TECHNOLOGY EVALUATION REPORT:
BIOTROL SOIL WASHING SYSTEM
FOR TREATMENT OF A
WOOD PRESERVING SITE
" Volume r "-
by
Science Applications International Corporation
McLean, VA 22102 and Paramus, NJ 07652
EPA Contract No. 68-03-3485, WA# 0-21
and 68-CO-0048, WA# 0-13
Project Officer
Ms. Mary K. Stinson
Risk Reduction Engineering Laboratory
Edison, New Jersey 08837
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
NOTICE
The information in this document has been funded by the U.S. Environmental
Protection Agency under the auspices of the Superfund Innovative Technology
Evaluation (SITE) Program under Contract numbers 68-03-3485 and 68-CO-0048 to
Science Applications International Corporation. It has been subjected to the
Agency's peer and administrative review, and it has been approved for publication
as an EPA document. Mention of trade names or commercial products does not
constitute an endorsement or recommendation for use.
-------
FOREWORD
The Superfund Innovative Technology Evaluation (SITE) program was
authorized in the 1986 Superfund amendments. The program is a joint effort
between EPA's Office of Research and Development and Office of Solid Waste and
Emergency Response. The purpose of the program is to assist the development of
hazardous waste treatment technologies necessary to implement new cleanup
standards which require greater reliance on permanent remedies This is
accomplished through technology demonstrations which are designed to provide
engineering and cost data on selected technologies.
This project consists of an analysis of one configuration of BIoTrol
Incs proprietary soil washing system. The demonstration took place af the
MacGillis & Gibbs Superfund site in New Brighton, Minnesota, a wood treatment
facility operating since approximately 1920. The demonstration effort was
directed at obtaining information on the performance and cost of the process
sequence for use in assessments at other sites. Documentation for the project
consists of this Technology Evaluation Report and an Applications Analysis
Report. The Technology Evaluation Report provides a detailed analysis of the
data acquired during the demonstration; discusses the technical aspects of the
technology; and describes the field activities and laboratory results The
Applications Analysis Report summarizes the results of the demonstration and
looks more broadly at the applicability of the technology to other sites and the
criteria (e.g., site and waste characteristics, economics, etc.) that would need
to be addressed.
For further information, please contact the Superfund Technology
Demonstration Division at the Risk Reduction Engineering Laboratory.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
iii
-------
ABSTRACT
carried out aTth ^ MacGmTs' * Pihh sl"rry bl°-reactor. The demonstration was
WacGillis & Gibbs Superfund site in New Brighton MN
this SITE . .
o The Soil Washer (SW) can achieve removal efficiencies
s°u
iv
-------
A small quantity of a fine particle cake (-7% of output solids) retaining
a major portion of the penta and PAH contamination (about 30% of the total
output) is also obtained. With the MacGillis and Gibbs soils, two
additional fractions containing significant amounts of penta and PAHs were
also obtained.
.The Slurry Bio-reactor (SBR) appears to be capable of achieving penta-
chlorophenol removal of >90% and PAH removals in the 70-90+% range;
however, the unit did not reach steady state operation and optimum
performance was not reached. ;
The BioTrol Aqueous Treatment System (BATS) can achieve up to 94%
degradation of pentachlorophenol in process water from the SW. Removal of
PAHs could not be determined because influent concentrations were below
detection limits.
The estimated cost to install and operate an integrated pilot-scale system
(SW, BATS, SBR), based on a 0.25-0.50 ton/hr Soil Washer operating for two
weeks to assess applicability for remediation is $745,000. For a full-
scale system the total cost (capital plus operating) to clean up 30,000
yd3, such as at the MacGillis and Gibbs site, is estimated at about
$168/ton, based on a 20 ton/hr Soil Washer, three 100 gpm BATS units, and
a 23 gpm SBR. This figure does include incineration of contaminated fine
and coarse oversize material, which constributes about 75% to the total
cost. Costs for a full-scale, operating system were not available at the
time this report was prepared.
-------
VOLUME I*
TABLE OF CONTENTS
PAGE
FOREWORD .......................................... ............ til
ABSTRACT ............ ............................................... lv
TABLE OF CONTENTS ................................. ____ . . ........... vi
LIST OF FIGURES ..........
""•«•••••"••••••••••••••••••••••••.»..»... H.
LIST OF TABLES [[[ xii
TABLE OF ABBREVIATIONS AND SYMBOLS ............................ ..... xv
CONVERSION FACTORS ............................... ............. ..... xvli
ACKNOWLEDGEMENTS ...................... ............................. xviii
SECTION 1. INTRODUCTION .......................... . ................. i
1 . 1 THE SITE PROGRAM ............................................. i
1 . 2 OBJECTIVES OF THE DEMONSTRATION ......... '.'.I'.]'.'.'.'.'.'.'.'.'.'. ...... 2
1 . 3 SITE DESCRIPTION ........ . ............... ....]................ 2
1 . 4 DESCRIPTION OF OPERATIONS ........... . ........ ........... ..... 4
i .5 PROJECT ORGANIZATION ......... '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.':'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. 6
SECTION 2 . EXECUTIVE SUMMARY ...................... ! ........... ...... 8
2 . 1 SITE DEMONSTRATION ........ . .................. ................ 8
2.2 SUMMARY OF RESULTS ........................... ................ 9
2.2.1 Soil Washer ............. ........ '.'.'.'.'.'.'.','.'.'.'.'.'.'.'.'.'.'.'.'.'. 9
2.2.2 BioTrol Aqueous Treatment System ...... , ................ 10
2.2.3 Slurry Bio-Reactor .................... ................ 10
2 . 3 CONCLUSIONS ..... . ............. ............ '...... 10
2.3.1 Soil Washer ____ ____ ....... '. . '.'.'.'.'.'. '. '. '. . ............... 10
2.3.2 BioTrol Aqueous Treatment System ...................... 11
2.3.3 Slurry Bio-Reactor ........................ H
2.3.4 Costs ........................... '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. 12
SECTION 3 . PROCESS DESCRIPTION .................... ' ................ 13
3 . 1 PRETREATMENT AND PROCESSING REQUIREMENTS ..... '...'.' ............ 13
3.1.1 Soil Excavation and Preparation ....... . ............... 13
3.1.2 Process Water ......................................... 13
3.1.3 Slurry for SBR ............. ............. .............. 13
*
Volume II is published in two separate document's - Part "A" and Part "B" ;
-------
TABLE OF CONTENTS (CONT'D)
PAGE
3.2 PROCESS DESCRIPTION 14
3.2.1 Introduction 14
3.2.2 Soil Washer 14
3.2.3 BioTrol Aqueous Treatment System 16
3.2.4 Slurry Bio-Reactor 20
3.3 ANCILLARY POLLUTION CONTROL EQUIPMENT 23
3.4 WASTE REMOVAL AND SITE CLOSURE 23
SECTION 4. THE BIOTROL FIELD DEMONSTRATION 24
4.1 BACKGROUND 24
4.1.1 Technology Description 24
4.1.2 Site Description 24
4.1.3 Demonstration Test Plan . 26
4.2 DEMONSTRATION PROCEDURES 26
4.2.1 Site Preparation 26
4.2.2 Field Operations 27
4.2.2.1 Soil Washer Demonstration 27
4.2.2.2 BioTrol Aqueous Treatment System Test 29
4.2.2.3 Slurry Bio-Reactor Test 28
4.2.3 Field Measurements 28
4.2.3.1 Soil Washer '.'.'.'.'. 28
4.2.3.2 BioTrol Aqueous Treatment System 28
4.2.3.3 Slurry Bio-Reactor 28
4.2.4 Sampling and Analysis 32
4.2.4.1 Predemonstration Sampling and Analysis 32
4.2.4.2 Demonstration Sampling and Analysis 34
4.2.4.3 Post Demonstration Sampling and Analysis-- .... 35
4.2.5 Health and Safety Protocols 38
4.3 MATERIAL CHARACTERIZATION 38
4.3.1 Feed Soils '.'.'.'.'.'.'.'.'.'.'.'. 38
4.3.2 Soil Washer Output Streams 44
4.4 PILOT SYSTEM PERFORMANCE RESULTS 52
4.4.1 Introduction 52
4.4.2 Soil Washer Performance 52
4.4.2.1 As -, is Material Balance 54
4.4.2.2 Dry Solids Material Balance 58
4.4.2.3 Penta Distribution - Low Penta SW Test 58
4.4.2.4 Penta Distribution - High Penta SW Test 64
4.4.2.5 PAH Distribution 67
4.4.2.6 Total Organic Carbon and Total Recoverable
Petroleum Hydrocarbons 77
4.4.2.7 Metals Behavior 77
vii
-------
TABLE OF CONTENTS (CONT'D)
PAGE
4.4.2.8 Flow Rate Stability 80
Low Penta SW Material Flow Rates
High Penta SW Material Flow Rates
Feed Soil Flow Rate Effect on Organics Removal
4.4.3 BioTrol Aqueous Treatment System Performance 92
4.4.3.1 Critical Analyses 92
4.4.3.2 Other Analyses 99
4.4.3.3 System Parameters Ill
4.4.4 Slurry Bio-Reactor Performance Ill
4.4.4.1 Critical Analyses Ill
4.4.4.2 Other Analyses 152
4.4.4.3 System Parameters 161
4.4.5 Dioxins 159
4.4.5.1 Introduction 169
4.4.5.2 CDD/CDFs in Soil Washer Streams 169
4.4.5.3 CDD/CDFs in the BATS 169
4.4.5.4 CDDs/CDFs in the Slurry Bio-Reactor 173
4.5 FACTORS AFFECTING THE DEMONSTRATION 176
4.5.1 Unit Operating Problems - Soil Washer 176
4.5.2 BioTrol Aqueous Treatment System - Problems . . . . '. 176
4.5.3 Slurry Bio-Reactor - Problems 177
4.6 POTENTIAL OVERALL BSWS PERFORMANCE 177
SECTION 5. ECONOMIC ANALYSIS 179
5 .1 INTRODUCTION 179
5.2 BASIS OF ECONOMIC ANALYSIS 179
SECTION 6. QUALITY ASSURANCE/QUALITY CONTROL 183
6.1 INTRODUCTION 183
6.2 CONCLUSIONS AND LIMITATIONS OF DATA 184
6.3 PROCEDURES DEFINING DATA QUALITY 185
6.3.1 Precision 185
6.3.2 Accuracy 186
6.3.3 Completeness 186
6.3.4 Comparability , 186
6.3.5 Representativeness 187
6.4 ANALYTICAL QUALITY CONTROL 187
6.4.1 Soil Pile Analyses . 187
6.4.2 Soil Washer Analyses 199
6.4.3 Aqueous Treatment System Analyses 210
viii
-------
TABLE OF CONTENTS (CONT'D)
PAGE
6.4.4 Slurry Bioreactor 218
6.4.5 Blank Results 229
6.5 SPECIAL STUDIES . 229
6.6 AUDITS AND RESULTING QAPj P DEVIATIONS 243
6.6.1 Biotrol Soils Audit 243
6.6.2 Laboratory Audit at 1st Subcontractor Laboratory
and Field Audit of Slurry Bioreactor 245
6.6.3 Technical Systems Review (TSR) at Second
Subcontractor Laboratory 245
6.6.4 Follow-up Audit of 2nd Subcontracting Laboratory 250
6.6.5 Audit of Soil Sieving Procedure for BioTrol Project .... 251
SECTION 7. BIBLIOGRAPHY 254
ix
-------
LIST OF FIGURES
PAGE
1-1. SITE DEMONSTRATION LOCATION AT MACGILLS & GIBBS 3
1-2. MACGILLIS AND GIBBS/BIOTROL SITE MAP 5
1-3. PROJECT ORGANIZATION '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. 7
3-1. FLOW DIAGRAM OF SOIL WASHER SYSTEM (SWS) 15
3-2. FLOW DIAGRAM OF BIOTROL AQUEOUS TREATMENT SYSTEM (BATS) ..... 17
3-3. BIOTROL INC. MOBILE AQUEOUS TREATMENT SYSTEM 18
3-4. SCHEMATIC OF BIOREACTOR ." 19
3-5. FLOW DIAGRAM OF THE SLURRY BIO-REACTOR (SBR) WITH
SAMPLE POINTS 21
3-6. SLURRY BIO-REACTOR PROCESS DIAGRAM !!!!!!!!!!!!!!! ':'. '.'. '. 22
4-1. FLOW DIAGRAM OF THE BIOTROL SOIL WASHING SYSTEM (BSWS) 25
4-2. PARTICLE-SIZE FRACTION ANALYSIS ! ! 43
4-3. PARTICLE-SIZE FRACTION ANALYSIS '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. 48
4-4. PARTICLE-SIZE FRACTION ANALYSIS '..'.'.'.'.','.'.'.'.'.'.'. 49
4-5. MASS DISTRIBUTION IN LOW PENTA SW TEST '.'.'.'.'.'.'.'.'.'. 56
4-6. DISTRIBUTION OF OUTPUT MASS - LOW PENTA SW TEST! '.'.'.'.'.'.'.'.'.'.'.'. 57
4-7. MASS DISTRIBUTION IN HIGH PENTA SW TEST 60
4-8. DISTRIBUTION OF OUTPUT MASS - HIGH PENTA SW TEST* !!!!!!!!!!!! 61
4-9. PENTA CONCENTRATION IN OUTPUT STREAMS - LOW PENTA SW TEST!"' 63
4-10. PENTA CONCENTRATION IN OUTPUT STREAMS - HIGH PENTA SW TEST ' 66
4-11. METALS IN LOW PENTA SW TEST 79
4-12. METALS IN HIGH PENTA SW TEST !.!!!!!!!!!!!!!!!! 79
4-13. SOLIDS STREAM RATES - LOW PENTA SW TEST! !!!!!!!!!.'!!!! 84
4-14. AQUEOUS STREAM RATES - LOW PENTA SW TEST !!!!!!! 8'5
4-15 . SOLIDS STREAMS - HIGH PENTA SW TEST '..!!!!!!!!!!!! 86
4-16. AQUEOUS STREAM - HIGH PENTA SW TEST !!!!!!!!!!!!!!!! 87
4-17. WASHED SOIL AND FEED SOIL PENTA CONG. - LOW PENTA'SW'TEST!!!! 89
4-18. WASHED SOIL AND FEED SOIL PENTA CONG. - HIGH PENTA SW TEST... 91
4-19. BATS - LOW PENTA TEST - PENTACHLQROPHENOL CONCENTRATION ..!!! 96
4-20. BATS - HIGH PENTA TEST - PENTACHLOROPHENOL CONCENTRATION 97
4-21. SBR - PENTA REMOVAL EFFICIENCY 123
4-22. SBR - PENTA IN LIQUID PHASE OF INFLUENT !!!!!!!!!! 124
4-23. SBR - PENTA IN SOLID PHASE OF INFLUENT !!!!!!!!!!!!!!!!!! 125
4-24. SBR - OVERALL PENTA REMOVAL EFFICIENCY ! ! ! . ! 128
4-25. SBR - ACENAPHTHENE INFLUENT SOLID PHASE !!!!!!!!!!!!!!!!! 136
4-26. SBR - FLUORANTHENE INFLUENT SOLID PHASE 1^,6
4-27. SBR - PYRENE INFLUENT SOLID PHASE !!!!!!!!!!!!!!! 136
4-28 . SBR - BENZO (A) ANTHRACENE INFLUENT SOLID PHASE ! 136
4-29. SBR - CHRYSENE INFLUENT SOLID PHASE ' 137
4-30. SBR - BENZO(A)PYRENE INFLUENT SOLID PHASE 137
-------
LIST OF FIGURES (CONT'D)
PAGE
4-31. ACENAPHTHENE REMOVAL EFFICIENCY 145
4-32. FLUORANTHENE REMOVAL EFFICIENCY 146
4-33. PYRENE REMOVAL EFFICIENCY 147
4-34. BENZO(A)ANTHRACENE REMOVAL EFFICIENCY 148
4-35. CHRYSENE REMOVAL EFFICIENCY 149
4-36. BENZO (B) FLUORANTHENE REMOVAL EFFICIENCY 150
4-37. BENZO (A) PYRENE REMOVAL EFFICIENCY. . 151
4-38. SLURRY BIO-REACTOR - pH DATA 166
4-39. SLURRY BIO-REACTOR - POWER USAGE 167
xi
-------
LIST OF TABLES
TABLE PAGE
2-1. SUMMARY OF CONTAMINANT REMOVAL EFFICIENCIES 9
4-1. SOIL WASHER FIELD MEASUREMENTS 29
4-2. BIOTROL AQUEOUS TREATMENT SYSTEM FIELD MEASUREMENTS 30
4-3. SLURRY BIO-REACTOR FIELD MEASUREMENTS 31
4-4. ANALYTICAL METHODS USED'DURING THE BSWS SITE DEMONSTRATION .. 33
4-5. SOIL WASHER SAMPLES 36
4-6. BIOTROL AQUEOUS TREATMENT SYSTEM SAMPLES 37
4-7. SLURRY BIO-REACTOR SAMPLES 37
4-8. LOW PENTA SOIL PILE CHARACTERISTICS 39
4-9. HIGH PENTA SOIL PILE CHARACTERISTICS 40
4-10. PARTICLE-SIZE ANALYSIS OF LOW PENTA SW 41
4-11. CHEMICAL ANALYSIS OF PARTICLE-SIZE FRACTION - LOW PENTA
CONCENTRATION SOIL SAMPLE 42
4-12. PARTICLE-SIZE ANALYSIS OF HIGH PENTA CONCENTRATION SOIL
SODIUM HEXAMETAPHOSPHATE SIEVING SOLUTION 45
4-13. PARTICLE-SIZE ANALYSIS OF HIGH PENTA CONCENTRATION SOIL -
ASTM TYPE II WATER SIEVING SOLUTION 45
4-14. CHEMICAL ANALYSIS OF PARTICLE-SIZE FRACTIONS - HIGH PENTA
CONCENTRATION SOIL SAMPLE SODIUM HEXAMETHAPHOSPHATE
SIEVING SOLUTION 46
4-15. CHEMICAL ANALYSIS OF PARTICLE-SIZE FRACTIONS - HIGH
CONCENTRATION SOIL SAMPLE ASTM TYPE II WATER SIEVING
SOLUTION 47
4-16. PARTICLE-SIZE ANALYSIS AND CHEMICAL ANALYSIS OF SOLID
PROCESS STREAMS - LOW PENTA SW TEST 50
4-17. PARTICLE-SIZE ANALYSIS AND CHEMICAL ANALYSIS OF SOLID
PROCESS STREAMS - HIGH PENTA SW TEST 51
4-18. COMPARISON OF AVERAGE AND WEIGHTED CONCENTRATIONS 53
4-19. FATE OF MATERIALS - LOW PENTA SW TEST 55
4-20. FATE OF MATERIALS - HIGH PENTA SW TEST 59
4-21. AVERAGE PENTA DISTRIBUTION - LOW PENTA SW TEST 62
4-22. AVERAGE PENTA DISTRIBUTION - HIGH PENTA SW TEST 65
4-23. AVERAGE PAH CONCENTRATION AND MASS - LOW SW TEST 68
4-24. AVERAGE PAH CONCENTRATION AND MASS - HIGH SW TEST 72
4-25. COMPARISON OF PENTA AND PAH DISTRIBUTION IN OUTPUT STREAMS... 77
4-26. AVERAGE METALS BALANCE-LOW PENTA SW TEST 78
4-27. AVERAGE METALS BALANCE-HIGH PENTA SW TEST 78
4-28. EP TOXICITY TEST RESULTS - LOW PENTA SW TEST 81
4-29. EP TOXICITY TEST RESULTS - HIGH PENTA SW TEST 82
4-30. FEED SOIL RATE - LOW. PENTA SW TEST 90
4-31. BATS - LOW PENTA TEST SYSTEM PERFORMANCE
AND MATERIAL INVENTORY-PENTACHLOROPHENOL 93
4-32. BATS - HIGH PENTA TEST SYSTEM PERFORMANCE
AND MATERIAL INVENTORY - PENTACHLOROPHENOL 95
4-33. BATS - LOW PENTA TEST SYSTEM PERFORMANCE
AND MATERIAL INVENTORY-ANTHRACENE 98
-------
LIST OF TABLES
PAGE
4-34. BATS - LOW PENTA TEST SYSTEM
PERFORMANCE AND MATERIAL INVENTORY - ARSENIC,
CHROMIUM AND COPPER 100
4-35. BATS - HIGH PENTA TEST SYSTEM
PERFORMANCE AND MATERIAL INVENTORY - ARSENIC, CHROMIUM
AND COPPER 10i
4-36. BATS PENTA MINERALIZATION ASSESSMENT '.'.'.'.'.'. 103
4-37. BATS - LOW PENTA TEST MATERIAL
INVENTORY-TOTAL SOLIDS 105
4-38. BATS - HIGH PENTA TEST MATERIAL
INVENTORY - TOTAL SOLIDS 106
4-39. BATS - LOW PENTA TEST MATERIAL
INVENTORY - TOTAL RECOVERABLE PETROLEUM HYDROCARBONS 107
4-40. BATS - HIGH PENTA TEST MATERIAL
INVENTORY - TOTAL RECOVERABLE PETROLEUM HYDROCARBONS 108
4-41. BATS - LOW PENTA TEST MATERIAL
INVENTORY - CHEMICAL OXYGEN DEMAND (COD) 109
4-42. BATS - HIGH PENTA TEST MATERIAL
INVENTORY - CHEMICAL OXYGEN DEMAND (COD) 110
4-43. BATS LOW PENTA TEST - SYSTEM FIELD MEASUREMENTS ... 112
4-44. BATS HIGH PENTA TEST - SYSTEM FIELD MEASUREMENTS ............. 113
4-45. SBR TIME-WEIGHTED MASS & VOLUME BALANCE 117
4-46. SBR LIQUID PHASE PENTA DATA '.'.'.'.'.'.'.'.'.'.'. 121
4-47. SBR SOLID PHASE PENTA DATA '.'.'.'.'.'.'.'.'.'.'.'.'.'. 122
4-48. OVERALL PENTA REDUCTION IN SBR '.'.'.'/.'.'.'.'.'.'.'.'.'.'.'.'. 126
4-49. SBR INFLUENT LIQUID PHASE PAH DETECTION LIMITS '.'.'.'.'.'.'.'.'.'. 129
4-50. SBR EFFLUENT LIQUID PHASE PAH DETECTION LIMITS '.'.'.'. 130
4-51. SBR INFLUENT SOLID PHASE PAH DATA 131
4-52. SBR EFFLUENT SOLID PHASE PAH DATA '.'.'.'.'.'.'.'.'. 132
4-53. SOLID PHASE PAH RESULTS SUMMARY 134
4-54. ACENAPHTHENE REMOVAL EFFICIENCY '.'.'.'.'.'.'.'.'. 137
4-55. FLUORANTHENE REMOVAL EFFICIENCY 138
4-56. PYRENE REMOVAL EFFICIENCY '.'.'.'.'.'.'.'.'.'. 139
4-57. BENZO(A)ANTHRACENE REMOVAL EFFICIENCY ','," 140
4-58. CHRYSENE REMOVAL EFFICIENCY '.'.'.'.'.'.'.'... 141
4-59. BENZO(B)FLUORANTHENE REMOVAL EFFICIENCY .'.'.'.'.'.'.'.'.'.'.','.'." 142
4-60. BENZO(B)FLUORANTHENE REMOVAL EFFICIENCY '.'.'.'. 143
4-61. PAH REMOVAL SUMMARY 153
4-62. SBR ARSENIC, CHROMIUM & COPPER MATERIAL BALANCES .'.'.'. . . . '.'.'.'. . 154
4-63. SBR TOTAL RECOVERABLE PETROLEUM HYDROCARBON MATERIAL
BALANCE • 155
4-64. SBR CHEMICAL OXYGEN DEMAND '.'.'.'.'.'.'.'.'.'.'. 156
4-65. SBR TOTAL SOLIDS '.'.'.'.'.'.'.'.'.'.'.'.'. 158
4-66. SBR TOTAL SOLIDS BALANCE '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.' 159
xiii
-------
LIST OF TABLES
PAGE
4-67. SBR TOTAL ORGANIC HALIDES MATERIAL BALANCE . 160
4-68. SBR CHLORIDE MATERIAL BALANCE 162
4-69. SBR EFFLUENT METALS CONCENTRATIONS '.'.'.'. 163
4-70. SBR EFFLUENT EP TOXICITY ANALYSIS 164
4-71. SBR PH DATA .' .' 155
4-72. SBR POWER USAGE '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. 167
4-73. TOTAL CDD/CDF DATA -LOW PENTA SW TEST '...'.'.'.'.'.'.'.'.'.'.'. 170
4-74. TOTAL CDD/CDF DATA -HIGH PENTA SW TEST 170
4-75. CDD/CDF DISTRIBUTION - LOW PENTA SW TEST 171
4-76. BATS - CDD/CDF CONCENTRATIONS
IN THE EFFLUENT STREAMS 174
4-77. SLURRY BIO-REACTOR - CDD/CDF CONCENTRATIONS IN THE
EFFLUENT STREAMS 175
5-1. "ORDER-OF-MAGNITUDE" COST ESTIMATE FOR A HYPOTHETICAL
DEMONSTRATION TEST AT A "GENERIC" SITE ISO
6-1. ACCURACY - SURROGATE SPIKES SOIL WASHER LOW
CONCENTRATION SOIL TEST 188
6-2. ACCURACY - SURROGATE SPIKES SOIL WASHER HIGH
CONCENTRATION SOIL TEST 189
6-3. PENTACHLOROPHENOL MS/MSD RESULTS - PRECISION '.'.'.'.'.'.'.'. 191
6-4. MS/MSD DATA FOR SOIL WASHER (SOIL AND AQUEOUS
MATRICES) PAH COMPOUNDS ONLY 192
6-5. QC SUMMARY: PRECISION AND ACCURACY - COPPER CHROMIUM
ARSENIC, NON-CRITICAL METALS, RESIDUE, EP TOXICITY SOIL
PILE SAMPLES (1) 198
6-6. COMPLETENESS SOIL PILES '.','.'. 200
6-7. QC SUMMARY: PRECISION AND ACCURACY - COPPER CHROMIUM
ARSENIC, NON-CRITICAL METALS, RESIDUE, EP TOXICITY SOIL
WASHER SAMPLES 202
6-8. LOW CONCENTRATION SOIL DEMONSTRATION - PRECISION
(FIELD DUPLICATES) 205
6-9. HIGH CONCENTRATION SOIL DEMONSTRATION - PRECISION
(FIELD DUPLICATES) 206
6-10. "COMPLETENESS SOIL WASHER"; LOW CONG. SOIL DEMONSTRATION 208
6-11. "COMPLETENESS SOIL WASHER"; HIGH CONG. SOIL DEMONSTRATION .. .- 209
6-12. ACCURACY - SURROGATE SPIKES, AQUEOUS TREATMENT SYSTEM
(AT INFLUENT AND EFFLUENT SAMPLES) 211
6-13. MS/MSDS FOR ORGANICS ON ATS EFFLUENT SAMPLES 212
6-14.
6-15. COMPLETENESS FOR AQUEOUS TREATMENT SYSTEM (ATS) 216
6-16. ATS FIELD MEASUREMENTS: COMPLETENESS '.'. 217
6-17. ACCURACY - SURROGATE SPIKES SLURRY BIOREACTOR
(SOIL FRACTION) 219
6-18. ACCURACY - SURROGATE SPIKES SLURRY BIOREACTOR
(WATER FRACTION) 220
6-19. MS/MSD DATA FOR SLURRY BIOREACTOR INFLUENT
(PAH COMPOUNDS ONLY) 221
xiv
-------
I
I
I
I
I
I
I
I
I
LIST OF TABLES
PAGE
6-20. QC SUMMARY: PRECISION AND ACCURACY - COPPER CHROMIUM
ARSENIC, NON-CRITICAL METALS, RESIDUE, EP TOXICITY SOIL
SLURRY BIOREACTOR SAMPLES 223
6-21. PRECISION SLURRY BIOREACTOR FIELD DUPLICATES 224
6-22. "COMPLETENESS" SLURRY BIOREACTOR 228
6-23. HOLDING TIME STUDY FOR FEED SOIL 232
6-24. HOLDING TIME STUDY FOR WASHED SAND 233
6-25. HOLDING TIME STUDY FOR COARSE OVERSIZED 234
6-26. HOLDING TIME STUDY FOR FINE PARTICLE CAKE 235
6-27. HOLDING TIME STUDY FOR FINE OVERSIZE 236
6-28. HOLDING TIME STUDY FOR SLURRY PHASE 240
6-29. HOLDING TIME STUDY FOR SLURRY AQUEOUS PHASE 241
1
1
I
I
XV
-------
TABLE OF ABBREVIATIONS AND SYMBOLS
AAS
ASTM
BATS
BOD
BSWS
CCA
CDE
CEC
COD
CV
DFTPP
DO
EP Tox
EPA
gal
GC
gpm
kg
kw
L
MDL
mg
MgS04-H20
ml
mm
MN
MS
MS/MSD
MPCA
MTLS
NBS
NIST
No.
PAH
PCDD
PCDF
penta
PFTBA
PH
PM
atomic absorption spectroscopy
American Society for Testing and Materials
BioTrol Aqueous Treatment System
biochemical oxygen demand
BioTrol Soil Washer System
chromated copper arsenate
Combined Dewatering Effluent
cation exchange capacity
chemical oxygen demand
coefficient of variation
decafluorotriphenylphosphine
dissolved oxygen
Extraction Procedure Toxicity (test)
Environmental Protection Agency
gallon(s)
gas chromatography
gallons per minute
kilogram(s)
kilowatt(s)
liter(s)
method detection limit
miiligram(s)
magnesium sulfate monohydrate
milliliter
millimeter(s)
Minnesota
mass spectrometry
matrix spike and matrix spike duplicate
Minnesota Pollution Control Agency
metals, other than arsenic, chromium and copper
National Bureau of Standards (now NIST)
National Institute of Standards and Technology
number
polynuclear aromatic hydrocarbon
polychlorinated dibenzodioxins (also CDD)
polychlorinated dibenzofurans (also CDF)
pentachlorophenol
perfluorotributylamine
- logarithm [hydrogen ion concentration]
Project Manager
xv
-------
TABLE OF ABBREVIATIONS AND SYMBOLS (continued)
ppb
ppm
PVC
QA
QAPJP
QAPP
QG
RCRA
RE
RPD
RREL
SAIC
SBR
SITE
STD
SW
SW-xxxx
TOG
TRPH
um
°C
parts per billion
parts per million
polyvinyl chloride
quality assurance
quality assurance project plan
quality assurance program plan
quality control
Resource Conservation and Recovery Act
relative error
relative percent difference
Risk Reduction Engineering Laboratory
Science Applications International Corporation
Slurry Bio-Reactor
Superfund Innovative Technology Evaluation (Program)
standard
Soil Washer
EPA SW-846 method number xxxx
total organic carbon
total recoverable petroleum hydrocarbons
microgram
micrometer(s)
degrees Celsius
less than
greater than
greater than or equal to
xvi
-------
CONVERSION FACTORS
Area:
Flow Rate:
English (US)
x
Factor
Length:
Mass:
Volume:
ft = foot, ft2 = square foot, ft3 = cubic foot
in = inch, in2 = square inch
yd = yard
Ib — pound
gal = gallon
gal/min (or gpm) = gallons per minute
Mgal/d (or MGD) — million gallons per day
m = meter, m2 = square meter, m3 — cubic meter
cm - centimeter, cm2 = square centimeter
L = liter
g = gram
kg = kilogram
m3/s = cubic meters per second
L/s = liters/sec
m3/d = cubic meters per day
Metric
1 ft2
1 in2
1 gal/min
1 gal/min
1 Mgal/d
1 Mgal/d
1 Mgal/d
1 ft
1 in
1 yd
1 Ib
1 Ib
1 ft3
1 ft3
1 gal
1 gal
X
X
X
X
X
X .
X
X
X
X
X
X
X
X
X
X
9.29 x
6.45
6.31 x
6.31 x
43.8
3.78 x
4.38 x
0.30
2.54
0.91
4.54 x
0.454
28.3
2.83 x
3.78
3.78 x
io-2
io-5
io-2
IO3
io-2
IO2
io-2
io-3
- m2
= cm2
= m3/s
= L/s
= L/s
- m3/d
= m3/s
= m
= cm
= m
= g
- kg
= L
= m3
- L
- m3
xvii
-------
ACKNOWLEDGEMENTS
This report was prepared under the direction and coordination of Mary K.
Stinson, EPA SITE Program Manager in the Risk Reduction Engineering Laboratory-
Cincinnati, Ohio. Contributors and reviewers for this report were Darryl Owens
of EPA Region V - Remedial Project Manager for the MacGillis & Gibbs Superfund
site; Jerry Vorbach; of the Office of Solid Waste and Emergency Response and
Gordon Evans, Guy F. Simes, John E. Brugger, Patrick Augustin and Mary K Stinson
of the Office of Research and Development; and Dennis Chilcote, Steve Valine and
Tom Chresand from BioTrol Inc.
This report was prepared for EPA's Superfund Innovative Technology
Evaluation (SITE) Program by Science Applications International Corporation
(SAIC), McLean, VA for the U.S. Environmental Protection Agency under Contracts
No. 68-03-3485 and 68-CO-0048, by William Ellis, Joe Evans, Omer Kitaplioglu,
Jorge McPherson, Venkat Rao, Susan Roman, Sanjiv Shah, Herbert S. Skovronek., and
Rita Stasik. Laboratory analyses were conducted by Mid-Pacific Environmental
Laboratory (formerly Acurex Analytical Laboratory), Mountain View, CA, Radian
Corporation, Austin, TX and Morrisville, NC, and SAIC Laboratory in La Jolla, CA.
The investigators also express their appreciation to Mr. A.J. Bumby of
the MacGillis & Gibbs Co. for his patience and cooperation during the course of
the project.
xviii
-------
SECTION 1
INTRODUCTION
1.1 THE SITE PROGRAM
In 1986, the EPA's Office of Solid Waste and Emergency Response (OSWER)
and Office of Research and Development (ORD) established the SITE Program to
promote the development and use of innovative technologies to clean up Superfund
sites across the country. Now in its sixth year, the SITE program is helping to
provide the treatment technologies necessary to implement new Federal and State
cleanup standards aimed at permanent remedies. The SITE program is composed of
three elements: the Demonstration Program, the Emerging Technologies Program,
and an Analytical Methods Development Program.
The major focus has been on the Demonstration Program, which is designed
to provide engineering and cost data on selected technologies. EPA and
developers participating in the program share the cost of the demonstration.
Developers are responsible for demonstrating their innovative systems at chosen
sites, usually Superfund sites. EPA is responsible for sampling, analyzing, and
evaluating all test results. The result is an assessment of the technology's
performance that can be used in conjunction with other data to select the most
appropriate technologies for the cleanup of Superfund and otherwise contaminated
sites.
Developers of innovative technologies apply to the Demonstration Program
by responding to EPA's annual solicitation. EPA also will accept proposals.at
any time when a developer has a treatment project scheduled with Superfund waste.
To qualify for the program, a new technology must be at the pilot- or full-scale
stage of development and offer some advantage over existing technologies. Mobile
technologies are of particular interest to EPA.
Once EPA has accepted a proposal, EPA and the developer work with the EPA
regional offices and State agencies to identify a site containing wastes suitable
for testing the capabilities of the technology. EPA then prepares a detailed
sampling and analysis plan designed to evaluate the technology thoroughly and to
ensure that the resulting data are reliable. The duration of a demonstration
varies from a few days to several months, depending on the type of process and
quantity of waste needed to assess the technology. On completion of a
demonstration, EPA prepares a Technology Evaluation Report and an Applications
Analysis Report.
The Technology Evaluation Report compiles and summarizes the results of
the SITE demonstration including the vendor's design and test data, a detailed
technology description, the site and waste used for the demonstration, and
-------
sampling and analysis during the test. Costs of the technology are estimated
based on data from the demonstration and other pilot- and full-scale
applications. The report discusses the factors, such as site and waste
characteristics, that have a major impact on costs and performance. The
Applications Analysis Report more briefly explains the technology and the results
obtained. This report strives to evaluate the applicability of the vendor's
technology to other sites and other wastes. Ultimately, with the Technology
Evaluation Report and the Applications Analysis Report, the Demonstration Program
leads to an analysis of the technology's overall applicability to Superfund
problems.
1.2 OBJECTIVES OF THE DEMONSTRATION
The objectives of this SITE demonstration of BioTrol's soil washing
technology at the MacGillis & Gibbs Superfund site were to evaluates the
following:
1. Performance of the three technologies that made up the treatment
train in this test, particularly in terms of pentachlorophenol
(penta) and polynuclear aromatic hydrocarbon (PAH) contaminant
reduction efficiencies.
2. Fate of contaminants in the SW through the use of a material
balance.
3. Fate of contaminants in the BATS and SBR through the use of a
material inventory.
4. Effect of process operating conditions on performance.
5. Potential health and safety impacts resulting from system operation.
6. Equipment and material handling problems.
7. Projected system economics and major contributing factors.
8. Applicability of the this arrangement of technologies to other
contaminants.
1.3 SITE DESCRIPTION
The Site demonstration of the BioTrol soil washing system was conducted
at the MacGillis & Gibbs Superfund site in New Brighton, Minnesota. The
MacGillis & Gibbs Company has been operating a wood treatment facility on this
site since approximately 1920. Contaminants present at the site include
polynuclear aromatic hydrocarbons (PAHs) from creosote, a coal-tar distillate,
penta, and chromium, copper, and arsenic from chromated copper arsenate (CCA)
used most recently for wood treatment. Wood preserving operations were on-going
-------
SITE-
Dsflionstnrtion Am
FICDBZ 1-1. SITZ Demonscrmeion U>c*eion »e H»eCillis &
-------
at this site throughout the course of this demonstration. Section 4 contains a
more detailed site description.
Detailed process descriptions of the three technologies are presented in
Section 3 along with flow diagrams. The sampling and analytical program for the
tests is summarized in Section 4.
1.4 DESCRIPTION OF OPERATIONS
Contaminated soils excavated from two locations at the MacGillis & Gibbs
site were processed by the BSWS. A soil with a high penta concentration (680 ppm
penta) was excavated from the disposal area shown in Figure 1-2 and a soil with
a low penta concentration (35 ppm) excavated from another area and mixed with
more highly contaminated soil previously excavated by BioTrol to form a soil with
an average of 130 ppm penta. Both soil piles were screened to remove particles
greater than 3/s" in size and thoroughly mixed to form the high penta
concentration feed soil pile and the low penta concentration Feed Soil pile. For
convenience these will be referred to as the "High Penta" soil (680 ppm penta)
and the "Low Penta" soil (130 ppm penta).
Two tests of the SW were conducted. The first test using the Low Penta
soil lasted for 48 hours (2 days). The second test using the High Penta soil
lasted for 150 hours (6 days and 18 hours) but testing of the system was carried
out over about 112 hours. The process steps included screening, mixing, froth
flotation, attrition/classification, thickening, and dewatering. Soils were fed
via conveyor belt at an average rate of 220 kg/hr (484 Ib/hr) for the Low Penta
SW test and 160 kg/hr (352 Ib/hr) for the High Penta SW test. Feed rates varied
as a result of necessary equipment adjustments.
The BATS demonstration also consisted of two tests. During the first
test, which lasted 89 hours (approximately 4 days) , contaminated SW process water
(about 15 ppm penta) from the Low Penta soil washer test was treated in the BATS.
During the second test, which lasted 148 hours (approximately 6 days),
contaminated SW process water from the High Penta soil washer test (about 45 ppm
penta) was treated in the BATS. The process steps included nutrient addition,
pH adjustment and a single pass through the BATS reactor. The treated water was
recycled to the SW starting on the first day of the High Penta SW test. Flowrate
through the BATS was approximately 10 L/min (3 gpm) for both BATS tests.
While the SW and the BATS were compatible in size, the SBR was
considerably smaller, consequently it was tested only on a portion of the fine
particle slurry during the High Penta SW test. On the third day of the High Penta
SW test, the underflow from the fine particle thickener was diverted from the SW
to a holding tank to be used as feed to the SBR. The test of the SBR lasted
fourteen days. The process steps included nutrient addition and pH adjustment
while the slurry was passing through the cascading reactor system and a final
dewatering step. Flow through the SBR was approximately 24 ml/min (0.006 gpm) and
the retention time in the SBR reactor was 4-6 days.
-------
Wood Piles
r^,....
High Soil
Excavaiion Area
Main Roadway
itiiiiiii
MacGillis and Gibbs Company
Process Buildings
and
Administrative Otfices
Main Roadway
(Washed Sand
- Suwage ->'
_ Dium Storage
Buildings
Sloiage
lanK Aiea
Building
±1
Polu Tank |
r
Concicle Pad
i f
Sloiaju Tank /
Aiej /
1 / r
1 J l
— r
- Building
Pole Tank
1
1 GW
Lab
Low Soil E»cavdlion Am
The dimensions ol the area
displayed are approximately
62011 noilh lo-soiilhaiuJ380U
easi-lo-wesi
I" - 56'
FIGURE 1-2. MacGillis and Gibbs/Biotrol Site Map
-------
1.5 PROJECT ORGANIZATION
Through a Cooperative Agreement between EPA and BioTrol Inc. , BioTrol was
responsible for operating their equipment while EPA was responsible for
conducting the demonstration through its SITE Contractor, SAIC. The
demonstration included the following activities:
o preparation of the demonstration test plan
o preparation of the test site to support testing, equipment setup,
and health and safety orientation of field staff
o excavation and preparation of feed soils
o predemonstration sampling and analysis of feed soils
o preparation for analysis of samples
o sampling during the tests of BioTrol's equipment
o Public information meeting held to review BioTrol's soil washing
technology and tests on soils from the MacGillis & Gibbs site.
o two SW tests, two BATS tests, and one SBR test
o Visitors' Day
o post demonstration sampling and analysis of the test staging area
soils
o site closure and disposal of waste materials
o preparation of a Technology Evaluation Report and an Applications
Analysis Report.
Figure 1-3 illustrates the project organization for this demonstration.
1.5.1 Key Contacts
Information useful to potential technology users can be provided by the
following sources:
BioTrol Inc.
11 Peavey Road
Chaska, MN 55318
Dennis Chilcote, Project Director
Thomas J. Chresand, Development Engineer
(908)448-2515
U.S. EPA-ORD
Releases Control Branch (MS-104)
Edison, NJ 08837
Mary K. Stinson, Demonstration Project Manager
(908)321-6683
-------
Bio Trol, Inc.
Bb Trol Management
I
Project Director
Dennis Chilccte
Demonstration
Project Manager
Steve Valirn
> Project Engineering
»Operations Enpinaerinfl
I
Demonstration Technicians
Randy Poftar
Jamei Hlgtfns
GaiyFoss
K«y:
Lines ol Direct
Communication
1
FIGURE 1-3. Project Organization
-------
SECTION 2
EXECUTIVE SUMMARY
2.1 SITE DEMONSTRATION
This demonstration evaluated the individual components of a soil
treatment system developed by BioTrol, Inc., of Chaska, Minnesota. For this
study, a Soil Washer (SW) , the BioTrol Aqueous Treatment System (BATS), and a
Slurry Bio-Reactor (SBR) were arranged to represent integrated operation to
provide a complete remedial alternative. Soil is fed to the SW and mixed with
water. This slurry is subjected to a series of screening, mixing, froth
flotation, attrition/classification, thickening, and dewatering steps. This
separates the relatively clean coarse particles from the more heavily
contaminated fine particles (clay and silt) and also provides scrubbing and
extraction. Contaminated woody debris is segregated during the soil washing.
Resulting process water is treated in the BATS, a three-celled, packed bed
reactor where contaminants are removed via biological degradation. The heavily
contaminated fine particles are biologically treated as a slurry in the SBR,
which consists of three aerated reactors arranged in a cascading system, this
unit was provided by EIMCO Process Equipment Company. While the SW and BATS are
compatible in capacity, the SBR is considerably smaller.
The demonstration was conducted at the MacGillis & Gibbs Superfund site
in New Brighton, Minnesota. The site has been an active wood preserving facility
since the 1920s. Operational and waste disposal practices (now improved) have
contaminated the soil and the groundwater at the site with the pentachlorophenol,
creosote-based polynuclear aromatic hydrocarbons, and copper, chromium, and
arsenic chemicals used as wood preservatives over the course of the facility's
history.
The demonstration lasted a total of 5 weeks with the tests of individual
technologies lasting approximately 2 weeks each. A total of 29,000 kg (32 short
tons) of contaminated soil was processed in the SW along with 148,000 L (38,900
gal) of process water in the BATS and 315 L (83 gal) of fine particle slurry in
the SBR.
The demonstration of the BSWS was conducted through the U.S.
Environmental Protection Agency's (EPA's) Superfund Innovative Technology
Evaluation (SITE) program to develop reliable performance and cost data. The
data collected from this SITE demonstration test will be used to make sound
decisions as to the applicability of the process to other contaminated sites.
Specifically, the information from this demonstration will be used to:
-------
o characterize the efficiency of the Soil Washer and the two companion
technologies
o evaluate potential uses of soil washing for other remedial actions
o identify pretreatment requirements
o isolate operational problems and potential solutions
o characterize process residues
o identify the need for secondary treatment
o develop operating costs for the pilot-scale unit and determine
scale-up costs for a commercial unit
o identify pertinent Government policy and regulatory requirements
o provide a basis for comparison to competitive technologies
2.2 SUMMARY OF RESULTS
2.2.1 Soil Washer
The Soil Washer (SW) was used to treat soils contaminated with
pentachlorophenol (penta), polynuclear aromatic hydrocarbons (PAHs), and heavy
metals (chromium, copper, and arsenic). The demonstration consisted of tests of
two soil piles: a Low Penta soil (130 ppm penta) and a High Penta soil (680 ppm
penta). Removal efficiencies were calculated as:
100 [1-(Washed Soil output/Feed Soil input)]
where input and output refer to concentration or mass of contaminant. The Washed
Soil is the major solids output stream. The results are summarized in the
following table.
TABLE 2-1. SUMMARY OF CONTAMINANT REMOVAL EFFICIENCIES
(based on concentrations)
test Feed Rate penta PAHs c-PAHs TOG Cu Cr As
kg/hr % % % % % % %
Low
High
Penta
Penta
220
160
89
87
83
88
81
87
84
81
59
72
46
66
61
56
Within the precision of the averaged results for each test, Soil Washer
removal efficiency was not seriously affected by fluctuations in either Feed Soil
contaminant concentration or Feed Soil flow rate. It should, however, be noted
that there were significant discrepancies in the penta mass balance, with much
more penta reported in the output streams than in the input. While not verified,
the difference in accessibility for extraction and analysis before and after the
soil washing may be a major contributor to these discrepancies.
-------
2.2.2 BioTrol Aqueous Treatment System
v The BioTrol Aqueous Treatment System was used to treat Soil Washer
process water. In the first test, using process water with about 15 ppm penta
from the Low Penta SW test, the BATS achieved a removal efficiency of 91% for
'penta (based on weighted masses) at an average flow rate of 10.2 L/min (2.69
gpm) . In the second test, using process water with about 44 ppm penta from the
High Penta SW test, the BATS achieved a removal efficiency of 94% for penta
(weighted masses) at an average flow rate of 10.3 L/min (2.72 gpm). The total
volumes treated in the two tests were 55,400 L (14,600 gal) and 92,200 L (24,400
gal), respectively. The treated effluent was recycled to the SW as process water
during the demonstration.
2.2.3 Slurry Bio-Reactor
The Slurry Bio-Reactor tested had a much smaller capacity than the other
two technologies. At a constant flow rate of 24 ml/min and a volume of 180
liters, the average retention time was approximately 5.2 days. However, complete
acclimation apparently was not achieved until well into the test and steady state
operation was not reached. Nevertheless, penta removal did reach at least 90% and
probably would have stabilized at or above this level had the test been
continued. PAH removal efficiency reached at least 70% for all compounds of
interest after acclimation. Heavy metals were unaffected.
2.3 CONCLUSIONS
2.3.1 Soil Washer
Based on the results of this demonstration, the following conclusions can
be reached regarding the operation of the Soil Washer:
o The largest solid output stream, the Washed Soil, retains only a
relatively small portion (-10%) of the original pentachlorophenol mass.
o Penta and PAH removal efficiencies of 87-89% and 83-88%, respectively,
can be achieved with soils of different initial concentrations.
o Contaminant concentrations in the Washed Soil output stream appear to be
steady within a narrow range for each Feed Soil tested and seem to be
independent of fluctuations in the concentration range of the Feed Soil.
o Pentachlorophenol accumulates largely in the process water (Combined
Dewatering Effluent, CDE) and the clay and silt fines (Fine Particle
Cake, FPC). Both of these output streams can be treated biologically
(e.g., BATS and SBR). Water-insoluble compounds such as PAHs gather
mostly on the Fine Particle Cake.
o The Soil Washer did experience some operational instability, primarily
with the feed delivery system. BioTrol feels that this problem is
correctable.
10
-------
2.3.2 BioTrol Aqueous Treatment System
Based on the results of this demonstration and previous tests conducted
by BioTrol, the following conclusions can be reached regarding the operation of
the BATS:
o Results of the tests at two different penta concentrations verify the
vendor's claim that the system can achieve removals of >90% of
pentachlorophenol extracted into the process water by the soil washing
process.
o The claim that polynuclear aromatic hydrocarbons are removed by the
system cannot be substantiated from this demonstration because the
concentrations of most PAHs, even in the influent, were below detection
limits (2-15 ppb in the Low Penta test and 1-400 ppb in the High Penta
test).
o The system is unaffected by increases (up to 90 ppm) in the
concentrations of copper, chromium, and arsenic that occurred over the
course of the 10 day test due to recycle of the BATS effluent to the SW.
Longer operation may require some form of metals removal or treatment.
o Once acclimated, operation of the system on a day-to-day basis requires
the attention of only one operator on a part-time basis.
2.3.3 Slurry Bio-Reactor
Although the system did not achieve steady state, it appears that once
steady state is achieved, penta removal for the liquid phase of the treated
slurry would stabilize at about 97%. For the solid phase, which contains the bulk
of the penta, the penta removal increased from approximately 65% to 92% but had
not yet stabilized during the test. Overall penta removal reached about 96% near
the end of the test, which would tend to support BioTrol's claim of 90%.
Although the data for PAHs are limited, similar behavior between solid
and liquid phases is observed. Overall removal efficiencies after nine days of
operation ranged between 70% and 99% for different PAHs.
Decreases in the liquid and solid phase influent penta and PAH
concentrations over the course of testing suggested that some biodegradatioii was
occurring in the holding tank, probably due to indigenous bacteria. However, it
is felt that this did not have any adverse effect on test results or conclusions
drawn from them since calculated removal efficiencies were based on influent and
effluent concentrations at the reactor.
Copper, chromium, and arsenic concentrations are unaffected by SBR
operation. At least during the nine day study, there was no evidence of
bioaccumulation of these metals in the biomass.
11
-------
A doubling of chloride ion concentration in the effluent compared to the
influent indicates that dechlorination is occurring. The data are, however, too
limited to draw any more definitive conclusions concerning mineralization.
2.3.4 Costs
A pilot-scale demonstration test, such as the one conducted here, is very
often done before a commercial-scale remediation is attempted. The test system
used by BioTrol as the basis for costs consists of a 0.25-0.50 ton/hr Soil Washer
followed by a 5 gpm BioTrol Aqueous Treatment System and a 0.006 gpm Slurr]r Bio-
Reactor. This does not mean that all three technologies must be used together.
The estimated cost for such a hypothetical demonstration test at a "generic" site
based on some of the costs incurred under the SITE program and on some costs
provided by BioTrol would be on the order of approximately $750,000. While this
is equivalent to $6000/ton for a two-week test, it is clearly impacted by such
factors as lease rate, mobilization and demobilization, more intense labor
requirements, and the increased sampling and analysis that would be done during
a test.
Based on information provided by BioTrol, the cost for a commercial-scale
Soil Washing System operating at 20 ton/hr (18.2 metric ton/hr) would be about
$185/metric ton ($168 per short ton) of soil treated, including both capital cost
and operating costs for the Soil Washer, three 100 gpm, 3-celled BATS units
operating in parallel to treat a portion of the water being recycled, and a 23
gpm SBR consisting of three parallel trains of 3 cells to treat the contaminated
fines slurry. Capital equipment is assumed to have a 10-year life with no salvage
value for these calculations. The cost figure also includes the cost to
incinerate the relatively small masses of contaminated coarse and fine oversized
material at a cost of about $l,200/short ton. Although the mass to be
incinerated is small, the impact of the cost of incineration is large.
Approximately 75% of the total BSWS cost is attributable to incineration of the
coarse and fine oversize material.
12
-------
SECTION 3
PROCESS DESCRIPTION
3.1 PRETREATMENT AND PROCESSING REQUIREMENTS
3.1.1 Soil Excavation and Preparation
Prior to the SW tests, soil was excavated from two areas at the MacGillis
& Gibbs site. In the case of the Low Penta soil, it was necessary to mix the
soil that had been originally selected with another soil previously excavated by
BioTrol to obtain the approximate concentration desired. The soils were screened
with a 3/8 inch vibrating screen to remove large particles and debris. After
screening, the soils were thoroughly mixed and formed into two rectangular piles,
which were covered with tarpaulins to await the start of the tests. The results
of the analyses of the two piles of soil are presented in Section 4.
3.1.2 Process Water
Process water generated from the SW tests was pumped into storage tanks
to await treatment in the BATS. It was then pumped directly to the BATS without
any pretreatment. It was evident that some settling of suspended particles was
occurring in the storage tanks but this was not considered a pretreatment step.
At the end of the demonstration, remaining process water was treated in the BATS
and polished with granular activated carbon before it was discharged to the
Minneapolis POTW.
Additional water was provided to the SW as needed to maintain the
approximate ratio of 5 kg water/kg soil, either from municipal water or from that
produced by the dewatering of fines.
3.1.3 Slurry for SBR
The fine particle slurry for the SBR test consisted of the fines
segregated in the Soil Washer dispersed in process water. It was collected from
the output of the Soil Washer over a 30 hour period during the High Penta SW test
and stored in a tank prior to the start of the SBR test. The contents of this
storage tank were mixed by recycling a stream with a pump to keep the particles
in suspension.
13
-------
3.2 PROCESS DESCRIPTION
3.2.1 Introduction
Prior tests conducted by BioTrol indicated that most of the contaminants
are associated with the fine particle fraction (less than 0.075 mm) of a bulk
soil. Separation of the fines fraction concentrates the contaminants into a
small portion of the original soil, which greatly reduces the amount of material
requiring disposal or subsequent treatment.
The SW separates slightly contaminated coarser particles from heavily
contaminated fine particles and, at this site, contaminated (woody) debris.. The
fine particles are treated in the SBR, which reduces contaminant concentration
via biological destruction. The process water used in the SW is treated in the
BATS prior to discharge or recycle. The actual arrangement and operation of the
individual technologies will depend on site characteristics and the contaminants
present. All three technologies can operate in an integrated, continuous feed
mode.
3.2.2 Soil Washer
The SW is an intensive, countercurrent scrubbing system designed to treat
excavated, contaminated soils. A generalized process flow diagram is shown in
Figure 3-1. The SW is built on a 42 foot long, semi-trailer with drop-down sides
using pilot-scale mineral processing equipment. Included on the trailer are a
multi-component feed system, a mixing trommel, vibrating screens, froth flotation
cells, attrition machines, hydrocyclones, spiral classifiers, and dewatering
equipment. Four 8,000-gallon storage tanks (not trailer-mounted) provide
capacity for process water inputs and outputs. The utility requirements for the
SW unit are only a 440-volt, 3-phase power supply of 350 amps and a water supply
of 10 gallons per minute.
Following excavation, large debris is removed from the soil by a
vibrating screen. The screened-soil is fed via conveyor to a mixing trommel
where it is mixed with water to form a slurry. The feed rate of contaminated soil
is maintained by an automatic feed-back control system which adjusts the feed
rate from the conveyor belt to the mixing trommel. Overall water use is about 5
kg/kg soil (-1200 gal/ton). The water flows are measured by tank height gauges
and rotameters. The slurry flows from the mixing trommel across another, finer
vibrating screen where coarse oversize (CO) material is removed. The oversize
product is stored in drums for disposal. The undersize material passing through
this screen is fed to a flotation unit where hydrophobic constituents are removed
in a froth phase. The underflow from the flotation unit then enters an
intensive, multi-stage, countercurrent scrubbing circuit consisting of attrition
and classification equipment. The intense scrubbing action of the attrition
equipment disintegrates soil agglomerates and separates adhering fines from the
coarser particles. Abrasion between the coarser particles provides additional
cleaning of their surfaces. The classification equipment separates the fines
from the coarse soil particles.
14
-------
Excavate
Soil
Screen
I
Oversize
Debris
Slurry -^
Jfc
I
[
L
Multi-Stage
Washing
Circuit
f1.
Contaminated
Water
1
Aqueous
Treatment
System (ATS)
t
Clear
Water
*-
-^
Washed
Sand
1
Contaminated
Silt/Clay
i
Slurry
Bio-Reactor
(SBR)
t
Dewater
t
Treated
Silt/Clay
FIGURE 3-1. Flow Diagram of Soil Washer (SW)
-------
The fine silt, clay, and woody matter retain considerable amounts of
contaminants, even after undergoing intensive attrition scrubbing. The fine soil
particles, suspended in the process water from the scrubbing circuit, are fed to
a thickening operation along with the froth from the flotation unit. Just before
thickening, an aqueous solution of a polymeric flocculating agent is added to
assist settling and separation of the solids from the process water. The
thickened solids (underflow) are then sent to the SBR for treatment as in this
study (Section 3.3.3) or are dewatered using a horizontal centrifuge to form a
fine particle cake (FPC) which is drummed for disposal. The FPC contains most
of the contaminants from the feed soil and requires further treatment. The water
removed during thickening and dewatering processes also contains significant
contamination and is .recirculated to the SW, eventually going to the BATS for
treatment.
3.2.3 BioTrol Aqueous Treatment System
The BioTrol Aqueous Treatment System (BATS) is a multi-cell, submerged,
packed-bed reactor where penta- and PAH-contaminated process water from the SW
is biologically degraded.
The BATS used in the demonstration is a mobile, trailer-mounted unit with
a. nominal 10 gpm hydraulic capacity. The 20 ft x 8 ft x 8 ft enclosed trailer
requires only a base capable of supporting 5 tons. Influent (process wastewater)
and treated effluent are stored in 8,000 gallon carbon steel storage tanks. The
utility requirement for the BATS is a 480-Volt AC, three-phase power source.
The process flow diagram is shown in Figure 3-2. After acclimation of the
biological growth, including the penta-specific Flavobacterium (1-2 weeks),
incoming wastewater is pumped on a time cycle to the 100 gallon tempering tank
inside the BATS trailer (Figure 3.3).
In the tempering tank, the pH of the contaminated water is adjusted to
approximately 7.3 by the addition of caustic or acid in response to on-line pH
instrumentation and a concentrate of inorganic nutrients (trisodium phosphate and
urea) is metered into the water at a predetermined rate to provide the needed
nutrients. The nutrient concentrate is prepared on a semi-weekly basis and stored
in a 50 gallon nutrient tank located inside the BATS trailer.
From the tempering tank the stream passes through a heat exchanger where
the water temperature is raised to 21°G (70°F), using an auxiliary heater if
necessary.
The stream is then pumped to the base of the first of three cells in the
BATS reactor by passing under an influent baffle (Figure 3-4) . Each of the three
cells is filled with a corrugated polyvinyl chloride (PVC) media which serves as
the support for microbial attachment. With the PVC media in place, each cell can
hold approximately 150 gallons. A positive displacement blower supplies air to
a sparger tube system mounted beneath the packing support grid of each cell. The
wastewater stream flows upward from the base of each cell and contacts the fixed-
film microbes. At the top of a cell, the wastewater spills over a weir into a
16
-------
To Atmosphere
i
Carbon
Canister
OH Gas
Contaminated
Water from
Soil Washer
-_ J
• »
Nutrient
Addition
and
pH Adjustment
f ^
Recycle to
Sail Washer
V >
w
Heat
Exchanger
I
\
Bag Filter
(Optional)
I
f
Treated Water
V
„
*
\
" "•»
/
t
AQLH
Treat
Sys
(AT
i
sous
ment
tern
rsi
Waste
/^
To Car
(End of
V
Solids I
ton Filler
POTW
Test Only]
\
/
FIGURE 3-2. Flow Diagram of BloTrol Aqueous Treatment System (BATS)
-------
•INFLUENT
HEAT EXCHANGER
EFFLUENT PUMP
TABLE
•BLOWEF)
TEMPER TANK
CONTROL
PANELS
FIGURE 3-3. BIOTROL, INC. MOBILE AQUEOUS TREATMENT SYSTEM
-------
Vent
Influent
VO
Air Dllluser Pipe
Packing
Overflow
Weir
fr-nEflfluent
FIGURE 3-4. SCHEMATIC OF BIOREACTOR
-------
narrow slot that directs the flow to the base of the next cell for further
treatment.
The lid over the BATS reactor is fitted with a 4-inch diameter piece of
flexible tubing to capture and funnel the offgases from the BATS through a carbon
adsorption canister that was added for the demonstration study. After passing
through the carbon canister, the air is exhausted to the atmosphere.
When the treated water exits the third cell of the BATS reactor, it
passes through the heat exchanger and any heat added during its pass through the
BATS reactor is returned to the influent. The treated water leaves the BATS
trailer and is pumped to a holding tank for recycle in the SW. As an added
precaution, a bag filter was added to remove sloughed biomass before the effluent
was discharged to the POTW at the end of test. If necessary, carbon adsorption
can be used for polishing prior to discharge.
3.2.4 Slurry Bio-Reactor
The Slurry Bio-Reactor (SBR) is a microbiological system for degrading
penta and PAHs adsorbed in or absorbed on the surface of woody material, silt,
and clay particles (Figure 3-5). The bacterial population consists of indigenous
bacteria supplemented by a penta-specific Flavobacterium. The units used in the
demonstration study were designed by and purchased from EIMCO, Inc.
The EIMCO SBR consists of three upright, continuously-stirred, stainless
steel reactors arranged in series, each with a capacity of 60 liters (Figure 3-
6) . The three reactors are arranged in a cascading system, permitting continuous
feed and overflow by gravity at a rate of about 24 ml/min (0.006 gpm) . The
slurry of clay and silt fines enters the first reactor where easily degraded
contaminants are consumed by the pre-inoculated and acclimated microbial
population. As the slurry flows to each successive tank, the more refractory
contaminants are eventually broken down. Each reactor contains a modified slurry
agitator incorporating an airlift pump which returns settled solids that have
been raked to the center back to the top and redistributes them. Flexible
membrane diffusers mounted on stainless steel rotating rake arms provide fine
bubble aeration and turbulence for mixing. The membrane diffusers are a non-clog
type constructed of an elastomeric material which is resistant to the
contaminants. In this manner, mixing is maintained while shear forces are kept
low to allow microorganisms to adhere and grow on the surface of the particles
in the slurry. The reactors are sealed and all gases are vented through an
activated carbon .canister as a precaution to prevent emission of organic
compounds into the environment.
Five variables must be controlled for proper operation of the Slurry
Bio-Reactor system. They are: the influent flow rate, temperature, dissolved
oxygen concentration, gas flow rate, and rake arm speed. The goal is to operate
the reactor system at steady state, which minimizes operator attention and
maximizes the biological degradation rate. The influent flow rate is controlled
by a variable speed peristaltic pump. The system is equipped with heaters and
a thermostat. The dissolved oxygen concentration, a function of the gas flow
rate, the oxygen concentration in the gas, and the rate of uptake by the
20
-------
^
Fine Particle
Slurry
from Soil
Washer
^ v
Feed Storage
Tank
Nutrients
Slurry
Bio-Reactor
i
Air
To Atmosphere
i
Carbon
Canister
Off Gas
Treated
Slurry
FIGURE 3-5. Flow Diagram of the Slurry Bio-Reactor (SBR) with Sample Points
-------
Fine
Particle
Slurry
from
Soil
Washer
N>
N>
Nutrient
Addition
Exhaust
Gas
Slirrer
Activated
Carbon
F
I
Sample Points
Fine Particle Slurry
I
— — Air
Product
Drum
Compressor
FIGURE 3-6. Slurry Bio-Reactor Process flow Diagram
-------
microorganisms, is controlled by the air flow rate and is measured using a
dissolved oxygen probe. All gas flow rates are monitored by rotameters. The
rake arm speed is controlled by a variable speed drive.
The only utility requirement for the SBR is a 110 volt, single phase
power source.
The SBR system that was tested is much smaller in hydraulic capacity than
the SW and BATS systems. Consequently, rather than feed a portion of the fine
particle slurry to the SBR from the SW on a continuous basis, the fine particle
slurry was diverted from the SW into a 500 gallon storage tank for one day during
the middle of the High Penta soil washing test. The storage tank was equipped
with a circulating stream to agitate the slurry and maintain particles in
suspension.
3.3 ANCILLARY POLLUTION CONTROL EQUIPMENT
To prevent accidental spills and splashing from contaminating the
demonstration area grounds, impermeable liners were installed underneath the
feed hopper, the feed conveyor, the SW trailer, and the SBR. The liners were
bordered by absorbent socks to control potential runoff. To minimize runoff due
to weather conditions, process areas were covered by temporary roof structures
or tarpaulins, as necessary. These areas included the feed hopper, the feed
conveyor-belt, the SBR, and the SBR storage tanks.
The air emissions from the BATS and the SBR reactors were passed through
carbon adsorption canisters to assure that no hazardous constituents were being
released. For this demonstration, carbon was used even though air monitoring
conducted during the previous BATS demonstration indicated that this was not
necessary. Sloughed biomass from the BATS was captured by a bag filter and
collected in 55-gallon drums for disposal only at the end of the test when
effluent was discharging to the local POTW. At other times the effluent was
recycled to the SW without filtration.
3.4 WASTE REMOVAL AND SITE CLOSURE
Process wastes generated during the demonstration and drummed for
subsequent disposal included Coarse Oversize, Fine Oversize, Fine Particle Cake,
bag filters, and carbon canisters. Non-process wastes generated during pre-
demonstration and demonstration phases of the study included health and safety
disposables (e.g., Tyvek® suits, gloves, etc.) and water and solvents from the
decontamination of sampling equipment. At this time, all of these materials are
containerized and awaiting disposal at an off-site facility meeting state and
Federal regulations.
23
-------
SECTION 4
THE BIOTROL FIELD DEMONSTRATION
4.1 BACKGROUND
4.1.1 Technology Description
The pilot-scale BSWS treatment train, shown schematically in Figure 4-1
as configured for this demonstration, consists of three technologies which are
designed to work in an integrated fashion to treat contaminated soils. The
technologies are:
o Soil Washer (SW)
o BioTrol Aqueous Treatment System (BATS)
o Slurry Bio-Reactor (SBR).
The SW is an intensive countercurrent scrubbing system which separates slightly
contaminated coarse soil particles from heavily contaminated fine particles. The
process water used in the SW is treated in the BATS prior to discharge or
recycle. The BATS, a multi-cell, submerged, packed-bed biological reactor,
removes contaminants from the process water. The contaminated fine clay and silt
particles are treated in the SBR. The SBR consists of three cascading aerated
reactors which provide an aerobic environment for microbial growth. Section 3
provided descriptions of the three treatment technologies. As noted earlier, the
SBR is significantly smaller in hydraulic capacity than the SW or BATS.
4.1.2 Site Description
The SITE demonstration of this configuration of the BioTrol Soil Washing
System was conducted at the MacGillis & Gibbs Superfund site located 7 miles
north of Minneapolis, at 440 5th Avenue, NW, New Brighton, Minnesota. The 24
acre site is bounded by 5th Avenue NW to the east, 1st Street to the south, the
Minnesota Transfer Railroad and Bell Lumber and Pole Company parcel No. 2
properties to the west, and private residential land to the north (Figure 1-1).,
The surrounding area consists of wetlands to the east, west, and north and lakes
both to the north and south of the site. Results of a previous Remedial
Investigation Report indicate that the soil is part of the New Brighton Formation
and is composed of a silty sand which is interlaid with organic matter in some
areas of the site and underlain by localized silt lenses and clay. Results of
a previous hydrogeological investigation indicate that an unconfined surficial
aquifer which is recharged through the infiltration of precipitation is
continuous across the site. Groundwater depths range from 3.2 to 13 5 feet (Twin
Cities Testing, 1986).
24
-------
I Contaminated
I Soil
>^ _s
to
Water
(folding
Tank
Aqueous
Treatment
System
( Municipal^
Water
V Supply )
I Coarse 1
1 Oversize 1
i
»
Mixing „ c,««.
«jij t. i • Screen — fc»^ nom
romrm.1 »• Fk)lat|on
f
: i
I Fine I
1 Oversize 1
1
MulU-l
^ Coiintni
"^ Aiirilion/Ck
Pin
4
* i
\
Stage
currant
issilicau
:uil
on
Combined
Dewaterlno
Effluent
Holding Tank
Thickening
Fine Silts. ^
Clays, and 1
Oroaiika 1
FIGURE 4-1. Flow Diagram of the Soil Washer System
Washed
Soil
Product
Cationic
Polymer
LOHiuyu \
Underflow or 1
Firm Particle
Slurry )
\
r^-
•
-, -
pewatering
J
M
Slurry
Bio-Reactor
Fine Particle |
^ Cake J
-------
The MacGillis & Gibbs Company has been operating a wood treatment
facility on this site since approximately 1920. Contaminants present at the site
include creosote, which is a coal-tar distillate, pentachlorophenol (penta), and
chromium, copper, and arsenic. Creosote was used as a wood preservative from the
time operations began until approximately 1950. During the late 1940s, the
MacGillis & Gibbs Company began using a 5 percent mixture of penta in fuel oil.
Penta was phased out in the mid-1970s and replaced by another wood preservative,
chromated copper arsenate (CCA).
The MacGillis & Gibbs site was placed on the National Priorities List
(NPL) in conjunction with the neighboring Bell Lumber and Pole site in September
1983 because of surface and groundwater contamination. As part of the NPL listing
process, each potential Superfund site is evaluated using the Hazard Ranking
System (MRS). Each site is given an HRS score, which represents a composite of
values assigned from a standard set of factors related to risks from potential
or actual migration of hazardous substances through groundwater, surface water,
and air. The site,-ranked 201 out of 799 on the NPL in June 1987, is one of 54
wood preserving sites currently listed.
4.1.3 Demonstration Test Plan
The Demonstration Test Plan delineates personnel schedules, procedures,
and steps involved in the collection of data and evaluation of the BioTrol
technologies. The data collected from the SITE demonstration test will be used
by potential users of the technology to make sound decisions as to the
applicability of the process. Since the data collected from the demonstration
test may be used to "prove" a technology, the procedures used to collect and
analyze samples must be thoroughly planned and must comply with guidelines set
forth by RREL's Category II Quality Assurance Project Plans (QAPjP). Data
quality objectives (DQO) were explicitly defined in the QAPjP, which is a
supplement to the Demonstration Test Plan. The Demonstration Test Plan along
with the QAPjP ensure that all the necessary information is collected to evaluate
the technology and that data generated and collected throughout the demonstration
are of a quality sufficient to support sound scientific conclusions. This report
includes a Quality Assurance Section that presents an assessment of data quality
and describes corrective actions taken during the course of the demonstration.
The QA Section summarizes QAPjP specifications for collected data and presents
an overall evaluation of the data quality in terms of the QAPjP requirements.
4.2 DEMONSTRATION PROCEDURES
4.2.1 Site Preparation
BioTrol had been conducting tests of their technology at the site over
the past year; therefore, the three technologies along with the 4 water storage
tanks used to control water flow through the system were already on-site. The
concrete pads needed to support the SBR and the drums for wastes also were
already in place at the MacGillis & Gibbs site and did not need to be installed.
Prior to the start of the site demonstration, all the necessary equipment
needed to evaluate the technologies was mobilized and installed on-site. This
26
-------
included testing equipment, personnel support facilities, and residuals handling
equipment. Soils for the demonstration tests were excavated from two areas of
the site and placed in two separate, covered piles.
4.2.2 Field Operations
4.2.2.1 Soil Washer Demonstration
The Soil Washer Demonstration consisted of two tests: one with soil
containing the low penta concentration (130 ppm) and one with the high penta
concentration (680 ppm) . As noted earlier, these tests and any derivative studies
or analyses will be referred to as Low Penta and High Penta. The Low Penta test
began on September 25, 1989 at 12:37 after Feed Soil (FS) had circulated through
the Soil Washer for approximately one hour. Sampling was initiated two hours
later at 14:37. Eight composite samples were collected over a period of 50
hours. The Low Penta SW test concluded at 15:07 on September 27, 1989. The
final composite sample was collected at 15:07 that day. A total of about 11,000
kg was treated, or an average of 220 kg/hr.
The High Penta test began on September 29, 1989 at 23:48, again after
Feed Soil had circulated through the Soil Washer for approximately one hour.
Sampling was initiated two hours later at 01:48 on September 30, 1989. Eighteen
composite samples were collected over a period of 150 hours. On October 1, 1989
at 23:40 sample and data collection was temporarily suspended (as scheduled)
while Fine Particle Slurry (the precursor to the Fine Particle Cake) was diverted
from the Soil Washer's centrifuge to a holding tank for use as feed in the Slurry
Bio-Reactor evaluation. Sampling of the High Penta test resumed thirty hours
later at 08:00 on October 3, 1989 and concluded at 03:44 on October 6, 1989. The
final composite sample was collected at 03:44 that day. A total of 18,000 kg of
"as is" soil was treated.
4.2.2.2 BioTrol Aqueous Treatment System Test
The .BATS study also consisted of two tests and was integrated with the
SW operations. The first test began on September 26, 1989 at 13:01 when the feed
to the BATS was switched from groundwater to SW process water generated during
the Low Penta Soil Washer test. Sampling was initiated at 18:55 on the same day.
Fourteen 6-hour composite samples were collected over a period of approximately
3 1/2 days and the test was completed on September 30, 1989 at 10:48.
The second test began when the BATS feed was switched to SW process water
generated during the High Penta Soil Washer test at 10:48 on September 30, 1989.
Sampling was initiated at .12:57 on the same day. Twenty-three 6-hour composite
samples were collected over a period of approximately 6 1/2 days. The test was
completed on October 6, 1989 at 16:15.
27
-------
4.2.2.3 Slurry Bio-Reactor Test
The SBR system that was tested is smaller in hydraulic capacity than the
SW and BATS and could not be integrated. One test was conducted of the SBR
system. Rather than feed a portion of the fine particle slurry to the SBR from
the SW on a continuous basis, which would have been very impractical, the fine
particle slurry was diverted from the SW into a 500 gallon storage tank for one
day during the middle of the High Penta soil washing test. The storage tank was
equipped with a pump that circulated a stream to agitate the slurry and keep
particles in suspension. The test began on October 16, 1989 at 14:08 when the
first grab sample of the first composite was collected. Fourteen composite
samples were collected over a period of two weeks. The test ended on October 31
1989 at 10:45.
4.2.3 Field Measurements
4.2.3.1 Soil Washer
Field measurements were collected at various intervals over the course
of the Soil Washer tests. The masses of Coarse Oversize (CO), Fine Oversize
(FO), Fine Particle Cake (FPC) , and Washed Soil (WS) .output streams were
determined by weighing drums of each stream as filled, rather than at specified
time intervals. Individual output stream flow rates dictated the frequency and
number of measurements of each. Every two hours, flow rates of Combined
Dewatering Effluent (CDE), Municipal Water (MW), BATS Treated Water (High Penta
Soil Washer test only) and Cationic Polymer Solution (thickener) were measured
by either recording the depth of liquid in the respective tank or by recording
a rotameter reading. Feed Soil flow rate was measured every two hours by
recording the feed system's total mass. Every two hours, the air temperature was
measured using a standard mercury thermometer. Every eight hours, pH
measurements of Combined Dewatering Effluent were made from grab samples and
power readings were recorded from a standard domestic electric power meter. Table
4-1 summarizes the field measurements that were obtained.
4.2.3.2 BioTrol Aqueous Treatment System
Field measurements were made at various intervals and frequencies over
the course of the BATS tests, as noted in Table 4-2. Influent, effluent, and
nutrient flow rates were determined every two hours by recording the depth of
liquid in the respective tank and calculating the changes in volume. Every eight
hours, pH and temperature were measured on grab samples taken at "T" joints in
the influent and effluent lines of the system. Power readings were recorded
every eight hours from a standard domestic electric power meter. The weight of
the carbon adsorption residue was measured using a direct-read floor scale.
4.2.3.3 Slurry Bio-Reactor
Table 4-3 lists the field measurements made to characterize the process.
A splitter box diverted flow from the recycle stream of the storage tank to the
SBR feed for six seconds out of every minute. The influent flow rate measurement
and all analytical samples were taken from the recycle stream to take advantage
28
-------
TABLE 4-1. SOIL WASHER FIELD MEASUREMENTS
to
Measurement
==============================
Flow Measurements:
Coarse Oversize (CD)
Fine Particle Cake (FPC)
Fine Oversize (FO)
Washed Soil (WS)
Feed Soil (FS)
Combined Dewatering Effluent (CDE)
Municipal Water
BATS Treated Water
(High Soil Test Only)
Cationic Polymer
Other Measurements:
Air Temperature
CDEpH
Power
A:\TABLE4-1WK1
Frequency
i=
Each Filled Drum
Each Filled Drum
Each Filled Drum
Each Filled Drum
2 Hours
2 Hours
2 Hours
2 Hours
2 Hours
2 Hours '
8 Hours
8 Hours
Device
==============
Platform Balance
Platform Balance
Platform Balance
Platform Balance
Totalizer
Tank Depth Gage
Rotameter
Tank Depth Gage
Rotameter
Mercury Thermometer
pH Meter
Power Meter
Units
Pounds
Pounds
Pounds
Pounds
Pounds
Inches
Gallons/minute
Inches
Gallons/hour
Celsius
pH Units
kW-hrs
-------
TABLE 4-2. BIOTROL AQUEOUS TREATMENT SYSTEM FIELD MEASUREMENTS
Measurement
Frequency
Measurement Device
Influent
Flow Rate
Effluent
Flow Rate
Nutrient
Flow Rate
pH Chemicals
Flow Rate
Influent pH
Effluent pH
Influent
Temperature
Effluent
Temperature
Power
Carbon Residue
Weight
2 hours
2 hours
2 hours
Beginning &
end of the test
8 hours
8 hours
8 hours
8 hours
8 hours
At the end
of the test
Level indicator
Level indicator
Tape Measure
Direct-read floor scale
Hand-held pH meter
Hand-held pH meter
Immersible thermometer
Immersible thermometer
Standard domestic
electric power meter
Direct-read floor scale
file: TAB4-2.WK1
30
-------
TABLE 4-3. SLURRY BIO-REACTOR FIELD MEASUREMENTS
Measurement
Frequency
Measurement Device
Influent
Row Rate
Influent
Density
Effluent
Density
Effluent
Total Mass
Influent pH
Effluent pH
Power
Carbon Residue
Weight
12 hours
12 hours
12 hours
60 min.
12 hours
12 hours
12 hours
At the end
of the test
1000 ml graduated cylinder
and stopwatch
1000 ml graduated cylinder
and analytical balance
1000 ml graduated cylinder
and analytical balance
Direct-read floor scale
Hand-held pH meter
Hand-held pH meter
Standard domestic
electric power meter
Direct-read floor scale
file: TAB4-3.WK1
date: March 2,1991
31
-------
of its higher flow rate. The influent density was measured at this point twice
a day. The effluent density was also measured twice a day, corresponding in time
to the influent density measurement. The effluent mass was measured by noting the
weight of a drum on a scale. Six grab samples taken at four hour intervals were
composited once every twenty-four hours for chemical analyses.
4.2.4 Sampling and Analysis
Sampling and analysis of all process streams was conducted according to
the sampling and analysis plan outlined in the Demonstration Test Plan.
Pentachlorophenol (penta) and Polynuclear Aromatic Hydrocarbons (PAHs) were
defined as "critical" parameters.
The following were considered as "secondary critical" parameters:
o Copper, Chromium, and Arsenic (CCA)
o Polychlorinated Dibenzodioxins/Dibenzofurans (PCDDs/PCDFs)
o Total Residue (TR).
These contaminants were deemed critical in an evaluation of the technology
because the BSWS was designed to tr'eat penta and PAHs; the metals were expected
to be present in the feed soils based on facility history; the presence or
absence of PCDDs/PCDFs would be a determining factor in selecting disposal
options; and TR was necessary to perform mass balance/material inventory
calculations. In addition to the critical contaminants, selected samples also
were analyzed for the following:
o Total Recoverable Petroleum Hydrocarbons (TRPH)
o Chemical Oxygen Demand (COD)
o Other Metals (Ba,Cd,Pb,Hg,Se.Ag) (MTLS)
o EP Toxicity (EP Tox)
o Toxicity Characteristic Leaching Procedure (TCLP)
o Total Organic Carbon (TOG)
o Chloride Ion Concentration (Cl)
o Total Organic Halides (TOX)
o Cation Exchange Capacity (CEC)
o Particle Size Analysis and Chemical Analysis of Particle Size
Fractions (PS).
These measurements were made for process characterization, soil characterization,
and waste disposal information. Table 4-4 lists the analytical methods selected
for each of these analyses. The results of all testing are presented later in
Section 4.4.
4.2.4.1 Predemonstration Sampling and Analysis
Soil samples from the low penta concentration soil pile and the high
penta concentration soil pile were collected on September 25, 1989 (after removal
of large sized debris) prior to the start of the demonstration test using a
fixed-grid sampling pattern. A full-depth core sample was obtained from each
of 8 locations on the soil pile with Low Penta and 12 locations on the High Penta
32
-------
Analyte Sample Preparation
Aqueous/Soil
Penta and PAHs
PCDDs and PCDFs
Metals (Ba,Cd,Cr,Cu,Pb)
Arsenic
Mercury
Selenium
Silver
Total Residue
Total Recoverable
Petroleum Hydrocarbons
Chemical Oxygen Demand
EP Toxic ity
Toxicity Characteristic
Leaching Procedure
Total Organic Carbon
Chloride Ion
Total Organic Halide
Cation Exchange Capacity
3520/3550
8280/8280
3010/3050
7060/3050
7470/7471
7740/3050
7760/3050
418.1/418.1
1310/1310
---/1311
---/3510
300/---
9020/---
-"/---
Particle Size Analysis /D421
and Chemical Analysis
of Particle Size Fractions
Sample Analysis
Aqueous/Soil
8270/8270
8280/8280
6010/6010
7060/6010
7470/7471
7740/6010
7760/6010
160.3/3550
418.1/418.1
410. I/-- -
---/8270
---/8280
---/9060
300/---
9020/---
---/9081
---/D422
Source
Document
SW-846
SW-846
SW-846
SW-846
SW-846
SW-846
SW-846
EPA- 600
EPA- 600
EPA- 600
SW-846
40 CFR 268,
App. I
SW-846
EPA- 600
SW-846
SW-846
SW-846
Source Documents:
ASTM Standard Methods D421-58(Reapproved 1978) and D422-63(Reapproved 1972),
American Society for Testing and Materials Annual Book of ASTM
Standards, Vol. 4.08.
40 CFR 268 Federal Register, Vol. 51, No. 114, June 13, 1986, pp. 21685-21693.
EPA-600/4-79/020 Methods for Chemical Analysis of Water and Wastes (EPA, 3/1979).
SW-846 Test Methods for Evaluation Solid Waste (EPA, 11/1986).
33
-------
soil pile. Each of the core samples was analyzed as a grab sample for penta,
PAHs, and TR. A composite sample for each pile (made up of the individual core
samples) was analyzed for penta, PAHs, CCA, MTLS, TR, EP Tox, TOC, TRPH,
PCDDs/PCDFs, CEC, and PS. The results of these tests, presented later in Section
4.3.1 and in the Appendices, are considered representative of the piles used in
the study.
In addition, samples of the Low and the High Penta soil samples were
subjected to particle-size analysis and classified according to the Unified Soil
Classification System (USCS). The USCS is the most universally accepted soil
classification system and has been adopted by the American Society for Testing
and Materials (ASTM). It is based on textural characteristics for soils with
small amounts of fine particles that do not affect the behavior of the soil, such
as the soil used in the BioTrol SITE Demonstration.
In the USCS, four characteristic soil fractions are used to designate the
size ranges of soil particles. These four fractions are as follows:
Fraction Size Range
Cobbles >76.2mm (3")
Gravel 76.2mm to #4 sieve (4.75mm)
coarse gravel 76.2mm to 19.05mm
fine gravel 19.05mm to #4 sieve (4.75mm)
Sand #4 to # 200 sieve (4.75mm to 0.075mm)
coarse #4 to #10 sieve (4.75mm to 2.0mm)
medium #10 to #40 sieve (2.0mm to .425mm)
fine #40 to #200 sieve (.425mm to 0.075mm)
Fines (silts and clays) Below #200 sieve (<0.075mm)
After sieving, the particle-size fractions were combined into four solid
fractions and one aqueous fraction and analyzed for pentachlorophenol (penta),
polynuclear aromatic hydrocarbons (PAHs) , and copper, chromium and arsenic (CCA) .
The fractions were grouped as follows for analysis:
o Material retained on the #3/8, #4, and #10 sieves
o Material retained on the #20, and #40 sieves
o Material retained on the #60, #140, and #200 sieves
o Solid material passing the #200 sieve
o Aqueous solution passing the #200 sieve.
While the data presented later in Section 4.4.2 for both the Feed Soil
and the output streams give considerable insight into the weight and contaminant
distribution, the reader is cautioned that these data are not directly comparable
to that observed in the pilot-scale Soil Washer.
4.2.4.2 Demonstration Sampling and Analysis
Since the SW is considered a continuous operation, sampling strategy
involved a fixed interval sampling. Likewise, fixed interval sampling was also
adopted for the BATS and SBR. The sampling schemes for the three technologies
are summarized in the following paragraphs.
34
-------
Soil Washer - - The sampling scheme for the SW is presented in Table 4-5.
All grab samples were collected using heavy duty glass beakers. Grab samples
were combined in glass composite containers kept on ice.
BioTrol Aqueous Treatment System -- The sampling scheme for the BATS is
presented in Table 4-6. The BATS influent and effluent samples were collected
using ISCO samplers that automatically collected 250 ml grab samples every 10
minutes and deposited them in an ice-chilled composite container. The BATS
carbon canister was sampled at the end of the test by manually compositing four
750 ml grab samples of the carbon.
Slurry Bio-Reactor -- The sampling scheme for the SBR is presented in
Table 4-7. The SBR influent and effluent samples were collected using heavy duty
glass beakers. Composite samples of the SBR influent and effluent consisting of
six grab samples taken with heavy duty glass beakers at four hour intervals were
stored in glass composite containers kept on ice. At the end of the test the
contents of the SBR carbon canister were split into two halves and composite
samples of each half were obtained.
4.2.4.3 Post-Demonstration Sampling and Analysis
Drum Samp1ing- -The nine types of wastes generated during the
demonstration were as follows:
o Fine Particle Cake/Oversize
o Coarse Oversize
o Decontamination water containing hexane
o Decontamination water containing methanol
o Combined Dewatering Effluent
o General debris
o F003-Contaminated Debris
o Acid preserved aqueous samples
Following the demonstration, samples were taken of these wastes stored in 55-
gallon drums. These samples were sent to a designated waste disposal facility
for treatability testing to provide proper and safe disposal of the wastes.
Composite samples or grab samples were collected of each of the waste streams.
Composite samples consisted of 2-3 grab samples. Liquids were sampled with
lengths of glass tubing and solids were sampled with a stainless steel ladle.
Dioxin Sampling--Samples of the Washed Soil, Fine Particle Cake, Fine
Oversize, Coarse Oversize, and Combined Dewatering Effluent were collected for
high resolution dioxin/furan analysis by SW-846 Method 8290. Solid samples were
composites and aqueous samples were grabs.
Site Area Sampling--Soil sampling of the test staging area was performed
after the demonstration was completed to insure that no contamination was caused
by demonstration activities. Grab samples collected at 10 different locations
on-site at depths of 6" and 18" with a stainless steel hand-held bucket auger
confirmed that the demonstration had not adversely affected the ground.
Measurable concentrations of penta were only found in some of the 18" deep
samples, probably reflecting the condition of the site.
35
-------
TABLE 4-5. SOIL WASHER SAMPLES
Sample
Feed Soil
Washed Soil
Coarse Oversize
Fine Oversize
Fine Particle
Cake
Combined
Oewatering
Effluent
BATS Recycle
Water
Municipal
Water
Thickener
Solution
Matrix Sample Type
Solid Manual
Composite
Solid Manual
Composite
Solid Manual
Composite
Solid Manual
Composite
Solid Manual
Composite
Aqueous Manual
Composite
Aqueous Manual
Composite
Aqueous Manual
Composite
Aqueous Manual Grab
Frequency
6-Hour Composite
from 2-Houriy Grabs
6-Hour Composite
from 2-Hourly Grabs
6-Hour Composite
from 2-Hourly Grabs
6-Hour Composite
from 2-Hourly Grabs
6-Hour Composite
from 2-Hourly Grabs
6-Hour Composite
from 2-Hourly Grabs
12-Hour Composite
from 4-Hourly Grabs
12-Hour Composite
from 4-Hourly Grabs
End of Test
Analytical Parameters
Penta, PAHs, CCA, TR,
TOC, TRPH
Penta, PAHs, CCA, TR,
TOC, TRPH, EP TOX, TCLP,
MTLS, PCDDs, PCDFs
Penta, PAHs, CCA, TR,
TOC, TRPH, EP TOX, TCLP,
MTLS, PCDDs, PCDFs
Penta, PAHs, CCA, TR,
TOC, TRPH, EPTOX, TCLP,
MTLS, PCDDs, PCDFs
Penta, PAHs, CCA. TR,
TOC, TRPH, EP TOX, TCLP,
MTLS. PCDDs, PCDFs
Penta, PAHs, CCA, TR,
TOC. TRPH, EP TOX, TCLP.
MTLS, PCDDs, PCDFs
Penta, PAHs, CCA, TR,
TOC, TRPH
Penta, PAHs, CCA, TR,
TOC, TRPH
Penta, PAHs
file: 4-5. WK1
date: March 4,1991
36
-------
Table 4-6. BIOTROL AQUEOUS TREATMENT SYSTEM SAMPLES
Sample
Matrix
Sample type Frequency
Analtyical Parameters
BATS Influent
BATS Effluent
Carbon Canister
Aqueous Automatic 6-Hour Composite
Composite
Aqueous Automatic 6-Hour Composite
Composite
Solid Manual End of Test
Compoiste
Penta, PAHs, CCA, TR,
TRPH.COD.TOX.C
Penta, PAHs, CCA, TR,
TRPH, COD, PCDDs, PCDFs,
MTLS. TOX, C
Penta, PAHs, TR, PCDDs,
PCDFs
TABLE 4-7. SLURRY BIO-REACTOR SAMPLES
Sample Matrix
SBR Influent Solid
SBR Effluent Solid
Carbon Canister Solid
Sample type
Manual
Composite
Manual
Composite
Manual
Composite
Frequency
24-Hour Composite
from 4-Hourly Grabs
24-Hour Composite
from 4-Hourly Grabs
End of Test
Analtyical Parameters
Penta, PAHs, CCA, TR
TRPH, COD, TOX, CL
Penta, PAHs, CCA, TR
TRPH, COD, PCDDs, PCDFs,
MTLS, TOX, CL
Penta, PAHs, TR, PCDDs
PCDFs
file: T4-6&4-7
date: April 2,1991
37
-------
4.2.5 Health and Safety Protocols
Based on screening of the site and knowledge of the contaminants present,
a "contaminated zone" including all process equipment was designated. This was
primarily based on concern over dust. All activities within this area were
carried out in modified Level D protection consisting of clothing to protect
against dust, dust masks, face shields, and gloves. A decontamination area where
personnel cleaned equipment, boots, etc., before leaving the area was designated.
4.3 MATERIAL CHARACTERIZATION
4.3.1 Feed Soils
Analyses of the several core samples taken from the Low Penta and the
High Penta soil piles were used to establish that the average penta
concentrations were as needed for the demonstration. In addition, a composite
sample was also made from the individual samples and reanalyzed. The results of
these analyses are summarized in Tables 4-8 and 4-9. The averages of the
individual samples (133 and 512)are consistent with the anticipated values and
with the values obtained later during the test program (130 mg/kg and 680 mg/kg,
respectively); however, the analysis of the composited sample of the Low Penta
soil pile was high at 450 mg/kg. The results are summarized in Tables 4-8 and 4-9
for the Low Penta and High Penta soil piles, respectively; results obtained for
'the Feed Soil during the demonstration are found in the discussion of each
parameter.
The results of particle-size analysis for the Low Penta soil sample (SPL-
C-OO-PS) are shown in Table 4-10. It was composed of 5.74% gravel, 71.2% sand,
and 4.54% silt and clay with 81.48% of the original sample weight accounted for.
Table 4-11 summarizes the results of the chemical analyses of the
particle-sized fractions derived from the Low Penta soil sample. While the size
distributions of the samples analyzed do not quite match the distribution in
Table 4-10, the highest concentrations of contamination do occur in the soil
fractions with the smallest grain sizes (<200 mesh), and the soil fraction with
the largest grain sizes, the coarse oversize (>10 mesh). The bar chart in Figure
4-2 illustrates the distribution of penta mass for the different size ranges of
soil particles and the percent of soil represented by each particle size range.
From this figure it is clear that the "gravel and coarse sand" contribute the
largest mass of penta and only account for 1/4 of the soil weight at the
MacGillis and Gibbs site. This is probably the result of penta- and PAH-
contaminated woody debris and tarry material. These tarry materials probably were
more efficiently broken up during the demonstration soil washing than in the
sieving studies.
38
-------
LO
VO
SAIC I
TABLE 4-8 . LOW PENTA SOIL ANALYSES - PEHTA AND PAHs
Benzo-
Phenan-
Anth-
Penta
Benzo(b)-
FluQr. Q
Fluorene Q threne q r.cene q .nthen. q Pyrene q thr.cene q Chryaene q .nthene
Fluor-
PAH,
SPL-01-00
SPL-02-00
SPL-03-00
SPL-04-00
SPL-05-00
SPL-06-00
SPL-07-00
SPL-08-00
SPL-CO-0
AVERAGE
STD DEV
150000
110000
190000
94000
120000
130000
120000
150000
133000
27893
7100 J
13000
11000
13000
20000
7300 MJ
14000
6800 J
19000
11525
4218
20000
37000
33000
41000
2600 *
21000 M
30000
18000
64000
25325
11618
86000
100000
110000
110000
250000
86000 M
110000
96000
38000 J
118500
50584
12000
21000
19000
21000
13000
14000 M
17000
14000
61000
16375
3389
29000
30000
36000
36000
23000
26000 M
24000
20000
78000
28000
5500
3000 J
5200 J
4600 J
4800 J
3700 J
4200 HJ
3300 J
4100 J
15000 J
4113
704
8000 J
11000
11000
12000
10000
8900 M J
8800
9500 J
22000 J
9900
1266
2600 *
2600 *
2600 *
2600 *
4400 J
2600 *
3000 Jh
4100 J
40000 U
3063
702
167700
219800
227200
240400
326700
170000
210100
172500
216800
216800
49194.20
FILE: SWLAVG
-------
Acenaph-
SAIC #
SPH-01-00
SPH-02-00
SPH-03-00
SPH-04-00
SPH-05-00
SPH-06-00
SPH-07-00
SPH-08-00
SPH-09-00
SPH-10-00
SPH-11-00
SPH-12-00
Composite
Average
Std. Dev.
Penta thene Q
130000
380000
660000
650000
170000
570000
560000
830000
500000
660000
590000
440000
600000
511667
196080
11000
21000
22000
14000
24000
23000
25000
21000
16000
16000
21000
21000
17000
19583
4132
Fluorene Q
11000
24000
26000
11000
27000
28000
28000
20000
15000
17000
22000
20000
18000
20750
5932
TABLE 4- 9 . PENTA AND PAH CONCENTRATIONS
HIGH PENTA SOIL
Benzo-
Phenan- Anth- Fluor- (a)an-
threne Q
47000
84000
86000
38000
92000
97000
110000
74000
53000
80000
80000
74000
78000
76250
20146
racene Q
22000
39000
41000
19000
45000
42000
46000
41000
31000
35000
34000
31000
38000
35500
8231
anthene Q
43000
75000
68000
62000
76000
76000
77000
76000
65000
61000
70000
85000
61000
69500
10436
Benzo(b)- Benzo(k)
Fluor- Q Fluor-
TOTAL
PAHs
Pyrene Q thracene Q Chrysene Q anthene anthene Q
44000
68000
92000
49000
97000
79000
77000
81000
86000
2600 *
100000
59000
78000
69550
26461
8000 J
16000
17000
15000
19000
19000
20000
19000
17000
15000
19000
20000
15000 J
17000
3215
16000
28000
26000
21000
31000
29000
29000
32000
28000
23000
30000
30000
22000
26917
4499
3000 *
11000
13000
16000
12000
11000
11000
13000
11000
8500 J
16000
13000
40000 UJ
11542
3288
2700 *
9500 J
12000
2700 *
14000
11000
11000
10000
12000
8500 J
14000
11000
40000 UJ
9867
3557
207700
375500
403000
247700
437000
415000
434000
387000
334000
266600
406000
364000
356458
73278
file: swhavg
-------
TABLE 4-10.
PARTICLE-SIZE ANALYSIS OF LOW
CONCENTRATION SOIL SAMPLE SPL-C-00-PS
(1075 gm dry weight)
Soil Fraction
Gravel
Coarse Sand
Medium Sand
Fine Sand
Fines (Silt/Cla^
Size Range
76.2mm to #4 sieve
#4 to #10 sieve
#10 to #40 sieve
#40 to #200 sieve
0 below #200 sieve
Dry Weight
of Fraction
(grams)
61.82
158.59
461.76
145 . 24
48.45
% of Soil
Retained
on Sieve
5.74 %
14.75 %
42.94 %
13.51 %
4.54 %
Recovery 875.86 81.5 %
41
-------
TABLE 4-11. CHEMICAL ANALYSIS OF PARTICLE-SIZE FRACTIONS LOW PENTA CONCENTRATION SOIL SAMPLE
Soil Fractions:
Description:
% of Sample*:
Organic Analyses:
Units:
Pentaohlorophenol
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)lluoranthene
Inorganic Analyses:
Units:
Arsenic
Chromium
Copper
Feed Soil
Composite
ug/kg
133000
10000
64000
38000
61000
78000
15000
22000
40000
13.1
15.S
13.0
•- •
Gravel and
Coarse Sand
Mat'l retained on
#3/8,#4.#10 selves
25.16
ug/kg
240000
ND
15000
140000
16000
29000
ND
12000
ND
mg/kg
13.3
14
13.2
Medium Sand
Mat'l retained on
#20,#40 sieves
52.75
ug/kg
42000
3400 J
7100
45000
4800
6400
1200 J
2800 J
1000 J
mg/kg
2.9
5.5
4.2
i • -
Fine Sand
Mat'l retained on
#60,#140,#200 sieve
16.58
ug/kg
45000
1100 J
3100
13000
2200
2800
ND
1300 J
ND
mg/kg
2.4
5.4
3.7
•
Fines
(Silts/Clays)
Solid mat'l
passing #200 sieve
5.53
ug/kg
240000
14000
31000
230000
21000
40000
ND
16000
ND
mg/kg
17.5
46
22.7
========
Aqueous SoTn
passing #200 sieve
NA
ug/kg
NA
NA
NA
NA
NA
NA
NA
NA
NA
mg/L
3.6
4 9
N>
J - Indicates an estimated value. Used when a compound meets identification criteria and the result is lees than the sample quantitation limit but greater than zero
N A - Not analyzed. Original extracts lost; insufficient volume for re-extraction.
ND - Analyzed, not detected.
* - The percent is based on the amount of sample recovered from the sieves.
file:TAB4-11.WK1
date: April 2,1991
-------
£••7
B>
e
TJ
o
TJ
b
x
-------
The original sample of High Penta soil (SPH-C-00-PS) was split into two
samples of approximately equal weights (-500 gm each) and sieved wet using sodium
hexametaphosphate or ASTM Type II water as the sieving solution. The results of
both particle-size sieving analyses for the High Penta soil sample are shown in
Tables 4-12 and 4-13 and indicate there was little difference in particle size
distribution. The sample sieved with sodium hexametaphosphate was composed of
2.53% gravel, 83.83% sand particles, and 6.53% silt and clay with 92.9% recovery
of the sample weight. The sample sieved with ASTM Type II water was composed of
2.53% gravel, 83.03% sand particles, and 7.40% silt and clay with 92.95% recovery
of the sample weight.
Tables 4-14 and 4-15 show the results of the chemical analysis of the
original High Penta soil and the particle-size fractions from the sodium
hexametaphosphate solution and ASTM Type II water, respectively. While the
contaminant distributions from the two solutions differ somewhat, the highest
contaminant concentrations still occur in the fractions with the smallest and
largest grain sizes. Figures 4-3 and 4-4 show the distribution of penta mass for
different particle size fractions and the % of soil represented by this particle
size range. Although each soil fraction contributes approximately the same penta
mass, the gravel and coarse sand and the fines (silts/clays) account for only
approximately 20% of the soil weight.
4.3.2 Soil Washer Output Streams
Samples of the solid process streams generated during the demonstration
of the BSWS in both the Low Penta and the High Penta Soil Washer tests also were
subjected to particle-size distribution and chemical analysis. These process
streams were as follows:
o Washed Soil (WS)
o Fine Particle Cake (FPC)
o Fine Oversize (FO)
o Coarse Oversize (CO).
Tables 4-16 and 4-17 summarize the particle-size and chemical analysis results
for the solid process output streams for the Low Penta and the High Penta soil
washer tests, respectively.
Using the Unified Soil Classification System, the Washed Soil would be
classified as a medium-grained sand with 93% and 96% of the solid mass falling
in the sand range in the Low and High Penta tests, respectively. The Fine
Particle Cake would be classified as fines (silts and clays) with 75% and 62%,
respectively, of the mass falling in the fines range.
The Fine Oversize would be classified as a medium-grained sand with 85%
and 77% of the particles falling in the sand range. However, the majority of the
mass (76% and 73%) was retained on the #20 (0.85mm) sieve, the largest used in
the particle-size analysis of this process stream. In addition, this material was
described as looking like "peat moss" and consisting of very small organic
fibers. Based on the laboratory description and using the USCS, the fine
oversize could be classified as a highly organic soil.
44
-------
TABLE 4-12. PARTICLE-SIZE ANALYSIS OF HIGH PENTA SOIL SAMPLE SPH-C-00-PS
(SODIUM HEXAMETAPHOSPHATE SIEVING SOLUTION)
(442.3 gm dry weight)
Soil Fraction
Gravel
Coarse Sand
Medium Sand
Fine Sand
Fines (Silt/Clay)
Size Range
76.2mm to #4 sieve
#4 to #10 sieve
#10 to #40 sieve
#40 to #200 sieve
below #200 sieve
Dry Weight
of Fraction
(grams)
11.19
27.13
135.87
207.8
28.88
% of Soil
Retained
on Sieve
2.53 %
6.13 %
30.72 %
46.98 %
6.53 %
Recovery 410.87 92.9 %
TABLE 4-13. PARTICLE-SIZE ANALYSIS OF HIGH PENTA SOIL SAMPLE SPH-C-00-PS
(ASTM TYPE II WATER SIEVING SOLUTION)
(464.0 gm dry weight)
Soil Fraction
Gravel
Coarse Sand
Medium Sand
Fine Sand
Fines (Silt/Clay)
Size Range
76 . 2mm to #4 sieve
#4 to #10 sieve
#10 to #40 sieve
#40 to #200 sieve
below #200 sieve
Dry Weight
of Fraction
.(grams)
11.74
28.47
155.34
201.43
34.34
% of Soil
Retained
on Sieve
2.53 %
6.14 %
33.48 %
43.41 %
7.4 %
Recovery 431.32 92.95 %
45
-------
TABLE 4-14. CHEMICAL ANALYSIS OF PARTICLE-SIZE FRACTIONS FOR HIGH PENTA CONCENTRATION SOIL
hullfiauim. 1 ; — PLE S°DIUM HEXAMETAPHOSPHATE SIEVING so. i mnu
\J\JII 1 lOL-HUIIt).
Description:
% of Sample*:
urganic Analyses:
Units:
Pentachlorophenol
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Inorganic Analyses:
Units:
Arsenic
Chromium
Copper
r-eea son
Composite
ug/kg
600000
17000
18000
78000
36000
70000
54000
17000
28000
10000
7300 J
53000 J
•
21.9
32.1
28.5
D - Indicates the compound was analyzed at a second
Gravel and
Coarse Sand
Mat'l retained on
#3/8,#4,#lO sieves
9.33
ug/kg
1200000 D
27000
25000
97000
66000
89000
99000
24000
40000
10000
ND
ND
1
mg/kg
41.1
64.5
55
iry dilution.
Medium Sand
Mat'l retained o
#20,#40 sieves
33.07
ug/kg
380000 D
10000
11000
40000
23000
40000
43000
8300 J
29000
ND
ND
ND
mg/kg
8.3
11.7
9.6
Fine Sand
Mat'l retained on
#60,#140,#200 sieve
- •
50.58
ug/kg
190000
ND
ND
22000
ND
19000
18000
ND
ND
ND
ND
ND
— - — --
mg/kg
5.4
8.5
7
— — "
Fines
(Silts/Clays)
Solid mat'l
passing #200 sieve
7.03
ug/kg
1900000 D
48000
44000
160000
68000
160000
150000
40000
62000
24000
20000
14000
mg/kg
33.5
41.6
33.2
Aqueous sol'n
NA
ug/L
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
mg/L
7.21
6.73
6C7
• Or
ND - Analyzed, not detected.
- The percent is based on the amount of sample recovered from the sieves.
-------
Soil Fractions:
Description:
% of Sample*:
TABLE 4-15. CHEMICAL ANALYSIS OF PARTICLE-SIZE FRACTIONS FOR HIGH PENTA CONCENTRATION
ASTM TYPE i! WATER SIEVING SOLUTION
SOIL SAMPLE
Organic Analyses:
Units:
Pentachlorophenol
Acenaphthene
2,3,5,6-Tetrachlorophenol
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzb(k)fluoranthene
Benzo(a)pyrene
Inorganic Analyses:
Units:
Arsenic
Chromium
Copper
, <-••—.j..vu M» u mjuyiiuni j UUUIIUII.
.-.ndica^
NA - Not analyzed. Original extracts lost; Insufficient volume for re-extraction.
ND - Analyzed, not detected.
- The percent Is base on the amount of sample recovered from the sieves.
Feed Soil
Composite
100
ug/kg
600000
17000
18000
78000
36000
70000
54000
17000
28000
10000
73000 J
53000 J
21.9
32.1
28.5
zed at a secondary dilu
Gravel and
Coarse Sand
Mat'l retained on
#3/8,#4,#10 sieves
9.32
ug/kg
1200000 D
27000
ND
25000
97000
66000
89000
99000
24000
40000
10000
ND
ND
mg/kg
41.1
64.5
55
lion.
I
Medium Sand
Mat'l retained on
#20,#40 sieves
36.02
ug/kg
250000 D
6600
15000
4600
18000
19000
29000
26000
6300
14000
4900
4500
2400 J
mg/kg
5.9
20.8
10.4
• • .
Fine Sand
Mat'l retained on
#60,#140.#200 sieve
46.70
ug/kg
160000 D
4700 DJ
ND
4400 DJ
17000 D
8500 D
21000 D
19000 D
4200 DJ
7400 DJ
ND
ND
ND
mg/kg
7.2
7.6
6.7
=====
Fines
(Silts/Clays)
Solid mat'l
passing #200 sieve
7.96
ug/kg
1100000 D
22000
51000
19000
130000 D
32000
200000 D
160000 D
18000
30000
13000
10000
8400
mg/kg
67
97.4
78.6
.
Aqueous sol'n
passing #200 sieve
NA
ug/L
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
ug/L
376
283
342
-------
oo
Fig 4-3. Particle-Size Fraction Analysis
High Penta Concentration Soil Sample
(442 gm sample)
Pentachlorophenol (mg)
% of Soil by Weight
% of Soil By Weight
Gravel
Coarse Sand
Medium Fine
Sand Sand
Soil Fractions
Fines
(Silts/Clays)
HeXflmAtanhftpnhato Cio./i.^ O~l'~
... •• —r»"'»'»»|*iiMiw wi^vuiu vjui ii
f ename: PCPA.CHT
-------
Fig 4-4. Particle-Size Fraction Analysis
High Penta Concentration Soil Sample
(464 gm sample)
Pentachlorophenol (mg)
% of Soil by Weight
1% of Soil By Weight
Gravel
Coarse Sand
Medium Fine
Sand Sand
Soil Fractions
Fines
(Silts/Clays)
100%
90%
80%
70%
60%
60%
40%
30%
20%
10%
0%
ASTM Type II Water Sieving Sol'n
fiie name: PCPB.CHT
-------
TABLE 4-16. PARTICLE SIZE ANALYSIS AND CHEMICAL ANALYSIS OF SOLID PHASE STREAMS
LOW PENTA SOIL TEST
Sample ID:
Process Stream:
Particle-Size Results
Sieve ft/Fraction Size
#3/8 / 9.50mm
#4 / 4.75mm
#10 / 2.00mm
#20 / .850mm
#40 / .425mm
#60 / .250mm
#140 / .106mm
#200 / .075mm
PAN
SWL-CO-O
Feed Soil
6%
15%
43%
14%
5%
SWL-WS-02-3
Washed Soil
Soil
on Sieve
+
+
+
3%
76%
7%
6%
1%
2%
SWL-FPC-02-1
Fine Particle Cake
Soil
on Sieve
+
+
+
+
8%
5%
3%
3%
75%
SWL-FO-01-
Fine Oversize
Soil
on Sieve
+
+
+
76%
3%
2%
3%
1%
6%
SWL-CO-01-3
Coarse Oversize
Soil
on Sieve
9%
21%
30%
38%
, 1%
1%
1%
1%
3%
total % Recovery 82% 95% 94% 91% 105%
Analytical Results*
Pentachlorophenol
Fluorene
Phenanthrene
Anthracene
rluoranthene
3yrene
3enzo(a)anthracene
Chrysane
3enzc
-------
TABLE 4-17. PARTICLE SIZE ANALYSIS AND CHEMICAL ANALYSIS OF SOLID PROCESS STREAMS
HIGH PENTA SOIL TEST
Sample ID:
Process Stream:
Particle-Size Results
Sieve #/Fraction Size
#3/8 / 9.50mm
#4 / 4.75mm
#10 / 2.00mm
#20 / .850mm
#40 / .425mm
#60 / .250mm
#140 / .106mm
#200 / .075mm
PAN
SWH-CO-
Feed Soil
+
3%
6%
+
31%
+
+
47%
6%
SWH-WS-05-2
Washed Soil
Soil
on sieve
+
+
+
4%
63%
19%
9%
1%
2%
SWH-FPC-06-3
Rne Particle Cake
Soil
on sieve
+
+
+
+
4%
12%
11%
11%
62%
SWH-FO-01-
Fine Oversize
Soil
on sieve
+
+
+
73%
1%
1%
1%
1%
24%
SWH-CO-02-1
Coarse Oversize
Soil
on sieve
5%
16%
28%
46%
2%
2%
3%
1%
3%
Total % Recovery 93% 98% 100% 101% 106%
Analytical Results*
Pentachlorophenol
Fluorene
Phenanthrene
Anthracene
Fluoranthene
3yrene
3enzo(a)anthracene
Chrysene
3enzo(b)fluoranthene
Copper (mg/kg)
Chromium (mg/kg)
Arsenic (mg/kg)
600000
18000
78000
36000
70000
54000
17000
28000
10000
28.5
32.1
21.9
110000
4300
16000
6700
17000
11000
3100 J
5600
1900 J
20.6
33
11.8
1500000 D
56000
240000 D
130000 D
180000 D
200000 D
59000 DJ
65000 DJ
31000
120
113
80.8
430000 D
14000
47000 D
42000 BD
39000
35000 D
9100
17000
6000
17.3
24.8
16.8
1000000 D
22000
81000 D
33000 DJ
74000 D
60000 D
14000
26000
7800 J
36.7
80.8
36.6
+ - Sieve not used in particle-size analysis of this sample.
* - All analytical results are reported in ug/kg unless otherwise noted.
B - Indicates the compound was found in the associated method blank.
D - Indicated the compound was analyzed at a secondary dilution.
E - Indicates the concentration of a compound exceeds the calibration range.
J - Indicates and estimated value. Used when a compound meets identification
criteria and the result is less than the sample quantitation limit but greater than zero.
51
-------
The Coarse Oversize would be classified as a coarse-grained sand based
on the predominant particle-size falling in the sand range (72% and 82%
respectively). The Coarse Oversize sample from the Low Penta Soil Washer test
also contained 30% gravel (9% coarse gravel and 21% fine gravel). The Coarse
Oversize sample from the High Penta Soil Washer test contained 21% gravel (5%
coarse gravel and 16% fine gravel).
The highest concentrations of contamination occur in the Coarse Oversize
(gravel and coarse sand) and Fine Particle Cake (fines - silts and clays). This
agrees with the results of the particle-size analysis of the Feed Soil, where the
contaminants were concentrated in the gravel, coarse sand, and fines (silts and
clays) fractions.
Thus, the chemical analyses indicate that the highest concentrations of
contamination occurs in the soil fractions with the smallest grain sizes (<200
mesh) as well as in the soil fraction with the largest grain sizes (>10 mesh).
4.4 PILOT SYSTEM PERFORMANCE RESULTS
4.4.1. Introduction
The results of the evaluation of the three technologies are complex and
multi-faceted. To assist the reader, the processes are presented in what seemed
to be a logical sequence, first addressing the Soil Washer, then the BATS, and
finally the SBR. Generally, the results for each technology is: (a) material
balance, (b) penta and PAH data, (c) TOG and TRPH results, (d) metals, beginning
with chromium, copper and arsenic and finishing with other metals, and (e) EP
toxicity data. The discussion of the Soil Washer performance also includes, after
the analytical results, a discussion of the flow rate stability and its impact
on the various output fractions and the contaminants in each. Within each
subsection the results for the Low Penta are always discussed before the High
Penta results.
Finally, after the discussions of the results of the three technologies
are essentially complete, the results for dioxin determinations in the three
technologies are reported and discussed in a separate section.
4.4.2 Soil Washer Performance
Considerable variation in feed and output stream rates was observed.
Consequently, analytical data for contaminant concentrations in the input and
output streams have been weighted on the basis of input or output stream flows
or masses. These weighted values were the basis for most of the calculations in
this report. This helps smooth the data, and it also makes material or mass
balances somewhat more indicative of overall system performance. Recognizing that
the Soil Washer is a complex mechanical operation and exhibited some of the
problems inherent when operating at the pilot-plant scale, this approach was
considered the most practical. Nevertheless, the following table (Table 4-18)
provides a comparison of average concentrations and weighted concentrations for
the Low Penta and the High Penta Soil Washer tests.
52
-------
TABLE 4-18. COMPARISON OF AVERAGE AND WEIGHTED AVERAGE CONCENTRATIONS
(mg/kg or mg/L)
Low Penta SW test High Penta SW test
avg. wtd avg avg wtd avg
Feed Soil 112.8+20 130 657.8+228 680
Washed Soil
Fine Oversize
Coarse Oversize
Fine Particle Cake
Combined Dewater. Eff .
12.5±4.5
85.7±34
155.1+51
241 . 1+74
12.6±3.4
14
96
170
270
14
85 . 3±31
854.4+360
1346 ±650
1274 +390
74.8±36
87
900
1400
1300
80
One of the problems recognized even before the project was initiated was
that the approved analytical methods for organic contaminant concentrations in
Feed Soil and the several output solid streams were measured by sonication of the
soil in an organic solvent (e.g., methylene chloride) followed by GC/MS of the
extracts. This sonication process probably achieves equilibration of the
contaminants between the solvent and the accessible soil particle surfaces.
However, BioTrol's Soil Washer uses high energy abrasion of soil particles
against each other in a water matrix, separating soil particles which are stuck
together and abrading organic matter from the particle surfaces. This may open
up the matrix of the soil more than the laboratory sonication process and
consequently allow more complete extraction of the contaminants during subsequent
analysis of the output streams.
A second aspect of the soil washing process is the extraction step that
unavoidably occurs during the' soil washing; the mass of contaminants found in the
output streams is the sum of the contaminants extracted by a water extraction
step (the soil washing process) and a solvent extraction step. This extra water
extraction step that the Feed Soil is not subjected to, and the hydration of the
soil matrix before solvent extraction, may also have contributed to the greater
mass of pentachlorophenol found in the output streams compared to the input soil.
Finally, to provide a common basis for discussion of results, all data
have been reduced to the processing of 1 kg of Feed Soil in both the Low Penta
and the High Penta tests. Contaminant removal efficiency for the Soil Washer was
assessed by comparing the concentration (or mass) of a contaminant in the "as is"
Feed Soil and Washed Soil, using the equation:
% Removal Efficiency = (1 - Cone, in Washed Soil) x 100
Cone, in Feed Soil
as defined by BioTrol in its original claims. Because of fluctuations in
material flows in some portions of the study, which suggested that concentrations
were not a true indicator of the character of the fractions or the distribution
of the contaminants, this was converted to a mass basis, using the equation:
53
-------
% Removal Efficiency = (1 - Mass in Washed Soil) x 100
Mass in Feed Soil
where mass refers to the weighted mass over the course of a test or, as
standardized in this report, per unit (1 kg) of Feed Soil.
While these equations do not take into consideration the difference in
mass between the Feed Soil and the Washed Soil product, they do, nevertheless,
give a good indication of the amount of contaminant that would be present in the
major output stream, the material expected to be returned to the site.
It is also interesting to note that even though large discrepancies in
the mass balance (i.e., closure) were observed, the effects on the % Removal
Efficiency are small. For example, if the % Removal Efficiency for the High Penta
Soil Washing test .is restated, using the ratio of the mass of penta in the Washed
Soil to the mass of penta recovered in all output streams, the change is only
from 83% to 89%. And, if % Removal Efficiency is calculated on the assumption
that the Feed Soil penta concentration is biased 50% low because of poor
accessibility and a Feed Soil value 50% higher is used (1.5 x 680 ppm = 1020
ppm), the calculated value is then 91.4%.
Basis % Removal Efficiency
Concentration, WS/FS 87
Mass, WS/FS 83
Mass, WS/Total Output 89
Cone., Feed Soil @ 1020 ppm 91
4.4.2.1 As-Is Material Balance
During the Low Penta SW test, carried out for ^50 hrs, a total of 11,000
kg of "as-is" Feed Soil was treated at rates ranging from 110 kg/hr to 300 kg/hr.
On that basis, the weighted feed rate was, then, 220 kg/hr. To achieve the
desired consistency for processing, a total of 70,100 kg of water was added,
mostly as municipal water. Thus, for each kg of Feed Soil, a weighted average of
6.27 kg of water was also processed. Since the as-is Feed Soil contained 11%
water, 1 kg contains 0.89 kg of dry soil. Table 4-19 summarizes the mass flows
in the Low Penta Soil Washer test.
Processing yielded - on an as-is basis - 1.18 kg of Washed Soil, 0.22 kg
of Fine Particle Cake, 0.145 kg of Coarse Oversize, and 0.059 kg of Fine Oversize
plus 5.0 kg of Combined Dewatering Effluent. On this basis, the material balance
indicates that 7.27 kg of input (Feed Soil plus water) produced 6.6 kg of output;
the difference is approximately -9% which, considering the scale of the
operations, is not considered excessive. Of key significance are the large mass
of Washed Soil (1.18 kg) that would be returned to the site if it meets
regulatory agency requirements and the small mass of Fine Particle Cake (0.22 kg)
that would require additional treatment or disposal. Figure 4-5 and 4-6 present
the input and output streams in a more graphical manner.
54
-------
TABLE 4-19. FATE OF MATERIALS-LOW PENTA SOIL WASHER TEST
(Average Input Rate: 220 kg/hr)
Input Stream
Feed Soil (FS)
Municipal Water (MW)
Thickener Solution (T)
TOTAL
Output Stream
Washed Soil (WS)
Coarse Oversize (CO)
Fine Particle Cake (FPC)
Fine Oversize (FO)
Combined Dewatering Eff. (CDE)
TOTAL
. .
Kg
as-is
1
5.45
0.82
7.27
Kg
as-is
1.18
0.14
0.22
0.06
5
6.6
•
% of Input
as-is
14
76
10
100
% of Output
as-is
17.4
2.1
3.2
0.9
76.4
100
Solids
Content
89%
260 mg/L
NA
•—••«
Solids
Content
73%
69%
30%
8%
650 mg/L
—
•
Solids
kg (dry wt)
0.89
0.001
NA
0.891
% of Input
99.9
0.1
100
Solids
kg (dry wt)
0.861
0.097
0.066
0.005
0.003
1.032
% of Output
~
83.3
9.4
6.4
0.5
0.3
Ul
Ul
-------
FIG. 4-5. MASS DISTRIBUTION IN LOW SOIL TEST
I
N
• i
P
U
T
Ul
en
Feed Soil
1 Kg As Is
89% Solids
0.89 Kg Solids
o
U
T
P
U
T
Coarse Oversize
0.145 Kg As Is
69% Solids
0.10 Kg Solid
•• i ,»
Fine Particle Cake
0.22 Kg As Is
30% Solids
0.065 Kg Solid
Municipal Water
5.45 Kg
260 mg/L TSS
•00 IKg Solids
Thickener Stream
0.82
NA
•NA -
Fine Oversize
0.059 Kg As Is
8% Solids
0.005 Kg Solid
r
Hashed Soil
1.18 Kg As Is
73% Solids
0.86 Kg Solids
Input Outout %Dilf.
TOTAL MASS
SOLIDS MASS
7.27 Ka
0 891 Kg
6.6 Kq
1.033
-9.2
+15.9
Combined
Dewatering Elduent
5 Kg As Is
650 mg/L
0.003 Kg Solids
-------
Percentage of. Total Output Mass
O
c
CO
fD
Q
3
w
o
o
o
CO
o
o
' O
en
O
CD
O
o
T~
CO
o
LD
O
O
o
m
I
CD
O
oo'
c
6'
D
O
1. o
s
o
c
~o
c
Q
CO
GO
-------
In the High Penta SW test, the average processing rate.over 112 operating
hours was 160 kg/hr, with considerable variation as discussed later in the
subsection concerning flow rate variability (Section 4.4.2.8).
Over the course of the test, each kilogram of as-is Feed Soil (with an
average 84% solids, or 0.84 kg solids) was mixed with an additional 5.09 kg of
water, composed of 4.12 kg of treated effluent from the BATS, 0.22 kg of
municipal water, and 0.75 kg of the cationic polymer solution used to assist
dewatering. The masses of the output streams shown in Table 4-20 and Figures 4-7
and 4-8. Based on the total output of 6.27 kg from an input of 6.1 kg, the
difference, approximately 3%, is not significant.
Comparison of the water input and output streams indicates that 1.2 kg
and 0.5 kg of water is retained by the solid output streams from the Soil Washer
during the Low Penta test and the High Penta test, respectively. This water
would have to be made up from some outside source if further recovery of water
is not possible by improved dewatering of solid streams.
4.4.2.2 Dry Solids Material Balance
Each 1.0 kg of as-is Feed Soil (Low Penta SW test) contains an average
of 89% solids or 0.89 kg based on Total Solids analyses. Similarly, the solid
output streams retain varying amounts of water. Based on Total Solids analyses,
the output streams contain 1.033 kg of solids, including the suspended solids in
the Combined Dewatering Effluent (CDE). Compared to the 0.891 kg of solids in the
input (including the suspended solids in the municipal water) this represents an
apparent 16% increase. While this is no longer insignificant, it may reflect
small changes in large weight values, i.e., the precision of the various
weighings, particularly the Feed Soil and the Washed Soil, or subsequent rounding
of values during calculations. While the results (summarized in Table 4-19 and
Figures 4-5 and 4-6) can be used to infer that 97% of the Feed Soil solids left
the system as dry Washed Soil in the Low Penta SW test, the reader is cautioned
against making such an interpretation because of the uncertainty created by the
16% increase in solids.
In the High Penta SW test, each 1.0 kg of as-is Feed Soil contains 0.84
kg of solids. The calculated total solids in the output streams amount to 1.086
kg, including the small amount of solids in the CDE (Table 4-20 and Figures 4-7
and 4-8). The change from the input is now 29%, which is large and difficult to
explain. As a result of this increase, a value can not be calculated for recover}'
of the Washed Soil since the Washed Soil output (on both the as-is and dry weight
basis) is greater than the input.
4.4.2.3 Penta Distribution - Low Penta SW Test
Samples of each input and output stream .were analyzed for penta. The
analytipal results were converted to mass on the basis of the weights for each
interval sampled and weighted concentrations then calculated from the total penta
mass and the total input or output stream mass. These weighted results are
presented in Table 4-21 and Figure 4-9. Using the change in weighted
concentration from the Feed Soil to the Washed Soil, the average penta Removal
58
-------
TABLE 4-20. FATE OF MATERIALS-HIGH PENTA SOIL WASHER TEST
Average Input Rate: 160 Kg/hr
Input Stream
Feed Soil (FS)
Municipal Water (MW)
Thickener Solution (T)
ATS Effluent (ATS)
TOTAL
Output Stream
Washed Soil (WS)
Coarse Oversize (CO)
Fine Particle Cake (FPC)
Fine Oversize (FO)
Combined Dewatering Effl.
TOTAL
file- 4-17 WK1
—
Kg
as-is
1
0.22
0.75
4.12
6.09
Kg
as-is
1.31
0.18
0.22
0.06
4.5
6.27
% of Input
as-is
16
4
12
68
100
% of Output
as-is
20
3
3
1
73
100
„
Solids
Content
84%
310mg/L
NA
480 mg/L
Solids
Content
69%
51%
35%
16%
740 mg/L
—
==— ============^==^=^======:
Solids
Kg (dry wt)
0.84
0
0.002
0.842
Solids
Kg (dry wt)
0.904
0.092
0.077
0.01
0.003
1.086
% of Input
99 R
w v*O
0.2
% of Output
83 2
WWlb
8 5
V* w
7 i
1 t 1
0 Q
V/tW
0.3
vo
-------
FIGURE 4-7. MASS DISTRIBUTION IN HIGH SOIL TEST
Municipal Water
0.22 Kg
310 mg/L TSS
0.00 Kg Solids
Coarse Oversize
0.1 8 as is
57% Solids
0.092 Kg solids
Fine Particle
Cake
0.22 Kg as is
35% Solids
0.077 Kg solids
Fine Oversize
0.06 Kg as is
16% Solids
0.010 Kg Solids
ATS Effluent
4.12 Kg
480 mg/L TSS
0.02 Kg Solids
Washed Soil
1.31 Kg as is
69% Solids
0.904 Kg Solids
Combined Dewatering
Effluent
4.50 Kg as is
740 mg/L
0.003 Kg Solids
TOTAL MASf
SOLID MASS
INPUT
6.09
0.842
OUTPUT
6.27
1.086
% DIFF
3.0
29.0
-------
19
Percentage of Total Output Mass
O
c
a
3
in
O
O
-q
Tl
O
O
o
o
m
ro
o
en
o
en
O
o
T
CD
O
ID
O
o
o
en
•ya
m
i
oo
a
co'
CT
C
6'
D
o O
CO r—r-
°. Q
t/>
O
C
Q
0)
C/)
-------
TABLE 4-21. AVERAGE PENTA DISTRIBUTION-LOW PENTA SOIL WASHER TEST
INPUT
Feed Soil
Municipal Water
Thickener Sol'n
TOTAL
OUTPUT
WS
CO
FPC
FO
CDE
TOTAL
Stream
As Is wt.
(Kg)
1.00
5.45
0.82
7.27
Stream
As Is wt.
(Kg)
1.18
0.14
0.22
0.06
5.00
6.60
Cone.
(ppm)
130
0
0
Cone.
(ppm)
14
170
270
96
14
Penta
Mass
mg
130
0
0
130
Penta
Mass
mg
16.5
23.8
59.4
5.8
70
175.5
Dist.
(%)
100
0
0
Dist.
(%)
9
13
34
3
40
Calc. Cone.
on Solids
(ppm)
144
•
Calc. Cone.
on Solids
(ppm)
14
239
953
1042
-
o
Penta Removal Efficiency = [1 - (14/130)] 100 = 89.2%
-------
Percentage of Total Output Mas
Concentration of Penta (mg/kg or mg/l)
M
O
M
01
OJ
O
Ol
Ol
ON
w
CO
c
-J
n>
I
IO
TI
CD
IS
r-t-
Q
(I
I o
tn C
2. rt-
"
3-
CD
Q
3
w
-------
Efficiency is 89% (84-92% range), which is very close to the vendor's 90% claim.
Using the weighted mass figures, the calculated % Removal Efficiency would be
87.3%. If the % Removal Efficiency is recalculated on the basis of the mass
contribution of penta on the Washed Soil to the total penta output mass, the
value is increased only to 90.6%.
Since the masses of the input and output streams differ, even on the as-
is basis, the mass of penta in each of the output streams must also be
considered. In this case, 16.5 mg (9%) of the combined penta mass remains in the
Washed Soil (1.18 kg) resulting from processing 1 kg of Feed Soil. The average
output penta mass (175.5 mg) is considerably greater (35%) than the input mass
(130 mg) , perhaps reflecting the previously noted improvement in extractability
brought about by the soil washing prior to the analyses.
While it would have been informative to know the distribution of penta
in the water and solid phases of a particular output stream, such analyses were
not obtained. Approximations of the contributions from solid and liquid phases
of a sample can, however, be calculated if one assumes that the associated water
is comparable in its penta concentration to that in the CDE (14 mg/L) . For
example, for the FPC, with a mass of 0.22 kg from 1.0 kg of Feed Soil arid
containing 30% solids and a weighted average penta concentration of 270 mg/kg,
the penta concentration in/on the solids would be:
(0.22 kg)(270 mg/kg) - (0.22 kg)(0.3)X + [0.22-(0.22)(0.30)](14 mg/L)
fine particle cake solid phase liquid phase
X - 953 mg/kg
And, reversing the logic, if the CDE contains even 1000 mg/L of suspended solids
(650 and 740 mg/L were found in the CDE from the Low and High Penta Soil Washer
tests) and these solids contain -1000 mg/kg of penta, the solids would contribute
only 5 mg to the 14 mg/L found.
4.4.2.4 Penta Distribution - High Penta SW Test
Samples of input and output fractions from the High Penta test also were
analyzed for penta. Averaged concentration data are presented in Table 4-22 and
Figure 4-10. Using the change in concentration from the Feed Soil to the Washed
Soil, the average penta Removal Efficiency is 87%, still quite close to the
vendor's 90% claim, but with a wide range: 51% to 94%. On the basis of mass, the
penta Removal Efficiency decreases to 83.2%.
In this case, only 114 mg (11%) of the combined penta output mass is in
the Washed Soil (1.31 kg) resulting from processing 1.0 kg of Feed Soil. While
the solids balance is reasonable good, the output penta mass (1066 mg) is
considerably greater than the input mass (680 mg). If % Removal Efficiency is
calculated on the basis of the penta mass contribution in the Washed Soil to the
total output penta mass, a value of 89.3% is obtained. Estimated penta
concentrations have also been calculated for the solids portion of the output and
are also presented in the table.
64
-------
TABLE 4-22. AVERAGE PENTA DISTRIBUTION-HIGH PENTA SOIL WASHER TEST
Calc. Cone.
on Solids
(ppm;
790
Feed Soil
Municipal Water
Thickener Sol'n
ATS Effluent
TOTAL
Calc. Cone.
on Solids
ppm)
90
2680
3440
5000
Penta Removal
Efficiency = [1 - (87/680)1100 = 87.2V
-------
Percentage of Total Output Mass
Concentration of Penta (mg/kg or mg/l)
(Thousands)
a
o
if
c
to
*>
I—«
o
TJ
Q
o
§
I
•5°
» CO
9 '' '"
•" ~\
? 0)
ft Q
•3
-------
4.4.2.5 Polynuclear Aromatic Hydrocarbon Distribution
The distribution of PAHs in the individual streams is presented in Tables
4-23 and 4-24 for Low and High Penta tests. Each of these large tables ends with
a summary of total PAHs, carcinogenic PAHs and non-carcinogenic PAHs. Based on
the results, the highest concentrations and masses of PAHs are found in the FPG
while some PAHs are not detected in one or more streams. Using the concentrations
and combined masses of PAHs for the Feed Soil and the Washed Soil, % Removal
Efficiencies were 83% and 80% in the Low Penta test and 88 and 85% in the High
Penta test. Removal Efficiencies for individual PAHs ranged from 45% to 100% in
the Low Penta test where detectable and measureable levels were detected in the
Washed Soil and 86% to 100% in the High Penta test, where the Washed Soil
retained detectable levels of all the PAHs except naphthalene.
Of considerable interest is the observation that the distribution of the
masses of total PAHs in the output streams is quite different from that obtained
for penta, with much more of the PAH mass in the Fine Particle Cake and much less
in the Combined Dewatering Effluent. While concentration data also indicates the
same behavior, they fail to take into consideration the mass of the individual
output streams. Over 55-60% of the PAHs are concentrated in an output stream that
amounts to about 22% of the original Feed Soil mass. This effect probably is a
result of the decreased solubility of the PAHs in the water and greater tenacity
to the surface of the fine particles. The results, taken from Tables 4-23 and 4-
24 and compared with those for penta from Tables 4-21 and 4-22, are summarized
in Table 4-25.
The "Total Carcinogenic PAHs" given in Tables 4-23 and 4-24 include those
listed below in Column I and derived from Group B2, Probable Human Carcinogens
from the National Primary and Secondary Drinking Water Regulations (FR, 55 #143.
30397 [July 25, 1990]). Other PAHs investigated included those in Column II,
which are derived from Group D, Insufficient Data to Assess Carcinogenic
Potential," and for acenaphthene and acenaphthalene, which are IARC Class 3:
I II
Benz(a)anthracene --L,H Acenaphthene --H
Benzo(b)fluoranthene --L,H Acenaphthylene
Benzo(k)fluoranthene --H Anthracene --L,H
Benzo(a)pyrene Fluoranthene
Chrysene --L,H Fluorene --L,H
Dibenz(a,h)anthracene Naphthalene
Indeno(1,2,3,-cd)pyrene Phenanthrene -L,H
Benzo(g,h,i)perylene
Pyrene --L,H
Those PAHs followed by an "L" or "H" were detected at some quantity in all
streams of the Low or High Penta test, respectively.
67
-------
TABLE 4-23. AVERAGE PAH CONCENTRATION AND MASS
1 Kg FEED SOIL BASIS—LOW PENTA SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
Total
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
—
Naphthalene
Mean
Cone.
(ppb)
ND
ND
ND
2033
ND
1913
ND
- NA
No. of
Samples
3
4
Mass
(mg)
0.0
0.0
0.0
0.0
0.3
0.0
0.1
0.0
0.4
NA
i
Acenaphthylene
Mean
Cone.
(Ppb)
ND
ND
ND
150
ND
ND
ND
NA
No. of
Samples
1
Mass
(mg)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
NA
—
Acenaphthene
Mean
Cone.
(PPb)
2225
ND
ND
2817
6250
1065
ND
100%
No. Of
Samples
4
6
6
4
Mass
(mg)
2.2
0.0
2.2
0.0
0.4
1.4
0.1
0.0
1.8
-17%
i ' •' i -...I I.
Fluorene
Mean
Cone.
(PPb)
12463
ND
6885
20789
38750
18500
42
45%
No. Of
Samples
8
2
9
8
8
10
Mass
(mg)
12.5
0.0
12.5
8.1
3.0
8.5
1.1
0.2
20.9
68%
00
ND = Not Detected
N/A = Not Applicable
B:\TABLE422.WK1
-------
TABLE 4-23 {CON'T). PAH COMPOUNDS, LOW PENTA SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
Total
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
Phenanthrene
Mean
Cone.
(PPb)
30389
ND
7680
50625
103375
42857
70
75%
No. Of
Samples
9
4
8
8
7
9
Mass
(mg)
30.4
0.0
30.4
9.1
7.4
22.6
2.5
0.4
41.9
38%
Anthracene
Mean
Cone.
(PPb)
136375
ND
17083
170250
456667
106000
280
87%
No. of
Samples
8
6
4
3
4
11
Mass
(mg)
136.4
0.0
136.4
20.2
24.8
99.6
6.3
1.5
152.3
12%
Fluoranthene
Mean
Cone.
(PPb)
17222
ND
3100
15111
41500
10663
22
82%
No. of
Samples
9
6
9
8
8
9
Mass
(mg)
17.2
0.0
17.2
3.7
2.2
9.1
0.6
0.1
15.7
-9%
Pyrene
Mean
Cone.
(PPb)
27889
ND
3450
24111
69250
13463
33
88%
No. Of
Samples
g
6
9
8
8
10
Mass
(mg)
279
0.0
27.9
4 1
3.5
15 1
0.8
0.2
23.7
-15%
o\
VO
ND =
N/A =
Not Detected
= Not Applicable
B:\TABLE422. WK1
-------
TABLE 4-23 (CON'T). PAH COMPOUNDS, LOW PENTA SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
Total
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
Benzo(a)anthracene
Mean
Cone.
(ppb)
3950
ND
817
3700
13025
2438
18
79%
No. Of
Samples
8
3
7
8
8
3
Mass
(mg)
4.0
0.0
4.0
1.0
0.5
2.8
0.1
0.1
4.6
16%
Chrysene
Mean
Cone.
(PPb)
11450
ND
2210
9089
30125
6375
18
81%
NO. Of
Samples
8
4
9
8
8
4
Mass
(mg)
11.5
0.0
11.5
2.6
1.3
6.6
0.4
0.1
11.0
-4%
Benzo(b)fluoranthene
Mean
Cone.
(PPb)
2425
ND
870
1850
6720
1783
ND
64%
No. Of
Samples
4
1
4
5
6
Mass
(mg)
2.4
0.0
2.4
1.0
0.3
1.5
0.1
0.0
2.9
18%
Benzo(K)fluoranthene
Mean
Cone.
(PPb)
1500
ND
ND
1300
7350
817
ND
100%
No. Of
Samples
1
3
6
3
Mass
(mg)
1.5
0.0
1.5
0.0
0.2
1.6
0.0
0.0
1.8
23%
ND = Not Detected
N/A = Not Applicable
B:\TABLE422. WK1
-------
TABLE 4-23 (CON'T). PAH COMPOUNDS, LOW PENTA SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
Total
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
Benzo(a)pyrene
Mean
Cone.
(ppb)
1100
ND
NO
2133
4800
928
ND
100%
No. of
Samples
1
3
1
4
Mass
(mg)
1.1
0.0
1.1
0.0
0.3
1.0
0.1
0.0
1.4
28%
Benzo(g,h,i)perylene
Mean
Cone.
(PPb)
ND
ND
ND
4700
ND
1100
ND
NA
No. of
Samples
2
1
Mass
(mg)
0.0
0.0
0.0
0.0
0.7
0.0
0.1
0.0
0.7
NA
Total PAHs
Mean
Cone.
(PPb)
246988
0
42095
308658
777812
207900
483
83%
Mass
(mg)
247.0
0.0
247.0
49.7
44.9
169.7
12.3
2.5
279.1
13%
Carcinogenic
PAHs
Mean
Cone.
(PPb)
20425
0
3897
18072
62020
12340
36
81%
Mass
(mg)
20.4
0.0
20.4
4.6
2.6
13.5
0.7
0.2
21.7
6%
Non-Carcinogenic
PAHs
Mean
Cone.
(PPb)
226563
0
38198
290586
715792
195560
448
83%
Mass
(mg)
226.6
0.0
226.6
45.1
42.3
156.2
11.6
2.3
257.5
14%
ND = Not Detected
N/A = Not Applicable
B:\TABLE422-WK1
-------
TABLE 4-24. AVERAGE PAH COMPOUND CONCENTRATION AND MASS
ONE KILOGRAM FEED SOIL BASIS, HIGH PENTA SOIL WASHER TEST
Soil Washer Stream
inputs.
Feed Soil
Municipal Water
ATS Treated Water
Totals
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
Naphthalene
Mean
Cone.
(ppb)
ND
ND
ND
ND
1300
4800
2600
26
N/A
No. of
Samples
1
2
3
3
(mg)
0
0
0
0.0
0.0
0.2
1.0
0.1
0.1
.5
N/A
Acenaphthylene
Mean
Cone.
(ppb)
1300
ND
ND
ND
ND
ND
ND
36
100%
No. of
Samples
1
1
(mg)
1.3
0
0
1.3
0.0
0.0
0.0
0.0
0.2
0.2
-87%
— —
Acenaphthene
Mean
Cone.
(Ppb)
21000
ND
ND
2400
28000
54000
25000
110
89%
No. Of
Samples
20
. 11
16
19
21
20
(mg)
21
0
0
21.0
3.1
4.8
117
1.4
0.5
21.5
2%
Fluorene
Mean
Cone.
(PPb)
24000
ND
ND
2700
31000
59000
28000
110
89%
No. of
Samples
16
10
12
17
18
11
(mg)
- 24
0
0
24 0
3.5
5.3
12.8
1.6
0.5
23.6
-2%
ND = Not Detected
N/A = Not Applicable
B:\TABLE423.WK1
-------
TABLE 4-24 (CON'T). PAH COMPOUNDS, HIGH PENTA SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
ATS Treated Water
Totals
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
Phenanthrene
Mean
Cone.
(PPb)
80000
ND
ND
10000
99000
230000
82000
370
88%
No. of
Samples
19
19
11
14
11
17
(mg)
80
0
0
80.0
12.8
17.1
49.8
4.6
1.7
85.9
7%
Anthracene
Mean
Cone.
(PPb)
46000
ND
ND
4700
70000
104000
60000
170
90%
No. of
Samples
19
19
16
19
17
21
(mg)
46
0
0
46.0
6.0
12.1
22.5
3.3
0.8
44.7
-3%
Fluoranthene
Mean
Cone.
(PPb)
69000
ND
110
9800
97000
210000
72000
370
86%
No. Of
Samples
19
1
19
13
14
11
22
(mg)
69
0
0.5
69.5
12.5
16.7
45.5
4.0
1.7
80.4
16%
Pyrene
Mean
Cone.
(PPb)
74000
ND
110
9300
95000
179000
85000
310
87%
No. of
Samples
20
1
19
17
15
21
22
(mg)
74
0
0.5
74.5
11.9
16.4
38.8
4.7
1.4
73.2
-2%
ND = Not Detected
N/A - Not Applicable
B:\TABLE423.WK1
-------
TABLE 4-24 (CON'T). PAH COMPOUNDS. HIGH PENTA SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
ATS Treated Water
Totals
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
Benzo(a)anthracene
Mean
Cone.
(Ppb)
17000
ND
ND
2200
27000
47000
22000
78
87%
No. of
Samples
20
12
17
19
21
21
(mg)
17
0
0.0
17.0
2.8
4.7
10.2
1.2
0.4
19.2
13%
i =
Chrysene
Mean
Cone.
(Ppb)
30000
ND
ND
3800
42000
75000
40000
120
87%
No. of
Samples
20
15
17
19
21
21
(mg)
30
0
0.0
30.0
4.9
7.2
16.3
2.2
0.5
31.1
4%
Benzo(b)fluoranthene
Mean
Cone.
(PPb)
13000
ND
ND
1400
14000
31000
13000
51
89%
No. of
Samples
14
6
16
19
21
15
(mg)
13
0
0.0
13.0
1.8
2.4
6.7
0.7
0.2
11.9
-9%
Benzo(k)fluoranthene
Mean
Cone.
(PPb)
11000
ND
ND
1500
13000
26000
11000
45
86%
NO. Of
Samples
11
5
9
13
21
12
(mg)
11
0
0.0
11.0
1.9
2.2
5.6
0.6
0.2
10.6
-4%
ND = Not Detected
N/A = Not Applicable
B:\TABLE423.WK1
-------
TABLE 4-24 (CON'T). PAH COMPOUNDS. HIGH PENTA SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
ATS Treated Water
Totals
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
i
Benzo(a)pyrene
Mean
Cone.
(Ppb)
8200
ND
ND
550
8200
19000
9800
49
93%
No. of
Samples
10
2
15
18
21
6
(mg)
8.2
0
0.0
8.2
0.7
1.4
4.1
0.5
0.2
7.0
-15%
Indenop ,2,3-cd)pyrene
Mean
Cone.
(Ppb)
3800
ND
ND
ND
4000
15000
6200
45
100%
No. of
Samples
3
4
1
21
1
(mg)
3.8
0
0.0
3.8
0.0
0.7
3.3
0.3
0.2
4.5
18%
Dibenz(a,h)anthracene
Mean
Cone.
(Ppb)
1700
ND
ND
ND
1400
1900
2300
ND
100%
No. of
Samples
3
4
"""1
8
(mg)
1.7
0
0.0
1.7
0.0
0.2
0.4
0.1
0.0
0.8
-54%
Benzo(g,h.i)perylene
Mean
Cone.
(Ppb)
4100
ND
ND
ND
4400
8300
4600
ND
100%
No. of
Samples
2
3
2
8
(mg)
4.1
0
0.0
4 1
0.0
0.8
1.8
0.3
0.0
2.8
-31%
ND = Not Detected
N/A = Not Applicable
B:\TABLE423.WK1
-------
TABLE 4-24(CON'T). PAH COMPOUNDS, HIGH PENTA
SOIL WASHER TEST
Soil Washer Stream
Inputs:
Feed Soil
Municipal Water
ATS Treated Water
'Totals
Outputs:
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewatering Effluent
Totals
Percent Gain or (Loss)
Removal Efficiency
Total
PAHs
Mean
Cone.
(ppb)
404100
0
220
48350
535300
1064000
463500
1890
88%
(mg)
404.1
0
0.9
405.0
61.8
92.2
230.5
25.8
8.6
418.9
3%
Carcinogenic
PAHs
Mean
Cone.
(ppb)
71000
0
0
8900
96000
179000
86000
294
87%
(mg)
71
0
0.0
71.0
11.4
16.5
38.8
4.8
1.3
72.8
3%
Non-Carcinogenic
PAHs
Mean
Cone.
(ppb)
333100
0
220
39450
439300
885000
377500
1596
88%
(mg)
333.1
0
0.9
334.0
50.4
75.7
191.8
21 0
7.3
346.1
4%
ND =
N/A>
Not Detected
= Not Applicable
B:\TABLE423.WK1
-------
TABLE 4-25. COMPARISON OF PENTA AND PAH DISTRIBUTION IN OUTPUT STREAMS
FROM SOIL WASHING TESTS
output stream Low Penta test ' High Penta test
PAHs Penta PAHs Penta
Coarse Oversize
Fine Particle Cake
Fine Oversize
Washed Soil
Combined Dewater. Effl.
16.1
60.8
4.4
17.8
0.9
13.6
33.8
3.3
9.4
39.9
22.0
55.0
6.2
14.8
2.0
23.6
26.8
5.1
10.7
33.8
on 100 x mass in output fraction
total output mass
The Removal Efficiency for Total Carcinogenic PAHs in the Low Penta and
the High Penta Soil Washer tests, respectively, were 83% and 88%, based on
differences in total concentrations between Feed Soil and Washed Soil.
4.4.2.6 Total Organic Carbon and Total Recoverable Petroleum
Hydrocarbons
The average TOG removal efficiency for the Low Penta Soil Washer Test was
84% (Feed Soil: 12,000 mg/kg; Washed Soil: 1,900 mg/kg) with a range of 57% to
93%. In the High Penta test, the TOG removal efficiency was 81% (Feed Soil:
17,000 mg/kg; Washed Soil: 3,200 mg/kg) with a range from -21% to 99%. The high
TOG values probably reflect the presence of woody material and the carrier oil
used in penta treatment of wood. In both tests, the wide range suggests that the
average removal efficiency is only useful in a very general sense. The data are
provided in Volume II.
The mean TRPH removal efficiency in the Low Penta Soil Washer test was
94% (Feed Soil: 3,800 mg/kg; Washed Soil: 210 mg/kg) with a range of 89% to 98%.
In the High Penta test, the efficiency was 92% (Feed Soil: 8,900 mg/kg; Washed
Soil: 690 mg/kg) with a range of 41% to 98%. The data are provided in Volume II.
4.4.2.7 Metals Behavior
Chromium, copper, arsenic -- Average results for the Low and High Penta Soil
Washer tests are summarized in Table 4-26 and 4-27 and Figures 4-11 and 4-12.
Using the same 1 kg of as-is Feed Soil basis, the average mass for each metal in
each process stream was calculated by multiplying the average measured (or
estimated) concentration (see Volume II for the individual data and statistical
analyses) for each metal by the corresponding process stream mass. In both the
Low and High Penta tests, the metals are quite consistently distributed, with the
Fine Particle Cake containing the highest concentration and mass of each metal.
On a concentration basis, Coarse Oversize was the next largest contributor, but
on a mass content the Washed Soil was either equal to or greater than the Coarse
Oversize, as noted below:
77
-------
================
INPUT
Feed Soil
Municipal Water
Thickener Soln
ATS Effluent
TOTAL
OUTPl IT
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewater Eff
TOTAL
% Removal Efficiency
=======
stream
kgae-is
1.00
5.45
0.82
0.00
7.27
1.18
0.14
0.22
0.06
5.00
6.60
TABL
—
Arsenic % of
ppm mg total
13.50 13.50 100
0.00 0.00 0
0.00 0
0.00 0
13.50 100
5.00 5.90 31
23.60 3.30 18
41.20 9.06 48
6.70 0.34 2
0.03 0.15 1
18.76 100
63.0 56.3
= 4-26. AVERAGE METAL
Chromium % of
ppm mg total
17.00 17.00 100
0.00 0.00 0
0.00 0
0.00 0
17.00 100
8.80 10.38 40
28.70 4.02 16
48.00 10.56 41
9.70 0.58 2
0.03 0.15 1
25.69 100
48.2
5 BALANCE- LOW PENT
Copper % of
ppm mg total
14.70 14.70 95
0.15 0.82 5
0.00 0
0.00 0
15.52 100
5.60 6.61 29
30.00 4.20 19
49.40 10.87 48
11.80 0.71 3
0.04 0.20 1
22.58 100
61.9
ASWTEST
• "
Barium % of
ppm mg total
no Info
13.50 15.93 24
236.50 33.11 49
80.30 17.67 26
6.70 0.40 1
0.12 0.60 1
67.71 100
•
Lead % of
ppm mg total
no Info
5.00 5.90 23
16.20 2.27 9
75.70 16.65 65
11.30 0.68 3
0.05 0.25 1
25.75 100
.
Mercury % of
ppm mg total
no Info
0.10 0.12 73
0.00 0.00 0
0.20 0.04 27
0.00 0.00 0
0.00 0.00 0
0.16 100
00
Hie: 4-24
=======
IMpl IT
Feed Soil
Municipal Water
Thickener Soln
ATS Effluent
TOTAL
OlITPI IT
Washed Soil
Coarse Oversize
Fine Particle Cake
Fine Oversize
Combined Dewater Eff
TOTAL
% Removal Efficiency
—
stream
kgas-is
1.00
0.22
0.75
4.12
6.09
1.31
0.18
0.22
0.06
4.50
6.27
TABL
Arsenic % of
ppm mg total
31.40 31.40 99
0.01 0.00 0
0.00 0
0.04 0.16 1
31.57 100
11.60 15.20 31
64.10 11.54 24
88.00 19.36 40
20.50 1.23 3
0.22 0.99 2
48.31 100
63.1 51.6
= 4-27. AVERAGE METAL
Chromium % of
ppm mg total
38.50 38.50 100
0.00 0.00 0
0.00 0
0.01 0.04 0
38.54 100
14.00 18.34 27
105.60 19.01 28
123.60 27.19 40
36.20 2.17 3
0.20 0.90 1
67.61 100
63.6 52.4
8 BALANCE -HIGH PENT
Copper % of
ppm mg total
33.80 33.80 100
0.03 0.01 0
0.00 0
0.02 0.08 0
33.89 100
10.10 13.23 24
63.20 11.38 21
123.10 27.08 50
26.30 1.58 3
0.29 1.31 2
54.57 100
70.1 60.9
ASWTEST
===========
Barium % of
ppm mg total
0.00 0
0.07 0.02 100
0.00 0
0.00 0
0.02 100
23.15 30.33 31
256.60 46.19 47
87.50 19.25 20
15.80 0.95 1
0.39 1.76 2
98.47 100
Lei
PI
2
4
1
file: 4-25
ad
pm
6.83
>1.40
16.90
1.10
0.50
mo
no Info
8.95
3.85
10.32
0.67
2.25
26.03
—— •
%of
total
34
15
40
3
9
100
Mercury
ppm mg
%of
total
no info
0.10
0.10
0.20
0.20
0.00
0.13
0.02
0.04
0.01
0.00
0.21
— —
64
9
21
6
0
100
i
-------
c
4>
5f
V
O.
<3
v
Q.
»
0
O
FIGURE 4-11, EFFLUENT METALS IN BIOTROL SOIL WASHER
Low PENTA SW TEST
IUU
WaaOed Soil
Araviie
Fine Over»OK
Chromium Copper
CDE IT*)?! Fine Particle Cake l\\l Coarse Ovei
FIGURE 4-12, EFFLUENT METAL.S IN BIOTROL SOIL WASHER
HIGH PE^TA SW TEST
100
An en ic
Washed Soil
Fine Overaas
Chromium Copper
CDE f?vT Fhe Portele Cake . l\\l Coarse Oversea:
7-9
-------
cone: FPC > CO > FO > WS > CDE
mass: FPC > WS > CO > FO > CDE
Large discrepancies between influent and effluent masses for both tests
may reflect the high variations in the original analyses (i.e., high standard
deviations) or may, as noted earlier, reflect improved accessibility of metals
for analysis in product streams. Removal efficiencies (based on metal masses in
the Washed Soil compared to those in 1.0 kg of the Feed Soil) in the Low Penta
test were: As: 56%; Cr: 39%; and Cu: 55%. In the High Penta test, the comparable
removals were: As: 52%; Cr: 52%; and Cu: 61%.
Other Metals -- Other metals were also analyzed in the various influent and
effluent streams (also reported in Table 4-26 and 4-27). The number of samples
was limited and there were large variations in relative standard deviations.
Therefore, only general trends should be discussed. Low concentrations of lead
were found and appeared to follow the same distribution as the other .metals.
Barium was consistently encountered, and appeared to be distributed differently
in the output streams,with the Coarse Oversize containing about half of the mass
of barium in both Low and High Penta tests. The source of these metals is
unknown. Cadmium, selenium, and silver were either not detected or found in very
low concentrations in any process streams.
EP Toxicity -- Tables 4-28 and 4-29 summarize the EP Toxicity test results of
output streams for the Low Penta and the High Penta Soil Washer tests,
respectively. Only chromium and copper were detected in the Low Penta test in
about equal concentrations in the Fine Particle Cake, Coarse Oversize, and Fine
Oversize. In the High Penta SW test, arsenic, chromium and copper were found,
primarily in the leachate from the Fine Particle Cake, with lower concentrations
in the leachates from the Fine Oversize and Washed Soil. No cadmium, lead,
mercury, selenium, or silver was detected in either test. Barium was present in
every process stream with the Washed Soil and the Fine Oversize being the largest
sources in both SW tests. Unfortunately, there are no EP Toxicity Test data for
the Feed Soils for comparison. Since barium is not commonly used in the wood
preserving industry, there is no explanation for its presence.
4.4.2.8 Flow Rate Stability
Low Penta SW Material Flow Rates -- The Feed Soil rate planned for the Low Penta
SW study was approximately 275 kg/hr (610 Ibs/hr) on an as-is weight basis. In
fact, the Soil Washer operated for the first 28 hours of the test at 250 to 300
kg/hr (550 to 660 Ibs/hr). Starting at hour 28, clogging, caused by Feed Soil
compaction at the base of the hopper, impeded the transfer of soil to the feed
conveyor. Efforts to eliminate the problem were unsuccessful. This resulted in
a sharp decline in the Feed Soil rate to about 125 kg/hr (280 Ibs/hr), or 45% of
planned operation. For the entire test the average Feed Soil rate was 220 kg/hr.
Washed Soil output rate closely followed the fluctuations in the Feed
Soil rate. This suggests that retention time within the Soil Washer did not
produce a significant response lag. The Washed Soil output rate was consistently
greater than the Feed Soil input rate because of water uptake during the soil
washing process. The minor solids output streams -- Coarse Oversize, Fine
80
-------
TABLE 4-28. EP TOXICITY TEST RESULTS - LOW PENTA SW TEST
(U9/L)
Process
Stream
Washed
Soil
Coarse .
Oversize
Fine
Oversize
Fine
Particle
Cake
Sample-ID
Test Day-Time Period
SWL-WS-01-03
SWL-WS-02-03
Average
Standard Deviation
SWL-CO-01-03
SWL-CO-02-03
Average
Standard Deviation
SWL-FO-01-03
SWL-FO-02-03
Average
Standard Deviation
SWL-FPC-01-03+
SWL-FPC-02-03
Average
Standard Deviation
Arsenic
< 100.0
< 100.0
< 100.0
< 100.0
< 100.0
< 100.0
< 100.0
< 100.0
Barium
522.0
553.0
538
22
148.0
201.0
175
37
420.0
469.0
445
35
314.0
434.0
374
85
Cadmium
< 5.0
< 5.0
< 5.00
< 5.00
< 5.0
< 5.0
< 5.00
< 5.00
Chromium
10.3
10.9
11
0
15.0
13.6
14
1
13.2
15.1
14
1
14.7
12.0
13
2
Copper
NA
< 10.0
16.0
314.0
165
211
13.0
13.2
13
0
10.9
12.3
12
1
Lead
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
Mercury
< 0.20
< 0.20
< 0.20
< 0.20
< 0.20
< 0.20
< 0.20
< 0.20
Selenium
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
Silver
< 10.0
< 10.0
< 10.0
< 10.0
< 10.0
< 10.0
< 10.0
< 10.0
+ Archive sample
filename: TAB4-28.WK1
-------
TABLE 4-29. EP TOXICITY TEST RESULTS - HIGH PENTA SW TEST
00
S3
Process
Stream
Washed
Soil
Coarse
Oversize
Fine
Oversize
Fine
Particle
Cake
Sample-ID
Test Day-Time Period
SWH-WS-01-01
SWH-WS-02-01
SWH-WS-04-01
SWH-WS-05-01
SWH-WS-06-01
Average
Standard Deviation
SWH-CO-01-D1
SWH-CO-02-01
SWH-CO-04-01
SWH-CO-05-01
SWH-CO-06-01
. Average
Standard Deviation
SWH-FO-01-01
SWH-FO-02-01
SWH-FO-04-01
SWH-FO-05-01
SWH-FO-06-01
Average
Standard Deviation
SWH-FPC-01-01
SWH-FPC-02-01
SWH-FPC-04-01
SWH-FPC-05-01
SWH-FPC-05-01
Average
Standard Deviation
Arsenic
< 100.0
< 100.0
< 100.0
< 100.0
< 100.0
100.0
< 100.0
< 100.0
< 100.0
< 100.0
< 100.0
< 100.0
120.0
130.0
139.0
130
10
214.0
219.0
148.0
174.0
174.0
186
30
Barium
69.2
439.0
301.0
449.0
312.0
314
153
83.6
169.0
123.0
461.0
121.0
192
154
348.0
423.0
252.0
359.0
311.0
339
63
272.0
220.0
120.0
207,0
207.0
205
55
Cadmium
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
< 5.0
Chromium
14.7
< 10.0
< 10.0
< 10.0
< 10.0
15
0
< 10.0
18.3
< 10.0
< 10.0
< 10.0
18
0
< 10.0
22.0
11.9
< 10.0
11.9
15
6
22.6
19.6
< 10.0
< 10.0
< 10.0
21
2
Copper
26.1
17.8
11.7
23.0
10.8
18
7
18.9
50.7
28.1
36.2
31.5
33
12
23.2
37.1
29.0
29.8
39.5
32
7
60.9
47.6
27.8
47.1
47.1
46
12
Lead
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
< 50.0
Mercury
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
Selenium
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
< 200.0
Silver
< 10.0
< 10.0
< 10.0
< 10.0
< 100
< 10.0
< 10.0
< 10.0
< 10.0
< 10 0
< 10.0
< 10.0
< 10.0
< 10.0
< 100
< 10.0
< 10.0
< 10.0
< 10.0
< 100
-------
Oversize, and Fine Particle Cake -- also reflected variations in Feed Soil rates
but to a lesser extent. Flow rates are graphically summarized in Figure 4-13.
Municipal Water (MW) was the primary aqueous input stream, averaging .20
liters per minute (5.3 gpm) after a few initial adjustments (through hour 10),
while the Cationic Polymer input stream (Floe) contributed a steady but minor
flow rate of 2.7 liters per minute (0.7 gpm). Aqueous flow rates are shown
graphically in Figure 4-14.
The Combined Dewatering Effluent flow rate closely followed fluctuations
in the Municipal Water input rate, again suggesting that retention time did not
create a significant response lag. The Combined Dewatering Effluent flow rate
was approximately equal to the Municipal Water input rate because the Cationic
Polymer input stream offset water uptake by the solids-bearing output streamss.
High Penta SW Material Flow Rates --In this test the Feed Soil rate was stable
between 140 and 150 kg/hr (310 and 330 Ibs/hr) for the first 40 hours until
mechanical problems (compaction) with the feed system forced operation at a
reduced rate for about 10 hours. After the scheduled break to deliver Fine
Particle Slurry to the Slurry Bio-Reactor (hours 50 to 80), a vibrating device
was attached to the outside of the feed hopper and corrected the soil compaction
problem. Feed Soil flow rate increased by 50 kg/hr (110 Ibs/hr) to approximately
200 kg/hr (440 Ibs/hr). The unit operated at this level for 30 hours. During
hours 110 to 150 intermittent mechanical failures again destabilized the feed
delivery system and caused a drop in the Feed Soil rate.
The Washed Soil flow rate closely reflected changes in the Feed Soil
rate, as during the Low Penta test. However, the difference between the two
flow rates was greater, reflecting the higher moisture in the Washed Soil during
this test. The smaller output solids streams behaved as in the Low Penta test.
Treated effluent from the BioTrol Aqueous Treatment System was the
primary aqueous input stream, averaging 11 liters per minute (2.9 gpm). However,
treated water flow rate never stabilized during the test, starting at about 14
liters per minute (3.7 gpm) and gradually decreased to 7 liters per minute (1.9
gpm). Municipal Water supplied 1 to 3 liters per minute (0.3 to 0.8 gpm) during
the first 90 hours of operation after which it was used only intermittently; over
the course of the test this source provided an average of 0.6 liters per minute
(0.2 gpm). The Cationic Polymer flow rate was, once again, constant throughout
the test at about 2 liters/minute (0.5 gpm).
Figures 4-15 and 4-16 summarize the flow rate behavior of the various
input and output streams during the High Penta SW test.
Feed Soil Flow Rate Effects on Organic Contaminant Removal -- As noted earlier,
during the Low Penta SW test the Soil Washer experienced a significant drop in
Feed Soil input rate. Analysis of the data presented in Table 4-30, which are
summarized graphically in Figure 4-17, suggests that:
(1) penta concentration in the Washed Soil was not affected by the drop in
Feed Soil flow rate during the Low Penta SW Test; and
83
-------
00
C.
0)
4-1
0
400
350
300
250
200
150 i-
100 -
50 -
0
0
Elapsed Time (Hours)
Figure 4-13.. Solids Stream Flows - Low Penta Soil Washer Test
-------
00
Ui
0)
•«•«
a
c
L.
0)
a
in
0)
o
a:
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
•FLOC
0
10
20
30
40
50
60
Elapsed Time (Hours)
Figure 4-14. Aqueous Stream Rates - Low Penta SW Test
-------
o
tt
350
300 -
250 -
200 -
150 -
too -
0 I 20 I 40 I 60 I 80 I 100 I 120 I 140
10 30 50 70 90 110 130 150
Elapsed Time (Hours)
Figure 4-15. Solids Stream Flows - High Penta Soil Washer Test
-------
00
3
C
I.
0)
a
(A
0)
o
or
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
t _
FLOC
0 ! 20 I 40 I 60 I 80 i 100 I 120 I 140
10 30 50 70 90 110 130 150
Elapsed Time (Hours)
Figure 4-16. Aqueous Streams - High Penta SW Test
-------
TABLE 4-30 COMPARISON OF AVERAGE AND WEIGHTED MASS FLOWS IN SOIL WASHER
(Kg/hr) .
stream
Feed Soil
Mun . Water
ATW
Thickener
Washed Soil
Coarse Oversize
Fine Oversize
Fine Particle Cake
Comb . Dewater . Ef f
Low
weighted
flow
220
1200
180
259.6
30.8
13.0
48.4
1100
Penta test
avg. std
flow
225.8
1156.0
161.2
265.3
32.7
12.8
50.7
1120.8
.
dev
60.6
163.8
10.5
45.2
5.4
0.7
15.2
118.8
High
weighted
flow
160
35.2
659.2
120
209.6
28.8
9.6
35.2
720
Penta test
avg. std
flow
157.9
75.5
664.7
113.6
212.1
31.8
9.8
36
743.8
dev
37.8
69
146.1
20
33.6
9.8
3.8
10
176.9
88
-------
00
O*
U
c
0
o
o>
Q.
260
240 h
220
200
180
160
140
120
100
80
60
40
20
0
MS
10
20
30
40
50
60
Elapsed Time (Hours)
Figure 4-17. Penta Concentration - Washed and Feed Soil
Low Penta Soil Washer Test
-------
(2) penta concentration in the Washed Soil appeared to remain stable between
10 and 20 mg/kg, averaging 14 ± 4.7 mg/kg (before data was weighted)
while concentrations in the Feed Soil averaging 126.7 +.20.4 mg/kg
(before weighting) and ranged from 80 to 160 mg/kg.
During the High Penta SW test, the Soil Washer operated under three
distinct Feed Soil flow rates: -150 kg/hr, ~200 kg/hr, and fluctuations, between
150 to 200 kg/hr. Analysis of variance between the three Feed Soil flow rates
and Washed Soil penta concentration data indicates (Figure 4-18) that:
(1) penta concentration in the Washed Soil was not significantly different
during the three Feed Soil flow rates; and
(2) penta concentration in the Washed Soil, with an average of 85.3 + 30.8
mg/kg (before weighting), appeared to remain stable between 50 and 100
mg/kg while the concentration in the Feed Soil ranged between 300 and
1100 mg/kg (non-weighted average: 657.8 ± 228.5 mg/kg).
It may be noted that these observations could be interpreted as indicating that
all penta that could be removed by the soil washing was being removed and
increased time in the soil washer during low feed rate periods does not improve
removal further.
A similar analysis of the PAH data from the Low Penta Soil Washer test
suggests that:
(1) PAH concentrations in the Washed Soil also were not affected by
decreases in Feed Soil flow rate during the Low Penta SW Test; and
(2) The total PAH concentration in the Washed Soil also appeared to
stabilize, at an average of about 42 mg/kg, and the carcinogenic PAH
concentration in the Washed Soil appeared to stabilize between 2 and 5
mg/kg (non-weighted average: 3.9 mg/kg) while Feed Soil concentration
ranged between 30 and 60 mg/kg.
Analysis of the High Penta SW test data indicates that:
(1) Carcinogenic PAH concentration in the Washed Soil was not significantly
different during the three Feed Soil flow rates; and
(2) Combined PAH concentration in the Washed Soil stabilized at an average
of 48.3 mg/kg (non-weighted). Feed Soil concentration averaged 404.1
mg/kg.
Similar results were observed for TOC and TRPH. In the Low Penta SW
test, TOC concentration in the Washed Soil stabilized between 1,400 and 2,300
mg/kg (average: 1,900 mg/kg) with Feed Soil concentration ranging between 4,000
and 21,000 mg/kg (average: 12,000 mg/kg). TRPH concentration in the Washed Soil
stabilized between 100 and 300 mg/kg (average: 210 ing/kg) with Feed Soil
concentration ranging between 2,000 and 10,000 mg/kg (average: 3,800 mg/kg).
90
-------
D>
\^
En
•o
*^^^ ^
Si
ofc
*<
C
Q.
1.2
1.1
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 I 20 I 40 I 60 I 80 I 100 I 120 I 140
10 30 50 70 90 110 130 150
Elapsed Time (Hours)
Figure 4-18.- Penta Concentration •• Washed and Feed Soil
High Penta Soil washer Test
-------
In the High Penta SW test, TOG in the Washed Soil stabilized between
1,000 and 5,000 mg/kg with concentrations in the Feed Soil ranging between 3,200
and 48,000 mg/kg. TRPH concentrations in the Washed Soil stabilized between 300
and 700 mg/kg with Feed Soil concentrations between 5,400 and 11,000 mg/kg.
4.4.3 BloTrol Aqueous Treatment System Performance
4.4.3.1 Critical Analyses
Pentachlorophenol (Penta) -- During the BATS test with the process water from the
Low Penta Soil Washer test, the weighted average influent concentration of penta
was 15 ppm and the weighted average effluent concentration was 1.4 ppm. The
calculated % removal is then 91%. These weighted values for the input or output
were calculated by (1) multiplying each concentration by the volume of water
during that sampling period to give a penta mass; (2) totalling all penta masses
over the course of the test, and (3) dividing the total mass of penta for all
input or output samples by the total volume of water treated (after excluded
certain samples, as noted in the tables) to give a single, weighted average
concentration in the influent or the output from the BATS.
During the second test, using process water from the High Penta SW test,
the weighted average influent concentration of penta was 44 ppm and the weighted
average effluent concentration was 3 ppm, corresponding to a 94% penta removal.
The data and mass balance calculations are presented in Tables 4-31 and
4-32 for the Low Penta and the High Penta tests, respectively.
The data indicate that "steady state" operation had not been achieved
when sampling was initiated, and, in the case of the test with the High Penta
process water, it required approximately 100 hours to stabilize. Nevertheless,
it is evident that the BioTrol Aqueous Treatment System is effective at removing
pentachlorophenol from the SW process water (Figures 4-19 arid 4-20) once full
acclimation and stabilization has been achieved.
Polynuclear Aromatic Hydrocarbons (PAHs) - - Concentrations of the various PAHs
were lower than anticipated in both the BATS influent and effluent. During the
BATS test with the Low Penta process water, only one PAH, anthracene, was
detected with any frequency along with sporadic detections of acenaphthene,
fluorene, phenanthrene, fluoranthene and pyrene. Almost all of the values
reported were given a "J" qualifier which means the value is estimated because
the compound was detected at a value greater than the method detection limit but
below the quantitation limit. Analyses of effluent samples during this test
yielded all non-detects. Method detection limits ranged from 2-15 ppb and were
dependent upon the specific PAH compound. The data for anthracene are presented
in Table 4-33. Calculations of removal were performed using the detection limit
when non-detects occurred in the effluent and this yielded a minimum removal for
anthracene of 69%.
Even though penta concentration was higher during the second BATS test,
there were only sporadic detections of acenaphthylene, acenaphthene, fluorene,
phenanthrene, anthracene, fluoranthene, and pyrene in the influent and effluent
samples and they all were estimated values. There were insufficient detected
92
-------
TABLE 4-31. BATS - LOW PENTA TEST
PERFORMANCE AND MATERIAL INVENTORY - PENTACHLOROPHENOL
SAIC#
Influent Data
ATSIN011DL
ATSIN012DL
ATSIN013DL
ATSIN014DL
ATSIN021DL
ATSIN022DL
ATSIN023REDL
ATSIN024DL
ATSIN031DL
ATSIN032DL
ATSIN033DL
ATSIN034DL
ATSIN041DL
ATSIN042DL
AVG
STDDEV
WEIGHTED AVG:
Effluent Data
ATSEFF011DL
ATSEFF012DL
ATSEFF013DL
ATSEFF014DL
ATSEFF021DL
ATSEFF022DL
ATSEFF023
ATSEFF024
ATSEFF031
ATSEFF032
ATSEFF033
ATSEFF034
ATSEFF041
ATSEFF042
AVG
STDDEV
WEIGHTED AVG:
Penta
Cone.
(ug/L)
12000
11000
14000
16000
39000
11000
11000
16000
11000
13000
15000
8800
11000
19000
15400
7635
15000
8600
3900
2400
2400
2600
1900
1100
1300
1100
820
1100
850
910
910
1449
649
1300
Data Time from
Qualifiers Beg. of Test
(1) (hr)
# 6
# 12
18
25
32
38
c 45
51
57
64
71
b 77
83
b 89
TOTAL INFLUENT:
# 6
# 12
18
25
32
38
45
51
b 57
64
71
77
83
89
TOTAL EFFLUENT:
Volume during
Composite period
(L)
3719
3739
3431
4894
3722
3423
4509
3980
3900
3731
4696
3867
3774
4001
55390
3572
3803
3344
3441
3626
3341
4071
4016
3782
3810
4536
3804
3806
3837
52790
PENTA MASS REDUCTION (g):
% MASS REMOVAL OF PENTA:
Penta
Mass/Period
(g)
45
41
48
78
150
38
50
64
43
49
70
34
42
76
740
31
15
8.0
8.3
9.4
6.3
4.5
5.2
4.2
3.1
5.0
3.2
3.5
3.5
60
680
91
FILENAME: ATSLPCPF
93
-------
TABLE 4-31. (CONTINUED)
(1) - Data qualifiers used in subsequent data tables:
A - Analysis of an archive sample
B - Analyte found in associated method blank
C - Corrected analyte concentration
D - Diluted sample
E - Beyond calibration range
I - Interpolated or extrapolated analyte concentration
J - Below calibration range
L - Estimated detection limit
M - Mean analyte concentration
R - Replaced analyte concentration value
U - Undetected at specified concentration
W - Weighted average to replace missing data
X - Manually entered data
* = Statistically determined analyte concentration
+ - Concentration switched due to mislabeliing of sample bottles
# - Biomass acclimation data not used in calculations
a - 1 to 28 days late to extract
b - 29 to 56 days late to extract
c - 57 to 84 days late to extract
d - 85 and over days late to extract
e - 1 to 28 days late to analyze
f» 29 to 56 days late to analyze
g - 57 to 84 days late to analyze
h - 85 and over days late to analyze
94
-------
TABLE 4-32. BATS - HIGH PENTA TEST
PERFORMANCE AND MATERIAL INVENTORY - PENTACHLOROPHENOL
SAIC0
Influent Data
ATSIN043DL
ATSIN044DL
ATSIN051DL
ATSINOS2DL
ATSIN053DL
ATSINOS40L
ATSIN061
ATSIN062DL
ATSIN063
ATSIN064
ATSIN071
ATSIN072DL
ATSIN073DL
ATSIN074DL
ATSIN081DL
ATSIN082DL
ATSIN083DL
ATSIN084DL
ATSIN091DL
ATSIN092
ATSIN093DL
ATSIN094DL
ATSIN101DL
AVG
STDDEV
WEIGHTED AVG:
Effluent Data
ATSEFF043DL
ATSEFF044DL
ATSEFFOS1DL
ATSEFF052DL
ATSEFF053DL
ATSEFF054DL
ATSEFF061DL
ATSEFF062
ATSEFF063
ATSEFF064
ATSEFF071
ATSEFF072
ATSEFF073
ATSEFF074
ATSEFF081DL
ATSEFF082
ATSEFF083DL
ATSEFF084
ATSEFF091DL
ATSEFF092DL
ATSEFF093DL
ATSEFF094DL
ATSEFF101
AVG
STDDEV
WEIGHTED AVG:
Penta Data
Cone. Qualifier*
(ug/L) (1)
48000 #
83000 #
49000 h
80000
39000 M,h
38000
44000 W
38000 g
50000 +,b,f
44000 W
31000 +,f
38000 g
47000 X
32000 h
37000 M,b,h
43000 M.b.e
54000 b,e
42000 b.e
50000 b
44000 W
38000 b.f
39000 b,e
28000 b.g
43095
10469
44000
9700 #
8300 #
3100 h
3200 h
3000 M,n
3200
2900 g
940 h
2600 +,g
2700 W
900 +,f
790
560 f
1400 h
2300 M.h
3800 M,b,e
7500 b,e
2700 W
2100 b
2900 b.g
2000 b.f
2300 b.e
2700 W
2552
1416
3000
Time from Volume during
Beg. of Test Composite period
(hr)
96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
,
TOTAL INFLUENT:
96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL EFFLUENT:
(L)
4461
3930
3757
4281
4670
4156
4938
3174
4533
3910
3416
3117
3404
3168
4258
4717
4687
4817
4624
4815
4552
3513
1263
92160
4315
3949
3786
4204
4493
4014
4812
3146
4478
4131
3395
3161
2773
3488
4090
4560
4655
4849
4486
5092
4251
3613
1158
90900
PENTA MASS REDUCTION (g):
% MASS REMOVAL OF PENTA:
Ponta
Mass/Period
(g)
210
330
180
340
180
160
220
120
230
170
106
120
160
100
160
200
250
200
230
210
170
140
35
3700
42
33
12
13
13
13
14
3.0
12
11
3.1
2.5
1.6
4.9
9.4
17
35
13
9.4
15
8.5
8.3
3.1
220
3480
94
FILENAME: ATSHPCPF
(1) A complete explanation of data qualifiers is provided in Table 4-27
95
-------
Figure 4-19. BATS-Penta Concentration in Low Penta Test
40
Gone, of Pentachlorophenol x 1000 (ug/L)
35 +
30
25 +
20
15
10
Biomass
Acclimation
Period
-B-influent ->K-Effluent
/r\ /IN
0
10 20 3.0 40 • 50 60 70
Time from Beginning of the Test (hrs)
80
90
filename; PCPLCONC
-------
Figure 4-20, BATS-Penta
Concentration in High Penta Test
Cone, of Pentachlorophenol x 1000 (ug/L)
80-
70 —
60-
50 +
40
30 +
.20
10 +
0
3iomass
climati
Period
-H-Influent -*-Effluent
90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240
Time from Beginning of the Test (hrs)
filename: PCPHCONC
-------
TABLE 4-33. BATS - LOW PENTA TEST
PERFORMANCE AND MATERIAL INVENTORY - ANTHRACENE
SAIC # Anthracene
Data Time from Volume during
Cone. Qualifiers Beg. of Test Composite period
Influent data
ATSIN011
ATSIN012
ATSIN013
ATSIN014
ATSIN021
ATSIN022
ATSIN023
ATSIN024
ATSIN031
ATSIN032
ATSIN033
ATSIN034
ATSIN041
ATSIN042
AVG
STDDEV
WEIGHTED AVG:
Effluent data
ATSEFF011
ATSEFF012
ATSEFF013
ATSEFF014
ATSEFF021
ATSEFF022
ATSEFF023
ATSEFF024
ATSEFF031
ATSEFF032
ATSEFF033
ATSEFF034
ATSEFF041
ATSEFF042
AVG
STDDEV
WEIGHTED AVG:
(ug/L)
69
41
- 45
35
32
23
43
38
34
32
22
7-4
7.4
.7.4
31
16
31
10
10
10
10
10
4
10
10
10
10
10
10
10
10
9.6
1.5
9.6
MINIMUM
(1) (hr)
6
J 12
J 18
J 25
J 32
J,X 38
45
J 51
J 57
J 64 :
J 71
*,b 77
83
*,b 89
TOTAL INFLUENT:
L 6
L 12
L 18
L 25
L 32
L 38
L 45
L 51
L,b 57
L 64
L 71
L 77
L 83
L.h 89
TOTAL EFFLUENT:
(L)
3719
3739
3431
4894
3722
3423
4509
3980
3900
3731
4696
3867
3774
4001
55390
3572
3803
3344
3441
3626
3341
4071
4016
3782
3810
4536
3804
3806
3837
52790
ANTHRACENE MASS REDUCTION (g):
% MASS REMOVAL OF ANTHRACENE:
Anthracene
Mass/Period
(g)
0.26
0.15
0.15
0.17
0.12
0.08
0.19
0.15
0.13
0.12
0.10
0.03
0.03
0.03
1.7
0.04
0.04
0,03
0.03
0.04
0.01
0.04
. 0.04
0.04
0.04
0.05
0.04
0.04
0.04
0.51
1.2
69-100
FILE NAME: ATSLANTF
(1) A complete explanation of data qualifiers is provided in Table 4-27
98
-------
values to calculate percent removal for any PAHs in this test. Method detection
limits ranged from 1-400 ppb.
The lack of actual analytical results above detection for both influent
and effluent makes it impossible to assess the removal of PAHs across the system.
The fact that all of the effluent samples analyzed from the first test (Low
Penta) were reported as non-detects while some PAHs (e.g., anthracene) were
measured in the influent suggests that some removal of these compounds was
achieved; however, this removal can not be quantified.
Copper. Chromium, and Arsenic (CCA) -- Samples were analyzed for CCA to determine
the fate of the metals in the BATS. The CCA data and material inventory for the
tests v;ith the process water from the Low Penta and High Penta SW tests are
presented in Tables 4-34 and 4-35. There may be an upward trend in the
concentration of arsenic and chromium in both the influent and the effluent
samples of the BATS, probably reflecting a build-up of solubilized metals in the
aqueous sample stream as it was recycled through the soil washer system.
While the system (SW or BATS) as tested is not designed to remove metals,
the results indicate a 48% removal of copper mass and lower removals of chromium
and arsenic. These removals, when none is expected, may suggest that there was
some accumulation of metals inside the reactor. In a commercial scale system
operated with full recycle of all process water for several months, metals could
possibly build up to concentrations toxic to the microorganisms. Some discharge
of water or treatment of metals in the system might then be required before
continued recycle of process water.
4.4.3.2 Other Analyses
Chloride Ion/Total Organic Halides -- Laboratory studies (Saber and Crawford,
1985; Crawford and Mohn, 1985) indicate that nearly complete mineralization
occurs during biodegradation of penta with the Flavobacterium inoculated into the
BATS. The penta is not simply converted to other organic compounds. Although
none of the potential metabolites of biodegradation of penta were sought during
the demonstration study (e.g., polychloroquinones, polychloro-dicarboxylic
acids), samples were analyzed for chloride ion and total organic halides to
ascertain whether the pentachlorophenol was being completely mineralized or only
partially degraded.
The data for chloride arid total organic halides (TOX) are presented in
Table 4-36 along with the corresponding pentachlorophenol data and the calculated
values for the theoretical chloride or TOX changes. For the first four samples,
Cl ion concentrations increased from influent to effluent as expected, but for
the last two samples Cl ion concentrations decreased while pentachlorophenol was
significantly decreasing from influent to effluent.
Using the change in penta concentrations for each pair of influent and
effluent samples, the increase in Cl ion concentration expected in the effluent
samples was calculated on the basis of complete mineralization. Every mole of
penta (266.5 gm) should produce 5 moles (177.5 gm) of chloride; this is
equivalent to 0.67 mg/L of chloride produced for every mg/L of penta degraded.
A correction was added to account for the chloride concentration in the influent.
99
-------
o
o
TABLE 4-34.
BATS - LOW PENTA TEST — MATERIAL INVENTORY - ARSENIC. CHROMIUM AND COPPER
SAIC# Arsenic Q Chromium Q Copper
(ug/L) (1) (ug/L) (1) (ug/L)
Influent data
ATSIN011
ATSIN012
ATSIN013
ATSIN014
ATSIN021
ATSIN022
ATSIN023
ATSIN024
ATSIN031
ATSIN032
ATSIN033
ATSIN034
ATSIN041
ATSIN042
AVG
STD DEV
WEIGHTED AVG:
Effluent data
ATSEFF011
ATSEFF012
ATSEFF013
ATSEFF014
ATSEFF021
ATSEFF022
ATSEFF023
ATSEFF024
ATSEFF031
ATSEFF032
ATSEFF033
ATSEFF034
ATSEFF041
ATSEFF042
AVG
STD DEV
WEIGHTED AVG:
17.6
17.6
17.0
19.7
13.1
12.6
14.0
13.0
12.9
12.1
18.1
9.9
18.7
24.5
15.8
3.8
15.9
15.8
17.2
18.0
21.0
14.2
12.6
14.1
13.8
12.7
12.2
17.7
17.4
16.3
16.3
15.7
2.4
15.7
6.80 •
6.80 "
13.0
6.80 *
6.80 '
6.80 '
10.1
6.80 "
6.80 *
6.80 '
19.1
13.0
18.3
19.5
10.5
4.9
10.6
5.92 *
5.92 *
5.92 *
5.92 *
5.92 "
5.92 '
5.92 *
5.92 *
5.92 '
5.92 "
16.3
5.92 *
13.9
12.6
7.7
3.5
7.9
35.1
33.3
34.6
26.2
23.6
24.9
23.7
25.9
23.1
22.9
19.8
27.6
17.4
19.0
25.5
5.4
25.3
8.61
15.1
16.7
18.8
12.3
13.4
14.5
17.9
13.5
13.1
8.61
13.8
8.61
8.61
13.1
3.3
13.0
Q Time from Volume during Arsenic Chromium Copper
Beg. of Test Composite period Mass/Period Mass/Period Mass/Period
(D (hr) (L) (mg) (mg) (mg)
6
12
18
25
32
38
45
51
57
64
71
77
83
89
TOTAL INFLUENT:
6
12
18
25
31
37
45
51
57
63
71
77
83
89
TOTAL EFFLUENT:
3719
3739
3431
4894
3722
3423
4509
3980
3900
3731
4696
3867
3774
4001
55390
3572
3803
3344
3441
3626
3341
4071
4016
3782
3810
4536
3804
3806
3837
52790
MATERIAL LOSS (mg):
% MASS REMOVAL:
65.5
65.8
58.3
96.4
48.8
43.1
63.1
51.7
50.3
45.1
85.0
38.3
70.6
98.0
880
56.4
65.4
60.2
72.3
51.5
42.1
57.4
55.4
48.0
46.5
80.3
66.2
62.0
62.5
826
54
1.5
25.3
25.4
44.6
33.3
25.3
23.3
45.5
27.1
26.5
25.4
89.7
50.3
69.1
78.0
589
21.1
22.5
19.8
20.4
21.5
19.8
24.1
23.8
22.4
22.6
73.9
22.5
52.9
48.3
416
173
26
131
125
119
128
87.8
85.2
107
103
90.1
85.4
93.0
107
65.7
76.0
1400
30.8
57.4
55.9
64.7
44.6
44.8
59.0
71.9
61.1
49.9
39.1
52.5
32.8
33.0
687
713
48
FILENAME: ATSLCCAF
(1) A complete explanation of data qualifiers is provided in Table 4-27
-------
TABLE 4-35. BATS - HIGH PENTA TEST -- MATERIAL INVENTORY - ARSENIC, CHROMIUM AND COPPER
SAIC # Arsenic Q
(ug/L) (1)
Influent data
ATSIN043
ATSIN044
ATSIN051
ATSIN052
ATSIN053
ATSIN054
ATSIN061
ATSIN062
ATSIN063
ATSIN064 .
ATSIN071
ATSIN072
ATSIN073
ATSIN074
ATSIN081
ATSIN082
ATSIN083
ATSIN084
ATSIN091
ATSIN092
ATSIN093
ATSIN094
ATSIN101
AVG
STDDEV
WEIGHTED AVG:
57.7
59.9
59.8
57.9
59.3
76.9
90.8
52.4
56.3
67.1
59.7
65.5
60.9
71.3
57.3
69.5
79.9
40.3
75.1
65.7
76.1
87.0
73.2
66.1
11.4
66.0
Chromium Q
(ug/L) (1)
13.2
18.5
10.5
10.4
12.1
27.7
17.0
8.72 "
11.1
10.2
8.72 *
23.0
19.6
16.6
18.5
18.8
17.6
16.1
17.2
19.6
19.6
20.4
17.3
16.2
4.8
16.2
Copper Q Time from Volume during Arsenic Chromium Copper
Beg. of Test Composite period Mass/Period Mass/Period Mass/Period
(ug/L) (1) (hr) (L) (mg) (mg) (mg)
32.3
26.7
28.6
32.6
33.4
24.7
26.2
32.5
27.1
22.3
20.6
45.1
43.6
39.4
43.8
47.0
57.5
47.6
55.3
57.4
61.2
68.8
65.0
40.8
14.3
40.4
96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL INFLUENT:
4461
3930
3757
4281
4670
4156
4938
3174
4533
3910
3416
3117
3404
3168
4258
4717
4687
4817
4624
4815
4552
3513
1263
92160
257
235
225
248
277
320
448
166
255
262
204
204
207
226
244
328
374
194
347
316
346
306
92.5
6080
58.9
72.7
39.5
44.5
56.5
115
84.0
27.7
50.3
39.9
29.8
71.7
66.7
52.6
78.8
88.7
82.5
77.5
79.5
94.4
89.2
71.7
21.9
1490
144
105
107
140
156
103
129
103
123
, 87.2
70.4
141
148
125
187
222
269
229
256
276
279
242
82.1
3720
-------
TABLE 4-35. BATS - HIGH PENTA TEST — MATERIAL INVENTORY - ARSENIC. CHROMIUM AND COPPER (CONTINUED)
O
t-0
Chromium Q
Copper
Q Time from Volume during
Arsenic Chromium
Copper
Beg. of Test Composite period Mass/Period Mass/Period Mass/Period
Effluent data
ATSEFF043
ATSEFF044
ATSEFF051
ATSEFF052
ATSEFF053
ATSEFF054
ATSEFF061
ATSEFF062
ATSEFF063
ATSEFF064
ATSEFF071
ATSEFF072
ATSEFF073
ATSEFF074
ATSEFF081
ATSEFF082
ATSEFF083
ATSEFF084
ATSEFF091
ATSEFF092
ATSEFF093
ATSEFF094
ATSEFF101
AVG
STD DEV
WEIGHTED AVG:
EMI C MAMC. ATCL
(ug/L) (1)
41.6
46.4
48.9
50.8
55.4
74.4
56.7
46.0
59.7
67.2
62.0
49.6
65.3
69.8
57.4
45.6
5.7
68.0
56.9
55.9
57.9
61.0
57.7
54.8
13.3
54.3
r»r»Arr '
(ug/L) (1)
8.79 '
12.9
8.79 *
8.79 *
8.79 *
14.8
17.3
8.79 *
14.9
18.3
18.7
11.5
11.0
11.8
13.2
10.7
11.3
11.3
11.7
11.2
10.9
14.4
12.8
12.3
2.9
12.3
(ug/L)
14.8
13.3
17.7
17.7
22.5
11.4
11.6
21.1
32.8
39.5
41.7
25.5
23.9
20.9
32.3
26.6
24.3
33.8
26.1
20.8
38.5
46.7
35.1
26.0
9.7
25.5
(D (hr)
- 96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL EFFLUENT:
MATERIAL LOSS (mg):
% MASS REMOVAL
(L)
4315
3949
3786
4204
4493
4014
4812
3146
4478
4131
3395
3161
2773
3488
4090
4560
4655
4849
4486
5092
4251
3613
1158
90900
(mg)
179
183
185
214
249
299
273
145
267
278
210
157
181
243
235
208
26.5
330
255
285
246
220
66.8
4940
1140
18
(mg)
37.9
50.9 .
33.3
37.0
39.5
59.4
83.2
27.7
66.7
75.6
63.5
36.3
30.5
41.2
54.0
48.8
52.6
54.B
52.5
57.0
46.3
52.0
14.8
1120
370
24
(mg)
63.9
52.5
67.0
74.4
101
45.8
55.8
66.4
147
163
142
80.6
66.3
72.9
132
121
113
164
117
106
164
169
40.7
2320
1400
37
(1) A complete explanation of data qualifiers is provided in Table 4-27
-------
TABLE 4-36. ATS PENTA MINERALIZATION ASSESSMENT
Influent
ATSIN
042
044
052
054
062
064
Penta
(mg/L)
19.0
83.0
80.0
38.0
38.0
44.0
Cl
(mg/L)
33.9
54.3
50.0
55.0
116.0
106.0
TOX
(mg/L)
7.7
17.0
19.0
15.0
11.0
2.4
Effluent
ATSEFF
042
044
052
054
062
064
Penta
(mg/L)
0.9
8.3
3.2
3.2
0.9
2.7
Cl
(mg/L)
43.4
83.7
82.8
87.5
100.0
58.4
TOX
(mg/L)
8.8
13.0
11.0
7.8
15.0
9.0
Change in Cl
found
9.5
29.4
32.8
32.5
-16.0
-47.6
calc*
12.1
49.8
51.2
23.2
24.7
27.5
Change in TOX
found
1.1
-4.0
-8.0
-7.2
4.0
6.6
calc*
12.1
49.8
51.2
23.2
24.7
27.5
o
W
FILENAME: ATSCLTOX
* Every mg/L of penta is equivalent to 0.667 mg/L of Cl- and TOX; every mg/L of penta removed should increase Cl- and
decrease TOX by 0.667 mg/L
-------
Actual and calculated chloride concentrations in the effluent are not in good
agreement and the effluent chloride values vary widely. Thus, this data can not
be used to confirm that the decrease in penta concentration was due to
mineralization of the penta.
Examination of the data for total organic halides (also shown in Table
4-36) indicates that only three of the samples show a decrease in concentration
of TOX between influent and effluent, the other three samples show an increase.
Consequently, no conclusions can be drawn from this data as to whether the
chloride leaving the system is purely inorganic.
Total Solids -- Analyses for total solids (suspended and dissolved) also were
performed on the influent and effluent samples of the BioTrol Aqueous Treatment
System during each study. Results for Total Solids along with material balances
are presented in Tables 4-37 and 4-38. The material balance data show only a.n
8.6% loss of solids during the Low Penta test and a 4% gain in solids during the
High Penta test. It is possible that there was a slight build-up of solids
during the Low Penta test that was later released during the High Penta test.
Alternatively, the slight differences are also well within the precision of the
Total Residue (TR) test.
Total Recoverable Petroleum Hydrocarbons - - The TRPH analyses were intended as
a measure of the oil (used as a carrier for penta during wood treating, operations
at.MacGillis & Gibbs) in the soil and consequently expected in the process water
leaving the Soil Washer and entering the BATS. Results for TRPHs are presented
in Tables 4-39 and 4-40. The TRPH values were in the low ppm range in all of the
aqueous process streams, indicating that there was not a significant oil content
in the feed soils or that the oil did not transfer to the aqueous process stream
during soil washing. Removals of TRPHs were only 18% and 42% in the Low Penta and
High Penta studies, respectively. If significant amounts of oil had been present
and there were reason to suspect that it were adversely affecting the BSWS, a
means of oil removal such as an oil water separator would need to be added as a
pretreatment step before the BATS.
Chemical Oxygen Demand (COD) - - Chemical Oxygen Demand was examined as a measure
of the total potential oxidizable material in the samples. COD was included
rather than biochemical oxygen demand (BOD) because of its simplicity and because
it avoids BOD test problems concerning sample toxicity, refractory contaminants,
and the need to develop an influent sample inoculation procedure. Since the BSWS
includes biological treatment, measurements of COD decrease reflect overall
reduction of organic contaminants including those being analyzed for as primary
and secondary contaminants.
Measurements of COD confirmed that oxidation was occurring, albeit not
to a very great extent. The data were very erratic (Tables 4-41 and 4-42) and the
decrease in COD from influent to effluent is not very pronounced (14% and 47% in
either test, respectively). One possible explanation is that the primary
contaminants, penta and PAHs, only constitute a small fraction of the total
organic contamination in the system.
104
-------
TABLE 4-37. BATS — LOW PENTA TEST MATERIAL INVENTORY - TOTAL SOLIDS
SAIC#
Influent Data
ATSIN01 1
ATSIN012
ATSIN013
ATSIN014
ATSIN021
ATSIN022
ATSIN023
ATSIN024
ATSIN031
ATSIN032
ATSIN033
ATSIN034
ATSIN041
ATSIN042
AVG
STD DEV
WEIGHTED AVG:
Effluent Data
ATSEFF01 1
ATSEFF012
ATSEFF013 ,
ATSEFF014
ATSEFF021
ATSEFF022
ATSEFF023
ATSEFF024
ATSEFF031
ATSEFF032
ATSEFF033
ATSEFF034
ATSEFF041
ATSEFF042
AVG
STD DEV
WEIGHTED AVG:
Total
Solids
(mg/L)
539
545
527
535
501
502
491
680
614
683
472
500
547
547
549
63
547
462
524
473
515
432
463
446
622
590
658
525
568
525
525
523
65
525
Data Time from
Qualifiers Beg. of Test
(1) (hr)
6
12
18
25
32
38
45
51
57
64
71
77
W 83
W 89
TOTAL INFLUENT:
A 6
12
18
25
32
38
45
51
57
64
W 71
77
W 83
W 89
TOTAL EFFLUENT:
Volume during
Composite period
(L)
3719
3739
3431
4894
3722
3423
4509
3980
3900
3731
4696
3867
3774
4001
55390
3572
3803
3344
3441
3626
3341
4071
4016
3782
3810
4536
3804
3806
3837
52790
MATERIAL LOSS (kg):
% MATERIAL LOSS:
Total
Solids/Period
(kg)
2.00
2.04
1.81
2.62
1.86
1.72
2.21
2.7
2.39
2.55
2.22
2
2.07
2.19
30.3
1.65
1.99
1.58
1.77
1.57
1.55
1.82
2.50
2.23
2.51
2.38
2.16
2.00
2.01
27.7
2.6
8.5
FILENAME: ATSLTRF
(1) A complete explanation of data qualifiers is provided in Table 4-27
105
-------
TABLE 4-38. BATS - HIGH PENTA TEST MATERIAL INVENTORY - TOTAL SOLIDS
SAIC*
Influent Data
ATSIN043
ATSIN044
ATSIN051
ATSIN052
ATSIN053
ATSIN054
ATSIN061
ATSIN062
ATSIN063
ATSIN064
ATSIN071
ATSIN072
ATSIN073
ATSIN074
ATSIN081
ATSIN082
ATSIN083
ATSIN084
ATSIN091
ATSIN092
ATSIN093
ATSIN094
ATSIN101
AVG
STDDEV
WEIGHTED AVG:
Effluent Data
ATSEFF043
ATSEFF044
ATSEFF051
ATSEFF052
ATSEFF053
ATSEFFOS4
ATSEFF061
ATSEFF062
ATSEFF063
ATSEFF064
ATSEFF071
ATSEFF072
ATSEFF073
ATSEFF074
ATSEFF081
ATSEFF082
ATSEFF083
ATSEFF084
ATSEFF091
ATSEFF092
ATSEFF083
ATSEFF094
ATSEFF101
AVG
STDDEV
WEIGHTED AVG:
Total
Data Time from
Solids Qualifiers Beg. of Test
(mg/L)
507 A
488 W
273 A
303 A
522
337
489 W
546 A
503
359
451
563
536 A
545 A
497 A
525 W.M
489 W
510 A
715 A
570
489 W
530
593 A
493
96
491
517 W
514
356 A
546 M
408 A
875
517 W
528 A
522 A
517
517 W
566
528
542 A
363 A
536 A
486 A
331 A
517 W
517 W
590
620
647 A
524
107
517
(1) (hr)
96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL INFLUENT:
96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL EFFLUENT:
Volume during
Composite period
(L)
4461
3930
3757
4281
4670
4156
4938
3174
4533
3910
3416
3117
3404
3168
4258
4717
4687
4817
4624
4815
4552
3513
1263
92160
4315
3949
3786
4204
4493
4014
4812
3146
4478
4131
3395
3161
2773
3488
4090
4560
4655
4849
4486
5092
4251
3613
1158
90900
MATERIAL GAIN (kg):
% MATERIAL GAIN:
Total
Solids/Period
(kg)
2.26
1.92
1.03
1.30
2.44
1.40
2.42
1.73
2.28
1.40
1.54
1.75
1.82
1.73
2.12
2.48
2.29
2.46
3.31
2.74
2.23
1.86
0.75
45
2.23
2.03
1.35
2.30
1.83
3.51
2.49
1.66
2.34
2.13
1.76
1.79
1.46
1.89
1.48
2.44
2.26
1.60
2.32
2.63
2.51
2.24
0.75
47
1.8
4
(1) A complete explanation of data qualifiers ie provided in Table 4-27
106
-------
TABLE 4-39. BATS - LOW PENTA TEST - MATERIAL
INVENTORY -TOTAL RECOVERABLE PETROLEUM
HYDROCARBONS (TRPH)
SAIC #
Influent Data
ATSIN011
ATSIN012
ATSIN013
ATSIN014
ATS1N021
ATSIN022
ATSIN023
ATSIN024
ATSIN031
ATSIN032
ATSIN033
ATSIN034
ATSIN041
ATSIN042
AVG
STD DEV
WEIGHTED AVG:
Effluent Data
ATSEFF011
ATSEFF012
ATSEFF013
ATSEFF014
ATSEFF021
ATSEFF022
ATSEFF023
ATSEFF024
ATSEFF031
ATSEFF032
ATSEFF033
ATSEFF034
ATSEFF041
ATSEFF042
AVG
STD DEV
WEIGHTED AVG:
TRPH
(mg/L)
4.82 J
6.98 J
9.79 J
8.38
2.96
5.00
4.57 J
5.47 J
6.80 J
5.00 J
4.23 J
6.62 J
5.05 J
3.37 J
5.65
1.82
5.64
4.01
3.16 J
3.92 J
3.65
5.30
4.30
9.45 J
5.01 J
11.00 J
4.95 J
3.47 J
4.69 J
2.67 J
1.85 *
4.82
2.41
4.84
Data Time from
Qualifiers Beg. of Test
(1) (hr)
6
12
18
25
32
38
45
51
57
64
71
77
83
89
TOTAL INFLUENT:
6
12
18
26
32
38
45
51
57
64
71
77
83
89
TOTAL EFFLUENT:
Volume during
Composite period
(L)
3719
3739
3431
4894
3722
3423
4509
3980
3900
3731
4696
3867
3774
4001
55390
3572
3803
3344
3441
3626
3341
4071
4016
3782
3810
4536
3804.
3806
3837
52790
MATERIAL LOSS (g):
% MASS REMOVAL:
TRPH/
Period
(g)
17.9
26.1
33.6
41.0
11.0
17.1
20.6
21.8
26.5
18.7
19.9
25.6
19.1
13.5
312
14.3
12.0
13.1
12.6
19.2
14.4
38.5
20.1
41.6
18.9
15.7
17.8
10.2
7.1
255
56.8
18
FILENAME: ATSLTPHF
(1) A complete explanation of data qualifiers is provided in Table 4-27
107
-------
TABLE 4-40. BATS - HIGH PENTA TEST— MATERIAL INVENTORY
TOTAL RECOVERABLE PETROLEUM HYDROCARBONS (TRPH)
SAIC#
Influent Data
ATSIN043
ATSIN044
ATSIN051
ATSIN052
ATSIN053
ATSIN054
ATSIN061
ATSIN062
ATSIN063
ATSIN064
ATSIN071
ATSIN072
ATSIN073
ATSIN074
ATSIN081
ATSIN082
ATSIN083
ATSIN084
ATSIN091
ATSIN092
ATSIN093
ATSIN094
ATSIN101
AVG
STDDEV
WEIGHTED AVG:
Effluent Data
ATSEFF043
ATSEFF044
ATSEFF051
ATSEFF052
ATSEFF053
ATSEFF054
ATSEFF061
ATSEFF062
ATSEFFOS3
ATSEFF064
ATSEFF071
ATSEFF072
ATSEFF073
ATSEFF074
ATSEFF081
ATSEFF082
ATSEFF083
ATSEFF084
ATSEFF091
ATSEFF092
ATSEFF093
ATSEFF094
ATSEFF101
AVG
STDDEV
WEIGHTED AVG:
Data Time from
TRPH Qualifiers Beg. of Test
(mg/L)
6.48 J
1.35 *
2.39 J
1.46 *
2.53 M,J
6.42 J
3.11 J
7.63 J
1.39 *
1.35 *
1.54 "
3.96 J
1.39 *
9.08 J
4.22 J
4.10 J
2.93 J
3.87 J
2.84 J
2.74 J
2.87 J
6.55 J
7.19 J
3.80
2.26
3.60
4.85 J
2.06 J
2.14 J
1.09 '
1.04 M.*
1.05 *
1.05 '
1.03 '
1.02 *
2.13 J
2.87 J
2.63 J
10.20
1.07 *
4.09 J
1.09 *
1.10 "
1.09 "
1.07 *
1.02 *
2.09 W
3.41 J
3.93 J
2.31
2.03
2.09
(1) (hr)
96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL INFLUENT:
96
102
108
115
121
128
135
140
146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL EFFLUENT:
Volume during
Composite period
(L)
4461
3930
3757
4281
4670
4156
4938
3174
4533
3910
3416
3117
3404
3168
4258
4717
4687
4817
4624
4815
4552
3513
1263
92160
4315
3949
3786
4204
4493
4014
4812
3146
4478
4131
3395
3161
2773
3488
4090
4560
4655
4849
4486
5092
4251
3613
1158
90900
MATERIAL LOSS (g):
% MASS REMOVAL:
TRPH/
Period
(3)
28.9
5.32
8.98
6.23
11.8
26.7
15.4
24.2
6.29
5.29
5.25
12.3
4.72
28.8
18.0
19.3
13.7
18.6
13.1
13.2
13.1
23.0
9.08
331
20.9
8.14
8.10
4.59
4.67
4.23
5.07
3.23
4.56
8.80
9.74
8.3
28.29
3.7
16.7
5.0
5.1
5.3
4.8
5.2
8.9
12.3
4.55
190
141
42
FILENAME: ATSHTPHF
(1) A complete explanation of data qualifiers is provided in Table 4-27
108
-------
TABLE 4-41. BATS - LOW PENTA TEST — MATERIAL
INVENTORY - CHEMICAL OXYGEN DEMAND (COD)
SAIC#
Influent Data
ATSIN01 1
ATSIN012
ATSIN013
ATSIN014
ATSIN021
ATSIN022
ATSIN023
ATSIN024
ATSIN031
ATSIN032
ATSIN033
ATSIN034
ATSIN041
ATSIN042
AVERAGE
STDDEV
WEIGHTED AVG:
Effluent Data
ATSEFF011
ATSEFF012
ATSEFF013
ATSEFF014
ATSEFF021
ATSEFF022
ATSEFF023
ATSEFF024
ATSEFF031
ATSEFF032
ATSEFF033
ATSEFF034
ATSEFF041
ATSEFF042
AVERAGE
STD DEV
WEIGHTED AVG:
COD
(mg/L)
220
180
190
250
1100
140
170
2200
170
1300
2100
160
170
190
610
722
630
950
130
140
870
710
100
1100
1800
110
120
1200
92
270
150
553
531
570
Data Time from Volume
during
Qualifiers Beg. of Test Composite period
(D (hr)
6
12
18
26
32
38
45
51
57
64
71
77
83
89
TOTAL INFLUENT:
6
12
18
25
32
38
45
51
57
64
71
77
83
89
TOTAL EFFLUENT:
MATERIAL LOSS (kg):
% MASS REMOVAL:
(L)
3719
3739
3431
4894
3722
3423
4509
3980
3900
3731
4696
3867
3774
4001
55390
3572
3803
3344
3441
3626
3341
4071
4016
3782
3810
4536
3804
3806
3837
52790
COD/
Period
(kg)
0.82
0.67
0.65
1.2
4.1
0.48
0.77
8.8
0.66
4.9
9.9
0.62
0.64
0.76
35
3.4
0.49
0.47
3.0
2.6
0.33
4.5
7.2
0.42
0.46
5.4
0.35
1.0
0.58
30
5
13
FILENAME: ATSLCODF
(1) A complete explanation of data qualifiers is provided in Table 4-27
109
-------
TABLE 4-42. BATS - HISH PENTA TEST—MATERIALINVENTORY
CHEMICAL OXYGEN DEMAND (COD)
Data Time from Volum* during
SAICf
Influent Data
ATSIN043
ATSIN044
ATS1NOS1
ATSIN052
ATSIN053
ATSIN054
ATSIN061
ATSIN062
ATSIN063
ATSIN064
ATSIN071
ATSIN072
ATSIN073
ATSIN074
ATSIN081
ATSIN082
ATSIN083
ATSIN084
ATSIN091
ATSIN092
ATSIN093
ATSIN094
ATSIN101
AVG
STDDEV
WEIGHTED AVG:
Effluent Data
ATSEFF043
ATSEFF044
ATSEFFOS1
ATSEFF052
ATSEFF053
ATSEFF054
ATSEFF061
ATSEFF062
ATSEFF063
ATSEFF064
ATSEFF071
ATSEFF072
ATSEFF073
ATSEFF074
ATSEFF081
ATSEFF082
ATSEFF083
ATSEFF084
ATSEFF091
ATSEFF092
ATSEFF093
ATSEFF094
ATSEFF101
AVG
STDDEV
WEIGHTED AVG:
COD
(mo/14
710
190
200
180
485
320
220
130
110
530
120
240
360
430
190
460
280
240
380
1000
380
240
310
335
202
350
120
170
150
160
115
140
120
110
184
130
300
420
320
110
140
270
280
160
260
120
170
210
120
186
81
180
Qufliifiofv Bog of Tort Cofnooi
(1) (nr)
98
102
108
115
M 121
128
135
140
146
153
159
165
172
179
185
19.1
198
204
210
217
223
229
237
TOTAL INFLUENT:
96
102
108
115
M 121
128
135
140
W 146
153
159
165
172
179
185
191
198
204
210
217
223
229
237
TOTAL EFFLUENT:
it0 DQftod
(L)
4461
3930
3757
4281
4670
4156
4938
3174
4533
3910
3418
3117
3404
3168
4258
4717
4687
4817
4624
4815
4552
3513
1263
92160
4315
3949
3786
4204
4493
4014
4812
3146
4478
4131
3395
3161
2773
3488
4090
4560
4655
4849
4486
5092
4251
3613
1158
90900
MATERIAL LOSS > 0.81
2.2
1.3
1.2
1.8
4.8
1.7
0.84
0.39
32
0.52
0.67
0.57
0.67
0.52
0.56
0.58
0.35
0.82
0.54
1.0
1.3
0.89
0.38
0.57
1.2
1.3
0.78
1.?
0.61
0.72
0.76
0.14
17
15
48
FILENAME: ATSHCODF
(1) A complete explanation of data qualifiers is proviaed in Table 4-27
110
-------
4.4.3.3 System Parameters
Field data collected over the course of both tests, including flow,
temperature, and pH, are summarized in Tables 4-43 and 4-44.
Influent and effluent flow rate data were collected at two hour intervals
over the course of the demonstration test. The measurements were in the form of
the depth of water in the water storage tanks as read from a level indicator
located outside each tank. Each depth measurement was then converted to a volume
measurement using the following equation for horizontal tanks (see Perry and
Green, 1984):
V - L-R2 [(B/57.3) -(sinB)(cosB)] B = cos^Cl-H/R) (radians)
V — volume of liquid (ft3)
R - tank radius (ft)
L = length of tank (ft)
H = depth of liquid (ft)
The change in volume was then calculated from two consecutive volume measurements
for the same storage tank. This change in volume divided by the time interval
over which the change occurred yields the flow rate.
The flow rates remained very steady for both the influent and the
effluent over the entire course of the test. The mean influent flow rate was
10.22+0.80 L/min and 10.35+2.64 L/min in the Low Penta and the High Penta test,
respectively. The mean effluent flow rate was 9.94+0.87 L/min and 10.13+2.58
L/min for the two tests, respectively. Any significant changes in flow rate were
due to equipment adjustments.
Grab samples for temperature measurements were taken 3 times per day and
measured with a previously-calibrated immersion thermometer. The average influent
temperatures were 16.5+2.0°C and 14.6+4.0°C and the average effluent temperatures
were 25.2+1.6°C and 24.7+1.8°C, in the two tests, respectively. The increase in
temperature from influent to effluent may be due to the biodegradation within
the reactor or solar heating of the trailer and the system, even though the study
was carried out in the Fall.
Measurements of pH were recorded from the on-line pH adjustment system
and from grab samples taken 3 times per day using a hand-held pH meter. The
vendor had specified a pH of approximately 7.3 as the ideal pH of the system and
measurements indicate that this was achievable through the on-line pH adjustment
system. The pH of the SW process water over the course of the test was in the
range of 6.64-8.03 standard units.
4.4.4 Slurry Bio-Reactor Performance
4.4.4.1 Critical Analyses
As noted earlier, a single, 14-day test of the SBR was carried out using
fine particle slurry taken during the High Penta Soil Washing test. Table 4-45
111
-------
I
I
I
I
TABLE4-43. BATS FIELD MEASUREMENTS-LOW PENTA TEST
I
I
I
I
I
I
I
I
I
I
I
I
Dal*
09ISB1K
09/26/8J
09/26/ss
09/27/8C
09/27/8!
09/27/8S
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89
09/27/89.
09/27/89
09/28/89
00/28/89
00/28/88
09/28/89
09/28/89
09/28/89
09/28/89
00/28/89
08/28/89
OS/28/89
08/28/89
09/28/89
09/28/89
09/28/89
09/28/89
09/28/89
09/28/89
09/28/89
09/28/89
09/28/89
09/28/89 !
09/28/89 :
09/29/89 (
09/29/89 C
09/29/89 C
09/29/89 C
09/29/89 C
09/29/89 0
Tlmi
I 22:1!
23:2!
01:2S
02O4
06:40
07:03
07:04
07:11
07:22
09:02
11:10
11:24
12:29
12:40
12:52
13:02
18:30
18:30
19:02
19:15
19:24
20:23
20:47
22:22
OO:30
00:56
02:25
02:30
02:35
03:04
04:23
06:24
06:29
08:26
08:40
10:60
13:23
13:56
15:47
16:55
18:16
20:47
22:00
22:41
22:65
22:59
0:29
0:36
2:25
2:29
4:15
4:34
TBTI* trocn
Btg.ofT*«
(hr)
) 3.S
3.4
'. 4.1
5.7
6.S
e.e
7.7
S.<
10.6
10.76
12.1
12.2
12.3
12.4
14.1
16.2
16.3
17.8
18.0
19.7
21.6
23.6
23.8
24-1
24.3
24.6
26.6
25.9
27.4
27.6
29.8
30.0
31.6
31.6
31.7
32.2
33.6
35.6
35.6
37.6
37.8
39.9
42.6
43.0
44.9
46.0
47.3
49.9
51.1
51.8
52.0
52.1
53.8
53.7
55.5
55.6
S7.3
57.7
t SAK5*
) BEQIN SAMPUN
L
ATSIN/EFF011
ATSIN012
ATSEFF012
ATSIN/EFF013
ATSIN/EFFOU
ATSIN/EFF021
ATSIN/EFF022
ATSIN/EFF023
ATSIN/EFF024
ATSIN/EFF031
Flow RUM
Inflinn
(Umln)
a
10.76
9.62
10.5C
9.71
9.67
9.52
9.11
12.17
11.10
9.77
9.99
9.35
9.19
9.15
8.26
10.99
9.63
11.31
11.18
10.86
9.92
10.66
Effluon
(Ultlftl)
10.2'
9.84
9.7C
10.11
9.70
9.87
8.38
6.34
10.10
9.57
9.61
9.42
8.85
9.26
10.22
11.70
10.68
10.37
9.85
Vohinw
t Influwi
(L)
3711
3739
3431
4894
3722
3423
4509
3980
3900
EfflUMl
(L)
3672
3803
3344
3441
3626
3341
4O71
4016
3782
(continued)
112
PH
1 On-Un
7.2-
7.31
71i
7.34
7.26
7.03
7.19
7.20
7.19
6.37
7.19
7.10
7.16
7.14
7.08
6.96
7.23
7.21
7.13
7.31
7.24
7.36
7.31
7.24
7.36
• lnflu»n
7.3
7.46
7.41
7.36
7.41
7.38
7.50
7.39
7.41
7.48
1 Efflucn
.
7.85
7.47
7.43
7.39
7.42
7.48
7.48
7.43
7.56
Tamptratur*
t lnftu«n
«3)
14.1
14.6
14.8
17.0
18.9
17.0
18.0
19.3
18.0
16.0
1 Effflnn
(C)
25.1
25.6
25.7
23.6
246
26.5
23.0
29.0
26.0
Tank Switch
t Influtnl
1
Efflumt
4
-------
TABLE 4-43. BATS RELD MEASUREMENTS -LOW PENTA TEST
Date
09/29/89
08/29/69
09/28/89
09/29/89
08/29/89
09/29/89
09/29/89
08/28/88
09/28/89
08/29/89
08/28/89
08/29/89
09/28/88
09/29/89
09/29/89
08/28/88
09/29/89
08/29/88
09/29/89
09/30/88
08/30/89
09/30/89
09/30/89
09/30/89
09/30/89
09/30/89
09/30/88
09/30/89
08/30/88
08/30/88
09/30/89
08/30/89
09/30/89
08/30/89
09/30/89
09/30/89
AVERAGE
STDDEV
Tim*
04:40
08:27
06:31
08:34
08:34
10:22
10:30
11:25
12:40
12:66
15:30
17:41
18:29
19:39
19:43
21:30
21:40
23:27
23:60
01:30
01:39
01:40
0230
02:35
03:28
03:31
06:38
05:40
)S:46
15:69
07:26
07:29
09:30
10:48
11:29
12:20
Time ffufii
8*0. of Teet
-------
TABLE4-44. BATS FIELD MEASUREMENTS-HIGH PENTATEST
Data
08/26/8!
OS/30/80
08/30/89
09/30/88
09/30/8S
09/30/89
09/30/89
09/30/80
09/30/89
09/30/89
09/30/89
09/30/89
09/30/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01789
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/01/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/02/89
10/03/89
10/03/89
10/03/89
10/03/89
10/03/89
Tims
18:65
11:29
13:32
16:3!
15:4C
17:28
18:66
19:01
19:35
21:33
22:47
23:28
23:38
01:05
01:30
03:29
04:37
04:44
04:46
05:30
07:12
07:29
09:36
11:30
13:15
13:40
14:16
16:30
15:35
17:21
19:27
20:21
22:00
2253
00:22
01:47
02:27
02:30
02:55
03:00
03:45
04:25
06:38
05:40
06:30
07:27
08:40
09:46
10:45
12:30
14:45
15:16
15:30
17:19
19:25
21:17
21:56
22:03
22:07
23:25
00:00
01:29
03:26
03:28
03:36
Time from
Bag. of Tea
(hr)
0.0
88.6
90.6
92.0
92.8
94.6
96.0
96.1
96.7
98.6
99.9
100.6
100.7
102.1
102.6
104.6
105.7
106.8
106.9
106.6
104.3
108.6
110.7
112.6
114.3
114.8
116.4
116.6
116.7
118.5
120.6
121.4
123.1
124.0
126.6
126.9
127.6
127.6
128.0
128.1
128.9
129.5
130.7
130.8
131.6
132.6
133.8
134.8
136.9
137.6
139.8
140.4
140.6
142.4
144.6
146.4
147.0
147.1
147.2
148.5
149.1
150.6
152.6
162.6
152.7
SAIC*
BEGIN SAMPL1NC
ATSEFF043
ATSIN043
ATSIN/EFF044
ATSIN/EFF051
ATSIN/EFF052
ATSIN/EFF053
ATSIN/EFF054
ATSIN/EFF061
ATSIN/EFF062
ATSIN/EFF063
ATSIN/EFF084
FtowRatee
Influent
(L/mln)
10.88
11.35
10.88
11.80
10.49
10.30
10.83
9.74
10.20
10.50
11.60
10.80
11.50
12.01
11.19
11.36
11.39
11.40
11.21
11.80
10.91
11.53
11.04
10.69
9.61
13.19
11.08
11.56
11.17
9.73
Effluent
(L/mln)
10.81
10.83
10.93
10.98
11.10
10.68
10.61
10.04
10.16
10.64
10.46
11.29
10.80
11.83
11.46
10.71
10.76
11.16
10.75
10.90
10.49
11.26
12.02
11.04
10.24
9.92
13.22
10.76
11.62
10.81
10.70
Volume
Influent
(L)
4461
3930
3757
4281
4670
4166
4938
3174
4533
3910
Effluent
(L)
4316
3949
3786
4204
4493
4014
4812
3148
4478
4131
pH
On-Une
PH
7.38
7.03
8.34
7.06
7.09
7.19
7.26
7.30
7.30
7.37
7.16
7.18
7.21
7.36
7.16
7.28
7.24
7.14
6.92
7.17
7.73
7.29
7.25
7.30
7.26
7.24
7.36
7.32
7.35
7.18
7.2S
7.36
Influent
6.89
7.26
6.64
7.26
7.50
9.19
7.19
7.03
6.94
6.77
6.60
7.32
7.03
9.07
7.28
7.00
Temperature
(O
22.0
20.0
19.0
19.3
19.0
17.5
16.6
13.0
(0
27.5
24.0
24.0
28.2
22.0
23.0
24.0
23.0
Influent
2
1
Effluent
3
4
3
(continued)
114
-------
TABLE 4-44. BATS FIELD MEASUREMENTS-HIGH PENTATEST
Dot*
10/03/8
10/03/8
10/03/8
10/03/89
10/03/69
10/03/89
10/03/89
10/03/8!
10/03/8!
10/03/8!
10/03/89
10/03/89
10/03/89
10/03/81
10/03/81
10/03/89
10/03/89
10/03/89
10/03/81
10/03/89
10/03/89
10/03/89
10/03/89
10/03/85
10/03/89
10/04/8S
10/04/89
10/04/8S
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/04/89
10/05/89
10/06/89
10/06/89
10/06/89
10/06/89
10/05/89
10/05/89
10/05/89
10/06/89
10/05/89
10/06/89
10/05/89
10/06/89
10/05/89
10/05/89
10/06/89
10/05/89
Tim*
05:2
06:2
06:52
06:55
07:27
07:30
09:46
10:2
10:2
11:3
12:16
14:06
14:11
16:41
18:06
17:28
19:21
20:26
21:3(
21:44
21:48
22:37
22:52
23:21
23:38
)1:25
)1:28
)3:30
14:06
04:12
05:26
16:35
17:26
08:10
19:33
1:46
2:16
4:20
6:00
5:40
7:35
8:15
9:38
20:18
1:00
1:26
23:50
0:36
1:42
3:30
3:33
6:34
5:38
6:22
7:06
7:38
9:40
1:53
3:05
4:15
4:40
4:50
5:20
>:47
Tim* from
Btg.ofT«*
-------
TABLE 4-44. BATS HELD MEASUREMENTS -HIQHPENTA TEST
Oat*
10/05/89
10/05/69
10/05/89
10/05/89
10/05/89
10/05/89
10/05/89
10/05/89
10/08/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
10/06/89
AVERAGE
STODEV
Time
16:40
17:28
19:34
19:62
21:26
21:34
21:38
23:27
0125
01:28
01:58
03:25
03:27
06:22
07:26
07:33
07:36
08:16
09:16
12:10
12:30
13:30
15:30
15:56
Tim* from
Beg.o
-------
TABLE 4~*5. SBR TIME-WEIGHTED MASS & VOLUME BALANCE
INFLUENT
DAY TIME
START 12:04
14:08
02:10
1 11:43
16:00
01:45
2 11:58
13:50
01:37
3 11:43
14:04
01:30
4 11:52
13:55
01:55
5 12:04
13:54
02:00
6 11:59
13:52
01:47
7 11:59
13:59
01:45
8 11:44
13:48
01:35
9 11:48
13:50
01:15
10 11:58
14:16
01:30
11 11:55
14:09
01:20
12 11:44
DATE
10/17/89
10/17/89
10/18/89
10/18/89
10/18/89
10/19/89
10/19/89
10/19/89
10/20/89
10/20/89
10/20/89
10/21/89
10/21/89
10/21/89
10/22/89
10/22/89
10/22/89
10/23/89
10/23/89
10/23/89
10/24/89
10/24/89
10/24/89
10/25/89
10/25/89
10/25/89
10/26/89
10/26/89
10/26/89
10/27/89
10/27/89
10/27/89
10/28/89
10/28/89
10/28/89
10/29/89
10/29/89
HOURS
FROM
START
0.00
2.07
14.10
23.66
27.93
37.68
47.90
49.92
61.55
71.65
74.00
85.43
95.81
97.85
109.85
120.00
121.83
133.93
143.92
145.80
157.72
167.92
169.92
181.68
191.67
193.73
205.52
215.73
217.77
229.18
239.91
242.20
253.43
263.85
266.08
277.27
287.67
DENSITY
g/ml
1.019
1.081
1.022
1.022
1.021
1.014
1.022
1.018
1.016
1.020
1.017
1.011
1.017
1.022
1.018
1.019
1.018
1.025
1.015
1.023
1.014
1.014
1.023
1.016
VOLUME
MASS
FLOWRATE PER 24 HRS.
ml/sec L g
0.426
31.06
0.359
0.384
33.48
0.394
0.398
34.09
0.400
0.407
38.01
0.425
0.407
36.05
0.420
0.412
36.06
0.427
0.412
36.05
0.421
0.424
35.94
0.417
0.412
35.99
0.419
0.403
35.35
0.407
0.403
34.93
0.407
0.398
34.12
31730
34200
34710
36650
36710
36570
38770
36590
36710
35990
35540
35030
EM-LUENT
TIME
12:04
14:08
02:10
12:02
16:00
01:45
11:53
13:59
01:37
11:45
14:04
01:30
11:40
13:55
01:55
11:49
13:64
02:00
11:28
13:52
01:47
11:50
13:59
01:45
11:16
13:48
01:35
11:34
'13:50
01:15
11:41
14:16
01:30
11:41
14:09
01:20
11:37
HOURS
FROM
START
0.00
2.07
14.10
23.98
27.93
37.68
47.82
49.92
61.55
71.68
74.00
85.43
95.60
97.85
109.85
119.76
121.83
133.93
143.41
145.80
157.72
167.77
169.92
181.68
191.20
193.73
205.52
215.50
217.77
229.18
239.62
242.20
253.43
283.63
266.08
277.27
287.55
DENSITY
g/ml
1.019
1.028
1.023
1.056
1.021
1.039
1.031
1.031
1.031
1.023
0.974
0.998
1.031
1.047
1.039
1.024
1.033
1.029
1.018
1.033
1.025
1.031
1.025
1.028
1.028
1.029
1.029
1.023
1.026
1.024
1.023
1.025
1.024
,1.017
1.026
1.022
VOLUME
MASS
PER 24 HRS.
L 3
31.18
33.64
33.70
35.06
44.59
25.86
36.13
34.87
35.22
34.82
34.17
38.66
32050
34850
34690
35060
46310
26590
37100
35830
36200
35670
34950
39550
(Continued)
117
-------
TABLE 4-45. SBR TIME-WEIGHTED MASS & VOLUME BALANCE
INFLUENT
DAY TIME DATE
13:58 10/29/89
01:20 10/30/89
13 11:51 10/30/89
14:08 10/30/89
01:10 10/31/89
09:47 10/31/89
14 11:51 10/31/89
TOTAL FOR NINE DAYS
AVERAGE
STANDARD DEVIATION
HOURS
FROM
START
289.90
301.27
311.79
314.07
325.10
333.72
335.78
335.78
RELATIVE STD. DEVIATION (%)
VARIANCE
(•!-{Vin/Vout))*100(%)
CHMin/Mout))*100 (%)
VOLUME MASS
DENSITY FLOWRATE PEH24HRS.
9/ml
1.039
1.027
1.036
1.042
1.033
1.024
0.01
1.28
0.000
ml/sec L g
0.3S6
0.407
34.59 35690
0.389
0.398
0.398
34.31 35590
314.7 320640
0.392
0.08
19.24
0.006
EFFLUENT
HOURS
VOLUME MASS
FROM DENSITY
TIME
13:58
01:20
11:36
14:08
01:10
09:47
11:51
START
289.90
301.27
311.54
314.07
325.10
333.72
335.78
335.78
g/ml
1.025
1.024
1.025
1.019
1.026
1.029
1.025
1.026
0.01
1.22
0.005
First 9 days of influent compared to last 9 days of effluent
Rrst 9 days of influent compared to last 9 days of effluent
PER24HRS.
L g
31.49 32240
35.26 36160
306.5 314300
2.62
1.98
FILENAME: TBL4-39
118
-------
is a temporal tabulation of density and flow rate measurements for both influent
and effluent streams. A statistical analysis of the variations in these
measurements is included at the bottom. Results of time-weighted average
calculations of the volume and mass as well as their respective closure
calculations are also shown.
The volumetric flow rate was quite constant at about 0.4 ml/sec (0.38
gph) with less than 1% variation. Since each of the three reactors had a volume
of 60 liters, the total SBR volume was 180 liters. At 0.4 ml/sec, the average
retention time was approximately 5.2 days. Because the SBR was operated in a
continuous feed mode, equalization was presumed to have occurred in each of the
three reactors. However, all of the mass of influent introduced on a particular
day would not necessarily be accounted for in the effluent mass 5 days later.
It would instead be most likely to be accounted for over several days with most
of the mass being accounted for on day 5. This distribution would depend on
reactor volume and flow rate. Relying on past experience, the manufacturer
contends that the more stages that are arranged in series, the more the system
approaches true plug flow conditions. Nevertheless, for purposes of this study,
true plug flow operation was assumed, i.e. , all influent mass corresponded to all
the effluent mass measured 5 days later.
In reviewing the analytical data, it is important to keep this 5 day time
lag in mind. To properly characterize the performance of the SBR, the influent
concentrations on a particular day should be compared to the effluent
concentrations measured 5 days later. Both the influent and effluent densities
were quite constant at about 1.025 gm/ml with about a 1% variation in the
measurements. The total volume and mass figures shown on the bottom are for the
first nine days of influent and days 6 through 14, the corresponding nine days,
of effluent flow. This is to properly account for the five day residence time
of the SBR. Volume and mass balance closure were excellent with a discrepancy
of about 2% between influent and effluent, indicating that all mass was properly
accounted for.
Penta and PAHs were the two primary contaminants of interest to determine
if BioTrol's claim of 90-95% reduction was valid. Because of the high solids
content in the slurry (10.5+2.9%), all influent and effluent composited samples
were separated into solid and liquid phases before the extractions were performed
(see later section on Total Residue for solid/liquid composition of the liquid
and solid samples) . The extracts from both phases were then recombined for the
chemical analyses. Problems with separation of the solid and liquid phases by
centrifugation were encountered. It was ultimately decided to incorporate an
additional filtration step, but this in turn delayed the subsequent extraction
and analyses, resulting in samples exceeding the holding times specified in the
QAPj P.
To determine if missed holding times' would have an effect on analyte
concentrations, a holding time study was done using a limited number of samples.
However, it was not sufficiently comprehensive to be conclusive and the effect
of missed holding times could not be determined from the limited amount of data
generated. Hence, it was not taken into account and data were used as obtained.
119
-------
Pentachlorophenol (Penta) -- Analytical results for penta in the liquid phase are
shown in Table 4-46 and results for the solid phase are presented in Table 4-47.
The data are positioned in the tables so that influent data can be compared with
the effluent data for 5 days later, based on the retention time in the system.
It should be noted that the filtration left some aqueous phase in the solids.
Process monitoring for the first five days of operation revealed that the SBR was
not functioning as expected and the acclimation period presumably extended beyond
day 5. This was primarily due to the fact that influent penta concentrations
were higher than anticipated and process parameters had to be adjusted when this
was recognized. No influent readings were taken on day 5 because process
parameters were still being adjusted. SBR performance comparisons were done
beginning with day 1 influent and day 6 effluent. Calculated penta reductions
for both liquid and solid phases are plotted in Figure 4-21. Negative reduction
values are not shown on Figure 4-21; it is believed that such anomalies were due
to incomplete equalization within the SBR reactor vessels, as discussed earlier,
and are not representative of SBR performance.
Penta concentrations in the solid phase were approximately two orders of
magnitude higher than in the liquid phase. One reason for this may be that penta
solubility in water is reported (Merck Index) as about 80 mg/L, but is pH
dependent. Additionally, penta concentration for the liquid phase of the
influent (Figure 4-22) decreased over the fifteen day test period from 80.5 to
-35 mg/L, whereas the influent penta concentration for the solid phase (Figure
4-23) remained fairly constant.
Considering the results after the extended acclimation period i.e. , after
day 5, it can be seen from the data in Tables 4-46 and 4-47 and Figure 4-21 that
penta reductions remained at around 97% for the liquid phase but continued to
increase asymptotically from approximately 65% to 92% for the solid phase, even
after 4 days of "steady state" operation (day 6 through day 9) . The most likely
reason for such performance probably can be traced back to the contaminant
concentrations discussed earlier. The contaminant concentration of the liquid
phase, being two orders of magnitude less than the concentration of the solid
phase, may insure that there are more than enough bacteria to consume just about
all of the contaminant. Conversely, the solid phase has such a high contaminant
concentration that bacteria must be generated during the acclimation process to
produce a population large enough or aggressive enough to consume all the
contaminants and/or overcome any toxicity from the high penta concentration. In
this case, the bacteria population or its growth rate may limit the biological
degradation reaction rate. Alternately, the solid phase may merely serve as a
reservoir or ballast assuring a constant penta concentration in the liquid phase
until a tolerable, non-inhibitory concentration is reached on the soil.
Table 4-48 summarizes the influent and effluent penta data for both solid
and liquid phases. Concentrations were taken from Tables 4-46 and 4-47. Split
samples taken on day 8 were averaged. Solid phase and total solid weight
fractions were taken from Table 4-65 (shown later). The water weight fraction
was calculated by subtracting the total solid weight fraction from 1. Penta
concentration in the solid and liquid phases had to be properly apportioned
according to their relative masses to evaluate SBR performance accurately. Penta
120
-------
TABLE 4-46. SBR LIQUID PHASE PENTA DATA (MG/L)
INFLUENT
DAY
1
2
3
4
5
6
7
8
9
10
11
12
13
14
SAIC#
SBRIN011LSJ3L
SBRIN021LS_DL
SBRIN031LS_DL
SBRIN041LS_DL
SBRIN061LS_DL
SBRIN071LS_DL
SBRIN081LSJ3L
SBRIN085LS-SP01
SBRIN085LS_DL-SP01
SBRIN091LS
SBRIN101LS_DL
SBRIN111LS
SBRIN121LS
SBRIN131LS
SBRIN141LS
AVERAGE
STD DEV
PENTA
71
88
87
46
79
69
87
59
43
46
48
62
32
37
59
30
59
20
Q
D
DJ
D
D
D
DJ
D
*
D
D
D+
EFFLUENT
SAIC#
SBREFF011LSDL
SBREFF021LSDL
SBREFF031LSDL
SBREFF041LSDL
SBREFF051LSDL
SBREFF061LSDL
SBREFF071LSDL
SBREFF081LSDL
SBREFF085LSDL-SP01
SBREFF091LS
SBREFF101LSDL
SBREFF111LS
SBREFF121LS
SBREFF131LSDL
SBREFF141LS
PENTA
230
60
170
280
160
170
100
67
71
75
49
180
6.8
2.2
2.3
1.4
111
14
Q
D
D
D
D
D
D
DJ
D
*
D
D
J
J
J+
J
REMOVAL
EFF.
(%)
-139
-14
18
-7
91
97
96
97
FILENAME: TBL4-42
+ Values interchanged with effluent based upon SAIC's QA officer
recommendation after lab. visit.
* Average value for 3 samples
Q Codes: D = diluted sample
J = estimated value
•121
-------
TABLE 4-»7 SBR SOLID PHASE PENTA DATA
INFLUENT
DAY
1
2
3
4
5
6
7
8
9
10 '
11
12
13
14
SMC*
SBRIN011SS
SSRIN021SS_DL
SBRIN031SS_DL
SBRIN041SS_DL
SBRIN061SS_DL
SBRIN071SS
SBRIN081SS_DL
SBRIN085SS_DL-SP01
SBRIN091SS_DL
SBRIN101SS_DL
SBRIN111SS_DL
SBRIN121SS_DL
SBRIN131SSJ3L
SBRIN141SS_DL
AVERAGE
STDDEV
PENTA
2,600
2,500
2.700
2,000
3,100
1,600
3.400
3,150
2,900
3,500
2.300
2.300
2.300
2.600
2,200
2571
506
Q
D
D
D
0
D
*
D
D
D
D
D
D
D
SMC*
S8REFF011SSOL
S8REFF021SSDL
S8REFF031SS
SBREFF041SSOL
S8REFF051SS
SBREFF061SSOL
SBREFF071SSOL
S8REFF081SSOL
SBREFF085SSDL-SP01
SBREFF091SSDL
SBREFF101SSOL
SBREFF111SS
SBREFF121SSDL
SBREFF131SS
SBREFF141SSDL
PENTA
2,100
1,600
1.300
1.500
1.600
2,300
2.100
2.300
2,350
2.400
2.600
1.800
1.100
620
660
280
1617
688
Q
~~D
D
D
D
D
D
*
D
D
D
0
D
EFF.
(%)
12
10
100
-30
es
61
79
92!
FILENAME: TBL4-43
* Average value for 2 samples
Q Codes: D = diluted sample
122
-------
£21
Percent Removol Based on Concentration
c
E*
TJ
5"
(O
o
o
£_
a
O
u>
o
m
i
ro
CO
|—
c:
^3
33
-<
DO
*•*«
O
I
^3
m
>
o
•H
O
m
m
2
O
-------
(Q
C
-J
-p>
ro
?enta Concentration (ppm)
O
O
O
O
O
CO
oa
73
I/I
ro
10
rt-
y
•*
c-t
fa
c
(D
3
rt-
C
_j.
Q.
-a
3-
Cu
in
fD
-------
I
ro
CO
to
CD
33
3
H-
CO
O
Q.
13
en
n>
Penta Concentration (Ppm)
IQ
C
n>
O
O
O
O
O
1 I
O
O
O
O
t-1
N>
Ui
-a
0)
3
rt-
OJ
a
-------
TABLE4-48. OVERALL PENTA REDUCTION SBR
INFLUENT
Ifc&l
DAY
1
2
3
4
6
e
7
8
9
SoMPhaaa
Cone.
(ppm)
2,600
2,600
2,700
2,000
3,100
1,800
3,150
3,500
Solid
WBt-Ffad
o.so
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PmtaMata
(am)
47.85
17.81
26.68
14.11
15.28
8.59
34.82
36.94
Liquid PhaM
Cone.
(ppm)
71
88
87
46
79
69
59
48
PantaMu*
(flm)
1.6O
2.71
2.82
1.61
2.71
2.35
1.85
1.52
Maaa
(gm)
31730
34200
34710
36660
36710
36570
36770
36680
38710
Malar Wgl
Fract.
0.710
0.900
0.860
0.898
0.938
0.927
0.858
0.862
Solid WgL
Fract.
0.290
0.100
0.131
0.102
0.062
0.073
0.142
0.138
TOTAL
Pantatlaa*
torn)
49.46
20.62
29.31
15.62
17.99
10.94
36.68
38.46
FwiuCone,
(ppm)
600
846
426
492
288
1002
1048
OVERALL
PENTA
REDUCTION
c*>
39
35
•-89
4«
48
86
96
* Average value
EFFFLUENT
i£&r
DAY
6
7
8
9
10
11
12
13
14
SoBdPhaM
Cone.
(ppm)
2,300
2,100
2.3SO
2,600
1,800
1,100
620
660
280
Solid
Wat Fract
0.46
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PontaMan
(gm)
20.11
9.97
17.38
27.64
18.53
9.02
6.01
4.56
1.50
Liquid PhaM
Cone.
(ppm)
170
100
71
49
180
6.8
2.2
2.3
1.4
PentaMaM
(gm)
3.85
3.51
2.30
1.53
5.61
0.21
0.08
0.07
0.05
TOTAL
MaM
torn)
26590
37100
35830
36200
35670
34950
39560
32240
36160
WhMfWgtJ
Fract
0.862
0.945
0.905
0.862
0.873
0.878
0.876
0.895
0.832
SoildWgt
Fract
0.148
0.056
0.085
0.138
0.127
0.122
0.125
0.105
0.068
PmtaMaaa
(gin)
23.97
13.47
18.69
29.16
24.14
9.23
6.09
4.63
1.54
PdntaCone.
(ppm)
901
363
550
806
677
264
154
143
43
FILENAME: TBL4-44
126
-------
mass for the solid phase was calculated by dividing the concentration of penta
in the solid phase by the solid phase weight fraction and then multiplying by the
total solid weight fraction and the total mass. A sample calculation for the
solid phase measurements on the first day follows:
2600 gm Penta
1 x 106 gm solid phase x 0.29 gm solid x 31,730 gm waste = 47.85 gm penta
0.5 gm solid gm waste
gm solid phase
Penta mass for the liquid phase was calculated by multiplying the liquid phase
penta concentration by the water weight fraction and the total mass. A sample
calculation for the first day liquid phase influent measurements follows:
71 gm penta x 0.71 gm water x 31,730 gm waste = 1.60 gm penta
1 x 106 gm water gm waste
The total penta mass was calculated by summing the penta mass of the
solid and liquid phases. Total penta concentration was calculated by dividing
the total penta mass by the total mass and multiplying by a million to get ppm.
Overall penta reduction was calculated by subtracting the ratio of effluent to
influent total penta concentration from 1 and multiplying by 100. The results
are plotted in Figure 4-24. Table 4-48 confirms that the majority of the penta
is found in the solid phase. Hence, the behavior of the solid weight fraction
has a significant influence on penta removal. This influence has been taken into
account by the calculation procedure just described and can be observed by
comparing Figure 4-24 with Figure 4-21. The high removal efficiencies observed
for the liquid phase are tempered by the relatively lower removal efficiencies
observed for the solid phase (Figure 4-21) primarily through the solids weight
fraction.
As mentioned earlier, the acclimation period extended at least through
the first five days of testing and this is shown by the relatively low removal
efficiencies (-40%). There is no satisfactory explanation for the negative
removal observed on day 4 except that BioTrol was still adjusting process
parameters to get optimum performance on day 5. Hence, the SBR was not operating
at steady state conditions for the scheduled 14 day test period as prescribed in
the Demonstration Test Plan. Had the SBR been allowed to operate for a longer
period of time, its performance might have stabilized at a steady state value
that would have been more indicative of BioTrol's penta removal claim of 90-95%.
It appears, however, that overall removal efficiency for penta (solid and liquid
phases) does reach 90% or above based on the last few data and the apparent trend
shown in Figure 4-24.
Polynuclear Aromatic Hydrocarbons (PAH) -- For the liquid phase, there were no
values above detection limits for either influent or effluent streams. Tables
4-49 and 4-50 present the PAH detection limits reported for the liquid phase
samples. Tables 4-51 and 4-52 show the influent and effluent solid phase results
for PAHs. It should be observed that the solid phase influent concentrations for
the seven PAH compounds of interest are of the same order of magnitude as the
penta influent concentration for the liquid phase, discussed earlier. If the
127
-------
831
Percent Removal Based on Total PCP Cone
c
CD
31
o
CD
D
D
o
ro
o
O-J
o
en
o
CD
O
o
00
o
CD
o
co
o
o
c
~-i
CD
K)
J\,
00
o
Q <
ro CD
o ~^i
Q
n
CD
CD
O
<
Q
-------
TABLE 4-48. 88H INFLUENT LIQUID PHASE PAH DETECTION LIMITS
DAY
2
3
4
5
6
7
8
9
10
11
12
13
14
SAICf
SBRIN011LS
SBRIN021LS
SDRIN031LS
SBRIN041LS
SBRIN061LS
SBRIN071LS
SBRIN081LS
SBRIN086LS-
SP01
SBRIN091LS
SBRIN101LS
SBRIN111LS
SBRIN121LS
SBRIN131LS
SBRIN141LS
Acanaph-
thylene
5100
2600
2800
2500
930
2400
2900
2900
5700
2300
2600
4900
540
2500
Acenaph-
Miane
5100
2500
2800
2500
.
930
2400
2900
2900
5700
2300
2600
4900
540
2500
Fluorine
5100
2500
2800
2500
930
2400
2900
2900
6700
2300
2600
4900
540
2500
Phanan-
thrana
6100
2500
2800
2500
930
2400
2900
2900
5700
2300
2600
4900
540
2600
Anth-
racene
5100
2600
2800
2500
930
2400
2900
2900
6700
2300
2600
4900
640
2500
Fluor-
anthana
6100
2600
2800
2500
930
2400
2900
2900
5700
2300
2600
4900
540
2500
Pyrana
5100
2500
2800
2500
^
930
'2400
2900
2900
6700
2300
2600
4900
640
2500
Banzo-
(a)-An-
thracena
6100
2500
2800
2500
930
2400
2900
2900
5700
2300
2600
4900
640
2500
Chrynana
5100
2600
2800
2500
830
2400
2900
2900
6700
2300
2600
4900
640
2600
Benzo(b)-
Fluor-
anthena
6100
2500
2800
2500
830
2400
2900
2900
6700
2300
2600
4900
640
2500
Banzo(k>-
Ftuor-
an thane
6100
2600
2800
2600
830
2400
2900
2900
6700
2300
2600
4900
540
2500
Banzo(a)-
Pyram
5100
2500
2800
2500
—___
830
2400
2900
2900
6700
2300
2600
4900
540
2600
Banzo
fo,h,fr-
Parylana
6100
2500
2800
2500
•
930
2400
2900
•2900
5700
2300
2600
4900
640
2600
N>
VO
FILENAME: TBL4-45
-------
TABLE 4-50. SBR EFFLUEKT LIQUID PHASE PAH DETECTION LIMITS
CO
O
TEST
DAY
1
2
3
4
5
6
7
8
9
10
11
12
13
14
FILEN/
• '
SAlCf
oontrrUl iLo
SBREFF021LS
SBREFF031LS
SBREFF041LS
SBREFF051LS
SBREFF061LS
SBREFF071LS
SBREFF081LS
SBREFF085LS-
SP01
SBREFF091LS
SBREFF101LS
.SBREFF111LS
SBREFF121LS
SBREFF131LS
SBREFF141LS
ME: TBL4-46
Acenaph-
thytene
4000
2600
2400
2800
2300
1200
2300
2600
2500
6700
2800
2600
530
2600
470
—
Aoanaph-
ttione
4000
2600
2400
2SOO
2300
1200
2300
2600
2500
5700
2600
2600
530
2600
470
Fluortnt
4000
2600
2400
2800
2300
1200
2300
2600
2500
5700
2800
2600
530
2600
470
—
Ptionan-
ttuena
4000
2600
2400
2800
2300
1200
2300
2600
2500
5700
2800
2600
530
2600
470
An ttl-
rac«n«
4000
2600
2400
2800
2300
1200
2300
2600
2500
5700
2800
2600
530
2600
470
.: . =
Fluof-
anttimt
4000
2600
2400
2800
2300
1200
2300
2600
2600
5700
2800
2600
530
2600
470
Pynna
4000
2600
2400
2600
2300
1200
2300
2600
2500
6700
2800
2600
530
2600
470
Banzo-
(oMn-
thracena
4000
2600
2400
2800
2300
1200
2300
2600
2500
6700
2800
2600
630
2600
470
:=^=zz^=^ — i
Chryoono
4000
2600
2400
2800
2300
1200
2300
2600
2500
5700
2800
2600
530
2600
470
B«nzo(b>-
Ftuw-
anttwna
4000
2600
2400
2800
2300
1200
2300
2600
2500
6700
2800
2600
630
2600
470
B»nzo(k)
Fluof-
•nthono
4000
2600
2400
• 2800
2300
1200
2300
2600
2500
6700
2800
2600
530
2600
470
••
Benzofa)-
Pyren*
4000
2600
2400
2800
2300
1200
2300
2600
2500
6700
2800
2600
630
2600
470
Bcnzo
(9,h,i)-
Paryfena
4000
2600
2400
2800
2300
1200
2300
2600
2500
6700
2600
2600
630
2600
470
-------
TABLE 4-61. S8R INFLUENT SOLID PHASE PAH DATA
(rng/kg)
Ihs
DAY
1
2
3
4
5
6
7
8
9
10
11
12
13
14
SAC*
SBRIN011SS
S8RIN021SS
SSRIN031SS
SBRIN041SS
SBRIN061SS
SBRIN071SS
SBRIM081SS
SBRIN085SS-
SP01
SBRIN091SS
SBRIN101SS
SBRIN111SS
SBRIN121SS
SBRIN131SS
SBRIN141SS
AVERAGE
STDDEV
Acenaph-
thene Q
190 U
88
78
50
46 J
18 J
17 J
48 U
7.7 J
49 U
49 U
4.1 J
50 U
50 U
53
44
Fluor-
antfteno Q
390
480
450
280
330
190
300
310
270
240
170
190
260
22 J
277
114
Pyrene C
400
430
400
360
420
270
490
340
570
410
310
380
280
27 J
363
121
Benzo-
-An-
thracene Q
190 U
96
87
64
74
36 J
51 J
39 J
38 J
30 J
49 U
25 J
50 U
50 U
63
41
ChryueneQ
140 J
180
160
110
120
68
93
82
80
63
43 J
50
50 U
5.2 J
89
47
Benzo(b)-
Fluor- Q
anthene
190 U
52
53
66
49 U
47 J
87
48 U
85
70
50
62
50 U
50 U
69
38
E!enzo(a}-
Pyrene Q
190 U
47 J
18 J
21 J
28 J
52 U
31 J
48 U
32 J
27 J
49 U
23 J
50 U
50 U
48
41
FILENAME: TBL4-47
Q Codes: U » undetected sample
J » eotimated value
131
-------
TABLE 4-62. S8H EFFLUENT SOUD PHASE PAH DATA
(mg/kg)
TEST
DAY
1
2
3
4
5
6
7
8
9
10
11
12
13
14
SAC*
SBREFF011SS
SBREFF021SS
SBREFF031SS
SBREFF041SS
SBREFFOS1SS
SBREFF061SS
SBREFF071SS
SBREFF081SS
SBREFF085SS-
SP01
SBREFF091SS
SBREFF101SS
SBREFF111SS
SBREFF121SS
SBREFF131SS
SBREFF141SS
AVERAGE
STDDEV
Acenapn-
thene Q
48 U
49 U
6.8 J
49 U
53 U
49 U
48 U
48 U
50 U
50 U
48 U
49 U
10 U
55 U
11 U
42
16
Fluor-
anthene Q
210
210
210
310
250
220
170
160
180
120
83
55
43
55 U
11 U
152
84
Pyrene Q
380
350
340
370
310
350
310
290
290
360
330
320
210
200
64
298
80
Bonzo-
(a)-An-
thracene Q
54
49
51
66
51 J
44 J
28 J
44 J
50 U
50 U
35 J
27 J
26
55 U
9.7 J
43
14
Chrysene Q
86
87
87
110
91
75
53
47 J
50 U
39 J
39 J
30 J
29
55 U
11
59
27
Benzo(b)-
Fluor- Q
anthene
71
49 U
61
51
53 U
68
58
48 U
50 U
65
57
59
53
55 U
41
56
8
Benzo(a>-
Pyrene Q
48 U
49 U
19 J
49 U
53 U
22 J
18 J
48 U
50 U
50 U
20 J
49 U
36
55 U
18
39
14
FILENAME: TABLE 4-48
Q Codes: U » undetected sample
J » estimated value
132
-------
same two order of magnitude relationship that was observed between the solid and
liquid phases for the penta data is assumed, then the PAH concentrations for the
liquid phase were too low to be detected. Hence, all PAH values in the liquid
phase were assumed to be insignificant when compared to the PAH concentrations
in the solid phase.
. Undetected PAH values reported by the laboratory for the solid phase are
shown in Tables 4-51 and 4-52 with a "U" next to the number to indicate that the
values are the practical quantitation limit (PQL) . The PQL lies in a range
somewhere between 5-10 times the standard deviation of the background noise level
and is a conservative number to minimize the effect of background noise on the
analytical measurement. A less stringent criterion is the method detection limit
(MDL) and is normally defined as 3 times the standard deviation of the background
noise level. It was decided that a good compromise between using 0 and the
reported PQL for undetected values was to use the MDL.
Unfortunately, the laboratory did not routinely report MDLs. Hence, it
was necessary to determine a fundamental factor relating PQLs to MDLs. From the
above description, this factor had to be between 1.67 and 3.33. The laboratory
did, however, report estimated values denoted by the letter "J" as a qualifier.
An estimated value is reported when a compound meets identification criteria but
the result is less than the PQL and greater than the MDL. Therefore, estimated
values would lie somewhere between the MDL and PQL. The lowest estimated values
would be the closest to the MDL. A cursory review of several semivolatile
compounds for low, medium and high concentration soils selected at random
revealed that a factor of 3 could be used to approximately relate MDLs to PQLs.
Table 4-53 summarizes the solid phase PAH results using the above interpretation
of the data where necessary.
Figures 4-25 through 4-30 show a distinct drop in influent concentration
for six of the seven carcinogenic PAHs as the test progressed (negative
efficiencies are not included in the graphs). As discussed earlier, the
concentrations of these PAHs in the solid phase of the influent are of the same
order of magnitude as the concentrations of penta in the liquid phase. The fact
that both sets of data (Figure 4-22 and Figures 4-25 through 4-30) show a
decrease over the fourteen day test period suggests that microbiological
degradation may be occurring in the holding tank prior to the SBR. This is
probably caused by bacteria that are indigenous to the soil at the demonstration
site. A similar drop in influent penta concentration for the solid phase may not
be evident because of the much higher concentration.
Tables 4-54 through 4-60 and Figures 4-31 through 4-37 show the 5 day lag
time removal efficiency for seven PAH compounds. Although there is a limited
amount of data, some general trends can be observed:
o performance in the first 4 days of operation appeared to peak on day
2 and then drop precipitously;
o after day 5, removal efficiency increased to between 70% and 99% on
day 9;
133
-------
sets i
29991
Z9991
ion
r 00*8
n oooss
00097
r ooozz
r ooosc
n 00009
n oooos
r ooovt
r OOOK
r ooott>
O p*in§i»M
(Bx/Cnimarma
2OOOI.
20091
ECE9L
ion
n 00005
n oooos
r oooss
n OOOBV
r ooooc
r ooosc
r oooec
r ooois
r oooec
OOOfrl
O P"*"U"II
(Oil/fin) wenpi)
>l
EL
Zl
11
01
e
8
8
L
e
A»a
ion
000*9
000003
000013
oooo2E
oooocc
00009C
ooooez
000062
OOOOIC
oooost
Opamtwun
(0>i/Cn)>o»nBQ
nan
r 000^2
000083
OOOOSE
0000 IE
OOOOlt
oooozs
oooow:
oooom-
OOOOZ2
000021-
D Dajn>*«f|
(BvOn)»uenuu|
VI
El
21
u
01.
6
8
s
L
g
ABQ
oat
CCE8L
nan
n ooou
n oooss
oooet
oooss
OOOE8
000021
000081
000091
OOOOil
000023
D pwnrn^
__
now
r 00023
000092
OOOOBl
OOOOZl
0000*2
0000£2
OOOOIC
oooooc
00006L
OOOOCC
DPunmn
,«_»
n
Cl
Zl
u
01
e
8
8
i.
9
*
JJtK
EEESl
CCCC
EEE91
law
n ooou
n oooss
n oooot
n oooot
DP*ira»n
(B)Vfifi)»uenBa
Z8991
ZB991
CCE9L
CCC9L
00091
nan
n oooos
n oooos
r ooi>
n ooost •
n oooot
r ooii
n ooost
r ooaii
r ooosi
r ooost
D pamnen
(BVfin) juenuuf
VI
CL
21
11
01
6
8
8
L
g
ABQ
sitnsaa nvd asvHd QITOS -£S-*
-------
TABLE 4-53. SOLID PHASE PAH RESULTS SUMARY
(CONTINUED)
Chrywn*
Day
8
7
8
8
9
10
11
12
13
14
IrafiiMnt (ug/kg)
Mauurad Q
120000
88000
93000
82000
80000
83000
43000 J
50000
50000 U
5200 J
MOL
27333
Effluent (ug/kg)
UauurodQ
75000
53000
47000 J
50000 U
39000 J
39000 J
30000 U
29000
55000 U
11000
MOL
16667
18333
BenzofW-Huoranthww
Day
8
7
3
8
9
10
11
12
13
14
ln*Mnt(ug/kQ)
UMWdQ
48000 U
47000 J
67000
4*000 U
•5000
70000
50000
62000
50000 U
50000 U
UOL
18333
16000
16887
16887
Effluent (ug/kg)
htouuradQ
fifiOOQ
58000
48000 U
50000 U
65000
57000
59000
53000
55000 U
41000
UOL
16000
16667
18333
Day
a
7
3
8
9
10
11
12
13
14
Influont (ug/kg)
Mauurad Q MOL
28000 J
52000 U
31000 J
48000 U
32000 J
27000 J
49000 U
23000 J
50000 U
50000 U
17333
16000
18333
16667
16667
EfBUMK(ug/kg)
Mouured Q MOL
22000 J
18000 J
48000 U
50000 U
50000 U
20000 J
49000 U
38000
55000 U
18000
16000
16667
16667
16333
18333
135
-------
9£I
Pyrene Concentration (ppb)
Acenaphthene Concentration (ppb)
O
LS
re
3
Benzo(A)Anthracene Concentration (ppb)
g
o
i-
o
Fluoranthene concentration (ppb)
e
g
o
e
o
o
10
c:
-5
Q
O.
on
° I
-------
Benzo(A)Pyrene Concentration (ppb)
Chrysene Concentration (ppb)
o
o
g
00
D
Q.
to
c
-s
<<
o
-h
ID
2
id'
-5
n>
.fl
o
to
ca
70
n>
1/1
OJ
n>
N
o
(D
3
ID
§
g
o
g
~T III I I i\
O
•<
a.
to
c
o
-h
-s
CO
I
ro
10
to
ca
73
ro
in
o
3-
1
(D
S
-------
TABLE 4-64. ACENAPHTHENE REMOVAL EFFICIENCY
INFLUENT
TEST
DAY
1
2
3
4
S
6
7
8
9
Solid Phase
Cone.
(PP«n)
63.3
88
76
50
46.0
18.0
16.5
7.7
Solid
WgtFract
0.50
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PAH Mass
(8«n)
1.16
0.63
0.75
0.35
0.23
0.10
0.18
0.08
TOTAL
Mara
(gm)
31730
34200
34710
36650
36710
36570
36770
36590
36710
Solid WQt
FraeL
0.290
0.100
0.131
0.102
0.062
0.073
0.142
0.138
PAH Maw
(gm)
1.16
0.63
0.75
0.35
0.23
0.10
0.18
0.08
PAH Cone.
(ppm)
36.7
18.3
21.6
9.6
6.2
2.6
5.0
2.2
OVERALL
PAH
REDUCTION
(%)
•86
80
84
40
38
69
21
7S
EFFLUENT
TEST
DAY
6
7
8
9
10
11
12
13
14
Solid Phase
Cone.
(ppm)
16.3
16.0
16.3
16.7
16.0
16.3
3.3
18.3
3.7
Solid
WgtFract.
0.45
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PAH Mass
(gm)
0.14
0.08
0.12
0.18
0.16
0.13
0.03
0.13
0.02
TOTAL
Mant
(gm)
26590
37100
35830
36200
35670
34950
39550
32240
36160
Solid Wgt.
Fract
0.148
0.055
0.095
.0.138
0.127
0.122
0.125
0.105
0.068
PAH Mara
(gm)
0.14
0.08
0.12
0.18
0.16
0.13
0.03
0.13
0.02
PAH Cone.
(ppm)
5
2
3
5
5
4
1
4
1
FILENAME: TBL4-50
138
-------
TABLE 4-65. FLUORANTHENE REMOVAL EFFICIENCY
INFLUENT
TEST
DAY
1
2
3
4
5
' 6
7
8
9
Solid Phase
Cone.
(Ppm)
390
480
450
280
330
190
305
270
Solid
WgtFract
0.50
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PAH Maes
(gm)
7.18
3.42
4.45
1.97
1.63
1.02
3.37
2.85
TOTAL
Mass
(am)
31730
34200
34710
36650
36710
36570
38770
36590
36710
Solid Wgt
Fract
0.290
0.100
0.131
0.102
0.062
0.073
0.142
6.138
PAH Mass
(gm)
7.18
3.42
4.45
1.97
1.63
1.02
3.37
2.85
PAH Cone.
(ppm)
226.2
100.0
128.2
53.9
44.5
27.7
92.1
77.6
OVERALL
PAH
RB3UCTION
(%)
68
78
73
35
71
62
96
99
EFFFLUENT
TEST
DAY
6
7
8
9
10
11
12
13
14
Solid Phase
Cone.
(ppm)
220
170
170
120
83
55.0
43.0
18.3
3.7
Solid
WgtFract.
0.45
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PAH Mass
(am)
1.92
0.81
1.26
1.28
0.85
0.45
0.42
0.13
0.02
TOTAL
Mass
(am)
26590
37100
35830
36200
35670
34950
39550
32240
36160
Solid Wat
Fract
0.148
0.055
0.095
0.138
0.127
0.122
0.125
0.105
0.068
PAH Mass
(am)
1.92
0.81
1.26
1.28
0.85
0.45
0.42
0.13
0.02
PAH Cone.
(ppm)
72
22
35
35
24
13
11
4
1
FILENAME: TBL4-51
139
-------
TABLE 4-66. PYRENE REMOVAL EFFICIENCY
INFLUENT
TEST
DAY
1
2
3
4
5
6
7
8
9
Solid Phaw
Cone.
(ppm)
400
430
400
360
420
270
415
570
Solid
WgtFract
0.50
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PAH Mam
(am)
7.36
3.06
3.95
2.54
2.07
1.45
4.59
6.02
TOTAL
Mac*
fam)
31730
34200
34710
36650
36710
36570
36770
36580
36710
Solid Wgt
Fract
0.290
0.100
0.131
0.102
0.062
0.073
0.142
0.138
PAH Mara
(gm)
7.36
3.06
3.95
2.54
2.07
1.45
4.59
6.02
PAH Cone.
(ppm)
232.0
89.6
113.9
69.3
56.6
39.4
125.4
163.9
OVERALL
PAH
REDUCTION
{%)
50
56
47
-53
-•33
-31
66
94
EFFFLUENT
TEST
DAY
6
7
8
9
10
11
12
13
14
Solid Phaea
Cone.
(ppm)
350
310
290
360
330
320.0
210.0
200.0
64.0
Solid
WgtFract
0.45
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PAH Mass
(gm)
3.06
1.47
2.15
3.83
3.40
2.62
2.04
1.38
0.34
TOTAL
Mas*
(gm)
26590
37100
35830
36200
35670
34950
39550
32240
36160
Solid Wgt
Fract
0.148
0.055
0.095
0.138
0.127
0.122
0.125
0.105
0.068
PAH Mass
(gm)
3.06
1.47
2.15
3.83
3.40
2.62
2.04
1.38
0.34
PAH Cone.
(ppm)
115
40
60
106
95
75
51
43
9
FILENAME: TBL4-52
140
-------
TABLE 4-57. BENZO(A) ANTHRACENE REMOVAL EFFICIENCY
INFLUENT
TEST
DAY
1
2
3
4
5
6
7
8
9
Solid Phase
Cone.
(PPH>)
63
96
87
64
74
36
45
38
Solid
WgtFract
0.50
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PAH Mass
(gm)
1.16
0.68
0.86
0.45
0.36
0.19
0.50
0.40
TOTAL
Maw
(gm)
31730
34200
34710
36650
36710
38570
36770
36590
36710
Solid WgL
Fract
0.290
0.100
0.131
0.102
0.062
0.073
0.142
0.138
PAH Man
(gm)
1.16
0.68
0.86
0.45
0.36
0.19
0.50
0.40
PAH Cone.
(ppm)
36.5
20.0
24.8
12.3
10.0
5.3
13.6
10.9
OVERALL
PAH
REDUCTION
(%)
60
82
75
60
36
-21
71
87
EFFFLUENT
TEST
DAY
6
7
8
9
10
11
12
13
14
Solid Phase
Cone.
(ppm)
44
28
30.3
16.7
35
27.0
26.0
18.3
9.7
Solid
WgtFract
0.45
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PAH Mara
(gm)
0.38
0.13
0.22
0.18
0.36
0.22
0.25
0.13
0.05
TOTAL
Mara
(gm)
26590
37100
35830
36200
35670
34950
39550
32240
36160
Solid Wgt
Fract
0.148
0.055
0.095
0.138
0.127
0.122
0.125
0.105
0.068
PAH Mara
(gm)
0.38
0.13
0.22
0.18
0.36
0.22
0.25
0.13
0.05
PAH Cone.
(ppm)
14
4
6
5
10
6
6
4
1
FILENAME: TBL4-53
141
-------
TABLE 4-68. CHRYSENE REMOVAL EFFICIENCY
INFLUENT
TEST
DAY
1
2
3
4
5
6
7
8
9
Solid Phase
Cone.
(ppm)
140
180
160
110
120
68
88
80
Solid
Wgt Fract
0.50
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PAH Mass
(am)
2.58
1.28
1.58
0.78
0.59
0.37
0.97
0.84
TOTAL
MOM
(am)
31730
34200
34710
36660
36710
36570
38770
36590
38710
Solid Wgt
Fract
0.290
0.100
0.131
0.102
0.062
0.073
0.142
0.138
PAH Man
(am)
2.58
1.28
1.58
0.78
0.59
0.37
0.97
0.84
PAH Cone.
(ppm)
81.2
37.5
45.6
21.2
16.2
9.9
26.4
23.0
OVERALL
PAH
REDUCTION
(%)
70
82
86
46
56
28
35
93
EFFFLUENT
TEST
DAY
6
7
8
9
10
11
12
13
14
Solid Phase
Cone.
(ppm)
75
53
31.8
39
39
30.0
29.0
18.3
11.0
Solid
Wgt Fract.
0.45
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PAH Mass
(gm)
0.66
0.25
0.24
0.41
0.40
0.25
0.28
0.13
0.06
TOTAL
Mas*
(am)
26590
37100
35830
36200
35670
34950
395SO
32240
36160
Solid Wgt
Fract
0.148
0.055
0.095
0.138
0.127
0.122
0.125
0.105
0.068
PAH Mass
(gm)
0.66
0.25
0.24
0.41
0.40
0.25
0.28
0.13
0.06
PAH Cone.
(ppm)
25
7
7
11
11
7
7
4
2
FILENAME: TBL4-S4
142
-------
TABLE 4-69 BENZO(B)-FLUORANTHENE REMOVAL EFFICIENCY
INFLUENT
TEST
DAY
1
2
3
4
5
6
7
8
9
Solid Phase
Cone.
(ppm)
63
52
53
66
16
47
52
85
Solid
WgtFract
0.50
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PAH Mara
(grn)
1.16
0.37
0.52
0.47
0.08
0.25
0.57
0.90
TOTAL
Man
(gm)
31730
34200
34710
36650
36710
36570
36770
36590
36710
Solid Wot
Fract
0.290
0.100
0.131
0.102
0.062
0.073
0.142
0.138
PAH Maes
(gm)
1.16
0.37
0.52
0.47
0.08
0.25
0.57
0.90
PAH Cone.
(ppm)
36.5
10.8
15.1
12.7
2.2
6.9
15.6
24.4
OVERALL
PAH
REDUCTION
(%)
39
32
55
-50
-530
-89
75
75
EFFFLUENT
TEST
DAY
6
7
8
9
10
11
12
13
14
Solid Phase
Cone.
(ppm)
68
58
32.7
65
57
59.0
53.0
18.3
41.0
Solid
WgtFract.
0.45
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PAH Mass
(gm)
0.59
0.28
0.24
0.69
0.59
0.48
0.51
0.13
0.22
TOTAL
Maes
(gm)
26590
37100
35830
36200
35870
34850
39550
32240
36160
Solid Wgt.
Fract
0.148
0.055
0.095
0.138
0.127
0.122
0.125
0.105
0.068
PAH Mass
(gm)
0.59
0.28
0.24
0.69
0.59
0.48
0.51
0.13
0.22
PAH Cone.
(ppm)
22
7
7
19
16
14
13
4
6
FILENAME: TBL4-55
143
-------
TABLE 4-60. BENZO(A) PYRENE REMOVAL EFFICIENCY
INFLUENT
TEST
DAY
1
2
3
4
5
6
7
8
9
Solid Phase
Cone.
(ppm)
63
47
18
21
28
17
24
32
Solid
WgtFract
0.50
0.48
0.46
0.53
0.46
0.50
0.47
0.48
PAH Mas*
(gin)
1.16
0.33
0.18
0.15
0.14
0.09
0.26
0.34
TOTAL
Mass
(gm)
31730
34200
34710
36650
36710
36570
36770
36590
36710
Solid Wgt
Fract
0.290
0.100
0.131
0.102
0.062
0.073
0.142
0.138
PAH Mam
(gm)
1.16
0.33
0.18
0.15
0.14
0.09
0.26
0.34
PAH Cone.
(ppm)
36.5
9.8
5.1
4.0
3.8
2.5
7.1
9.2
OVERALL
PAH
REDUCIION
(%)
80
76
34
-21
-1
-1256
45
71
EFFFLUENT
TEST
DAY
6
7
8
9
10
11
12
13
14
Solid Phase
Cone.
(ppm)
22
18
16.3
16.7
20
16.3
36.0
18.3
18.0
Solid
WgtFract
0.45
0.43
0.46
0.47
0.44
0.52
0.51
0.49
0.46
PAH Mass
(gm)
0.19
0.09
0.12
0.18
0.21
0.13
0.35
0.13
0.10
TOTAL
Mam
(gm)
26590
37100
35830
36200
35670
34050
39550
32240
36160
Solid Wgt
Fract
0.148
0.055
0.095
0.138
0.127
0.122
0.125
0.105
0.068
PAH Mass
(gm)
0.19
0.09
0.12
0.18
0.21
0.13
0.35
0.13
0.10
PAH Cone.
(ppm)
7
2
3
5
6
4
9
4
3
FILENAME: TBL4-56
144
-------
Percent Removal Based on Total PAH Cone
c
CD
3!
o
CD
W
r*-
O
D
-^ ro
o o o
CJi
O
O
O
CO
o
O
O
O
ro
U-J
CD
CD
c
^
0)
Ul
^ >
H D
O —>
i Q
ro
Q
n
CD
Z5
CD
CD
O
<
Q
-------
Percent Removal Basea on Total PAH Cone
00 O O
13
—«-,
c"
3
CD
O
D
c
~^
CD
°
oo
c
- o
o w
0)
D
CD
D
CD
CD
o
<
Q
-------
Percent Removal Based on Total PAH Cone
c
CD
o
CD
0)
r-t-
O
Q
O
O
ro
o
en
O
CD
O
CO
o
ID
o
o
o
00 —
(£> —
c
~^s
CD
Jx
OJ
CD
o'
I
CD —*)
o CD
2 =5
CD
O
<
Q
-------
8"1?!
Percent Removal Based on Total PAH Cone
O
O
Cn
O
O')
O
>J
.0
CO
O
LD
O
c
CD
Z5
O
CD
C/)
rt-
o
D
CO
O
Q
CD
G:
CD
_J
N
O
CD
T
>
Q
n
CD
13
CD
CD
O
<
Q
-------
61?!
Percent Removal Based on Total PAH Cone
c
CD
CD
M
r-H
O
D
O
o
NJ
o
G-J
o
O
CD
O
•CO
o
o
C7)
00
o
o
•S
CD
OJ
w CJi
r^
-^
*~^
o
CD
D
CD
CD
O
<
Q
-------
Percent Removal Based on Total PAH Cone
c
CD
D
CO
I-*-
a
Q
O
o
o
GJ
o
o
Cn
O
O
00
o
10
O
ro
oo -
O
O
I
CD
°
CD
•CD
_)
N
O
Q
CD
D
CD
CD
O
<
Q
-------
1ST
Percent Removal Based on Total PAH Cone
:c
n>
n>
en
O
D
ro
o
o
O
tn
O
en
O
CO
o
o
CD
CX3
o
o
CD
CD
CD
Z5
N
~D
CD
_)
,CD
CD
O
<
Q
-------
o for most compounds, there was a" noticeable drop in removal
efficiency on day 7. This drop was more pronounced for some
compounds than for others.
These trends could be a manifestation of the same phenomenon that was
observed for penta; i.e. , contaminant concentrations are so high that the initial
bacteria population is insufficient to consume it and must acclimate and grow.
Thereafter, removal efficiency increases with increasing bacterial population.
Maximum PAH removal efficiency and PAH removal efficiencies after 9 days of
operation at steady state are summarized in Table 4-61. These compounds had the
lowest influent concentrations on day 7, accounting for the drop in reduction
that was observed. Thereafter, however, a pronounced improvement in reduction
efficiency is evident. •
Copper. Chromium and Arsenic (CCA) -- Table 4-62 summarizes the influent and
effluent CCA data. The first datum point for the influent stream was an outlier
and hence was not used in calculating the statistical parameters at the bottom
of the Table. In this case, solid and liquid phases were analyzed together.
Variations in the concentration of constituents are comparable between
influent and effluent streams. Mass balances between influent for the first 9
days and effluent for the last 9 days of operation for all constituents were
within 2%. Therefore, there was no substantial change in CCA within the SBR, as
was expected.
4.4.4.2 Other Analyses
Total Recoverable Petroleum Hydrocarbons (TRPH) — TRPH is a measure of the
relatively non-biodegradable oils found in a waste. Table 4-63 presents influent
and effluent TRPH data. It should be noted that several points were non-detects.
In these instances TRPH concentrations were assumed to be 50% of the detection
limit for calculation of statistical parameters. Since the first point for both
influent and effluent streams were outliers, they were not used in the
statistical calculations shown at the bottom.
The fluctuations in both influent and effluent concentrations are
substantial. Average TRPH concentrations show a 30% increase between influent
and effluent. This is reflected in a 300% discrepancy for the mass balance when
influent for the first 9 days are compared with effluent for the last 9 days of
operation. Given the large amount of uncertainty in the data, it is difficult
to draw any conclusions.
Chemical Oxygen Demand (COD) - - The chemical oxygen demand is a measure of how
much of the organic content of a sample is susceptible to oxidation by a strong
chemical oxidant. Table 4-64 summarizes the influent and effluent COD
measurements. There is not a significant change in the mean COD concentration
between the influent and effluent (2,113 mg/L vs. 2,053 mg/L). The larger
variation observed for the influent (39%) than for the effluent (23%) may be
attributable to better mixing within the SBR than in the storage tank. The
limited amount of data available after day 5 and the large variation in COD
concentrations make it difficult to draw definitive conclusions.
152
-------
TABLE 4-61. PAH REMOVAL SUMMARY
PAH
COMPOUND
Acenaphthene
Fluoranthene
Pyrene
Benzo(a)Anthracene
Chrysene
Benzo(b)Fluoranthene
Benzo(a)Pyrene
MAXIMUM
REMOVAL
EFFICIENCY
(%)
89
99
94
87
93
75
80
REMOVAL
EFFICIENCY
ON DAY 9
(%)
75
99
94
87
93
75
71
FILENAME: TBL4-57
153
-------
TABLE 4-62. SBa ARSENIC, CHROMIUM & COPPER MATERIAL BALANCES
ARSENIC CHROMIUM COPPER
DAY SAMPLE ID ARSENIC CHROMIUM COPPER VOLUME MASS MASS MASS
2
3
4
6
6
7
8
6
9
10
11
12
13
14
SBR-IN-02-1-CCA
SBR-IN-03-1-CCA
SBR-IN-04-1-CCA
SBR-IN-05-1-CCA
SBR-IN-08-1-CCA
SBR-IN-07-1-CCA
•SBR-IN-08-1-CCA
•SBR-IN-08-6-CCA
SBR-IN-08-1-CCA
SBR-IN-10-1-CCA
SBR-IN-11-1-CCA
SBR-IN-12-1-CCA
SBR-IN-13-1-CCA
SBR-IN-14-1-CCA
(ug/L)
46700
20100
20200
18600
18100
18600
21200
20200
20200
20700
15000
20200
16100
26600
25600
(ugfL)
49400
22000
21300
20700
18500
18700
21300
20100
20200
20700
15300
20800
16300
26800
26400
(ug/L)
64600
24300
23400
22800
21600
21700
23400
21800
22100
22600
20400
24300
22200
28800
30000
(L)
31.06
33.48
34.09
36.01
36.05
36.06
36.05
36.94
36.94
35.99
35.35
34.93
34.12
34.59
34.31
(mg) (mg) (mg)
1419
673
689
702
689
707
764
726
746
630
706
616
882
878
1634
737
728
745
703
710
768
724
745
641
727
556
896
906
1696
814
788
821
779
783
844
791
813
721
849
767
1000
1029
TOTAL (llrst 9 days) 71U ^^ 8137
AVERAGE 20167 20728 23543
STANDARD DEVIATION 2848 2847 2628
REL. STD. DEV. (%) 14.13 13.74 11.16
VARIANCE 8113878 8107766 6803878
EFFLUENT
ARSENIC CHROMIUM COPPER
SAMPLE ID ARSENIC CHROMIUM COPPER VOLUME MASS MASS MASS
(ug/L) (ug/L) (ug/L) (L) (mg) (mg) (mg)
8BR-EFF-01-1-MET
SBR-EFF-02-1-MET
SBR-EFF-03-1-MET
SBR-EFF-04-1-MET
SBR-EFF-05-1-MET
SBR-EFF-06-1-MET
SBR-EFF-07-1-MET
•SBR-EFF-08-1-MET
•SBR-EQB-08-5-MET
SBR-EFF-09-1-MET
SBR-EFF-10-1-MET
SBR-EFF-11-1-MET
8BR-EFF-12-1-MET
'SBR-EFF-13-1-MET
8BR-EFF-14-1-MET
(laal 9 days)
16500
20600
24300
24000
24700
24200
24900
25400
23900
23800
23100
23200
23100
22100
23100
16200
23700
27500
26000
26900
26100
26100
27100
26200
24700
24700
24300
23300
22800
23600
16900
23500
27200
26400
27200
26600
26600
27300
26000
26100
25900
27600
26800
26600
26700
31.18
33.64
33.70
35.06
44.59
26.86
36.13
34.87
34.87
35.22
34.82
34.17
33.63
31.49
36.26
614
693
819
841
1101
626
900
860
838
804
793
893
696
816
7224
60S
797
927
877
1165
649
907
912
870
860
830
901
718
832
7479
627
791
917
926
1213
688
957
929
919
902
940
1023
806
941
8111
23127 24280 26733
2106 2483 2638
8.10 10.27 8.86
4432622 6216267 6440889
Ul
FILENAME: TBL4-58
* Mass calculations used average of split sample
-------
TABUE4-63. S6RTOTAL RECOVERABLE PETROLEUM HYDROCARBON MATERIAL BALANCE
DAY
01
02
03
04
06
06
07
08
OB
09
10
11
12
13
14
SAMPLE ID
SBR-IN-01-1-TRPH
SBR-IN-02-1-TRPH
SBR-IN-03-1-TRPH
SBR-IN-04-1-TRPH
3BR-IN-06-1-TRPH
3BR-IN-08-1-TRPH
SBR-IN-07-1-TRPH
SBR-IN-08-1-TRPH
SBR-IN-08-S-TRPH
SBR-IN-09-1-TRPH
SBR-IN-10-1-TRPH
SBR-IN-11-1-TRPH
SBR-IN-12-1-TRPH
SBR-IN-13-1-TRPH
SBR-IN-14-1-TRPH
TRPH
(mg/L)
4
46
60
50
429
50
50
60
60
228
214
690
670
412
660
QUALIFER
<5X
ND
ND
ND
<6X
ND
ND
ND
ND
<6X
<6X
<6X
lETECTIOh
LIMnr
(mg/L)
2
92
too
100
100
100
100
100
100
100
100
too
100
100
100
VOLUME
0-)
31.06
33.48
34.09
36.01
36.06
36.06
36.06
35.94
35.84
36.99
35.35
34.93
34.12
34.69
34.31
MASS
(mg)
126
1640
1706
1301
16466
1803
1802
1797
8206
7565
24102
22860
14261
22646
TOTAL (first 9 days) 342^4"
AVERAGE 261
STANDARD DEVIATION 250
REL. STD. DEV. (%) 96
VARIANCE 62728
EFFLUENT
SAMPLE ID
8BR-EFF-01-1-TRPH
SBR-EFF-02-1-TRPH
SBR-EFF-03-1-TRPH
SBR-EFF-04-1-TRPH
8BR-EFF-06-1-TRPH
SBR-EFF-06-1-TRPH
SBR-EFF-07-1-TRPH
SBR-EFF-08-1-TRPH
SBR-EFF-08-6-TRPH
SBR-EFF-09-1-TRPH
SBR-EFF-10-1-TRPH
SBR-EFF-11-1-TRPH
8BR-EFF-12-1-TRPH
SBR-EFF-13-1-TRPH
SBR-EFF-14-1-TRPH
TRPH
(mg/L)
3
448
60
160
114
60
60
60
60
418
630
700
690
660
840
(last 9 days)
QUALIFIER
<6X
<6X
ND
<6X
<6X
ND
ND
ND
ND
<5X
DETECTOr,
LIMIT
(mg/L)
2
100
100
100
100
100
100
100
100
100
100
100
100
100
100
VOLUME
(L)
31.18
33.64
33.70
35.06
44.69
26.86
36.13
34.87
34.87
35.22
34.82
34.17
38.66
31.49
36.26
MASS
(mg)
98
16071
1686
6610
6083
1293
1807
1743
14722
18465
23919
26676
20469
29618
138701
343
287
64
82565
MASS CLOSURE (%) - first 9 days compared to last 9 days _306 M
+ NOT INCLUDED IN STATISTICAL CALCULATIONS BECAUSE REPORTED VALUE WAS OUTLIER
QUALIFIERS: <5X - lass than 5 times the practical quantltatlon limit (PQL)
ND -Not Detected
Ul
Ul
FILENAME: TB4-69
-------
TABLE 4-64. S8R CHEMICAL OXYGEN DEMAND
INFLUENT
DAY
01
02
03
04
05
00
07
08
0»
09
10
11
12
13
14
SAMPLE ID
S8R-IN-01-1-COD
SBR-IN-02-1-COD
S8R-IN-03-1-COD
S8R-IN-04-1-COD
S8R-IN-05-1-COD
SBR-IN-06-1-COD
SBR-IN-07-1-COD
S8R-IN-08-1-COD
SBR-IN-08-6-COD
SBR-IN-09-1-COD
SBR-IN-10-1-COD
SBR-IN-11-1-COD
SBR-IN-12-1-COD
SBR-IN-13-1-COD
SBR-IN-14-1-COD
COD
(mg/lL)
2300
1500
MO
1100
1100
3300
2000
1900
2700
2100
2300
4200
2400
1600
2200
AVERAGE 2113
STANDARD DEVIATION 828
RELATIVE STD. DEV. (%) 39
VARIANCE 685,000
EFFLUENT
SAMPLE ID
S8R-EFF-01-1-COD
SBR-EFF-02-1-COD
SBR-EFF-03-1-COD
SBR-EFF-04-1-COD
SBR-EFF-05-1-COD
SBR-EFF-06-1-COD
SBR-EFF-07-1-COD
SBR-EFF-08-1-COD
SBR-EFF-08-6-COD
SBR-EFF-09-1-COD
SBR-EFF-10-1-COD
SBR-EFF-11-1-COD
SBR-EFF-12-1-COD
SBR-EFF-13-1-COD
SBR-EFF-14-1-COD
COD
, (m0/U
1800
1600
1300
1700
1500
2800
2000
2400
2300
2700
1800
2700
1700
2000
2500
2053
465
23
216,000
FILENAME: TBL4-58
156
-------
Total Residue or Total Solids -- Table 4-65 presents the influent and effluent
solids results. Total sample weight was calculated by multiplying the sample
volume (50 ml) by the appropriate density. The density was assumed to be the
average of the two daily measurements shown in Table 4-45. Each 50 ml sample was
separated into two components by filtration; a filtercake, which was
predominantly but not completely solid, and a filtrate, which was predominantly
but not completely water. The laboratory measured the weights and percent solids
of the filtercake. For the filtrate, solids were measured as total suspended
solids (TSS), which amounted to <0.5% of the solid in all samples. These
measurements were then used to calculate the total percent solids in the original
sample as follows:
Filtercake
o solid wt. = (filtercake wt) x (solid mass fraction)
Filtrate
o filtrate wt — sample wt - filtercake wt.
o solid wt = (TSS) x (filtrate wt)/(lgm/cc x lOOOcc/L x lOOOmg/gm)
Total Sample
o solids wt = filtercake solid wt + filtrate solid wt
o % solids - (solids wt/total sample wt) x 100
The total solids mass balance is shown in Table 4-66. For the influent stream,
the solids weight percent for day 1 was an outlier point. This could have been
due to the fact that influent was stored in a tank that allowed solids to settle
to the bottom resulting in an unrepresentative first sample. Data for day 5 were
not available so the average value calculated for the other days was used. All
statistical parameters were determined without using these two days. The solids
mass was calculated by multiplying the mass by the corresponding solids weight
fraction.
The average influent solids weight percent is 10.5 with a RSD of 28% and
the average effluent solids weight percent is 10.8 with a RSD of 30%. This would
indicate that the solids content was unaffected by the SBR, which is not
surprising since the solids are inert clay and silt. The large day-to-day
variations in the solids weight percent for both influent and effluent do,
however, indicate that there is poor mixing in the storage tank. The solids mass
closure between influent and effluent was good with a discrepancy of about 6%.
Total Organic Halides (TOX) -- Table 4-67 shows the influent and effluent TOX
concentrations. The TOX concentrations, particularly for the effluent, show a
great deal of variability, based on the limited data. Although there appears to
be a reduction in TOX across the SBR in terms of both concentration and mass, it
is difficult to say what the level of reduction is, given the wide variability
of the effluent concentrations, questions about detection limits and matrix spike
recoveries, and operating difficulties in the first five days of testing and the
resultant lack of data. No conclusions can be drawn.
157
-------
TABLE 4-65. TOTAL SOUDS
Sample-ID
SBR-IN-01-1
-02-1
-03-1
-04-1
-06-1
-07-1
-08-1
-08-5
-09-1
-10-1
-11-1
-12-1
-13-1
-14-1
SBR-EFF-01-1
-02-1
-03-1
-04-1
-05-1
-06-1
-07-1
-08-1
-08-5
-09-1
-10-1
-11-1
-12-1
-13-1
-14-1
TOTAL SAMPLE
(50ml)
Density
(gm/ml)
1.050
1.022
1.018
1.020
1.014
1.020
1.019
1.019
1.022
1.019
1.014
1.020
1.033
1.039
1.024
1.039
1.031
0.999
1.039
1.029
1.026
1.028
1.028
1.029
1.025
1.024
1.022
1.025
1.023
Wgt.
(am)
52.5
51.1
50.9
51.0
50.7
51.0
50.9
50.9
51.1
50.9
50.7
51.0
51.7
51.9
51.2
51.9
51.6
49.9
51.9
51.4
51.3
51.4
51.4
51.4
51.2
51.2
51.1
51.2
51.1
RLTERCAKE
Wgt.
(9m)
30.40
10.66
14.49
9.85
6.78
7.38
15.40
15.30
14.70
6.57
11.40
9.20
13.10
9.70
24.90
11.47
8.40
13.67
6.78
16.90
6.46
10.90
10.00
15.10
14.80
12.00
12.47
11.00
7.56
%
Solids
50
48
46
53
46
50
46
48
48
44
42
51
53
50
33
50
51
47
44
45
43
53
39
47
44
52
51
49
46
Solid Wgt.
fem)
15.20
5.12
6.67
5.22
3.12
3.69
7.08
7.34
7.06
2.89
4.79
4.69
6.94
4.85
8.22
5.74
4.28
6.42
2.98
7.60
2.78
5.78
3.90
7.10
6.51
6.24
6.36
5.39
3.48
FILTRATE
Wgt.
(am)
22.10
40.44
36.39
41.15
43.92
43.60
35.53
35.63
36.38
44.38
39.30
41.78
38.55
42.25
26.28
40.46
43.15
36.26
45.17
34.53
44.82
40.50
41.40
36.32
36.43
39.20
38.61
40.23
43.57
TSS
(mg/L)
155
102
78
*
103
100
136
127
*
*
*
. *
*
*
678
575
571
470
420
434
413
479
482
*
*
400
420
*
ft
Solid Wgt.
(am)
0.0034
0.0041
0.0028
0.0045
0.0044
0.0048
0.0045
— —
0.0178
0.0233
0.0246
0.0170
0.0190
0.0150
0.0185
0.0194
0.0200
0.0157
0.0162
—. —
TOTAL SAMF1£
Solids Wgt.
(am)
15.20
5.12
6.67
3.12
3.69
7.09
7.35
— —
8.23
5.76
4.31
6.44
3.00
7.62
2.80
5.80
3.92
6.26
6.38
%
Solids
28.96
10.02
13.11
10.24 +
6.16
7.25
13.92
14.43
13.81 +
5.67 +
9.44 +
9.20 +
13.44 +
9.34 +
16.09
11.09
8.36
12.90
5.78
14.82
5.45
11.28
7.63
13.80 +
12.71 +
12.22
12.48
10.52 +
6.80 +
FILENAME: TBL4-61
No TSS data for these samples.
% solids does not include TSS from the filtrate.
158
-------
TABLE 4-66. SBR TOTAL SOLIDS BALANCES
INFLUENT
SAMPLE ID
SBR-IN-01-1-TSS
SBR-IN-02-1-TSS
SBR-IN-03-1-TSS
SBR-IN-04-1-TSS
SBR-IN-05-1-TSS
SBR-IN-06-1-TSS
SBR-IN-07-1-TSS
SBR-IN-08-1-TSS
SBR-IN-08-5-TSS
SBR-IN-09-1-TSS
SBR-IN-10-1-TSS
SBR-IN-11-1-TSS
SBR-IN-12-1-TSS
SBR-IN-13-1-TSS
SBR-IN-14-1-TSS
TOTAL OF 14 DAYS
SAMPLE
MASS
(gm)
31730
34200
34710
36650
36710
36570
36770
36590
36590
36710
. 35990
35540
35030
35690
35590
SOLIDS
WEIGHT
%
29.0
10.0
13.1
10.2
10.5
6.2
7.3
13.9
14.4
13.8
5.7
9.4
9.2
13.4
9.3
SOLIDS
MASS
(gm)
9189
3427
4550
3753
3840 +
2253 *
2666
5187
5070
2041
3355
3223
4797
3324
56673
AVERAGE 10.46
STD. DEV. - 2.92
REL STD. DEV. % 27.89
VARIANCE 8.52
SAMPLE ID
SBR-EFF-01-1-TSS
SBR-EFF-02-1-TSS
SBR-EFF-03-1-TSS
SBR-EFF-04-1-TSS
SBR-EFF-05-1-TSS
SBR-EFF-06-1-TSS
SBR-EFF-07-1-TSS
SBR-EFF-08-1-TSS
SBR-EFF-08-5-TSS
SBR-EFF-09-1-TSS
SBR-EFF-10-1-TSS
SBR-EFF-11-1-TSS
SBR-EFF-12-1-TSS
SBR-EFF-13-1-TSS
SBR-EFF-14-1-TSS
SAMPLE
MASS
(gm)
32050
34850
34690
35060
46310
26590
37100
35830
35830
36200
35670
34950
39550
32240
36160
SOLIDS
WEIGHT
%
16.1
li.1
8.4
12.9
5.8
14.8
5.5
11.3
7.6
13.8
12.7
12.2
12.5
10.5
6.8
SOLIDS
MASS
(gm)
5157
3865
2900
4523
2677
3941
2022
3388 *
4996
4534
4271
4936
3392
2459
" 53058
10.80
3.19
29.51
10.15
MASS CLOSURE % 6'38
FILENAME: TBL4-62
* Average of split sample measurement
+ Solids Weight % assumed to be average value
159
-------
TABLE 4-«7. TOTAL ORGANIC HALIDES MATERIAL BALANCE
~ ~~ INFLUENT ~
DAY
05
06
08
08
19
SAMPLE ID
SBR-IN-05-1-1
rox
SBR-IN-06-1-TOX
AVERAGE
TOX
(mg/L)
0.97000
1.29999
1.20000
QUALIFIER
Q
DETECTION
LIMIT
(mg/L)
0.1000
0.1000
0.1000
VOLUME
(L)
36.01
36.05
36.06
TOX
MASS
fag)
34.93
46.86
43.27
108.12 125.0?
1.16
STANDARD DEVIATION 0.14
RELATIVE STD. DEV. {%) 11.94 2803
VARIANCE
0.02
(%)
rrr • — — . 1
EFFLUENT ~~~~ ~
SAMPLE ID
SBR-EFF-08-1-TOX
SBR-EFF-08-5-TOX
SBR-EFF-10-1-TOX
SBR-EFF-12-1-TOX
TOX
(mg/L)
0.74000
0.11999
1.39999
0.40000
QUALIFIER
<6X
DETECTION
LIMIT
(mg/L)
0.1000
0.1000
0.0100
0.0100
SAMPLE
VOLUME
(U)
34.87
34.87
34.82
38.66
TOX
MASS
(mg)
25.80
4.18 "
48.75
15.46
108.35 90.0154
0 66
0 48
71 85
0.23
L_ 28
CTl
O
FOOTNOTE: Q denotes spike recoveries outside of control limits
* Not accounted for in mass balance
< 5X = less than 5 times practical quantitation limit (PQL)
-------
Chloride Ion (CD -- Table 4-68 presents the influent and effluent chloride ion
data. A significant increase in both concentration and mass is observed. As
with the TOX data, the variations in influent concentration are far lower (-3%)
than for effluent concentration (-43%). The average chloride mass almost doubles
between influent and effluent. Comparing the chloride results with the total
organic halide results, it can be seen that a reduction in TOX corresponds to an
increase in chloride. Although this is not a 1:1 correspondence, it could be
presumed that it is the result of converting relatively toxic chlorophenols to
the more benign inorganic form. What is not clear is the exact magnitude of the
conversion. This dechlorination process is verified by the reduction in penta
concentration discussed earlier, but it cannot be concluded from the chloride
data that total mineralization of penta takes place.
Other Metals - - Concentrations of other metals in the SBR effluent are shown in
Table 4-69. Non-detected values for particular metals were not included in the
calculation of statistical parameters. No cadmium or selenium was detected,, The
average barium concentration was 18 ppm with a deviation of 11%; as noted in
earlier sections, the source of the barium is unknown. The average lead
concentration was 10 ppm with a deviation of 11%. The average mercury
concentration was 8.6 ppb with a deviation of 63%. The average silver
concentration was 180 ppb with a deviation of 23%.
Since concentrations of these other metals in the influent stream were
not measured, it is impossible to state whether the SBR affected them in any way.
EP Toxicity -- From the limited amount of EP Toxicity data presented in Table 4-
70 it appears that very little cadmium, lead, mercury, selenium or silver leached
from the effluent stream. The average arsenic concentration was 1.2 mg/L with
a deviation of 32%. Barium showed the largest variation with a deviation of 60%
and an average concentration of 117 ug/L. Average chromium concentration was 30
ug/L with a deviation of 10%. Average copper concentration was 79 ug/L with a
deviation of 9%.
4.4.4.3 System Parameters
No penta or PAHs were detected on the carbon from the adsorption
canister. This would indicate that very little, if any, of the target semi-
volatile organic compounds escape from the unit as vapor.
pH measurements over the course of the test are shown in Table 4-71 and
Figure 4-38. The first readings on day four are suspect. The influent pH was
less than that for the effluent pH; presumably due to the addition of caustic.
(The reasonably stable pH suggests that pH fluctuations is not a major cause of
variations in penta concentrations.)
Power usage on a day-by-day basis is shown in Table 4-72 and plotted in
Figure 4-39. Over the 14 day test period, 16.7 kw-hr of electricity was
consumed. On this basis, the average daily consumption is approximately 1.2 kw-
hr. And, since -490 liters (130 gallons) or 499 Kg (1100 Ib) of slurry were
treated during this period, the electrical usage can be estimated to be
0.034kwh/L (0.13 kwh/gal).
161
-------
TABLE 4-68. CHLORIDE MATERIAL BALANCE
INFLUENT
DAY
04
05
06
08
08
10
12
SAMPLE ID
SBR-IN-04-1-CL
SBR-IN-05-1-CL
SBR-IN-06-1-CL
a
(mg/L)
79.2
83.9
77.5
VOLUME
(L)
36.01
36.05
36.06
a
MASS
(mg)
2851.99
3024.60
2794.65
TOTAL 8671.24
AVERAGE 80.20
STANDARD DEVIATION 2.71
REL STD. DEV. (%) 3.38
VARIANCE 7.33
thR-UENT
SAMPLE ID
SBR-EFF-08-1-CL
SBR-EFF-08-5-CL
SBR-EFF-10-1-CL
SBR-EFF-12-1-CL
a
(mg/L)
79.4
78.3
203.0
194.0
VOLUME
(L)
34.87
34.87
34.82
38.66
a
MASS
(mg)
2749.50 *
7068.46
7500.04
17317.99
138.68
59.91
43.20
3589.31
MASS CLOSURE (%) -99.72
FILENAME: TBL4-64
FOOTNOTE: * Mass calculation based on average of split sample
162
-------
TABLE 4-69. EFFLUENT METALS CONCENTRATION
(ug/l)
Sample-ID
Test Day-Time Period
SBR-EFF-01-1-MET
SBR-EFF-02-1-MET
SBR-EFF-03-1-MET
SBR-EFF-04-1-MET
SBR-EFF-05-1-MET
SBR-EFF-06-1-MET
SBR-EFF-07-1-MET
SBR-EFF-08-1-MET
SBR-EQB-08-5-MET
SBR-EFF-09-1-MET
SBR-EFF-10-1-MET
SBR-EFF-11-1-MET
SBR-EFF-12-1-MET
SBR-EFF-13-1-MET
SBR-EFF-14-1-MET
AVERAGE
Time
Sampled
1015
1100
1100
1100
1045
1100
1115
1100
1100
N/A
1100
1105
1115
1100
1100
Acurex
ID
8910121
8910137
8910144
8910147
8910147
8910157
8910161
8910176
8910176
8910180
8910183
8910187
8910187
8910195
8911001
Barium
12000
18000
20700
19000
19300
19100
19000
20000
17800
18300
18400
17300
16600
16400
17600
Cadmium
<50
<50
<50
<50
<50
<50
<50
<50
<50
<50
<50
<50
<50
<50
<50
Lead
6080
9350
11100
10500
10500
10400
10400
10600
10200
10000
9930
10100
9720
9310
9510
Mercury
5.7
8.3
5.0
<0.2
7.5
9.4
<2.0
8.6
7.9
5.3
8.6
18.2
3.1
22.2
2.4
Selenium
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
Silver
<100
170
180
250
130
210
140
<100
<100
<100
<100
<100
<100
<100
<100
17967 9847 8.6 0
STD. DEV. 1960 1116 5.4 0
REL STD. DEV. (%) 11 11 62.7
VARIANCE - 3840000 1245000 29.3 0
N/A - Not Available
FILENAME: TBL4-65
163
-------
TABLE 4-70. EFFLUENT E P TOXICITY ANALYSIS
(ug/0
Test Day-Time Period
SBR-EFF-08-1-MET
SBR-EFF-10-1-MET
SBR-EFF-12-1-MET
Sampled
1100
1100
1115
ID
8910176
8910183
8910187
Arsenic
929
930
1760
Barium
215
82
54.9
Cadmium
< 5.0
< 5.0
< 5.0
Chromium
29.5
27.2
34.2
Copper
71
88.8
76.8
Lead
< 50
< 50
< 50
AVtHAlat 1206 117 30 79
Mercury
0.23
<0.2
<0.2
Selenium
<200
<200
<200
Silver
< 10
< 10
< 10
STD.DEV. 392 70 37
RELSTD.DEV.(%) 32 eo 10 9
VARIANCE 1 53000 4900 8 55
CTi
FILENAME: TBL4-66
-------
TABLE 4-71. SBRpHDATA
DAY
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
* suspect reading
DATE
10/17/89
10/18/89
10/18/89
10/19/89
10/19/89
10/20/89
10/20/89
10/21/89
10/21/89
10/22/89
10/22/89
10/23/89
10/23/89
10/24/89
10/24/89
10/25/89
10/25/89
10/26/89
10/26/89
10/27/89
10/27/89
10/28/89
10/28/89
10/29/89
10/29/89
10/30/89
10/30/89
10/31/89
10/31/89
PH
TIME INFLUENT EFFLUENT
14:08
02:10
16:00
01:45
13:59
01:37
14:04 *
01:30
13:55
01:55
13:54
02:00
13:52
01:47
13:59
01:45
13:48
01:35
13:50
01:15
14:16
01:30
14:09
01:20
13:58
01:20
14:08
01:10
09:47
6.67
6.15
6.31
6.19
6.39
6.01
6.91 *
6.33
6.26
6.06
6.54
6.12
6.35
5.82
6.34
5.90
6.30
6.13
6.31
5.86
6.23
5.95
6.17
6.29
6.21
6.38
6.44
6.53
6.45
7.27
7.10
7.08
7.29
7.19
7.19
7.99
6.97
6.96
6.73
7.37
6.99
6.96
6.85
6.91
6.91
6.96
6.94
7.06
6.86
6.74
6.54
6.60
6.58
6.42
6.88
6.43
6.37
6.23
t-LAnSED
TIME
(hr)
0.00
12.03
25.87
35.62
47.85
59.48
71.93
83.37
95.78
107.78
119.77
131.87
143.73
155.65
167.85
179.62
191.67
203.45
215.70
227.12
240.13
251.37
264.02
275.20
287.83
299.20
312.00
323.03
331.65
FILENAME: TBL4-67
165
-------
Figure 4-38. Slurry Bio-Reactor
5.5 -
Influent & Effluent pH Data
100
150
200
250
300
350
Elapsed Time (hr)
D Influent pH O Effluent pH
-------
TABLE 4-72. SBR POWER USAGE
DAY
1
1
1
1
1
2
2
2
2
2
2
2
2
3
3
3
4
4
4
5
5
5
5
5
6
6
6
,7
7
7
8
8
8
8
8
DATE
10/17
10/17
10/17
10/17
10/18
10/18
10/18
10/18
10/18
10/18
10/18
10/18
10/10
10/19
10/19
10/20
10/20
10/20
10/20
10/21
10/21
10/21
10/22
10/22
10/22
10/22
10/23
10/23
10/23
10/24
10/24
10/24
10/24
10/25
10/25
TIME
12:28
17:10
20:10
21:00
02:06
10:40
12:20
12:50
14:55
16:03
17:00
20:45
06:00
10:40
21:35
01:47
10:22
13:10
21:00
14:12
17:00
19:50
01:55
05:30
17:10
19:35
02:28
13:35
17:37
01:48
12:45
17:15
20:15
01:49
05:14
ELAPSED
TIME
(HR)
0.0
4.7
7.7
8.5
13.6
22.2
23.9
24.4
26.5
27.6
28.5
32.3
41.5
46.2
57.1
61.3
69.9
72.7
80.5
97.7
100.5
103.4
109.4
113.0
124.7
127.1
134.0
145.1
149.1
157.3
168.3
172.8
175.8
181.3
184.8
METER
READING
(KW-HR)
59.3
59.5
59.7
59.8
60.1
60.5
60.6
60.6
60.7
60.8
60.8
61.0
61.6
61.8
62.3
62.6
63.1
63.3
63.6
64.5
64.6
64.7
65.05
65.47
65.8
65.9
66.2
66.8
66.9
67.2
67.8
68.0
68.1
68.4
68.5
POWER.
USAGE
(KW-HR)
0.2
0.2
0.1
0.3
0.4
0.1
0
0.1
0.1
0
0.2
0.6
0.2
0.5
0.3
0.5
0.2
0.3
0.9
0.1
0.1
0.35
0.42
0.33
0.1
0.3
0.6
0.1
0.3
0.6
0.2
0.1
0.3
0.1
DAY DATE
9 10/25
9 10/26
9 10/26
9 10/26
10 10/26
10 10/26
10 10/26
10 10/27
10 10/27
10 10/27
11 10/27
11 10/27
11 10/27
11 10/28
11 10/28
11 10/28
12 10/28
12 10/28
12 10/28
12 10/29
12 10/29
12 10/29
13 10/29
13 10/29
13 10/29
13 10/30
13 10/30
13 10/30
14 10/30
14 10/30
14 10/30
14 10/31
14 10/31
14 10/31
15 10/31
TOTAL
AVERAGE
STD. DEV.
TIME
18:25
01:35
05:20
09:15
13:53
18:00
21:39
01:15
05:43
09:06
14:16
18:03
21:38
01:30
06:15
09:14
14:09
18:07
21:40
01:25
06:02
09:30
14:07
18:04
21:34
01:17
05:00
09:00
14:13
18:00
21:32
01:18
05:37
09:00
10:45
ELAPSED
TIME
(HR)
197.9
205.1
208.9
212.8
217.4
221.5
225.2
228.8
233.2
236.6
241.8
245.6
249.2
253.0
257.8
260.8
265.7
269.7
273.2
277.0
281.6
286.0
289.7
293.6
297.1
300.8
304.5
308.5
313.7
317.5
321.1
324.8
329.1
332.5
334.3
MtfbH
READING
(KW-HR)
69.08
69.45
69.6
69.8
69.96
70.1
70.25
70.5
70.6
70.8
71.0
71.2
71.4
71.5
71.5
71.8
72.2
72.4
72.6
72.8
73.0
73.2
73.37
73.56
73.75
74.0
74.25
74.4
74.7
74.92
75.1
75.35
75.63
75.8
75.99
REL. STD. DEV. (%)
VAR.
POWER
USAGE
(KW-HR)
0.58
0.37
0.15
0.2
0.16
0.14
0.15
0.25
0.1
0.2
0,2
0.18
0.17
0.15
0
0.3
0.42
0.18
0.18
0.17
0.25
0.2
0.17
0.19
0.19
0.25
0.25
0.15
0.3
0.22
0.18
0.25
0.28
0.17
0.19
16.69
0.24
0.16
65.43
0.03
FILENAME: TBL4-68
167
-------
Figure 4-39. Slurry Bio-Reactor
POWER USAGE
MO)
D
L.
OJ
5
0
d
0
0
50
100
150
200
300
350
Elapsed Time (hr)
-------
4.4.5 Dloxins/Furans
4.4.5.1 Introduction
Removal of chlorinated dioxins and furans was not a stated objective of
the vendor; rather the data were collected to guide the project team in
determining disposal for waste materials from the project. To determine the
contaminant reductions during the demonstration, the distribution of various CDD
and CDF congeners was measured for each technology. An operational term, 'Total
CDD/CDFs', is used to describe the sum of all CDD (tetra-, penta-, hexa-, hepta-,
octa-) and CDF (tetra-, penta-, hexa-, hepta-, octa-) congeners and isomers.
This term, Total CDD/CDFs, is then used to estimate the changes between inptit and
output streams for each technology.
4.4.5.2 CDDs/CDFs in Soil Washer Streams
The presence of CDDs in the Feed Soil and all output streams from the
BioTrol Soil Washing System was measured, including:
2,3,7,8-tetrachloro-dibenzo(p)dioxin (2,3,7,8-TCDD),
other tetrachlorodibenzodioxin congeners (TCDD),
pentachlorodibenzodioxins(PeCDD),
hexachlorodibenzodioxins (HxCDD),
heptachlorodibenzodioxins (HeCDD),
octachlorodibenzodioxins (OCDD),
and the following CDFs:
tetrachlorodibenzofurans(TCDF),
pentachlorodibenzofurans (PeCDF),
hexachlorodibenzofurans (HxCDF),
heptachlorodibenzofurans (HpCDF)
and octachlorodibenzofurans (OCDF)
The number of samples analyzed are limited, usually to two for a particular
stream, and the average results for the Total CDD/CDFs in all streams in the Low
Penta SW test and the High Penta SW test, respectively, are summarized in Table
4-73 and 4-74 . Concentrations of individual congeners in each stream from each
test are also presented in Table 4-75.
Comparison of various stream profiles for CDD/CDF distribution reveals the
following:
(a) OCDD is the most significant among the CDD/CDF congeners present in
all the stream samples, accounting for about 61-66% of the Total
CDD/CDFs, followed by HpCDD (13.2 - 13.5%), HpCDF (8.4 - 11.8%),
OCDF (7.8 - 11.6%), HxCDF (2.1 -4.7%), and HxCDD (0.8 - 1.0%). TCDD,
PeCDD, TcCDF and the congener usually of primary interest, 2,3,7,8-
TCDD were not detected.
(b) A comparison of the concentrations of individual CDDs and CDFs in
various process stream samples reveals that the pattern of the
169
-------
TABLE 4-73. TOTAL CDD/CDF DATA - LOW SOIL TEST *
Soil Washer
Streams
INPUT STREAMS
Feed Soil
Municipal Water
ATS effluent
Thickener Stream
TOTAL
OUTPUT STREAMS
Washed Soil
Fine Oversize
Coarse Oversize
Fine Particle Cake
Combined Dewater.Eff
TOTAL
Rate
kg/hr.
220.0
1200.0
N/A
180.0
1600.0
260.0
13.0
32.0
48.0
1148.0
1501.0
Input or
Output/kg
Feed Soil
1.00
5.45
—
0.82
7.27
1.18
0.06
0.15
0.22
5.22
6.82
Cone.
(ppm)
1.370
0.000
-
0.000
0.110
1.820
1.040
3.130
0.002
Mass/Fraction
(mg/kg)
1.37
1.37
0.13
0.11
0.15
0.68
0.01
1.08
Percent of
Output
100
100
12.0
9.9
14.0
63.1
1.0
100.0
file: tab!469
* CDD/CDF average concentrations represents mean of at least duplicate samples.
TABLE 4-74. TOTAL CDD/CDF DATA - HIGH SOIL TEST *
Soil Washer
Streams
INPUT STREAMS
Feed Soil
Municipal Water
ATS effluent
Thickener Stream
TOTAL
OUTPUT STREAMS
Washed Soil
Fine Oversize
Coarse Oversize
Fine Particle Cake
Combined Dewater.Eff
TOTAL
Rate
kg/hr.
160.0
36.0
660.0
120.0
976.0
210.0
9.1
28.0
36.0
720.0
1003.1
Input or
Output/kg
Feed Soil
1.00
0.23
4.13
0.75
6.10
1.31
0.06
0.18
0.23
4.50
6.27
Cone.
(ppm)
2.508
0.000
0.000
0.000
2.508
0.078
1.235
2.319
6.818
0.008
Mass/Fraction
(mg/kg)
2.51
2.51
0.10
0.07
0.41
1.53
0.04
2.15
Percent of
Output
100
100
4.8
3.3
18.9
71.4
1.7
100.0
file: tab!470
CDD/CDF average concentrations represents mean of at least duplicate samples.
• 170
-------
TABLE 4-75. SOIL WASHER: CDD/CDF CONCENTRATIONS IN ALL STREAMS
Soil
Test
1. Low Soil Test
Input
Feed Soil*
% of total input
Municipal Water $
Flocculant $
ATW$
Output
Washed Soil **
Coarse Oversize
Rne Oversize
Fine Particle Cake
CDE
% of Total CDD/CDFs output
II. High Soil Test
Input
Feed Soil #
% of Total CDD/CDFs input
Municipal Water $
Flocculant $
ATW$
Output
Washed Soil *
Coarse Oversize
Fine Oversize
Fine Particle Cake
CDE
46 of Total CDD/CDFs output
CDD/CDF Congeners Concentration @
2378-
TCDD
0.00
0.0
-
-
-
0.00
0.00
0.00
0.00
0.00
0.0
0.00
0.0
-
-
-
0.00
0.00
0.00
0.00
0.00
0.0
TCDD PeCDD
0.00
0.0
-
-
-
0.00
0.00
0.00
0.00
0.00
0.0
0.00
0.0
-
-
-
0.00
0.00
0.00
0.00
0.00
0.0
0.00
0.0
-
-
-
0.00
0.00
0.00
0.00
0.00
0.0
0.00
0.0
-
-
-
0.00
0.00
0.00
0.00
0.00
0.0
HxCDD
13.00
1.0
-
-
-
1.00
7.60
14.00
27.30
0.01
0.8
21.00
0.8
-
-
-
0.47
18.00
6.60
53.50
0.05
0.8
HpCDD
180.00
13.2
•-
-
-
11.60
125.00
230.00
455.00
0.10
13.5
330.00
13.2
-
-
-
10.00
315.00
171.50
900.00
1.05
13.4
OCDD
830.00
60.8
-
-
'-
71.02
670.00
1150.00
1850.00
1.16
61.3
1600.00
63.8
-
-
-
53.44
1505.00
805.00
4500.00
5.50
65.7
TCDF PeCDF
0.00
0.0
-
-
-
0.01
0.03
0.04
0.10
0.00
0.0
0.00
0.0
-
-
-
0.00
0.00
0.00
0.00
0.00
0.0
3.20
0.2
-
-
-
0.24
3.55
6.70
2.05
0.00
0.2
2.10
0.1
-
-
-
0.05
2.35
2.05
0.55
0.01
0.1
HxCDF
58.00
4.3
- .
-
-
4.62
30.50
74.00
167.50
0.08
4.7
53.00
2.1
-
-
-
1.67
45.50
32.50
140.00
0.18
2.1
HpCDF
160.00
11.7
-
-
-
11.60
115.50
100.00
405.00
0.10
11.8
212.00
8.5
-
-
-
6.00
103.00
106.50
575.00
0.66
8.4
OCDF
120.00
8.8
-
-
-
8.80
82.00
160.00
222.50
0.16
7.8
200.00
11.6
-
-
-
6.43
240.50
110.50
640.00
0.73
0.5
TOTAL
ODD
1023.00
74.8
-
-
-
83.62
802.60
1394.00
2332.30
1.36
75.5
1851.00
77.8
-
-
-
63.81
1838.00
883.10
5453.50
6.60
78.8
TOTAL
CDF
342.20
25.1
-
-
-
25.27
240.58
430.74
787.24
0.42
24.5
557.10
22.2
-
-
-
14.15
481.35
251.55
1364.55
1.57
20.2
TOTAL
CDD/CDF
1365.20
100.0
-
-
-
108.80
1043.18
1824.74
3120.54
1.78
100.0
2508.10
100.0
-
-
-
78.06
2310.35
1234.65
6818.05
8.17
100.0
Unless otherwise stated, all concentrations in ug/L or ug/kg and represent duplicates.
# Data from a single sample
* Mean of seven samples
* * Mean of five samples
$ Not analyzed for CDD/CDFs
-------
relative proportions of all ODD and CDF congeners tested is uniform
for all streams and, largely, for both the Low Penta and the High
Penta tests. Although of different magnitude, the patterns for the
various CDDs and CDFs are superimposable where measureable.
(c) Since the congener pattern for all input and output streams is
superimposable for all 11 CDDs and CDFs, estimation of process
efficiency on the basis of the major dioxin congener, OCDD, may
adequately describe process characteristics and performance.
Low Penta SW Test -- The levels and pattern of distribution of CDDs and CDFs in
the Low Penta SW process streams indicate that OCDD is the most significant of
the CDD/CDFs in the soil samples and represents about 61% of the total CDD/CDF
mass in all the process streams. This is followed by HpCDD (13.5%), HpCDF
(11.8%), OCDF (7.8%), HxCDF (4.7%), HxCDD (0.8%), and PeCDF (0.2%). The
distribution of all isomers is presented in Table 4-75.
The Total CDD/CDF concentration in the Washed Soil (0.11 mg/kg)
represents 8% of the Total CDD/CDF concentration in the Feed Soil (1.365 mg/kg),
equivalent to a % Removal Efficiency of 92% of the input CDD/CDFs. While the
Total CDD/CDFs in the Washed Soil represents only 12% of the cumulative Total
CDD/CDF mass in all outputs (Table 4-73), it must be recognized that there is a
significant discrepancy between the calculated input mass (1.37 mg/kg of Feed
Soil) and the combined output mass (1.08 mg) of Total CDD/CDFs (mass balance:
79%).
The Coarse Oversize fraction, which contributes approximately 2.1% of the
total output mass, contains 13.9% of the Total CDD/CDF output mass in the Low
Penta SW test. The Fine Oversize (8.7% of the output mass) contains about 10.2%
of the Total CDD/CDF mass. The Fine Particle Cake, containing the highest
concentration of Total CDD/CDFs in 3.2% of the output mass, also contributes the
largest portion to the output mass of Total CDD/CDFs, 63%.
The Combined Dewatering Effluent, CDE, is the largest of the stream
outputs, with 76% of the total output mass, but accounts for only 0.9% of the
Total CDD/CDF output mass. This is probably attributable to the low solubility
of the CDD/CDFs in water and, in fact, the amount of dioxins/ furans present may
only be due to suspended solids in the water.
High Penta SW Test -- In the High Penta SW test, an exceptionally high 97%
Removal Efficiency is calculated based on the Total CDD/CDF concentration in the
Feed Soil (2.508 mg/kg) and that in the Washed Soil (0.078 mg/kg). The Washed
Soil, representing 20.9% of the total output mass, contains only 4.6% of the
output mass of Total CDD/CDFs and 4% of the input CDD/CDFs mass (Table 4-74) . In
this test, the Total CDD/CDF mass input and output are in closer agreement, with
a mass balance of 86%.
The Coarse Oversize, (2.9% of output mass) contains 19% of the Total
CDD/CDF output mass and the Fine Oversize (0.96% of output mass) contains 3.2%
of the output mass of Total CDD/CDF. Combined they account for 19.1% of the Total
CDD/CDF input mass. The Fine Particle Cake, representing only 3.7% of the output
172
-------
mass, contains 71% of the output Total GDD/CDF mass and 61% of the input Total
CDD/CDF mass.
As expected, the CDE process stream contained insignificant concentrations
of CDD/CDFs (total: 8.2 jig/L) . Although CDE mass is 72% of the total output mass
per period, this stream contains only 1.9% of the Total CDD/CDF output mass and
1.6% of the input mass from the Feed Soil. As noted above, this probably is due
to the hydrophobic character of CDD/CDFs and CDD/CDFs in the CDE process stream
also may be due to suspended particulates with adsorbed CDDs/CDFs. Although the
CDE masses are comparable in both the High Penta (72%) and Low Penta (76%) tests,
the CDE from the High Penta test contained almost twice as much, 1.8%, of the
Total CDD/CDF output; however, this observation is based on very limited data and
may be due to suspended solids carryover, differences in the soil composition,
CDD/CDF sources, analytical precision, etc.
Materials/Mass/Balance -- Mass balances were carried out for both the Low Penta
and the High Penta SW tests. Based on averaged input and output flows, it is
evident from the totals in Tables 4-73 and 4-74 that satisfactory material
balances (6% and -3% for the Low and High Penta tests, respectively) were
achieved. For Total CDDs/CDFs, the mass balances were 79% and 86%, respectively.
However, the BATS effluent (ATS) was not analyzed for CDD/CDFs. This could
significantly affect the calculations for the High Penta -SW test since BATS
Effluent represents 81% of the liquid input and 68% of the total input mass.
4.4.5.3 CDD/CDFs in the BioTrol Aqueous Treatment System
CDD/CDF concentrations in the BATS effluent samples are as shown in Table
4-76. The mean daily Total CDD/CDF concentration in the BATS effluent stream was
194.8 ng/L (range: 12-503 ng/L) . The effluent Total CDD levels ranged from 12-
403 jig/L (mean: 140 Hg/L) . Among the CDD congeners, only HxCDD and OCDD were
detected in the effluent streams; OCDD was detected in all the effluent samples
from BATS. Likewise, the mean daily effluent output of CDF congeners was 54.84
Hg/L (range: 0-99.8 ng/L). Penta-, hexa-, hepta-, and octachlorodibenzofurans
were detected in at least one effluent sample from the BATS. Note that 2,3,7,8-
TCDD, TCDD, PeCDD, HpCDD, and TCDF, and PeCDF were not detected in any of the
BATS effluent samples. In the absence of CDD/CDF congeners concentrations for
the influent to the BATS reactor, no conclusions concerning removal by the system
are possible.
4.4.5.4 CDD/CDFs in the Slurry Bio-Reactor
The results of the eight analyses for 6 CDD and 5 CDF congeners in the
SBR effluent streams over the course of the fourteen day test with the fine
particle slurry from the High Penta SW test are as shown in Table 4-77. The
average Total CDD/CDF concentration in the SBR effluent samples was 6,300 ng/kg
(range: 0.0-9,272 ng/kg). CDD congeners with a mean total output of 4,960 Hg/kg
(range 0.0-7,260 ng/kg) constitute 78 percent of the total CDD/CDF load in the
173
-------
TABLE 4-76. AQUEOUS TREATMENT SYSTEM: CDD/CDF CONCENTRATIONS IN THE
EFFLUENT STREAMS
CDD/CDF Concentrations *
CDD/CDF
2378-TCDD
TCDD
PeCDD
HpCDD
HxCDD
OCDD
TCDF
PeCDF
HxCDF
HpCDF
OCDF
TOTAL CDD '
TOTAL CDF
TOTAL CDD/CDF
eff-01
0.00
0.00
0.00
0.00
0.00
37.00
0.00
0.00
0.00
0.00
0.00
37.00
0.00
37.00
eff-03
0.00
0.00
0.00
0.00
27.00
180.00
0.00
0.00
8.90
31.00
25.00
207.00
64.90
271.90
eff-05
0.00
0.00
0.00
0.00
0.00
41.00
0.00
0.00
0.00
0.00
0.00
41.00
0.00
41.00
eff-07
0.00
0.00
0.00
0.00
63.00
340.00
0.00
0.00
6.80
42.00
51.00
403.00
99.80
502.80
eff-09
0.00
0.00
0.00
0.00
0.00
12.00
0.00
0.00
0.00
0.00
0.00
12.00
0.00
12.00
Mean
Daily
Output
0.00
0.00
0.00
0.00
18.00
122.00
0.00
0.00
3.14
14.60
15.20
140.00
32.90
172.90
file: 4-71
* Unless otherwise stated, CDD/CDF concentrations data were obtained from a single sample
All data is in ng/L
174
-------
TABLE 4-77. SLURRY BIO-REACTOR: CDD/CDF CONCENTRATIONS IN THE EFFLUENT
STREAMS
CDD/CDF
2378-TCDD
TCDD
PeCDD
HxCDD
HpCDD
OCDD
TCDF
PeCDF
HxCDF
HpCDF
OCDF
TOTAL CDD
TOTAL CDF
TOTAL CDD/CDF
CDD/CDF Concentrations *
eff-01
0.0
0.0
0.0
41.0
780.0
4200.0
0.0
10.0
150.0
580.0
520.0
5021.0
1260.0
6281.0
eff-03
0.0
0.0
0.0
41.0
780.0
4200.0
0.0
11.0
150.0
570.0
540.0
5021.0
1271.0
6292.0
eff-05
0.0
0.0
0.0
50.0
940.0
3900.0
0.0
17.0
170.0
650.0
580.0
4890.0
1417.0
6307.0
eff-07
0.0
0.0
0.0
46.0
900.0
5000.0
0.0
20.0
200.0
760.0
640.0
5946.0
1620.0
7566.0
eff-09#
0.0
0.0
0.0
47.0
900.0
4710.0
0.0
16.5
160.0
715.0
670.0
5657.0
1561.5
721S8.5
eff-11
0.0
0.0
0.0
52.0
940.0
4900.0
0.0
24.0
190.0
770.0
600.0
5892.0
1584.0
7476.0
eff-13
0.0
0.0
0.0
60.0
1100.0
6100.0
0.0
32.0
250.0
1000.0
730.0
7260.0
2012.0
9272.0
eff-14#
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Average
Output
(8 days)
0.00
0.00
0.00
42.13
792.50
4126.25
0.00
16.31
158.75
630.63
535.00
4960.88
1340.69
6301.56
* Unless otherwise stated, CDD/CDF concentration data were obtained from a single sample.
Concentrations of CDD/CDF are in ug/kg.
# Duplicate samples analyzed
175
-------
SBR output streams. Hexa-, hepta-, and octa- CDDs were detected in all effluent
samples except day 14. Among the CDF congeners penta-, hexa- , hepta-, and octa-
CDFs were detected in all the effluent samples. The mean total CDF output was
1,340 ng/kg (range: 0.0-2,012 jig/kg) and represents 22 percent of the Total
CDD/CDF load in the SBR-treated slurry. Data are not available on the slurry of
fines fed to the SBR and the data for the Fine Particle Cake obtained after
dewatering are not a good representation of the slurry; therefore, removal cannot
be evaluated.
4.5 FACTORS AFFECTING THE DEMONSTRATION
4.5.1 Unit Operating Problems - Soil Washer
Several relatively minor and fully expected difficulties were encountered
during the demonstration. These are described below both to document how the
problems were addressed to minimize impact on the study and to alert the
developer and potential users to changes in equipment or its design that may be
called for. The major operating problem encountered during the SW tests was in
the transfer of soil from the feed hopper to the conveyor belt. The feed hopper
was a new feed system used for the first time during this demonstration. Because
of a higher than expected moisture content, the soil was coating the screws in
the feed hopper, making it necessary to increase in the auger rate from 10% to
80% to maintain the same feed rate. This higher auger rate is subject to greater
variability. Attempts were made to aerate the feed soils to decrease their
moisture content and to modify the feed hopper with a vibrating device and
inclined, wooden walls to help keep the augers clear. This enabled the
demonstration tests to be completed. However, in a commercial scale system a
different means of delivering soil to the conveyor belt will be needed.
Minor problems that were encountered included blown fuses, a broken shim
on an attrition machine, the overall centrifuge operation, and failure of various
pumps. In these cases the equipment was either repaired or replaced. In a
commercial facility, back-up equipment or parts would need to be readily
available to avoid shut-down of the system or two Soil Washers might need to be
run in a parallel configuration to allow for the shut-down of a unit for routine
maintenance.
4.5.2 BioTrol Aqueous Treatment System Problems
Operational problems encountered during the demonstration included a
leaking influent pump, a leaking recycle line, worn bearings in the influent
pump, and the overheating of the BATS reactor. Repairs of the pumps and recycle
line were relatively minor but did require the cut-off of feed to the system for
short periods of time. Having replacement pumps on hand would avoid any loss of
feed for more than a few minutes. The overheating of the BATS reactor occurred
due to a major decrease in flow rate (from 3 gpm down to 0.5 gpm) when the
bearings in the influent pump had worn out. This decrease in flow rate should
have been accompanied by a corresponding decrease in the thermostat setting on
the heater but subsequent overheating occurred when the thermostat was not
adjusted. The result was deactivation of the biomass due to extreme temperatures
in the BATS reactor. This occurred on the 10th day of the High Penta
176
-------
demonstration test and sufficient data had already been collected so that the
test'was considered concluded at that time.
4.5.3 Slurry Bio-Reactor Problems
Operating problems encountered during the demonstration of the SBR
included clogging of the lines connecting the cells of the reactor, rupture of
the line in the effluent pump, blown circuit breakers, and ambient temperatures
below freezing. A rubber mallet was used to loosen material caught between
reactor cells; in a larger system this is not expected to be a problem. The line
in the effluent pump was replaced following each occurrence of a rupture. The
tripped circuit breaker was reset as soon as it was discovered, which was
immediately upon occurrence. The feed system was dependent upon a timer which
diverted flow 6 seconds out of each minute. If power were cut off for any
extended period of time, this timer would need to be reprogrammed. Although the
reactor cells are equipped with automatic temperature controls, the feed tank was
not. A propane, heater was used on extremely cold nights inside the temporary
structure to keep the feed from freezing.
4.6 POTENTIAL OVERALL SYSTEM PERFORMANCE
If all units were fully integrated in size, operation of the SW system
would return Washed Soil to the site, segregate Coarse and Fine Oversize, and
biologically treat the Combined Dewatering Effluent in the BATS and the Fine
Particle Cake (as slurry) in the SBR. Using the normalized 1.0 kg of Feed Soil,
and making the following assumptions, the overall fate of soil and penta then can
be estimated.
Assumptions
Solid mass balance is good
Penta mass balance is good
BATS operates at 90% removal efficiency
SBR operates at 90% removal efficiency
Effluent from BATS is recycled
Treated slurry from SBR is returned to site
With a soil such as that used in the Low Penta test, 1.0 kg of Feed Soil,
containing 175 mg of penta (based on the output penta mass rather than the lower
input mass), produces the following:
o 0.06 kg Fine Oversize (containing 5.8 mg penta) and 0.14 kg Coarse
Oversize (containing 23.8 mg penta) to be disposed of off-site, probably
by incineration;
o 1.18 kg of Washed Soil (16.5 mg penta) that it is presumed could be
returned to the site;
o 0.22 kg of Fine Particle Cake (59.4 mg penta) that would be treated in
the Slurry Bio-Reactor to remove 90% of the penta, leaving only 5.9 mg.
It is assumed that the slurry then could be returned to the site.
177
-------
o 5.0 kg of Combined Dewatering Effluent (70 mg penta) that would be
treated in the BATS to achieve a 90% penta reduction to 7.0 mg; the
treated effluent would be recycled.
With a soil such as that used in the High Penta test, 1.0 kg of Feed
Soil, containing 1066 mg of penta (output penta mass rather than the
lower input mass), produces the following:
o 0.06 kg Fine Oversize (54 mg penta) and 0.18 kg Coarse Oversize (252 mg
penta) to be disposed of off-site;
o 1.31 kg of Washed Soil (114 mg penta). At a concentration of 87 mg/kg,
it is questionable whether this material would be suitable for return to
the site without further treatment, even though -90% of the penta has
been removed.;
o 0.22 kg of Fine Particle Cake (1300 mg penta) that would be treated in
the Slurry Bio-Reactor to remove 90% of the penta, leaving 130 mg. The
treated slurry, then containing about 590 mg/kg, probably could not be
returned to the site without further treatment. If SBR treatment
achieved 99% at steady state, the resulting slurry would have only 13 mg
and, consequently, about 59 ppm.
o 4.5 kg of Combined Dewatering Effluent (360 mg penta) that would be
treated in the BATS to achieve a 90% penta reduction to 36.0 mg; the
treated effluent would then be recycled.
Thus, the overall effect is a segregation of the penta (and PAH)
contamination into three relatively small streams: the Coarse and Fine Oversize,
which are disposed of off-site, and the Fine Particle Cake, which is treated in
the SBR, to degrade approximately 90% of its contaminants. The Combined
Dewatering Effluent is treated in the BATS for reuse or is polished further with
carbon before discharge. The bulk of the material returned to the site, the
Washed Soil (and the treated Fine Particle slurry), would contain only a small
portion of the original contamination. As noted earlier, the decision concerning
return of material to the site remains a regulatory one.
It also may be feasible to recycle at least a portion of the CDE as input
water to the Soil Washer without treatment. Even though the CDE contributes 35-
40% of the penta mass in the two tests, its contribution to any of the solid
output streams would not be large. Since penta concentration would, presumably,
be limited to its solubility (-80 mg/L) , its contribution would be limited to 80
mg for each kilogram of water used. Experiments would be needed to evaluate the
buildup of penta, PAHs, and metals.
178
-------
SECTION 5
ECONOMIC ANALYSIS
5.1 INTRODUCTION
A pilot-scale demonstration test, such as the one conducted here, is very
often done before a commercial-scale remediation is attempted to prove the
applicability of the technology for a specific waste matrix and/or a specific set
of waste characteristics at a particular site. An "order-of-magnitude" cost
estimate for a hypothetical demonstration test at a "generic" site is presented
here based on costs incurred under the SITE program, on costs estimated by
BioTrol, and data obtained from other sources. Breakdown of these costs into the
same 12 categories that are used in the Applications Analysis has been attempted
for comparison purposes.
5.2 BASIS OF ECONOMIC ANALYSIS
The demonstration system used by BioTrol to quote costs consists of a 225-
450 kg/hr (0.25-0.50 ton/hr) Soil Washer followed by a 19 L/min (5 gpm) BioTrol
Aqueous Treatment System and a 0.02 L/min (0.006 gpm) Slurry Bio-reactor. It was
not possible for BioTrol to break down costs further according to the three
different technologies tested because of the generic nature of this analysis.
This does not mean to imply, however, that all three technologies must be used
together. Any one of the three technologies supplied by BioTrol may be used
alone or in conjunction with other processes that may or may not be supplied by
BioTrol for a particular application. No attempt has been made here to consider
the costs of such system configurations.
Table 5-1 shows the breakdown into the 12 cost categories used in the
Applications Analysis. Feed soil preparation such as excavation, debris removal,
blending etc. are not included. BioTrol's costs also include overhead and
profit. Site preparation, and permitting and regulatory requirements combined
are estimated to be $125,000. Based on experience with the SITE program,
permitting and regulatory requirements are minimal. Hence the majority of this
cost is associated with site preparation including shipping/freight, mobile crane
service, labor, travel and per diem, and installation of utilities. It should
be noted that this is a gross figure and includes many unknowns such as distance
to the site, the contaminants, required process modifications, and number of pre-
mobilization site visits and meetings with the client and/or regulatory agencies.
This figure, $125,000, was half of the average cost proposed by BioTrol,
$150,000-250,000, for mobilization and demobilization. It was felt that the
upper limit of this range was more realistic due to the inclusion of the SBR in
\
179
-------
the total clean-up system. The other half of this average figure is included in
item 12, demobilization, and is dependent on decontamination requirements.
TABLE 5-l."ORDER-OF-MAGHITUDE" COST ESTIMATE
FOR A HYPOTHETICAL DEMOHSJTRATIOH TEST AT A "GENERIC" SITE
Cost: Category
Technology
Cost
SW
BATS
SBR
1 Site Preparation
2 Permitting and Regulatory Req.
3 Capital Equipment
4 Start-Up
5 Labor
6 Consumables and Supplies
7 Utilities
8 Effluent Treatment & Disposal
9 Residuals/Waste Shipping & Handling
10 Analytical Services
11 Maintenance & Modification
12 Demobilization
Total
$125,000
$45,000/wk $25,000/wk
$6000/day
included in items 3-7 above
$1200/day
$117,000
$7200/day
$125,000
BioTrol estimated the cost of operation for the SW/BATS combination at
$35,000-45,000 per week. For a two week demonstration test, the upper end of
this range was used and included capital equipment (including rental of earth
moving equipment, office trailer with toilet, and decontamination trailer),
start-up, labor (including travel and per diem), consumables and supplies
(including flocculent, nutrients, and protective clothing/equipment) , utilities,
maintenance and modification.
For the SBR, BioTrol estimates operating costs at $15,000 - $25,000 per
week. Since the flow through the SBR is much lower than that in the SW/BATS
combination, it would have to be operated for a longer period of time to treat
the effluent from the 2 week SW/BATS demonstration test. For a 4-6 week SBR
demonstration test preceded by a 5-10 day acclimation period, costs would again
be in the upper portion of this range. Operating costs for the acclimation
period have been included in the site preparation costs. It was assumed that SBR
operation would begin after the Soil Washer was operated for 1 week.
Based on the SITE demonstration, effluent treatment and disposal, and
residuals/waste shipping and handling were estimated at $117,000. It should be
noted that the Fine and Coarse Oversize woody debris from the Soil Washer was not
incinerated during the demonstration test, but was drummed for off-site landfill.
This was less costly than incineration because of the smaller volume of waste
generated.
Analytical services was the cost category that was the most difficult to
estimate. Costs from the demonstration test would not be indicative of a
180
-------
commercial application because of the extensive analytical QA/QC requirements of
the SITE program. Analytical costs are highly dependent upon project scope,
i.e. , the contaminants, the number of bulk soil samples to test with the process,
the length of the test, and decontamination requirements imposed by the
regulating agency. During operation, the Soil Washer may generate 6-12 composite
samples per day. The BioTrol Aqueous Treatment System and Slurry Bio-Reactor may
each generate fewer samples per day. The total number of samples and the
parameters to be analyzed will determine the analytical costs. A cost-saving
step would be to choose one or two "indicator" compounds and do the analyses only
on these compounds during preliminary testing. Once the process is optimized,
analyses for all contaminants of interest can be done. For purposes of this
discussion, it was assumed that a "generic" group of analyses of a sample would
cost $600 including the containers, shipping and handling, QA/QC analyses, and
reporting. It was further assumed that 8 samples per day would be taken from the
Soil Washer and 2 samples per day (influent/effluent) for the BioTrol Aqueous
Treatment System and 2 samples per day (influent/effluent) for the Slurry Bio-
Reactor. The SW/BATS would generate 10 samples/day for 2 weeks while the SBR
would generate 2 samples/day for 4-6 weeks. Thus the total analytical costs are
estimated to be:
SW/BATS 10 samples/day x 15 days x $600/sample = $ 90,000
SBR 2 samples/day x 40 days x $600/sample = $ 48.000
$138,000
The total cost for operating a typical pilot-scale demonstration test would
then be broken down this way:
Percent of
Total Cost Total Costs
Mobilization
SW/BATS Operation ($45,000/wk
SBR Operation ($25,000/wk x 6
x 2 wk)
wk)
Residual shipping/handling and disposal
Analytical
Demobilization
$125
90
150
117
138
125
,000
,000
,000
,000
,000
,000
16.
12.
20.
15.
18.
8
1
1
7
5
TOTAL $745,000 100
The largest cost item is equipment operation (32%), followed by analytical
(18.5%) and mobilization/demobilization (16.8%) expenses. This would indicate
that waste site characteristics and test objectives play an important role in
determining costs. It is important, therefore, to define these parameters at the
beginning of the test. Capital cost for equipment lease has been included in the
operating cost of the testing program. Assuming continuous operation of the SW
at 0.375 ton/hour during 2 weeks, a total of 114,500 kg (126 tons) of soil would
be treated. The unit cost would then be about $6.50/kg or $6,000/ton.
The reader must be cautioned that this is for a pilot-scale operation and
does not relate to a full-scale installation. The cost per ton of soil treated
for full scale, commercial operation would drop dramatically from $6.50/kg to
$0.l9/kg ($168/ton). There are two major reasons for this. First, it was assumed
that the full-scale equipment was purchased outright and then amortized over a
181
-------
10-year life cycle period rather than leased. Under these circumstances, capital
equipment costs were about $40,000/month, excluding labor, as opposed to
$280,000/month, including labor, to lease the pilot-scale unit. This represents
a seven-fold difference. Second, the amount of soil treated increased two orders
of magnitude, from 115 metric tons (126 short tons) to 34,724 metric tons (38,273
short tons). The processing rate also increased proportionally, by two orders of
magnitude, but the associated equipment costs did not. This occurs because it
requires larger volumes which only cost incrementally more, rather than orders
of magnitude more. Hence, the cost in terms of $/ton goes down.
The full-scale treatment system would consist of an 18.2 metric ton/hour (20
short ton/hour) Soil Washer with a cost of $170/metric ton ($154/short ton) or
$257/m3 ($197/yd3) of soil treated; three 378.5 L/min (100 gpm) BATS units
connected in parallel with a cost of $1.73/metric ton ($1.57/short ton) of soil
or $0.44/1000 L ($1.65/1000 gal) of water treated; and an 87 L/min (23 gpm) SBR
with a cost of $2.92/metric ton ($2.65/short ton) of soil trated or $9.22/1000
L ($34.39/1000 gal) of slurry treated. These costs are based on the treatment of
34,724 metric tons (38,273 short tons) or 22,938 m3 (30,000 yd3) of contaminated
soil. In addition, the cost for incineration of the fine and coarse oversize
material has been included in the cost of the soil washer; it contributes about
80% to that cost figure or 75% to the overall cost.
182
-------
SECTION 6
QUALITY ASSURANCE/QUALITY CONTROL
6.1 INTRODUCTION
The demonstration test of the Biotrol soil washing process, designed for
removing pentachlorophenol (penta) and polynuclear aromatic hydro-carbons (PAHs)
from contaminated soil, was conducted in accordance with a detailed Quality
Assurance Project Plan (QAPjP). The QAPjP contained numerical specifications for
data quality in terms of precision, accuracy, completeness, and also required the
use of standardized methods to insure data comparability and representativeness.
The intent of this section is to review these objectives and report on the
quality of the data in terms of these defined objectives.
In order to satisfy data quality objectives, as defined above, a
sophisticated quality control process was implemented. This process focussed on
controlling measurement error associated with both sampling and field activities
through the use of laboratory spikes (surrogate and matrix), replicate sampling,
field and laboratory blanks, and strict adherence to QA/QC requirements of the
methods specified in the QAPjP. When appropriate, these methods were modified
to satisfy field requirements.
In addition to these controls, frequent laboratory and field audits were
conducted by EPA and by contractor QA personnel. As discussed in the following
summary, several corrective action procedures were required as a result of these
audits. Numerous obstacles were noted throughout the course of this project which
potentially or directly affected the quality of the collected data. In some
instances special studies were instituted to learn the effect of these
difficulties upon data quality. In addition, detailed analyses have been
conducted to determine the data quality impact of potentially negative results
obtained from data quality indicators. The presented discussions are intended
to inform the reader of potential impacts upon data quality, and the evaluation
in Section 6.2, summarizes what were considered to be the more critical data
quality problems. This section also explains how these problems were reconciled
during the course of the project.
Certain key measurements were used to determine process removal efficiencies
as they related to project objectives for penta and PAHs. These included penta
and PAH analyses in feed soil and washed soil for the soil washer, influent and
effluent streams for the slurry bioreactor, and influent and effluent streams for
the aqueous treatment system. These were all defined as critical measurements
for this demonstration. In addition, for purposes of obtaining a mass balance and
to learn the fate of other contaminants and their effect upon removal
efficiencies of penta and PAHs, other analyses were defined as secondary critical
183
-------
measurements. These included copper (Cu), chromium (Cr), and arsenic (As)
analyses, polychlorinated dibenzo-dioxins/ polychlorinated dibenzo-furans
(PCDDs/PCDFs), total residue, and measurements of perita and PAHs in matrices
other than those defined above. All other measurements were considered non-
critical and results of these analyses were not directly related to project
obj ectives.
6.2 CONCLUSIONS AND LIMITATIONS OF DATA
As presented in this entire QA/QC section as well as in the main portion of
the report, several occurrences during the course of this demonstration could
have a significant impact on the quality of the data. These problems are
discussed in detail in the remainder of section 6 and are summarized below. In
review, the most significant problems were; 1) holding times for extractions and
analyses which exceeded QAPjP and analytical specifications, 2) the lack of
MS/MSD data for all soil and slurry matrices, 3) the lack of sufficient data to
properly characterize the efficiency of the slurry bioreactor, and 4) the
inability to reconcile a mass balance between influent and effluent contaminants.
Because of the negative potential of each of these problems to affect data
quality, significant efforts have been made to determine the seriousness of each
of these occurrences and the resulting impact upon data generated during the
course of the project.
Holding times were examined in detail by conducting a separate holding time
study to determine if a holding time bias existed. Because holding times were
exceeded for a large portion of the samples, several sample replicates were re-
extracted or reanalyzed and concentrations were compared to original sample
analyses conducted within the specified holding time to determine if a change had
occurred. In addition to the samples chosen for reanalysis, a significant number
of replicate samples analyzed by the laboratory were evaluated such that one
sample was analyzed within holding time and another was analyzed outside the
holding time. While this often occurred only by chance and was not deliberate
planning, these sample comparisons added to the information gained by the holding
time studies.
In brief, holding time study results showed no detectable concentration bias
for the period of time over which the study was conducted. In addition, several
sample concentration comparisons of replicate samples analyzed by the laboratory
also showed similar data with no significant sample bias due to holding time.
Two solid matrices (fine particle cake and fine oversize) showed a possible
holding time bias, however, this proved to be inconclusive. In addition, these
two matrices were not critical for determining the efficiency of the process, but
were used only as part of the mass balance calculation. While holding time bias
cannot be completely ignored, because the number of total samples used in the
study is small and not statistically significant, the study does show that this
bias was not measurable within the observed sampling and analytical precision.
Impact upon the quality of the final data set is therefore considered to be
insignificant.
MS/MSD data are of little use for determining the extraction efficiency of
pentachlorophenol in solid matrices because the concentration of penta spiked
184
-------
into the matrix is often much lower than the native concentration. Usually, the
spike was as much as ten times below the native penta concentration. This
problem occurred because samples were not properly screened by the laboratory
prior to being spiked with the compound of interest. While matrix spikes cannot
be used to determine accuracy or extraction efficiency, results of surrogate
spiking suggest that extractions were within QAPjP limits, and in fact, extra
efforts were made by the laboratory to re-extract all samples which did not fall
within surrogate spiking specifications. Results from any re-extracted samples
which did not meet the specifications stated for surrogate spikes were not used
for any of the data manipulations or calculations. Accuracy of the penta
extraction is therefore somewhat in question, but the surrogate spike information
suggests that extraction efficiency is within specification. Additional detail
is included in section 6.4.
Because the micro-organisms within the slurry bioreactor were not actively
degrading the penta during a portion of the demonstration, several samples taken
from the SBR showed no significant change in penta concentration. Given the
difficulty in determining exactly when penta degradation is occurring (and
because samples could not be immediately analyzed) , the problem with the SBR was
not determined during the sampling period, but only after analyses were received
from the laboratory. It was also obvious from these analyses that the SBR began
functioning properly part-way through the demonstration. Results therefore show
a trend in efficiency, but because the efficiency is rapidly climbing during the
four days over which the samples were taken, a maximum or peak efficiency for the
process cannot be determined. Results reported therefore show an average
efficiency for these four days and also show a trend upward suggesting that the
process can maintain a higher degradation.
Mass balance between influent and effluent contaminants remains somewhat
unresolved. As noted in the report, it is believed that the increase in
contaminant concentration in effluent samples could be due in part to increased
extraction efficiency in the laboratory as a result of further particle breakdown
from the scrubbing action of the treatment process.
6.3 PROCEDURES DEFINING DATA QUALITY
6.3.1 Precision
Precision is the reproducibility or degree of agreement among replicate
measurements of a single parameter. Precision for a single parameter was
expressed as a percentage of the difference between results of duplicate samples
for a single parameter. This relative percent difference (RPD) was calculated
as:
RPD = Maximum Value - Minimum Value x 100
[(Maximum Value + Minimum Value)/2]
For data sets greater than two points, the coefficient of variation (CV) was used
to assess precision. The %CV is calculated as:
185
-------
%CV = Standard Deviation x 100
Mean
Precision for all measurements was calculated through examination of results
for laboratory matrix spike/matrix spike duplicates (MS/MSD) and other laboratory
duplicate samples. Critical measurements for organic compounds also employed
precision calculations for results of field replicate sample analysis to assess
combined field and laboratory variability.
6.3.2 Accuracy
Accuracy is the agreement between a measurement value and a theoretical or
known value. Accuracy can be expressed as the relative error (%RE) where;
%RE - Measured Value - Theoretical Value x 100
Theoretical Value
Accuracy for all measurements was calculated for analytical results of analytes
spiked into the sample matrix. Analytical systems were also checked routinely
for system accuracy through use of calibration check samples and control samples.
Matrix spike recoveries were calculated using the following equation:
%Recovery (spike) - C33 - Cu. x 100
Csa
Where: Cas = Analyte concentration in spiked sample
Cua = Analyte concentration in unspiked sample
Csa — Analyte concentration added to the sample
6.3.3 Completeness
Measurement data completeness is a measure of the amount of valid data
produced compared to the total amount of data originally planned for the project.
Project measurement completeness is discussed in detail in section 6.4.
6.3.4 Comparability
Comparability expresses the extent with which one data set can be compared
to another. This was accomplished by use of standard, accepted reference methods
and adherence to protocols specified in the QAPjP and generation of quality
control data. Non-standard methods used were well documented to ensure that
procedures are repeatable by other investigators.
186
-------
6.3.5 Representativeness
Representativeness refers to the degree with which analytical results
accurately and precisely represent actual conditions present at locations chosen
for environmental sampling. Throughout the life of this project, regular, timely
reviews of analytical and quality control data were conducted to ensure that
results generated could be considered representative of the actual site
conditions.
6.4 ANALYTICAL QUALITY CONTROL
6.4.1 Soil Pile Analyses
Low concentration pentachlorophenol (penta) and high concentration penta
soil piles were analyzed for the following parameters: Penta, polynuclear
aromatic hydrocarbons (PAHs), Copper(Gu), Chromium(Cr), Arsenic(As), total
residue, polychlorinated dibenzo-dioxins and polychlorinated dibenzo-furans
(PCDDs/PCDFs), total metals, total organic carbon (TOG), total recoverable
petroleum hydrocarbons (TRPH) and cation exchange capacity (CEC). For the low
concentration soil pile, 8 separate soil sample cores were analyzed, and for the
high concentration soil pile, 12 separate soil sample cores were analyzed. In
addition, one composite sample was analyzed for both the high concentration and
low concentration soil piles.
ACCURACY
PENTACHLOROPHENOL AND PAHs
Surrogate spike summary data for semivolatile analyses of all soil matrices
(aqueous and solid) including re-extracted samples, are presented in Tables 6-1
and 6-2. As noted in these tables, soil matrices analyzed from the two soil
piles (high and low penta concentration ) were generally within specified
accuracy limits as determined from surrogate spike results. It should be noted
that Terphenyl-dl4 and 2-Fluorophenol in the aqueous matrices for the high
concentration soil test, are surrogates which more often than others fell outside
acceptable recovery ranges. (Specific detail regarding the surrogate spike
criteria used to determine if the analyses are acceptable is presented in section
6.4.2. Data outside these specified recovery ranges were not used in the
calculation of project objectives.)
Per QAPjP specifications, accuracy was also to be determined from matrix
spike recoveries. Matrix spikes are intended to show the actual recovery
percentage for the primary compound being analyzed in the matrix of interest.
For this demonstration the critical compounds of interest were penta and PAHs.
In particular, because of its high concentration in the feed soil, penta was more
critical than any of the PAHs which were in relatively low concentrations.
Because the concentration of penta was extremely high in the feed soil (and
consequently in each of the effluent soil matrices as well as the slurry for the
slurry bioreactor process), matrix spike concentrations were too low for
187
-------
TABLE 6-1. ACCURACY - SURROGATE SPIKES SOIL WASHER
LOW CONCENTRATION SOIL TEST
Total # Within # Outside
Analyte No. of % Recovery Control Control Control
(Matrix) Analyses Ranges Limits (%) Limits Limits
(Aqueous )
NBZ
FBP
TPH
PHL
2FP
TBP
FLT
(Solid)
NBZ
FBP
TPH
PHL
2FP
TBP
FLT
19
19
19
19
19
19
19
52
52
52
52
52
52
52
58-105
58-114
34-121
18-76
41-78
53-119
58-120
31-95
44-113
34-130
28-99
40-91
29-100
0-123
35-114
43-116
33-141
10-94
21-100
10-123
20-130
23-120
30-115
18-137
24-113
25-121
19-122
15-140
19
19
19
19
19
19
19
52
52
52
52
52
52
51
0
0
0
0
0
0
0
0
0
0
0
0
0
1
NBZ Nitrobenzene-d5
FBP 2-Fluorobiphenyl
TPH Terphenyl-dl4
PHL Phenol-d5
2F.P 2-Fluorophenol
TBP 2,4,6-Tribromopheno1
FLT Fluoranthene-dlO
188
-------
TABLE 6-2. ACCURACY - SURROGATE SPIKES SOIL WASHER
HIGH CONCENTRATION SOIL TEST
Total
Analyte No. of %
(Matrix) Analyses
(Aqueous )
NBZ
FBP
TPH
PHL
2FP
TBP
FLT
(Solid)
NBZ
FBP
TPH
PHL
2FP
TBP
FLT
41
41
41
41
41
41
41
101
101
101
101
101
101
101
# Within # Outside
Recovery Control Control Control
Ranges Limits (%) Limits Limits
12-113
18-115
10-137
20-99
4-100
31-119
29-128
0-99
0-120
14-125
0-91
0-92
0-122
55-127
35-114
43-116
33-141
10-94
21-100
10-123
20-130
23-120
30-115
18-137
24-113
25-121
19-122
15-140
39
39
34
40
43
41
41
100
99
100
99
99
100
101
2
2
7
1
8
0
0
1
2
1
2
2
1
0
NBZ Nitrobenzene-d5
FBP 2-Fluorobiphenyl
TPH Terphenyl-dl4
PHL Phenol-d5
2 FP 2-Fluoropheno1
TBP 2,4,6-Tribromophenol
FLT Fluo ranthene-dlO
189
-------
accurate information to be obtained. Laboratory personnel assumed too low of a
native penta concentration in aqueous and soil matrices when preparing matrix
spike solutions. Spiking concentrations were based upon experience with other
soil matrices and not upon actual penta concentration in the solid being analyzed
for this project. Consequently, the penta matrix spike concentration was often
ten times or more lower than the native compound concentration. QA personnel
were made aware of this problem well after analyses were complete. In fact,
during the final QA audit of data quality, laboratory personnel were still unsure
about actual penta spiking concentrations. Corrective action in terms of
respiking proved impractical due to the exceedingly high concentrations required,
holding time considerations, and the fact that analyses had long been completed.
Matrix spike information was therefore of little use in determining percent
recoveries for penta, and the native concentration data presented from these
MS/MSD results are used only for purposes of laboratory precision.
Table 6-3 presents the penta MS/MSD data. This table covers both soil and
water matrices for the soil washer and slurry bioreactor. Generally precision
is within acceptable laboratory variability. Those matrices which showed large
variability were those which were also more difficult to extract due to their
slurry rather than solid or aqueous nature (i.e., fine particle cake).
Matrix spike information for PAH compounds and 2,3,5,6- tetrachlorophenol
in all soil washer matrices (aqueous and soil input and output streams) is
presented in Table 6-4. These data indicate, that in general, extraction
efficiencies were within specified limits for all spiked PAH compounds except
pyrene. Because pyrene was often outside QC limits there appeared to be a matrix
problem with this compound which may have an effect upon conclusions concerning
removal efficiency or mass balance for pyrene and compounds similar in chemical
structure and behavior.
Included in Table 6-4 are commonly used matrix spikes for SW-846 Method 8270
analysis and one added spiking compound 2, 3,5, 6-tetrachlorophenol (2,3, 5, 6-TCP) .
This was an additional surrogate for better determining the recovery efficiency
of penta. Anticipating high levels of penta and knowing this could cause matrix
spike complications 2,3,5,6-TCP was proposed during preparation of the QAPjP as
an additional check of recovery efficiency for penta. As noted by the results
in Table 6-4, 2,3,5,6-TCP recovery efficiency was somewhat variable as may be
expected given its chemical proximity to penta and knowing from previous proj ects
that penta can be difficult to extract from water or soil matrices. While the
laboratory had no specified QC limits for recovery efficiency for 2,3,5,6-TCP
(because it is not a commonly used laboratory spike) general recoveries do appear
to be acceptable if compared to accepted matrix spike recovery QC limits for
penta. There are some solid matrices, however, which have unacceptable recovery
limits and variability. The reason this occurred is because 2,3,5,6-TCP matrix
spike concentrations were based upon the samples containing no native 2,3,5,6-
TCP. It was later discovered that several of the matrices did contain native
2,3,5,6-TCP, and therefore, as with the penta spikes, several of the samples were
spiked at too low of a level to accurately determine recovery efficiency.
190
-------
TABLE 6-3. PENTACHLOROPHENOL MS/MSD RESULTS - PRECISION
MS (ug/mL) MSD (ug/mL)
On- Column
Sample Type
WATER SAMPLES
(All Treatment Processes)
SWL-MW
SWH-MW
SWH-ATW
SBR-EQB
SBR-IN
SBR-IN
SOIL SAMPLES
(All Treatment Processes)
SPL-C
SPH
SWL-FS
SWL-FPC
SWH-FPC
SWH-FS
SWH-WS
SWL - Soil Washing Low Cone
SWH - Soil Washing High Cone
SBR - Slurry Bioreactor
SPL - Soil Pile Low Cone
SPH - Soil Pile High Cone
Cone
92.7
67.1
199.5
93.8
115
139
95.8
105.1
173.9
81.5
107.9
84.2
189.4
MW
ATW
EQB
IN
C
FS
FPC
WS
On- Column
Cone
81.5
63.7
211.7
,103.2
143
266
91.5
118.6
130.5
57.3
65.8
75.3
189.9
RPD
%
12.9
5.2
5.9
9.5
21.7
62.7
4.6
12.1
28.5
34.9
48.5
11.2
0.3
- Municipal Water
- ATS Recycled
Water
- Equipment Blank
- Influent
- Composite
- Feed Soil
- Fine Particle
- Wash Soil
Cake
191
-------
TABLE 6-4. MS/USD DATA FOR SOIL WASHER (SOIL AND AQUEOUS MATRICES) PAH COMPOUNDS ONLY
Matrix
# of
Analyses
X Rec
Range
Control
Limits
# Within # Outside
Limits Limits
RPD
Range
RPD Control
Limits
# Within
Limits
# Outside
Limits •
SOIL MATRICES
ATS-CAR
SBR-CAR
SBR-IN
SPL
SPL-CO
SPH
SWL-CO
SWH-CO
SUH-FO
SWH-FPC
SWL-FS
SWH-FS
SWH-WS
2
4
4
4
2
6
2
4
2
6
12
2
6
24-26
7-20
59-89
44-51
72-78
42-74
60-62
69-77
59-82
57-63
56-108
69-70
44-77
30-140
30-140
30-140
30-140
30-140
30-140
30-140
30-140
30-140
30-140
30-140
30-140
30-140
- - Naphthalene -
0
2
4
4
2
6
2
4
2
6
12
2
6
-
2
2
0
0
0
0
0
0
0
0
0
0
0
8
22-25
3-12
15
8
3-7
3
3-11
33
2-8
0-39
1
4-13
50
50
50
50
50
50
50
50
50
50
50
50
50
1
2
2
2
1
3
1
2
1
3
6
1
3
0
0
0
0
0
0
0
0
0
0
0
0
0
- - 4-Chloro-3-methylphenol - -
ATS-CAR
SBR-CAR
SBR-IN
SPL
SPL-CO
SPH
SWL-CO
SWH-CO
SWH-FO
SWH-FPC
SWL-FS
SWH-FS
SWH-WS
2
4
4
0
2
4
2
4
2
6
12
2
6
33-35
11-29
73-85
-
84-87
46-57
55-61
65-68
44-66
50-72
48-87
54-55
46-73
26-103
26-103
26-103
26-103
26-103
26-103
26-103
26-103
26-103
26-103
26-103
26-103
26-103
2
1
4
-
2
4
2
4
2
6
12
2
6
0
3
0
-
0
0
0
0
0
0
0
0
0
6
9-27
4-10
-
4
5-16
10
2-5
40
3-8
0-29
2
7-13
50
50
50
50
50
50
50
50
50
50
50
50
50
1
2
2
-
1
2
1
2
1
3
6
1
3
0
0
0
-
0
0
0
0
0
0
0
0
0
192
-------
TABLE 6-4. (CONTINUED)
# of
Matrix Analyses
SOIL MATRICES
ATS-CAR
SBR-CAR
SBR-3N
SPL
SPL-CO
SPH
SUL-CO
SUH-CO
SWH-FO
SWH-FPC
SWL-FS
swH-ns
SWH-WS
ATS-CAR
SBR-CAR
SBR-IN
SPL
SPL-CO
SPH
SWL-CO
SUH-CO
SWH-FO
SWH-FPC
SWL-FS
SWH-FS
SWH-WS
2
4
4
4
2
6
2
4
2
6
12
2
6
2
4
4
0
2
4
2
4
2
6
12
2
6
% Rec
Range
20-21
5-17
94-121
67
83-85
68-87
68-81
52-113
64-121
66-107
30-104
78
34-72
12-16
0-12
41-97
-
104-114
18-77
56-68
61-186
58-71
19-99
42-106
23-44
49-96
Control # Within # Outside
Limits Limits Limits
. -
31-137
31-137
31-137
31-137
31-137
31-137
31-137
31-137
31-137
31-137
31-137
31-137
31-137
•
11-114
11-114
11-114
11-114
11-114
11-114
11-114
11-114
11-114
11-114
11-114
11-114
11-114
- Acenaphthene •
0
0
4
4
2
6
2
4
2
6
11
2
6
- 4-Nitrophenol
2
1
4
-
2
4
2
2
2
6
12
2
6
• -
2
4
0
0
0
0
0
0
0
0
1
0
0
- -
0
3
0
-
0
0
0
2
0
.0
0
0
0
RPD RPD Control # Within # Outside
Range Limits Limits Limits
5
18-27
8-11
0
2
6-22
17
32-74
62
3-31
4-70
0
6-26
29
0-120
26-64
-
9
5-70
19
18
20
30-108
2-58
63
10-30
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
1
2
2
2
1
3
1
1
0
3
5
1
3
1
1
2
-
1
1
1
2
1
2
5
0
3
0
0
0
0
0
(I
(I
1
1
0
\
0
0
0
1
0
-
0
1
Cl
0
0
1
1
1
0
193
-------
TABLE 6-4. (CONTINUED)
Matrix
# of
Analyses
X Rec
Range
Control # Within # Outside
Limits
Limits Limits
RPD
Range
RPD Control
Limits
# Within
Limits
Limits
SOIL MATRICES
ATS-CAR
SBR-CAR
SBR-IN
SPL
SPL-CO
SPH
SUL-CO
SUH-CO
SWH-FO
SWH-FPC
SUL-FS
SUH-FS
SUH-WS
ATS-CAR
SBR-CAR
SBR-IN
SPL
SPL-CO
SPH
SUL-CO
SUH-CO
SWH-FO
SUH-FPC
SUL-FS
SUH-FS
SUH-WS
2
4
4
4
2
6
2
4
2
6
12
2
6
2
-4
4
0
2
4
2
4
2
6
10
2
6
3-4
0-6
35-168
60-65
70-71
23-110
73-91
-413-61
185-275
-26-171
2-187
31-51
-12-65
5-7
0-8
7-50
-
42-45
28-69
16-18
-26-87
-(110-30)
15-56
-49-111
24-30
-13-32
-
35-142
35-142
35-142
35-142
35-142
35-142
35-142
35-142
35-142
35-142
35-142
35-142
35-142
.
**
**
**
**
**
**
**
**
**
**
**
**
**
- Pyrene - -
0
0
3
4
2
4
2
0
0
1
8
2
2
- 2,3,5,6-TCP -
-
-
-
-
-
-
-
-
-
-
-
-
-
2
4
1
0
0
2
0
4
2
5
4
0
4
-
-
-
-
"
-
-
-
-
-
-
-
-
-
29
0-200
43-96
8
1
0-32
22
387-453
39
94-233
2-109
49
2-169
33
NA
4-83
7
0-49
12
NA
NA
24-81
NA
22
NA
50
50
50
50
50
50
50
50
50
50
50
50
50
**
Wit
**
**
**
1t1t
W*
4rA
llrik-
**
**
**
**
1
1
1
2
1
3
1
0
1
0
2
1
1
• "
0
1
0
0
0
0
2
3
s m
194
-------
TABLE 6-4. (CONTINUED)
# of
Matrix Analyses
AQUEOUS MATRICES
SWL-CDE
SUL-HW
SWH-HW
SWH-ATU
2
2
4
4
% Rec
Range
56-60
59-64
72-81
28-80
Control
Limits
-
30-120
30-120
30-120
30-120
# Within # Outside
Limits Limits
- Naphthalene -
2
2
4
3
-
0
0
0
1
RPD RPD Control # Within # Outside
Range Limits Limits Limits
7
8
4-5
1-76
40
40
40
40
1
1
2
1
0
0
0
- 1
- - 4-Chloro-3-methylphenol - -
SWL-CDE
SWL-HU
SWH-MU
SUH-ATW
SWL-CDE
SWL-HU
SWH-HW
SWH-ATW
SWL-CDE
SWL-MW
SWH-MW
SWH-ATW
SWL-CDE
SWL-MW
SUH-HW
SWH-ATW
2
2
4
4
2
2
4
4
2
2
4
4
2
2
4
4
54-59
69-72
67-76
38-53
69-70
70-81
80-85
64-83
59-91
114
67-92
40-84
42-49
83-89
88-97
33-73
23-97
23-97
23-97
23-97
.
46-118
46-118
46-118
46-118
.
10-80
10-80
10-80
10-80
-
26-127
26-127
26-127
26-127
2
2
4
4
- Acenaphthene -
2
2
4
4
- 4-Nitrophenol
1
0
2
2
- Pyrene - -
2
2
4
4
0
0
0
0
-
0
0
0
0
. .
1
2
2
2
0
0
0
0
9
4
4-5
2-33
1
15
2-6
6-16
43
0
1-9
1-28
15
7
4-7
16-33
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
1
1
2
2
1
1
2
2
0
1
2
2
1
1
2
2
0
0
0
0
0
Cl
Cl
0
1
0
0
0
0
0
0
0
195
-------
TABLE 6-4. (CONTINUED)
Matrix
AQUEOUS
# of
• Analyses
MATRICES
X Rec
Range
Control
Limits
# Within
Limits
# Outside
Limits
RPO
Range
RPO Control
Limits
# Within
Limits
# Outside
Limits
- - 2,3,5,6-TCP - -
SUL-CDE
SWL-MW
SUH-MW
SWH-ATW
ATS
SBR
SPL
SPH
SWL
SWH
2 70-125 **
2 80-85 **
4 41-76 **
4 39-67 **
Aqueous Treatment System
Slurry Bioreactor
Soil Pile Low Concentration
Soil Pile High Concentration
Soil Washing Lou Concentration
Soil Washing High Concentration
-
-
-
-
CAR
IN
CO
FO
FPC
FS
WS
COE
MW
ATW
56 **
6 **
16-58 **
10-16 . **
Carbon Canister
Influent
Coarse Oversize
Fine Oversize
Fine Particle Cake
Feed Soil
Wash Soil
Combined Dewatering Effluent:
Municipal Water
ATS Recycled Water
196
-------
Matrix spiking procedures are intended to mimic native compound behavior in
the matrix of interest. While this may work well for aqueous matrices, (provided
spiking concentrations are well below solubility limits), they are of more
questionable value for soil matrices. Specifically, there are problems with
soils because: 1) a spiking compound cannot be evenly distributed over and into
the soil, and 2) the question of the compound reaching some kind of equilibrium
in the soil before being extracted cannot be fully answered. Given these
uncertainties associated with the soil spiking procedure, and the facts that
surrogate recoveries were always within QC specifications and 2,3,5,6-TCP
recoveries were generally within the acceptable range for penta recoveries, it
can be postulated that laboratory extraction and analysis procedures were within
acceptable limits.
COPPER, CHROMIUM, ARSENIC
Accuracy for copper, chromium, and arsenic (CCA) analyses of composited
high- and low-concentration soil piles, and the particle size fractions of these
samples, was assessed by laboratory matrix spike results (see Table 6-5). .One
arsenic recovery, associated with the analysis of some of the particle size
fraction samples, was below the control limit. All other values were within
QAPjP specifications indicating acceptable laboratory recovery and precision.
TOTAL RESIDUE
During the technical systems review of the laboratory, analytical balances
used for weighing total residue were checked to insure calibrations met
specifications. Standard operating procedure required a calibration check by
calibrated laboratory weights used to ensure accuracy specifications did not
exceed QAPj P requirements. All balances were then checked regularly during the
course of the project to ensure total residue measurements were within specified
requirements. These calibration checks served as a means of determining the
accuracy of the total residue analysis.
PCDDs/PCDFs ,
Surrogate recoveries for all PCDD/PCDF analyses were within QAPjP
specifications. Surrogates used in the PCDD/PCDF analysis are C13 derivatives
of the congeners of interest. These surrogates therefore chemically represent
total efficiency for extraction and analysis without need of matrix spike
determinations.
NON-CRITICAL ANALYSES
There were no accuracy checks for the CEC analysis. Matrix spikes for total
metals and EP Toxicity are reported in Table 6-5. As noted by this table, metal
digestions were generally within specified control limits. Spikes for total
recoverable petroleum hydrocarbons (TRPH) and total organic carbon (TOG) were
all within specifications.
197
-------
Parameter
Hetals:
Arsenic
Baritin
Cadmium
Chromium
, Copper
Lead
Mercury
Se lent un
Silver
VO
0° TOTAL
RejBjue
fP Toxicjty:
Arsenic
Bariun
Cadmium
Chromiui
Copper
Lead
Hercury
Selenium
Silver
Total
No. of
Analyses
2
1
1
3
3
1
1
1
1
U
N/A
X Recovery
Ranges
55-87
108
100
104-107
87-91
107
66
80
98
—
119
122
102
101
us
119
too
125
95
• ACCURACY
* within
Control Control
Limits (X) Limits
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
13
.-
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
1 Outside
control
Limits
1
0
0
0
0
0
0
0
0
1
• •
0
0
0
0
0
.0
0
0
0
Total
Ho. of
Analyses
2
1
1
3
3
1
1
1
1
1*
2
RPO
Ranges
1.3-5.7
18
4.3
0.53-8.6
2.1-9.4
0.83
7.8
6.0
3.8
1.4-7.0
3.0
4.6
7.7
2.5
1.3
1.8
19
5.3
2.0
PRECISION
Control
Limits
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-30
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
tfUithin « Outside
Control Control
Limits Limits
2 0
0
0
3 0
3 0
0
0
0
0
14 0
2 0
0.
0
0
0
0
0
0
0
0
( } Soil pile sables include the high and lot* composite samples, as Nell as the particle size fractions of the high end lot, soil pile sanples.
-------
PRECISION
Because the soil piles were composited prior to the demonstration all 12 of
the high concentration soil pile samples and all 8 of the low concentration soil
pile analyses were considered to be the same sample. For some of the analyses,
however, split field samples were obtained for precision estimates. Table 6-5
shows the precision in terms of relative percent difference (RPD) for this split
sample for the analyses of metals, residue, and EP Toxicity. All analyses shown
in this table were within QAPjP specifications.
Analytical precision for penta and PAH analyses are shown by MS/MSB results
in Tables 6-3 and 6-4. Field precision for penta as determined by the 12
replicate samples for the high concentration soil pile in terms of relative
standard deviation is well within QAPjP specifications of 50%. Precision for the
feed soil matrix is presented in the section on soil analyses. The feed soil is
the same matrix as the soil pile, and therefore, these results are also
comparable and provide additional verification of analytical precision for semi-
volatile analyses in the soil piles.
COMPLETENESS
Completeness data for the soil pile are presented in Table 6-6. As shown
in this table, all analyses were 100% complete.
6.4.2 Soil Washer Analyses
Two separate demonstrations were performed to evaluate the effectiveness of
the BioTrol process on two separate soils with low and high concentrations of
pentachlorophenol (400 and 1100 PPM, respectively). Characteristics of these two
different soils are discussed in detail in the main body of the report. Both
solid and aqueous matrices were produced as a result of the BioTrol soil washing
procedure. The separate fractions were designated as feed soil, washed soil,
fine particle cake, coarse oversize, fine oversize, and combined dewatering
effluent. Municipal water feeding into the system was also analyzed to complete
the mass balance equations. Analyses performed for each of these tests included
the following parameters; penta, PAHs, Cu, Cr, As, total residue, PCDDs/PCDFs,
total metals, total organic carbon (TOG) , total recoverable petroleum hydrocarbon
(TRPH), EP Toxicity, and the Toxicity Characteristic Leaching Procedure (TCLP).
ACCURACY
PENTACHLOROPHENOL AND PAHs
Surrogate spike summary data are presented in Tables 6-1 and 6-2 for all
semivolatile analyses. Some samples were outside QAPjP specifications. Matrix
effects are probably why there were problems with surrogate recovery. The two
surrogate compounds which gave particular problems were 2-fluorophenol arid p-
terphenyl-d!4. The criteria used for acceptability of data based upon surrogate
recovery are explained in the following paragraph. If surrogate spike data did
199
-------
TABLE 6-6. COMPLETENESS SOIL PILES
High Concentration Soil Pile
Low Concentration Soil Pile
ro
o
o
I
Analysis Projected
- - Critical - -
Pentachlorophenol 13
PAHs 13
Cu, Cr, As 1
Total Residue 13
Dioxin/Furan 1
- - Non-Critical - -
Total Metals
TOC
Total Rec Petro Hydro
Cation Exchange Capacity
EP TOX
Satisfactory
Analysis
13
13
1
13
1
1
1
1
1
1
Percent
Complete
100
100
100
100
100
100
100
100
100
100
QAPjP
Objective
1
90
90
90
90
90
90
90
90
90
90
Projected
.
9
9
1
9
1
1
1
1
1
1
Satisfactory
Analysis
9
9
1
9
1
t
1
1
1
1
1
Percent
Complete
100
100
100
100
100
100
100
100
100
100
QAPjP
Objective
90
90
90
90
90
90
90
90
90
90
-------
not meet these specifications, results from those specific samples were not used
in determining proj ect obj ectives.
Data were considered acceptable for use if one surrogate was outside its
acceptable recovery range or if two surrogates were outside the acceptable range,
provided they were not both the same class of compounds, (e.g. , two acids or two
bases). For example, if penta or an acid PAH compound concentration was being
evaluated then two base/neutral surrogates could be outside the recovery range,
provided all acid surrogates are within acceptable limits. This same rule
applies for base/neutral compounds of interest but instead of base/neutral
surrogates being outside their acceptable recovery range, surrogate recoveries
for two acid surrogates can be outside recovery specifications and all base/
neutral surrogate compounds must be within acceptable limits. Because* the
laboratory always performed necessary re-extractions when surrogates were outside
limits of acceptability, it is probable that matrix interferences account for the
sometimes poor recoveries of 2-fluorophenol and p-terphenyl-d!4.
Matrix spike/matrix spike duplicate (MS/MSD) data are presented in Tables
6-3 and 6-4. As noted in the section on soil pile analyses, matrix spike data
are not useful for determining recovery efficiencies for penta on any of the soil
matrices because spiking concentrations were well below native penta
concentrations. This occurred for almost all matrices analyzed from the soil
washer. This was not the case for the PAHs; however, recovery results are of
less interest because little information was gained from PAH analyses.
Specifically, only very low concentrations of PAHs were found in any of the
native feed soils, and consequently removal efficiencies for PAHs were only
determined for very few compounds. The data for penta MS/MSD results are useful
however for determining analytical precision. Additional information on MS/MSD
data is presented in the section on soil pile analyses.
COPPER, CHROMIUM, ARSENIC
Accuracy for copper, chromium, and arsenic (CCA) was determined by the
analysis of laboratory matrix spikes. Results are summarized in Table 6-7 for
both aqueous and solid streams of the soil washer.
All but one arsenic and one chromium spike recovery were within the control
limits defined by the QA obj ectives.
TOTAL RESIDUE
As described previously accuracy for total residue analyses was determined
by balance calibration checks performed as part of the technical systems review
as well as routine calibration checks conducted during the course of the project.
Analytical balances were always within specification.
PCDDs/PCDFs
Accuracy for PCDD/PCDF analyses was based upon the surrogate spike
recoveries. As noted in the previous section under soil pile analyses, matrix
spike data were not required for PCDD/PCDF analyses since surrogates were
201
-------
TABLE 6-7. QC SUHHARY:
PRECISION AND ACCURACY - COPPER CHROHIUN ARSENIC. NON-CRITICAL HETALS. RESIDUE.
EP TOXICITY SOIL UASNER SAMPLES
O
Parameter
Soil Samp|es'"'
Hetals:
Arsenic
Barium
Caclniun
Chromium
Copper
Lead
Mercury
Selenium
Silver
TOTAL
Residue (X Solids)
EP Toxlcity;
Arsenic
Buriun
Cadmium
Chromiun
Copper
Lead
Mercury
Selenium
Silver
Aqueous Samp) eg/**
Hetals:
Arsenic
Barium
Cadmium
Total
No. of X Recovery
Analyses Ranges
10
8
8
8
8
8
3
6
8
67
N/A
2
2
2
2
2
2
2
2
2
10
8
8
(1) Soil samples include all
(2) Aqueous samples Include
(3) The MS with OX
N/A Not applicable
recovery
75-110
75-117
96-119
76-125
76-120
92-119
86-9S
44-103
87-115
--
84-119
103-122
90-102
101-103
93-115
105-119
85-100
105-125
93-95
67-94
96-104
98-113
solid streams front
aqueous streams from
Mas reported with a
ACCURACY -•
# Within
Control Control
Limits (X) Limits
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
--
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
10
8
8
7
8
8
3
4
8
64
--
2
2
2
'2
2
2
2
2
2
9
8
8
the low and high pi la
the low (SUL-CPE/HU)
footnote
stating that
» Outside
Control
Limits
0
0
0
1
0
0
0
2
0
3
" --
0
0
0
0
0
0
0
0
0
1
0
0
soil Hashing
and high
-------
TABLE 6-7. OC SUMMARY:
PRECISION AMD ACCURACY - COPPER CHROMIUM ARSEMIC. MOM-CRITICAL METALS. RESIDUE
EP TOXICm SOIL UASHER SAMPLES '
(CONTINUED)
Total
No. of
Analyses
X Recovery
Ranges
ACCURACY --
# Within
Control control
Limits (X) Limits
* Outside
control
Limits
Total
No. of
Analyses
RPD
Ranges
•—PRECISION
* Within
Control Control
Limits Limits
* Outside
control
Limits
Parameter
Aqueous Samples*^
Metals: (Cont'd)
Chromium
Copper
Lead
Mercury
Selenium
Silver
w
TOTAL
a
8
8
J
9
2
64
Residue (Total Solids) N/A
(1)
<2i
N/J
105-113
96-106
98-109
85-100
0-102(3)
95-106
75-125
75-125
75-125
75-125
75-125
75-125
B
a
B
3
6
2
60
'
fnlLf,rPi!Si lnCiUl? 2" S°lld 8trearas frow the |OM and
h'e Ss "„ h^ STr co ± rerth, .
lot * X H reP°rted Wlth ' '°°tn0te statlna
B
B
B
3
6(3)
2
63
K
0.17-5.5 0-20
0.83-5.4 0-20
0.67-10 0-20
3.3-12 0-20
8.1-17 0-20
6.5-U 0-20
8
B
8
3
8
2
0.37-21 0-30
63
K
•oil Mashing tests (SWL/SUH-CO/WS/FO/FPC/FS).
a posfdiaestion spike indicated matrix Interference. No RPD was reported.
-------
chemical counterparts of congeners of interest. All soil samples except: one,
SWH-WS-05-03, had acceptable recoveries for all surrogate compounds. This sample
had recoveries below the lower control limit for all surrogates. Recoveries
averaged around 20% (lower control limit = 40%) indicating that the results for
this sample may be biased low due to inefficient extraction. This same sample,
however, was re-extracted and results were about 40% higher. This extraction was
still below acceptable QC limits indicating a possible matrix effect. Because
this was the only sample with poor surrogate recovery very little impact is
expected upon data quality.
NON-CRITICAL ANALYSES
Accuracy for total metals is shown by matrix spike recoveries in Table 6-7.
Almost all analyses were within QAPjP specifications. Average recovery for TOG
for all solid analyses as computed from surrogate spike data was 83%. This was
well within the QAPjP objective of 65-135%. TOG recoveries for all aqueous
matrices of the soil analyses were within the QAPjP specified control limits of
50-150%. Matrix spikes for TRPH solid and aqueous samples were within the
specified control limits of 60-120%. EP Toxicity accuracy was determined by
matrix spikes for metal analyses as shown in Table 6-7. Surrogate spike results
for TCLP analyses are presented with the semivolatile matrices in Tables 6-1 and
6-2. As previously noted, surrogate data which were outside specified control
limits were not used.
PRECISION
PENTACHLOROPHENOL
Precision for these two analyses was determined by taking the relative
percent difference (RPD) of field and/or laboratory duplicates or the relative
standard deviation (RSD) if 3 or more replicates were sampled and analyzed.
Results of replicate measurements for the pentachlorophenol data for soil
analyses are presented in Tables 6-8 and 6-9. In addition, as noted in the
discussion on accuracy, MS/MSD data served as an additional measure of analytical
precision (see Tables 6-3 and 6-4 for semivolatile data and Table 6-7 for other
non-critical analyses).
As shown in Table 6-8, all but one group of replicate penta measurements
were within the QAPjP precision specifications for every matrix in both the high
and low concentration soil tests. The single replicate measurement that was
outside the specifications was for the washed soil matrix in the low concentra-
tion soil demonstration. The projected precision specified in the QAPjP was based
upon anticipated sampling and analytical variability associated with the
semivolatile measurements. The observed variability was well under these
specifications for one of the washed soil replicate measurements and well above
specifications for the other duplicate pair. Because it is expected that there
are no real matrix differences between these two tests, it appears that this one
duplicate result is the only measurement outside the normal range of variability
for the washed soil matrix.
204
-------
TABLE 6-8. LOW CONCENTRATION SOIL DEMONSTRATION - PRECISION
(FIELD DUPLICATES)
Sample # of
Type Replicate
(Matrix) Msmt Pairs
FS
US
CO
FO
FPC
MU
CDE
FS
WS
CO
FO
FPC -
MM
CDE
FS CU
Cr
As
US Cu
Cr
As
CO Cu
Cr
As
FO Cu
Cr
As
FPC Cu
Cr
As
RPD or QAPjP Total Total
RSD Range Control Uithin Outside
(X) Limits Limits Limits
- - Pentachlorophenol - -
2
2
1
1
1
ND
1
ND
3
2
1
ND
ND
ND
1
1
1
3
3
3
1
1
1
1
1
1
1
1
1
8-26
6-151
7
21
11
32
- -
1-22
5-10
8
- - Copper,
6.7
6.2
49
46-88
15-64
3-25
15
16
141
43
20
22
20
16
17
30 2
30 1
30 1
30 1
50 1
50
50 1
Total Residue - -
30
30 3
30 2
30 1
30
30
30
Chromium, Arsenic Analyses - -
20 1
20 1
20 0
20 0
20 1
20 2
20 1
20 1
20 0
20 0
20 1
20 0
20 1
20 1
20 1
0
1
0
0
0
-
0
0
0
0
-
-
-
0
0
1
3
2
1
0
0
1
1
0
1
0
0
0
Msmt - Measurement
FS - Feed SoiI
US - Uashed Soil
CO - Coarse Oversize
FO - Fine Oversize
FPC - Fine Particle Cake
MU - Municipal Water
CDE - Combined Dewatering Effluent
ND - Not Detected
205
-------
TABLE 6-9. HIGH CONCENTRATION SOIL DEHONSTRATION - PRECISION
(FIELD DUPLICATES)
Sample # of
Type Replicate
(Matrix) Msmts (pairs)
SPH
FS
US
CO
FO
FPC
MU
COE
ATW
FS
WS
CO
FO
FPC
MM
COE
ATW
WS Cu
Cr
Ar
FO Cu
CP
Ar
RPD or
RSD Range
(%)
QAPjP Total
Control Within
Limits Limits
Total
Outside
Limits
- - Pentachlorophenol - -
12
1
1
1
1
1
ND
2
1
2
2
2
3
2
ND
ND
1
38
20
29
14
7
37
-
14-25
19
- - Total
1-22
3-8
1-2
13-67
5-6
-
-
4
.
30 1
30 1
30 1
30 1
50 1
50 -
50 2
50 1
Residue - -
30 2
30 2
30 2
30 2
30 2
30
30
30 1
-
0
0
0
0
0
-
0
0
0
0
0
1
0
-
-
0
- - Copper, Chromium, Arsenic Analyses - -
1
1
1
1
1
1
9
25
70
63
38
15
20 1
20 0
20 0
20 0
20 0
20 1
0
1
1
1
1
0
Hsint - Measurement
SPW - Soil Pile High Concentration PCP
FS - Feed Soil
WS - Washed Soil
CO - Coarse Oversize
FO - Fine Oversize
FPC - Fine Particle Cake
MW - Municipal Water
CDE - Combined Dewatering
ATM - ATS Recycled Water
ND - Not Detected
Effluent
206
-------
Because this single result is high and variability measurements by duplicate
samples were limited, it would suggest that overall variability may be higher
than originally anticipated for the washed soil matrix. CCA measurements also
showed somewhat higher variability for this matrix (see below) . This increased
variability, however, will not affect project objectives. The project design
anticipated this type of problem by increasing the number of samples collected
and analyzed so that increased sampling and analytical variability could be
tolerated. While the number of valid sample analyses is less than the number
planned (see completeness objectives below) the statistical evaluation as
presented in Appendix C of the QAPjP shows that even with these increased sample
and analytical variabilities, project objectives for evaluating process
efficiency could still be achieved for the soil washing system. This statistical
design, as described in the QAPjP, is of critical importance in determining the
minimum number of required samples. Analysis of the washed soil samples showed
project objectives for determining penta removal were within specified confidence
limits. (See final report conclusions.)
OTHER CRITICAL AND NON-CRITICAL ANALYSES
Precision for total residue, copper, chromium, and arsenic is also presented
in terms of RPD in Tables 6-8 and 6-9. In general, all measurements were within
QAPjP specifications. (Again, some discrepancies were observed in the washed
soil matrix.) Table 6-7 shows precision for additional non-critical measurements
including total metals, and EP Toxicity and additional metals found in the
aqueous matrices. All analyses were within QAPjP specifications for precision.
In addition PCDD/PCDF analyses met required precision specifications for
duplicate samples. The PAH concentrations were so low that duplicate samples
showed non-detectable PAHs. So few PAHs were detected in any of the samples that
the lack of this precision estimate for this one parameter should have little
effect upon the quality of the data collected in terms of project objectives.
TOG, TRPH were non-critical measurements and therefore no precision data are
available. TCLP data (a non-critical measurement), in terms of semivolatiles,
also had primarily non-detectable compounds in the duplicate samples, and
therefore, precision data are not considered critical for this parameter.
COMPLETENESS
Tables 6-10 and 6-11 present completeness values for all critical analyses.
As rioted by these tables, all analyses did not meet the specified completeness
objective. The primary reason completeness objectives were not achieved was
because of the reduced number of samples collected. This was because the soil
had to be mixed in one large batch prior to starting the tests, and the estimated
volume of soil fell short of the required volume to complete the demonstration.
Therefore, once the soil was completely used up, new soil could not be obtained
because it would be a new batch of soil and would not be representative of the
test being conducted. Given the volume of soil required to complete the entire
demonstration it is not unusual that the estimated volume fell short of the
required testing volume. Even though the total number of samples was less than
planned this shortage did not affect project objectives. It became apparent to
project and QA management, after the project began, that the number of samples
needed in order to obtain project objectives was less than the completeness value
207
-------
Analysis
TABLE 6-10. '-COMPLETENESS SOIL WASHER"; LOW COHC. SOIL DEMONSTRATION
Proposed Collected Satisfactory Analysis Percent Conplete
QAPjP Objective
'
- - Critical - -
Pentach loropheno I
PAHs
Cu, Cr, As
Total Residue
Dioxins/Furans
- - Non-Critical - -
Total Metals
TOC
Total Rec Petro Hydro
EP TOX
TCLP semi dioxin/furan
N>
o
CO
-. - Critical - -
Pentachlorophenol
PAHs
Cu, Cr, As
Total Residue
Dioxins/Furans
- - Non-Critical - -
Total Metals
TOC
Total Rec Petro Hydro
FS WS CO
11 11 11
11 11 11
11 11 11
11 11 11
2 2
2 2
999
999
2 2
2
MW
7
7
7
7
2
5
5
FO FPC
11 11
11 11
11 11
11 11
2 2
2 2
9 9
9 9
2 2
CDE
11
11
11
2
2
9
9
- - Collected Solids - -
FS WS CO FO FPC FS WS CO FO FPC FS US CO FO FPC FS WS CO
8888 8 88888 7373737373 90 90 90
8 7 8 8 8 87888 7364737373 90 90 90
90 90 90
8888 8 8888 8 7373737373 90 90 90
90 90 90
- - Collected Liquids - -
MU CDE MW CDE MW CDE MW
7 11 7 11 100 100 90
7 11 . 7 11 100 100 90
90
38 38 43 73 90
90
FO
90
90
90
90
90
CDE
90
90
90
TU
90
90
FPC
90
90
90
90
90
WS - Washed Soil
CO - Coarse Oversize
FO - Fine Oversize
FPC - Fine Particle Cake
MW - Municipal Water
CDE - Combined Dewatering Effluent
-------
Analysis
TABLE 6-11. "COMPLETENESS SOIL UASHER"; HIGH COHC. SOIL DEMONSTRATION
Proposed Collected Satisfactory Analysis Percent Conplete
FS - Feed Soil
US - Washed Soil
CO - Coarse Oversize
FO ^ - Fine Oversize
FPC - Fine Particle Cake
ATU - ATS Recycled Water
HW - Municipal Water
CDE - Combined Dewatering Effluent
QAPJP Objective
- - Critical - -
Pentachlorophenol
PAHs
Cu, Cr, As
• Total Residue
Dioxins/Furans
- - Non-Critical • -
Total Hetals
TOC
Total Rec Petro Hydro
EP TOX
S3
O
vo
- - Critical - -
Pentachlorophenol
PAH&
Cu, Cr, As
Total Residue
Dioxins/Furans
- - Non~Critical - -
Total Metals
TOC
Total Rec Petro Hydro
FS WS
27 27
27 27
27 27
27 27
3
6
25 25
25 25
6
ATW
15
15
15
15
13
13
CO
27
27
27
27
3
6
25
25
6
MW
15
15
15
15
3
13
13
- - Collected Solids - -
FO FPC FS WS CO FO FPC FS WS CO FO FPC
27 27 21 21 21 21 21 20 19 18 21 19
27 27 21 21 21 21 21 20 16 18 21 19
27 27
27 27 21 21 21 21 21 21 20 21 22 20
33
66
25 25
25 25
6 6
- - Collected Liquids - -
CDE ATW HW CDE ATW HW CDE
27 12 12 22 11 12 22
27 12 12 22 11 12 22
27
27 12 12 21 11 8 14
3
6
25
25
FS WS CO FO FPC FS WS
U 70 67 78 70 90 90
74 59 67 78 70 90 90
90 90
78 74 78 81 74 90 90
90 90
ATW HW CDE ATW
73 80 81 90
73 80 81 90
90
73 53 52 90
90
CO
90
90
90
90
90
HW
90
90
90
90
90
FO
90
90
90
90
90
CDE
90
90
90
90
90
ppc
90
90
90
90
90
-------
specified in the QAPjP. As demonstrated by the conclusions presented in the
report, objectives for the soil tests in terms of penta removal efficiency were
not affected by the shortage of samples obtained and analyzed.
6.4.3 BioTrol Aqueous Treatment System Analyses
The BATS was run simultaneously with the soil washer. Water effluent from
the soil washer was diverted into a holding tank for storage. This effluent was
later used to demonstrate the effectiveness of the aqueous treatment system. The
test was scheduled to last 12 days but was shortened to 10 days due to mechanical
pump failure. Even though the test was shortened by several days, a sufficient
number of samples was still collected to determine the efficiency of the system.
The following analyses were conducted: Penta, PAHs, Cu, Cr, As, total residue,
PCDDs/PCDFs, total metals, TRPH, total organic chloride, chloride ion, and
chemical oxygen demand.
ACCURACY
Penta and PAH COMPOUNDS
Accuracy for the aqueous treatment system was evaluated by percent recovery
for surrogate and matrix spike samples. Results for surrogate spikes for PAH and
penta analyses are presented in Table 6-12. As noted in the discussion for soil
analyses, the same criteria for evaluating surrogate spike recovery were used for
the aqueous treatment samples. If the sample did not meet specified recovery
criteria the result was not used for determining recovery efficiency.
As with the soil analyses, Terphenyl - d!4 and 2-Fluorophenol were
particularly difficult surrogates to recover. Because re-extractions were
performed whenever a single surrogate was outside its recovery range, these poor
recoveries should be attributed to matrix interferences.
In addition to the ^surrogate spike recoveries MS/MSD results are shown in
Table 6-13. (Pentachlorophenol data are not included in this table.) Similar
to the soil analyses, most PAH recoveries are within specified limits; however,
the PAH removal efficiency for the aqueous treatment system is not critical or
even very meaningful because most native PAH compounds were at extremely low
concentrations. Therefore, they were of little significance. (Removal efficiency
for PAHs in the aqueous treatment system was not calculated due to the low
concentrations of compounds.)
Penta concentrations were very high in the native sample and matrix spikes
were spiked too low to accurately determine recovery efficiencies. (See
discussion in section on soil analyses.) Table 6-3, however, presents precision
for penta recovery from MS/MSD data. Matrix spike data for penta are of little
use in determining accuracy based on percent recovery.
210
-------
TABLE 6-12. ACCURACY - SURROGATE SPIKES, AQUEOUS TREATMENT SYSTEM (ATS)
INFLUENT AND EFFLUENT SAMPLES
Total
Aqueous No. of % Recovery
# Within # Outside
Control Control Control
Analyte Analyses Ranges Limits (%) Limits Limits
NBZ
FBP
TPH
PHL
2FP
TBP
FLT
NBZ
FBP
TPH
PHL
2FP
TBP
FLT
98
98
98
98
98
98
98
Nitrobenzene -d5
2 - Fluor obiphenyl
Terpheny 1 - d!4
Phenol -d5
2 - Fluoropheno 1
2 , 4 , 6 - Tr ib r omopheno 1
Fluoranthene - dlO
0-98
51-110
8-73
7-96
0-80
0-110
22-130
,
35-114
43-116
33-141
10-94
21-100
10-123
20-130
97
98
64
96
74
96
98
1
0
34
2
24
2
0
211
-------
TABLE 6-13. MS/MSDS FOR ORGANICS ON ATS EFFLUENT SAMPLES
MS % MSD %
REC REC
%
RPD
QAPjP
QC Limits
Accuracy
(% REC)
No.
Within
QC Limits
MS - EPA Sample No. ATSEFF073
Water
Naphthalene
4-Chloro-3-methylphenol
Acenaphthene
4-Nitrophenol
Pyrene
2,3,5,6-TCP
59
60
68
54
49
38
88
71
90
63
18
45
39
17
28
15
93
17
40-120
23-97
46-118
10-80
26-127
10-110
2
2
2
2
1
2
MS - EPA Sample No. ATSEFF091
Water
Naphthalene
4-Chloro-3-methylphenol
Acenaphthene
4-Nitrophenol
Pyrene
2,3,5,6-TCP
72
66
75
96
66
54
82
56
82
106
52
54
13
16
9
10
24
0
40-120
23-97
46-118
10-80
26-127
10-110
2
2
2
0
2
2
MS - EPA Sample No. ATSEFF083
Water
Naphthalene
4-Chloro-3-methylphenol
Acenaphthene
4-Nitrophenol
Pyrene
2,3,5,6-TCP
77
59
72
90
51
NA
67
65
68
75
64
NA
14
10
6
18
23
NA
40-120
23-97
46-118
10-80
26-127
10-110
2
2
2
1
2
Not Spiked
MS - EPA Sample No. ATSEFF072
Water
Naphthalene
4-Chloro-3-methylphenol
Acenaphthene
4-Nitrophenol
Pyrene
2,3,5,6-TCP
94
84
118
53
103
56
101
94
123
46
114
68
7
11
4
14
10
19
40-120
23-97
46-118
10-80
26-127
10-110
2
2
1
2
2
2
Precision objectives specified in QAPjP are for field, not
analytical, duplicates.
212
-------
As with the soil analyses, 2,3,5, 6-TCP was used as a surrogate compound for
penta recovery. While penta matrix spikes are non-representative due to the
spiking concentration being too low, 2,3,5,6-TCP recovery should be considered
as a surrogate for penta (this was noted in the QAPjP). Similar to the soil
matrices, 2,3,5,6-TCP recoveries are within acceptable limits if compared to QC
limits for penta. While it would be better to have penta recovery information
for determining accuracy for penta, having designated 2,3,5,6-TCP in the QAPjP
as an additional spiking compound proved beneficial because of the problem
encountered with penta spiking. Based upon the recovery efficiencies for
surrogate spikes, as well as the recovery efficiencies for PAH matrix spike
compounds and for 2,3,5,6-TCP, it is reasonable to conclude that accuracy for
penta and PAHs in the aqueous treatment samples was well within QAPjP
specifications. Therefore, conclusions concerning removal efficiency for penta
in the aqueous treatment system were calculated using acceptable quality data.
COPPER, CHROMIUM, ARSENIC and TOTAL RESIDUE
Accuracy for copper, chromium and arsenic (CCA) analysis of influent and
effluent ATS samples was assessed by laboratory matrix spike results. Results
are summarized in Table 6-14. Arsenic was the only metal having an outlier
recovery value.
Accuracy for total residue measurements is not readily assessed by routine
spiking procedures. The accuracy of the values may be affected by analysis
performed past the method-recommended holding time. Most ATS samples were
analyzed for total solids between one and 41 days past the hold time. However,
these residue measurements were not used in the evaluation of the ATS
performance, and therefore, they had minimal impact on the project technical
objectives.
PCDDs/PCDFs
As with other matrices, accuracy for PCDDs/PCDFs was determined using
recovery results for C13 isotopes for each of the congeners. All results were
within QAPjP specifications. While these analyses were designated as secondary
critical measurements, the data were not used in determining proj ect obj ectives.
The reason for dioxin analyses was to evaluate the quality of the effluent water;
these analyses were not designated as part of the developer claims.
NON-CRITICAL ANALYSES
Spiked samples were used to assess accuracy for the non-critical analysis
of chloride and additional metals (barium, cadmium, lead, mercury, selenium and
silver, see Table 6-14). Only selenium showed any potential bias in recovery
with four of ten spike recoveries below control limits. Since selenium was not
detected in any of the effluent samples analyzed, the impact is limited to a
potential bias in the reported detection limit. Two equipment blanks contained
barium at 11.2-11.3 jig/1; sample concentrations ranged from 3-4 times higher.
213
-------
TABLE 6-14. QC SUMMARY:
PRECISION AND ACCURACY - COPPER CHROHIUH ARSENIC. NON-CRITICAL HETALS. RESIDUE.
CHLORIDE AQUEOUS TREATMENT SYSTEH SAMPLES
Parameter
Total
No. Of
Analyses
X Recovery
Ranges
- 1
Control
Limits (X)
RbUJKACT ---
* Within
Control
Limits
« Outside
control
Limits
Total
Ho. of
Analyses
RPD
Ranges
PRECISION
« Within
Control Control
Limits Limits
* Outside
Control
Limits
Arsenic
Barium
Cadniui)
Chromium
Copper
lead
Hercury
Selenium
Silver
TOTAL
Residue
Chloride
NOTES:
(1)
11
a
a
8
3
10
2
66
N/A
2
67-94
96-104
98-113
105-113
96-104
98-109
85-100
0-102
95-106
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
10
8
8
8
8
8
3
6
2
96.6-108 N/A
61
11
a
8
a
a
a
3
(1)
65
15
1
1.2-10
fl.61-4.9
0.9-8.7
0.17-S.5
0.83-5.<
0.67-10
3.3-12
8.1-17
6.5-14
0.37-21
1.3
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-30
N/A
11
a
8
8
8
8
3
9
2
65
15
1
The HS with OX recovery Has reported with • footnote stating that • post-digestion spike indicated matrix interference. No RPD was reported.
-------
One chloride analysis exceeded the 28-day method holding time by 30 days.
This analysis has little impact upon overall data quality.
PRECISION
PENTACHLOROPHENOL AND PAHs
Precision for these analyses was determined from laboratory and field
splits. RPD and RSD determinations were obtained from duplicate and replicates
samples. Results of these analyses showed all replicates were within specified
QAPjP limitations. A table has not been included because there were no
deviations.
COPPER, CHROMIUM, AND ARSENIC
Precision for CCA analyses was measured by the relative percent difference
(RPD) between spiked duplicates. As summarized in Table 6-14, all RPD values
were within control limits.
TOTAL RESIDUE
Precision for total solids was determined by the analysis of duplicate
samples. As noted in Table 6-14, all reported RPD values were within QAPjP
specifications.
Of the three field duplicates collected from the ATS influent and effluent
streams, only one RPD exceeded the precision control limits.
PCDDs/PCDFs
Precision for this measurement was evaluated by replicate sampling. One
field duplicate for the aqueous treatment system samples had an RPD of 71%. This
was outside QAPjP specifications. Because these analyses were not used in the
determination of removal efficiency this result should have no effect on project
obj ectives. The reason PCDD/PCDF analyses were performed was to determine the
quality of the effluent water.
NON-CRITICAL ANALYSES
Precision measurements for the additional metal and chloride were all within
control limits. (See Table 6-14.)
COMPLETENESS
Completeness results for the aqueous treatment system are presented in Table
6-15. Completeness for the field measurements associated with the aqueous
treatment system are presented in Table 6-16. As seen by the data presented in
these two tables, completeness fell short of the QAPjP objective. During the
demonstration the test had to be shortened by two days because of pump failure,
and consequent toxic shock experienced by the micro-organisms. While the test
was shortened and proposed data were not collected, the results obtained were
used to compute an average efficiency and associated range. As presented
215
-------
TABLE 6-15. COMPLETENESS FOR AQUEOUS TREATMENT SYSTEM (ATS)
Sample
Location
Influent
Effluent
Carbon
Bag Filter
# of Samples for
Analytical which Results
Parameter are Satisfactory
Penta
PAHs
CCA
Total Residue
TRPH
COD
TOX
CL
Penta
PAHs
CCA
Total Residue
TRPH
COD
PCDD
MTLS
TOX
CL
Penta
PAHs
Total Residue
PCDD
No samples were
32
32
37
30
37
37
6
6
33
33
37
29
36
36
5
9
6
6
4
4
4
0
collected of the
if of Samples for
which Results
were Planned *
48
48
48
48
48
48
6
6
48
48
48
48
48
48
6
12
6
6
1
1
1
1
bag filter residue
Completeness
(%)
67
67
77
63
77
77
100
100
69
69
77
60
75
75
83
75
100
100
400
400
400
0
QAPjP
Objective
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
* This number is based on a test of 12 days, however the test lasted only 10 days.
is the primary reason completeness (%) did not meet the QAPjP objective.
This
216
-------
TABLE 6-16. ATS FIELD MEASUREMENTS: COMPLETENESS
Number of Number of
Measurement Measurements Measurements Completeness
Taken Planned *
Influent
Flow Rate
Effluent
Flow Rate
Nutrient
Flow Rate
pH Chemicals
Flow Rate
Influent pH
Effluent pH
Influent Temp
Effluent Temp
Power
Carbon Residue
Weight
106
100
98
1
32
32
32
32
27
1
144
144
144
2
24
24
24
24
24
1
74%
69%
68%
50%
133%
133%
133%
133%
113%
100%
*This number is based on a test of 12 days, however the actual
test lasted only 10 days.
217
-------
previously in the report, sufficient data were collected to complete the primary
objective and determine the efficiency of the system for degrading penta.
Because the concentrations of PAHs were often below the analytical detection
limits, data were calculated for these compounds.
6.4.4 Slurry Bioreactor
The slurry bioreactor was the third system tested during this demonstration.
Once a sufficient slurry had been collected from the soil washer (fine particle
cake mixed with combined dewatering effluent) this slurry was used to
demonstrate the efficiency of the slurry bioreactor for penta and PAH removal.
(The main section of the report discusses the composition of this slurry.) The
following critical analyses were performed: penta, PAHs, Cu,Cr,As, total residue,
and PCDD/PCDFs. In addition the following non-critical analyses were also
conducted: TRPH, total organic chloride, chloride ion, and chemical oxygen
demand.
ACCURACY
PENTACHLOROPHENOL AND PAHS
Surrogate spike data for semivolatile analyses (water and soil fractions)
are presented in Tables 6-17 and 6-18. All surrogate spikes were within QAPjP
specifications. Similar to the soil analyses all matrix spikes were too low to
determine recovery efficiency. The MS/MSD data are therefore presented for
purposes of demonstrating analytical precision only. MS/MSD data in terms of RPD
are presented in Table 6-3.
MS/MSD recoveries for PAH compounds and 2,3,5,6-TCP are included in Table
6-19. In general recoveries were within QAPjP specifications. As with the soil
matrices. 2,3,5,6-TCP was used as a surrogate for penta recovery. Because
recovery of this compound was within specified limits for penta, and because
2,3,5,6-TCP should be chemically similar to penta, these data along with the
surrogate spike data indicate that penta values for the slurry bioreactor samples
are within acceptable limits.
TOTAL RESIDUE
Residue determinations were performed on the slurry samples after filtering
through a 0.7 |im glass fiber filter. The resulting filtrate was analyzed for
total solids (actually dissolved), and the filtered solid (filter cake) was
analyzed for percent solids. .Accuracy for residue measurements is not readily
assessed by routine spike and control sample analysis. Accuracy for the total
residue analyses was therefore determined by calibration checks for the
analytical balance used to weigh the samples. Balance calibration was confirmed
during the technical systems review at the subcontracting laboratory, and all
determinations were within QAPjP specifications.
218
-------
TABLE 6-17. ACCURACY - SURROGATE SPIKES
SLURRY BIOREACTOR (SOIL FRACTION)
Analyte
Nitrobenzene-d5
2- F luorobi pheny I
Terphenyl-d14
Phenol-d5
2-Fluorophenyl
2 , 4 , 6- T r i bromopheno I
Fluoranthene-d10
Total
Measurements
29
29
29
29
29
29
29
Percent
Recovery
Range
28-73
41-100
39-99
32-66
38-75
55-98
40-95
QC
Control
Limits
23-120
30-115
18-137
24-113
25-121
19-122
15-140
Total #
Within
Limits
29
29
29
29
29
29
29
Total #
Outside
Limits
0
0
0
0
0
0
0
219
-------
TABLE 6-18. ACCURACY - SURROGATE SPIKES
SLURRY BIOREACTOR (WATER FRACTION)
Analyte
Nit r obenzene - d5
2-Fluorobiphenyl
Terphenyl-d14
Phenol-d5
2-Fluorophenyl
2,4,6-Tribromophenol
Fluoranthene-d10
Total
Measurements
29
29
29
29
29
29
29
Percent
Recovery
Range
41-93
48-92
44-83
28-90
37-91
51-105
47-92
QC
Control
Limits
35-114
43-116
33-141
10-94
21-100
10-123
20-130
Total #
Within
Limits
29
29
29
29
29
29
29
Total #
Outside
Limits
0
0
0
0
0
0
0
220
-------
TABLE 6-19. MS/MSD DATA FOR SLURRY BIOREACTOR INFLUENT (PAH COMPOUNDS ONLY)
# of % Rec Control # Within # Outside RPD RPD Control # Within # Outside
Compound Analytes Range Limits Limits Limits Range Limits Limits Limits
AQUEOUS MATRICES
Naphthalene 6 55-70 30-120 6 0 1-4 40 3 0
4-Chloro-3-methylphenol 6 47-75 23-97 6 0 4-14 40 3 0
Acenaphthene 6 56-71 46-118 6 0 0-10 40 3 0
4-Nitrophenol 6 50-104 10-80 4 2 1-35 40 3 0
Pyrene 6 45-74 26-127 6 0 7-13 40 3 0
2,3,5,6-TCP 6 28-49 ** - - 9-22 40 -
221
-------
The accuracy of the residue determinations may be affected by analyses
performed beyond the method-recommended holding times. All SBR filtrates were
analyzed for total solids between 25-55 days past the hold time; all SBR solids
were analyzed for percent solids between 13-39 days after the holding time
expiration. Residue measurements were used in mathematically determining the
organic constituent (penta and PAHs) concentrations in the total sample (after
analyzing the filtrate and the filter cake separately). It is not possible to
estimate the effect of the exceeded holding times; however, sealed samples were
stored at 4°C at all times prior to analysis, thus minimizing moisture losses due
to evaporation.
COPPER, CHROMIUM, ARSENIC
Accuracy was determined by the analysis of laboratory matrix spikes for
copper and chromium. Table 6-20 shows these analyses to be under control.
Arsenic was analyzed by method SW-6010, and there were not any spikes associated
with the sample analyses of the SBR samples. All these slurry samples were
treated as aqueous samples due to the nature of the samples and the low solids
content.
NON-CRITICAL ANALYSES
Accuracy for additional metals (barium, cadmium, lead, mercury, selenium and
silver), chloride and metals on the EP Toxicity leachates was determined by the
analysis of matrix spikes (see Table 6-20). The one selenium spike (selenium not
in any of the samples) associated with the effluent samples had a low recovery,
reported with a note attributing it to matrix effect. All other recoveries were
within control limits.
Two samples exceeded the 28-day hold time for chloride by five and seven
days.
No accuracy data are available for COD or TRPH. Because these data were not
used to determine project objectives no impact is expected on overall data
quality. Matrix spikes for TOX on four samples resulted in two recoveries
slightly above the control limits, but other results were within specifications.
PRECISION
All precision data for the analyses of the SBR are presented in Table 6-21.
These data are based upon field duplicates, and therefore indicate combined
sampling and analytical variability.
PENTACHLOROPHENOL
Precision was calculated by determining the relative percent difference
(RPD) of measurement pairs for liquid and solid fractions of both the influent
and effluent samples. RPDs are presented in Table 6-21. In addition, analytical
precision is presented in Table 6-3 which shows the results of MS/MSD pairs. As
shown by the results presented in Table 6-21, all but one sample was within the
QAPjP specified precision. This specification was based upon the best estimate
222
-------
TABLE 6-20. 8C SUSU87: PfiECSSItS AND ACCURACY - COPPER. EKSCMHH. ARSEHIC. ttg-CSITICAL METALS. RES1MJE, CHLORIDE.
EP TOXICW SLURRY BIOREACTOt SAMPLES
w
Parameter
Hetals:(1)
Arsenic
Barium
Caciniun
Chromium
Copper
Lead
Hercury
Seleniun
Silver
TOTAL
Residue-filtrate
-solids
EP Toxicity:
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Selenium
Silver
Chloride
NOTES:
'** fillirrw eamnl«« t
Total
No. of X Recovery
Analyses Ranges
0
2
2
2
2
2
2
1
2
15
N/A
N/A
3
:»A« *B*_ *BB
-s(2>
B2-106
85-93
88-98
76-83
91-100
92:106
30(1J
103-113
--
—
90
97
91
107
96
101
113
105
105
B5.9-103
- ACCURACY
« Uithir « Outside
Control control control
Limits (X) Limits Limits
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
75-125
--
--
75-125
75-125
75-125
75-125
75-125
W-12S
75-125
75-125
75-125
75-125
0
2
2
2
2
2
2
0
2
14
--
--
3
0
0
0
0
0
0
0
1
0
1
--
--
., 0
0
0
0
0
0
0
0
0
1 0
Total
Mo. at
Analyses
0
2
2
2
2
2
2
1
2
15
1
2
3
PRECISION
RPD Control
Ranges Limits
HS(2)
2.8-3.6
2.6-14
2.3-4.3
2.7-4.3
1.9-3.9
5.3-17
24
6.2-6.4
0.7
2.0-4.1
4.3
17
S.5
6.5
2.1
1.7
S.6
0.38
0.76
1.1-5.3
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-30
0-30
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
t Uithln * Outside
Control Control
Limits Limits
0
2
2
2
2
2
2
0
2
14
1
2
3
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
o
0
0
0
0
0
0
(J) No spike recoveries were reported utth the arsenic results of the SB« satf>ie*~'anaiyied by 511044 Hethod 6010.
This recovery Mas reported with a footnote stating that the low recovery Has due to matrix effect.
-------
TABLE 6-21. PRECISION SLURRY BIOREACTOR
FIELD DUPLICATES
Sample # Msmt
type Pairs
Liq Effluent
Liq Influent
Solid Effluent
Solid Influent
Liq Effluent
Liq Influent
Solid Effluent
Solid Influent
Influent
Effluent
Compounds
Detected
QAPjP Total Total
RPD Control Within Outside
% Limits Limits Limits
- - Pentachlorophenol - -
1
1
1
1
1
1
1
1
1
1
1
1
undetected
undetected
fluoranthene
pyrene
fluoranthene
pyrene
b anthracene
chrysene
11
62
4
16
- - PAHs - -
12
0
3
36
27
13
50
50
30
30
50
50
50
50
50
50
1
0
1
1
- •
-
1
1
1
1
1
1
0
1
0
0
-
-
0
0
0
0
0
0
- - Copper, Chromium, Arsenic Analyses - -
1
1
1
1
1
1
As
Cr
Cu
As
Cr
Cu
0
1
1
6
7
5
20
20
20
20
20
20
1
1
1
1
1
1
0
0
0
0
0
0
b - Benzo
MSMT - Measurement
RPD - Relative Percent Difference
224
-------
TABLE 6-21. PRECISION SLURRY BIOREACTOR
FIELD DUPLICATES (CONTINUED)
Sample # Msmt
Type Pairs
Effluent 1
1
1
1
1
I 1
I 1
I
QAPjP Total
Compounds RPD Control Within
Detected % Limits Limits
- - Dioxins/Furans - -
HxCDD 8
HpCDD 11
OCDD 16
PeCDF 30
HxCDF 12
HpCDF 32
•OCDF 18
40 0
40 0
40 0
40 0
40 0
40 0
40 0
Total
Outside
Limits
0
1
0
0
I - - Total Residue - -
I
Influent | 1
i
i
Effluent | 1
i
i
I
i
I
Effluent | 1
I 1
I
I
I
I
Effluent | 1
. I
I
i
influent | 1
i
l
Effluent | 1
I
4
30
- - Metals - -
Barium 12
Lead 3.9
Mercury 8.5
- - TOX (NC)
144
--CO
35
4
30 1
30 1
20 1
20 1
20 1
—
20 0
20 1
40 1
40 1
0
0
0
0
0
1
0
0
0
i
I -- Chloride Ion (NC)
i
i
Effluent | 1
1
10 1
0
Msmt - Measurement
NC - Non Critical Measurement
Note - Total Recoverable Petrolem Hydrocarbons (TRPH) were not
detected in the field duplicate samples.
225
-------
of sampling and analytical variability available when the QAPjP was written. As
noted by the statistical calculations in appendix C of the QAPjP, however, a much
larger variation was still acceptable in order to accomplish project objectives
provided a large number of samples were obtained. (See section below on
completeness.)
Conclusions concerning the efficiency of the process are not affected by
this single precision result which was outside QAPjP specifications. Conclusions
presented in the report on the efficiency of the slurry bioreactor are only an
estimate of the best observed efficiency and are not used as supporting evidence
of the developer's claim for the process. Only a small number of representative
samples were obtained during the operation of the slurry bioreactor because the
system was nutrient starved for a majority of the sampling period. This
occurrence was only evident after samples had been collected and analyzed by the
laboratory and the demonstration had been completed. Additional information
concerning the slurry bioreactor efficiency can be found in section 6.2, entitled
"Conclusions and Limitations of Data", and in the main body of the report.
PAHs
As shown in Table 6-21, all precision determinations are within the
specified objective for any of the PAH compounds detected. It is therefore
anticipated that variability due to precision error will have no effect upon
conclusions concerning the efficiency of PAH destruction.
COPPER, CHROMIUM, ARSENIC
All precision determinations were well under the specifications presented
in the QAPjP. No significant effect is therefore anticipated with relation to the
conclusions presented (see Table 6-21).
TOTAL RESIDUE
Total residue precision for both influent and effluent samples were within
QAPjP specifications based upon RPDs of duplicate pairs.
PCDD/PCDF
One field duplicate was analyzed for PCDDs/PCDFs as shown in Table 6-21.
Sampling and analytical precision was well within QAPjP specifications for all
congeners. A slurry matrix laboratory duplicate was also performed, resulting
in RPD values for the various congeners ranging from 21-82%; the average RPD =
37.8%. (The QAPjP objective for method precision of the PCDD/PCDF analysis was
an RPD = 40%.) PCDD/PCDF analysis was conducted only to estimate relative
dioxin/furan levels in the effluent sample. While it was noted as a secondary
critical measurement, it was not used to determine process efficiency. In
addition, several (7) effluent samples were analyzed, so the concentration of
dioxin/furan potentially exiting the SBR is well characterized. While the method
precision for the laboratory duplicate shown above was not within QAPjP
specifications for all congeners, the duplicate field sample showed total
variability to be within specified limits. Based upon these results no
significant data quality problem is expected.
226
-------
NON-CRITICAL MEASUREMENTS
All duplicate samples for the non-critical measurements were within QAPjP
specifications except for total organic chloride. This is the only measurement
outside specified objectives, and it is not considered critical because
conclusions presented in this report are based only upon data from critical
analyses.
No data existed for precision of the total recoverable petroleum hydrocarbon
analysis because all samples reported undetectable levels for this analysis.
COMPLETENESS
Completeness objectives are presented in Table 6-22 for all measurements
performed during the evaluation of the slurry bioreactor. Based upon the
original QAPjP calculations, completeness objectives were not achieved for all
critical measurements. The definition for completeness as presented in RREL's
more recent QA guidance documents suggests that completeness objectives naist be
based upon a statistical evaluation in relation to project needs for
accomplishing project objectives in terms of decision making criteria. It is
this latter definition of completeness that is more relevant to this project.
As previously noted, Appendix C of the QAPjP presents a statistical evaluation
which defines the number of samples required for achieving project objectives in
terms of evaluating process efficiency based upon the analytical and sampling
variability associated with those samples. Using this evaluation, project and
QA personnel have now agreed that original completeness objectives were higher
than necessary given the actual variability of these measurements. This
statistical evaluation was also reviewed with RREL's QA Contractor prior to the
demonstration and was found to be acceptable. These conclusions suggest that the
completeness achieved from the analytically valid SBR data was within statistical
specifications for determining whether the process was meeting the developer's
claims.
REPRESENTATIVENESS
For purposes of this demonstration two samples were collected. Influent and
effluent contaminant concentrations were measured and reported for all the above
analyses. While the designated number of samples were collected and analyzed,
it was discovered that a majority of these samples were collected prior to slurry
activation. Since this is a biological system, organisms require an activation
start-up period based in part upon residence time of the reactor. Calculations
to determine this residence time and consequently to determine peak performance
were used so that samples could be taken during the period of maximum
performance. Unfortunately, it was discovered after analysis of these samples
that peak performance could not be determined because this activation start-up
period had been incorrectly estimated. The micro-organisms had experienced toxic
shock, and process parameters had to be altered. Several samples were taken
227
-------
ro
- - Critical - -
Pentachlorophenol
PAHs
Cu, Cr, As
Total Residue
a) % Solids; Filter cake
b) TDS; Filtrate
Dioxins/Furans
EP TOX *
- - Non Critical - -
Total Rec Petro HC
Tot Organic Chloride
Chloride Ion
Chem. Oxygen Demand
Pentachlorophenol
PAHs
Total Residue
Dioxins Furans
Influent
18
18
18
18
18
16
3
3
16
18
18
18
18
18
8
16
3
3
16
15
15
15
15
15
15
3
3
15
# Samples with
Satisfactory
Analyses
nt Influent Effluent
15
15
15
15
15
7
US
14/14
14/14
15
15
7
US
15/15
15/15
15
15
11
7
3
15
3
3
15
15
3
3
15
- Additional SBR Samples (Carbon Canister)
1
15
3
3
15
Percent
Complete
Influent
78
78
83
83
39
94
100
100
94
Effluent
83
83
83
83
61
88
94
100
100
94
100
100
100
100
90
90
90
90
90
90
90
75
75
75
90
90
90
90
•Collected; however this sample was not specified in original QAPiP
US - Liquid Fraction/Solid Fraction
CAR Carbon Canister
-------
prior to this activation start-up period, and it became obvious from the final
effluent samples collected that process efficiency was continuing to rise.
Therefore, maximum efficiency could not be determined. While contaminant
concentrations were believed to be representative of the matrix, they were not
representative of peak sample performance as had been specified in the QAPjP.
As a result of this problem, peak performance of the system could not be
determined.
ADDITIONAL SBR SAMPLES
An additional SBR sample was collected from the carbon canister for analysis
of pentachlorophenol, PAHs, total residue, and PCDDs/PCDFs. Completeness and
precision objectives were achieved as specified for this sample. There were no
matrix spikes performed for this sample, nor were they specified in the QAPjP,
because this was not a critical measurement and results were not used for
determining proj ect obj ectives.
6.4.5 Blank Results
A series of equipment and method blank determinations were performed for all
critical analyses. As noted in the QAPjP there were 7 equipment blanks collected
and analyzed for pentachlorophenol, PAHs, and PCDDs/PCDFs. Of the analyses that
were performed, no significant amount of contamination was detected except for
one of the equipment blanks analyzed for PCDD/PCDF. The compound detected (OCDD
at 28 ng/1) was less than five times the reporting limit and significantly below
the sample concentrations reported for OCDD. Since this concentration was so low
and was detected in only one of seven blanks, it is not expected to adversely
affect the results. All other analyses for equipment blank and method blank
samples indicated no detectable contamination.
»
6.5 SPECIAL STUDIES
During the course of the follow-up technical systems review conducted at the
subcontractor laboratory by the contractor QA Manager, it was discovered that a
significant number of samples had been extracted past their holding time. (The
reasons for this occurrence are discussed in section 6.6 describing audit
findings.) Because this is a cause of major concern, the contractor QA Manager
and RREL QA Officer decided it was necessary to measure missed holding-time
effects for each soil matrix sampled to determine if the data were still of
sufficient quality for a SITE Demonstration project.
In order to evaluate the holding time effect, it was necessary to re-extract
and re-analyze samples which were originally extracted within their specified
holding times and then compare the analytical results. Holding time effects,
however, need to be distinguished from sampling and analytical variability. This
could be accomplished only by re-extracting replicate samples which were
originally extracted within specified holding times.
Matrices which exceeded their required extraction holding times included
washed soil, feed soil, fine particle cake, fine oversize, and coarse oversize,
229
-------
all of which are solid matrices. Except for the slurry bioreactor samples,
aqueous samples were not significantly affected, according to data received
during this first review. In reference to the slurry bioreactor, all those
samples were extracted past their holding times at intervals ranging from a few
days to 3 weeks. This occurred due to problems encountered in first attempting
to centrifuge and then switching to pressure filtration of samples prior to
extraction. Problems of sample extraction were due to the unusual matrix
encountered.
The contractor QA Manager specified that several soil samples be re-
extracted. These samples represented all five solid matrices. In addition,
several slurry samples were re-extracted for both solid and aqueous phases.
With the exception of the slurry samples, all the re-extracted samples were
replicates of samples initially extracted and analyzed within holding time
specifications, and they represented all matrices. Re-extraction of these
replicate samples provided data representative of extractions up to six weeks
past the holding time. Re-extracted slurry samples were treated somewhat
differently as noted below.
The preceding paragraphs describe the holding time problems and methods
which were used to evaluate analyte concentrations with respect to holding time.
Results of these re-extractions were compared to results of the original
extractions in order to determine if the time period past the holding time caused
significant changes in target analyte concentrations. It was known that a large
portion of the samples including each soil matrix was affected.
Briefly reviewing project objectives, it was stated that their purpose was
to determine Pentachlorophenol (penta) and polynuclear aromatic hydrocarbon (PAH)
removal efficiency in the soil washing, aqueous treatment, and slurry bioreactor
processes. To determine removal efficiency, the input and output analyte
concentrations must be known. Given the steady state nature of the process,
final values of grab samples were to be averaged and sample variation was to be
included in this final calculation. Statistical variation from sample to sample
was to be accounted for by taking a large sample set. The number of samples
obtained is therefore critical with respect to this final calculation.
The soil washing and slurry bioreactor samples were affected because these
solid extracts were known to have missed their holding time extraction date.
Because the total number of samples is critical with respect to determining input
and output concentrations, questionable data with respect to holding time could
not be discarded. If possible therefore, these data must be used and included
in the efficiency calculation for all systems examined.
As discussed in the main section of this report, analyte concentrations in
two of the soil matrices must be known to determine efficiency of the soil
washing system. These include the feed soil and the washed soil matrix. The
feed soil is obviously what goes into the system, and while four solid matrices
come out of the soil washer, the washed soil is what previously has been
designated as the "clean fraction." The other solid matrices, which include the
coarse oversize, the fine particle cake, and the fine oversize, are not used to
calculate process efficiency. The purpose of collecting these other fractions
230
-------
was to determine overall mass balance, which was also a primary project
obj ective.
Because all solid matrices, except the feed soil, were products of the
treatment system, analyte concentrations may be expected to decrease over time
(following collection) , as biological and physical/chemical processes within any
treated matrix (containing inocula and nutrients) may be expected to continue.
Once extracted, however, these processes slow down and target, analyte
concentrations are expected to be more stable. Consequently, holding times
specified in the QAPjP were primarily based on three conditions: 1) what was
known about these samples, 2) what were thought to be reasonable time p€iriods
needed to extract the samples, and 3) specifications for holding times noted in
SW-846.
The rate of decrease over time is unknown and of course will vary with
respect to storage conditions. Keeping the samples at 4 degrees centigrade
should slow compound degradation, but it will not stop it altogether. If this
rate of decrease is significant when compared to sample concentrations, and
degradation occurs in the time period between the specified holding time and the
time the samples were actually extracted, then data from the extractions would
not give accurate analyte concentrations at the time of sample collection.
Efficiency of the process is based upon conditions of the sample in the field at
the time of collection.
Results of the holding time studies are presented in Tables 6-23 to 6-27.
Analyte concentrations in each of the solid matrices for the soil washing system
are shown. The slurry matrix for the slurry bioreactor are not presented in
these tables but are discussed below. Samples extracted within the holding time
are compared to samples extracted several weeks passed their holding time. The
time frame chosen for samples purposefully extracted past their holding times is
a little longer than the actual extraction time of real samples, and it is a
compromise between actual extraction conditions and a reasonable time period in
which the laboratory could perform these re-extractions. Actual conditions for
samples extracted past their holding times are always less than the time frame
presented in these tables generated from the holding-time studies. These data
therefore represent a worse case scenario.
Because the Demonstration was not originally set up to determine the effects
of missed holding times, the number of samples we could use for this comparison
was limited. Replicate samples were chosen when available, and fortunately most
of the replicate samples in the study were extracted within the required holding
time so that this comparison could be made. The number of replicates, however,
is still limited, and therefore absolute statistical comparisons are difficult.
The tables present mean values for each analyte concentration, along with the
relative percent difference (RPD) or, if possible with triplicate samples, the
relative standard deviation (RSD). Except where indicated, concentrations were
within the instrument calibration range and surrogate concentrations were within
the specified recovery ranges.
Feed soil concentrations represent analyte concentrations before treatment
and therefore, as discussed, are critical for determining the process efficiency.
As presented in Table 6-23, duplicate samples were available for comparison. As
231
-------
TABLE 6-23. HOLDING TIME STUDY FOR FEED SOIL
Analyte Concentration
(Extracted Within Hold Time)
Pentachlorophenol
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)Anthracene
Chrysene
Benzo(b)Fluoranthene
Benzo(k)Fluoranthene
SWH-FS-05-5
•1420254
15180
19937
949694
36189
76556
50081
13172
22132
"8755
"9176
PPB
SWH-FS-05-6
"840000
15280
20480
92800
15600
76640
49920
12880
23840
16400
—
mean
1130127
15230
20208
93747
25895
76598
50000
13026
22986
12578
9176
RPD%
51
1
3
20
80
1
1
2
8
61
N/A
Analyte Concentration
(Extracted 50 Days After Hold Time)
SWH-FS-05-5
881352
"23149
—
100561
40513
75180
60797
** 17329
"25233
__
--
PPB
SWH-FS-05-6
864889
"18803
"18385
74500
45030
69577
51301
"12571
"24418
—
mean
873121
20976
18385
87531
42772
72379
56049
14950
24826
RPD%
2
21
N/A
30
11
8
17
32
4
N:
* Value above calibration curve (no greater than 2X the highest calibration point)
* * Value below calibration curve but above the method detection limit
— Non detected values because dilutions were extremely large
-------
TABLE 8-24. HOLDING TIME STUDY FOR WASHED SAND
(S3
CO
CO
Analyte Concentration
(Extracted Within Hold Time)
PPB
Pentachlorophenol
Acenaphthane
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)Anthracene
Chrysene
Benzo(b)Fluoranthene
Benzo(k)Fluoranthene
SWH-WS
05-5
162000
3400
3700
18000
7300
14000
13000
3400
6100
2000
2000
SWH-WS SWH-WS
05-7 05-7(DUP)
124000 124000
3200 "3200
3200 "3100
13200 13200
6100 -7600
11400 . 12200
11000 10000
2000 "2300
4000 *4400
1300 -1500
2000 *2000
mean
136700
3270
3300
14800
7000
12500
11300
2600
4800
1600
2000
RSO%
16
4
10
19
11
11
13
29
23
23
0
SWH-WS
05-5
128000
15000
78600
36000
710000
135000
M3000
29000
•13000
•10000
Analyte Concentration
(Extracted 44 Days Afer Hold Time)
PPB
SWH-WS SWH-WS SWH-WS
05-5(DUP) 05-7 05-7(DUP)
244000 83000 94400
12000 — 5300
— 5100 '5600
67000 19200 21000
30000 9200 10500
60000 20000 19000
113000 9000 11000
— 4000 *4300
26000 6400 '7100
9600 2400 '2600
12000 2300 *2500
Analyte Concentration
(Extracted 71 Days After Hold Time)
PPB
mean
137400
10800
5400
46500
21400
42500
67000
7100
17100
6900
6700
RSD%
54
46
7
66
63
63
99
72
70
76
75
SWH-WS
05-5
116000
2800
14000
5600
10000
13000
SWH-WS
05-7 mean RPD %
164000 140000 34
3700 3250 28
4000 4000 N/A
15000 14500 7
7700 6650 32
16000 13000 33
12000 12500 8
3000 3000 N/A
5000 5000 N/A
-
Duplicates represent analytical duplicates; samples 05-5 and 05-7 are field duplicates
Numbers above the Instrument detection limit but below the calibration curve
— Nondetected values because dilutions were extremely large
-------
TABLE 6-25. HOLDING TIME STUDY FOR COARSE OVERSIZE
Analyte Concentration
(Extracted Within Hold Time)
Pentachlorophenol
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)Anthracene
Chrysene
Benzo(b)Fluoranthene
Benzo(k)Fluoranthene
SWH-CO-05-5
857796
23766
24717
122866
53084
76188
61701
16355
36397
8773
8610
PPB
SWH-CO-05-6
1194077
26488
33379
127417
03807
86043
78752
21241
61436
11520
0121
SWH-CO-05-7
016910
16682
20368
08044
122769
64189
57802
12359
29912
6891
B076
mean
989594
22312
26155
116109
89887
75773
66085
16652
42582
0061
8605
RSO%
18
23
25
14
39
15
17
27
39
26
6
Analyte Concentration
(Extracted 44 Days After Hold Time)
SWH-CO-05-5
1029609
21497
—
82965
100865
80313
60803
20009
138559
""11885
"0039
PPB
SWH-CO-05-6
1035484
"31505
—
80862
85694
121340
82808
"23037
41433
"15414
"14700
SWH-CO-05-7
785951
30259
30531
80603
99418
02622
47628
18758
20890
•0678
•9652
mean
950348
27754
30531
81477
95326
98095
63746
20631
69961
12325
11130
RSD
15
20
N/A
2
9
22
28
11
85
24
28
W
• Value above calibration curve (no greater than 2X the highest calibration point)
** Value below calibration curve but above the method detection limit
— Ncn detected values because dilutions were extremely large
-------
TABLE 6-26. HOLDiNG TIME STUDY FOR FINE PARTICLE CAKE
Analyte Concentration
(Extracted Within Hold Time)
Pentachlorophenol
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)Anthracene
Chrysene
Benzo{b)Fluoranthene
Benzo(k)Fluoranthene
SWH-FPC-05-5
2458704
55619
71478
'295570
120366
•234977
* 21 01 50
63535
90219
39892
34117
PPB
SWH-FPC-05-7
790000
27605
34519
140440
54164
113225
102085
24584
40321
18122
"13956
mean
1624352
41612
52999
218005
87265
174101
156118
44060
65270
29007
24037
RPD%
103
67
69
71
76
70
69
88
76
75
84
Analyte Concentration
(Extracted 44 Days After Hold Time)
SWH-FPC-05-5
432437
49170
-
86113
108552
72258
55284
54715
87587
41654
19668
PPB
SWH-FPC-05-7
635470
50248
60185
100614
58695
123134
88646
55094
79955
30991
23101
mean
533954
49709
60185
93364
83624
97696
71965
54905
83771
36323
21384
RPD%
38
2
N/A
15
60
52
46
1
9
29
16
S3
Ui
Ul
* Value above calibration curve (no greater than 2X the highest calibration point)
** Value below calibration curve but above the method detection limit
— Won detected values because dilutions were extremely large
-------
TABLE 6-27. HOLDING TIME STUDY FOR FINE OVERSIZE
Analyte Concentration
(Extracted Within Hold Time)
Pentachlorophenol
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)Anthracene
Chrysene
Benzo(b)Fluoranthene
Benzo(k)Fluoranthene
SWH-FO-05-5
14250000
43828
50797
"240796
136066
190792
150325
38451
73286
25202
24388
PPB
SWH-FO-05-6
1571423
42552
52801
•249916
153210
•185941
155780
35023
66490
22824
23526
SWH-FO-05-7
2206666
42382
53490
"237121
95168
150367
127808
45581
84420
21152
21477
mean
2E+06
42911
52363
242611
128148
. 175700
144637
39685
74732
23059
23130
RSD%
24
2
3
3
23
13
10
14
12
9
6
Analyte Concentration
(Extracted 44 Days After Hold Time)
SWH-FO-05-5
946390
"30314
—
94390
92758
116220
83759
**22035
"38340
••12619
* "12407
PPB
SWH-FO-05-6
1281840
34769
36987
108963
77639
139737
81342
26248
44833
•"15268
••15767
SWH-FO-05-7
1944560
"38657
—
126622
114691
147269
83496
"28872
53594
"15122
"15134
mean
1E+06
34580
36987
109992
95029
134409
82866
25718
45589
14336
14436
RSD
37
12
N/A
15
20
12
2
13
17
10
12
N5
* Value above calibration curve (no greater than 2X the highest calibration point)
" Value below calibration curve but above the method detection limit
— Non detected values because dilutions were extremely large
-------
shown, relative percent differences which account for both sampling and
analytical variability, are well within expected deviations for almost all
compounds. As noted, no significant difference is detected between analyte
concentrations in samples which were extracted within holding time compared to
concentrations obtained for samples extracted 50 days past their holding time.
Because standard deviations were not computed for duplicate samples, a
formal statistical test such as "Student's t" was not performed. It was
necessary to compare each compound separately in order to detect patterns of
degradation if they existed. For the feed soil matrix, this seemed to be the
best approach, and all that was needed was a comparison of the means taking into
account the relative percent difference for each sample set. Because feed soil
is untreated, it is expected that holding time would be of less significance.
Further statistical evaluations may be performed; however, they probably remain
unnecessary and also may require assumptions about the data which may not be
valid.
While a "t test" was performed for other matrices when possible, other
observations about the data were required. For example, since no gross
contamination was detected in the blank samples and micro-organisms are not
expected to produce penta or PAHs, apparent increases in analyte concentrations
are not presumed to be statistically significant. In addition, general
observations showed that if one compound degraded in the time between
extractions, almost all other compounds also degraded. This helped to determine
if apparent losses of specific analytes were real. Additional observations and
the implications of these observations are presented below. In cases where the
"t test" was used to compare means, an alpha value of 0.05 was assumed
significant.
Because feed soil did not show significant losses, data from the feed soil
should be valid even if extracted past the original holding time. As discussed
previously, all extractions were performed within the time period of the holding
time study, and therefore these tables present a worse case situation.
The other critical matrix for determining treatment efficiency is the washed
soil, in that this is what BioTrol claims to be "cleaned soil". Efficiency of
the process is determined by comparing results of the washed soil to the feed
soil. Table 6-24 shows results of the washed-sand holding-time study. All
compounds detected showed no significant decrease in concentration between
extraction periods. Relative percent differences were not as good as the soil
matrix, and therefore a much wider variation is observed. In fact, some PAH
compounds showed significantly higher concentrations when re-extracted. The
reasons for this are unknown and are not easily explained. Comparisons of the
mean concentrations for penta, however, showed no significant change. Since
washed soil did not show significant losses, data should be valid even if samples
were extracted past the original holding time. Statistical significance testing
for the data generated during this phase of the holding time study was not
performed because only duplicate samples could be compared. It is also difficult
237
-------
to label any selected data as outliers. Therefore, concentrations of all
compounds are presented.
Other matrices in the treated soil are important for mass balance
determinations but do not affect treatment efficiency. Mass balance is difficult
under the best conditions, and several factors can affect final calculations.
For the coarse oversize soil fraction (see Table 6-25) only one compound
showed significant degradation as determined by Student's "t test" (alpha -
0.05). Phenanthrene appeared to have degraded by a minor amount, and in fact,
showed only a slight difference in concentration when values were compared for
pre- and post- extractions. Because this was the only compound concentration out
of eleven compounds that significantly decreased, and because the micro-organisms
or physical/chemical processes in question are not known to have any -particular
affinity for degrading phenanthrene, it appears more likely that no significant
compound degradation occurred for this sample.
The fine particle cake fraction (see Table 6-26) showed a much larger
variation between replicate samples, and in this regard, it was similar to what
was observed for washed soil. Three compounds appeared to have experienced some
degradation including penta, phenanthrene, and possibly pyrene. These
degradations, however, may not be statistically significant, as only duplicate
samples were compared and relative standard deviations were not computed.
Nevertheless, because there are three compounds which appear to have degraded,
and holding time is suspect, it is more difficult to rule out the possibility of
biased data. Because the default position is that holding time affects compound
concentration, (and these data do not prove otherwise), and because penta
concentrations apparently decreased, it should be assumed that holding time
affected the reported compound concentrations for a significant portion of these
samples.
The final washed soil matrix which was examined was fine oversize (see Table
6-27). For this data set a statistically significant difference (alpha = 0.05)
between the means is observed for all but one compound, anthracene. Not only do
the mean concentrations differ significantly, but in all cases where there was
a significant difference, individual analyte concentrations within a single
sample went down over time. The evidence appears almost conclusive that
degradation occurred with respect to time, and ranged from 30% to 50% depending
upon the compound.
Given this additional information concerning the fine oversize and fine
particle cake samples, determining the mass balance of target analytes within
each of the soil matrices could be biased by the effect of missed holding times.
Certainly mass balance is a difficult accomplishment in a project such as this,
and missed holding times.appear to cause additional confusion and possible bias
in the final calculation.
For the above soil matrices, samples were extracted past their holding time,
and therefore appropriate steps were taken to evaluate the effect. In several
cases the evidence is not conclusive, but it certainly appears that holding time
does not present a problem in determining treatment efficiency, although it will
cause some difficulties in calculating mass balance. This mass balance
238
-------
calculation is already known to be somewhat approximate. Unfortunately, studies
such as these are not likely to produce conclusive results because they are often
initiated after sample collection, and they are generally never included in the
original project design. Therefore, decisions regarding utilization of available
data are required based upon partial evidence. Recommendations concerning
holding time effects have been clearly stated in the previous paragraphs, and are
based solely upon the evidence presented.
Tables 6-28 and 6-29 show results of the holding time study conducted for
the slurry samples. Extraction of these samples proved to be much more difficult
than expected; penta concentrations were unknown and these types of samples had
not been previously extracted. Originally, the samples were to be centrifuged,
the solid and water portions were then to be extracted separately, and these
extracts were to be combined for analysis. Centrifugation proved to be useless
as solid and water portions could not be separated. Filtration was therefore
conducted, but in performing this extra step, all samples were extracted past
their initial holding time.. Consequently, samples were extracted anywhere
between several days to two weeks past holding times specified in the QAPjP.
Once samples were extracted, analysis was delayed until further examination
of these extracts could be performed by the project QA Manager. This examination
needed to be performed because the laboratory believed that concentration
differences between the solid and water extracts differed greatly, and therefore
it would probably be unwise to combine these extracts for analysis. In addition,
these extracts looked extremely dirty, so the laboratory recommended a GPC clean-
up prior to analysis. The QA Manager agreed with both of these recommendations.
This decision was made in mid-December, 1989, and at the same time, laboratory
personnel re-extracted the specified slurry samples for the holding time study.
(It should be noted that the holding time study for the slurry samples compares
samples extracted 1-2 weeks after their specified holding time to samples
extracted approximately 6 weeks after their specified holding time.) During the
analysis of these samples, the laboratory supervisor indicated that because the
samples were so highly concentrated with contaminants, analyses were much more
difficult than expected. Consequently, because of the analytical problems,
results of the study were not available until mid-January, 1990, and therefore
holding time evaluations were delayed.
Penta concentrations in the slurry samples were extremely high, and in order
to get analyte concentrations within the linear range of the instrument, sample
dilutions of 300X to 600X were required. As a result, surrogate information for
each of these samples in the initial extraction set were lost. Samples could not
be run at dilutions to quantitate the surrogate spikes since penta concentrations
were extremely high. In attempting to perform surrogate quantitation., the
laboratory supervisor soon realized that instrumentation would be unable to
handle the overload, and resultant instrument down time became a critical issue.
As a result, samples were diluted to attempt to quantify only the penta
concentrations. Concentrations of other analytes and surrogates were not
quantified.
239
-------
TABLE 6-28. HOLDING TIME STUDY FOR SLURRY SOILIDS PHASE
INFLUENT
Pentachlorophenol
Fluoranthene
Pyrene
EFFLUENT
Pentachlorophenol
Fluoranthene
Pyrene
Original Extract
(13 Days Past the Holding Time)
PPB
SBR-IN-Q8-1 SBR-IN-08-5 mean RPD%
2600000 3200000 2900000 21
150000 280000 * 215000 60
150000 230000 * 190000 42
Original Extract
(13 Days Past the Holding Time)
PPB
SBR-EFF-08-1 SBR-EFF-08-5 mean RPD%
2900000 2600000 2750000 11
89000 89000 N/A
10000 10000 N/A
Re-extract
(77 Days Past the Holding Time)
PPB
SBR-IN-08-1 SBR-IN-08-5 mean
3400000 2900000 3150000
209000 * 350000 •* 279500
154000 * 380000 * 267000
Re-extract
(77 Days Past the Holding Time)
PPB
SBR-EFF-08-1 SBR-EFF-08-5 mean
2300000 2400000 2350000
RPD%
16
50
85
RPD%
4
M
-f>
O
* Result is uncertain due to high dilutions
~ Non detected, but 300X to 600X dilutions were required to maintain PCP in calibration range
file: 6-2S&29
-------
TABLE 6-29. HOLDING TIME STUDY FOR SLURRY AQUEOUS PHASE
INFLUENT
Pentachlorophenol
EFFLUENT
Pentachlorophenol
=====================================
Original Extract
(1-2 Weeks Past the Holding Time)
PPB
SBR-IN-08-1 SBR-IN-08-5 mean RPD%
32000 33000 32500 3
Original Extract
(1 -2 Weeks Past the Holding Time)
PPB
SBR-EFF-08-1 SBR-EFF-08-5 mean RPD%
110000 97000 103500 13
========================^
Re-extract
(6 Weeks Past the Holding Time)
PPB
SBR-IN-08-1 SBR-IN-08-5 mean
87000 43000 ERR
Re-extract
(6 Weeks Past the Holding Time)
PPB
SBR-EFF-08-1 SBR-EFF-08-5 mean
67000 75000' 71000
—
Rpnoh
65
RPDMi
11
-------
It was decided it would be best to analyze all the initially extracted
samples for penta, and then perform a re-extraction for the entire sample set.
The purpose of the re-extraction was to extract smaller amounts of ssimple,
thereby yielding extracts where surrogate spikes could be quantified at
appropriate concentrations. This procedure was performed on a few selected
samples which were then analyzed, and results were examined to ensure that the
proper sample size was selected. The remaining sample set was then re-extracted
and analyzed using these more appropriate sample sizes; however, the major
drawback with this approach was that samples were extracted several months after
their holding time. Nevertheless, these were the data which were eventually used
for determining project obj ectives.
Presented in Tables 6-28 and 6-29 are slurry results from the first
extraction sets and holding time studies which have no surrogates associated with
the reported concentrations. Only duplicate samples were extracted, and the
relative percent difference, between the two samples is very large. Given these
problems, concentrations can only be considered as approximations. Nevertheless,
these data suggest no effect associated with holding time.
Penta concentrations in the aqueous slurry may also represent a maximum
solubility threshold. Penta solubility in water is approximately 80,000 PPB,
(reference: Merck Index, pH dependent). If degradation is occurring in the
slurry sample, and penta concentrations are well above the maximum solubility of
penta in water, the water portion of the slurry may always contain the same
amount of penta, regardless of when it is extracted. Slurries are separated into
water and solid samples by filtration just prior to extraction. Levels of penta
are always well above the 80,000 PPB range in the solid samples, (see Table 6-
28) . Penta concentrations in the aqueous portion of the slurry range from 32,000
PPB to over 100,000 PPB. Penta extraction data from MS/MSD samples of aqueous
solutions for the BioTrol liquids project range from 40-100%. Given this
variability, along with instrument analytical variability, it is likely that the
penta concentrations in Table 6-29 represent a maximum saturation in the aqueous
portion of the slurry sample.
Because penta concentration levels differ by no significant amounts between
the two slurry solid extractions, it is likely that holding time had no
significant effect on concentrations for all the slurry samples. This is
difficult to quantify because only duplicate samples were available for
comparison and relative percent differences were fairly large. While it cannot
be conclusively stated that holding time exhibits no sample bias, this effect is
not noticeable within normal sampling and analytical variability. In addition,
it is not believed that concentrations continued to decrease over time.
Therefore, the slurry samples that were re-extracted at levels where surrogate
information was available are not expected to experience holding time bias even
though extractions are several months past their holding time. As noted, these
were the samples that were used in the main body of the report.
In response to holding time concerns noted above, an additional data quality
audit was performed several months after this study was conducted to determine
if any additional data bias could be evaluated. During a data review conducted
prior to this final audit, several concerns were noted and resolutions were
242
-------
discussed between project and QA Managers which resulted in a better
understanding of the conclusions to be designated in the final report.
Of primary concern was the issue of holding times which had been exceeded
for several sample matrices. The holding time study described above had shown
that holding time effects were not significant for the more critical soil
matrices nor were they significant for the slurry samples. No holding time
studies were conducted for aqueous samples because records obtained from the
laboratory had indicated that aqueous samples had not exceeded their designated
holding times. Review of the more recent project data received in the
laboratory's final report showed that conclusions concerning soil and slurry
matrices were still correct and holding time problems did not appear to be
significant. In addition, laboratory records now showed that about 1/3 of the
aqueous samples had exceeded their holding time.
The number of aqueous samples which were extracted past their holding time
is extremely significant, therefore it was decided that holding time effect
should be considered in order to save otherwise questionable data. A review of
results for the same aqueous samples which had been extracted at two separate
time periods was important to determine if differences in these data could cause
bias in the reported conclusions. While no special study had been conducted, by
coincidence because some surrogates did not meet specified recovery criteria,
several aqueous samples were re-extracted past their holding time and results
from these two extractions were compared. The Deputy Project Manager's
statistical study of this somewhat limited group of samples indicated no
significant effect due to holding time. While it could not be proven that
holding time caused no bias, it was determined that if a holding time bias
existed, this bias was within analytical and sampling variability. In addition
to this statistical analysis, a review by the QA Manager of replicate samples
analyzed within and outside their holding times indicated that effluent samples
showed no effect due to holding time, but there did appear to be a bias caused
by holding time for influent aqueous samples. (This difference could be due to
the different nature of these two matrices.)
6.6 AUDITS and RESULTING QAPjP DEVIATIONS
6.6.1 BioTrol Soils Audit
An audit of the BioTrol Soils Project was conducted on September 26t'h and
27th, 1989. The EPA Auditor and Contracting QA Manager were present during the
pre-test soil sampling and calibration operations.
There were two concerns and recommended corrective actions noted by the EPA
Auditor. The first concern involved a change from the hourly collection of
sample drum weights while sampling was in progress. Instead, full sample drums
were weighed after collection. This change did not affect the collected data;
however, it was also decided that 1 in 5 drums were to be weighed in duplicate
in order to obtain a precision estimate of drum weights.
The second concern was the loss of measurement data on the oxidation-
reduction potential of the combined dewatering effluent. A suitable redox meter
243
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
and. probe did not arrive until the final day of sampling. Because this is a non-
critical measurement, the loss of these data was not a major concern.
Other minor concerns noted were that split samples were performed in the
laboratory and not in the field, contrary to specifications in the QAPjP, and
less accurate alcohol thermometers had to be replaced by Hg thermometers. The
EPA Auditor also recommended use of a sample log book and sample data sheets to
insure a more permanent and traceable record.
Other changes in the QAPjP were required once field operations began. These
changes included the following:
• Only a water and distilled water rinse were performed for
decontamination between samples since only one auger was available.
No methanol rinse or drying was performed between sampling since
cross-contamination of composited soil samples was not considered a
critical problem.
• Although rotameter calibration was carefully conducted with readings
taken every 2 hours, it was of minimal benefit due to variable water
flow through the meter. All water through the soil washing system,
except municipal water, was measured more accurately by tank volume
measurements. Municipal water accounts for 10% of total water when
system is running at steady state conditions and was not expected to
contain a significant amount of pollutants.
• The amount of nutrient added to the Aqueous Treatment System (ATS) was
measured as total volume and the caustic used for pH adjustment was
measured as total weight.
• The soil feed hopper was calibrated by hanging 3 separate weights from
the belt scale. The manufacturer provided a conversion factor to
obtain the measurements in pounds per hour. The speed of the feed
belt was carefully timed to insure that static weighing would be close
• to the actual pounds per hour.
• The soil washing system was shut down several times due to higher
creosote concentration causing the soil to cake inside the feed
hopper. Samples were not taken during shutdown.
• Only 75% of the total data expected was collected because BioTrol ran
out of soil and the ATS was shut down early due to pumping problems.
Project and field managers however, believed that sufficient data were
available to fulfill overall project objectives. This problem is
discussed in detail in section 6.4.
The EPA Auditor noted that several changes made during the course of the project
were not approved by the RREL QA officer; however, these changes were necessary
and approved by the contractor QA Manager and are not expected to affect data
quality. Although some decisions need to be made before the RREL QA officer can
be consulted, these decisions are always discussed with the Project and QA
Manager.
244
-------
6.6.2 Laboratory Audit at 1st Subcontractor Laboratory and Field Audit of Slurrv
Bioreactor
The audit for the subcontracting laboratory was conducted on October 19th
and 20th, 1989 by EPA and contractor QA personnel. Analyses being performed
included total organic carbon (TOG), total organic halides (TOX) , cation exchange
capacity (CEC), chemical oxygen demand (COD), total recoverable petroleum
hydrocarbons (TRPH), and analysis of chlorinated dibenzo dioxins/furans (PCDD).
The only critical measurement was the analysis of PCDD. There were no analytical
concerns for PCDD; however there were sampling concerns for PCDD and other
concerns for the non-critical measurements.
The laboratory was instructed to perform PCDD analysis on 10 soil field
splits. Dioxin analyses were not performed on the 10 split samples so there
could be no field duplicates to assess precision. To resolve the oversight, 4
additional split samples, previously archived samples, were analyzed for dioxin.
The samples chosen included one of each soil fraction: washed soil, fine
particle cake, fine oversize, and coarse oversize.
A review of other PCDD analysis showed no .field splits were scheduled for
either the aqueous or slurry samples. Since sampling for the slurry bioreactor
was ongoing, collection of the sampling plans and the field crew was notified.
An additional aqueous sample was retrieved from the collected archives.
Corrective action was initiated in response to these concerns.
The audit for on-site sampling of the slurry bioreactor was conducted by the
EPA Auditor on October 23rd. Field personnel were conducting all operations as
specified, and the audit rating was satisfactory.
6.6.3 Technical Systems Review (TSR) at Second Subcontractor Laboratory
On October 9th and 10th, 1989, a TSR was performed by EPA QA contractor.
Concerns are noted below.
SAMPLE LOG-IN AND CUSTODY
Samples were properly logged into a bound notebook and entered into an
electronic system. All shipment information was retained and the in-laboratory
chain of custody forms were properly maintained. It was noted that one of the
refrigerators used to store samples for this project was maintained ,at a
temperature of 8°C, outside the specified range of less than 4°C.
LEACHING PROCEDURES
TCLP and EP forms were well organized and relatively complete except for
information relating to the leaching fluid determination step for TCLP,
documentation of extraction fluid preparation, and information regarding the
initial liquid/solid separation step (e.g. volume of leachate). There were no
means for monitoring or documenting the temperature of the room where leaching
is performed. TCLP requires that the 18-hour leaching be performed around 22°C.
245
-------
It was recommended that a recording thermometer be placed in the room where
leaching occurs, and the chart documentation be retained.
EXTRACTION LABORATORY
The BioTrol SITE sample flow caused the laboratory to operate near or at
their maximum capacity. In an effort to meet sample extraction holding times,
two procedures were implemented that did not comply with Method 3550. Some soil
samples were weighed and mixed with Na2S04 in glass beakers with aluminum foil
covers up to 24 hours prior to extraction. These were stored in a refrigerator
at 4°C. This could lead to cross-contamination from external sources, and loss
of unknown amounts of the more volatile PAH target analytes due to the inadequate
aluminum foil seal. The second procedure involved concentrating 53 soil samples
using a rotary evaporator. Method 3550 indicates only the use of the K-D
technique for concentration of soil extracts. The main concern with using rotary
evaporator concentration is that the recovery behavior of the targets may riot be
the same when compared to the K-D concentration technique. This would adversely
affect data quality with respect to precision. In addition, method blank, data
for those extracts concentrated by rotary evaporation should be carefully
examined because high-level sample extracts can cause analyte carryover into
succeeding concentrations. Subsequent data analyses showed,, however, that
adequate recoveries of some analytes, including penta, were obtained using the
rotary evaporator for extract concentration. It was recommended that the samples
which did not comply with Method 3550 be re-extracted and use of the rotary
evaporator discontinued.
Other minor concerns included: the storage of standards with some BioTrol
samples in one refrigerator located in the extraction laboratory, two loose pages
from the calibration notebook were found that did not contain proper analytical
balance calibration information, references to both Method 8270 and Method 625
were found in the extraction laboratory paper work although these Methods are not
entirely comparable, and the extraction logs were missing the date the samples
were received into the extraction laboratory, the extract final volume, and (at
times) the technician's initials. Finally, there was concern for the storage of
extracts in round-bottom flasks for up to one . week before drying and
concentrating the samples. Without drying the extract with Na2SOA prior to
storage, there could be a loss of some components due to hydrolysis.
DETERMINATION OF PAHs AND PENTACHLOROPHENOL USING GC/MS
The GC/MS instrumentation appeared to be properly set up and maintained.
A GC/MS maintenance log properly documented problems and routine maintenance of
the system. Every entry was clearly written, dated, and initialed. The analyst
stated that maintenance such as changing septa, liners, and breaking off of a
few inches of column to optimize chromatography was being done on a routine
basis, as required. There were two deviations from recommended GC/MC operating
parameters listed in Method 8270. The mass spectrometer was being scanned from
45-500 amu, the recommended scan range is 35-500 amu. This difference in mass
range scanning should not adversely impact data for this project because all PAH,
surrogate, and MS analytes have quantitation ions well above m/e 45. A faster
ramp rate (14°C/min) was also incorporated in the temperature program than is
recommended in Method 8270 (10°C/min). A review of a GC/MS 50 ppm standard
246
-------
analysis indicated that adequate chromatographic resolution was being achieved
on the DB-5 0.32-mm bore, 30-meter, thick-film,fused silica capillary column
which was used.
Surrogate recoveries met the objectives given in the QA Project Plan in
every sample and blank reviewed. A total of seven surrogate compound recoveries
were studied, including D10-fluoranthene, which was added to more closely mimic
the behavior of the PAHs of interest in this project. A review of MS/MSD
recovery data yielded two compounds that were out of the QA Project Plan
specified recovery control limits.
Pentachlorophenol and 4-nitrophenol recoveries were high in a MS sample
analysis and 4-nitrophenol recovery was non-compliant in the MSD sample. The
corresponding sample which was not spiked had not yet been analyzed by GC/MS at
the time of the audit. It is possible that these analytes were present in the
sample which could explain the outlier recovery data. In some cases, samples
with MS/MSD pairs were not analyzed during the same shifts. It was recommended
that a sample and its corresponding MS/MSD pair should be analyzed on the same
shift so as to minimize variability of the data.
A QC check sample was not prepared or available for GC/MS analysis for this
project. Method 8270 states that QC standards prepared from an independent
source should be carried through the analytical system so as to provide
information on the laboratory's ability to meet specified method accuracy
criteria and that this must be done before sample analysis begins. QC check
samples prepared with standards obtained from an independent source provide
verification of calibration standard accuracy. Without a check standard prepared
from a second source, it cannot be known whether the calibration standards are
accurate. Hence, it might be stated that all ensuing analyses are of unknown
accuracy. It was recommended that a check standard be prepared and analyzed in
accordance with Method 8270.
RESIDUE DETERMINATIONS
Residue determinations are critical measurement parameters for this project.
The QA Project Plan specified the use of EPA Method 160.3 for these
determinations.
A serious problem not relating to laboratory performance was noted during
this TSR. Some of the residue determinations were performed on samples which
were preserved with acid. The laboratory was forced to take this approach since
no separate aliquot without preservative was received, except for the PAH/penta
aliquots, and many of these samples were completely consumed during
extraction/analysis. Acidification of samples will likely produce biased results
for the residue determinations through a variety of mechanisms. This was noted
when comparing residue results for samples which were acidified with samples
which were not acidified. Results between the two groups varied by a factor of
four with the acidified samples having the higher results. Laboratory personnel
noted that they would attempt to locate enough sample for the residue
determination from either the extractable fraction or from the archive sample,
and not perform any further residue analyses on acidified samples.
247
-------
The balance used for weighing the residue samples was not checked for
calibration on each day of use. Checking calibration of the balance is important
since the balance is the primary standard for gravimetric procedures. It was
recommended that the laboratory check and document calibration on each day the
balance is used for residue determination.
No temperature 'record was kept for the drying oven used for the residue
determinations. Method 160.3 requires that the samples are to be evaporated in
the oven at 103-105°C. A check of the oven temperature during the TSR indicated
that the temperature was 100°C. Since drying temperature can have significant
effects on the accuracy of the method, it was recommended that the oven
temperature be adjusted within the correct range and documented at the beginning
and end of each batch of residue determinations.
Section 7.3 of Method 160.3 requires that a sample is to be evaporated,
dried for one hour in the oven, then dried and desiccated to a constant weight.
The procedure used was to dry the samples for 24 hours, desiccate for one hour,
and weigh the samples once for the residue determination. This method is a
significant deviation from the specified method. The impact to data quality with
respect to accuracy due to this deviation is unknown since no QC samples for the
evaluation of accuracy were analyzed (the method and QA Project Plan do not
require the analyses of these samples). It was recommended that the laboratory
adhere to method requirements with respect to drying procedures. Unfortunately,
reanalysis of improperly processed samples may not be possible in all cases due
to expiration of holding times. ;
One minor concern is an inadequate collection of duplicate samples on
October 5th. A total of 41 samples were analyzed with only one duplicate
analysis when the QA Project Plan specified one duplicate for every twenty
samples.
TRACE METALS DETERMINATIONS
Copper, chromium, and arsenic determinations were considered critical
measurements whereas Ba, Cd, Pb, Hg, Se, Ag were considered non-critical.
Analytical runs were examined for all of the critical elements.
Two minor concerns which should not affect the quality of the determinations
are that matrix spiking solution preparation information was not recorded, and
the digestion logbook did not indicate the method used for sample
digestion/preparation.
CONCLUSIONS
Since major concerns were noted with some of the critical measurements, this
TSR indicated a rating of Conditional for the analytical efforts conducted at
this subcontracting laboratory for the BioTrol Soils SITE project. Corrective
actions and the initial TSR findings were discussed with the laboratory during
a closing briefing. A Corrective Action Recommendation form was completed and
is summarized below.
248
-------
Nature of Concern:
1. Balance not checked for calibration (residue determination).
2. Drying to constant weight not performed (residue).
3. Improper storage of soil samples prior to preparation (foil covered
beakers).
4. Samples not dried promptly after extraction.
5. Use of Roto-Vap for sample concentration.
6. No QC check standards prepared/analyzed for GC/MS.
Recommended Action:
1. Check and log balance calibration daily.
2. Dry samples to constant weight per method.
3. Discontinue this storage practice.
4. Dry samples promptly after extraction.
5. Use Kuderna-Danish apparatus in place of Roto-Vap as required.
Document which samples by Roto-Vap.
6. Prepare/analyze QC check standards per method.
RESPONSE TO EPA PROJECT AUDIT DEBRIEFING
The laboratory implemented several changes in response to the concerns
outlined during the audit debriefing. All changes were implemented with the
concurrence of the contractor QA manager.
Responses to Identified Concerns:
1. Logbooks were started for the top loading balance to document that the
calibration checks have taken place before each use.
2. Drying times were shortened and a constant weight was achieved.
3. All samples in the foil covered beakers (this included only those
samples seen during the TSR and not prior samples) were prepared from
the original sample bottles. The foil covered samples were not used.
4. All samples clearly showing any water were dried within two days of
the TSR. Those not showing visible water were dried within a week.
The heavy demands on the extraction laboratory staff precluded
immediate drying of all samples. In the future, if samples need to
sit in solvent, they will be dried before storage.
5. A total of 53 samples were concentrated by the Roto-vap procedure. It
was not feasible to re-extract those samples within hold times so
analyses were allowed to continue. As an independent check,
laboratory personnel spiked solvent blanks with a subset of the
analytes of interest and concentrated them with the roto-vap. This
provided information on how the roto-vap affects recoveries. Since
method blanks were already included with the samples, any
contamination introduced by the technique could be determined. Prior
experience indicates that recoveries should be good and contamination
249
-------
minimal, but this approach allowed documentation for the compounds of
interest.
6. Subsequent to the audit, QC check samples were prepared and analyzed.
6.6-4 Follow-up Audit of 2nd Subcontracting Laboratory
A follow-up audit was conducted, (November 2, 1989), by contracting QA
management to document that the appropriate corrective action had been taken in
response to concerns raised during the EPA audit, October 9, 1989.
The follow-up audit, in short, concluded that the laboratory had taken
appropriate corrective action for each concern noted. Specifically, laboratory
personnel were no longer using the roto-evaporator for any sample extracts, and
they were in the process of assessing the effect roto-evaporation (compared to
K-D concentration) would have on the specified samples. A study involving
spiking blank extracts with project-specific target analytes was conducted.
Given the time involved to extract 50 samples, it did not seem feasible to
perform re-extractions, and in fact re-extracting, these samples would cause
additional problems in meeting sample holding times. Problems with holding times
are noted below. This approach was discussed during the audit as a means of
corrective action.
A new concern was uncovered during the follow-up audit. It was learned, that
approximately 60 samples were not extracted within the specified holding time.
The reasons this occurred are listed below:
• Several samples arrived at the laboratory simultaneously.
• Two personnel in the extraction laboratory had quit.
• The October 1989 San Francisco earthquake caused the laboratory
shutdown of two to three days.
• Other samples were extracted ahead of BioTrol soil samples apparently
because a higher priority had been placed upon these other samples.
• Newly implemented sample tracking procedures were not fully functional
and caused some delay in sample extraction.
As a result of samples not being extracted within specified holding times, the
laboratory initiated an additional study to assess this effect on the samples.
A list of samples that missed holding times was created for the contracting QA
Manager's preliminary review. The contracting QA Manager instructed laboratory
personnel to re-extract several archived samples (previously extracted) with
holding times nearly equal to those of samples which were not extracted within
holding-time specifications. The results of these studies show the effect missed
holding times had on these samples (section 6.5).
250
-------
6.6.5 Audit of Soil Sieving Procedure for BioTrol Prolect
An audit was conducted by the EPA QA Contractor on October 12th on the soil
sieving procedure performed by another subcontracting laboratory. A conditional
rating was received for the procedure which was a non-critical measurement as
specified by the QA Project Plan.
The procedure chosen for soil sieving was a combination of ASTM methods 421
and 422. This procedure was developed by the project manager and was not
previously reviewed by EPA before being used.
The samples chosen were separate composites of the low and high
concentration soils before treatment. One sample of each soil was sent to the
laboratory. The entire sample was sieved per specifications. Except for the
first sieving separation, the procedure uses wet sieving--soaking the soil in
sodium hexametaphosphate (calgon solution) which has a pH between 8 and 9. Since
the pKa of penta is acidic, and given the amount of time the soil is soaked in
the calgon solution (16 hours) it was anticipated that most of the penta would
end up in the wash solution. (This conclusion was shown to be false as indicated
by results of penta in each of the soil fractions; see report.) Since PAH
concentrations are relatively low, it was also believed that accurate
determinations of PAH percentage in the various soil fractions would be
difficult. It was anticipated that PAH concentrations would drop below
analytical detection limits, except for fractions where the PAH concentration was
likely to be highest.
Although the following concerns were considered minor, corrective action was
warranted since many of these are standard laboratory operations which are used
for all laboratory measurements.
• Daily balance calibration checks were not performed on the analytical
balance in the containment laboratory.
• The pH of the calgon solution should be checked and recorded prior to
usage.
• A record of reagent preparation should be recorded in the laboratory
notebook.
• A record of each soil fraction weight needs to be recorded even if it
is zero to avoid the appearance of a missing record.
• A daily log for the drying oven is needed.
• Percent Moisture determination should be calculated after a constant
dry soil weight has been obtained.
A final issue concerned the largest sized soil fraction which appeared to
be primarily "sticky clumps" of soil. The method specified use of a mortar and
pestle for separation of soil clumps observed in the original sample. This was
not possible since the soil had a high creosote concentration and had a gummy
texture. It was decided that separation of these soil clumps could not be
251
-------
achieved by mechanical means. Therefore, the soil clumps remained as a part of
the largest sized soil fraction. It is likely that a high concentration of penta
and PAHs are contained in this fraction due to its gummy texture and tar-like
appearance. (This conclusion proved to be correct as noted in the report.)
Although several concerns resulted from last minute procedural changes,
research should continue on SITE projects and additional measurements should not
be discouraged. Procedures should be approved along with the QAPjP and tested
prior to use, especially if they are procedures which have not been previously
verified. There are provisions for pre-demonstration testing of analytical
methods or potentially troublesome matrices, and such method development in this
case appears to have been warranted. This approval process is not meant to
prevent additional information from being gained, but should be used to insure
objectives of the measurement can be obtained.
RESPONSE TO TSR DRAFT REVIEW SUMMARY OF SOIL SIEVING AUDIT
Initially this procedure was only briefly detailed in the QAPjP. Since it
is a non-critical parameter, the QAPjP was approved without requiring a clearly
defined procedure. This procedure was later developed by the project manager and
sent to the laboratory only a few days prior to actually using the method on
field samples. A blank was run first, followed by the two field samples. Once
the first field sample was sieved, problems in the procedure were noted and some
changes were required. The laboratory was in constant communication with the
contracting QA and Project Manager regarding additional steps which needed to be
taken and alternative methods which could be used.
Since these samples needed to be analyzed as quickly as possible (holding
times had already been exceeded) and since no pre-validation studies had been
performed, the method used was a compromise method for obtaining project
objectives. It is expected only to give a gross indication of chemical
partitioning for the separate soil fractions, and it is understood that the
largest soil fraction also contains sticky clumps of soil which could not be
separated. These clumps are not necessarily representative of larger particle
sizes.
The method specified use of a mortar and pestle for separation of soil
clumps observed in the original sample. This was not possible since the soil had
a high creosote concentration and was a gummy texture. It was decided that
separation of these soil clumps could not be achieved by mechanical means. Other
types of separating processes, such as chemical separation, were discussed but
not pursued due to technical considerations.
The following corrective actions were initiated in response to minor
concerns previously noted:
• The balance used for percent moisture was checked to insure
calibration was within proj ect obj ectives.
• Preparation of the hexametaphosphate solution was properly recorded.
252
-------
Holding times were not within specifications for analysis since
sieving procedures needed to be performed. Samples were sieved and
analyzed as quickly as possible to insure a minimal effect in
exceeding specified holding -times.
253
-------
SECTION 7
BIBLIOGRAPHY
ASTM, 1988; Annual Book of ASTM Standards. Section 11 - Water and Environmental
Technology, Volume 11.01 - Water (I); Publication Code Number 01-110188-16;
Philadelphia, PA.
ASTM, 1989; Annual Book of ASTM Standards. Section 4 - Construction, Volume
4.08 - Soil and Rock; Building Stones; Geotextiles; Publication Code Number
01-040889-38; Philadelphia, PA.
BioTrol, Inc., 1988a; SITE Proposal: Soils Washing Technology Treatment
Technology. March 7, 1988. (NOTE: Confidential Business Information)
BioTrol, Inc., 1988b; Progress Report. Project No. 305. Soil Treatment System
Development. April 29, 1988. (NOTE: Confidential Business Information)
Cochran, William G., 1963; Sampling Techniques, p. 158.
Crawford, R. L. , and W. W. Mohn, Microbiological Removal of Pentachlorophenol
From Soil Using a Flavobacterium; Enzyme Microbiology and Technology 1_,
617-620; 1985.
Dixon, J.W., 1953; "Processing Data for Outliers," by J.W. Dixon, Biometrics,
Volume 9, pages 74 to 89, November 1, 1953.
Eichelberger, J.W., L.E. Harris, and W.L. Budde, 1975; Analytical Chemistry.
Volume 47, p. 995.
Perry, Robert H. and Don Green, 1984; Perry's Chemical Engineers' Handbook. 6th
Edition, McGraw-Hill Book Company, New York, NY.
Saber, Diane L. , and R. L. Crawford, Isolation and Characterization of
Flavobacterium Strains That Degrade Pentachlorophenol; Applied
Environmental Microbiology 50. 1512-1518; 1985.
Scholz, Robert, and Joseph Milanowski, 1983; Mobile System for Extracting Spilled
Hazardous Materials from Excavated Soils. United States Environmental
Protection Agency Project Summary, EPA-600/S2-83-100, December, 1983.
Twin City Testing Corporation, 1986; Remedial Investigation Report. MacGillis
and Gibbs Company Site New Brighton. Minnesota. June 25, 1986.
254
-------
Twin City Testing Corporation, 1987; Remedial Investigation Report. MacGillis and
Gibbs Company. New Brighton. Minnesota. February 17, 1987.
USEPA, 1979; Methods for Chemical Analysis of Water and Wastes. EPA-600/4-79/020,
EMSL, Cincinnati, OH, March, 1979.
USEPA, 1984; Test Method: The Determination of Inorganic Ions in Water by Ion
Chromatography - Method 300.0; Environmental Monitoring and Support
Laboratory, Cincinnati, OH 45268; EPA-600/4-84-017; March, 1984
USEPA, 1986; Test Methods for Evaluating Solid Waste. SW-846, Third Edition,
Office of Solid Waste, U.S. Environmental Protection Agency, Document
Control No. 995-001-00000-1.
Valine, Steven B. Memorandum to Mary Stinson (EPA) and William Ellis (SAIC)
regarding analysis of MacGillis and Gibbs soil piles, August 31, 1989.
Weast, Robert C. , and Melvin J. Astle, 1982; CRG Handbook of Chemistry and
Physics. 63rd Edition, CRC Press, Boca Raton, FL.
255
------- |