issued for
alternative destruction processes generally will require that all treated
materials and by-product waste streams must have PCB concentrations of less
than 2 ug/g resolvable chromatographic peak (2 ppm).  If this condition is not
•et, the effluents containing 2 ppm or greater must be disposed as if they
contained the PCB concentration of the original influent material.  (Neulicht,
1986).  If the PCB feed material being treated by the process is over 50 ppm
PCB. then the resulting effluents must be incinerated unless an analysis is
conducted and indicates that the PCB concentration is below 2 ppm per PCB
peak.
    In accordance with these policy and treatments requirements, we have
selected £ 2 ppm PCB as the standard of cleanup for alternative treatments.
3.1.2  Probable Cost of Treatment After Performance is Proven
    The probable cost of treatment is presented as the cost per cubic meter of
sediment treated, baaed on a system sufficiently large to process 380.000 m3
of Hudson River sediments in 2.5 years.  By focusing on a specific site and
size of cleanup task, each process could be assessed using data from the same
feed materials, and comparative 'cost estimates for a specific application
could be obtained.  The sediments from the Hudson River also meet the
requirement for use of a variety of soil/sediment types in testing PCB-
treatment processes (Section 3.2).
    Treatment process requirements determined capital, energy, and maintenance
costs.  Labor rates, overhead, contingency, profit, and health and safety were
costed uniformly for all processes.
    Since no full-scale systems exist for the processes under assessment .
capital costs were estimated by designing a full-scale system in collaboration
with the developer utilizing the data available as a basis.  Equipment costs
were then obtained as planning estimates from manufacturers or developers, or
estimated using the method of exponents:
                                   Ct
                                      10

-------
3.1.3  T and E Effort Required
    The test and evaluation effort required has been estimated based on a
comparison of available process data with the requirements for thorough test
and evaluation.  A checklist of information requirements was developed to
identify the data categories to be supplied to qualify the processes for a
permit to test.  The checklist. Appendix A, identifies the following informa-
tion as basic to assessment of each process:
    1.  Waste characteristics;
    2.  Process engineering description:
    3.  Sampling and monitoring plan:
    4.  Accident and spill prevention and countermeasure; and
    5.  Demonstration test plan.
For these assessments. Hudson River sediments were selected as the character-
ized wastes.
    Hudson River sediment material has been classified according to its con-
tent of clay, silt, muck, muck and wood chips,.sand, sand and wood chips.
coarse sand, and coarse sand and wood chips (Tofflemire and Quinn, 1979).
Sediments have been shown to range from clay to cobbles, with the largest mass
fraction being sand.
    The coarse fraction (>0.42 mm) of the sediments, sampled and characterized
by Normandeau Associates, Inc. (NAI), typically contained wood chips, sawdust.
shale chips, cinders, and coal fragments.  The fine size fractions contained
some fragments of the above, plus sand (containing quartz and feldspar), silt.
clay, and organic material.
    The highest PCB concentration was the muck with wood chips class, which
typically had over 30 percent silt and clay, high volatile solids and some
small but visible wood chips.  The size lowest in PCB was medium sized sand or
gravel without wood chips.
    Table 2 shows the characteristics of a representative sample, portions
dried overnight at 60 *C and 100 *C and sieved.  Additional sidiment charac-
teristics, including metals analyses, are shown for four selected grab samples
in Tables 3 and 4.
                                      12

-------
                         TABLE 3.  HUDSON RIVER SEDIMENTS GRAB SAMPLES  PROPERTIES
                                       Texture                Volatile
Sample       Sp.      Bulk             primary/                 solids        >2  ••      2-0.074       <0.074
Nu*ber       Gr.      ga/cc            secondary                 *            %         ••.  *       mm.  \


  25         2.4      0.44       Buck A fine wood chips/         17.4            19          51         30
                                 coarse sand


  26        2.35      0.74       Buck, find sand/                9.3          0.5        59.4        40.1
                                 silt


  27        2.83      1.26       sand & wood chips/              4.4          25.0        69.0         5.1)
                                 silt


  29        2.06      0.78       coarse sand & wood/             13.5          38.7        61.0         0.3
                                 sand


Inference:  Toff)e»ir
-------
    PCB concentration was positively correlated with Csl37, lead, and volatile
solids, and negatively correlated with total solids.  The parameter most high-
ly related to PCB was Csl37, but the simplest field test to relate to PCB was
total solids.  Equations developed from core-sample data are:
                   log PCB - 1.494 * 1.4 log Cs, R - 0.82
                                     and
              log PCB - 2.56 - 0.009 (total solids). R - -0.67.
The equation for total solids explained approximately 40 percent of the total
variance in log PCB concentration.  However, the large confidence intervals
about the mean preclude the use of this equation, or other equations develop-
ed, as good predictors of resultant PCB values.
    Process engineering descriptions were developed for each process assessed.
These vary in completeness because the processes vary in stage of development
from conceptual (Section 3.4) to field tested (Section 3.3).  For example,
some processes will require tests to determine material and energy balances;
others to confirm estimated balances and cleanup performance.
    While unit operations have been identified-and described for all proc-
esses, the descriptions are based only on performance requirements.  Detailed
equipment specifications have not been made, except where necessary to obtain
cost estimates (e.g., high pressure compression and slurry pumps).
    The descriptions include process flow diagrams and identify all product
and waste streams.  Additional process information includes summaries of bench
tests, pilot tests, and field tests, if available.
    Sampling and monitoring plans are given, based on the scale of process
tests required, the purposes of the tests, and the extent of data needed to
characterize the process performance and scaleup the system to full-scale.
Some of the processes, when the developers' prior experience justifies it. can
be scaled-up from bench-scale tests.  Thus the size of system indicated for T
and E is the size the developer feels can be scaled-up with confidence.  For
the needed tests, the extent of sampling and analyses is indicated.  Methods
of analysis are spe'ified and their costs estimated.
    Accident and spill prevention and counter-measures needs have been identi-
fied.   Part of the estimated cost is allocated to these factors.
                                      16

-------
    Sediments exist in many types ranging from hard, dense, large pieces of
rock through gravel, sand. silt, and clay to organic deposits of soft compres-
sible peat.  All of these materials may occur over a range of densities and
water contents.  A number of different sediment types «ay be present at any
given site, and the composition may vary with depth over intervals as little
as a few inches.
    Soils have been classified as zonal, intrazonal. and azonal in efforts to
organize their morphology relative to a particular set of soil forming factors
(Mitchell, 1976).  Zonal soils are characterized by the dominating influence
of climate.  If climatic conditions are reasonably uniform and continuous and
erosion is not too rapid, then soils from similar climates become alike re-
gardless of parent materials.  Intrazonal soils are associated with zonal
soils but reflect the influence of some local conditions (e.g.. poor drainage,
alkali salts, etc.).  Azonal soils are soils without profile development.
There is little or no alteration of the parent material because of their youth
or environmental setting.
    Sediments are soil and rock debris that have been transported and deposit-
ed away from their zones of formation, being carried by streams, currents.
winds,  ground water, and glaciers, and (in the case of lagoons) by transport
of solid and waste material.  A broad definition of sediments, from the point
of view of hazardous wastes, is any solid or sludge under water.  Sediments
have been classified geographically because the environment of deposition  is
important in determining their properties.  The environment determines the
complex of physical, chemical, and biological conditions under which a sedi-
ment accumulates and consolidates.  These conditions can be important in
creating the sediment's characteristics.  Effects of transportation on sedi-
ments include size reduction, shape and roundness, surface texture, and sort--
ing.  The method of deposition can also affect the sediment characteristics.
Deposition by slow settling tends to create a more consolidated and well
sorted deposit, whereas rapid deposit where a stream moves into a larger body
of water may create a poorly sorted, unconsolidated deposit.  The main types
of sediments are terrestrial (above tidal reach), mixed continental and
marine,  and marine (below tidal limits).  Sediments can contain components
                                     18

-------
    The sediment characteristics most useful in determining a behavior under
cleanup are particle size distribution,  mineralogy,  organic and clay content,
water content, permeability, elemental composition (including heavy metals).
alkalinity. pH, Atterberg limits, cohesion, electrical conductivity, shrink/
swell potential, cation exchange capacity, surface chemistry,  and total
cations.  These characteristics are important for making sure that soils/
sediments used in testing of PCB destruction processes are or are not the same
between processes and between process efficiencies.
    The characteristics that may turn out to have the most influence on PCB
retention and removal are particle size distribution, clay content, and miner-
ology.  Mitchell (1976) states that the greater the quantity of clay mineral
in a soil or sediment, the higher the plasticity, the greater the potential
shrinkage and swell, the lower the permeability, the higher the compressibili-
ty, the higher the true cohesion, and the lower the true angle of internal
friction.  Surface forces and their range of influence are small relative to
the weight and size of silt and sand particles; whereas, the behavior of small
and flaky clay mineral particles is strongly influenced by surface forces.
Only a maximum of about one-third of the soil solids need be clay in order  to
have a condition where the clay is likely to dominate the behavior by prevent-
ing direct interparticle contact of the granular particles.   There is a
tendency for clay particles to coat the granular particles in many sediments.
possibly causing the clay to exert a significant influence on properties at
even a lower content.  Clay mineralogy is important due to the extreme changes
in properties created by the presence of montmorillonite versus illite.
chlorite, or kaolinite for example.  The presence of a Ca-montmorillonite
provides a soil with an enormous shrink/swell potential as compared to kaoli-
nite which has very little shrink/swell potential.
    Sposito (1984) interprets surface phenomena from the point of view that
the adsorbing solids are inorganic and organic polymers bearing surface func-
tional groups whose reactivity determines the adsorptive characteristics.
    The solid phases that exhibit surface activity in soils are to be found
primarily in the clay and organic fractions.  The most important structural
units in the inorganic polymers found in clays are the silica tetrahedron
Si044~ and the octahedral complex MX6*~6b comprising a metal cation. Mm", and
six anions, Xb~.  Both of these units can polymerize to form sheet structures.
                                     20

-------
    If. on the other hand, isomorphic substitution of Si4* by Al3* occurs in
the tetrahedral sheet, the excess negative charge can distribute itself pri-
marily over just the three surface oxygen atoms of one tetrahedron, and much
stronger complexes with cations and dipolar molecules become possible because
of this localization of charge.
    Inorganic hydroxyl groups occur exposed on the outer periphery of phyl-
losilicates, amorphous silicate minerals, metal oxides, oxyhydroxides, and
hydroxides.  These groups commonly occur coordinated to one, two, or three
metal cations.  For example, three types of OH groups are found on the sur-
faces of the mineral goethite:  OH groups coordinated with one, two, or three
Pe3* cations.  The first can be protonated to form a Lewis acid site and then
exchanged to allow the formation of an inner-sphere complex with the HP042~
ion.  The OH in the o-phosphate unit and the oxygen ions coordinated to the
Fe3* cations are hydrogen-bonded to the goethite surface.
    When phyllosilicate crystallites are broken apart, singly coordinated OH
groups are exposed on the new edge surfaces.  On some mineral surfaces (e.g.,
kaolinite). Al(III) • H20, a Lewis acid, is found at the edge of the octa-
hedral sheet.  The hydroxyl group associated with the site can form a complex
with a proton at a low pH or with an hydroxide anion at a high pH.  Also, at a
high ptt, the water molecule bound to the AI3* cation can be expected to be
replaced by an hydroxide anion.  In contrast to the coordinated Al(III), OH
groups at the edge of the tetrahedral sheet are singly coordinated to Si4*
cations.  Because of the greater valence of the silicon, these OH groups tend
to complex only hydroxide ions.
    Some of the organic functional groups present in the compounds that poly-
merize to form the humic substances in clays would likely ultimately reside on
the interfaces between solid organic matter and the fluid phases in sediments.
The more prominent organic surface functional groups in well oxidized soils
are carboxyl, carbonyl, and phenylhydroxyl groups.  The stabilities of
complexes between these key groups and protons range from weak (uncharged
carbonyl) to very strong (phenolic OH. which does not ionize until about pH
9).  For this reason, it is entirely conceivable that the properties of
organic surface functional groups are not well defined, but instead can be
characterized only by a range of values.
                                     22

-------
been determined, then the extreme or optimum conditions can be tested at pilot
scale.  In order to choose sediment types, a sediment classification system
similar to one used for soils, must be developed.  The "Unified Soil Classifi-
cation System" (USCS) was developed in 1952 as a modification of Professor
Casagrande's Airfield Classification System (U.S. Department of Interior Earth
Manual. 1974).  This system takes into account engineering properties of the
soils and can be used on field soils or soils mixed in a laboratory.  The
system is based on the size of the particles, the amounts of various sizes,
the characteristics of the very fine grains, plasticity, and compressibility.
USCS divides soils into three major groups:  coarse-grained soils, fine-
grained soils, and highly organic (peaty) soils  (see Table 5).  Using particle
size, sediments can be classified with this system.
    The range of sediments of most probable importance to this study will be
the sands, silts, and clays.  Possibly a gravel should be tested in order to
confirm its effect on the PCB destruction process.  Sediments of smaller grain
size will probably be the most difficult to remove PCBs from in extraction
processes.  Other characteristics of the sediments themselves, in combination
with process characteristics will affect the processes.  Therefore, setting up
a matrix with the most probable characteristics of importance for a given
process,  and selecting sediments representative of these conditions would be
necessary for testing.  Another method of obtaining representative sediments
would be to obtain the components with the chosen characteristics and mix them
by hand in the lab.  Use of natural sediments may provide more real results.
whereas use of prepared sediments will provide more control in the testing
process.
    It is not anticipated that sufficient data will be made available to show
the effects of sediment type on treatment rate constants, or even to measure
these constants.  These concepts of sediment behavior will, however, help
considerably to identify the data needs and further experimental work needs
for the processes to be assessed.  The summary of characteristics given here
focuses on those that- may serve to distinguish hard to treat sediments  from
easy to treat sediments.  The theory helps in identifying the different types
of sediments that should be included in a test and evaluation program.  It
helps explain the need to identify in a laboratory the removal rates which a
given sediment dictates so that field testing of the processes can be planned
accordingly.
                                     24

-------
3.3  BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.)  PROCESS.  RESOURCES
     CONSERVATION COMPANY
3.3.1  Availability of System to Test
    The Basic Extraction Sludge Treatment (B.E.S.T.)  process  has  been develop-
ed by the Resources Conservation Company (RCC)  of  3101  N.E. Northup Way.
Bellevue. Washington 98004.   The company Mas founded  in 1968  and  is a subsidi-
ary of Reading and Bates.  The process uses a solvent having  an inverse
critical solution point in water to remove water and  oily material  from solid
matter (U.S. Patents 3899419, 3925201, 4002562. and 4056466).  It has been
applied to clean up PCB-contaminated oily sludges  at  a CERCLA site  (General
Refining Site, Savannah, Georgia).
    The company has bench units to perform glassware simulations  of the proc-
ess, required to establish parameters for its application to  a particular
sediment.  The company also has a test system sufficiently large  to process 91
kg of sediment feed in seven days (Figure 1), and a large-scale skid mounted
unit designed to process 91 metric tons per day (24 hours) of feed (Figure 2).
3.3.2  Process Description
    The process is described generally as it is applied to PCB-contaminated
sediment, sludge, or other feed material containing solid matter, oily con-
taminants, and bound and unbound water.  The feed is first pretreated with an
alkaline composition, then admixed with triethylamine  (TEA) while cool in?
below the critical solution temperature (CST).  A single liquid phase  is form-
ed from which the solid matter is separated.  The liquid is  then heated to a
temperature above the CST, to form an amine phase and  a water phase, after
which the water phase is decanted from the amine phase.  The amine phase con-
tains substantially all of the oily material including organic contaminants.
It is processed to recover the oil and contaminants, and the TEA is  recycled
for the processing of additional feed material.  The pretreatment of tho feed
with an alkali reduces substantially  the amount of residual  amine carriea over
into the solid and water products.
    Figure 3 shows a process  flow diagram  for  the full-scale unit', with  two-
stage extraction added as an  addable  option  to  the operations  available  on  the
full-scale unit.  The number  of stages can  be  increased  if necessary.  The
feed sludge or sediment with  free water  is  mixed with  TEA  in a mixor desig
                                      26

-------
   Figure 2.  Basic Extraction Sludge Treatment (B.E.S.T.),
large-scale skid-mounted unit (Resources Conservation Company).
                               28

-------
using the settling characteristics of the sediment.  The ratio of TEA to the
sediment feed must be high enough for all of the bound and unbound water in
the solid matter to be completely nixable in a single phase at or below a
predetermined temperature, and may range from 1 to 7 parts by weight TEA to 1
part by weight of water.
    The mixture is then mechanically separated, by centrifuging, into a solids
fraction and a liquid fraction containing the TEA. oil. contaminants, and
substantially all of the water.
    The liquid fraction is heated, usually to 60 *C or higher, whereupon the
liquid forms an oil/solvent phase and a water phase.  The former contains most
of the oil and contaminants.  The two phases are separated by decantation.
    The TEA is recovered from the oil/solvent fraction by flash evaporation,
countercurrent steam stripping, and heating of the oil residues containing the
oil-soluble contaminants to remove water.  TEA is also removed from the water
layer by flash evaporation and steam stripping.  Recovered TEA is chilled and
recycled.
    The separated solids are subjected to one or more additional extractions
with TEA, after which they are separated by centrifuging. and dried to remove
any residual TEA.
3.3.3 "Information from Prior Studies
    RCC conducted preliminary tests of the application of B.E.S.T. treatment
to samples of a PCB-contaminated soil supplied by EPA Region 10 (RCC, 1986).
The composition of the contaminated soil is shown in Table 6.  The B.E.S.T.
treatment was applied as described below.

       TABLE 6.   BASIC EXTRACTION SLUDGE TREATMENT (B.E.S.T.) PROCESS
                 PCB-CONTAMINATED SOIL COMPOSITION ANALYSIS
             Component                                    wt. %

            Oil                                             1.5
            Water                                         11.0
            Solids                                        87.5
            PCBs (mg/kg)                                  1500a

         aTotal  sample basis.   RCC data.
                                     30

-------
         TABLE  7.   BASIC  EXTRACTION  SLUDGE TREATMENT  (B.E.S.T.) PROCESS
               PCB-CONTAMINATED SOIL MULTIPLE  EXTRACTION WITH TEA
Soil










>50


extraction number
1
2
3
4
5
6
7
8
9
10
(Soxlet extraction of
raw sample with TEA)
Recovered Oil
Residual PCB, Concentration, mg/kg
310
93
35
53
63
19
32
22
19
20

23
598.000
    Reference:  Resources Conservation Co.

3.3.4  Field Tests
    The B.E.S.T. process has just been field te.sted at the General Refinery
site near Savannah, Georgia.  The CERCLA site cleanup was completed March 6.
1987.  PCB-containing oily wastes and sludges froa an oil-recycling plant were
cleaned up using the 91 Metric ton per day unit  (Figure 2).  In preparation
for the cleanup, samples collected by RCC, Weston. and Haztech were used to
characterize all materials onsite (RCC, 1986).
    Figure 4 shows a site map prepared by Roy F. Weston. Inc. to identify the
locations of samples.  Barrel samples of sludge and any free water were taken
from four lagoons located on the site.  Lagoon 1 was shallow and lacked the
free water layer found in the other ponds.  Core sediment samples were taken
in the vicinity of the barrel sample locations.  Samples of site well water,
oil tanks, and several soil core samples adjacent to the ponds were also
taken.  Filter cake samples were taken from the solids piles labeled GRSC*1
and 2 on the site map.
    From these samples, the following were selected as representative of feed
stocks to the process and were subjected to B.E.S.T. glassware simulation
testing (Phase 1 testing):
    Pond 1,  Ponds 2-4 surface, Ponds 2-4 subsurface. Ponds 2-4 surface and
    free water in proportionate quantifies, Filter Cake. Back filled  lagoon,
    and Ponds 1-4 sediment.
                                     32

-------
3.3.4.1  Phase 1. Simulations—
    The laboratory glassware testing involved extraction of oil and water and
removal of particulate solids by repeated centrifugation/decantations with
TEA.  The solids obtained were dried at 105 *C.  The oil/water/TEA extract was
then heated and decanted to effect separation of the aqueous fraction from the
solvent/oil fraction.  After separation, each fraction was steam stripped
using a Buchi roto-evaporator apparatus with steam injection into the flask
contents.
    The resulting oil, solids, and treated water were analyzed to determine
the ultimate distribution of contaminants and the basis for disposition of the
materials generated from the processing.  The water treatment applied to the
process effluent consisted of a two-stage clarification system.  The first
stage reduced oil and grease by addition of sulfuric acid and an emulsion
breaker.  The second stage reduced heavy metals using lime, a coagulant and a
coagulant aid.
    Table 8 identifies the types of analyses applied to the samples as
received and the products of simulated treatment.  All analyses were conducted
using EPA methods.  Hater quality analyses included halogens, sulfate, total
dissolved solids, total organic carbon, pH, turbidity (NTU). and conductivity.
Oil quality analyses included sulfur, sediment and water, pour point, flash
point, specific gravity, and heat value.  All analyses have been reported by
RCC in their Phase 1 test report.  The PCS results were reported as combined
Araclor 1242 and Araclor 1260, and are shown here.
3.3.4.2  Phase 2. Testing—
    RCC conducted further testing of the process in their Components Testing
unit to establish operating conditions for the treatment and to determine the
quality of the products, namely, recovered oil, water, and solids.  Two dif-
ferent types of feed stocks were prepared for processing, each using different
proportions of the total contaminated material present at the site.  The
B.E.S.T. water effluent was treated by a two stage, coagulation process to
reduce the amount of oil, metals, and other contaminants.
    Feed Composition—Calculations were made to estimate the blend of site
materials to achieve a representative feed stock which would allow the proc-
essing of all site materials at a constant feed composition.  Two feed stocks
were established as blends of the materials, as shown in Table 9.
                                      34

-------
                                                  TAIH.E 8.   (Continued)
Sample
Total
•etals
Water
quality
PCBs
Ug/L
Composition
EP Tox.
leachate
Oil
quality
Total
organ i cs
      Ponds  2-4  surface plus
      3  parts  free water
        Solids product

        Treated  water
        Oi 1  product
X
X
insufficient solids
   for analysis
        <5
      Ponds  1-4  sediment
       Oil  product
       Solids product
                        46
CO
o>
      tiar.kf i 1 led lagoon
      Storage  tank oil
     ah'rom  RCC Phase 1  test report.
     ^X  indicates that  analyses were conducted.

-------
        TABLE 10.  BASIC EXTRACTION SLUDGE TREATMENT  (B.E.S.T.) PROCESS
                COMPONENTS TESTING, ANALYSIS OF SOLIDS PRODUCTS
I tea
As
Ba
Cd
Cr
Cu
Fe
Pb
Mn
He
Ni
Se
Ag
Zn
TEA
PCBs
As
Ba
Cd
Cr
Cu
Fe
Pb
Mn
Hg
Ni
Se
Ag
Zn
04G
Method
EP Tox (Bg/kg)
EP Tox (Bg/kg)
EP Tox (Bg/kg)
EP Tox (Bg/kg)
EP Tox (Bg/kg)
EP Tox (Bg/kg)
EP Tox (Bg/kg)
EP Tox (»g/kg)
EP Tox (Bg/kg)
EP Tox (mg/kg)
EP Tox (Bg/kg)
EP Tox (Bg/kg)
EP Tox (Bg/kg)
G.C. (Bg/kg)
G.C. (Bg/kg)
Total Digest (ag/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Total Digest (Bg/kg)
Feed 1
<0.2
0.1
<0.01
0.02
<0.01
0.48
1.7
0.13
<0.05
<0.02
<0.3
<0.01
0.66
<100
0.14
7
300
0.6
12
14
3.000
3.000
13
<1.0
2.0
7
<0.2
58
0.4%
Feed 2
<0.2
0.1
<0.01
<0.02
<0.01
7.1
4.8
0.11
<0.05
<0.02
<0.3
<0.01
2.5
<100
0.02
<5
90
0.3
5.3
7.7
1.700
1.200
7
<1.0
1.0
<6
<0.2
r>o
0.5*
Resources Conservation Coapany
                                     38

-------
 accomplished  by providing a  recycle of  the oil-rich stripper bottoms with the
 incoming  feed to  the stripper.

 3.3.4.4   Field Test Results--

    The cleanup was completed March 6,  1987 using a full-scale system designed

 based on  the  data given  in Section 3.3.4.3.  Analytical data are not yet

 available for the sampling program carried out during the cleanup.  When

 available, these  data can be analyzed to provide a basis for confirmation of
 the performance of this  process in removing PCBs from sediments.

 3.3.5  Additional Data Needs

    For application to PCB sediment, the following data needs have been

 identified.   Laboratory  testing of the  sediments is necessary to determine:


    1.  The number of extraction stages required to achieve <2 ppm PCBs in the
        treated sediments;

    2.  The amount and type of alkali to add;

    3.  The suitability  of the process  for the range of particle size (up to
        0.64  cm dia particles are readily accommodated);

    4.  The settling characteristics of the sediments in the single liquid
      . phase;

    5.  The best ratios  of TEA/water to employ;

    6.  The need  to add  a suitable oil  to the feed to enhance the extraction
        of PCBs from the sediments; and

    7.  The need for post-treatment of process water effluent.

    Based on the data of Section 3.3.3.  RTI concludes that the extraction of

PCBs may be more efficient for sediments containing some oil than for those

containing little or no oil.1   If laboratory tests confirm this, oil could be
   The Resources Conservation Company disagrees somewhat with this conclusion
   (Tose. M. K. 1987):  "This may be true in the comparison of the data
   presented, but I do not believe that there is enough data to say with
   confidence that this is so."  RTI believes that, should the extraction
   without oil reach a limit, as shown in Table 20. then the process should
   not be rejected without trying oil addition.  The low PCBs in the solids
   from treatment of sludges containing oil (Table 23) are the basis for RTI' s
   proposal that oil addition be tried if needed.
                                     41

-------
           Water
              Steam strip for solvent recovery
           -  Test for PCBs,  TOC,  oil and'grease,  total  solids,  total  dis-
              solved solids,  and residual solvent
           -  Nitric acid digest for metals
           -  Remaining water sample available for client  analysis

        •   Oil
           - . Steam strip for solvent recover
           -  Test for PCBs,  residual solvent and  water
           -  Oil  tests;  API  gravity. BSAW
           -  Metals analysis following xylene dilution
              Remaining sample available for client analysis

    3.   Conclusions of Glassware Evaluations:

        •   Final report prepared containing  all analyses and  relevant  observa-
           tions.

        •   RCC engineering and laboratory personnel available for discussions.

           Upon successful results,  proceed  to pilot testing.

    Basic  Extraction Sludge Treatment process pilot testing would include  the
following  elements.

    1.   Component  Tests,  Pilot Units:

        The sediment sample will be  treated  as follows:

        •   PCB extraction of  the sediment with solvent will be accomplished by
           mixing and centrifuging the solvent/sediment mixture.   The  centrate
           and centrifuge solids from the extraction will  be  collected for
           further processing and analysis.

        •   The centrate will  be heated and decanted and the two recovered
           fractions segregated.  The upper  soil/solvent layer will be
           stripped of solvent in a  column.   The  lower water  layer will also
           be stripped of residual solvent in a separate column.   If neces-
           sary, the recovered water will be post  treated to  achieve the
           desired level  of purity.

    2.   Analytical Tests:

        Basic Extraction  Sludge Treatment process  separated components will be
        analyzed as follows:

        •   Recovered oil  will be analyzed for:
           -  PCBs
           -  Water content
              Residual solvent
           -  Sediment.
                                     43

-------
                  TABLE 13.  SAMPLES/ANALYSES FOR BASIC EXTRACTION SLUDGE TREATMENT  (B.E.S.T)  PROCESS,
                                                    PRELIMINARY TESTS
on
Sample Nimber
Feed 1 composite


Successive 4 solids
washings
4 water
4 oil
Total
Analyses
1
2
2
4
4
4
4
4
4
Method
Volatiles0 (T04)
Pestlcldes/PCBs (w/cog. scan)
Metals (24 by ICP. AA)
Pestlcides/PCBs (608/8080 w/
cog. scan)
Metals (24 by ICP. AA)
Pesticides/PCBs (608/8080)
Metals (24 by ICP, AA)
Pesticides/PCBs (608/8080)
Metals (24 by ICP. AA)
Estimated cost
of analyses'1
$ 125
500
370
1000
740 .
550
740
550
740
$5.315
      "Costs based on standard costs hy California Analytical Labs.
      UEI'A 60O/4 84 04

-------
amount to 26.2 kg (=62.4 L or 16.5 gallons).  Although the cost estimate
includes the cost of utilities,  it is assumed that installation of elerr.r i--ai
service to the site(s) is provided by others.  Also, cooling water is assumed
to be readily and freely obtainable from the Hudson River.  All permits,
licenses, a level site, concrete foundations and containment for the B.E.S.T.
unit will be provided by others.
    The estimated cost is $133.30. using a standard 50 percent profit (RCC
used 49.9 percent).   This estimate is detailed in Table 15.  Note that the
capital costs are amortized over the first 380,000 m3 of sediment treated.
Thereafter, the treatment cost would reduce to S73.31 m3 of treated sediment.
3.3.7  Environmental Characteristics
    Process wastewaters may contain undesirably high TEA concentrations and be
toxic to fish.  Overdesign of the water stripper is necessary to remove any
excess TEA.  In addition, wastewater treatment may be required, and has been
included in the cost estimates.   Emissions of TEA from vents, etc., require
control by condensation and/or scrubbing with cold TEA.  The decanters require
a nitrogen blanket to insure against unsafe vapor concentrations.
                                              •
3.3.8  Health and Safety Characteristics
    The sediments feed and the concentrated PCB-containing oil are hazardous
and require special  handling, with provision for personnel protection.  The
solvent. TEA. is handled safely in many commercial operations.
    Secondary containment under the full-scale unit is planned, with any
spills conveyed to storage.
3.3.9  When Process  Can Be Hade Available
    Full-scale processing of Hudson River sediments could begin in 14 to 19
months per the following schedule submitted by RCC.
          Test and Evaluation                           1 month
          Report                                        2 months
          Approval by EPA                               3 months
          Design. Procure, fabricate and ship          10 months*
          Installation                                  1 month
          Checkout/startup                              2 months
          Total                                        14 to 19 months
        to 5 months of this effort could be carried out in parallel  (e.g..
     design and procurement).  This would reduce  the  total time to  14 montns

-------
 3.4   UV/OZONE/ULTRASONICS AND UV/HYDROGEN/ULTRASONICS TREATMENT. OZONIC
      TECHNOLOGY,  INC.
 3.4.1   Availability  of  System for Test
    •The LARC  process employing ultraviolet  (UV) energy and hydrogen was
 studied under Phase  1.   It  is not now under  further consideration by the
 developers  (Atlantic Research Corporation).  Another fir*, Ozonic Technology.
 Inc.,  90 Herbert  Avenue, P.O. Box 320, Closter, N.J., has capabilities and
 interest in conducting  tests of PCB-contaainated sediments.  Ozonics utilizes
 ultrasonics together with UV/ozone treatments  in several commercial applica-
 tions  and has applied for a patent on their  process.  The use of ultrasonics
 to  increase the rate and extent of extraction  of PCBs from sediments and to
 increase the  rate of destruction of PCBs in  subsequent UV/ozone or UV/hydrogen
 treatment offers  the potential for substantial savings in the cost of treat-
 ment .
    The  technology to be demonstrated consists of three steps:  extraction.
 solids  separation, and  UV/ozone treatment of the extracted PCBs.  The sedi-
 ments would be treated  in a water slurry.
    Ozonics has a bench-system suitable to demonstrate performance and deter-
 mine parameters for a commercial size treatment system.
 3.4.2   Process Description
    The  process is described based on parameters chosen from the technical
 literature.   As shown in Figure 5, the sediments are mixed with sufficient
 water to  dissolve the contained PCBs.  For the flow diagram, sediments are
 assumed  to contain 300  ppm PCBs, equivalent  to 0.504 kg per m3.  In an
 ultrasonic-assisted extraction of PCBs with  an aqueous surfactant. Smith and
 Sitabkhan (1986) obtained a solution concentration of 44 mg/L (0.044 kg/™3).
 This concentration was  used to estimate the  ratio of water to sediment for the
 process:

         0.504 kg PCB/m3 of sediment
           0.044 kg/.3  of water       '  U'45 "3 water/"3 of sediment

    The required treatment rate set to define  the process for use on Hudson
river sediments was 21.7 m3 of contaminated  sediments per hour  (520 m3 per
day).  Figure 5 shows one of five required stirred tank ultrasonic extractors.
                                     49

-------
 each  holding  10  m3  of  slurry  feed.  The extractor volume was set to provide an
 average  residence  time of  nine minutes with  stirrings at 200 rpm (Levenspiei.
 1962).
    The  reactors were  sized conservatively,  using 44 mg/L as the limit of
 solubility.   PCBs have been extracted from soils using a 1. percent Tween 80
 surfactant  (Scholz  and Milanowski.  1984).  The tests were conducted with soils
 dosed up to 26.000  ppm PCBs.  Ultrasonic power was not used.  While extraction
 was incomplete relative to the < 2  ppm residuals standard, the supernatant
 liquid had a  PCB concentration of 366 mg/L.
    Ultrasonic energy  input was set at 13.2  watts/L.  This energy level would
 be achieved using suitable transducers to convert 60 Hz power to 23 - 43 kHz
 acoustic power at 80 percent  efficiency.  The treatment time required to
 achieve  the desired PCBs removal cannot be specified with precision until
 suitable data are obtained in preliminary tests.  A range of 9 - 18 minutes
 has been estimated  to  be necessary, based on Information from studies of
 ultrasonic-assisted extraction reported in the literature (Fogler, 1971.
 Schunn and Sole, 1967).  Fogler summarizes four such studies each assessing
 the results of'the use of ultrasonics on different, but relevant criterion:
                                              *
       Extraction                 Comparison                  Results
 1.  Alkaloids from          Ultrasound vs Soxhlet      15 sec w/ultrasound -
    jaborandi leaf          'extraction                 5 hours w/Soxhlet
                                                       extraction
 2.  Oil  from cotton-        Ultrasound vs no          830 percent increase
    seed                   ultrasound                 in amount extracted
 3.  Bitters from beer       Hops consumption with      40 percent reduction
    hops                    ultrasonics vs with-       with ultrasound
                            out ultrasonics
 4.  Perfume extraction     Payout  time for  use        Less than 1 year
                            of ultrasound

    The  data from extraction  of oil from cottonseed (Schurig and Sole,  1967)
 provide  guidance in regard to the impact of  ultrasonics power on the overall
 rate of  extraction and on the rate  of diffusion of the oils through the porous
membranes in which they are held.   The authors used a small plug flow reactor
with a fixed bed of solids, and developed a  correlation between extraction
 rate and ultrasonic power per cm^ of extraction base area:
                                     51

-------
    The very  fine suspended solids are expected to have adsorbed PCBs on. their
 surfaces,  and are allowed  to remain  in the water while the UV/ozone treatment
 is applied.   PCB degradation using UV/hydrogen has been demonstrated in the
 presence  of particulates  (Kitchens et al. . 1984).
    The UV/ozone treatment unit has  been projected from a detailed design and
 cost  study (Hacknan.  1978).  Their recommendations were followed for residence
 time,  required 03, and  required UV lamps.  The residence tine was set at  265
 minutes,  based on data  from batch studies conducted by the Houston Research
 Corporation.  Ozone requirements were set at 1127 kg/day.  This provides  an
 estimated 4 mg/L 03 in  the effluent  from the reactor plus 0.1 mg/L 03 per
 minute of residence time to account  for auto decomposition of 03. and 3 kg
 03/kg of  COD.  The sediments are expected to contain non-PCBs COD at an
 estimated 30  mg/L.  For the sake of  conservatism. non-PCBs COD was assumed to
 be completely oxidized  before PCBs could be oxidized.  It was also assumed
 that  this  can be done within the residence time projected for PCBs removal.
 The number of lamps (43-watt) was set at 7800  (one lamp per 0.14 m3 of reactor
 volume).1   As shown in Figure 5, the lamps are stacked vertically throughout
 the reactor to provide  continuous irradiation of the water throughout the path
 of flow.
    This  projection of  UV/ozone treatment is based on demonstrated technology.
 and a conservative ozone dissolution efficiency of 0.7 as recommended by  Evans
 (1972).  This efficiency could be increased by using ultrasound in this reac-
 tor.   Ultrasound would  also be expected to help keep the UV lamps clean,  and
 thus maintain their performance.  Ultrasonic energy was not applied in the
 projection for this stage of treatment because no suitable data were found on
which to  base the projection.  The estimated cost of treatment may be
 correspondingly high.
    The treated water would be recycled, or discharged after conditioning to
 remove any residual suspended particles.
3.4.3  Information from Prior Studies
    PCBs have been removed from high-energy metallic surfaces by ultrasonic-
assisted extraction with aqueous surfactant solutions (Smith and Sitabkhan.
1986).  Using Nu-Clear  at 10 percent, metal coupons coated with PCBs (Aroclor
      number of lamps provides 0 30 watts of UV per liter of reactor, which
 slightly exceeds the 0.27 watts L recommended by Glaze et al.  (1984) for
 UV/ozone removal of trihalomethdne precursors.
                                     53

-------
       To compare the performance of UV/ozone/ultrasonics with UV/hydrogen,
       ultrasonics; and
     •  To set parameters and performance standards for a full-scale process.
    A series of 25 extraction tests and 21 UV/ozone (or UV/hydrogen) test are
expected to be required to meet these purposes.  The extraction tests would be
designed to compare the residence time requirements for adequate removal of
PCBs both with and without ultrasound.  At least two different ultrasound
power levels would be tested.  The maximum sediment concentration in the
slurry that can be adequately extracted, and the effects of various ultrasound
frequencies, power, and ozone treatment on this concentration would be identi-
fied using small sets of experiments sequentially designed.
    The UV tests would determine whether ozone or hydrogen were required to
effect the necessary destruction of solubilized PCBs. and the impact of ultra-
sonic and UV power on the treatment time requirement.  The resulting data
would determine the performance and operating requirements for the sequenced
treatment.   Factors to be observed would include:
    •  The extent of extraction vs treatment time, the residual PCB in the
       sediments, and the concentration of the*liquid after extraction at
       varying ultrasonic power loadings (including none);
    • "The relative merits of UV/ozone/ultrasonics versus UV/hydrogen/ultra-
       sonics:
       The extent of PCB destruction in the separated liquid extract as a
       function of ozone level. UV power level, ultrasound power level, and
       time;
    •  The effectiveness of the ultrasound in reducing or preventing fouling
       of the UV lamp surfaces; and
       The required reagent usage rates per m3 of treated sediment.
    The sampling requirement for these tests are shown in Table 16.  The
additional  data would be obtained utilizing the bench-scale capabilities anu
equipment of Ozonics Technology, Inc.  A permit would be required.  The cost
of conducting the tests is estimated at $55,000 not counting support services
provided by EPA.
    The total estimated T and E cost is $151,000 (analyses - S21.000: T and E
support,  permits, and report - 375,000; system operation - 535,000).
                                     55

-------
 3.4.5  Probable Cost of Treatment
     The probable cost of treatment,  estimated  from  the  process  as  projected  in
 Section 3.4.2.  is 590 to S120  per  m3 of  containated sediment  treated.   The
 capital cost  for ultrasonic  extraction was  estimated based  upon the  cost  of
 ultrasonic  transducers at 52,900 per kW  of  power  input,  and the cost of
 stirred tank  extractors.   These main equipment items were allotted 40  percent
 of  the  total  required investment.  Additional  capital costs were included at
 the following proportions:   installation 6  percent,  piping  5  percent.
 electrical  5  percent,  building and services 5  percent,  engineering 10  percent.
 construction  expenses 12  percent,  contractor fee  2  percent, and contingency  15
 percent.
     The capital  cost  for  solids separation was estimated based  on  the  cost of
 six hydroclones  and  a manifold (Dorr-Oliver).   Additional elements of  capital
 cost  were allocated  as  for the extraction system.
     The UV/ozone  treatment cost was  estimated  based on  the  reactor flow rate
 of  water separated from  the  sediments, 4.17 m3/minute.   Capital  costs  detailed
 by  Hackman  (1971) were  equated to  this flow rate  and updated  to 1986.   For
 this  system,  electrical costs  were allotted 15'percent  of the cost of  the
 ozone production  unit.  The  number of ultraviolet lamps  was estimated  one oer
 0.14  m3 of  reactor volume.
    The details of the capital cost  estimated  are given  in  Table 17.   The
 total estimated  treatment cost is  shown  in Table  18.  Electric  power for
 ultrasound  generation, ozone generation, ultraviolet generation, and materials
 handling is $20.59/m3 for 9-minutes  extraction time, and S22.63/m3 for  18
 minutes.  Labor costs are based on 7 operators per  shift, plus  1 foreman  and
 one chemist, with 1 general manager  for  the project.
 3.4.6   Environmental  Characteristics
    The process would have vents for exit gases which would require  monitoring
 and control of PCBs or other volatiles.  Unused hydrogen would  be  recycled.
Unused  ozone would be decomposed by  treatment  with  a suitable reducing  agent.
Feed sediments are hazardous and require special  handling.  The  treated sedi-
ments,  if cleaned to <2 ppm PCBs.  would  not be  considered hazardous  with
respect to PCB-content.
    The effluent waters would  have been  treated for  PCBs. a factor inherent  in
the  process.  Further treatment for  residual surfactants mijht  be  required
                                     57

-------
3.4.7  Health Characteristics
    Hydrogen, if used, is flammabJe and would require special handling.
Special clothing would be required for sediment handling, as is the case with
all processes assessed.
3.4.8  When Process Can Be Made Available
    The fir* involved. Ozonics Technology, Inc.,  based on experience gained
over several years in designing ultrasonic systems handling ozone, believes
the process could be scaled-up based on the bench-scale tests.  If ozone 'is
used, suitable generators are commercially available.  If hydrogen is used.
some provision for its recycle would permit the use of purchased  liquid
hydrogen, avoiding the cost of a hydrogen plant.   Without recycle, an onsite
hydrogen plant using steam/ methane reforming or methanol cracking would be
required.  The following schedule shows the estimated time to full-scale
operation:
    Test and Evaluation
    Report
    Approval by EPA
    Process, design, fabrication and
      shipment
    Installation
    Checkout/Startup
    Total
 3 months
 2 months
 3 months
13 months (10 months if
     ozone is used)
 2 months
 1 month
21 - 24 months
    Ozonics has indicated that all technical work involving their systems can
be carried out without delay, and that therefore, this schedule could be
shortened significantly (Pedzy. 1987)
3.5  NATURALLY-ADAPTED MICROBES PROCESS, BIO-CLEAN, INC.
3.5.1  Availability of a System to Test
    The Bio-Clean Naturally-Adapted Microbe process has been developed by Bio-
Clean. Inc., Suite 130G, Burnsville. Minnesota 55337.  A patent is pending on
the process.  Bio-Clean is a company engaged in developing process sys.;ms to
                                     60

-------
 Dredged
S«dimentt
                              Cauitic
                              Storage•
            Ammonium
 Acid        Phosphate
Storage        Storage
                                Boiler
                             Make Up Water
                                                Natural
                                                Gas or
                                                Other Fuel
                                                                                                          Overflow
                                                                                                          to River
                    Figure  6.   Schematic of  Bio-Clean Naturally-Adapted Microbe process
                                            as applied  to  sediments.

-------
              240
o>
E
Q.
a
c

a.
O
a.
              220
              200
              180
              160
              140
              120
                                                        Soil - wet - ideal conditions
              100 -
                 0   2   4  6  8  10  12  14  16  18 20 22  24  26 28  30 32 34 36  38 40 42  44 46 48 50  52  54 66 58  60

                                                                Days


                      Figure 7-   Typical pentachlorophenol decay rates vs.  time (Bio-Clean,  Inc.).

-------
ppm to 7.5 ppra on the soil after three days.  A repeated test using culture
from the first test reduced the soil PCP level from 200 ppm to 18 ppm in less
than 2 days  (46 hours).  Subsequent tests showed reduction to below 5 ppm in
48 hours.
    The process was tested further in larger equipment:
    •  A 0.95 a3 rotating dru» system;
    •  A 16.7 m3 horizontal tank with a top-Mounted stir agitator; and
    •  A pilot-scale Bepex ribbon blender (Bepex Corporation. Minneapolis.
       Minnesota).
All three systems gave PCP destruction to less than 5 ppm.  The rotating drum
system would require special seals to prevent leaks.  The horizontal tank
system provided incomplete mixing, and unloading of the treated soils was a
problem.  The Bepex blender performed satisfactorily, but needed design modi-
fications to provide better temperature control.
    These tests provided the process parameters needed for purchase of
commercial-scale equipment, as follows:
    Parameter                  Specifications and Ranges of Operation
    Soil                   Screened to 1/2" or less
    Water                  Recycle to produce slurry
    Caustic                To produce pH 11.0 ± 1.0 (0.001 N to 0.01 N NaOH)
    Heat                   80 *C (176 •?) or more for 1 hour
    Nitrogen               To produce 0.6* by wt.
    Acid                   To pH 7.2 ± 0.4
    Water                  To cool and dilute to 200 ppm of PCP
    Bacteria               To produce 2 million cells/ml
    Ferment                With air, at 30 «C (86 •?) for 48 hours
    Discharge              Empty and repeat cycle

3.5.5  Process Design Basis for PCP Cleanup
    Bio-Clean developed a design basis for a system to clean up a site con-
taining 7646 m3 (10,000 cubic yards) of PCP-contaminated soil in nine months.
A system with three 91 m3 (20,500 gal) digesters operating at 90 percent
utilization would be required.  This would permit the cleanup of 30.6 m3 of
soil per day, using the following batch formula.
                                     66

-------
to dissolve 270 mg/L of Che contaminant.  Data for water solutions show a
range of 72-412 mg/L, depending on whether inorganic or organic soil was
extracted (Scholz and Milanowski, 1984).
    The best microorganism for the digestion of PCBs in this process needs to
be determined.  Suggested candidate microorganisms for testing include the
Alcaligenes eutrophus H850 and the Pseudomonas Putida LB-400 shown by Bedard
to be very effective (Bedard et al., 1985).  Samples of these strains can be
obtained from General Electric Research, Schenectady, New York (Finkbeiner,
1987).  All organisms tested could, if required, be selected from those
approved by the U.S.D.A. Fish and Wildlife-approved list.  Bio-Clean suggests
that one of the selected microbe cultures be taken from Hudson River Isolates
themselves.  It is estimated that one to two weeks will be required, initial-
ly, to condition the microbes 'for these sediments.
    Initial process tests would use the conditioned cultures.  Thereafter, in
subsequent tests, batches will be inoculated with 3-5 percent of the liquor
from the previous batch.  New culture would be expected to be supplied once
per month, from a full-scale fermenter dedicated to this service.  Digestion
time,  temperature, and nutrient needs will be determined using the New
Brunswick fermenter.  Periodic sampling of the "Well mixed slurry, followed by
phase-separation and analysis of both phases for PCBs will provide data to
time the digestion phase of the treatment cycle.
    During the heating cycle, vapors are condensed.  Exit gases will be sam-
pled and analyzed for PCBs and other toxic-emissions.  After seeding, there is
no recycle of exit gases.   Only sterilized air is fed in order to prevent the
entry of unwanted microorganisms while providing needed oxygen for microbe
growth.   Based on oxygen rates used in the field tests, the needed air rate
could be up to 0.5 kg mol of air per minute per cubic meter of digester charge
(3.4 kg 02/min x m3 of charge).  The initial tests will determine the required
aeration rates.
    Ammonium phosphate is added as needed as a nutrient and buffer at a rate
of 0.1-1 percent by weight.  It is added to the batch after the high-
temperature period, while the batch is still hot and before adjusting the pH
downward from =11 to 7.2 with sulfuric acid.  The digester charge should be
sampled after these additions, and cooling of the charge.
    For  the PCP field test, the hot mixture was cooled by dilution with
sterilized water and the final solids volume fraction was 0.36.  For Hudson
                                     68

-------
       TABLE 19.  SAMPLES/ANALYSES  FOR  BIO CLEAN NATURALLY ADAPTED MICROBE PROCESS.  PRELIMINARY TESTS
Sample
Sol ub i 1 i ly/
Extraction

Sediment
grinding*1
Microbe
Select ion

Process
Parameters

Exit Gas
Samples

Feed


Total
Number Analyses
4 sediments x 2 each -8 8 s. x 2
(1 sediment, 1
liquor) = 16
2 sediments x 2 each = 4 4 s. x 2 = 8

4 cultures x 1 Bedard 4 a. x 2
standard test each =4 (1 sediment, 1
liquor) - 8
2 temp, x 9 tests = 18 36
(solids and liquid) 2
2
18 tests x 2 each = 36 36
(1 during extraction; 36
1 during digestion)
1 composite 1
2
2


Estimated cost
Method of analyses'*
Pcsticides/PCBs 608/8080 $ 2.200


Pesticides/PCBs 608/8080 1,100

Pest icides/PCBs 608/8080 1.100


Pestlcides/PCB 608/8080 4.950
Metals (ICP, AA) 370
PCB 608/8080 w/cogener scan 500
Volatilesb (T04) 4.500
Pesticides/PCB 608/8080 4,950

VolatilC8c (T04) 125
Pesticides/PCBs (w/cog. scan) 500
Metals (24 by ICP, AA) 370
$2O . f>(>f)

•'Costs basrd on  standard costs by California Analytical Labs.
^CullcctKtl emissions  from tin: cHiiding operation.
(l I'A (i()()/4 84 04

-------
    The preliminary testing is estimated by the developer to cost 515,000:
the pilot demonstration, S40.000.  This does not include the sampling and
analyses given  in Tables 19 and 20.
    The total estimated T and E cost is $165,800 (analyses - $35,800; T and E
support, permits, and report - $75,000; system operation -.$55.000).
3.5.7  Probable Cost of Treatment After Demonstration
    The probable cost of treatment of Hudson River Sediments using the Bio-
Clean Naturally-Adapted Microbe process has been estimated by the developer
for a commercial-sized system designed to treat 650 m3 per day of sediment.
This capacity will, at 65 percent utilization, treat 380.000 m3 of contami-
nated sediment  in 2.5 years or less.  A low utilization allows for process
shutdown during freezing weather.  The treatment system would consist of sets
of three digesters, as shown in Figure 6.  The floating process would use
river water for non-contact cooling, and could be used at other locations.
The estimate assumes a laboratory onsite, as part of the system, so that
treated sediments could be tested, certified, then discharged back to the
river.  A post-treatment of wastewater at $2.20 per m3 of sediment treated has
been added to Bio-Clean's estimate as a contingency.  The percent profit.
cited by Bio-Clean at 40, has been increased to 50 for uniformity.
    The estimated cost is $156/m3.  The estimate is detailed in Table 21.  The
labor cost includes operating and maintenance labor, and laboratory testing
costs for operational control.
3.5.8  Environmental Characteristics
    The treatment process utilizes naturally occurring microorganisms.  The
organisms are adapted to PCBs by their exposure to these chemicals as food.
Their action is expected to result in complete mineralization of the PCBs.
with the final products of the process being carbon dioxide, water, and sodium
chloride.   This requires confirmation since the degree of competence in
degrading PCBs varies with the strain (Unterman. 1985).  The organisms can be
selected from these approved by the U.S. Department of Agriculture Fish and
Wildlife Division.
    During operation of the process, air feed as a source of oxygen  is ex-
hausted to the atmosphere.   This stream is passed through a condenser to
remove all condensible components.  Vent gases from the condenser will be
                                     72

-------
tested for any hazardous contaminants.  The stream is continuously monitored
for COg as a part of process control.
    The treated slurry would contain acceptably low levels of PCBs (or none)
in both the water and sediment phases together with spent microorganisms and
should on this basis be dischargeable back into the river.
3.5.9  Health and Safety Characteristics
    The sediment feed is hazardous and requires special handling, with provi-
sion for personnel protection.  The only reagents used are sulfuric acid and
sodium hydroxide.  These would be stored in tanks with containment beneath.
and provision to pump any spills to a holding basin for neutralization.  The
microorganisms are natural to the environment, but may be a health risk to
workers using the process.
3.5.10  When Process Can Be Made Available
    Upon successful demonstration, the process could be made available, given
sufficient funding. In 19 months from the start of preliminary testing.
according to Bio-Clean.  The preliminary testing and pilot tests would require
an estimated nine months.  Construction of the plant would require approxi-
mately 10 months.  The following schedule shows the estimated time to full-
scale operation.
               Preliminary tests                    1.5 months
               Test and evaluation                  2.5 months
               Report                                 2 months
               Approval by EPA                        3 months
               Process design, fabrication.
                 and shipment                        10 months
               TOTAL                                 19 months
3.6  POTASSIUM POLYETHYLENE GLYCOLATE (KPEG) WITH DMSO PROCESS BY GALSON
     RESEARCH CORPORATION
3.6.1  Availability of System for Test
    The Potassium Polyethylene Glycolate (KPEG) with Dimethyl Sulfoxide
process has two potential applications in the treatment of PCB-contaminated
sediments:  the treatment of the sediments themselves and the treatment of
concentrated PCBs from extraction processes.  The former is assessed herein  in
                                     74

-------
                                  Water
                                  Vapor
                           Reagent
                           Makeup
                                                                                       Carbon Adsorption Unit
                                    Condenser
        PCB
        Soil
        Water
O)
Reaction
(160°C)
Decant
Reagent
                                             Heater
First
Soil
Wash
                                                                    Fresh
                                                                    Water
Second
 Soil
 Wash
NonPCB
Soil,Reagent
(trace)
                                                                        Storage and
                                                                     Possibly Byproduct
                                                                       Removal with
                                                                      Activated Carbon
                                         Figure  9.   Schematic of KPEG with DM30 process.

-------
3.6.4  More Recent Data
3.6.4.1  Preliminary Tests—
    The KPEG with DMSO process is to be demonstrated for treatment of PCB-
contaminated soil on Guam.  Preliminary work on process operating conditions
has been completed, and EPA is acquiring a 1.5 m3 treatment system for this
evaluation.  Data acquired for an operating permit application are summarized
here (Research Demonstration Permit Application, 1987).
    Replicate samples of soil to be treated showed the following analyses:
           PCB Type
    1.  Aroclor 1260, ppm
    2.  Aroclor 1260, ppm
        Site No. 19
            2950
            4450
Site No.  22
     300
    2000
    The soil is sandy in texture, and contains about 17 wt. percent moisture.
Laboratory-scale KPEG treatments were applied by Kornel and Galson (Kornel,
1986).  Kornel reduced the PCB levels to 17.5 ppm by GC quantification (28.3
ppm by MS quantification) by treating the soil 5 hours at 115 to 120 *C.
Residual PCBs were qualitatively identified as penta- and hexa-chloro bi-
                                               *
phenyl.  These congeners had been reduced 75 percent and 60 percent, respec-
tively 'by the treatment.  Galson reported reduction from 1800 to 2.3 ppm by
treatment at 150 *C for 2 hours (Peterson, 1986).
    The reagent medium selected on the basis of these tests consists of:
        2 parts by wt.:
        2 parts by wt.:
        1 part:
Polyethylene glycol 400 (MW 400)
Dimethyl sulfoxide
50 percent aqueous potassium hydroxide
The reagent is applied to an equal volume of sediment.
3.6.4.2  Toxicity of KPEG Reaction Products—
    The basic nucleophilic substitution chemistry of the KPEG process yields
substituted biphenyls rather than ultimate products of decomposition. C02-
H20,  and KC1.   A large number of byproducts may be formed in processing the
PCB contaminants.  Thus, a great deal of painstaking analytical chemistry may
                                     78

-------
                                                     TABLE  22.   TOXICITV TESTS OP REACTION  PRODUCTS OP KPEO WITH DMSOa
00
o
Contaalnant Treated

Te t rach 1 orobeniene



2.3.7.8 Trtrachloro
dlbrnzo p dloxln )
(TCDO)




Test Species Treated

BloacciMulatlon Fathead Minnow Saaple T
control
O.S ppai TCB
0.3 ppa TCB/KPEG
Results
TCB 1 n
i«ye».i_"B/i

1733
<0. 1 to 0.3
Nutaccnclty Salaonella Ratio Colony Count*: Test Mate
(Kudo Test) typhlaurlua.
Strain TA9B Material
Mlth SB rat-
liver extract TCDD/KPEG
KPEO
DSMO
Strain TA98
S9 {to S9

1.3 1.2
I.I 0.7
1.2 0.1
Anthraclne 41.6 —

TCDO




TCDD









2 Nllrof liioronc
Aquatic Toxlclty Carp Material
30-days .
KPtr., ru>\> byproducts
KPEG. 25O ppai

Manhattan Guinea pig MMerlo.l
Toxlclty. oral
administration KPEn/TCIH) byproduct
SO -days


KI'EG rcBcrlil



14.9


, 250 ppai

Dose

BloaccuajUlallon
Factor
0
6199
0
rlal/Controli
Strain TA98
SB So 89

1.0 I.I
I.I 1.0
1.0 1.0
6.1
— 	
""•berJHed

0.0
00

*>liiE!c.»._2U'-B5o. De_aths

O.S
1
2
ino
0.5
1
2
100

0
0
0
0
0
0
U
0
                           "Rrsr.trch llrauimli ill Ion 1'rialt Appl lent Ion,  IUB7

-------
products of PCB reaction, substituted biphenyls, were not necessarily present
since previous tests involved products of reaction with other chemicals.
    A demonstration test is needed.  The Galson pilot system would meet
requirements for a test system.  Operating parameters and reagent composition
would be defined by the data from the preliminary tests.  This system is
designed to treat 45.4 kg (100 Ibs) of solid particulate per batch.  To
acquire scaleup data, five to ten complete batch treatments would be made.
Each treatment would take an estimated four to six hours of which the sedi-
ments would be reacted at prescribed temperatures from one to three hours.
The remainder of the runs would be devoted to startup, stabilization of the
system, and shutdown.  Allowing time for cleanup, preparation for the next
run, any repairs or modifications, and process data compilation, the total
test program would require three to four weeks.
    While fewer runs might suffice to demonstrate the performance of the
treatment, addition of selected tests to determine scaleup needs will help
ensure a better full-scale system.
    Scaleup to a commercial size system consisting of multiple reactors for
treatment of at least 14 m3 sediment per batch will require:
                                                    1
                                               .      <
    1.  Sizing of carbon filters for vent gases;
    2.  Selection of reactor mixer;
    3.  Choice of heating plant:  steam or hot oil; and
    4.  Selection of controls.
    Estimated sampling and analysis requirements for the demonstration  tests
are presented in Table 24.
    The preliminary tests and system operation for T and E are estimated  by
the developer to cost $100,000.  Adding analyses at $21.000, and support  at
S75.000 gives a total estimated cost of $196.000.
3.6.6  Probable Cost of Treatment After Demonstration
    The cost of KPEG with DMSO process treatment depends upon the size  of
commercial units used and the water content of the sedirents to be  treated.
To clean up the Hudson River sediments in 2.5 years, a system consisting  of  3
sets of 3 reactors (14 m3 capacity each) are estimated to be required.  These
reactors would be mounted in five modules, served by a single utility module
                                     82

-------
                       TAIII.K  24.   SAMPLES/ANALYSES  FOR KPEK WITH DNSO PROCI-SS DEMONSTRATION TKSTS
                                                      (Basis 8 Runs)
            Sample
      Nuaher
      Kccd (raw scdincnts)
   8 cu»pos He-
Analyses
                                                                                  Method
                                                                 8

                                                                 8
Estimated
 Cost of
Analyses'1
                 Pesticides/PCBs             $ 1.100
                 (608/8080)

                 Volatiles  (T04)b              1,000

                 Metals  (34,  ICP,  AA)          1.480
oo
CJ
      Trisil 
-------
                   Reagent Makeup
                   136 kg/Batch
                        I
   Water Vapor
   13.608 kg/Batch
                        Reagent Tank
                       13.608 kg/Batch
00
o>
                                   I
                                    Heater
                                  31.676,100
                                   kJ/Batch
            PCB Soil - 14 m3
            14.696 kg Sediment  ^
            9.798 kg Water      ^
            7.71 kg PCB
       Condensor
 React PCBs/
Decant Reagent
    Phase
                     Wash Tank 2
                    13.608 kg/Batch
 First Water
Wash of Soil
                     Wash Tank 1
                    13,608 kg/Batch
Second Water
Wash of Soil
Nor, PCB Soil
14,696 kg Sediment
9,798 kg Water
136 kg Reagent
                                                                              Biphenyls Removal
                                                                                7.71 kg/Batch
                               Figure  10.   Schematic of scaled-up KPEG with  DMSQ process.

-------
    Waste disposal - 2 kg activated carbon m3 soil  at S2.16/kg * 54.32.
    The total cost of application of the treatment  process is estimated as
follows:
                                                 Cost.  S/m3
         Cost Item                     4-hr cycle
    Capital                               18.82
    Utilities                             39.05
    Chemicals                             25.66
    Labor                                 15.40
    Maintenance                            1.88
    Supplies and safety equipment          0.80
    QA/QC                                  1.00
    Destruction  of PCBs/waste
      disposal                             4.30
        Subtotal                        S106.91
        Profit                            53.46
    Total                               $160.37
6-hr cycle
   27.25
   39.05
   25.66
   26.66
    2.72
    0.30
    1.00
    4.30
 S127.44
   63.72
 S191.16
3.6.7  Environmental Characteristics
    The KPEG with DMSO process operates with a closed system except for con-
denser vents and storage tank vents, which are controlled using adsorption by
activated carbon.  Wastewater and spent reaction mixes have shown no toxicity
to living organisms for the treatment of hazardous materials TCDD and chloro-
benzene.  Some further tests are prescribed to confirm that this non-toxicity
still holds when PCBs are treated,  and to quantify all discharges from the
process.
3.6.8  Health and Safety Characteristics
    The process treats hazardous wastes.  This requires a site safety plan and
a personnel training plan.  Spills  need to be contained in capture basins
beneath the reactors and reagent systems, with provision for pumping to hold-
ing tanks.  Reagents used are strong bases, especially when not diluted with
w-ter.  Safe handling requires wearing of protective clothing.  Except for
requirements for safety in handling hazardous wastes, the process should pre-
sent no health hazards.
                                    87

-------
    5.  Stream stripping of acetone from the sediment.
    The PCB solution, product of Step 5. would be treated with a reagent such
as KPEG, or incinerated (Brenner, Rugg. and Steiner. 1986).  The recovered
hydrophoblc solvent would be recycled.
    New York University proposes to develop the process over the next two
years by developing an analytical model and constructing an 50 kg/hr bench-
scale system for testing and analysis of the unit components involved.  This
would provide the basis for design of a 1.1 m3/hr unit for test and demonstra-
tion.  The University will need support for this development.
3.7.2  Process Description
    The process is described in terms of the extraction of PCB* using acetone
                   -t -                     -
as the hydrophylic solvent and Kerosene as the hydrophobic solvent.  Figure 11
shows a preliminary material balance and flow diagram.- The balance is based
on treatment of one m3 (1680 kg) of sediments.  The sediment is fed as a 5 wt.
percent slurry to a horizontal belt filter (Block 2) where 94.7 percent of the
water is removed to yield a solids fraction containing 50 percent sediments.
Based on the partition coefficient for PCBs between sediments and water, the
sediments will likely contain 98 pifrcent of the PCB content of the total feed.
    In.the second step on the process (Block 4), PCB-contaminated oil is ex-
tracted from the sediments using a hydrophylic-solvent (acetone) in^counter-
current extractions.  The number of stages required for the extraction of an
original contamination level to a prescribed residual level caa^tw determined
using experimentally measured partition coefficient* and stage efficiencies.
Acetone is removed from the decontaminated sediments by steam stripping (Block
6).
    In the third step of the process,  the PCB-containing stream (Block 4) is
contacted in a liquid-liquid extractor (Block 11) with a hydrophobic solvent
(kerosene) and additional water from acetone recovery, if needed,  to drive the
PCBs into the kerosene.  This step separates the PCBs from the water contain-
ing phase and concentrates them.  The resulting more concentrated kerosene
solution is more suitable for a final chemical destruction treatment.
    The two streams which leave this step are the PCB-containing stripping
solvent which proceeds to concentration and final destruction,  and the
                                     89

-------
acetone/water mixture containing traces of PCBs.  The acetone/water mixture
goes to a distillation column (Block 8) where the acetone is recovered and
returned to the leaching process (Block 4).  The water, contaminated with
trace amounts of PCBs, is recycled to the front of the liquid-liquid extractor
or pumped to the adsorption unit (Block 15) where it is adsorbed onto clean
sediment to close the cycle.
3.7.3  Information from Prior Studies
    Research completed thus far has -been directed toward:
    1.  Studying the effectiveness of various solvents for the leaching of
        PCBs from sediments;
    2.  Studying the settling behaviors of sediment in various solvents;
    3.  Selecting appropriate hydrophylic and hydrophobic solvents;
    4.  Studying the stripping (liquid/liquid extraction) unit operation using
        the selected solvent pair;
    5.  Developing a mathematical model for the most relevant steps in the
        process; and
    6.  Obtaining a preliminary economic analysis.
    Leaching experiments conducted with Waukegan Harbor sludge resulted in
very high efficiencies for the 'solvents acetone, methanol, and isopropanol.
Kerosene showed much lower efficiencies except in the case of dry sediment.
The results are shown in Figure 12.
    Settling experiments conducted with dry topsoil in acetone, methanol. and
isopropanol. containing various degrees of water, gave the following order of
settling rate, with sediments settling the fastest in acetone:
                      Acetone > Nethanol > Isopropanol
    Based on the results from the leaching and settling experiments, the find-
ings from the first year of research, and taking into consideration various
physical, chemical, and toxicological properties, a solvent pair has been
selected:  acetone as the hydrophylic solvent, and kerosene as the hydrophobic
solvent.
    Extraction experiments conducted with these two solvents with varying
water/acetone ratios have shown a partition coefficient between kerosene and
                                     91

-------
acetone of 30, which allows for the completion of the stripping operation with
a very small number of stages.            •
    For the countercurrent leaching process, a mathematical model based on the
well known Kremser equation (Treybal. 1968)  was derived.  The efficiency is
                         - DNp)/CF -
where:    nj - efficiency, fraction
          Cp • concentration of PCB. wet feed sediment, wt. fraction
         Dfjp - concentration of PCB, final leached sediment, wt. fraction
           L » leaching factor » the ratio R/ME;
                 where:  R » solvent and solute in leaching solution, mass/hr.
                         E - solvent and solute Kith leached solids, mass/hr,
                         M - slope of equilibrium curve; concentration of
                             solute in mixture vs concentration of solute in
                             solution.
          Np - final leaching stage number.
    The leaching factor is expressed in terms of a partition coefficient and
process parameters by:
                   L - {Zs (Pj * 1) - 1)/US (K! - 1) + 1}
where: .   Zs * mass fraction of PCBs in solids;
         P! * mass flow ratio:  hydrophylic solvent/dry sediment; and
         KI = partition coefficient.
    The liquid-liquid stripping process was modeled, based on the Alders equa-
tion (Alders, 1955):
                         E2 - 1 - [(H-1)/(HN*1 - 1)]
where:   E2 * efficiency, stripping process;
          H - P2 x K2:
               ?2 • mass flow ratio - hydrophobic solvent/hydrophylic solvent
               K2 = partition coefficient, stripping process:
        N»l » number of stripping stages plus 1.
3.7.4  Additional Data Seeds
3.7.4.1   Pilot System--
    Further development of the process beyond that discussed in Section 3.7.3
is required.   Additional data are required to support the selection of
                                     93

-------
I
                                                                                        JJ
                                                                                        rA
                                                                                         0)
                                                                                          C
                                                                                           o
                                                                                           M
                                                                                           •H
                                                                                           P
                                                                                           O
                                                                                                                        «n
                                                                                                                         c>
                                                                                              bO

-------
This fundamental  information (e.g., the height of an equivalent theoretical
plate) does not need to be determined using PCBs.   After its capabilities are
defined, the system would be utilized to define operating parameters for a
field test to demonstrate PCB-decontaaination.
3.7.4.2  Field Tests--
    The preliminary test for PCB field tests would define the extent of unit
operation needs for each phase of the process:

           Process Operation                       Data Needs
         Liquid/Solids Separation      Feed rates, handling parameters to
                                       achieve 50% solids.
         Leaching                      Number of stages for reduction to <2
                                       ppm PCBs.  Feed ratios, solvents/
                                       sediments.
         Extraction                    Solvent feed ratios.  Number of
                                       theoretical plates.
         Adsorption                    Adsorption column operating parameters.
                                       Sediment capacity for treatment of PCB-
                                       contaminated water.
         Solvent recovery       .       Steam stripping operating parameters.
                                       Distillation operating parameters.

    The estimated sampling and analyses for the preliminary tests are shown in
Table 25.   The single composite feed sample would be composited from the hori-
zontal filter after its operation was set to produce a 50 percent solids feed
to the process.   Treated sediments are sampled after steam stripping of the
solvents.   An estimated four stages of extraction would be tested.  Three
ratios of kerosene to water/acetone are assumed to be tested in the liquid-
liquid extraction stage.  Similarly, three tests are estimated for the waste-
water cleaning to set the amount of sediment required.
    The field test sampling requirements (Table 26) are based upon two weeks
of continuous operation of a 1.1 m3 demonstration-size system.  Feed samples
are composited for each week of operation.   Other samples are composited
da j ly
    The development of this process through field test and evaluation is
estimated  by Mew York University to require 5827,000.  This includes the
                                     97

-------
                                 TAIII.F  2fi.   SAMPLES/ANALYSES FOR LOW ENEMY  EXTRACTION PROCESS

                                                            FIELD TESTS

-------
3.7.5.2  Energy Requirements and Cost —
    The amount of energy required for solvent recovery is a function of proc-
ess parameters and capacity.  The cost of energy is assumed to be $0.26/liter
of *2 fuel oil and the energy value to be 37.7 MJ/liter.
3.7.5.3  Cost of Labor —
    The labor cost is based on an automated industrial chemical processing
plant.  Operator hours per day and processing step are calculated as follows:
where:   ej - Operator hours per day and processing step of reference case;
         62 m Operator hours per day and processing step of case 2:
         Qj - Process capacity of reference case;
         Q2 " Process capacity of case 2; and
          n * Empirical constant.

The values used in this evaluation are:       e± - 18 h/d x step;
                                              Qj - 9.07 mt/d
                                               n - 0.22

    The number of foremen and chemists are taken to be 15 percent of the num-
ber of operators.  In addition to these workers, there is one site manager.

    The hourly wages are assumed to be:       Operators:  15 $/hr
                                              Foreman:    18 $/hr
                                              Chemist:    25 S/hr
                                              Manager:    60 S/hr

    Based on these assumptions, the staff requirements and the labor cost per
m3 of treated sediment can be estimated.  The results are given in Table 28.

        TABLE 28.  LOW ENERGY EXTRACTION PROCESS, LABOR REQUIREMENTS
                                                               Labor cost of
   Process capacity            Staff per 8-hour shift             Treated
-    -    •                       Sediment
metric ton/day  m3/day    Operators  Foremen  Chemists  Mgr.        S/m3

      958         570         11        1         1      1          9.60
                                      101

-------
TABLE 30.  LOW ENERGY EXTRACTION PROCESS
            COST OF TREATMENT
Cost I tea
Capital
Energy
Labor
Maintenance
Solvent/acetone, kerosene
Supplies and safety equipment
Waste treatment (KPEG)
QA/QC
Subtotal
Profit
Total
Incineration of
extracted PCBs
as a 50% solution
S 9.50
11.00
8.60
1.40
0.08
0.80
0.77
1.00 ^
S33.15
16.58
$49.73
Incineration of
extracted PCBs
as a 10% solution
S 9.50
11.00
8.60
1.40
0.08
0.80
5.40
1.00
S37.78
18.89
$56.67
                  103

-------
facility in Natick, Massachusetts and at the CECOS International Falls site.
Niagara Falls, New York.
3.8.2  Process Description
    The process is described based on discussions with the developers and data
supplied by the*.  Figure 15 shows a schematic flow sheet for the MOOAR proc-
ess as it would be applied to sediments.  A full-scale system would include
screening to remove rocks and large pebbles if this were not done as part of
the dredging operation.  The feed pump (e.g., Gardner-Denver mud pump) would
handle the range of sediment particle size up to about 2 mm diameter, at 20-40
percent solids.
    Feed to the process is controlled to an upper limit of heating value of
4187 kJ/kg (1800 Btu/lb).  The Hudson River sediments lack sufficient heating
value, therefore fuel addition will be necessary.  A combination of preheat by
exchange with process effluent and fuel addition is a more cost-effective
option.  As shown, a portion of the supercritical process effluent may be
recycled to the reactor by a high-temperature, high-pressure pump to raise  the
combined fluids to a high enough temperature to maintain rapid oxidation reac-
tions in the continuously fed reactor.
    Oxygen, stored as a liquid, is pumped to system pressure, preheated, and
metered into the reaction vessel.  Alternatively, air can be compressed and
used as the oxidant.
    When wastes contain organic heteroatoms which produce mineral acids (HC1
in the case of PCBs) and it is necessary to neutralize these acids, caustic is
injected into the feed system to form appropriate salts.
    Feed lines would be sized to provide a flow velocity sufficient to keep
the solids suspended.  In the reactor, at a temperature of 400 to 650 *C and a
pressure of 22.1 to 25 MPa, the oxidant is completely miscible with the solu-
tion and the sediments are suspended in a single homogeneous fluid.  Organic
contaminants are oxidized rapidly.  A residence time of less than a minute  is
expected.  A second-stage reactor, as shown, is used to insure complete con-
version of residual CO to 003.
    Inorganic salts have a low solubility in supercritical water and will fail
to the bottom of the solids separator where they are removed with the treated
sediments.
                                     105

-------
    Gaseous products of reaction leave the reactor along with supercritical
water.  The reactor effluent is cooled to discharge carbon dioxide and water
at atmospheric conditions.
    Heat regaining in the effluent stream after the slurry preheat exchanger
can be used for lower level heat requirements or be dissipated.
    The cooled effluent fro* the process separates into a liquid water phase
and a gaseous phase, the latter containing primarily carbon dioxide along with
oxygen which is in excess of the stoichiometric requirements, and nitrogen
when air is the oxidant.  Pressure letdown and separation is carried out in
multiple stages in order to minimize erosion of valves as well as to optimize
equilibria.  Clean sediment and salts may be removed from the separator as a
cool brine/slurry through multiple letdown stages and are either dried or
discharged as a brine/slurry depending upon operating requirements.
    Key parameters monitored for the process include the effluent gas CO. and
02 concentration, the liquid effluent TOC, and the liquid effluent chloride
concentration.  NOX compounds are monitored in the gas. but have never been
detected at the operating temperatures employed.
3.8.3  Information from Prior Studies
    MOOAR has successfully conducted laboratory experiments decontaminating
dioxin tainted soil.  They claim to have achieved reduction to background
levels.   This work was conducted under an agreement of confidentiality with a
client,  therefore it has not been shared with us.
3.8.4  Pilot and Field Tests
    MODAR,  together with CECOS International of Buffalo, New York, have com-
pleted a field, pilot-scale demonstration of the process for the destruction
of hazardous organic waste materials.  Two waste streams were destroyed in the
field tests:  an aqueous-based waste contaminated with several organic prior-
ity pollutants, and an organic transformer dielectric fluid contaminated with
PCBs.   The demonstration tests were performed at the CECOS'  Niagara Falls. N'ew
York Hazardous Waste Treatment and Disposal Facility.
    The description of the test of the PCB-contaminated fluid given below is
taken from the report prepared by Carl N. Staszak, K. C. Malinowski. and W.  R.
Killilea (1987).   A schematic flow sheet of the process as applied to liquid
wastes is shown in Figure 16.   Figure 17 shows a plot plan of the installation
                                     107

-------
                                                                  Enisting Containment Derm
	 - -- _ 	 . — _ 	
Instrument Air Compressor
gj I 	 H*
CO ,.- x— v Oxidation Uni
1 88 r—O
ot> V J ill
t — ,.,iy j iT .. . 1^1 f-| — |
brum Storage/ 1 	 1 f ^A I ' i' i'|
Farm Work Area De-Ionized V / II I 1
Water System . n.i.™ f
Aqueous urum i
Waste Rack V
Tank

(
.


t
1 	

h-i

Effluent
Neut. Tanks
i


CO i
co2 j
j | Oz Meters



1
1
Operation 1
Trailer 1
1
1
1
Effluent I
Holding Tank I
1
Safety Fence
  Figure  17.   NODAR Supercritical  Water Oxidation process demonstration plot  plan.

-------
 Organic
  Waste
 Module
 Aqueous
  Waste
 Module

\  Air Introduction Module  |
                               L_
         Reactor 1
                '
                     —11
                    D   !
                         I
                    ")   I
                                                               I                        I
                                                 Reactor
                                                 Module
                                             Reactor 2
                                   r	i	1
I   H
                                      L3
                                               a
                           I        !
t
                                                  3

V
r Water Preheat Module  j  I
                                                                 r-D
                               Liquid
                               Sample
                                                        Liquid
                                               /*-*<     Outlet
                                                              L	
        Figure 18.  Schematic showing major components  of the MODAR  pilot unit.

-------
                   TABLE 31.  WASTE DESTRUCTION EFFICIENCY
                MODAR/CECOS DEMONSTRATION ORGANIC WASTE TEST
Contaminant
PCB
Feed rate
(g/min)
9.1x10-2
Liquid
effluent
rate (g/«in)
<3.1xlO~7
Gaseous
effluent
rate (g/min)
4.7 x 10~6
Destruction
efficiency
(*)
>99.995
                  TABLE 32.  ELEMENTAL MASS BALANCE SUMMARY
                                  PCB WASTE
Element
                 Quantity in
Quantity out
  (g/min)
Balance Closure
      (*)
c
0
Cl
49
251'
0.055 .
48
278
0.071
98
111
129
3.8.5  Additional Data Needs
    Discussions with MODAR determined that it would be best to demonstrate
viability on their bench-scale unit.  This would avoid the costs of modifica-
tion of the pilot system for slurry handling and onsite demonstration until
the process had been proven out.  Bench tests with the PCB-laden sediments to
be treated would be conducted after modification of the bench unit to n con-
figuration geared to solids handling rather than for liquid feeds as it is now
set up to support the company's commercial activities.  The sediment feed
probably would require grinding and sizing to 38 microns, maximum, to permit
use of the Lewa reciprocating pumps employed on this unit.  A pilot- or fuil-
scale unit would utilize larger pumps such as are used in oil field work, with
valves that could handle the total (screened) sediment feed directly.  Addi-
tional solids-handling equipment downstream from the reactor would be provided
for the bench unit.   Previous work has included the pumping of feces and urine
                                     113

-------
    MOOAR is evaluating a conceptual feed introduction design which may allow
bench-scale testing of a broader range of particle sizes than is mentioned
above.  If particle size reduction and classification is required. MODAR
suggests that it be carried out by a laboratory with proper equipment to
accomplish this type of work.  Any discarded sediments from the preparation
process could be subject to sampling and testing for PCBs.
    The performance test program as described would require about 20 kg (dry
basis) of contaminated sediment.  If these contaminated sediments must be
manifested for shipment then permits will be needed for MODAR to receive the
shipment.  If no permits are required, then MODAR may be able to conduct the
tests under the provisions of Massachusetts Department of Environmental
Quality Engineering Regulation 310 CMR paragraph 30.353:  Insignificant
Wastes.  The scaleup test program would required about 400 kg (dry basis) of
uncontaminated sediments with characteristics similar to those of the con-
taminated sediments.
    MODAR estimates the cost of the performance tests to be $50,000 - $75.000.
The cost for the scaleup tests will be $325,000 - $350.000.  The total cost of
the program is approximately $400,000.
    Table 33 shows the sampling and analyses estimates for conducting perform-
ance tests.  The scaleup tests will not require sampling and analyses. Four
feed samples are prescribed to determine the PCBs in any oversized residue
from the grinding of sediments preparatory to their processing in the test
unit.  The total estimated cost of T and E is $483,000.  This total cost esti-
mate may be broken down into costs for performance testing (analyses - 58,000;
T and E support - $50.000; operating cost - $75.000: total - $133,000) and
scaleup costs ($350.000).
3.8.6  Probable Cost of Treatment After Demonstration
    The MODAR Supercritical Water Oxidation process can be scaled to several
sizes depending on the magnitude of the application.  For use in cleaning
Hudson River sediments, a total of 380,000 m3 are to be treated over a 2.5-
year period.  At a 73-percent utilization rate projected by the developer.
this would require a nominal capacity of 570 i  ' per day of sediment.  This
utilization is lower than the 85 percent used for KPEG to allow for more
frequent shutdowns for safety inspection and any modifications that may be
                                     115

-------
required.  The capacity could be supplied by a pipe reactor, with a 17.3 cm
(6.8 in.) to 21.6 cm (8.5 in.) ID pipe.
    The process flow rates for a system of this size are shown in the flow
sheet.  Figure 15.  This volume provides a residence tine of less than one
•inute for the total feed at 25.51 NPa and 550 to 600 *C (3700 lb/in2 and 1022
to 1112 *F).  The total feed is:  0.4 m3/min sediments;  0.998 »3/«in water;
0.029 «3/«in fuel; and 0.082 »3/«in liquid oxygen.
    The material balance assumes complete conversion of the fuel (No. 2 fuel
oil) and contaminants to carbon dioxide, and retention of 10 wt. percent
moisture in the discharged sediment.  For this balance,  sediments were con-
sidered to be inert and to pass through the process without loss.
3.8.6.1  Energy Usage—
    Energy requirements for the process are largely involved with heating the
reaction mix to the required temperature.  The processing of one m3 of
sediments would require an estimated 3.2 x 106 kJ of energy, calculated as
follows:

      Mass of sediment/m3                                 1680 kg
      Mass of water processed with sediment               2520 kg
      Mass of fuel             .                          64.4 kg
      Mass of oxygen                                       236 kg
      Electrical energy
        to pump sediment, water, fuel.
          and oxygen                                   0.27 x 106 kJ
      Fuel energy
        to heat reactor fluid                           2.9 x 106 kJ
      Total energy                                      3.2 x 106 kJ

    The feed is preheated to 400 *C by exchange with the reactor effluent.
Additional energy is required to attain the prescribed 600 *C reaction temper-
ature.
    Volatile solids in the sediments could provide as much as 100% of the
required fuel energy.  For sediments with negligible heating value,  the cost
of fuel, assuming an energy value of 45,100 kJ/kg (19,400 BTU/lb), a density
of 0.876 kg/L. and a heat loss of 10 percent would be:
                                     117

-------
             Cost Item                                Cost S/m3
    Capital equipment                               $23.68 - $35.14
    Utilities
      Electricity                                         6.84
      Fuel Oil                                        0.00-20.98
    Chemicals                                            17.55
    Labor                                                 3.55
    Maintenance                                       2.37 -  3.51
    Supplies/safety equipment                             0.80
    Water treatment                                       1.40
    QA/QC                                                 1.00
        Subtotal        .                            $57.19 - $90.77
        Profit                                       28.60 - 45.39
    Total                                           $85.79 - S136.16

    The range in capital cost is reported by the developer to be due primarily
to the uncertainty in the size (and therefore cost) of the heat exchanger
design.  There is a trade-off of reduction in fuel value and oxygen require-
ments with the heat exchanger cost.  This limits the optimum preheat tempera-
tures to 400 to 500 *C.
3.8.7  'Environmental Characteristics
    The MODAR Supercritical Water Oxidation process is projected to handle
sediments at a 40 wt. percent concentration.  Should the dredged sediments
require filtration to remove excess water, the removed water would be subject
to treatment before discharge to remove any PCBs.  This has been included in
the estimated treatment  costs.
    Effluent gases are to be monitored for CO as a process control parameter.
They are not expected to contain any PCBs. however the commercial operation
would require monitoring for PCBs to insure against the emission of volati'-
ized PCBs in the event of a process upset, or during shutdown.
    Should pretreatment  of the sediments by grinding to reduce parti CM- si7f
be required, this operation could be a source of PCB-contaminated emissions.
Suit  Die controls should be applied to thjs operation.
                                     119

-------
water slurry of the sediment.  The treated slurry is discharged after separa-
tion from the liquid propane which contains dissolved contaminant.  The pro-
pane solution is fed to a separator where the solvent is removed by vaporiza-
tion and recycled.  The contaminants are drawn off as a concentrate for final
treatment.  The process has been tested for PCB-containing refinery sludge.
The PCB content of the solids component of the sludge was reduced to 5 ppm.
Additional extractions may be required to achieve the desired 2 ppm level.
    The company has a small portable 1-liter test unit for preliminary evalu-
ations of the potential of the process using 0.56 kg of feed, and will have a
mobile propane pilot system by July, 1987.  Preliminary tests of Hudson River
sediments using the portable test unit would be conducted by C. F. Systems at
their expense.  Larger-scale tests would require financial support.
3.9.2  Process Description
    The CFS Propane Extraction process is illustrated by the simplified flow
chart,  Figure 19.   Applied to the decontamination of sediments, a slurry would
be fed into the top of the extractor.  Propane, condensed by compression at
approximately 20 *C flows upwards through the extractor, making non-reactive
contact with the slurry.   The propane is allowed to accumulate until a pres-
sure of 1034 to 1379 kPa (ISO to 200 lb/in2) is attained.  The propane dis-
solves the oils in the sediment, including the PCBs, and extracts most of
these materials from the water.   Because of the low viscosity of the propane
and its low density, the separation of phases is expected to be rapid and
essentially complete.   The cleaned sediments and water are withdrawn from the
extractor.  Depending upon the material and the level of cleaning to be
attained, one or more extractions may be necessary.  A typical cycle of opera-
tion consists of charging the reactor, adding the propane, agitating for 5
minutes,  allowing to settle for 5 minutes, and removal of the top (propane)
layer while refilling with propane, in order to maintain the set pressure of
operation.
    As the propane from the first extraction leaves the extractor, it passes
to a separator through a valve where the pressure is partially reduced.  The
pressure  may typically be 345 kPa (50 Ib/in^) after th  valve.  In the separa-
tor, the  propane is vaporized and recycled as fresh solvent.  The extracted
PCBs and  other organics are drawn off from the separator for further treatment
and destruction.
                                     121

-------
    The proportion of propane vaporized, condensed,  and recycled can be upward
of 90 percent.   The remaining 10 percent or less retains the extracted con-
taminant .
    As an optical step, the extracted PCBs may be reacted with a reagent while
dissolved in or Mixed with the propane at the higher pressure (Modell, 1978).
    An important aspect of this process is the use of propane vapor recompres-
sion which restores the propane to its solvent-condition and utilizes the
overhead vapor enthalpy as the boiler heat source.  In order to accomplish
this, the temperature at which the heat is delivered from the vapor must be
raised sufficiently to provide a AT driving force for heat transfer to the
still bottoms in the boiler.  This is achieved by vapor compression, so that
the condensation and enthalpy release will occur at a temperature higher than
the boiling point of the boiler liquid.
    The process as described operates below the critical state for propane
(96.8 *C,  4118 kPa).  Where the solubility characteristics of the solvent are
favorable to the use of such lower temperatures and pressures, the costs of
the pressure vessels are reduced. .
                                                 •
3.9.3  Information from Prior Studies
    The following results were obtained in treating a PCB-contaminated
refinery sludge composed of 60 wt. percent solids, 20 wt. percent water, and
20 wt. percent oils.

                     Component                PCB, ppm
                   Sludge feed                   62a
                   Extract                      192b
                   Residue Solids                 5b
                   aBy material balance
                   bBy analysis

    These results were obtained using a bench-scale reactor of 1-L capacity.
The cylindrical reactor was half filled with sludge, and liquid propane was
pumped in so as to flow upward through the sludge and accumulate until a pr  s-
sure of 1034 kPa was attained.  The phases were mixed for 5 minutes, allowed
to settle, and the bottom layer withdrawn for analysis.  The extract was also
                                     123

-------
                                    TAIILE  34.   SAMPLES/ANALYSES  FOR CVS EXTRACTION PROCESS
to
en
Sa <>le Number
I' «•«''! 1 (composite)


Propane extract 3 (3 stages)

Treat ed sediments 3 (1 each
extraction)

Vent gases 3 (1 each
extraction)
Wastcwaler (water dis 1 (composite)
charged with sediments)

Total per lest
Tot a 1 . four 1 esls
Analyses
1
2
2
3
3
3

3
3

1

1


Method
Volatilesb (T04)
Pesticides/PCBs (w/cog. scan)
Metals (24 by ICP. AA)
Pesticides/PCBs (w/cog. scan)
Metals (24 by ICP. AA)
Pesticides/PCBs (608/8080 w/
cog. scan)
Metals (24 by ICP. AA)
Volutiles (T04)

Pesticides/PCBs (608/8080 w/
cog. scan)
Metals (24 by ICP. AA)


Estimated
cost
of analyses"
$ 125
500
370
750
555

750
555
375


250
185
$4.415
$17.600
       •'( usis li.iscd on standard rosls by I ,i I i Torn i a  Analytical Laboratories

       ''I I'A  (.(10/4 »l 01.

-------
            TABLE 35.  CFS EXTRACTION PROCESS,  LABOR REQUIREMENTS
Type
Operator
Foreman
Chemist
Manager
Total
S/m3 sediment treated
Number
per shift
6
1
1



Total
18
3
3
1


Hours/day
' 144
24
24
8


S/day
2,160
432
600
480
3,672
6.45
    Treatment of the extracted PCBs could be accomplished chemically or by
incineration.  Generally, if the concentration of PCBs in the material to be
treated is <1 percent, KPEG or other chemical treatment would be appropriate
cost wise.  For concentrations between 1 and 10 percent, such treatment would
likely still be appropriate.  For concentrations of PCBs in oil or other
liquid ranging to 50 percent PCBs, incineration is a cost-effective method of
destruction (Peterson, 1987).
    The full-scale treatment of sediments at 570 i3/day would yield 28? kg/day
of PCBs from a feed with a PCB-concentration of 300 ppm.  The weight of 50
percent solution would be 574 kg/day (1262 Ib/day).  This quantity could be
stored in 4 55-gallon drums.  Each filled drum would weigh an estimated 445
Ibs (allowing 45 Ibs for the drum).  The estimated cost of incineration of
this PCB extract is $0.77/m3 of sediment treated, computed using an incinera-
tion cost of S0.45/lb and a transportation cost of S3.75 per loaded mile (SCA
Chemical Services, 1987).

  1 m3 sediment yields 0.504 kg PCBs
  1 drum 50% concentrate holds 90.9 kg PCBs « 180.4 m3 of treated sediments
  1 truckload x 40.000 Ibs/load x 1/445 « 90 drums * 16.234 m3
  Transportation cost:
                        S3.75/mi x 800 mi
SO.lS/nr
                            16.234 m3

  Incineration Cost:   0.504 x 2.2 Ib/kg x S0.45/lb = S0.50/m3
  Barrel Cost:  S15/barrol x 1/180.4 m3 = 50.09'm3
  Total Incineration  Cost:  SO.18 * 0.50 - 0.09 = S0.77'm3.
                                     127

-------
      TABLE 36.  COST ESTIMATES. C. F. SYSTEMS ORGANIC EXTRACTION SYSTEM
                               S/m3 OF SEDIMENT
     Cost Iten
S5 per barrel8
                                                Cost/m3
S10 per barrel
Capital equipment
Utilities
Fuel oil
Chemicals
Labor
Maintenance
Supplies/ safety equipment
QA/QC
Destruction of PCBs
Subtotal
Profit
Total
S100.68

10.43
11.00
5.45
10.07
0.80
1.00
0.77
$140.22
14.72b
S154.92
S201.37

' 10.43
11.00
5.45
20.14
0.80
1.00
0.77
S250.96
14.72b
S265.68
aC.  F.  Systems'  estimate for 42 gallon barrel of waste for treatment.

bProfit on capital and maintenance is included in the amounts listed.  The
 remaining items are costed at the uniform rate of 50%.
                                     129

-------
Convective currents within the melt distribute the wastes evenly.  During tue
process, off gases emitted from the molten mass carry a small percentage of
volatilized organics (typically 0.05% of the inventory being vitrified).  The
gases are collected by a hood over the area and routed to a treatment system.
When power to the system is turned off, the molten volume begins to cool,
producing a block of glass and crystalline material that resembles natural
obsidian or basalt.  The subsidence that occurs can be covered with uncontami-
nated backfill.
    The principle of ISV operation is based on Joule heating, which occurs
when an electrical current passes through the molten mass.  As the molten mass
grows, resistance decreases; so to maintain the power level high enough to
continue melting the soil, the current must be increased.   This is accom-
plished by a transformer equipped with multiple voltage taps and a saturable
reactor power controller.  The multiple taps allow for more efficient use of
the power system by maintaining the power factor (the relationship between
current and voltage) near maximum.  The process continues until the appro-
priate depth is reached.  Melt depth is limited as the heat losses from the
melt approach the energy level that is deliverable to the molten soil by the
electrodes.
3.10.3  Information from Prior Studies
    The process has been tested for PCB-contaminated soils on a small scale
shown schematically in Figure 20 dimmerman. 1986).
    1.  A 20-cm-diameter by 30-cm-deep zone of loamy-clay soil, containing 500
        ppm PCBs was centrally located in a sealed metal container.  The
        contaminated soil was surrounded by noncontaminated soil and was 25 cm
        beneath the surface.
    2.  Four cylindrical molybdenum electrodes were set on a 23-cm square to
        surround the contaminated soil.  The electrodes extended to a depth of
        61 cm.
    3.  A path for electric current was established by placing a small amount
        of graphite and glass frit mixture between the electrodes on the soil
        surface.
    4.  Off-gases were collected and passed through a dual-stage activated
        carbon filter to contain any PCBs or decomposition products released.
        Online grab samples of exit gases were taken from a sample port.  the
        samples were analyzed for chlorine and hydrogen chloride.
                                     131

-------
    5.  Electric power was applied for a 6-hour period.  A vitrified block
        (220 kg) and 0.14 m3 was produced.  The melt extended to a depth of 81
        cm.
    6.  In addition to sampling off-gas emissions, residues in off-gas lines
        and containment equipment, the migration of PCBs into the surrounding
        soil, and the residual level of PCBs in the vitrified block were
        monitored.
    Data from off-gas release and soil container smears provided the most
quantitative values on the release from the melt during and after processing.
Information collected from the adsorption tubes and the smear sample extrac-
tions indicated a 4.4-mg total off-gas emission. 1.1 mg of which was deposited
on container surfaces.  These off-gas releases accounted for 0.05 wt% of the
initial PCB quantity, corresponding to a greater than 99.9 percent thermal
destruction and removal efficiency for the ISV process.  This does not include
the removal efficiency of the off-gas system; therefore, a system ORE cannot
be calculated form the available data.  Activated carbon filters can effec-
tively contain any of the off-gas emissions.
    Analyses of the florisil adsorber also indicated a small amount of furan
(PCDF) and dioxin (PCDO) generated in total quantities of 0.4 ug and 0.1 ug,
respectively.  Only the tetra and penta isomers of the PCDF were detected, and
only the hepta and octa Isomers of the PCDD were detected.  These small
quantities are less than the reported amounts typically generated from a PCB
fire and do not represent a hazardous operational concern.
    The vitrified mass showed no detectable residual level of PCB. which is to
be expected considering the high process temperatures.  Also, no PCB contami-
nation was detected in the majority of soil surrounding the vitrified block.
indicating that migration outside the vitrification zone was not a significant
problem.  A few samples directly adjacent to the block contained measurable
concentrations up to 0.7 ppm, which is acceptable.  These results indicated
that the vitrification rate must be higher than the diffusion rate of vola-
tilized PCBs in soil, thus overcoming migration away from the hot molten mass.
    The product of this process is a solid glass and crystalline block.  This
form may be more costly to redeposit.   There may be fewer options than would
be available for ordinary sedimental material.  The nature and extent of emis-
sions from the melt would likely vary from one type of sediment to another.
Sediments  containing significant amounts of organic matter would lose ail of
                                     133

-------
                             TAIII.I- 37.  SAMPLES/ANALYSES HATTEU.E  IN  SITU VITRIFICATION PROCESS,

                                                      DEMONSTRATION TESTS
CO
en
Sample Number Analyses
Feed (raw sediaent) 2 y,' ,{»• ' 2
l< \, '•! *
Off gas. raw 8b ft Bflorisil
8 scrub solution
.if I or scrubber 2 2
2
2
after HEPA filter 2 2
2
2
2
si rnl»l)i-i water 2 2
2
2
Method
Pest ir ides /PCDs. (608/8080
:' w/cogener scan)
Metals (24 by ICP. AA)
Pesticidos/PCIIs. (608/8080
w/cogener scan)
Metals ('24 by ICP. AA)
Sulfates. Nitrates
Pesticldes/PCBs. (608/8080
w/cogener scan)
Metals (24 by ICP. AA)
Sul fates. Nitrates
Pestlcidcs/PCBs, (608/8080
w/cogeiier scan)
Metals (24 by ICP. AA)
Sulfates. Nitrates
Poly chlorinated dibenzo
fimuns and dihenzo p dioxins
HRGC/HRMS
Pesticides/PCBs. (6O8/8080
w/r:ot;ener scan)
Metals (24 ICP. AA)
Sulfates, Nitrates
Estimated
cost
of analyses'1
$ 500
370
2,000
1.480
576
500
370
144
50O
370
144
2 . 200
500
370
I'M
                                                                                                              (ConI iiiueii)

-------
                    $1,519.000 x 12 X (319/325.3)
                   	 » $47.04
                               380,000


where:   319/325.3 - the Chem. Eng. plant cost index ratio:
                     April. 1987/Annual. 1985.
    Labor costs were calculated using the standardized rates for this study.

applied to labor requirements projected by Battelle for a single treatment

system.  For the set of twelve systems, the management requirements were

reduced to a staff of two instead of one per system,  and an extra operator was

added to provide two per shift per system.  Six chemists were included for

qualify control, one for each set of six units per shift.  Maintenance was

estimated at one man for each two systems per shift.   The total daily labor

and labor per m3 of soil treated, was estimated as follows:


            Labor category          Hours/day          Cost. S/day

            Manager                     16               $   960
            Maintenance .               144                 2.160
            Operators                  576       "         8.640
            Chemists                    48                 1,200

            Total                                        $12.960

            Cost/m3 treated (5% moisture)
                S12,960/(12 X 52.4) -                     $20.61

            Cost/m3 treated (25% moisture)
                $12,960/(12 X 41.8) -                     $25.84


Consumable costs were estimated using the standardized electricity cost of

$0.09/kWH. and other consumables as estimated by Battelle:

       Electrodes                                          $98/m3
       Secondary waste                                     $1.85/m3
       Electricity. 5* moisture
         (392.000 kWH x 48 settings x $0.09)/13,500 m3     $125.44/m3
       Electricity, 25X moisture
         (414.000 kWH x 48 settings x $0.09)/12.200 m3     $146.60/m3


    A site cost of S2/m3 was included as Battelle's estimated for transport of

equipment, site clearing,  and acquiring/applying backfill material as needed

    The total cost of application is estimated as follows:
                                     137

-------
 scrubber and HEPA filters were effective in removing the residual zinc and
 other elements entrained in the off-gas.   A schematic of the off-gas treatment
 system for this process is shown in Figure 21.
     During ISV operations,  the mass mean particle diameter ranged from <0.l to
 0.8 urn.   When combustible wastes were present,  the entrained particle size was
 larger,  averaging 1.4  urn.
     During ISV of a  waste site,  any solid combustible inclusions  within the
 soil are pyrolyzed into combustible gases at the high temperatures of the
 •elt.  The pyrolysis gases  move upward through  the molten zone, expanding as
 they are heated.   Combustion does not occur until the pyrolyzed gases contact
 air at the surface of  the molten soil,  since the molten glass  is  reducing in
 nature.   With a cold cap or an insulated  surface,  the release  of  gases occurs
 primarily near the electrodes,  because  the glass is hotter and has a lower
 viscosity in the  areas  of  highest current density.   When a cold cap is not
 present,  release  of  gases  is more uniform over  the molten surface.   Several
 effects  of these  gas releases must  be considered in establishing  the design of
 the hood  and off-gas system:
     •  Pyrolysis  gases  carry with them  to the off-gas system a portion of
       elements associated  with  the combustible waste.   (Note  that only the
       contaminants  associated with the combustibles  are available for release
       -  those already  incorporated in  the melt remain in the  vitreous mass.)
     •  The protective,  subsided  cold  cap  may be broken up by active gas
       releases,  thereby increasing heat  losses and hood temperatures.
     •  Pyrolysis  gases  that  are  superheated in  the  molten zone burn in the
       hood  plenum,  thereby  creating  high temperatures in the  hood and
       increasing the heat  removal  requirements of  the off-gas system.
     •  The gas generation rate of buried  combustibles  and the  air  required to
       combust the gases determine  the  maximum  off-gas  flow  rate  required for
       that  application.
    The magnitude of the effects  of the gas  releases  is  directly  proportional
 to  the rate  at which pyrolyzed gases  are  generated  and  released.    3attelle's
engineering-scale and pilot-scale tests have  shown  that  combustible  gas
release is sporadic and  may  occur in  a  very  short  time  period.  During an
engineering-scale test,  0.2 kg of simulated  combustible waste  was  piarori
inside a metal canister.  Active  surface  combustion of  the pyrolyzed  gases
occurred over an  18-min  period during i 12-h  test.  Thus  the release  period
                                     139

-------
was only 2.5* of the total ISV time.  Similar observations have been made
during the pilot-scale tests.
    Vitrified soil blocks were analyzed to determine their chemical durability
with a series of tests including 24-h soxhlet leach tests. Materials Charac-
terization Center tests (MCC-1) (NCC 1981), and hydration.  The soxhlet leach
rate for all radionuclides was less than 1 x 10~5 g/c»2/day, which is an
acceptable value.  A 28-day MCC-1 test was also conducted on a contaminated
soil sample that was vitrified in the laboratory at 1600 *C.  The overall
leach rate of the vitrified soil (2 x 10~7 £/cm2/day) was higher than the
rates for the borosilicate and aluminosilicate glasses.  Longer vitrification
times at temperatures like those experienced in the field are expected to
lower the observed Pu leach rate making it more comparable to HLW glasses.  As
an example of the excellent leach characteristics of field samples, TRU leach
rates from the vitrified block produced during the pilot-scale radioactive
test (PSRT) were too low to be detectable.
3.10.8  Health and Safety Characteristics
    With the control? applied to exit gases from the treatment, the process
offers good health and safety characteristics, especially for sediments
containing no radionuclides.   Since radioactive cesium is present in the
Hudson river sediments,  Battelle's safety analysis for potential exposure to
radionuclides is summarized here.
    To analyze the occupational and public safety of routine and nonroutine
ISV operations for both the short and the long term. Battelle selected a
representative transuranics (TRUs) contaminated waste site as a reference (Oma
et al.  1983).  Radionuclide release rates from the soil during vitrification
were estimated.   Ten times the waste inventory for the site was the basis for
the radionuclide source term to account for concentrated TRUs around the
distribution pipe(s).
    Tables 38 and 39 give the radiation doses from routine operations in  the
short term for the ISV worker and the public, respectively.  For all routine
exposures, radiation doses are estimated to be well below the federal guide-
lines set by the Department of Energy (DOE).  Of all activities assoc.jted
with ISV operations, the maximum occupational dose is expected to occur while
the worker is placing electrodes in the soil.  The low exposure levels can oe
seen in Table 38. where the occupational dose for this activity is compared tj
                                     141

-------
           TABLE 40.   OCCUPATIONAL DOSES FROM ACCIDENTAL RELEASES
              (120-h run.  15 settings,  concentrated inventory)
Accident
Uncontrolled venting
Break in off-gas line
Excess overburden
removal
Number of
personnel
1
1

2
Length
of
exposure
1 nin
5 ain

10 *in
Ist-year dose commitment
to each worker, rera
Total
1 X
6 X

3 x
body
ID'3
10-3

ID'3
Bone
2 x 10~2
1 x 10-1

4 x 10"2
Lung
2 x 10°
1 x 101

5 x 10°
   TABLE 41.  PUBLIC DOSE COMMITMENTS FROM POSTULATED ABNORMAL OCCURRENCES
                                      Maximum exposed
                                      individual,  rem
                    Population,
                      •an-rera
    Uncontrolled Venting
      ist-yr dose (lungs)
      50-yr dose (bone)
5 x 10'5
5 X 10-4
2 X 10'1
2 x 10°
Off-Gas Line Break
Ist-yr dose (lungs)
50-yr dose (bone)
Excessive Overburden Removal
ist-yr dose (lungs)
50-yr dose (bone)

3 x 10~2
3 x 10-1

1 x ID'2
9 x 10~2

1 x 102
1 x 103

3 x 10A
3 x 102
3.10.9  When Process Can Be Made Available

    Using the standardized T and E time requirements (7.5 months) and adding
12 months to build and demonstrate a full-scale system (Timmerman 1987). the
projected schedule to ready this process for this application is 19-24 months,
depending on whether all 12 systems are constructed at once,  or 11 are con-
structed after demonstration of the first.
         Test and Evaluation
         Report
         Approval by EPA
         Design,  procure,  fabricate full-
           scale  system and conduct onsite
           demonstration
         Construct additional units

         Total
            2.5 months
              2 months
              3 months
           11.5 months
              5 months

           19 - 24 months
                                     143

-------
    Value of d
      1.0-0.99       Represents the ultimate level of the characteristic y.
                     Improvement beyond this point would have no appreciable
                     value.
     0.99-0.80       Acceptable and excellent.  Unusually good performance.
     0.80-0.63       Acceptable and good.
     0.63-0.40       Acceptable.  Some improvement is desirable.
     0.40-0.30       Borderline acceptability.
     0.30-0.01       Unacceptable.  This one characteristic could load to
                     rejection of the process.
    The scale of d so developed is a dimensionless scale to which any charac-
teristic may be transformed so that it may be interpreted in terms of its
desirability for the intended application.  In this evaluation, the most cost-
effective final process was sought that could be available in the shortest
reasonable time.
    A characteristic assessed on a numerical scale was transformed to the
scale of "d" by the basic equation:
                             _C0.77941 [(-y^ •<

In this equation:   yj is a value of a treatment process characteristic i;
                       is the acceptable valuable of yi; and
                       is the borderline value of y< .
    Table 42 shows the acceptable and borderline values of yj for each charac-
teristic rated.
           TABLE 42.  ACCEPTABLE AND BORDERLINE VALUES FOR PROCESS
                               CHARACTERISTICS
Characteristic
Probability of cleaning to £ 2 ppm
Probable cost of treatment $/m3
T and E effort. S/1000
Test system availability, rating
Time to provide commercial system, months
Acceptable
Value3
0.9
100
300
0.9
18
Borderline
Valueb
0.3
300
900
0.3
36
ad = 0.63 for these vaJues.
bd = 0.37 for these values.
                                     145

-------
                      TABLE 43.  OVERALL DESIRABILITY OF INBJIATE T AND E OF THE EIGHT CANDIDATE PROCESSES
KPSG,
Qalson
Probability of clean-
Ing to < 2 ppe 0.9
d racing 0.63
Probable ml or
traataent, I/a3 160-191
d racing 0.54*
T and E effirt
$1000 216
d rating 0.66
Availability of a
jyitei for a taet
future purchaae by
govern, reouired
future pjrchaee by
govern, not required 0.9
d rating O.S3
Libly future aveil-
abi nty of the praoeat
aantht 19.S
d racing 0.62
Overall darirability, 0
earl last future avail. 0.61S
lataet future avail. 0.815
average 0.61S
UV/Ozane-
Ptadar Hydrogen/
Supercritical Ultraamia CF5 Lar€nergy In Situ
Hatr SitKlean Technology Extraction B.E.S.T. Extraction Vitrification

0.8 0.8 0.8 0.8 0.8 0.9 0.9
O.S9 0.63 O.S9 O.S9 0.59 0.63 0.63

86-135 156 90-120 153-264 133 50-57 U3-483
0.62 0.57 0.63 0.50 0.59 0.6S 0.16

«3 166 151 123 149 170-827° 400
0.56 0.68 0.69 0.69 0.69 O.S4 0.59
*


0.8 0.8

0.9 0.9 0.9 0.9 0.9
0.63 0.63 0.63 0.63 0.63 0.59 O.S9


21.5 19 21-24 25 19 25 19-24
0.53 0.62 0.59-0.55 0.54 0.62 0.54 0.62-0.55

0.60 3.617 0.625 0.59 0.623 0.614 0.46
0.60 0.617 0.616 0.59 0.623 0.514 0.15
0.60 0.617 0.621 0.59 0.623 0.614 0.46
a*wage coat used for rating.




'tost of $170,000 if developed by spmsoring firm.  A cost of  $280,000 ws used in tN evaluation TO allca for the \jeertaTty.
                                                         147

-------
 (Table 43).  indicating that all the processes might reasonably be expected  :o
 meet the requirement.  Tho availability of a test system was not considered as
 important as the total test and evaluation cost.  The  time required  to make a
 commercial process available showed a range of only six months, and  was judged
 of  lesser importance than the two major costs assessed.  All ratios  among the
 five factors that resulted from these assignments are  shown below as a matrix.
 For example, the ratio (test system availability)/(T and E cost) is  shown as
 the intersection of Row 4 and Column 3 as 0.2.
Clean to 2 ppm
Cost
T and E Cost
Test system availability
Early con. availability
Clean
to
2 ppm
1
5
5
1
1.25


Cost
0.2
1
1
0.2
0.25

T & E
Cost
0.2
1
1
0.2
0.25

Test System
Availability
1
5
5
1
1.25
Early
Commercial
Availability
0.8
4
4
0.8
1
    From these ratios and the following tabular algorithm, the factor weights
(W) were generated.
             Factors

    Clean to 2 ppm
    T and E Cost
    Future commercial proc.
    Test system availability
    Cost
Ratios
0.2000
                w
4.000
1.25
0.2000
5.000
0.20
1.00
0.25
0.20
Weights. V

   0.0755
   0.3774
   0.0943
   0.0755
   0.3774
                                                    2.65

The procedure for weight generation is as follows:

    •   Construct an intermediate weighting scale  (the w-column) by tht-
       following procedure.  Opposite that last factor enter a "1".  The
       remaining numbers in this column are formed by the product of its
       predecessor and Ratio value opposite it in a sort of zigzag route up
       the column.  For example, the first w-value, 0.20 is the product of the
       second w-value (1.00) and the first Ratio-value (0.2000).

    •   Total  the w-values.   This total is 2.65.  Construct a column of
       standardized weights by dividing each element of the w-column by this
       total  to obtain the  W-column.  The elements in the W-column will.
       perforce,  total one.
                                     149

-------
    d-SSYS requires s comparison between two simple lotteries for each factor
rated.
                           ;   50% chance of most undesirable rating.
             Lottery 1  *  .
                           I   50\ chance of most desirable rating.
                           i
             Lottery 2  -  \   X value of the rating for certain.
    Using probable treatment cost as an example, RTI selected for Lottery 1:
                  50* chance of a treatment cost of S313/m3
                  50* change of a treatment cost of S80/m3
and an X value equal to the mathematical expectation of Lottery 1 for
Lottery 2:
                   (0.5 X S313) * (0.5 x 80) * $196.50/m3.
    The value of $196.50/m3 on the y1 scile is
                                $313 - $196.5
                                  $313 - S80
                                                 0.5
    The utility of Lottery 2 is easily determined, since it is equal to the
utility of Lottery 1:
           (0.5)(utility of $313/m3) - (0.5)(utility of $80/m3) =
                       (0.5 x 0.0) • (0 5 x 1) « 0.5.
From Equation 8:
                           f = (In utility)
-------
types of materials.  The UV energy must penetrate the treatment media suffi-
ciently to reach the molecules of the hazardous compound.  Filtration or
cyclone-separation may be required following extraction in order to provide a
suitable medium for treatment.
    The Low Energy Extraction process should apply generally to any organic
waste that can be dissolved into a hydrophilic solvent.
    The MODAR process should apply generally to any organic waste that can be
oxidized at supercritical conditions.
    The CFS Extraction process has been applied to the removal and concentra-
tion of oily contaminants for which propane is the recommended solvent.
Hazardous solvents and oxygenated compounds are extracted using pressurized
carbon dioxide.
                                     154

-------
DeWolf, G., et al.  The Cost Digest:  Cost Summaries of Selected Environmental
   Control Technologies, EPA-600/8-84-010. U.S. Environmental Protection
   Agency. Research Triangle Park, NC. 1984, 92 pp.

Evans. L.  Ozone  in Water and Wastewater Treatment, Ann Arbor Science. Ann
   Arbor, Michigan (1972).

Excalibur Enterprises, Inc.  Ozone/Ultraviolet/Ultrasonics, 314 W. 53rd St.,
   New York, New York 10019, 1987.

Finkbeiner, H. L.  Personal Communication. General Electric Research,
   Schenectedy, New York, May 14, 1987.

Folger, H. S.  Recent Advances in Sonochemical Engineering, Chem. Eng.
   Progress Symp. Series 109, Vol. 67, 1974, Pages 1-12.

Glaze, W. H., G. R. Peyton, B. Sohm. and D. A. Meldrum, Pilot-Scale Evaluation
   of Photolytic Ozonation for Trihalonethane Precursor Removal, 600/52-84-
   136, Municipal Environmental Research Laboratory, U.S. Environmental
   Protection Agency. Cincinnati, Ohio, 1984.

Hackman. E. E. III.  Toxic Organic Chemicals, Destruction and Waste Treatment.
   Noyes Data Corp..  Park Ridge,  New Jersey. 1978.

Harrington, E. C.. Jr.  "The Desirability Function. Industrial Quality
   Control."  April 1965, pp. 494-498.

Kitchens, J.  Dehalogenation of Halogenated Compounds. U.S. Patent 4,144.152.
   March 13. 1979.

Kitchens, J. F., G. L. Anspach, L. B. Mongoba. and E. A. Kobylinski.  Cleanup
   of Spilled Chlorinated Organics with the LARC Process.  In:  Proceedings of
   the 1984 Hazardous Materials Spills Conference:  Prevention, Behavior,
   Control, and Cleanup of Waste Sites, Nashville. Tennessee, April 9-12.
   Government Institute, Rockville.  Maryland, pp. 110-115, 1984.

Klee. A. J.  d-SSYS.  A Computer Model for the Evaluation of Competing
   Alternatives. HWERL.  U.S. Environmental Protection Agency. Cincinnati,
   Ohio. 1987.

Kornel, A.  Memorandum of December 4. 1986 to Charles J. Rogers. APEG
   Treatment of Guam Contaminated Soil.

Levenspiel. 0.  Chemical Reaction Engineering. J. Wiley and Sons, New York.
   1962.

Mitchell. J. K.. Fundamentals of  Soil Behavior, J. Wiley and Sons. Inc.. N>w
   York, 1976.

Model 1. M.  Process for Regenerating Adsorbants with Supercritical Fluids.
   L' S  Patent 4.124,528. November 7. 1978.
                                     156

-------
Staszak, C. M., K, C. Malinowski. and W. R. Killilea.  The Pilot-Scale
   Demonstration of the MODAR Oxidation Process for the Destruction of
   Hazardous Organic Waste Materials.  Environmental Progress. 6, 1. February
   1987. pp. 39-43.

Timmerman, C. L.   In Situ Vitrification of PCB-Contaminated Soils, Battelle
   Pacific Northwest Laboratories, Electric Power Research Institute, Project
   RP 1263-64, Electric Power Research Institute, 3412 Hillview Ave.. Palo
   Alto, California, October 1986.

Timmerman, C. L.   Letter of July 9, 1987 to Don Wilson.

Tofflemire, T. J., and Quinn, S. 0., PCB in the Upper Hudson River:  Mapping
   and Sediment Relationships, Technical Paper No. 56, New York State
   Department of Environmental Conservation. April 1979.

Tose, M. K.  Letter of June 24, 1987 to Donald Wilson. Review of Draft Report.

Treybal. R. E., Mass Transfer Operations. 2/c, McGraw-Hill, Inc., New York,
   1968.

U.S.  Congress, Office of Technology Assessment, Superfund Strategy, OTA-ITE-
   252,  April 1985.

U.S.  Department of the Interior Water and Power Resources Service  Earth
   Manual:   A Water Resources Technical Publication. 2nd edition, 1974.

U.S.  Environmental Protection Agency, Guidance on Feasibility Studies Under
   CERCLA,  EPA/540/G-85/003, U.S. Environmental Protection Agency, Cincinnati,
   Ohio, June 1985.

U.S.  Patent 3899419, Method for Chemical Fractionation, Defalting and
   Dewatering of Solids and Suspensions, C. F. Emanuel, Belleville, WA.
   Assigned to Resources Conservation Company, August 12, 1975.

U.S.  Patent 3925201, Method of Dewatering Material Containing Solid Matter and
   Bound and Unbound Water, R. K. Ames,  Seattle, Washington.   Assigned to
   Resources Conservation Company, December 9, 1975.

U.S.  Patent 4002562. Oil Emulsion Processing.  R. K.  Ames and H. H. Peters.
   Seattle, WA.  Assigned to Resources Conservation Company,  January 11.   1977.

U.S.  Patent 4056466, Method of Dewatering Material Containing Solid Matter ana
   Bound and Unbound Water, H. H. Peters. Federal Highway.  WA.   Assigned  to
   Resources Conservation Company, November 1. 1977.

Unterman.  R.,  et al.   Microbial Degradation of Polychlorinated Biphenyls.
   Proc.  10th Conf. on New Frontier? for Hazardous Waste Management.
   Pittsburgh.  Pennsylvania.  Septemuer 15-18,  1985,   EPA/600-9-85-025.
                                     158

-------
              TABLE A-l.   HUDSON RIVER SEDIMENT CHARACTERISTICS
Size.
Sample No. mm
11PCBT01
Dried at 60 »C +2.0
-1.18
+0.059
+0.042
+0.210
+0 . 075
-0.075
Dried at 100 *C +2.0
+1.18
+0 . 595
+0.420
+0.210
-0.210
Weight
% on
Sieve

12.0
12.1
32.1
15.5
19.0
7.1
2.2
10.1
12.1
37.6
17.2
15.4
7.0
Float-
* able
Volatile Solids
Solids *

23.8
34.4
12.7




25.76 57.0
9.33 27.5
5.94 9.8
1.59 0.7
1.61 2.4
3.58 2.5
Total
Weight of
Extracted
Material.
ppm

2227
2958
862
94
550
1700
1680






Total
PCB.
PP»

36.9
36.1
41.1
4.85
5 . 73
25.85
35.55






Note:  Extractions were Made with hexane acetone.  Volatile solids were
       determined by ashing in a muffle furnace at 580 *C.
                                     161

-------
                          TAIII.h A-3.  HUDSON  KIVKK SEDIMENTS. METAL CONCENTRATIONS AND PCB CONTENT
O)
CJ
S.imple
Nnmher
25
20
27
29
K. Ca. Ti. Mn. Fe.
% * * UB/B *
8.0 4.4 1.7 1930 6.5
8.6 5.7 2.2 1125 5.5
8.1 4.8 1.5 950 4 . 5
3.9 1.8 0.6 580 2 . 1
Cu, Zn.
UB/E UB/g
120 750
80 280
45 200
10 <10
Rb.
UB/K
130
140
140
110
Sr,
Ug/g
385
410
390
310
PCB
ug/e
240
80-34
36-66
142-37
Cr.
Ug/g
7000
1540
760
170
Pb.
UB/K
1600
525
400
120
Ni.
ue/B
60
60
45
20
      Ann lysis  by  X i iiy fluorescence

-------
Automatic Waste Feed Cutoff System:

    «  Description of the automatic waste feed cutoff system when process
       conditions deviate beyond the safe operating limits and aelay time
       prior to cutoff.

    •  Description of the procedures to shut off the waste feed line and the
       whole process in the event of an equipment malfunction.

Destruction System:

    •  Narrative description of the destruction system (e.g., description of
       chemical reactions, stoichiometry. reagents, catalysts, process design
       capacity, etc.).

       Engineering diagrams.

       List of products and by-products and their concentrations.

       Description of how essential parameters (e.g., temperature pressure.
       flow rate, etc.) are monitored and the design values.

       Description of reactant/oxidant/fuel/catalyst/feed rates and how they
       are monitored.

    •  Design capacity of the system.

    •  Detailed description of the unique engineering features of the process
       (e.g.. high temperature, pressure, long residence time, heat transfer.
       etc.).

    •  Description of any regeneration/recycling processes applied in the
       process.

Pollution Control System (PCS):

    •  A description of the pollution control system for process effluents
       (air emissions,  liquid effluents, sludge, solid waste, etc.)

       Design parameters.

    •  The important operating parameters of the PCS and how they will b«»
       monitored.

Summary of Process Operating Parameters;

    Provide a summary which lists target values as well as upper and lower
    boundaries for all  major measured operating parameters, inst ument sot-
    tings, and control  equipment parameters.  All values must be reported  in
    common, consistent  units.   The application must also describe the action
    to be taken whenever the parameter deviates outside the control limits
    These actions may include adjusting the operating conditions, stopping tne
    PCB feed, shutting down the process, etc.  The time allowable for correc-
    tive action before  shut-down or other action must be specified.
                                      165

-------
                                    T
T
Monitor ing location

       Ueotioa
                                                                  fcik Storogo
  Figure  A-l. Schematic of sampling and monitoring  locations  for a chemical
             dechlorination process.  (Guidelines  for  Permit Applications
             and Demonstration Test Plans for PCB  Disposal  by Alternate
             Methods.   U.S. EPA, Office of Toxic Substances, Contract
             Number 68-02-3938-6. )
                                      167

-------
     •  The sampling design for each unit.  This may require a mathematical
       sampling design or simply a reference to a standard protocol.  The
       frequency (e.g., every 15 min), size (e.g.. 10 m3), timing (e.g.. any
       time after reaching steady-state), number of replicates (e.g.. tripli-
       cates for 10% of the samples or 2 samples, whichever is greater), num-
       ber of surrogate-spiked samples, and total number of samples should be
       listed for each sample type.

     •  An estimate of the sample representativeness.   This may be based on
       data (e.g..  historical data on replicates) or scientific/engineering
       judgment (e.g., a sample from an actively mixed feed tank could be
       characterized as "highly" representative).

     •  Contingencies for action if samples cannot be collected according to
       plan (e.g.,  alternate sites or times or an entirely new sampling plan).

Sampling Procedures:

    Details of the sampling methods to be used on a routine basis should be
    discussed in this section.  Include an explanation of the apparatus, cali-
    bration procedures, and maintenance procedures, if applicable.

    When "standard methods" will be used, they may be referenced and included
    as an appendix.   However, any deviations from standard procedures must be
    noted.  Furthermore, when the standard method allows different procedural
    variations to be use,  the developer must be specific as to the procedures
    which will be followed.

    The discussion of sampling and analysis methods should include the follow-
    ing.

    •  Sampling equipment.
    •  Sampling equipment calibration.
    •  Sampling procedures.
    •  Sample recovery, storage, and preservation.
    •  Sample transport and custody.
    •  Analytical equipment.
       Reagents.
    •  Reagents preparation.
    •  Calibration standards.
       Calibration procedures.

Sample Analysis Procedures:

    Summarize the analytical  procedures (including sample preparation) which
    will  be used for each sample.   The summary should include the analytical
    method,  apparatus,  data reduction procedures, data storage, equipment
    calibration,  and equipment maintenance.  Specific details of the analyti-
    cal procedures  need not be inciuoea in this section, but should be refer-
    enced (if standard published procedure) or should be included as an
    appendix,  if unpublished  or if the publication is not readily available.
                                     169

-------
       Pollution control system.
       Process alarms.
       Fire extinguisher system.


5.  Spill Prevention Control and Countermeasures Plan

    Describe the procedures (including system design) which will be used to
prevent spills of PCBs.  Also describe the procedures which will be followed
should a spill occur.  Coast Guard regulations specifying spill prevention
control and counter-measure plans (40 CFR 112.7) can be used as an example for
the type of information which should be addressed; however, the plan provided
in the permit application need not be in the format or detail specified in
40 CFR 112.7.

Safety Plan:

    This section addresses the safety program which will be initiated to pro-
    tect workers and other humans from PCB exposure or other health hazaras.
    Identify specific items (e.g.,  protective clothing) of the program for
    ensuring safe routine operations.  Procedures for preventing worker/
    population exposure in the case of an equipment malfunction also should be
    addressed; procedures for stopping waste feed, shutting down the process,
    and controlling emissions in the event of a malfunction should be address-
    ed.  Provisions for prevention and control of fires, explosions, electri-
    cal outages, etc., also should be addressed.  .                                    )

Training Plan:

    Present a description of the training program which will be initiated to
    assure workers are trained in items appropriate to their jobs including
    the following.

       Number of persons to be trained and time required.
       Equipment operation (in accordance with standard operating procedures).
    •  Emergency shut-down procedures.
    •  Use of protective clothing.
    •  Waste handling.
    •  Spill prevention/control.
    •  Fire control.
       Hazards of PCBs.


6.  Demonstration Test Plans

    Briefly summarize the plans for conducting a demonstration test.  Summary
information which should be presented in this section include the following.

    •   Testing time, days.
       Tentative location for the test.
       Parameters to be tested.
       Type waste to be used.
       Earliest date test could be  made.
                                     171

-------
    «   Preventative maintenance procedures and frequency.
    •   Specific routine procedures to assess accuracy, precision, and
       completeness.
    •   Procedures for corrective action.
    •   Quality assurance reports to management.

Standard Operating Procedures:

    A summary of the standard operating procedures (SOP) should be included.
    The SOP should consist of the procedures available to the facility oper-
    ators for use in plant operations.  A process operating manual, if avail-
    able, will be satisfactory.

    The SOP:

    •   Assures that applicants have reviewed the operations in detail;

    •   Gives  EPA opportunity to review and become familiar with the operations
       prior  to the on-site audit; and

    •   May be used as a tool for training new employees, which gives some
       assurance that the employees have received a minimum of training.

    An SOP should be a step-by-step procedure; however, details of procedures
    such as the use of sampling or monitoring equipment may be omitted but
    must be referenced.  Divergence from the SOP during trials or commercial
    runs should be documented and significant modifications should be sub-
    mitted to EPA.   For convenience of use. lab procedures should be separate
    from system operational procedures.

    The SOP should be part of the training plan.   Each employee should sign
    and date  a statement indicating that the employee has read and understood
    the SOP.
    Reference

    Tofflemire,  T.  J..  and Quinn.  S.  0..  PCB in the Upper Hudson River:
      Mapping and Sediment Relationships. Technical Paper No. 56. New York
      State Department  of Environmental Conservation. April 1979.
                                                    .
                                                Region 5, Library
                                                77  YVES t Jackson
                                                Chicago, IL  o;J
                                     173

-------