EPA/540/R-92/015
September 1992
TECHNOLOGY EVALUATION REPORT
SITE PROGRAM DEMONSTRATION OF A
TRIAL EXCAVATION AT THE
MCCOLL SUPERFUND SITE
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 Contract No. 68-02-4284 and the Superfund Innovative Tech-
nology Evaluation (SITE) Program. It has been subjected to the Agency's peer review
and administrative review and it has been approved for publication as a U.S. EPA
document. Mention of trade names or commercial products does not constitute an
endorsement or recommendation for use.
ii
-------
FOREWORD
The Superfund Innovative Technology Evaluation (SITE) program was autho-
rized 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 describes the trial excavation performed at the McColl Hazardous
Waste Site. Excavation at this site presented unique problems due to the high poten-
tial for release of sulfur dioxide and volatile odorous compounds contained in the
waste. The excavation demonstration was used to obtain information on the utilization
of an enclosure and associated air treatment systems around the excavation to mini-
mize air emissions, and the use of foam vapor suppressants to reduce emissions from
the waste during excavation. In addition, information was obtained on processing the
tar fraction of this waste by mixing with cement and fly ash.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
iii
-------
ABSTRACT
A trial excavation of approximately 137 cubic yards of waste was performed at
the McColl Superfund Site in Fullerton, CA to better determine the nature of this waste,
any treatment needed to improve its handling characteristics, and the extent of air
emissions that might occur during excavation. This type of information is necessary to
plan full-scale remediation of this highly acidic petroleum refinery waste buried at this
site. The trial excavation was conducted within a temporary enclosure with air ex-
hausted from the enclosure through a sodium hydroxide-based wet scrubber and
activated-carbon bed adsorber to reduce air emissions of sulfur dioxide and organic
compounds. Foam was used in an attempt to suppress atmospheric releases from
the raw waste during excavation, storage, and processing. The air exhaust was moni-
tored for total hydrocarbons and sulfur dioxide before and after the air emission
control system. In addition, total hydrocarbons and sulfur dioxide were monitored
along the site perimeter to determine potential impact of air emissions on the nearby
community.
This waste consisted of layers of overburden, mud, tar, and char. Excavation
was conducted with a track hoe and the waste was separated into stockpiles of mud,
tar, and char for subsequent study and experimentation. Upon completion of the
work, the waste was replaced into the excavation pit and covered with topsoil to mini-
mize atmospheric releases.
This report describes the equipment used to measure and control emissions,
the excavation progress, and resulting emissions before and after the air control
system. An assessment of the foam vapor suppressants is also provided as well as
information on the costs incurred for the field portion of this study. The information
contained in this report will assist in planning the full-scale remediation of the McColl
site and other similar waste sites throughout the country.
iv
-------
CONTENTS
Page
Foreword iii
Abstract iv
Figures vii
Tables x
Abbreviations and Symbols xiii
Acknowledgments xvi
1. Executive Summary 1
Objectives 2
Excavation and waste processing 4
Air emission controls 5
Waste characterization 6
Community impact 7
Costs of excavation and tar processing 7
Conclusions 7
Observations 11
2. Description of Technologies 13
Enclosure and exhaust air control system 13
Foam vapor suppressants 21
Waste Treatment Techniques 22
3. Held Demonstration 27
Site description and waste characteristics 27
Waste Excavation 31
Waste treatment tests 45
Foam application and efficiency tests 47
Air sampling and analysis 49
v
-------
CONTENTS (continued)
Page
4. Results of Field Demonstration 76
Enclosure air emission control system monitoring . . 76
Perimeter air monitoring 135
Air speciation data 135
Foam efficiency evaluation 137
Tar processing 150
Equipment-Related Operational Factors 157
Waste characterization 162
5. Quality Assurance Summary 168
Enclosure exhaust air monitoring for S02 and Hydrocarbons 168
Air speciation measurements 181
Waste characterization 182
Flux chamber measurements 186
6. Trial Excavation Study Costs 188
Enclosure and air exhaust controls 188
Foam costs 190
Tar processing costs 190
Excavation costs 191
Air monitoring 191
Cost summary 193
References 194
Appendices
A. Daily Operating Logs and Geotechnical Observations
B. Enclosure Air Exhaust Monitoring Data
C. Perimeter THC and S02 Monitoring Data
D. Air Emission Control System Design Basis
E. Projected Excavation Rate
F. Waste Analysis
G. Meteorological Data
H. Project Audit Overview
I. Quality Assurance Project Plan
{Note: Appendices A through I, as referenced here and In the body of the report, are available In limited number. Please
mall your request to: Jack Hubbard, U.S. Environmental Protection Agency, 25 West Martin Luther King Drive,
Cincinnati, OH 45268.) .
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
ge
14
15
17
18
20
24
24
26
28
30
32
34
34
36
36
37
38
38
51
53
55
59
FIGURES
Enclosure Plan and Section
Excavation Site Enclosure
Ventilation Air Cleaning Equipment General Arrangement
Ventilation Air Cleaning Equipment and Ducting Layout
Scrubber Cross Section
Pug Mill
Pug Mill Paddles During Tar Processing
Char/Mud Crusher Schematic
Site Location Map, McColl Site, Fullerton, California
McColl Site
Approximate Cross Section of L-4 Sump Based on Boring Logs
Caterpillar Model 2156 DLC Trackhoe
John Deere Model 710 Loader/Backhoe
Bobcat Model 743 Loader
Roll-Off Bin
Small and Large Trench Shields
Level B Personal Protection Equipment
Level A Personal Protection Equipment
McColl Inlet Site
McColl Outlet Site
Outlet and Inlet S02 Sample Systems
S02 Sampling Train Assembly
Outlet and Inlet THC Sample Systems
vii
-------
FIGURES (continued)
Number Page
24 McColl Site Showing Perimeter Monitor Locations 66
25 Plot of S02 Concentrations vs Time for June 9, 1990 79
26 Plot of THC Concentrations vs Time for June 9, 1990 80
27 Late Afternoon Portion of S02 Outlet Strip Chart for June 9 81
28 Late Afternoon Portion of THC Outlet Strip Chart for June 9 82
29 Legend for Data Plots on Figures 30 Through 65 87
30 Plot of S02 Concentrations vs Time for June 9, 1990 88
31 Plot of S02 Concentrations vs Time for June 12, 1990 89
32 Plot of S02 Concentrations vs Time for June 13,199D 90
33 Plot of S02 Concentrations vs Time for June 14, 1990 91
34 Plot of S02 Concentrations vs Time for June 15, 1990 92
35 Plot of S02 Concentrations vs Time for June 18, 1990 93
36 Plot of S02 Concentrations vs Time for June 26, 1990 94
37 Plot of S02 Concentrations vs Time for June 28, 1990 95
38 Plot of S02 Concentrations vs Time for June 29,1990 96
39 Plot of S02 Concentrations vs Time for July 2, 1990 97
40 Plot of S02 Concentrations vs Time for July 10, 1990 98
41 Plot of S02 Concentrations vs Time for July 11, 1990 99
42 Plot of S02 Concentrations vs Time for July 12, 1990 100
43 Plot of S02 Concentrations vs Time for July 13, 1990 101
44 Plot of S02 Concentrations vs Time for July 14, 1990 102
45 Plot of S02 Concentrations vs Time for July 16, 1990 103
46 Plot of S02 Concentrations vs Time for July 17, 1990 104
47 Plot of THC Concentrations vs Time for June 7,1990 109
48 Plot of THC Concentrations vs Time for June 8,1990 110
49 Plot of THC Concentrations vs Time for June 9, 1990 111
50 Plot of THC Concentrations vs Time for June 12, 1990 112
viii
-------
FIGURES (continued)
Number Page
51 Plot of THC Concentrations vs Time for June 13, 1990 113
52 Plot of THC Concentrations vs Time for June 14, 1990 114
53 Plot of THC Concentrations vs Time for June 15, 1990 115
54 Plot of THC Concentrations vs Time for June 18, 1990 116
55 Plot of THC Concentrations vs Time for June 26, 1990 117
56 Plot of THC Concentrations vs Time for June 28,1990 118
57 Plot of THC Concentrations vs Time for June 29, 1990 119
58 Plot of THC Concentrations vs Time for July 2, 1990 120
59 Plot of THC Concentrations vs Time for July 10, 1990 121
60 Plot of THC Concentrations vs Time for July 11, 1990 122
61 Plot of THC Concentrations vs Time for July 12, 1990 123
62 Plot of THC Concentrations vs Time for July 13,1990 124
63 Plot of THC Concentrations vs Time for July 14, 1990 125
64 Plot of THC Concentrations vs Time for July 16, 1990 126
65 Plot of THC Concentrations vs Time for July 17, 1990 127
66 McColl Site in Fullerton, California 136
67 Treated Tar Pellets Broken Into Two Pieces 155
ix
-------
TABLES
Number Page
1 Maximum and Average Trial Excavation Rates 5
2 Waste Characteristics, As-Received Basis 6
3 Summary of Onsite Costs 7
4 Scrubber and Fan Specifications 19
5 Specifications for Carbon Bed Adsorber 21
6 Properties of Foam Reagents 22
7 Fly Ash and Portland Cement Properties 23
8 Chemical Characterization of Waste in Pit L-4 33
9 Excavation Quantity Estimates 41
10 Maximum and Average Excavation Rates 43
11 Tar Treatment Test Feed Quantities .46
12 Tar Treatment Test Operational Data .48
13 Sampling Procedures, Equipment, and Methods 52
14 Monitor Response for Various Molecular Structures 63
15 Field Equipment Calibration Summary 65
16 Field Checks of Sampling Equipment 65
17 Summary of Air Speciation Bag Collection Activities 70
18 Background Air Data .77
19 Example S02 and THC Data for June 9 78
20 S02 Inlet and Outlet Concentrations 84
21 THC Inlet and Outlet Concentrations 107
22 Excavation Area S02 Data 130
x
-------
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
TABLES (continued)
Page
Summary of Exhaust Gas Conditions 133
Effect of Foam Use During Mud Excavation and Movement 139
Effect of Foam Use During Tar Excavation and Movement 140
Effect of Foam Use During Char Excavation and Movement ......... 141
Five-Minute Data Averages 143
Tar and Total Material Processing Rates 151
Tar Treatment Product Characteristics 153
Proximate Analysis Comparisons Among Raw and Treated Tar 155
Emission Changes During Tar Treatment Tests 156
Operational Factors Affecting Trial Excavation 158
Trial Excavation Waste Sample Analysis - Proximate and Ultimate 163
Bore Samples - Proximate and Ultimate Analysis 164
BTEX Content of Waste 164
Sample Extraction Tests for Metals 166
TCLP Organic Compounds 167
California Wet Test Semivolatile Organic Compounds 167
RT1 CEM Sample System Audit . . . 170
QA Objectives for Precision, Analyzer Calibration Error, Completeness,
and Method Detection Limit 172
CEM Precision Data 173
S02 Calibration Summary 174
THC Calibration Summary 175
CEM Response Time Results 176
Comparison of Strip Chart Hourly Data with ODESSA Hourly Data
For S02 177
Comparison of Strip Chart Hourly Data with ODESSA Hourly Data
For THC 178
Summary of S02 Emission Data Comparison 179
xi
-------
TABLES (continued)
Number Page
48 Field Equipment Calibration 180
49 Summary of Laboratory S02 Audit Results 181
50 Waste Characterization Method Detection Limits, Metals 184
51 Waste Characterization Method Detection Limits, Organics 185
52 Sample Matrix Spike Recovery Data, Metals 186
53 Example Matrix and Surrogate Spike Data, Organics 186
54 Costs Associated With Temporary Enclosure 189
55 Costs for Air Control System 189
56 Foam Applications Costs 190
57 Tar Processing Costs 191
58 Excavation Labor and Equipment Costs 192
59 Air Monitoring Costs 192
60 Summary of Trial Excavation Field Costs 193
xii
-------
AA
acfm
ARAR
ASTM
BNA
CAM-WET
CEM
CERCLA
CFR
cm/s
cfm
ft3
yd3
DAS
DSA
EPA
EP Tox
FID
ft/s
g/mL
GC/MS
h
hp
ABBREVIATIONS AND SYMBOLS
Atomic Absorption
Actual cubic feet per minute
Applicable or Relevant and Appropriate Requirements
American Society for Testing and Materials
Base neutral/acid (extractable)
California Wet Extraction Test for Metals
Continuous Emission Monitor
Comprehensive Environmental Response, Compensation, and
Liability Act of 1980
Code of Federal Regulations
Centimeters per second
Cubic feet per minute
Cubic feet
Cubic yard
Data acquisition system
Drum storage area
Environmental Protection Agency
Extraction Procedure Toxicity Test-leach test
Flame ionization detector
Feet per second
Grams per milliliter
Gas chromatograph/mass spectrometer
Hour
Horsepower
xiii
-------
ABBREVIATIONS AND SYMBOLS (continued)
HSWA
Hazardous and Solid Waste Amendments to RCRA-1984
ICP
Inductively-coupled plasma
kW
Kilowatt(s)
Ib/min
pounds per minute
m/s
meters per second
MDL
Method detection limit
mg/kg
Milligrams per killogram
mg/m2/min
Milligrams per square meter per minute
mg/L
Milligrams per liter
mL/g
Milliliters per gram
NOP
National Contingency Plan
NPL
National Priorities List
ORD
Office of Research and Development
OSHA
Occupational Safety and Health Act
OSWER
Office of Solid Waste and Emergency Response
PAHs
Polycyclic aromatic hydrocarbon
Pb
Lead
PID
Photo ionization detector
PPb
Parts per billion
PPE
Personal protection equipment
ppm
Parts per million
psi
Pounds per square inch
QAPP
Quality Assurance Project Plan
RCRA
Resource Conservation and Recovery Act of 1976
RI/FS
Remedial Investigation/Feasibility Study
RPD
Relative Percent Difference
RREL
Risk Reduction Engineering Laboratory
xiv
-------
ABBREVIATIONS AND SYMBOLS (continued)
SARA Superfund Amendments and Reauthorization Act of 1986
SITE Superfund Innovative Technology Evaluation Program
S02 Sulfur dioxide
TCEP Totally encapsulating chemical suit
TCLP Toxicity Characteristic Leaching Procedure
THC Total hydrocarbons
TNMHC Total nonmethane hydrocarbons
TOC Total organic carbon
TSCA Toxic Substances Control Act of 1985
^m Micrometer
ng/L Micrograms per liter
Mg/m3 Micrograms per cubic meter
VOC Volatile organic compound
xv
-------
ACKNOWLEDGMENTS
This document was developed jointly by EPA Region IX's Superfund Program,
EPA Office of Research and Development's (ORD) Superfund Innovative Technology
Evaluation (SITE) Program and ORD's Air and Energy Engineering Research Labora-
tory (AEERL) with assistance provided by PEI Associates, Inc., in fulfillment of contract
No. 68-02-4284. Extensive subcontract support was provided by Edward Aul and
Associates, Inc., and IT Corporation.
EPA technical oversight was provided by Joe McSorley (AEERL), Jack Hubbard
(SITE), Pam Wieman (Region IX), and John Blevins (Region IX).
Significant contributions to the technical scope of the project was provided by
Dr. William A. Vance from the California Department of Health Services and by Edward
F. Aul of Edward Aul and Associates, Inc.
Valuable administrative assistance was provided by the following people from
AEERL: Malcolm Huneycutt, Alice Gagnon, and Judith Ford.
Without the management support of the following people, the project would not
have occurred: Steve James and John Martin (SITE); Robert Hall (AEERL); and Dave
Jones, Jerry Clifford, and Jeff Zelikson (Region IX).
Peer review of the project was provided by the following persons: Paul de
Percin, Mary Gaughn, and Robert Stenburg (SITE); Dan Opalski and Dante Rodriguez
(Region IX); and Z.M. Michael Wang (a professional engineer from North Carolina).
xvi
-------
SECTION 1
EXECUTIVE SUMMARY
Region IX of the U.S. Environmental Protection Agency (EPA), in cooperation
with EPA's Air and Energy Engineering Research Laboratory (AEERL), and EPA's
Superfund Innovative Technology Evaluation (SITE) Program, and with assistance from
the California Department of Health Services (DHS), conducted a trial waste excavation
project at the McColl Superfund site in Fullerton, California.
The information gained from the trial excavation will be presented in two
separate reports. This report (Technical Evaluation Report) will present the raw data
collected during the trial excavation and analysis. An Application Analysis (AA) Report
will follow this report. The AA report will analyze the data presented in this report in
more detail, and will evaluate the application of technologies used during the trial exca-
vation to the final remediation of McColl and other Superfund sites.
In the early to mid-1940's, the McColl site was used for disposal of acidic refin-
ery sludge, and in 1982, it was placed on the National Priorities List (NPL). The
McColl waste is known to release volatile organic compounds (VOCs) and sulfur diox-
ide (S02) whenever disturbed. Since 1984, the entire site has been covered with soil
in an attempt to minimize atmospheric emissions of VOCs and S02.
In February 1989, EPA and DHS issued a proposed plan for the McColl project
selecting thermal destruction, either on or off site, as the preferred remedy. An impor-
tant component of this remedy is the excavation and waste-handling activities that
must occur as a precursor to thermal destruction. The overall goal of the trial excava-
tion was to obtain information pertaining to these activities to support the selection of
thermal destruction as the preferred remedy and to aid in the design of a thermal de-
struction remedy after its selection in a Record of Decision (ROD).
EPA determined that the trial excavation was necessary to ascertain if the
McColl waste could be excavated with conventional equipment without releasing signif-
icant amounts of VOCs and S02 to the surrounding community. The trial excavation
was also necessary to define the treatment needed, if any, to improve the handling
characteristics of the waste as a precursor to thermal destruction.
1
-------
t.1 Objectives
The trial excavation was conducted on a portion of the Los Coyotes sump L-4
(see Figure 10 in Section 3). The objectives of the trial excavation are presented
beiow. Based on the goal and objectives of the project, EPA believes that the trial
excavation was successful and significant information was obtained that will be useful
in the design phase of the McColl remediation process.
Objective 1: To excavate approximately 100 yards of waste to assess waste-
handling characteristics and to determine if any treatment is required to improve
handling characteristics as a precursor to thermal destruction.
More than 130 solid cubic yards of waste material (mud, tar, and char) was
excavated under the enclosure using conventional excavation methods (Table 9).
During the trial excavation, it was determined that the mud and char material did
not need further treatment. For the mud, it was apparent that the waste could be
easily sized to the nominal 2-inch-diameter thermal destruction requirement. For the
char, it was determined that more than 50 percent of the excavated char was under
2 inches in diameter and that the remaining material could easily be sized using con-
ventional methods [i.e., pug mill, shredder (Section 4.5.3)].
The tar material was determined to require additional treatment to allow for
future processing into a thermal destruction unit. This was accomplished by mixing
the tar with cement or fly ash and water in a pug mill. The result of this treatment
process was pellets that were less than 2 inches in diameter (Section 4.5.1).
Objective 2: To determine the atmospheric emissions resulting from the
excavation activities.
This objective was only partially achieved during the trial excavation. Data for
S02 and total hydrocarbons (THC) are reported; however, no data for organic species
or reduced sulfur species are reported.
High quality data were obtained for S02 and THC emissions exiting the
enclosure exhaust treatment system. Five-minute averages for S02 emissions were
maintained at less than 1 ppm throughout the project. The highest 5-minute average
for THC emissions was 98.1 ppm (Sections 4.2 through 4.4).
Samples for organic and reduced sulfur compounds were collected from the
stack and analyzed, but were determined invalid by an EPA audit (Section 4.3).
2
-------
it is known that benzene (a known carcinogen), toluene, ethyl benzene, and
xylenes are the major constituents of the THC concentrations reported, but no quantifi-
able concentrations for these compounds can be reported for the reason listed above.
Objective 3: To assess the degree of S02 and THC emission control achieved
through the use of an enclosure and an enclosure exhaust treatment system.
This objective was achieved by erecting an enclosure around the excavation
area and exhausting the ventilation air through an enclosure exhaust treatment system
consisting of a sodium-hydroxide wet scrubber and an activated carbon unit.
The daily average removal efficiency for S02 ranged from 71.8 percent to 99.9
percent with the majority of the days achieving over 90 percent removal (Table 20).
The daily average removal efficiency for THC ranged from 15.8 percent to 90.7
percent with the majority of the days achieving over 50 percent removal (Table 21).
Objective 4: To determine the emission levels for S02 and VOCs at the fence-
line of the McColl site as an indicator of impacts on the local community.
This objective was partially achieved for the reasons outlined in Objective 2.
Reliable data for S02 and THC emissions were collected at four perimeter monitoring
stations (see Figure 66 in Section 4) with no levels being detected that would
adversely affect the surrounding community (Section 4.2).
Samples for organic and reduced sulfur compounds were collected at the
fence-line and in the community and analyzed, but were determined invalid by an EPA
audit (Section 4.3).
It is known that benzene (a known carcinogen), toluene, ethyl benzene, and
xylenes are the major constituents of the THC concentrations reported, but no quantifi-
able concentrations for these compounds can be reported for the reason listed above.
Objective 5: To assess the effectiveness of vapor-suppressing foam.
This objective was partially achieved. Reduction efficiency rates have been
calculated for dynamic conditions. Reduction efficiency rates could not be calculated
for static conditions because analytical data were determined invalid by an EPA audit.
In dynamic conditions, it has been estimated that the vapor-suppression foam
can be up to 80 percent effective for S02 control and 60 percent effective for THC
control (Section 4.4).
3
-------
Static flux chamber measurements were conducted on the mud, tar, and char
within the enclosure. The gas streams from these tests were analyzed for organic
compounds and reduced-sulfur compounds. An EPA audit, however, determined the
data to be invalid (Section 4.4.2).
Objective 6: To assess potential problems that might occur during excavation.
Assessments were made regarding problems that occurred due to the follow-
ing: higher-than-expected emissions of S02 and THC from the tar and char; high
particulate diesel emissions; heat gain; working in Level B and Level A; excess water
in a confined space; and seepage of tar material {Section 4.6).
1.2 Excavation and Waste Processing
Removal of overburden and excavation of the underlying waste were readily
performed with a trackhoe equipped with an extended boom and a 1-yd3 bucket. The
waste, which was found to be fairly well segregated into layers, was placed in roll-off
bins or piles for subsequent use. Removal of the overburden proceeded routinely, fol-
lowed by excavation of a 3-foot-thick mud layer. A 4-foot-thick tar layer was excavated
next. After the tar was removed, a trench shield was placed in the excavated area to
reduce seepage of additional tar into the opening. After the tar layer was excavated, a
hard, coal-like, char layer was encountered. This material was broken up and excavat-
ed with the trackhoe.
During the tar excavation, S02 and THC levels within the enclosure increased
dramatically and reached 5-minute average values of 1000 (June 12th) and 492 ppm
(June 14), respectively. The enclosure exhaust treatment system removed up to 99.9
percent of the S02 and 60 percent of the THC during this excavation period. The use
of the enclosure and enclosure exhaust treatment system prevented any significant
amounts of these pollutants from reaching the site perimeter, as evidenced by the low
concentrations measured there. Detailed air concentrations data are presented in
Section 4 and Appendix B. The higher-than-expected concentrations within the enclo-
sure required an upgrading of personal protection equipment to Level A (completely
encapsulated suit with supplied air).
Char excavation was also accompanied by high concentrations of S02 and
THC, reaching 5-minute average values of 755 and 350 ppm, respectively. The
enclosure exhaust treatment system operated efficiently during the entire study with up
to 99 percent removal of the S02 and up to 90.7 percent removal of the THC.
Higher-than-expected levels of S02 and THC within the enclosure were caused
by the failure of vapor-suppressing foams to form an impermeable membrane over the
exposed wastes. The foams reacted with the extremely acidic waste, which severely
impacted the foam's ability to suppress emissions.
4
-------
This ability was improved somewhat, however, when the concentration of foam
reagents in water was increased. Though difficult to estimate, the overall reduction
achieved by applying foam was estimated at up to 80 percent for S02 and 60 percent
for THC, based on concentrations measured at the enclosure exhaust treatment sys-
tem inlet during excavation activities with and without foam.
In all, 137 yd3 of waste and 101 yd3 of overburden were excavated. Maximum
and average trial excavation rates are summarized in Table 1.
TABLE 1. MAXIMUM AND AVERAGE TRIAL
EXCAVATION RATES
(yd /h)
Component
Maximum
Average
Overburden
51
7.6
Mud
65
4.1
Tar
58
4.3
Char
9
2.6
The average excavation rates achieved during this trial excavation will be in-
creased considerably during fuli-scale excavation as fewer observations and measure-
ments would be needed. Anticipated average excavation rates that could be achieved
during full-scale excavation are estimated at 49, 32, and 25 yd3/h for overburden and
mud, tar, and char, respectively.
The tar waste was further processed to reduce its size and to form a solid and
easier-to-handle pellet. This was accomplished by mixing the tar with cement, fly ash,
and water in a pug mill. Ten test runs were made within the enclosure at various ra-
tios of tar, cement, fly ash, and water. A ratio of 1 part tar to between 2.3 and 7 parts
cement and fly ash and from 0.26 to 1 part water formed a solid, easy-to-handle pellet,
Tar processing rates of approximately 3 tons/h were achieved during the trial excava-
tion, and it is estimated that this rate could be increased by up to a factor of 2 with a
more continuous operation. Indications were evident that tar processing with alkaline
materials such as cement and fly ash reduced the amount of S02 released by the tar.
The mud and char waste fractions did not require further processing, but could have
been fed through the pug mill, if necessary.
1.3 Air Emission Controls
Previous investigations at the McColl site indicated that the waste has the
potential to emit significant amounts of VOCs, organic sulfur compounds, and S02.
For this reason, excavation of the waste could potentially have significant air emission
impacts on workers and the community.
5
-------
For the trial excavation, this potential air emission impact was mitigated from a
community impact standpoint by the erection of a temporary enclosure 60 feet wide,
160 feet long, and 26 feet high at the center over the excavation area. Air from the
enclosure was vented through an enclosure exhaust treatment system consisting of a
sodium-hydroxide-based wet scrubber and an activated-carbon adsorber in series
before being released to the ambient air.
For the trial excavation, this potential air emission impact was mitigated from a
worker impact standpoint by having workers wear Level B or Level A protection at all
times while inside the enclosure.
Concentrations of S02 and THC were continuously monitored before and after
the enclosure exhaust treatment system. As part of a supplemental sampling and
analytical effort, samples for speciation of organic and reduced sulfur compounds
occurred at the stack inlet and outlet, fence-line, and in the community. However,
these data are not reported because of invalidation by an EPA, audit. The basis for the
enclosure exhaust treatment system design is presented in Appendix D.
1.4 Waste Characterization
Samples of excavated waste were analyzed to determine heat value and the
concentrations of selected constituents. The information obtained by these analyses is
summarized in Table 2, based on information presented in Subsection 4.7.
TABLE 2. WASTE CHARACTERISTICS, AS-RECEIVED BASIS
Mud Tar Treated tar Char
Moisture, %
13.2
11.6
fi.l
21.2
Sulfur, %
0.8
10.6
3.6
4.5
Fixed carbon, %
0.2
16.9
2.0
4.0
Ash, %
82.9
1.6
7Ł.9
54.7
Benzene, ppm
<0.7
240
NAC
97
Toluene, ppm
1.5
580
NA
150
Xylene, ppm
Ethyl benzene, ppm
8.6
910
NA
220
0.9
140
NA
35
Heat value, Btu/lb
<500
9160
2200
5200
a NA « Not analyzed. Use of cement additive would reduce con-
centrations found in raw tar sample.
Toxicity characteristics of the raw tar and char were determined by the Toxicity
Characteristics Leaching Procedure (TCLP) and California Wet Test. No metal
constituents exceeded the regulatory limit in either case. Benzene in the tar and char
waste extract exceeded the EPA TCLP limit of 500 Mg/liter bv greater than a factor
of 2.
6
-------
1.5 Community Impact
Perimeter air monitoring for S02 and THC was conducted continually during this
study. Windspeed and direction were also recorded continually at the site. This
information was obtained to comply with the Community Contingency Plan, which
mandates that all site work be stopped if S02 levels at the perimeter exceed 0.5 ppm
for 5 minutes or if THC levels exceed 70 ppm for 30 seconds. These levels were
never reached during this study. The maximum 1-hour readings obtained at any pe-
rimeter station in June, which was the period of highest emissions from the waste,
were 0.08 ppm for S02 and 21.9 ppm for THC.
Specific compounds in the air at the perimeter and in the neighborhood were
sampled and analyzed; however, these data are not included in this report because
they were deemed invalid by an EPA audit.
1.6 Costs of Excavation and Tar Processing
The costs for the field aspects of this trial excavation work consisted of those
involved with the enclosure and enclosure exhaust treatment system, actual excavation
labor and equipment, foam application, tar processing, and air monitoring. Much of
the equipment for this project (e.g., enclosure framework, scrubber, and excavation
machinery) was rented on a monthly basis; therefore, total costs were comprised of
the monthly machinery charges, labor, and fixed costs required to mobilize and demo-
bilize. These costs, which are presented in Section 6 of this report, are summarized in
Table 3 for the 2-month duration of the field work.
TABLE 3. SUMMARY OF 0NSITE COSTS
Item
Total cost, $
Enclosure
70,976
Air exhaust control system
40,415
Foam vapor suppressants
89,591
Excavation
82,512
Tar processing
17,367
Air monitoring
100.160
Total
401,021
a Based on 18 days of excavation.
1.7 Conclusions
This section presents the conclusion reached from the data obtained during the
trial excavation. It is important to remember that the trial excavation was a research
7
-------
project designed to gather information for use in the design of the final remediation for
Superfund sites and specifically for the McColl Superfund site in Fullerton, California.
The overall goal of the trial excavation was to obtain information pertaining to
excavation and waste-handling activities to support the selection of thermal destruction
as the preferred remedy and to aid in the design of a thermal destruction remedy after
selection in a Record of Decision (ROD). An important question to answer with infor-
mation from the trial excavation is whether the McColl waste can be excavated with
conventional equipment without significant impacts to the community. Based on this
goal, the following conclusions have been reached pertaining to the trial excavation:
Excavation of more than 130 solid cubic yards of waste material from
sump L-4 was accomplished with conventional excavation equipment
without significant adverse impacts on the community.
Excavation under an enclosure is technically feasible.
The enclosure used during the trial excavation was successfully operated
at or near negative pressure, which allowed for emissions generated
during the excavation activities to be processed through an enclosure
exhaust treatment system consisting of a sodium hydroxide wet scrubber
and an activated carbon unit.
Although unanticipated problems during the trial excavation impeded the
ability to excavate under the enclosure, EPA believes that these problems
can be resolved by engineering practices during the design of the final
remedy. The most important impediment of the trial excavation was the
higher-than-expected THC and S02 emissions within the enclosure.
The higher-than-expected emissions of S02 and THC during the trial
excavation necessitated upgrading the personal protective equipment for
the workers within the enclosure from Level B protection to Level A pro-
tection (completely enclosed chemical-resistant suit with supplied air).
During the final remediation of McColl, the high emission rates of S02
and THC could still occur and the collection and treatment of the emis-
sions generated using an enclosure and enclosure exhaust treatment
system is appropriate and recommended.
Excavation and waste handling activities are not technically feasible with-
out an enclosure and enclosure exhaust treatment system.
S02 emissions generated from the excavation activities can be effectively
treated (up to 99% removal efficiency) using existing technologies.
8
-------
The high S02 emissions entering the sodium hydroxide wet scrubber
were efficiently treated to less than 1 ppm throughout the trial excava-
tion. The removal efficiencies were greater than 95 percent for the ma-
jority of the trial excavation and reached up to 99 percent removal.
THC emissions generated from the excavation activities can be effectively
treated (up to 90.7% removal efficiency) using existing technologies.
Although the THC emissions were not controlled as effectively as expect-
ed (greater than 90%) with activated carbon, the removal efficiency
ranged from 40 to 90.7 percent throughout the trial excavation. EPA
believes that the less-than-expected removal efficiencies can be remedied
during the design phase of tie final remediation. Therefore, based on
other experiences with activated carbon, it is an appropriate technology
for removal of organics entering the enclosure exhaust control system.
The waste material was successfully treated to improve its handling char-
acteristic so as to allow easy processing into a thermal destruction unit, if
desired.
Visual observations indicated that the char and mud fraction of the waste
did not need further size reduction to ensure easy processing into a
thermal destruction unit. It was determined that more than 50 percent of
the char material excavated was less than 2 inches in diameter (the
criterion for solids processed into a thermal destruction unit) and that the
remaining other material could be easily processed with conventional
equipment (i.e., pug mill, shredder, etc.).
The tar material was treated with mixtures of cement, fly ash, and water
in a pug mill. Both the cement and fly ash worked well in improving the
handling characteristics of the tar material. The field test has shown that
the optimum ratio (pound/pound) of tar to char/fly ash has ranged from
1:7 to 2.3:7. Mixtures within this range produced an easily handled solid
with the majority of the material being less than 2 inches in diameter.
Excavation and treatment of the waste material at McColl can be effec-
tively performed by workers in protective clothing.
All excavation and treatment during the trial excavation were performed
with workers in either Level B or Level A protective clothing within the
enclosure. Although productivity is affected when employee work is
conducted in protective clothing, the trial excavation demonstrated that
conventional excavation equipment can be effectively operated in both
9
-------
Level B and Level A protective gear. Processing of the tar material also
was conducted effectively in both levels of protection.
The trial excavation had no significant adverse impacts (i.e., exceedance
of health-based levels established in the McColl Contingency Plan) on the
surrounding community.
Based on personnel observations during the trial excavation, the noise
level related to the excavation and treatment activities was minimal. At
no time during the trial excavation were the health-based levels (estab-
lished in the McColl contingency plan for S02 and THC) exceeded at the
fence-line monitoring stations. Although odor complaints were received
during the trial excavation period, they were not excessive. Most of the
complaints were received after the trial excavation/treatment activities
were completed for the day, and may not have been related to the exca-
vation/treatment activities.
The vapor-suppressing foam did not perform as anticipated in controlling
S02 and THC emission within the enclosure, and cannot be relied upon
exclusively to control emissions during activity-related waste disturbanc-
es.
Because analytical data were deemed invalid as a result of an EPA audit,
no results are available concerning the static effectiveness of the vapor-
suppressing foam in relation to the McColl waste.
Visual observations and dynamic-condition calculations, however, indi-
cate that the vapor-suppressing foam was not as efficient as anticipated
in controlling emissions from activities related to excavating and process-
ing the waste. Visual observations indicated that the foam chemically
reacted with the McColl waste, which inhibited its ability to form a vapor-
suppressing seal on the waste. This reaction caused the foam to
change color (from yellow to red and orange) and to disintegrate before
forming a seal on the waste.
Dynamic-condition calculations indicated that the effectiveness of the
vapor-suppressing foam ranges from 50 to 80 percent, depending on the
activity and the compound of concern.
Based on toxicity characteristic leaching procedure (TCLP) results on the
raw waste, it has been determined that the waste would not pass the
RCRA Land Ban toxicity characteristics (TC Rule) requirement for
benzene in the char and tar.
10
-------
1.8 Observations
This section lists observations made regarding activities that occurred during
the trial excavation. These observations are qualitative in nature, and no quantitative
data exists to support them; however, they represent best engineering judgment in
relation to activities related to the trial excavation.
it is believed that community communication on this project was effective,
and a necessary part of the project. The community interaction was
important to the success of the trial excavation and to the passive odor
and health effects survey conducted by the California Department of
Health Services.
Excess water introduced into the enclosure through the foaming activities
had a significant impact on operations within the enclosure. The excess
water made the ground surface slippery for both workers and equipment.
Based on visual observation and qualitative calculations, it has been
determined that the trench shield was a very effective tool in minimizing
the amount of tar material that could seep into the excavation area. It
was also determined that the trench shield was not needed to shore up
the soils or char material within the excavation area.
Having workers in Level A protective gear adversely affected their
productivity and communication, but did not make excavation activities
unfeasible.
Lower airflow rates through the activated carbon unit increased the THC
removal efficiencies. This supports the theory that residence time is a
critical factor in the ability of activated carbon to remove organic com-
pounds in an air stream.
Contrary to original plans, EPA had to move major equipment into and
out of the enclosure during operations. It has been determined that no
adverse impacts on the community or workers resulted from opening the
enclosure for short periods of time (under 1 hour) to allow for efficient
equipment movement.
It is believed that the pug mill, based on the results of the tar processing,
could be used to process the char and mud fractions of the McCoII
waste effectively for use in a thermal destruction remedy.
11
-------
Due to tar seepage into the excavation area, approximately 100 cubic
yards of material could not be replaced into the excavation pit at the
completion of the project. This material was stockpiled on site in the
staging area under a plastic liner covered by topsoil.
It is believed that under true field conditions (not research-restricted
conditions) the excavation rating associated with the equipment could be
achieved. The excavation rates achieved during the trial excavation were
artificially constrained to allow for data collected and visual observations.
The excavation rates were also constrained by unexpected high S02 and
THC emissions rates.
The observation camera used was an invaluable tool in observing/
recording activities that occurred within the enclosure. The camera also
allowed all workers to be observed from a health and safety standpoint.
The camera also assisted in reducing the number of employees neces-
sary within the enclosure, which allowed for more efficient operations and
reduced the risk of employee accidents.
12
-------
SECTION 2
DESCRIPTION OF TECHNOLOGIES
A number of measures were implemented during excavation operations to en-
sure that these operations did not create a public health impact. These measures
were aimed at controlling air emission releases from the operations, since this was the
only potential source of impact expected. The following measures were implemented
for this purpose:
o Use of enclosure structure
° S02 scrubber
e Activated carbon unit
° Use of vapor-suppressing foam
Waste processing technologies planned during this program consisted of size
reduction by crushing the char and mud wastes, and tar solidification using cement
and fly ash mixtures.
2.1 Enclosure and Exhaust Air Control System
2.1.1 Excavation Enclosure
A rigid-frame, PVC-covered enclosure structure was erected over part of the L-4
sump and adjoining land prior to the start of excavation. Before its erection, the site
was graded to provide a smooth, level area. The enclosure, supplied by Sprung In-
stant Structures and shown in Figures 1 and 2, was nominally 60 feet wide by 157 feet
long, and 26 feet high at the center. The white, opaque PVC cover was 26 mils thick
and impervious to gaseous emissions. The lower edge was covered by 12 to 18
inches of soil along the ground level to prevent air leakage. Translucent panels
located along the roof peak allowed light to enter. Personnel entry was through an
airlock door which minimized fugitive emissions during entry. Equipment was moved
inside the enclosure through a sliding door that was 14 feet high and 9 feet, 5 inches
wide.
The volume of the enclosure was approximately 192,000 ft3 and air was drawn
through the building at a rate of approximately 1,000 flf/min. This air entered the
13
-------
156 ft 11-1/4 in.
DOOR
60 ft
SLIDING
DOOR
5 ft x 9II 6 in. STD. VESTIBULE ENTRANCE
PLAN
4 In. * 6 In.
ALUMINUM
I BEAM
26 ft
60 ft
SECTION
Figure 1. Enclosure plan and section.
-------
View from east side.
r
': ¦¦ :: . ¦ ¦: ¦ 1 ¦¦ 1.
^
'¦*v->k
• '
|'-:i '¦ >v
View from west side showing air emission control system and monitoring trailer.
Figure 2. Excavation site enclosure.
15
-------
building through five small adjustable slot type air vents and was exhausted through
three dampered openings along the west side of the building. This exhaust system
provided an air turnover rate of about 7 air changes per day arid maintained a slight
negative pressure of about 0,005 inch of water inside the enclosure. This ventilation
air rate was based on maintaining the S02 level in the enclosure below 100 ppm. This
was in turn based on an estimated S02 release from the exposed waste and a 95 per-
cent reduction in these releases by use of foam suppressants.
The enclosure proved to be very effective in preventing the escape of any air
emissions and proved quite satisfactory even though it made for a confined work
space and increased temperatures approximately 20 *F above the outdoor tempera-
ture.
2.1.2 Air Emission Control System
The enclosure ventilation air was passed through an emission control system
consisting of a wet scrubber and activated carbon bed in series, followed by a fan and
vent stack as shown in Figures 3 and 4. The basis for design of the air control system
is provided in Appendix D.
Wet Scrubber
A counter-flow packed-bed wet scrubber that used sodium hydroxide (NaOH) in
water mixture was used to control sulfur dioxide emissions. Tine system was designed
for a nominal gas flow rate of 1,000 ft3/min at 1000*F and a maximum outlet S02
concentration of 2 ppm. The maximum inlet S02 concentration was estimated to be
200 ppm and the required control efficiency was therefore 99 percent. A maximum
pressure drop of 10 inches of water was specified. The scrubber selected based on
these specifications was supplied by Interel Corp. in Englewood, Colorado. The
specification for the actual scrubber and fan are shown in Table 4 and the scrubber
cross-section is shown in Figure 5.
In operation, scrubber liquid was initially maintained at a pH of 10 to 13. Con-
siderable scrubber liquor foaming was encountered at this pH level, and the pH was
reduced to the 7 to 10 range after operation showed that high S02 removal could be
maintained in this range without foaming. The nominal liquor recirculation rate of 20
gpm provided a liquor-to-gas ratio (L/G) of 20 gallons per 1000 f^/min.
Activated Carbon Bed
To reduce emissions of volatile organic compounds, a granular activated
carbon bed was installed after the wet scrubber. A knockout chamber was inserted
between the scrubber and carbon bed to trap any liquid carryover from the scrubber.
Specifications for this adsorber called for a 95 percent minimum removal of total
16
-------
DAMPERED
. OPENINGS
DOOR
INLET
VENT A
FAN (W/ STACK)
ACTIVATED CARBON
CABINET
KNOCK OUT POT
WET SCRUBBER
FILTER HOLDER
INLET
VENT B
SLIDING
DOOR
XT
-N-
TT
AIR
LOCK
DOOR
INLET
VENT C
DOOR
INLET
VENTE
INLET
VENT D
Figure 3. Ventilation air cleaning equipment general arrangement.
I
-------
DUCTING
WET
SCRUBBER
ENCLOSURE
STACK
CONNECTIONS
SAMPLE
PORTS
a
CD
SAMPLE
PORTS
DUCTING
DUCTING
DUST
FILTER
AND
HOLDER
JS91
DUCTING
FROM
OPENINGS
1 AND 2
KNOCK
,6uf POT. ,
' (55 gal f
DRUM)
AC I iVA i ED
CARBON
CABINET
MOTOR
FAN
PLYWOOD
BASE ""
iuiiu«umuiw.wj;
3UND LEVEL^^
PLYWOOD
BASE '
Figure 4. Ventilation air cleaning equipment and ducting layout.
-------
TABLE 4. SCRUBBER AND FAN SPECIFICATIONS3
Scrubber
Scrubber size
Design flowrate
Diameter
Sump capacity
Circulation rate
Pump motor
Type of packing
Packing height
Scrubber overall height
Type of mist eliminator
Empty weight
Operating weight
PURCHASE PRICE
GWX 1200
1,200 ft3/min
24 inches
190 gallons
25 GPM
1.5 hp
2-inch hollow spheres
11 feet
17 feet
Chevron
650 lb
2350 lb
$22,600.00
Fan
Material of construction
Corrosion resistant coating
Gas flowrate (standard air density)
Static pressure (Neg/Pos)
Motor rating
PURCHASE PRICE:
Steel
Polyurethane
1,200 ft/min
20 inches WG
7.5 hp
$2,200.00
Supplied by Interel Corp. 5/14/90
19
-------
GAS OUTLET-
MIST
ELIMINATOR
SPRAY
NOZZLE
PACKED
BED
ELECTRICAL
PANEL
RECIRCULATION
PUMP
LIQUID
RESERVOIR
GAS INLET
NOT TO SCALE
Figure 5. Scrubber cross section.
20
-------
organics at a flow rate of 1000 ft3/min at 100* F and a pressure drop not to exceed 5
inches of water.
A radial flow type packed bed carbon adsorber was selected from Tigg Corp. in
Pittsburgh, PA (NIXTOX Model 1500). Specifications for this unit are shown in
Table 5.
TABLE 5. SPECIFICATIONS FOR CARBON BED ADSORBER3
Flow rate, max.
Temperature, max.
Connections
Diameter/height
Adsorbent fill
Minimum contact
Shipping weight
Materials of construction
1500 ft /min
350 * F
7-in. duct
32-in./44-in.
300 lb virgin TIGG 5C 0410 (coal-based)
0.4 seconds
475 lb
Coated mild steel with 316
stainless steel screen
Purchase price $2450 FOB plant (including initial
carbon fill)
Lease payment per month $700
Virgin TIGG 5C 0410, per fill $600
a From Tigg Corporation, March 3, 1990.
2.2 Foam Vapor Suppressants
Two types of water-based commercially available foam supplied by 3-M Corpor-
ation were selected for this study: a temporary foam that is effective for up to about 1
hour, and a stabilized more permanent foam that is effective for at least one day.
These foam reagents are mixed with water and sprayed onto the waste through a
hand-held nozzle. The temporary foam is a mixture of 6 percent concentrate and 94
percent water, and the more permanent foam is produced by adding 6 percent of the
stabilizer to the temporary foam mixture. The foam was generated in a self-contained,
trailer mounted system (Boots & Coots Model 100) outside of the enclosure and
pumped through a hose that passed under the enclosure's edge to an air-aspirating
nozzle. The temporary foam was sprayed on freshly excavated waste surfaces in the
excavation pit and on waste in storage areas. Stabilized foam was sprayed on all
exposed waste after completing each day's work. According to 3M, 200 gallons of
foam concentrate (FX 9162) and 200 gallons of foam stabilizer (FX 9161) are required
to form a 1-inch-thick layer over 1 acre of surface, or about 0.9 gallon per 100 fr. The
properties of the two types of foam used in this work are shown in Table 6.
21
-------
TABLE 6. PROPERTIES OF FOAM REAGENTS3
FX-9161 foam
FX-9162 foam
Properties
stabilizer
concentrate
Appearance
Yellow, clear
liquid solution
Amber liquid
solution
Density, lb/gal
8.99
8.51
Viscosity at 77*F (25*C), cps
1500
2300
Specific gravity at 77*F (25*C)
1.08
1.02
pH at 77 * F (25*C)
-
7.8
Flash point, *F
200
-
Freeze point, *F
-
28
Minimum use temperature, *F
-
32
Storage temperature, *F
40 to 100
35 to 120
Non-corrosive
Yes
Yes
Moisture sensitive
Yes
No
Price, $/1b
4.65
2.55
3 From 3M Corp., St. Paul, MN,
2.3 Waste Treatment Techniques
2.3.1 Tar Treatment
Because of its viscous nature and size (as excavated), H; was anticipated that tar
would require some type of solidification and size reduction before it could be fed to a
thermal destruction system. The two solidification agents most widely used with haz-
ardous waste are portland cement and lime-based pozzolana (Arniella 1990). In
addition to stabilization, these agents were expected to reduce the acidity of the low-
pH tar to mitigate S02 emissions during processing. Both of these agents were
evaluated during McColl tar treatment operations.
Pozzolana are materials which contain aluminum and silica and which harden at
ambient temperatures in the presence of lime and water (by themselves, however,
they display no cementing reactions). The two most common pozzolanic materials are
fly ash and cement kiln dust. Fly ash from a nearby power plant was used for the
McColl tests since it was readily available (cement kiln dust is itself considered a
hazardous material in California and therefore more difficult to transport and use). The
chemical and physical properties of the fly ash and portland cement delivered to the
McColl Site are summarized in Table 7.
22
-------
TABLE 7. FLY ASH AND PORTLAND CEMENT PROPERTIESa
Portland
Fly ash
cement
Silicon dioxide, %
Aluminum oxide, %
Iron oxide, %
Sulfur trioxide, %
61.04
18.59
5.16
1.07
22.61
3.78
3.25
1.84
Calcium oxide, %
Loss on ignition, %
5.97
0.29
65.15
0.88
Bulk density*3, lb/ft3
86
Class F
78
Type V
Classification
a From Amcal Minerals Corporation, 1990.
k From field measurements.
Excavated tar was combined with portland cement, fly ash, and water in a pug
mill to both mix these materials and reduce the size of tar lumps. The pug mill used
for this project was a Barber Green Mixer (Model 848), shown in Figure 6, that
reportedly was built during the 1950's. Tar, cement, and fly ash were charged at one
end of the mill into a small feed hopper with a capacity of approximately 1.2 yd3.
Material moved down through the hopper and flowed onto a moving belt. The
clearance between the bottom of the hopper and the belt was almost 8 inches. The
belt transported the material to the head of the pug mill where water was added manu-
ally. The mill consisted of two shafts fitted with short heavy paddles which rotated in
the opposite inward direction (from the bottom towards the top) in an open half cylin-
der. The mixing/conveying action of the paddles pushed the material from the head
of the mill to its tail, where the mixed material fell into a small product hopper
(approximately 2 yd3 capacity). The hopper, in turn, emptied directly onto the ground.
The feed belt and paddle shafts were powered by a 175-horsepower diesel engine.
The pug mill cylinder was approximately 10 feet long, 45 inches wide, and 27
inches deep, corresponding to an overall volume of 5.1 yd3. The paddles were 7
inches long and 4 inches wide at the tip. Two paddles (set at 180 degrees from each
other) were set every 6 inches along the two tapered shafts; this resulted in a
clearance of 2 inches between paddle sets. As shown in Figure 7, each set was offset
90 degrees from adjoining sets. The throughput capacity of the mill was reported to
be almost 100 tons/hour.
23
-------
Figure 6. Pug mill (with product hopper in foreground).
Figure 7. Pug mill paddies during tar processing.
24
-------
2.3.2 Char and Mud Treatment
The objective of the char and mud processing operations planned for this
project was to reduce the size of these materials to less than 2 inches so they would
be suitable for feed to a thermal destruction system. The crusher brought on site for
this purpose was a Masterskreen Explorer, manufactured by M&KK Quarry Plant Ltd.
in 1989. With this system, material is dumped into a 4-yd3 tray feed hopper fitted with
6-inch stationary bars. From the hopper, material is transported by a feed belt into the
jaws of the crusher. After passing through the crusher, material is picked up by a
product conveyor and transported to a vibrating screen with 2-inch square openings.
Undersize material passes through the screen to the ground while oversize material
rolls off the screen to another pile on the ground. The conveyor belts, crusher, and
hydraulic control system were powered by a diesel engine.
The crusher was expected to operate on both char alone and a mixture of char
and mud. A schematic of the crusher is shown in Figure 8. The overall dimensions of
the unit were 51 feet long, 7 feet wide, and 17 feet high.
25
-------
2-»n.
VIBRATING
SCREEN
BAR SCREEN FEED HOPPER
MAIN
CONVEYOR
BELT v
CRUSHER
$ CRUSHER jmiimm
OVER
SIZE
DIESEL
ENGINE
PRODUCT
Figure 8. Char/mud crusher schematic.
-------
SECTION 3
FIELD DEMONSTRATION
This section presents a brief history of the site, a description of waste char-
acteristics, and a description of the various aspects of work conducted during this trial
excavation study.
3.1 Site Description and Waste Characteristics
3.1.1 Site Description arid History
The McColl hazardous waste site is an inactive waste disposal facility located at
2650 Rosecrans Avenue in the city of Fullerton, Orange County, California (Figure 9).
The site was used in the early and mid-1940s for the disposal of acidic refinery sludge,
a byproduct from the production of aviation fuel. A series of pits or sumps were ex-
cavated on the site to receive the refinery sludge at that time. Onsite disposal of
refinery sludge ceased in 1946. From 1951 through 1962, fill material (soil) and drilling
mud from oil exploration activities near the Coyote Hills were deposited in some of the
pits in an attempt to make the site suitable for future development.
By 1962, the Upper Ramparts area had been covered with soil, and has existed
since that time as unoccupied open space. In the eariy 1980s, a clay cap was placed
on the Lower Ramparts area to reduce odors. The Los Coyotes area was covered
with 4 to 5 feet of soil and developed as part of the Los Coyotes Country Club golf
course.
Areas east of McColl site were subdivided and developed for residential housing
in the late 1970s and the early 1980s. Recreational facilities were constructed west of
the site at the Ralph B. Clark (formerly Los Coyotes) Regional Park. As the population
increased and development continued, residents began complaining of odors emanat-
ing from the site. Odor complaints were first received by the Orange County Health
Department in 1978. Subsequent environmental investigations at the site identified
extensive contamination.
In 1982, the McColl site was placed on the National Priority List (NPL), which
made remediation of the site eligible for funding through CERCLA.
27
-------
LA MtRADA
nawauin
COTOTE HIIXS
¦OMiemjCTs
. MeCOU. SITE
Vf
niLLCHTOM
mLLCNTON
AIRPORT
i«l
\
MN»(*UM, MWT
WTVCItttOCFWW
I
I
SCALE r-01 MILES
VICINITY MAP
Figure 9. Site location map, McColl Site, Fulierton, California.
28
-------
3.1.2 Site Characteristics
The site is approximately 20 acres in size, of which approximately 8 acres
contain waste in pits or sumps. As shown in Figure 10, the site is divided into two dis-
tinct areas, the Ramparts area and the Los Coyotes area. The Ramparts area com-
prises the eastern portion of the site and contains six buried waste pits or sumps. The
Los Coyotes area, located immediately southwest of the Ramparts area, also contains
six pits (L-1 through L-6). The six pits in this area were covered with soil during the
construction of the golf course. The site is bordered by the West Coyote Hills Oil Field
to the north, housing developments to the east and south of the Ramparts area, Los
Coyotes Country Club golf course to the south, and the Ralph B. Clark Regional Park
to the west. All pits are covered with soil, and the site is secured with a chain-link
fence and 24-hour guard.
The trial excavation work in this project was conducted in the L-4 pit in the Los
Coyotes area. This pit contains approximately 3300 cubic yards of waste, of which
approximately 100 cubic yards was planned to be excavated.
3.1.3 Waste Characteristics
Three major waste types are present at the McColl site: 1) hard, black char-like
asphaltic wastes; 2) viscous, black, tar-like wastes; and 3) gray, sludge-like drilling
mud. The predominant waste type found at the site is a black asphaltic waste that is
apparently the result of chemical and physical changes in acidic refinery sludge that
have occurred over the last 40 years. This asphaltic waste has a low pH (acidic) and
contains elevated levels of organic compounds. When disturbed, the waste emits
sulfur dioxide (S02) and hydrocarbon vapors. Based on field observations, the
following changes have occurred since this waste was disposed of (Radian 1983).
o Sulfuric acid reactions with oil constituents have continued, as evidenced
by large amounts of S02 found on site.
« Some separation in the oil/acid emulsion has occurred, which has result-
ed in a contamination of the soil/sand formation below the waste. The
soil below the waste shows pH values from about 2 to 4 and contains
acid waste and salts.
o Solidification, polymerization, and coagulation of hydrocarbon constitu-
ents have occurred, which has resulted in zones or layers of very hard,
asphaltic-type waste.
o Some fluid components in the waste, including entrained oil, have sepa-
rated from the waste and are very mobile, and they occasionally seep to
the surface.
29
-------
ramparts area
McCOLL
SITE
Ralph b. clap*
regional park
L08 COYOTES
QOLff. COURSE
CITY *0F
1 fullefIton
Figure 10. McCoII Site.
30
-------
Borings previously made in the L-4 sump showed that both tar and char were
present in fairly segregated layers under a layer of moist soil or mud, which was in
turn under approximately 8 feet of overburden soil. The anticipated waste configura-
tion based on these boring is indicated in Figure 11; this information was used to plan
the excavation work for the study (Geotechnical Professionals 1989).
Table 8 lists the waste characteristics of Pit L-4, based on the analytical results
of two waste samples collected in 1989 (CH2M Hill 1989).
During previous studies at this site, two types of air emissions were observed
when the waste was disturbed. Initial disturbance generally caused a high level or
"puff" release of contaminants, followed by a rapid decline to lower levels (Radian
1983). These steady-state emission levels were then observed for longer periods of
time and gradually decreased over several hours. The emission potential in the
Ramparts area ranged from 130 to 130,000 mg/m2 per min for S02 and 10 to 3600
mg/m2 per min for total hydrocarbons (THC) for all disturbed waste types. Average
steady-state emissions from asphaltic waste were 5200 mg/m2 per min for S02 and
190 mg/m2 per min for THC. Hydrocarbon analysis of air samples showed an
average composition of 60 percent aliphatic and oxygenated species, 30 percent
aromatic species, and 10 percent organic sulfur species. The waste composition did
vary, however, from sump to sump and even with depth within a sump (Schmidt
1989).
3.2 Waste Excavation
This section describes the equipment used for waste excavation activities and
presents a summary of the chronology of excavation operations. This summary is
based on the daily operations log contained in Appendix A. This information and
related geotechnical information provided the basis for excavation rates calculated for
the various materials encountered.
3.2.1 Excavation Equipment
The equipment used under the enclosure for the trial excavation fell into three
general categories: excavation, transport, and storage. In addition, operating person-
nel were equipped with personal protection equipment (PPE) in response to the
hazards present inside the enclosure.
Excavation equipment included a diesel-powered (Caterpillar Model 215 DLC)
tractor-mounted backhoe (or trackhoe), shown in Figure 12. This trackhoe was equip-
ped with an extended boom and 1-ycr bucket. With the extended boom, the trackhoe
had an effective digging depth of 20 feet. The trackhoe was used for excavation of
overburden, mud, tar, and char. In addition to the trackhoe, the backhoe portion
31
-------
DEPTH,
FEET
12
15
25-28
L«4 SUMP
TOP SOIL
MOIST/TARRY SOIL
CHAR
ORIGINAL SAND/SOIL
Figure 11. Approximate cross section of L-4 sump based on boring logs
(Geotechnical Professionals 1939).
32
-------
TABLE 8. CHEMICAL CHARACTERIZATION OF WASTE IN PIT L-4*
Constituent Concentration range, mq/kq
Oraanics
pH 1.0 - 1.4
Methylene chloride 0 - 0.32
Acetone 0 - 0.36
Chloroform ND"
2-Butanone 0-49
Benzene 0.2 - 11
2-Hexanone ND
Toluene 0.4 - 24
Ethyl benzene 0.1-8
Total xylenes 0.5 - 41
2-Methyl phenol ND
4-Methyl phenol ND
Benzoic acid ND
Naphthalene ND - 14
2-Methyl naphthalene ND - 17
Diethyl phthalate ND
Fluorene ND
Phenanthrene ND
Di-n-butyl phthalate ND
Bis(2-ethylhexyl)phthalate ND - 11
Thiophene ND - 21
Others' 13 - 865
Total organic carbon 3 - 62%
Metals
A1umi num 7900
Arsenic ND
Bari um 63
Beryl 1i urn ND
Calcium 2120
Chromium 19
Copper 7.5
Cobalt 5.7
Iron 9300
Lead 8.2
Manganese 126
Nickel ND
Tin NO
Zinc 32
Magnesium 1910
Potassium 859
Sodium 150
Vanadium 25
* Source: Supplemental Reevaluation of Alternatives, Final Draft, Appendices
(pp. D-8, D-9, D-ll, and D-12), February 1, 1989.
* ND « Not detected.
' Organic compounds not on the Hazardous Substances List.
33
-------
Figure 12. Caterpillar Model 215 DLC trackhoe
Figure 13. John Deere Model 710 loader/backhoe.
34
-------
of a diesel-powered (John Deere Model 710) loader/backhoe (see Figure 13) was also
used for some excavation operations. At 1/3 yd3, the backhoe bucket was smaller
than the trackhoe bucket and was used primarily during placement of the trench
shield. The effective digging depth of the wheel-mounted backhoe was about 15 feet.
The loader portion of the loader/backhoe was used for material transport
operations such as carrying excavated waste from the pit staging area to stockpiles.
The 1-yd3 loader bucket was approximately 7.5 feet in width. A diesel-powered Bobcat
Model 743 loader was also used for materials transport when the floor of the enclo-
sure became too muddy and slippery for the John Deere loader/backhoe to operate
properly. The muddy conditions were caused by the water in the foam vapor suppres-
sants and by blowdown water associated with foam application. As shown in Fig-
ure 14, the Bobcat was fitted with special chain-like tracks over its wheels to provide
adequate traction on the wet floor. The bucket of the Bobcat was 5 feet wide and had
a capacity of 1/4 yd3.
For material storage, standard 40-yd3 steel roll-off bins were utilized. These
bins, shown in Figure 15, were of the same (internal) dimensions: 6 feet high,
21.8 feet long, and 7.4 feet wide. The bins were mounted on steel rollers. At the
front, two doors swung open to provide access to material transport equipment. The
top of the door frame was 82 inches above the bottom of the bin, which limited access
by the loader/backhoe. In most cases, the bins were lined with polyethylene sheeting
before material was placed inside. A total of five bins were used during the trial
excavation for storage of mud, tar (two), fly ash, and Portland cement.
Additional equipment used during the excavation was the trench shield (see
Figure 16). The trench shield was designed to prevent collapse of overburden and
side walls during excavation operations. For the McColI trial excavation, the trench
shield also proved to be valuable in mitigating tar seepage into the excavated pit after
the pit level had reached the tar layer. The trench shield employed during this
program had dimensions of approximately 15 feet long, 10 feet wide, and 10 feet high.
Figure 16 shows that the trench shield was bolted together at two ends with cross-
members. After the trench shield frame was lowered into the pit, the open ends were
closed off by slipping two steel plates into position and securing them with backfill.
Excavation work inside the enclosure was conducted either in Level B or Lev-
el A personal protective equipment. Level B equipment consisted of supplied-air
respirators, coated Tyvek overalls, steel-toed boots, inner and outer gloves, and a
hard hat. Air bottles were mounted on the trackhoe, loader/backhoe, and Bobcat for
operator air supply; other members of the crew used air lines supplied from air
cylinders located outside the enclosure. Level A requirements added a totally encap-
sulated chemical protective (TECP) suit to the preceding equipment list. Air supplies
to these suits were either from air lines (as previously discussed) or from
35
-------
Figure 14. Bobcat Model 743 loader.
Figure 15. Roll-off bin.
36
-------
Figure 16. Small and large trench shields (without end plates).
self-contained breathing apparatus worn inside the suits. Level B and Level A
ensembles are pictured in Figures 17 and 18, respectively.
3.2.2 Excavation Operations Chronology
This section summarizes the daily progress of the excavation operations be-
tween June 7 and June 26. More detailed information on these activities is provided in
the daily operations log summaries and the Summary of Geotechnical Observations in
Appendix A.
Date Excavation activity
June 7/Th Trackhoe and loader operated without digging in the morning for
background emissions data. Overburden excavation started in the
afternoon.
June 8/F Overburden excavation continued until mud layer was encountered at
the 8-foot level. After moving mud bin into position, mud excavation
started. Excavation halted in the afternoon because of leakage at the
foam-mixing trailer and plugging of foam nozzles.
June 9/Sa Repairs made to foam system. Mud excavation completed. Tar
excavation started at the 11-foot level after tar bin was moved into
position. Tar excavation halted when top of char layer was reached
at the 15-foot level.
June 10/Su Onsite activity avoided on Sundays.
37
-------
Figure 17. Level B personal protection equipment.
Figure 18. Level A personal protection equipment.
38
-------
Date Excavation activity
June 11/M In attempts to improve foaming effectiveness, surface of tar bin was
leveled and platform was installed for applying foam to bin. New
layer of foam was applied. No materials were excavated.
June 12/Tu After moving trench shield and bins into position, discovered that
excavated tar and associated mud had seeped into pit during the
previous two days. Excavation was halted due to high emission
levels.
June 13/W Second tar bin was brought into enclosure and moved into position.
Excavated tar and associated mud had seeped into pit overnight.
Began overexcavation of pit walls to make room for trench shield.
Trench shield was lowered into pit, but became hung up on one side.
June 14/Th Overexcavation of pit walls continued until trench shield was moved
into position. Trench shield end plates were positioned in pit.
June 15/F Excavated small amount of tar that had seeped under trench shield
into pit overnight. Began char excavation.
June 16/Sa Changed caustic soda scrubber solution. Replaced coal-derived
activated carbon with coconut-derived activated carbon. No
excavation activity.
June 17/Su Onsite activity avoided on Sundays.
June 18/M No excavation operations. Site safety plan being revised/reviewed.
June 19/T No excavation operations. Site safety plan being revised/reviewed.
June 20/W No excavation operations. Site safety plan being revised/reviewed.
Refueled equipment. Revised decontamination line for Level A.
June 21/Th Overburden excavated on west side of pit to a depth of 5 feet, to
form ramp-down bench. Trackhoe positioned on bench.
June 22/F No excavation operations because of failure of Level A suits to pass
pressure test.
June 23/Sa Level A suits were being repaired. No excavation operations.
June 24/Su Onsite activity avoided on Sundays.
June 25/M Instrument problems and the need to refill foam tank delayed start of
activities inside enclosure. By midday, heat stress concerns related
to Level A suits warranted against further operations for the day.
39
-------
Date
Excavation activity
June 26/T Excavated small amount of tar seepage and overburden that had
fallen into the pit during ramp-down. Char excavation recommenced
and continued until high emission levels caused work to be halted at
the 25-foot level. Excavation operations completed.
All of the excavation operations described above were conducted in Level B
gear until the final two days, June 25 and 26; on these days, operating personnel
worked in Level A suits.
On June 27, the trench shield was pulled from the pit and preparations for tar
treatment operations began on June 28.
The following descriptions of the excavated overburden and waste materials are
taken from the geotechnical observations report in Appendix A and are presented here
to avoid subsequent confusion when the terms are discussed.
° Overburden - Olive brown, sandy/silty clay, dry, stiff to hard, with a thin
lens of uniformly coarse sand at the 5-foot level.
° Mud - Greenish grey silty clay, moist, plastic, firm, with inclusions of tar
chunks.
° Tar - Black, no soil inclusion, deformability varies with temperature and
overburden pressure, mostly stiff when freshly excavated, becomes
sticky and flowing after exposure to warmer temperatures.
° Char - Black, cemented coal-like material, no soil or tar inclusions, very
hard, nonplastic,
3.2,3 Excavation Quantities and Rates
Excavation of overburden and waste was conducted over 9 work days. The
quantities and volumes of materials excavated on these days are summarized in Ta-
ble 9. A total of 137 solid cubic yards of waste (mud, tar, and char) are estimated to
have been removed during the trial excavation together with 101 solid cubic yards of
overburden. (Solid cubic yards refers to the volumes of material in the ground, and
corresponds to the dimensions of the pit spaces from which materials are removed.)
After removal, solid materials generally tend to expand volume, or bulk. (The
bulking factors are the estimated ratios of material volumes before excavation to
material volumes after excavation. Multiplying solid cubic yards by bulking factors
40
-------
TABLE 9. EXCAVATION QUANTITY ESTIMATES3
Date
(1990)
Excavation
quantity, solid cubic yards
Overburden
Hud
Tar
Char
Totalb
6/7
27
0
0
0
27
6/8
10
20
<2
0
31
6/9
0
<2
20
<2
22
6/12
<2
5
25
<2
32
6/13
<2
5
20
<2
27
6/14
20
7
7
<2
35
6/15
0
0
2
7
9
6/21
42
0
<2
0
43
6/26
0
0
<2
11
12
Total solid cubic yards
101
38
77
22
238
Estimated bulking factor3
1.5
1.5
1.2
1.2
-
Total bulk cubic yards0
152
57
92
26
327
Bulk density, lb/ft3d
120
84
33
74
-
Excavation tonnagee
245
65
41
26
377
3 Based on information in Appendix A.
L
For totals, 1 cubic yard assumed for <2-cubic-yard estimates.
c Solid cubic yards x bulking factor.
d Based on field measurements (see Appendix F).
e Bulk cubic yards x 27 ft3/yd3 x bulk density/2000 lb/ton.
41
-------
yields bulk cubic yards.) Waste materials thus amounted to 175 bulk cubic yards after
excavation.
Bulk cubic yards can be converted to a weight basis using the bulk density
measurements obtained from the field and appropriate conversion factors. Applying
these factors results in estimates of 132 tons of waste excavated and 245 tons of over-
burden.
Based on the operating time recorded in the operating log summaries, and the
excavation volume estimates in the geotechnical report (Appendix A), Table 10 was
prepared to summarize calculations of excavation rates. For each day on which mate-
rial was excavated, the elapsed time for excavation has been calculated by using the
beginning and ending times for excavation operations. In addition, the lengths of the
effective workdays inside the enclosure have been calculated based on the initial entry
and final exit times. The effective workday lengths include nonoperating periods such
as lunch breaks, work breaks, and equipment-related stoppages.
A "maximum" excavation rate has been calculated as excavation volume divided
by the elapsed excavation time. This is referred to as a maximum rate because it
does not include time for normal work stoppages (e.g., lunch breaks). An "average"
excavation rate has also been calculated as the excavation volume divided by the ef-
fective workday length inside the enclosure. This rate is always lower than the maxi-
mum excavation rate because it includes nonproductive time for work breaks and
other stoppages. Also, since this was a trial, excavation breaks were required for
observation and measurement.
Considering first the maximum excavation rates, the data in Table 10 show that
maximum rates for overburden removal ranged from 12 to 81 yd3/hour; the higher and
more representative rate associated with the ramp-down operations on June 21. The
two rates calculated for mud excavation were 15 and 66 yd3/hour. Of these, the sec-
ond is more representative because the first was calculated over a short time period (4
minutes). The most representative days for tar excavation rates were June 9, 12, and
13 when at least 20 yd were excavated on each day and tar was the major com-
ponent of the material excavated. The average maximum tar excavation rate for these
three days was 56 yd3/hour. The two maximum rates calculated for char excavation
were 4 and 9 yd3/hour. The first rate was lower than the second because it was the
first time that char had been encountered and because more time was taken for visual
observations and for flux chamber measurements on char samples. As a result, the
second or higher rate is considered more representative of trial excavation operations.
In calculating average excavation rates it was necessary, in most cases, to
combine the excavated volumes of two or more types of materials. The average
excavation rate for overburden on June 7 was 7.6 yd3/hour. The average excavation
rates for the remaining 8 days, which included excavation of char, tar, mud, and over
42
-------
TABLE 10. MAXIMUM AND AVERAGE EXCAVATION RATES
Effective work day inside
Excavation time enclosure Excavation rate, ydVh
Period
Date
Elasped
Material
Initial
Final
Day
Excavated volune,
f
No.
(1990)
Begin
End
time, h
type*
entry
exit
length, h
solid yd'
Maximum**
Average
1
June
7
1327
1418
0.85
0
1045
1418
3.55
27
32
7.6
2a
June
8
0800
0952
0.87
0
0800
1530
7.50
10
12
4.in
2b
June
8
1021
1040
0.32
H/T
21
66
4.3*
3a
June
9
1033
1037
0.07
M
1033
1543
5.17
1
15
3b
June
9
1345
1429
0.73
T/C
21
29
4.0*
4
June
12
1408
1449
0.68
T/M/O/C
0725
1520
7.92
32
47
5a
June
13
1606
1623
0.28
T/N/C
0652
1925
12.55
26
92
2.2n
5b
June
13
1623
1651
0.47
0
1
2
4.4^
6
June
14
1015
1115
1.00
O/M/T/C
0920
1720
8.00
35
35
7
June
15
1035
1115
0.67
T
0900
1300
4.00
2
3
2.3++
1115
1300
1.75
C
7
4
8
June
21
1000
1032
0.53
0/T
0831
1423
5.87
43
81
7.3
9a
June
26
0647
0710
0.38
T/H
0628
1110
4.70
1§
3§
2.6n
9b
June
26
0710
0744
0.57
C
11*
9
9c
June
26
0920
1000
0.67
C
0 * Overburden, M « Hud, T * Tar, C« Char.
**
Excavation volume/excavation elapsed time.
^ Excavation volume/effective work day length.
| |
Average rate for all materials excavated on the subject day.
$
Rates apply to Periods 9b and 9c coofcined.
-------
burden, fell within a narrow range of 2.2 to 4.3 yd3/hour. The most representative
average excavation rates for mud, tar, and char were 4.1, 4.3, and 2.6 yd3/hour,
respectively, based on those days when a majority of each waste type was excavated.
Excavation rates were also estimated based on the use of a tractor-mounted
backhoe with a 1 yd3 bucket under routine excavation conditions for these types of
waste materials. These estimates, as contained in Appendix E, indicated that average
excavation rates of approximately 49, 32, and 25 yd3/h could be obtained for over-
burden and mud, tar, and char, respectively.
3.2.4 Use of Trench Shield
As discussed in Section 3.2.1, the trench shield was employed during the trial
excavation primarily as a means of minimizing tar seepage into the pit. In the opinion
of the geotechnical engineer observing the excavation, the walls of the pit at the over-
burden and mud levels were stable enough that the shield was not necessary to
prevent sloughing or collapse of the walls. However, after the excavation depth had
reached the char layer, the trench shield became necessary to prevent the pit from
refilling with tar fat least to the tar level). Before the trench shield was positioned, an
estimated 25 ydr of seeped tar had to be excavated on June 12 and 20 yd3 of seeped
tar had to be excavated early on June 13. After the trench shield frame (two sides)
was lowered partially into position on June 13, the flow was reduced to about 7
yd3/night, which was excavated on June 14. After the shield was lowered into the final
position and the end plates were installed on June 14, tar seepage was reduced to
minor amounts (less than 2 yd3/night).
A smaller trench shield was available on site to use for shoring of the pit walls at
the char level. However, because the char proved to be very hard and the walls quite
stable, the smaller trench shield was not required.
3.2.5 Waste Sample Collection/Shipping Procedures
Samples of char, raw tar, treated tar, and mud were collected in 8-ounce and 4-
ounce glass containers from double-lined, 55-gallon sample drums. The sample
drums had been filled inside the enclosure with the front-end loader and bobcat and
then moved outside the enclosure into a staging area prior to transportation to the on-
site drum storage area. Since the char, raw tar, and mud in the sample drums had
already been mixed during excavation, storage, and movement operations prior to
drumming, grab samples from single drums were collected for these materials.
Samples of treated tar (discussed in the next subsection) we re collected from the
drum containing the product material from Run 10 of the tar treatability tests.
The samples were collected in the middle of the afternoon of July 16, 1990.
Within one hour after collection, the glass containers were placed in insulated shipping
44
-------
boxes and packed with ice. A sample chain-of-custody form was filled out for the
samples and sent with them. They were picked up by an express delivery van
between 5:00 and 5:30 P.M. and shipped via overnight delivery to PEI Associates, Inc.,
in Cincinnati, Ohio.
3.3 Waste Treatment Tests
Treatment tests were conducted on tar wastes with the pug mill and feed addi-
tives discussed in Subsection 2.3. No treatment tests were conducted on the char
and mud wastes with the crusher because most of these waste components were less
than 2 inches in size after excavation, and there was no need for further crushing.
This section discusses the test conditions and procedures used for the tar treatment
tests.
3.3.1 Tar Treatment Test Conditions
The tar treatment tests were designed to investigate a range of mixing ratios for
tar, portland cement (i.e., cement), fly ash (i.e., ash), and water. In addition, it was
decided to operate the pug mill at both a nominal throughput rate and a maximum
throughput rate. For all tests, the desired end product was a free-flowing material with
most, if not all, particles less than 2 inches in size.
A total of 10 treatment tests were conducted within the enclosure over the 4-day
period July 10 through July 13. Table 11 summarizes the volumes and weights of tar,
cement, water, and ash fed into the pug mill during these 10 tests. The material
weights shown are based on the known volumes of the backhoe bucket and water
pails and the results of the field bulk density measurements, as reported in Appendix
F. The backhoe was equipped with a 1 /3-yd3 bucket. The 5-gallon water pails were
approximately 90 percent full when their contents were added to the mill, for an
effective volume of 4.5 gallons/pail.
The first nine runs in Table 11 were conducted at the low idle paddle rotation
speed, whereas the tenth run was conducted at maximum paddle rotation speed. Run
1 was a "dry run," conducted without water, to investigate the operability of the pug
mill while processing tar without the complications that water addition might cause.
Runs 2, 3, 4, and 8 were tests in which only cement and water were added to the tar
at low paddle idle speed. In Run 3, the water/cement ratio was increased over that in
Run 2. In Run 4, the tar/cement ratio was increased over that in Run 2. At the end of
Run 4, the paddle speed was increased to about 3/4 maximum speed to observe the
effects on treated tar. A tar/cement ratio between the Run 2 and Run 4 levels was
selected for Run 8. Run 10 was a repeat of Run 2, except that the paddle was set to
the maximum level.
45
-------
TABLE 11. TAR TREATMENT TEST FEED QUANTITIES
Feed quantities to pug mill, lb
Run
No.
Date
(1990)
Tar
Cement
Water
Fly ash
Total
1
7/10
294
1390
0
0
1684
2
7/10
294
2085
351
0
2730
3
7/11
294
2085
585
0
2964
4
7/11
588
1390
273
0
2251
5
7/11
294
1390
312
766
2762
6
7/11
294
695
312
1532
2833
7
7/12
588
695
156
766
2205
8
7/12
588
2085
312
0
2985
9
7/12
588
0
234
3064
3886
10
7/12
294
2085
312
0
2691
Fly ash was added to the feed mix in Runs 5, 6, 7, and 9. The ash/(cement
+ash) ratio was doubled between Runs 5 and 6. The tar/(cement+ash) ratio was in-
creased in Run 7 over the levels of the two preceding runs. Finally, in Run 8, tar was
combined with fly ash and water alone (i.e., no cement).
3.3.2 Tar Treatment Testing Procedures
The operating procedures used to conduct the pug mill tests were essentially
the same from one test to the next. The first step was to use the backhoe to load
thefeed materials into the pug mill feed hopper. Cement and ash were removed from
their respective roll-off storage bins by the Bobcat and placed on the ground in stag-
ing areas. From there, the cement and ash were picked up by the backhoe bucket
and loaded into the pug mill feed hopper. Tar was loaded by the backhoe, either
directly out of the tar bins or from a staging area where it was separated from
associated mud.
The feed materials were layered into the feed hopper. A bucketful of cement or
ash was loaded first, followed by tar, and then by any remaining cement or ash. With
this approach, there was always some cement or ash in the mill before the tar reached
the mill. Water was manually added to the head of the mill in 5-gallon pails, which
were filled from a hose that ran from outside the enclosure to the pug mill inside.
Water addition did not begin until solid materials from the feed hopper had reached
the mill, and it continued at regular intervals during solids feeding.
46
-------
After the feed hopper was loaded with solids in the desired ratios and the water
pails had been filled, the pug mill diesel engine was started up. After engine operation
steadied (generally within a minute), the control rod on the mill was engaged. This
action simultaneously started the turning of the feed belt below the hopper and the
rotation of the paddies in the mill. Feed belt movement transported materials from the
feed hopper to the head of the mill. Material was mixed and pushed through the mill
by the rotating action of the paddles. After passing over a small dam at the tail end of
the mill, treated material rolled through the open product hopper and onto the ground
below.
If material became plugged in the feed hopper, the product hopper, or the mill,
the paddles and feeder belt could be stopped by using the manual control rod. After
the material flow out of the mill was stopped or reduced to a trickle, the paddles and
feeder belt were idled and immediately thereafter the diesel engine was stopped. This
completed the test run. The only remaining operations were to move the treated ma-
terial from under the product hopper into either a sample drum or the excavation pit;
both operations were accomplished with the Bobcat.
As previously indicated, the pug mill was operated in a batch mode during the
tar treatment tests to provide a high level of control over feeding operations. In addi-
tion, the amount of tar, cement, and ash available for processing and the time availa-
ble for testing dictated a batch operations approach. Had there been a desire to
process more tar or to continue a given set of conditions over a longer time period,
the pug mill could have been operated in a continuous mode as well.
The starting and stopping times for the tar treatment tests are summarized in
Table 12, along with the number and reasons for feeder belt/paddle stoppages during
the testing periods. In Subsection 4.5, these start/stop times and the feed weights
from Table 11 are translated into material throughput rates.
3.4 Foam Application and Efficiency Tests
This section describes the foam application and control efficiency tests.
3.4.1 Foam Application
As described earlier in Subsection 2.2, temporary and stabilized foams supplied
by 3M Corporation were mixed with water in a self-contained portable mixing/pumping
system (i.e., foam generator), which was powered by a small gasoline engine. The
foam reagents [either temporary foam (FX 9162) or temporary foam with stabilizer (FX
9161)] were mixed with water and pumped through a hose and nozzle for manual ap-
plication to the waste as required. At least two people were required to operate this
system-one at the foam generator and one at the nozzle inside the enclosure. During
most periods of char excavation, two men were positioned inside the enclosure with
47
-------
TABLE 12. TAR TREATMENT TEST OPERATIONAL DATA
Run
No.
Date
(1990)
Start
time
Stop
time
Elapsed
time,
min
Stoppages
Cause
1
7/10
0817
0832
15
1
Feed hopper blockage
2
7/10
1350
1407
17
0
N/Aa
3
7/11
0738
0755
17
3
1 - Hopper doors closed
2 - Clear material in
product hopper
4
7/11
0821
0830
9
2
Clear material in mill
and product hopper
5
7/11
1056
1108
12
3
Clear material in mill
6
7/11
1328
1337
9
1
Clear material in mill
7
7/12
0847
0852
5
0
N/A
8
7/12
0911
0917
6
0
N/A
9
7/12
1104
1118
14
1
Feed hopper blockage
10
7/13
0659
0703
4
0
N/A
a N/A = Not applicable.
foam nozzles--one in the excavation pit area and one in the stockpile area. In the pit
area, temporary foam was applied to exposed waste surfaces in the pit and to each
bucketful of waste after it was removed from the pit and placed in the staging area
next to the pit. This material was picked up by the front-end loader and carried to a
waste stockpile or storage bin, where exposed waste surfaces were again covered
with temporary foam from the second nozzle.
After an application of stabilized foam, the hose was flushed with water to pre-
vent plugging. Temporary foam was also flushed from hoses prior to extended work
breaks, such as for lunch or during equipment malfunctions. Normally these flushes,
which lasted for 30 to 60 seconds, were directed into 55-gallon drums or a water stor-
age tank. If a drum was not immediately available, however, the flush water was
directed onto the waste pit/pile or ground nearby. In addition, when foam use was
started, water alone was sprayed from the nozzles during the first 30 to 60 seconds;
this water was also directed onto the waste or ground nearby . As a result of this
water flush at the beginning or end of the foaming operation and the breakdown of
foam applied to char, a considerable amount of water collected on the ground inside
the enclosure. This had the following adverse effects: it made footing difficult for
workers, it made movement of loaded storage bins impossible, and it made movement
48
-------
of wheel-mounted equipment difficult to control. As a result, the wheel-mounted front-
end loader had to be replaced with a smaller, tractor-mounted Bobcat.
This system operated satisfactorily after making initial practice runs and adjust-
ments to the foam generation. The temporary foam was usually applied to waste
surfaces in the pit, on the staging pile, and on waste stockpiles/bins, as previously
discussed. During some periods of operation, however, misunderstandings between
personnel inside the enclosure, due to the difficulty of communicating in Level A and
Level B protective clothing, resulted in waste being excavated without foam being
applied for periods of several minutes up to a half-hour. During the workday, the sta-
bilized or "permanent" foam was applied to waste surfaces only if emission levels ex-
ceeded safety plan action levels. At the end of each workday, the stabilized foam was
applied to all exposed waste surfaces. Beginning June 21, the stabilizer reagent was
increased from 6 to 10.5 percent in the permanent foam in an effort to form a more
impermeable layer of foam.
3.4.2 Foam Efficiency Tests
In an attempt to quantify the foam's ability to reduce emissions, testing was
conducted over 3 days on the following three types of waste materials: tar (June 13,
1990), mud (June 14, 1990), and char (June 15, 1990). The tar and mud were ex-
cavated prior to the testing, placed in steel bins, and covered with stabilized foam.
The char was excavated on the day of testing. Flux measurements were performed
by a flux chamber method per the EPA guidance document (U.S. EPA 1986). In total,
11 flux chamber measurements were performed: blank measurement, uncontrolled
tar, duplicate uncontrolled tar, temporary foam on tar, stabilized foam on tar, uncon-
trolled muds, temporary foam on muds, stabilized foam on muds, uncontrolled char,
temporary foam on char, and stabilized foam on char. Grab samples were collected
for each measurement and submitted for offsite analyses. Evacuated stainless steel
canisters were used to sample for the speciation of volatile hydrocarbons, and Tedlar
bags were used to sample for S02 and reduced sulfur compounds.
The results of these tests did not yield consistent results and the analytical data
did not meet the Agency's quality controls. This information is contained in the raw
data file maintained by EPA Region IX, but is not included in this report.
3.5 Air Sampling and Analysis
3.5.1 Enclosure Exhaust Air Stream Monitoring - Total Hydrocarbons (THC) and
Sulfur Dioxide (SOŁ Operating Procedures
The trial excavation at the McColl site was conducted within an enclosure which
exhausted air emissions through a filter, wet scrubber, and carbon bed adsorber. The
filter controlled large particulate, the scrubber controlled particulate/S02 emissions,
49
-------
and the carbon bed controlled hydrocarbon emissions. Continuous emission monitors
(CEM) for S02 and THC were used to monitor air emissions at the inlet and outlet of
the air emission control system (filter, scrubber, and carbon bed). Sampling was con-
ducted continuously from the start of excavation (June 7} through the completion of
the project (July 18). Five hours of each day was used for CEM calibration and sys-
tem maintenance; the balance (19 hours) was the goal for daily on-line monitoring.
Daily Site Activities
Each morning there was a pretest meeting to discuss the day's strategy and
activities. Enclosure S02 and THC levels were reported at this time. Target concen-
tration levels for S02 and THC were then set to indicate when excavation should con-
tinue, slow down, or stop. During any enclosure work, communication between the
trailer and CEM operator was continuous.
Morning calibrations (4-point) were conducted immediately after the meeting.
The first enclosure work session would begin immediately after calibrations. Strip
charts were marked according to type of instrument, concentration range, date, sam-
ple location (inlet/outlet), chart speed, and time (24-h). Log books recorded all strip
chart calibration data (chart response versus gas concentrations), dates, time, and en-
closure events. Calculations of linear regression equations, calibration error, and cor-
relation coefficients were also entered in the notebook. Also included were span and
zero checks, comparisons of computer and strip chart data, and changes in instru-
ment ranges. After the morning test session, the data from the data logger cartridge
was copied to the computer hard drive. The transfer was checked and then the car-
tridge was erased and put back on-line. The data were then transferred to a floppy
disk. Records of data were recorded on the strip charts, hard drive, and floppy disk.
After completion of daily enclosure activities, the CEM's were span and zero
checked. Sample conditioning systems were cleaned, calculations completed, and the
previous days computer hourly and calibration data were printed out. These data
along with the log book were placed in the central file.
Inlet Sample Location (Site 1)
The inlet sample location, Figure 19, is located outside the site enclosure, just
prior to the scrubber. This duct contained all exhaust gases from the excavation
process. Sampling ports were installed to meet EPA Method 1 sample location re-
quirements. CEM sample probes were located at the inlet sample site. The initial
ranges were 0 to 50 ppm S02 and 0 to 200 ppm THC. These ranges were changed
at times when emissions went offscale. Changes were documented on strip charts
and in the field log books. Initial inlet and outlet S02/THC concentration ranges were
estimated based on readings taken around soil borings during a site survey.
50
-------
SAMPLE PORTS
FILTER
TO SCRUBBER
b-
~T~
12 in. I.D.
1
ENCLOSURE
(-*30 in.
-42 in.
CROSS SECTION
VERTICAL PORT
•
•
L12
•
•
•
•
•
1 j
HORIZONTAL PORT
TRAVERSE
FRACTION OF
POINT NO.
STACK I D.
1
2.1
2
6.7
3
11.8
4
17.7
5
250
6
35.6
7
64.4
8
75.0
9
82.3
10
88.2
11
93.3
12
97.9
Figure 19. McColl inlet site.
-------
Outlet Sample Location (Site 2)
The outlet sample location, Figure 20, is located after the scrubber and carbon
bed control system. This duct contained all treated gases from the excavation
process. Sampling ports were installed to meet EPA Method 1 sample location re-
quirements. The S02 and THC CEM sample probes were located at the outlet sample
site. The initial ranges were 0 to 50 ppm S02 and 0 to 100 ppm THC. The S02 moni-
tor range was never changed. The THC range was changed periodically and docu-
mented as mentioned previously.
CEM Trailer
The CEM trailer was located within 60 feet of both the inlet and outlet sample
locations. The trailer contained the CEM systems, sample pumps, strip chart record-
ers, and ODESSA data system. The trailer was climate controlled for the proper oper-
ation of the instruments.
Sampling Strategy and Sampling Procedures
The sampling procedures are summarized in Table 13.
TABLE 13. SAMPLING PROCEDURES, EQUIPMENT, AND METHODS
Sample
stream
Access
Analytical
parameter
Sampling equipment
Procedure/per
run frequency
Inlet/outlet
Duct
S02
EPA Hethod 6C
- Chemiluminescent analyzer
- Strip chart recorder
- EPA Protocol 1 calibration
gases
Continuous extraction
and analysis. Daily
operation 19 h. Four
point calibrations
every 24 hours.
Single span and zero
checks during daily
excavations.
Inlet/outlet
Duct
Total orgamcs
EPA Hethod 25A
- Flame ionization analyzer
- Master gas certified gases
- Strip chart recorder
Same as above
The field testing consisted of continuously monitoring S02 and THC emissions
at the inlet and outlet sites for 19 hours per day during the entire field program. Ap-
proximately 5 hours per day was required for calibration of instruments and system
maintenance (filter changes, cleaning, and changing sample pumps). Included in the
field testing program was background sampling which involved sampling ambient air
prior to excavation procedures. This ensured system operation and supplied back-
ground information on S02 and THC levels.
52
-------
PORTS
FLOW
CROSS SECTION
FROM CONTROL SYSTEM
TRAVERSE
FRACTION OF
POINT NO.
STACK I D.
1
2.1
2
6.7
3
11.8
4
17.7
5
25.0
6
35.6
7
644
8
75.0
9
82.3
10
88.2
11
93.3
12
97.9
Figure 20. McColl outlet site.
-------
Inlet-Outlet S02 Sampling System
Sampling and analysis for S02 emissions was accomplished using a continuous
extractive pulsed fluorescent analyzer. The sampling apparatus is shown in
Figure 21 and meets all design specifications of Method 6C. It consisted of:
Probe - 0.375-in. stainless steel. Probe was hooked directly to 0.25-in.-o.d.
Teflon sample line.
Sample Line - 60-ft, 0.25-in.-o.d. Teflon line was used to transport sample gases
and calibration gases.
Sampling Conditioning Apparatus - A glass wool filter was used in the stack to
eliminate particulate. Two Teflon impingers immersed in an ice bath were used
to remove moisture. Moisture removal is essential to the proper operation of a
S02 monitor.
Sample Pump - Leak-free vacuum pumps provided a constant steady flow of
sample gas through the analyzer.
Transfer Sampling Pump - A leak-free Teflon pump was used to transport gas
60 ft from the stack to the manifold system.
Rate Meter - A rotameter was used to measure air flow through the analyzer.
Manifold - A Teflon manifold was used to deliver sample gases to the monitor.
Bubble Meter - A bubble meter was used to check flow rates leaving the mani-
fold and to ensure sample and calibration gas flows.
Calibration Gases - S02 standards in air will be used to calibrate the instru-
ments. Ambient grade zero air is used to zero the monitors. Three EPA Proto-
col 1 gases were used in the sampling range. The following are gases used for
this program:
Inlet, ppm
Outlet, ppm
Zero
Zero
1.9
1.9
24.6
24.6
45.3
45.3
400
54
-------
GLASS WOOL
FILTER
INLET
DUCT
3/8 In. SS
PROBE
3/0 In. SS
PROBE
1/4 In.)
TEaON LINE
5011
TRAILER
INLET
SO»
ZERO
1.9 ppm
24.6 ppm
45.3 ppm
400.0 ppm
CALIBRATION
LINE
1/4 In. TEFLON
OUTLET
SO*
ZERO
1.9 ppm
24.6 ppm
45.3 ppm
GLASS WOOL
FILTER
:;
OUTLET
DUCT
1/4 In.
TEFLON LINE
5011
TEaON
PUMP
TEaON
PUMP
EXHAUST
EXHAUST
SO* MONITOR
SO? MONITOR
TEFLON
(MPWGtflS
ODESSA
DSM 3260
CHART RECORDER
CHART RECORDER
NOTE: EXHAUST GASES WILL BE VENTED
BACK INTO THE EXCAVATION AREA
Figure 21, Outlet and inlet S02 sample systems
-------
Source monitors were chosen because they have a variety of ranges, and the
minimum detection limit can be as low as 1 ppm. Hie additional 400 ppm cali-
bration gas made it possible to quantitate concentrations above 50 ppm. High-
er concentrations were not expected and additional calibration gases were not
available for this study.
Data Recorder: A Yokogawa strip chart type recorder was used to continuous-
ly record data. This will provide a permanent record of data. An ODESSA DSM
3260 data logging system was used to record all data including calibrations.
Transfer of data was done on a central processing computer. This system was
used for all data reductions.
S02 Monitor
A Thermo Electron pulsed fluorescent analyzer was used to measure S02 con-
centrations at the inlet and outlet test sites. The fluorescent analyzer uses pulsating
ultraviolet light focused through a filter into the fluorescent chamber.
The UV light excites the S02 molecules and they give off a characteristic decay
radiation. A second filter allows only this radiation to fall on a photomultiplier tube.
The light energy is then electronically processed and displayed in a reading which is
directly proportional to the S02 concentration.
The instrument has detection ranges of 0 to 50, 0 to 100, 0 to 500, 0 to 1000,
and 0 to 5000 ppm. The Thermo Electron S02 analyzers meet EPA Method 6C per-
formance guidelines.
Daily S02 Instrument Operation
Prior to the start of daily excavation, the S02 system was calibrated with three
calibration gases and zero air. The calibration was checked for linearity; an accepta-
ble correlation coefficient was >0.999. Filters and condensers were checked/changed
prior to sampling. Strip chart recorders were marked listing sample locations, range,
calibration gases, chart speed, date, and time of sampling. The ODESSA data acquisi-
tion system was used to record data every minute and calculate hourly averages.
When range changes were made, a factor was used to reduce computer data based
on the range (range change from 50 to 500 used a 10 multiplier). During daily excava-
tions, the system was span and zero checked and filters and condensers changed.
Emissions were also monitored during nonexcavation periods.
56
-------
EPA Method 6* was used to perform audits on the Method 6C sampling sys-
tems. One Method 6 sample was collected at both the inlet and outlet test sites to
verify monitor data. These data were collected June 13, 20, and 29, and are sum-
marized in Section 5.1.
Method 6 Sampling Apparatus
The S02 sampling train used in these tests meets design specifications estab-
lished by the Federal EPA and was assembled by PEI personnel. The Method 6 sam-
pling was conducted to compare with the CEM Method 6C results. It consists of:
Probe - 0.375-in. stainless steel probe. A plug of glass wool is placed in the
end of the probe to remove particulate matter.
Impingers - Four impingers connected in series with glass ball joints. The first
and third impingers are of the Greenburg-Smith design. The second and fourth
impingers are modified by replacing the tip with a 1/2-in.-i.d. glass tube
extending to within 1 /2 in. from the bottom of the flask. Glass wool is placed in
the first U-joint to prevent sulfuric acid mist carryover.
Metering System - Vacuum gauge, leak-free pump, thermometers capable of
measuring temperature to within 5 • F, dry gas meter with 2 percent accuracy,
and related equipment to maintain a constant sampling rate and to determine
total sample volume.
Barometer - Aneroid type to measure atmospheric pressure to ± 0.1 in.Hg.
Sampling Reagents
Water - Deionized, distilled to conform to ASTM specifications D1193-74,
Type 3.
Isopropanol. 80%
Hydrogen Peroxide. 3%
Analytical Reagents
Water - Deionized, distilled
Isopropanol. 100%
* 40 CFR 60, Appendix A.
57
-------
Thorin Indicator
Barium Perchlorate Solution. 0.01 N
Sulfuric Acid Standard. 0.0100 N
Ammonium Sulfate Standard. 0.0100 N
Sampling Procedure
After the sampling site and the required traverse points were selected, the stack
pressure, temperature, moisture, and range of velocity head were measured according
to procedures described in Methods 1 through 4.*
One sampling point was chosen and the train was assembled as follows: 30 ml
of 80 percent isopropanol in the first impinger, 30 ml of 3 percent hydrogen peroxide
in both the second and third impingers, and 30 g of silica gel in the fourth impinger. A
portion of each reagent was retained for use as a blank. The train was assembled as
shown in Figure 22. The sampling train was leak-checked at the sampling site before
and after each test by plugging the inlet to the first impinger and pulling a 10-in.Hg
vacuum. All leaks were corrected before sampling began. The probe heater setting
was adjusted during sampling to prevent any visible condensation. Crushed ice was
placed around the impingers and more ice was added during the test to keep the
temperature of the gases leaving the last impinger at 68 • F or less.
At the completion of each test, the train was removed from the stack and
purged for 15 minutes.
Sample Recovery
The contents of the first impinger were measured and discarded. The contents
of the second and third impingers were measured and placed in a leak-free polyethyl-
ene container. The second and third impingers and connecting glassware were then
rinsed with distilled, deionized water and this rinse was added to the same polyethyl-
ene container. The container was sealed and identified, and the liquid level was
marked.
Analytical Procedures
The volume of each sample was recorded and diluted to 100 ml with deionized
distilled water. An appropriate aliquot of this solution was pipetted into a 250-ml
40 CFR 60, Appendix A.
58
-------
STACK WALL
THERMOMETER
GLASS
WOOL
SILICA GEL
DRYING TUBE
STEEL PROBE
(END PACKED WITH
GLASS WOOL)
i SILICA
GEL
IMPINGER CONTENTS
30 mL 00% ISOPROPANOL
30 mL 3% H?0*
30 mL 3% HjO»
200 g SILICA GEL
VACUUM
LINE
IMPINGERS
ICE WATER BATH
THERMOMETERS
BY-PASS
VALVE
VACUUM GAUGE
MAIN VALVE
DRY TEST
METER
AIR-TIGHT PUMP
Figure 22. S02 sampling train assembly.
-------
Erlenmeyer flask. Deionized, distilled water was added to bring the volume to 20 ml.
Eighty ml of 100 percent isopropanol and two to four drops of thorin indicator are
added. The solution was titrated to a pink end point using 0.01 N barium perchlorate.
A blank was titrated in the same manner as the samples. An EPA Method 6 audit
solution was analyzed as a check on the analysis procedure.
Inlet-Outlet THC Emissions
Sampling and analysis for THC emissions was accomplished using a flame
ionization analyzer (FIA). The sampling apparatus is shown in Figure 23 and meets all
design specifications of Method 25A. It consists of:
Particulate Filter - A short piece of 1 /2-in. diameter stainless steel pipe packed
with glass wool and attached to the end of the sample probe.
Sample Probe -1 /4-in. stainless steel tubing inserted into the gas stream being
sampled.
Sample Line -1 /4-in. o.d. heated Teflon line with controller to maintain a sample
temperature of 250 *F. Heated sample line is required to prevent condensation
of hydrocarbon emissions.
Calibration Gases - Methane standards in air and zero air (less than 0.1 ppm
THC) are used to calibrate the monitor. Master Gas Certified gases were used
for this program. EPA Protocol 1 gases are not available in methane standards.
Gases are certified to be ±2 percent of label values.
The following gases were chosen for this program:
Inlet, ppm
Outlet, ppm
Zero
Zero
49.9
25.1
91.5
49.9
180.0
91.5
900
These monitors and gases were chosen to allow for measurements on a variety
of ranges. If instruments went offscale on the original scale, a 900-ppm calibra-
tion gas was available to calibrate on higher ranges. Higher concentrations
were not expected and additional calibration gases were not available for this
study.
60
-------
3-WAY HEATABLE
r VALVE
3-WAY HEATABLE
VALVE
GLASS WOOL
FILTER
1/4 In. SS
PROBE
GLASS WOOL
FILTER
OUTLET
DUCT
1/4 ln.SS
PROBE
INLET
DUCT
CALIBRATION
LINE
1/4 In, TEFLON
1/4 In. HEATED (250 F)
TERON LINE
1/4 In. HEATED (250 F)
TEFLON LINE
INLET
OUTLET
TRAILER
EXHAUST
EXHAUST
ODESSA
DSM 3260
THC MONITOR
CHART RECORDER
THC MONITOR
CHART RECORDER
METHANE
ZERO
49.9 ppm
91.5 ppm
180.0 ppm
900.0 ppm
METHANE
ZERO
25.1 ppm
49.9 ppm
91.5 ppm
NOTE: EXHAUST GASES WILL BE VENTED
BACK INTO THE EXCAVATION AREA
Figure 23. Outlet and inlet THC sample systems.
-------
Fuel and Air - A cylinder of 100 percent hydrogen and a cylinder of compressed
air were used to provide fuel and an air supply for the analyzer's flame.
Chart Recorder - Heath strip chart recorder is used to provide a permanent
record of hydrocarbon concentration data. An ODESSA DSM 3260 was used
to record data including calibrations. Transfer of data v/as done on a central
processing computer. This system was used for all data reductions.
THC Monitor
A Ratfisch total hydrocarbon analyzer that works on the principle of flame ion-
ization was used. All critical sample-handling components of the analyzer were con-
tained in a heat-controlled oven. The oven temperature was maintained at 250 °F.
The following analyzer specifications were provided by the manufacturer:
Full-scale sensitivity: Adjustable from 5 ppm methane to 100,000 ppm
(percent methane)
Response time (0 to 95 percent of full scale): Less thai 1.5 seconds
Electronic stability: ± 1 percent of full scale per 24 hours, with ambient
temperature change of less than 10* F.
Precision: 0.1 percent of full scale
Zero drift: ±0.5 percent of full scale/24 hours
Reproducibility: ± 1 percent of full scale for successive identical samples
Output: 0 to 10V, 4 to 20 MA
The magnitude of the analyzer response to carbon atoms depends on the
chemical environment of this atom in its molecule. Typical ratios of monitor response
relative to methane for carbon atoms in various molecular structures are listed in
Table 14.
62
-------
TABLE 14. MONITOR RESPONSE FOR
VARIOUS MOLECULAR STRUCTURES
Response relative
Molecular structure to methane, %
Aliphatic compound
Aromatic compound
Olefinic compound
200
100
95
130
0
30
Acetyl enic compound
Carbonyl radical
Nitrile radical
THC Monitor Setup and Calibration
The monitor setup and checkout procedures outlined below were performed
prior to sampling. The monitor was calibrated by introducing zero and high-level cali-
bration gases to the calibration port of the sampling manifold. The predicted response
for low-level and mid-level calibration gases was calculated, assuming that the monitor
response is linear. The low-level and mid-level gases were then introduced into the
monitor. If actual responses for the gases differed from the predicted responses by
more than 5 percent, the monitoring system was inspected and repaired before sam-
pling begins.
Once the monitor was calibrated, a system integrity check was performed.
Zero air and one of the methane standards were sampled through the sample probes
and lines to make sure that the sampling system was not diluting or contaminating the
samples. A stainless-steel tee with a leg left open to the atmosphere was placed on
the end of the probe during this step so that calibration gases being sent from the
cylinders did not pressurize the sampling system.
Once the sample lines were checked out, a response-time test was performed.
This test consisted of introducing zero gas to the probes and switching to the high-
level calibration gas when the system was stabilized. The response time is the time
from the concentration change until the measurement system response is equivalent
to 95 percent of the response for the high-level calibration gas. The test was per-
formed three times, and the results were averaged.
Daily THC Instrument Operation
Daily calibrations for each range were performed with three calibration stan-
dards (low-level, mid-level, and high-level) and zero air. Each calibration range was
checked by linear regression calculations, which indicate linear responses and are
used to reduce field data.
63
-------
When sampling was completed, a calibration drift check was performed on the
monitor by introducing the zero and span gas to the monitor.
CEM Data Reduction
The ODESSA DSM 3260 data acquisition unit was used to collect, average, and
compute 5-minute and hourly averages of the four OEM's. The system was set up to
summarize all calibration points and reflect the accuracies of each calibration. The
system reduces data, based on full scale and zero input values. Each calibration gas
value was also programmed into the computer. Deviations from calibration gas actual
values versus expected values were then computed for each calibration point. Calibra-
tion data were marked with a C and were not used in the data averages.
The following were programmed readings for each instrument:
S02 inlet - Full scale 48.2 ppm
S02 outlet - Full scale 48.2 ppm
THC inlet - Full scale 209.3 ppm
THC outlet - Full scale 104.0 ppm
When range changes were made, a multiplier was used to adjust the computer data.
A range change on the S02 monitor from 50 to 500 ppm scale would use a multiplier
of 10 to adjust the data. Strip chart calibration responses were used to validate
computer data. Strip chart calibration data was calculated using linear regression
equations. This data was also used to check instrument linearity and drift. One hour
of reduced strip chart data was compared daily to one hour of computer data during
this program.
Sampling Equipment Calibration Procedures
All sampling equipment was calibrated before the test program according to the
procedures contained in the Quality Assurance Handbook for Air Pollution Measure-
ment systems, Volume III EPA-600/4-77-027b, or as specified in the method. In addi-
tion, onsfte calibration checks were made prior to the start of testing to ensure that
equipment was not damaged during packing and transport to the field site. Table 15
summarizes the sampling equipment calibrations that were performed for this project
and Table 16 summarizes the field checks that were performed.
3.5.2 Site Perimeter Air Monitoring
In an effort to determine the ambient impact of the trial excavation, and to moni-
tor the concentrations of pollutants emanating from the site, a network of ambient air
monitors was established at locations selected by EPA and the South Coast Air Quality
Management District. Total hydrocarbons (THC) and sulfur dioxide (S02) were moni-
64
-------
TABLE 15. FIELD EQUIPMENT CALIBRATION SUMMARY*
Allowable
Equipment Calibrated against difference
EPA Method 6 box
Bubble meter
Posttest
Y ±0.05 Y
S02 monitor
EPA protocol gases
±1%
THC Monitor
Master Gas Certified
±2%
Digital temperature
indicator
Millivolt signal
±0.5%
Thermocouple
ASTM-3F thermometer
±1.5"F
Impinger (or condenser)
thermometer
ASTM-3F
±2°F
Dry-gas thermometer
ASTM-3F
±5'F
Balance
Type-S weights
±0.5 g
Barometer
NBS-traceable barometer
±0.1 in.Hg
a As recommended in "Quality Assurance Handbook for Air Pollution
Measurement Systems: Volume III." Stationary Source Specific
Methods. EPA-600/4-77-027b. August 1977.
TABLE 16. FIELD CHECKS OF SAMPLING EQUIPMENT
Equipment
Checked against
Allowable difference
Pitot tube
Inspection
No visible damage
Digital temperature
Millivolt signal
±0.5%
Indicator
Thermocouples
ASTM 2F or 3F
±1.5%
S02 monitor
Method 6
±10%
tored at each of the locations shown in Figure 24. Additionally, air samples were
taken for subsequent analysis for organic and reduced sulfur compounds at these
locations utilizing a Tedlar bag instantaneous sampling technique. This air monitoring
was performed to ensure compliance with the Community Safety/Contingency Re-
source Plan established by the Interagency Committee (IAC). This plan had estab-
lished criteria levels for implementing actions to protect the community from possible
effects of excessive air concentrations, as follows:
65
-------
Rosecrans Avenue
~
n
MET
TOWER
' I EQUIP.
DECON
AREA
Sumps
ft S
ft-1
Excavation
Area
Ol
y.
• Perimeter monitoring sites
Figure 24. McColl site showing perimeter monitor locations.
66
-------
Action levels, ppm and time period
Compounds Shut down Shelter Relocation
S02 0.5 15.0 20.0
(5 min) (2 min) (2 min)
THC 70.0 80.0 145.0
(30 s) (2 min) (2 min)
Thus, if a concentration of 0.5 ppm S02 was reached and sustained for 5 minutes, all
site work would stop immediately and foam would be applied to reduce emissions.
Shelter and relocation criteria were designed for short-term situations when
emission levels could not be immediately reduced by onsite mitigation, and evacuation
or taking shelter would be necessary to prevent population exposures. These values
were based on acute toxicity data in humans for S02 and THC. Levels detected at the
perimeter of the site were assumed to be the same as levels in the ambient air of the
community. However, because emissions would be diluted as they move away from
the site, the actual community levels would be lower.
Sulfur Dioxide
Monitor Labs Model 8850 S02 analyzers, set on a 0- to 10-ppm S02 range,
were deployed at each of the four sampling locations. The monitors were operated on
a 24-hour, continuous basis with associated strip chart recorders. Each monitor was
connected to a visual and audible alarm system to provide an indication of sustained
concentrations of S02 for a 5-minute duration above the 0.50-ppm S02 by volume
alarm level. The contingency plan has established this 0.5-ppm level sustained over
5 minutes as the shutdown level for general public safety.
The S02 instruments were calibrated by challenging each instrument with
known concentrations of S02 at zero and three upscale points using National Bureau
of Standards (NBS) traceable gases. Generally, the three upscale points were estab-
lished at 20, 40, and 90 percent of the full-scale range of the S02 monitoring instru-
ment. Calibrations were performed at the beginning of the study, at the midpoint of
the study, and at the completion of the study. On a daily basis, a zero and a single
point span check at approximately 0.60 ppm S02 were performed and the instrument
was adjusted, as necessary. All instrumentation adjustments were documented in the
project and instrument log books, as well as on the strip charts. The 0.60-ppm con-
centration was chosen so that the alarm system would be activated and tested daily,
as well.
67
-------
Total Hydrocarbons
Foxboro Model 128 Organic Vapor Analyzers were used to measure total hy-
drocarbons at each of the sampling locations. The monitors were operated on the 0-
to 100-ppm scale. The THC instruments were calibrated with known concentrations at
zero and three upscale points (approximately 20, 50, and 75 percent of full scale)
using NBS traceable methane in air.
The monitors were operated on a continuous basis, 24 hours a day. Each
monitor had a strip chart recorder and the monitors were connected to a visual/
audible alarm to provide an indication of sustained concentrations above 70 ppm of
THC for 30 seconds duration. This action level was established in the contingency
plan to shut down excavation activities.
On a daily basis, a zero and single point span check, at approximately 75 ppm
THC, were performed and the instruments adjusted as necessary. All instrumentation
adjustments were documented in the project and instrument log books as well as on
the associated strip chart. The 75-ppm concentration was selected so that the alarm
system would be activated and tested daily as well.
Organic and Reduced Sulfur Compounds
Five-liter Tedlar bags were collected over approximately 5-minute time periods
at each of the four perimeter site locations. The Tedlar bag samples are collected
twice per week for the analysis of organic and reduced-sulfur compounds. The grab
bag sampling train consisted of a Teflon sample probe, a 5-liter Tedlar bag contained
in an airtight 5-gallon container, and a foot-operated bellows-type pump to evacuate
the rigid container. Tedlar bags were leak checked at the laboratory by inflating the
bag with zero nitrogen and observing the bag for deflation over a 24-hour period.
The bags were filled in approximately 5 minutes to capacity, which represents
the total sample volume. Upon completion of sampling, the bag was disconnected,
sealed, and uniquely labeled including date and time of collection and sample location.
The sample technician also recorded the ambient temperature and barometric pres-
sure for each sample collection. A courier delivered the samples to Performance Ana-
lytical, Inc., for analyses within 24 hours of collection.
Meteorological Parameters
A meteorological monitoring system was also established during the course of
this pilot excavation. The meteorological parameters monitored included wind speed,
wind direction, and standard deviation of wind data.
68
-------
A Met One Model 024A wind direction .sensor was installed on a 10-meter
tower. The sensor was operated on a 1- to 540-degree compass range. The sensor
was installed and an initial orientation was performed using four compass points (90-,
180-, 270-, and 360-degree compass) to properly align the sensor. A Met One Model
014A wind speed sensor was installed on the 10-meter tower. The sensor was
operated on a 0- to 100-mile-per-hour (mph) range. The sensor was calibrated in
PEI's audit system verification center (ASVC) prior to shipment to the site. A Met One
Model 076B Temperature Sensor with Radiation Shield was installed on the 10-meter
tower. The sensor had an operating range of +50"C to -50 *C. The sensor was fac-
tory calibrated prior to shipment to the McColl site. This unit, however, did not
operate properly and temperature data were not valid.
All meteorological sensors were connected to a Met One Model 120 Translator
for output in analog form to an Odessa Model DSM 3260/AQM Data Acquisition Sys-
tem (DAS). The DAS recorded hourly averages for wind speed, wind direction and
ambient temperature. The DAS also calculated the standard deviation of the wind di-
rection, with respect to wind speed in degrees compass.
3.5.3 Air Speciation Measurements
Beginning June 12, 1990, grab bag samples were collected at the enclosure
control system inlet and outlet, selected ambient locations (Perimeter Stations 3 and
4), and the surrounding neighborhood (Cul-de-Sac-Tiffany Court). Three samples
were subjected to both a gas chromatography-mass spectroscopy (GC-MS) and gas
chromatography with flame photometric detector (GC-FPD) analyses for select
organics, thiophene, and sulfur compounds. The primary target analyte list of
compounds included: tetrahydrothiophene, benzene, toluene, m.p-xylene, o-xylene,
naphthalene, methyltetrahydrothiophenes, dimethyltetrahydrothiophenes, ethyltetrahy-
drothiophenes, dimethylthioether, diethylthioether, 1-methylthiopropane.
Table 17 summarizes sample collection activities including dates/times and
location, and a brief summary of waste processing.
Sampling Procedures
A grab bag sample system was used to collect samples at each location. The
system consisted of a Teflon sample probe, a 5-liter Tedlar bag contained in a sized
container and a foot-operated bulloros-type pump to fill the bag. Prior to sample
collection, the system (probe and pump connections to container) was visually
inspected to ensure all connections were in place and tightly sealed. This sample
probe and pump connector hose were leak checked periodically by connecting the
probe to the pump hose, plugging the probe end, and evacuating the line with the foot
pump until the pump was flat and no upward movement in the pump was observed.
69
-------
TABLE 17. SUMMARY OF AIR SPECIATION BAG COLLECTION ACTIVITIES
Date (1990)
Designated
and
sample run
Time (24 H)
Sample location
number
Waste processing activity
Comments
6/12 - 1415
Scrubber inlet
SI-1
Tar excavation w/o foam
6/12 - 1418
Scrubber outlet
0-1
Tar excavation w/o foam
6/12 - -
-
8-1 (blank)
6/13 - 1157
Scrubber inlet
SI-2
None
6/13 - 1206
Scrubber outlet
0-2
None
6/13 - 1418
Perimeter (Sta.
4)
P-l
Enclosure sliding door open
6/13 - 1423
Perimeter (Sta.
33
P-2
Enclosure sliding door open
Sample taken at
6/13 - -
-
8-2 (blank)
seepaqe Sta. 3
6/14 - 1128
Scrubber outlet
0-3
Tar material loaded to second
6/14 - 1137
Scrubber inlet
SI -3
bin; temp, foam applied
6/14 - 1203
Perimeter (Sta,
3)
P-3
6/14 - 1211
Nei ghborhood
NH-1
6/14 -
-
B-3 (blank)
6/15 - 1112
Scrubber inlet
SI—4
Seeped tar cleanout; foam
6/15 - 1118
Scrubber outlet
0-4
applied as excavated; char
6/15 -
-
B-4 (blank)
excavation
6/19 - 1025
Scrubber outlet
0-5
None - foaming, moving soil.
6/19 - 1031
Scrubber inlet
SI-5
refueling
6/19 - 1108
Perimeter (Sta.
3)
P-4
6/19 - 1123
Nei ghborhood
NH-2
6/19 -
-
B-5 (blank)
6/26 - 0730
Scrubber outlet
D-6
Tar/char excavation
6/26 - 0738
Scrubber inlet
SI-6
6/26 - 0806
Perimeter (Sta.
3)
P-5
6/26 - 0813
Neighborhood
NH-3
6/26 -
-
B-6 (blank)
6/29 - 1057
Perimeter (Sta.
3)
P-6*
• - RTI audit
6/29 - 1130
Scrubber inlet
SI—7#
Char transfer into pit
samples
6/29 - 1155
Scrubber outlet
0-7
6/29 - 1303
Nei ghborhood
NH-3A*
6/29 -
-
B-7 (blank)*
7/2 - 1011
Scrubber outlet
0-8
Hud transfer from bin to
• - Duplicate
7/2 - 1016
Scrubber outlet
0-8-D1*
stockpile
sample
7/2 - 1028
Scrubber inlet
S1—8
7/2 - 1030
Scrubber inlet
SI-8-D1*
7/2 - 1040
Perimeter (Sta.
3)
P-7
7/2 - 1044
Perimeter (Sta.
3)
P-7-D1*
7/2 - 1055
Neighborhood
NH-4
7/2 - 1100
Neighborhood
NH-4-01*
7/2 -
-
B-8 (blank)
7/11 - 1103
Scrubber outlet
0-9
Tar processing (Run 5)
• - Duplicate
7/11 - 1108
Scrubber outlet
0-9-02*
samples (stabi-
7/11 - 1117
Scrubber inlet
SI—9
lity check)
7/11 - 1122
Scrubber inlet
SI-9-D2*
7/11 -
-
1-9 (blank)*
* - Blank air
bubbled through
H O prior to col-
lection
(continued)
70
-------
TABLE 17 (continued)
Date (1990)
Designated
and
sample run
Time (24 H)
Sample location
number
Waste processing activity Ccximents
7/12 - 0908
Scrubber outlet
0-10
Tar processing (Run 8}
7/12 - 0912
Scrubber inlet
Si-10
7/12 - 0930
Perimeter (Sta, 3)
P-8
7/12 - 0940
Nei ghborhood
NH-5
7/12 -
-
B-10 (blank)
71
-------
No attempt was made to leak check the sample bags to preclude the possibility of
sample contamination. For the scrubber inlet and outlet tests, the sample probe was
placed at the duct centroid and purged with stack gas using a Thomas pump before
connecting the sample bag. Samples were collected over approximately a 5-minute
period or until the bag was full. At each location, the date/time, barometric pressure,
wind direction, and temperature were recorded. Each sample was uniquely identified
(see Table 17) and chain-of-custody forms were filled out for each set of samples
collected. Samples were then couriered to Performance Analytical, Inc. Canoga Park,
California for analysis within a 24 hour period. Blank samples were collected daily by
filling a bag with zero air and submitting for analysis with each sample lot. Additional-
ly, duplicate samples were collected to access field precision and sample stability.
Analytical Procedures
Each sample was subjected to a GC-MS and GC-FPD analysis. The volatile
organic compounds were analyzed using procedures described in EPA Method TO-14.
Analyses were performed using a Finnigan 4500 Quadrupole GC/MS/DS system inter-
faced to a Tekmar 501067 automatic desorber and a Dessage GS 3/2 gas sampler.
Sulfur-bearing organics and thiophene were semi-quantified from the mass spectra
data generated during these analyses. These compounds were quantified by cor-
relating their response to tetrahydrothiophene and assuming a 1:1 response relative to
that compound. A separate "aliquot" from each bag was directly injected into a GC-
FPD to quantify reduced sulfur compounds. The reduced sulfur analysis generally fol-
lowed the GC-FPD criterion established in EPA Reference Method 16. In addition to
the method blank, field blank, and duplicate sample analyses, internal standards and
surrogate spike compounds were added to each sample analyzed by T014 to access
data accuracy. The internal standards and surrogates are those specified in Method
8240 Of EPA SW846.
Analytical procedures were audited by Research Triangle Institute. As dis-
cussed in Section 5, the Agency determined that the resulting data did not meet the
necessary quality control criteria, and these data were not used. This information is
contained in the EPA Region IX raw data file.
3.5.4 Odor Survey
To evaluate the possibility of fugitive emissions entering the neighborhood adja-
cent to the McColl site, the California State Department of Health Services (CDHS)
planned to conduct an odor surveillance study in the McColl community. There are
several highly odoriferous chemicals characteristic of the acidified asphaltic petroleum
sludges buried at McColl. One of these, tetrahydrothiophene (THT) and its alkylated
(methyl-, dimethyl-, and ethyl-) derivatives, can be detected by most human noses at
1 part per billion (ppb). This is below the analytical detection limit for most laboratory
equipment; consequently, people can be better detectors than machines. Thus, by
72
-------
involving community members in reporting odors characteristic of McColl, the agen-
cies could assess possible community exposures to emissions resulting from the exca-
vation of the buried wastes. To verify potential exposures, select members of the
community would also be recruited to collect grab samples (5-liter Tedlar bags) of
ambient or indoor air during odor "episodes." This section of the report describes the
plan that was developed by CDHS and what was accomplished.
To maximize the number of neighborhood air grab samples that would contain
measurable amounts of odorous chemicals, it was deemed necessary to train a panel
of community members to evaluate odor intensity. This was to be accomplished by
utilizing the professional services of Odor Science and Engineering (OSE), Hartford,
Connecticut. In brief, OSE would train people to judge the intensity of an odor by
matching it to one of eight serial dilutions (from 1:16 to 1:2000) of butanol in water.
Ideally, a member of the odor panel would compare the odor he or she detected from
McColl with that from one of the eight bottles and only collect an air sample that would
be characterized as "strong" (low dilution factor). OSE would also equip and train
select members of the odor panel on how to collect an air sample. The equipment
consisted of a sealed 5-gallon bucket containing a Tedlar bag vented to the outside air
with Teflon tubing. By use of a foot pump, one can evacuate the air from the bucket
and cause the bag to inflate, thereby collecting a grab sample. Each designated odor
panel member would initially be given three Tedlar bags and an inflation device.
Selection of odor panel members was to be based on several criteria. In order
to collect the most concentrated air sample, the member should live very near the site.
Initially, this was defined as residing on Fairgreen Drive or Tiffany Place. The panel
member should also be at home during the hours of planned excavation. Also, be-
cause some people have very poor odor thresholds, which would not be desirable, the
panel member must have successfully completed the odor intensity training. In order
to increase the probability of recruiting at least six households that would meet these
criteria, the area solicited was expanded to include homes on Fairford Drive, Crown
Way, and Regency Circle. Ideally, one adult member from nine households would be
trained, with six being selected as primary sample collectors and the other three as
alternates. While initial community interest seemed high, actual volunteers were tew.
Therefore, in addition to letters being sent to each eligible household, 47 families were
also contacted by telephone. Only two households could be recruited that met the
requirements previously outlined. Unfortunately, this would not have provided ade-
quate coverage of the perimeter of the site, so the odor panel training was canceled.
The lack of participation could be attributed to attempting to recruit for a period of time
when many people take summer vacations (June and July) and the fact that many
people work during the day.
In order to still assess the frequency of odor complaints originating from the
site, the CDHS opted to perform a "passive" surveillance of the McColl neighborhood.
While the initial odor survey would have required active participation on the part of cer-
-------
tain community members, a passive surveillance is designed to allow large numbers of
residents to respond when they choose. Thus, CDHS set up an "800" telephone num-
ber through which the community could telephone in odor complaints any time after
experiencing odorous emissions from the site. The telephone number remained in
service from May 29 through August 15, providing essentially 24-hour coverage
before, during, and after the trial excavation. Actual odor complaints were received
and documented by the Orange County Department of Environmental Health and
relayed to CDHS, EPA, and the South Coast Air Quality Management District
(SCAQMD).
To determine whether emissions were reaching the community, the EPA's SITE
Program implemented a limited community air sampling program. Five-liter air
samples were collected on random days near the perimeter air monitoring stations and
in the community at the cul-de-sac of Tiffany Place. These ambient air grab samples
were analyzed as described in Subsection 5.5.3. The reduced organosulfur
compounds quantified by Performance Analytical, Inc. (PA!) were specifically chosen
because they had previously been identified by GC/MS analysis of a Tedlar bag sam-
ple collected in response to an odor complaint at 10:30 p.m. on April 16,1982, at
1837 Fairgreen Drive (TRC, July 1982). None of the eight perimeter samples or the
five neighborhood air grab samples contained detectable amounts of the reduced
organosulfur target compounds. These data did not, however, meet analytical quality
controls, and were deemed invalid by the Agency. There were no odor complaints
from the community on the days the neighborhood air samples were collected.
Odor complaints were reported on the 2 days prior to the last round of EPA's
neighborhood air sampling (July 12). The reports of odors originated from homeown-
ers immediately adjacent to the site, near the intersection of Tiffany Place and Fair-
green Drive. Examination of meteorological data at the time of the complaints sug-
gests that wind conditions (274 degrees/10 mph and 200 degrees/<3 mph) were
such that emissions from the stack could be transported or diffused in the direction of
these homes. Examination of the Site Activity Logs indicates that stack emission flow
rates from the air scrubber were reduced from 1100 ft^/min on July 9 to 350 ft3/min
on July 11. This reduction in exhaust airflow probably resulted from a buildup on the
packing balls in the scrubber. This problem was remedied on July 11. However, ex-
haust airflow rates were reduced on July 10, the first day of odor complaints. The
effect of reduced airflow rates is to decrease the vertical dispersion of emissions,
allowing for increased horizontal transport (the winds were from 274 degrees at 10
mph the evening of July 10). The essentially calm wind conditions on the evening of
July 11 could have allowed diffusion of these odorous compounds into the neighbor-
hood. However, because of the presence of the large seep at Pit L-1 and a smaller
one near perimeter Monitoring Station 3 (which are nearer the affected homes than the
stack and yet are in line with the stack and the homes), it is difficult to determine the
contribution of the stack emissions to the odors detected in the neighborhood.
Interestingly, there were other times during the trial excavation when meteorological
74
-------
conditions would have allowed the diffusion of odorous emissions from the seeps into
the community, but odor complaints were not registered. However, it should be noted
that odor perception/detection requires that a person be in the right place at the right
time; the absence of an odor complaint may simply reflect the absence of a human
receptor.
The very low number of odor complaints registered by the McColl community
during the course of the trial excavation may have several explanations. Control mea-
sures may have reduced concentrations of organosulfur compounds such that air dis-
persion would allow sufficient dilution to nondetectable levels. Also, daytime winds
during the course of the trial excavation were predominantly from the south-southwest,
which would tend to disperse the stack emissions toward the Los Coyotes oil fields
and away from the residential areas. Summertime temperatures were also in excess
of 100* F, which would tend to keep most residents indoors with their air conditioners
operating. As noted earlier, there may not have been many residents home during the
daytime due to adult members of the household being at work or possibly on summer
vacations. In this regard, it will be difficult to assess whether nighttime or weekend
excavation activities would have affected the incidence of odor complaints, since exca-
vation did not occur at these times. There is also the possibility that the almost contin-
uous presence of rather large seeps at McColl induced a tolerance effect on the per-
ception of odors by nearby residents. Nearby residents may have become acclimated
to low levels of McColl odors such that it may take significant increases above back-
ground for these odors to become objectionable. For example, some people notice
the ever-present odors at McColl more readily after returning from vacations out of the
local area.
Although the CDHS original odor survey was not implemented, the CDHS pas-
sive odor surveillance and the meteorological measurements did provide for limited
odor monitoring during the trial excavation. The results indicate that fugitive emissions
from the site did not result in excessive odor complaints during the 6-week trial exca-
vation.
75
-------
SECTION 4
RESULTS OF FIELD DEMONSTRATION
This section presents arid discusses the results of the activities performed
during the trial excavation. The topics covered in this section are as follows: inlet and
outlet S02 and THC emissions; performance of the wet scrubber and activated carbon
unit; excavation pit emissions; exhaust gas flow rates; perimeter S02 and THC ambient
air levels; air speciation data; foam efficiency evaluation; tar processing evaluation;
equipment performance evaluation; and waste characterization.
Also, at the beginning of this chapter is a discussion of the method used to
reduce the inlet and outlet S02 and THC emission data.
4.1 Enclosure Air Emission Control System Monitoring
4.1.1 Background Ambient Air and Diesel Engine Testing Concentration Data
Prior to the start of excavation, background testing was conducted on 1) ambi-
ent air outside the enclosure, 2) air from an empty enclosure with no activity, and
3) air from the enclosure while diesel engines were running. During the diesel engine
measurements, the air exhaust system was in operation and the samples were taken
at the air control system inlet site. These data give an indication of S02 and THC
levels prior to any excavation activity. The background data are presented in
Table 18.
The ambient air and empty enclosure S02 and THC levels were low as expect-
ed. During the diesel engine background testing, both the S02 and THC concentra-
tions gradually increased. The THC levels started at 5.5 ppm (0645), peaked at
18.2 ppm (1035), decreased to 4.6 ppm (1150), and were in the 4- to 6-ppm range
thereafter. Although the S02 concentrations also increased gradually during the diesel
engine/background testing, some of the increase may have resulted from monitor drift
during these initial tests. The drift was attributed to moisture condensation when the
sample gas entered the monitoring trailer. This drift was specific to the S02 analyzers.
The problem was corrected when discovered.
76
-------
TABLE 18. BACKGROUND AIR DATA
Avg. concentration, ppm
Date
Sample conditions
(1990)
Time (24-h)
SO,
THC
Ambient air
5/5
1525-1555
<1.0"
2.0b
Empty enclosure
6/2
0955-1035
<1.0
3.4
Enclosure with diesel
6/7
0700-0800
<1.0
9.6
engines operated6
0800-0900
2.0
13.3
0900-1000
5.7
16.1
1000-1100
7.74
14.7
1100-1200
<1.0
5.7
1200-1300
1.6
4.2
a Method Detection Limit was 2% of full scale (1 ppm).
L
Method Detection Limit was 2% of full scale (2 ppm).
c Engines started at 0645 and stopped at 1035.
^ S0? readings are believed to be biased high due to drift as a result of
moisture condensation in the system.
4.1.2 Inlet and Outlet Data Reduction Process
The most important data for the evaluation of the air emission control system
are the S02 and THC results before and after the caustic scrubber and activated
carbon air units. The S02 and THC concentrations were measured in the enclosure
ventilation air leaving the enclosure and entering the scrubber-carbon system (inlet),
and in the air discharged from the stack (outlet) into the atmosphere.
The following discussion presents examples of the S02 and THC data from the
enclosure exhaust control system. The example data are from the S02 and THC con-
tinuous emission monitors and represent a short-time interval of approximately 2 hours
on June 9 when the S02 and THC emissions in the enclosure rapidly increased during
the excavation activities. Table 19 represents the 2-hour time interval of 1330 (1:30
PST) to 1540 (3:40 PST) on June 9, 1990. It reflects a continuous record of 5-minirte
average data points that were generated by the computerized automatic data
aquisition system that was calibrated to accept the inputs from the two S02 monitors
and the two THC monitors. Figures 25 and 26 represent graphical plots of the inlet
and outlet S02 and THC data highlighted during the same time interval and con-
structed from the 5-minute data shown in Table 19. Figures 27 and 28 are segments
of the continuous strip chart records that are also highlighted to show the same time
interval on June 9,1990. The strip chart records contain the actual observations and
-------
TABLE 19. EXAMPLE S02 AND THC DATA FOR JUNE 9
S02 (ppm)
Inlet
S02 (ppm)
THC (ppm)
THC (ppm)
Time
Outlet
Inlet
Outlet
*13:30
0.1
0.0
12.5
4.2
13:35
0.1
0.0
12.3
4.0
13:40
0.0
0.0
12.2
4.7
13:45
0.1
0.0
13.3
4.6
13:50
0.2
0.0
16.0
4.5
13:55
1.0
0.0
19.0
4.8
14:00
4.3
0.0
23.9
5.9
14:05
14.8
0.0
29.5
6.2
14:10
27.0
0.0
34.6
7.5
14:15
41.6
0.0
40.0
8.1
14:20
34.4®
37.6*
0.0
46.6
10.9
14:25
0.0
49.9
11.8
14:30
37.4®
0.0
49.7
12.8
14:35
38.1?
0.0
61.1
14.8
14:40
12.4P
0,0
81.7
18.8
14:45
13.6?
0.0
91.6
23.0
14:50
12.8?
0.0
88.0
25.7
14:55
10.2?
0.0
79.8
22.7
15:00
8'7b
0.0
103.4
28.1
15:05
8-2b
0.0
107.8
29.9
15:10
7.7,
0.0
96.3
26.6
15:15
28.5®
0.0
93.0
25.0r
15:20
34.0®
0.0
87.9
6.0^
15:25
31.8®
0.0
84.5_
41.0C
15:30
30.7®
0.0,.
22.n
23.1
15:35
27.4® H
28.1 *
0.0^
79.6d
26.3
15:40
14.0C
95.5
37.2
8 Scale change from 0-50 ppm to 0-100 ppm. Data must be
multiplied by 2 to obtain real value (i.e., 0 14:20
34.4 x 2 = 68.8).
b
Scale change from 0-100 ppm to 0-500 ppm. Data must
be multiplied by 10 to obtain real value. Multiply by
2 for change from 0-50 ppm to 0-100 ppm and then multiply
by 5 for change from 0-100 ppm to 0-500 ppm (i.e., §
14:40 12.4 x 2 x 5 - 124).
c Calibration period. Data reported is calibration data
point which represents the concentration of the calibra-
tion standard for SO- and THC and does not represent
true data points.
d Span run. Data reported is span check quality control
check point and does not represent a true data point.
78
-------
140
14:45
120
100
80
60
40
20
0
o
o
00
o
o
o
o
o
0
01
O
o
o
o
to
o
o
00
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Figure 25. Plot of S02 concentrations vs time for June 9,1990.
-------
Figure 26, Plot of THC concentrations vs time for June 9,1990.
-------
JM MBJ
!\
! \
Figure 27. Late afternoon portion of S02 outlet strip chart for June 9.
-------
wim} m, i i =qp
Figure 28. Late afternoon portion of THC outlet strip chart for June 9.
82
-------
notations by the on-site analyst who was continually maintaining and operating the
S02 and THC monitors during all periods of activity within the enclosure.
Please note on Table 19 at 14:45 the S02 inlet concentration is shown as 13.6
ppm [which is actually 136 ppm due to scale change (Footnote B)]. Figure 25 also
identifies the 136 ppm S02 at 14:45 and Figure 26 (strip chart record) also shows 136
ppm S02 at 14:45. The outlet S02 concentrations being emitted to the atmosphere
were all less than 1 ppm S02.
4.1.3 Sulfur Dioxide Concentrations Before and After the Enclosure Exhaust System
On June 7,1990, sampling began for sulfur dioxide at the inlet and outlet of the
enclosure exhaust control system. The continuous emission monitors (CEMs) ran
continuously until the completion of the program on July 18, 1990. For a chrono-
logical listing of excavation activities during this time period, see Section 3.2.2 of this
report, and refer to the daily operating logs presented in Appendix A.
Table 20 presents the S02 data generated. The table lists the enclosure
exhaust control system inlet and outlet concentrations (ppm) expressed as daily
averages, maximum values, second-highest values, and minimum values. The
maximum and minimum values are 5-minute average values. All of the data were
generated from the DAS generated 5-minute averages.
Also presented in this table are S02 removal efficiencies calculated for each
5-minute data set and averaged for each day. These data will be discussed in sub-
section 4.1.4. Figure 29 is a legend of the data plots for Figures 30 through 65.
Figures 30 through 46 present plots of inlet and outlet S02 concentrations. Additional
plots of S02 concentrations versus time are presented in Appendix B-1. The days
presented were chosen becaused of work activity occuring within the enclousre and
the resulting S02 and THC emissions.
4.1.4 Wet Scrubber Performance
The S02 scrubber used during the trial excavation was designed to achieve an
outlet S02 concentration of 2 ppm on a continuous basis, assuming that the average
inlet S02 concentration would be near 10 ppm and that the maximum inlet S02 con-
centration would be 200 ppm. The data in Table 20 show that the 2-ppm outlet S02
concentration limit was met with few exceptions. One exception was a 50-minute peri-
od on June 13 when the scrubbing liquor pH was inadvertently allowed to drop to 2.9,
well below the specified control range of 10 to 13. During this period, the outlet S02
concentration rose to a 5-minute average maximum of 12 ppm. The achievement of
the outlet S02 design criteria was especially impressive in light of the high inlet S02
concentrations experienced during a large portion of the operation.
83
-------
TABLE 20. SO, INLET AND OUTLET CONCENTRATIONS
9 a
SO, Inlet, ppm SO, outlet, ppm
Date
(1990)
Dally
average
Max)mum
2nd
highest
Minimum
Dally
average
Maximum
2nd
highest
Minimum
Average datjy
efficiency
Activity
6/7°
<1.0
11.4
8.0
<1,0
2.4
9.0
8.0
<1.0
d
Background data
6/8°
<1.0
<1.0
<1.0
<1.0
<1.0
2.0
2.0
<1.0
d
Excavate overburden
6/9
8.4
135.9
128
<1.0
<1.0
<1.0
<1.0
<1.0
96.7
Excavate tar/foam
6/10
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
d
-
6/11
<1.0
2.5
2.3
<1.0
<1.0
1.1
1.1
<1.0
d
Excavate tar/foam
6/12
121.8
1000.0
641.5
<1.1
<1.0
1.1
<1.0
<1.0
99.8
Excavate tar/foam
6/13*
127.5
466.1
463.9
19.9
<1.0
11.9
11.5
<1.0
99.5
Excavate tar/foam
6/14
127.6
637.7
564.4
12.6
<1.0
<1.0
<1.0
<1.0
99.9
Moved tar/foam
6/15
203.6
754.8
744.2
21.2
<1.0
<1.0
<1.0
<1.0
99.7
Excavate tar/char/foam
6/16
68.8
126.8
126.3
<1.0
<1.0
<1.0
<1.0
<1.0
99.9
-
6/17
50.5
61.3
60.3
5.8
<1.0
<1.0
<1.0
<1.0
99.8
-
6/18
40.0
65.0
59.4
24.9
<1.0
<1.0
<1.0
<1.0
99.2
-
6/19
29.8
42.1
41.4
21.0
<1.0
<1.0
<1.0
<1.0
99.4
-
6/20
18.7
31.6
31.2
<1.0
<1.0
<1.0
<1.0
<1.0
98.2
-
6/21
7.1
25.4
24.8
<1.0
<1.0
<1.0
<1.0
<1.0
97.0
Overburden/foam
6/22
5.3
10.5
10.5
1.5
<1.0
<1.0
<1.0
<1.0
96.1
Stab, foam
6/23
5.9
11.6
11.4
2.3
<1.0
<1.0
<1.0
<1.0
37.6
-
6/24
6.1
11.7
11.4
3.1
<1.0
<1.0
<1.0
<1.0
97.5
-
6/25
6.2
10.7
10.5
3.2
<1.0
<1.0
<1.0
<1.0
94.6
-
6/26
115.3
593.3
576.5
3.6
2.9f
39.9f
38.9f
<1.0
94.9
Excavate tar/char/foam
6/27
41.5
83.2
79.1
13.6
<1.0
2.4
2.4
<1.0
97.8
Tar seepage
6/28
19.0
82.4
81.9
2.5
<1.0
<1.0
<1.0
<1.0
98.5
Moved char/tar
6/29
35.4
213.8
213.5
5.6
<1.0
<1.0
<1.0
<1.0
97.4
Moved char
(continued)
-------
TABLE 20 (continued)
a a
SO, Inlet, ppm $0, outlet, ppm
Date
(1990)
Dally
average
Max1mum
2nd
highest
Minimum
Dally
average
Maximum
2nd
highest
Minimum
Average dalj^y
efficiency
Activity
6/30
7.3
10.6
10.4
3.4
<1.0
1.1
1.1
<1.0
87.5
-
7/1
6.6
12.6
12.3
2.4
1.1
1.7
1.7
<1.0
78.8
-
7/2
29.3
162.9
160.3
2.4
<1.0
1.4
1.4
<1.0
89.1
Moved tar/mixed waste
7/3
1.2
2.7
2.7
<1.0
<1.0
<1.0
<1.0
<1.0
71.8
Equipment activity
7/4
<1.0
1.7
1.7
<1.0
<1.0
<1.0
<1.0
1.1
d
-
7/5
<1.0
1.7
1.7
<1.0
<1.0
1.5
1.5
<1.0
d
-
7/6
<1.0
<1.0
<1.0
<1.0
g
-
-
-
d
-
7/7
<1.0
<1.0
<1.0
<1.0
g
-
-
-
d
-
7/8
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
d
-
7/9
<1.0
1.2
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
d
-
7/10
4.9
38.8
37.4
<1.0
<1.0
<1.0
<1.0
<1.0
99.8
Run 1/Run 2 - tar
7/11
2.1
8.9
8.4
<1.0
<1.0
1.1
<1.0
<1.0
98.4
Runs 3, 4, 5, & 6 - tar
7/12
1.9
9.0
8.6
<10
<1.0
<1.0
<1.0
<1.0
92.5
Runs 7, 8, & 9 - tar
7/13
<1.0
<1.0
<1.0
<1.0
<1.0
2.1
1.2
<1.0
d
Run 10 - tar
7/14
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
d
-
7/15
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
d
-
7/16
6.4
40.8
39.4
<1.0
<1.0
1.2
<1.0
<1.0
91.5
Move tar
7/17
14.9
85.3
81.7
<1.0
<1.0
2.7
2.7
<1.0
82.0
Move tar
7/18
<1.0
1.1
1.0
<1.0
<1.0
4.5
4.3
<1.0
d
Material handling
(continued)
-------
TABLE 20 (continued)
8 Method Detection Limit (HOL) was 2% of scale (0 to 50 ppm) or 1 ppm, Daily average value Is based on 24-hour period. Maximum and minimum
values are from 5-minute average values.
b Inlet S-minute averages - outlet 5-minute averages
Efficiency - inlet 5-minute average * 10°-
5
SO, Inlet and outlet data on 6/7 and SO, outlet data on 6/8 are biased high from moisture, which caused the instruments to drift. Values are
assumed to be below the detection limit.
Efficiencies were not calculated when inlet values were at the method detection limit (1 ppm).
e
SO, outlet values increased on 6/13 because scrubber pH went down to 2.9, which is below the normal operating pH of ? to 10.
SO, outlet values are biased high as a result of drift caused by moisture in the monitor.
g
Monitor off line as a result of moisture in monitor.
NOTE: Inlet site is air from the enclosure system (untreated). Outlet site is exhaust air from the scrubber and carbon treatment systems.
-------
1. The full calibration period is the injection of three calibra-
tion gases and zero air. The calibration is performed to insure the
system is operating properly and to enable data reduction. The calibra-
tion takes approximately one hour.
2. Reading data storage cartridge is the transfer of CEM data from the data
acquisition system (DAS) to the computer disk. The transfer of data
temporarily impedes data to be recorded to the DAS. Transfer of data
takes approximately 30 minutes.
3. Span and zero checks are conducted on each CEM when there was a
convenient break in excavation. These checks are conducted to insure
proper system operation. Span and zero checks take 30 minutes.
4. The probe was taken out of the sampling port for several reasons; a) to
clean the sampling system, b) to inject calibration gases through the
sample system, and c) to replace the sample filters. Also, at certain
times the probe fell out of the stack from fan vibration.
5. Condensers were emptied on the S02 instruments. These instruments
require water condensers to prevent moisture from entering the instru-
ment.
° The above occurrences show up on the graphs as "gaps" in the data.
The gaps in data are times when data were not being recorded to the
data acquisition system. The six items are abbreviated for use in the
text.
° S02/THC In - Designates sample location at the inlet to the air treatment
system. The inlet site samples untreated gases from the enclosure.
» S02/THC Out - Designates sample location at the outlet to the air
treatment system (scrubber and carbon bed). The outlet site samples
treated gases from the enclosure.
o The vertical graph axis is the concentration of the pollutant
(S02/THC) in parts per million (ppm).
° The horizontal axis is the time of day in two-hour increments.
Figure 29. Legend for data plots on Figures 30 through 65.
87
-------
See Figure 29 for explanation of legend.
Figure 30. Plot of SO2 concentrations vs time for June 9,1990
-------
1000-
See Figure 29 for explanation of legend.
•
¦
S02!N
_ S02 OUT
-
1
-
-
V 1
© 1
-
\
-
-
©
o
o
o
o
o
o
CM
O
O
O
s
o
o
«D
o
o
o
00
o
o
<Ł>
o
o
o
CM
o
o
<Ł>
O
Q
00
o
o
o
OJ
o
o
cvi
CM
Figure 31. Plot of S02 concentrations vs time for June 12,1990.
-------
500
See Figure 29 for explanation of legend.
450
400
350
SCL OUT
300
q. 250
Q.
200
150
100
0-I
o
o
o
o
«D
o
o
CO
o
p
o
CM
o
o
o
o
o
o
o
o
o
o
0
01
o
o
o
o
o
o
o
o
o
OI
o
o
o
Figure 32. Plot of S02 concentrations vs time for June 13,1990.
-------
700
See Figure 29 tor explanation of legend.
600
500
400
E
CL
0.
300
200
100
0-J
oooooooooooo
oo
-------
800
700
600
500
L 400
x
300
200
100
See Figure 29 for explanation of legend
SCUN
S02 OUT
Figure 34, Plot of S02 concentrations vs time for June 15,1990.
-------
See Figure 29 for explanation of legend.
o
o
00
o
o
o
o
o
o
csi
o
o
o
o
CO
o
o
CD
o
o
o
CO
o
o
o
o
o
o
o
o
o
CM
o
Figure 35. Plot of S02 concentrations vs time for June 18,1990.
-------
600
See Figure 29 for explanation of legend.
500
400
q. 300
Q.
200
100
0-J
o
o
(0
O
o
d
o
o
o
o
p
o
CM
o
o
o
o
o
o
CO
o
o
cvi
CM
o
o
o
o
o
o
00
o
o
p
d
o
o
o
Figure 36. Plot of S02 concentrations vs time for June 26,1990.
-------
90-
80-
70-
60
50
CO
en
E
CL
Q.
40-
30-
20
10
See Figure 29 for explanation of legend.
A
¦
-
so2in
_ S02 OUT
-
0
A
\
\
J
P©
*\rt
Aw
\°Ji
0
•
o S
8 8
o
es
o
o
p
to
o
o
p
00
o
o
o
o
p
C4
o
o
o
o
CD
o
p
CO
o
p
o
CM
O
p
c\i
CM
Figure 37. Plot of SOz concentrations vs time for June 28,1990.
-------
250
See Figure 29 for explanation of legend.
200
150
E
Q.
Q.
100
0-I
o
O
O
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
d
CM
o
o
o
o
o
CM
Figure 38. Plot of S02 concentrations vs time for June 29,1990.
-------
180
See Figure 29 for explanation of legend.
160
140
120
100
0-J
o
O
CD
O
O
66
o
o
o
CM
o
o
o
o
c\i
CM
o
o
o
o
ci
o
o
o
p
00
o
o
o
o
o
o
o
CD
Figure 39. Plot of S02 concentrations vs time for July 2,1990.
-------
See Figure 29 for explanation of legend.
S0„ OUT
o-l
O
p
00
o
p
o
CM
O
O
o
o
o
p
<0
o
p
CO
o
o
p
o
o
o
o
o
o
o
o
o
CM
o
Figure 40. Plot of S02 concentrations vs time for July 10,1990.
*
-------
See Figure 29 for explanation of legend.
S02!N
S02 OUT
Figure 41. Plot of S02 concentrations vs time for July 11,1990.
-------
See Figure 29 for explanation of legend
so2in
S02 OUT
Figure 42. Plot of SO2 concentrations vs time for July 12,1990.
-------
See Figure 29 for explanation of legend.
Figure 43. Plot of SO2 concentrations vs time for July 13,1990.
-------
0.25
0.15
E
Q.
a.
See Figure 29 for explanation of legend
SO, IN
SO. OUT
0.05
Figure 44, Plot of S02 concentrations vs time for July 14,1990.
-------
o
G1
See Figure 29 for explanation of legend.
s°2 IN
S02 OUT
Figure 45. Plot of S02 concentrations vs time for July 16,1990
-------
90-
80
70
60-
50
40
30
20
10
; See Figure 29 for explanation of legend.
¦
-
I
so2in
M. S02 OUT
-
-
-
-
V
-
I
€»
o
o
©
o
o
o
csi
©
o
p
"
-------
As a result of these high inlet and low outlet concentrations, the S02 removal
efficiency of the scrubber was higher than expected. For the operating days in
Table 20 for which daily average S02 inlet concentrations were above 10 ppm, the
daily average S02 removal efficiencies were always above 95 percent. On many of
these days, S02 removal efficiencies exceeded 99 percent.
The normal operating range for the scrubber liquor pH was established at 10 to
13 by the scrubber manufacturer prior to the trial excavation. It was noted, however,
that operation near the high end of this range often caused excessive foaming of the
scrubber liquor near the bottom of the packed tower, with subsequent overflow of
liquor out through the inlet duct and into the filter box. In light of the high S02 removal
levels demonstrated by the scrubber, the decision was made to reduce the pH operat-
ing range to 7 to 10. This change had the effect of eliminating the liquor foaming and
overflow problem, while maintaining consistently low outlet S02 concentrations.
Besides foaming, the only other operational problem encountered with the S02
scrubber was occasional restrictions in the tower, which caused low ventilation airflow.
The first episode occurred on June 15 and was diagnosed as excessive solids passing
through the filter (upstream of the scrubber) and building up in the scrubber packing.
The low airflow conditions were relieved by blowing down the scrubber liquor, washing
down the packing, and increasing the frequency of filter inspections and changes.
The filter system used during the trial excavation was a low-efficiency, field-fabricated
system which relied upon residential furnace filters as the filter media.
The second episode of low airflow occurred on July 11. The solids content of
the scrubber liquor at this time was much lower than during the first episode. Inspec-
tion of the packing balls through the lower access port showed that many contained a
buildup of black, soot-like material that appeared to be composed of very fine particu-
late matter. By experimenting, it was found that the airflow could be returned to nor-
mal levels by decreasing the liquor recirculation flowrate from its normal range of 15 to
20 gallons/minute to near 5 gallons/minute. Since the outlet S02 concentration re-
mained low even at the lower liquor recirculation flowrate, this rate was maintained for
the duration of the program.
At the conclusion of operations, the scrubber was shut down and opened at the
top cone and the bottom access port for inspection. At the top of the scrubber, the
demister pad was clean and free of any buildup. The packing balls at the top of the
scrubber were in a similar condition. At the bottom of the scrubber, packing balls near
the access port were found to be partially obstructed with the black buildup described
previously plus a white crystalline material speculated to be crystallized sodium hy-
droxide. Together, the combined solids filled approximately 25 percent of the volume
of these packing balls. However, after the first 6 inches of balls were removed from
the lower access port, it was clear that the packing balls in the center of the tower
were free of significant buildup. The air-distribution grid at the bottom of the packed
105
-------
tower was also free of solids buildup. Thus, the cause of the second incident of low
ventilation airflow could not be identified. All other portions of the scrubber were in
good working order at the completion of program operations.
With respect to a final remediation scrubber, one change recommended as a
result of trial excavation operations would be the installation of a high-efficiency indus-
trial particulate collection device upstream of the scrubber. The device should be de-
signed to capture both large and fine particulate (e.g., diesel engine emissions) to a
high degree and thereby prevent the buildup of solids in the scrubber liquor and pack-
ing material. Alternatively, a more open S02 absorber design could be employed,
such as a spray tower. In addition, an automatic pH control system should be added
that will maintain the desired pH range by addition of caustic soda, as opposed to the
manual system employed during the trial excavation.
4.1.5 Total Hydrocarbon Concentrations Before and After the Enclosure Exhaust
Control System
As mentioned in Section 4.1,2, the CEMS at the inlet and outlet of the enclosure
exhaust control system operated continuously from June 7 until July 18. In addition to
SOp monitoring, total hydrocarbon (THC) monitoring took place at the inlet and outlet
to the enclosure exhaust control system. A chronological listing of excavation activities
during this time period is included in Section 3.2.2 of this report, and is also contained
in the daily operating logs in Appendix A.
Table 21 presents the THC data generated. The table lists the enclosure
exhaust control system inlet and outlet concentrations (ppm) expressed as daily
averages, minimum and maximum values. The maximum and minimum values are 5-
minute average values. All of the data were generated from the DAS generated 5-
minute averages.
Also presented in this table are THC removal efficiencies calculated for each 5-
minute data set and averaged for the day. These data will be discussed in detail in
Section 4.1.6.
Graphs were generated from the 5-minute data sets displaying THC inlet and
outlet concentrations versus time. Figures 47 through 65 display the variations of THC
inlet and outlet concentrations for selected days during the trial excavation.
Additional plots of THC concentrations versus time are presented in
Appendix B-1.
106
-------
TABLE 21. THC INLET AND OUTLET CONCENTRATIONS
THC inlet,* ppm THC outlet.** ppm
Date
(1990)
Dally
average
Maximum
2nd
highest
Minimum
Dally
average
Maximum
2nd
highest
Minimum
Average daijy
efficiency
Activity
6/7
7.6
18.2
18.0
<4.0
2.7
6.2
5.4
<2.0
61.8
Background data
6/8
9.9
20.6
20.3
4.2
3.5
5.9
5.8
<2.0
61.2
Excavate overburden
6/9
27.1
107.8
103.4
5.5
10.9
37.2
32.7
3,0
54.2
Excavate tar/foam
6/10
29.2
40.5
40.5
21.0
11.9
15.1
15.0
4.3
59.6
-
6/11
31.3
58.9
58.3
<4.0
13.6
17.2
17.1
2.6
55.5
Foam applfed
6/12
69.9
305.6
273,1
12.8
25.8
62.5
61.4
8.9
53.1
Excavate tar/foam
6/13
103.7
221.8
219.1
22.9
52.5
117
110
22.3
50.0
Excavate tar/foam
6/14
193.1
492.3
464.0
7.5
82.6
284.3
224.2
49.3
49.4
Moved tar/foam
6/15
225.2
341.4
340.5
10.7
98.1
151.3
151.3
14.9
56.0
Excavate tar/char/foam
6/16
92.9
228.9
222.0
<4.0
20.9
85.3
81.1
<2.0
84.0
Change to coconut carbon
6/1?
87.3
288.6
104.2
5.3
8.7
27.6
25.7
<2.0
90.7
-
6/18
82.7
101.2
101.2
42.4
15.6
22.4
22.3
11.6
81.4
-
6/19
90.3
117.4
110.4
77.8
25.0
37.3
36.9
18.1
72.5
-
6/20
67.3
98.6
98.5
<4.0
27.9
36.6
36.6
20,5
67.9
-
6/21
53.4
131.9
128.0
<4.0
31.1
70.7
70.6
14.9
58.4
Overburden/foam
6/22
57.3
74.9
74.8
27.7
22.3
32.3
32.1
13.7
60.4
Stab, foam applied
6/23
54.9
72.0
71.8
40.1
20.6
29.9
29.6
3.5
61.9
-
6/24
47,2
61.6
61.4
35.2
19.1
25.3
25.2
15.9
58.6
-
6/25
47.7
66.2
65.5
<4.0
14.2
26.2
26.2
<2.0
72.3
-
6/26
121.9
345.3
326.5
33.6
29.9
76.7
75.3
<2.0
78.1
Tar/excavate char/foam
6/27
95,8
278.4
359.5
<4.0
37.6
71.9
71.6
26.1
60.3
Tar seepage
6/28
80.2
161.2
158.5
52.4
42.8
67.5
66.6
4.4
44.4
Moved char and tar
6/29
83,6
714.8
649.9
<4.0
46.5
83.4
82.8
24.6
39.2
Moved char
(continued)
-------
TABLE 21 (continued)
THC Inlet, pprn
THC outlet, ppm
Date
{1990)
Dally
average
Maximum
2nd
highest
Minimum
Daily
average
Maximum
2nd
highest
Minimum
Average daijy
efficiency
Activity
6/30
48.5
65.8
65.7
4.3
37.0
76,7
68.0
11.2
24.?
-
7/1
55.2
86.0
84.7
37.2
33.3
51.9
51.8
25.3
38.6
-
7/2
98.6
174.0
173.8
44.5
49.7
92.6
92.2
21.7
47.7
Moved tar and mixed
waste
7/3
43.4
67.8
67.3
18.3
30.5
45.0
44.7
16.6
39.4
Equipment activity
7/4
46.8
60.3
60.1
39.2
25.6
42.0
41.3
17.0
44.4
-
7/5
41.6
50.3
49.3
35.9
22.7
36.4
36.2
14.8
44.4
-
7/6
13.0
39.5
39.5
<4.0
9.2
24.6
24.6
<2.0
Offline
-
7/7
<4.0
<4.0
<4,0
<4.0
<2.0
<2.0
<2.0
<2.0
Offline
-
7/8
<4.0
<4.0
<4.0
<4.0
<2.0
<2.0
<2.0
<2.0
Offline
-
7/9
21.4
37.9
37.5
<4.0
15.4
29.8
29.7
<2.0
31.3
-
7/10
108.5
193.4
192,6
19.3
29.7
117.8
92.7
12.4
55.9
Pug mill Runs 1 & 2
7/11
193.3
358.0
358.0
88.2
51.1
d
106.4
d
106.4
25.5
71.6
Pug mi 11 Runs 3, 4,
5, & 6
7/12
149.2
548.5
278.3
18.0
84.4
132.1
131.6
<2.0
42.1
Pug mill Runs 7, 8,
8. 9
7/13
75.2
134.1
132.9
18.5
67.6
106.4d
d
106.4
32.0
15.8
Pug mill Run 10
7/14
58.1
78.1
77.5
40.0
54.9
65.6
65,5
29.0
23,6
-
7/15
47.1
54.0
54.0
33.3
37.7
55.2
55.2
1 M
C ,
CO o
•J J , I
-
7/16
84.4
175.5
172.9
12.3
56.9
d
106,4
d
106.4
2.1
45.4
Move tar
7/17
84.5
278.3
278.3
<4.0
34.7
121.7
120.6
<2.0
73.3
Move tar
7/18
39.1
94.5
89.2
<4.0
34.5
71,5
71.5
<2.0
24.2
Material handling
Inlet detection limit was 4 ppm, Inlet site is air from the enclosure system (untreated).
Outlet detection limit was 2 ppm. Outlet site is exhaust air from the scrubber and carbon treatment systems.
Efficiency
Inlet 5-winute average - outlet 5-mlnute average
Inlet 5-minute average
x 100.
Values are low because the voltage signal was at Its maximum (10 volts).
NOTE: Maximum and minimum values are from 5-minute average values.
-------
Figure 47. Plot of THC concentrations vs time for June 7,1990.
-------
25-
See Figure 29 tor explanation of legend.
THC IN
THC OUT
©
o
o
o
o
o
o
ej
o
o
o
s
o
o
CO
o
o
p
66
o
o
o
o o
p p
c\i ^
o
p
CO
o
p
66
o
o
o
p
p oi
c\j eg
Figure 48. Plot of THC concentrations vs time for June 8,1990.
-------
120
See Figure 29 for explanation of legend.
too
THC IN
THC OUT
80
60
40
20
Sr- m i ¦ *2
0
o
o
o
o
o
CM
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Figure 49. Plot of THC concentrations vs time for June 9,1990.
-------
See Figure 29 for explanation of legend.
Figure 50. Plot of THC concentrations vs time for June 12,1990.
-------
250
See Figure 29 for explanation of legend.
200
THC IN
THC OUT
150
100
50
0
o
o
to
o
o
00
o
o
o
o
o
o
cvi
o
o
o
p
to
o
p
ci
OJ
§
o
o
g
o
o
to
o
o
o
o
Figure 51. Plot of THC concentrations vs time for June 13,1990.
-------
500
450
q. 250
Q.
See Figure 29 for explanation of legend.
THC IN
THC OUT
Figure 52. Plot of THC concentrations vs time for June 14,1990.
-------
350
See Figure 29 for explanation of legend.
300-
250
THC IN
200
E
Q.
Q.
THC OUT
150
oo
100
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
CM
o
o
d
o
o
o
§
o
o
o
p
to
o
o
CM
Figure 53. Plot of THC concentrations vs time for June 15,1990.
-------
120-
See Figure 29 for explanation of legend.
100
THC IN
THC OUT
40-
o
o
o
o
o
o
csi
o
o
o
o
o
o
to
o
o
o
CO
o
o
o
o
o
csi
o
o
o
o
to
o
p
CO
o
p
o
CM
o
o
Figure 54. Plot of THC concentrations vs time for June 18,1990.
-------
350
See Figure 29 for explanation of legend.
300
THC IN
250
THC OUT
200
150
100
50
o
O
o
CM
O
p
csi
CM
o
p
o
o
o
o
csi
00
CM
O
CD
O
O
O
Figure 55. Plot of THC concentrations vs time for June 26,1990.
-------
180
See Figure 29 for explanation of legend
THC IN
THC OUT
Figure 56.
Plot of THC concentrations vs time for June 28,1990.
-------
300
See Figure 29 for explanation of legend.
250
THC IN
THC OUT
200
o. 150
Gl
100
50
0
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
CM
o
o
o
o
o
o
CD
o
o
Figure 57. Plot of THC concentrations vs time for June 29,1990.
-------
180
160
140
See Figure 29 for explanation of legend.
120-
100
THC IN
THC OUT
Figure 58. Plot of THC concentrations vs time for July 2,1990.
-------
200
See Figure 29 for explanation of legend.
180
160
THC IN
140
THC OUT
120
80
60
40
20
0
o
o
00
o
o
©
o
o
o
o
o
<6
o
o
o
CM
o
0
01
CM
o
o
CD
O
o
o
00
o
o
o
o
c\i
o
8
o
o
Figure 59. Plot of THC concentrations vs time for July 10,1990.
-------
See Figure 29 for explanation of legend.
Figure 60. Plot of THC concentrations vs time for July 11,1990.
-------
300
See Figure 29 for explanation of legend.
THC IN
THC OUT
a 150
Figure 61. Plot of THC concentrations vs time for July 12,1990.
-------
140
See Figure 29 for explanation of legend.
100"
THC IN
THC OUT
Figure 62. Plot of THC concentrations vs time for July 13,1990.
-------
See Figure 29 for explanation of legend.
Figure 63. Plot of THC concentrations vs time for July 14,1990.
-------
See Figure 29 for explanation of legend
THC IN
THC OUT
\
Figure 64. Plot of THC concentrations vs time for July 16,1990,
-------
450-
See Figure 29 for explanation of tegenc
400-
350
300-
THC IN
THC OUT
Figure 65. Plot of THC concentrations vs time for July 17,1990.
-------
4.1.6 Carbon Adsorber Performance
As discussed in Subsection 2.1, two types of granular activated carbons were
used in the carbon adsorber to remove hydrocarbon pollutants from the ventilation air
stream. The first was a coal-based carbon installed on June 5. This carbon was used
during the first 9 days of excavation operations between June 7 and 15. On June 16,
the coal-based carbon was replaced with a coconut-based carbon. The coconut-
based carbon was used during the remaining operation period until system shutdown
on July 18, for a total of 32 operating days.
To assess the performance of these carbons, the hydrocarbon removal efficien-
cies associated with the maximum 5-minute average inlet THC concentrations shown
in Table 21 were calculated and compared over time for the two carbon types. These
data show that the average daily hydrocarbon removal efficiency for the coal-based
carbon ranged from 61.8 percent (fresh carbon) to 49.4 percent over a nine-day
period from June 7 to June 15. For the coconut-based carbon, average hydrocarbon
removal efficiency ranged from 90.7 on June 17 (first full day of operation on new
carbon) to 58.4 percent over the first nine days of operation. By comparison, the
performance of the coconut-based carbon was slightly superior to that of the coal-
based carbon with respect to both initial activity and activity over a nine-day period.
For the remainder of the coconut-based carbon operating period, average
hydrocarbon removal efficiency declined from 78.1 percent on June 26 to 24.2 percent
on July 18. The exception to this trend was an increase in average removal efficiency
to 55.9 percent on July 10 and 71.6 percent on July 11. During other short-term
periods on those days, hydrocarbon removal efficiencies reached 93 percent on July
10 and 92 percent on July 11. The high removal efficiencies on July 11 corresponded
closely to the periods of low airflow rates measured on this day; after the airflow rate
was returned to normal levels (by adjustment of the scrubber recirculation rate), the
hydrocarbon removal efficiencies decreased. Although no airflow rate data are
available for July 10, the hydrocarbon removal efficiency data suggest that the flow
rate was also low on this day.
Post-operative inspection of the activated carbon unit showed no visible dam-
age to nor buildup on the spent carbon particles. Water corrosion was evident on the
steel rollers at the bottom of the accumulator cabinet, however. It is unlikely that this
water came in the form of carryover water droplets from the wet scrubber, since the
scrubber demister packing was in good order at the end of operations and the knock-
out pot (installed between the scrubber and carbon unit) showed very little water ac-
cumulation when checked regularly. A more likely source of water was air moisture
condensation on the inside of the accumulator cabinet during the cool nighttime and
early morning hours. The air entering the cabinet was no doubt saturated after pass-
ing through the packed-bed scrubber. Contact of this saturated gas with cold carbon
cabinet walls would be sufficient to cause water condensation and accumulation.
128
-------
Such condensation and accumulation was noted on the top inside panel of the accu-
mulator cabinet during periodic field inspections. The presence of water in the carbon
unit was also supported by the hard black powdery deposits found on the fan vanes
and housing after operations were completed. These deposits were likely formed by
the combination of moisture and fine pieces of carbon that attrited from the activated
carbon unit.
The presence of moisture in the carbon unit helps to explain the lower-than-
expected hydrocarbon removal performance of this system during the trial excavation.
The design specifications for this system were 95 percent THC removal. However, the
inlet THC concentration was much higher than expected due to the low vapor sup-
pression effectiveness of the foam. Nevertheless, the manufacturer of the carbon unit
still expected performance levels to be above 90 percent removal. Moisture conden-
sation onto carbon particles with subsequent reduction in active surface area remains
the most likely explanation for under design performance. This explanation is consist-
ent with the gradual loss of carbon THC removal efficiency observed over time, as well
as the increase in removal efficiency that occurred when the airflow rate was
significantly reduced on July 10 and 11.
A number of options would be available to eliminate moisture condensation
problems for a final remediation activated-carbon unit. These include installation of an
air dryer upstream of the carbon unit to lower scrubbed ventilation air moisture con-
tent, use of a dry scrubber in place of the wet scrubber used for the trial excavation,
adding insulation/heaters to the accumulator cabinet, and operating a duct heater
upstream of the carbon unit to maintain ventilation air temperature above the stream's
dewpoint. The technical and economic merits of these options should be evaluated,
either separately or in combination, for potential application during final remediation.
4.1.7 Sulfur Dioxide Monitoring at the Excavation Area
As a result of the revised Health and Safety Plan, a Thermo Electron S02 ana-
lyzer sampling system similar to the scrubber monitoring system, was set up to moni-
tor the work environment at the excavation area. The monitor was calibrated using a
span and zero gas to determine a parts-per-million (ppm) -per-chart-division (CD) for-
mula. The monitor was calibrated on the 0- to 1000-ppm scale of the instrument.
Only during periods when the S02 level exceeded the detection limit (2 percent of full
scale or 20 ppm) were the strip-chart data reduced. Table 22 presents the excavation
area S02 concentration levels including the arithmetic average. This instrument was
put on line June 26 and ran through July 18.
4.1.8 Exhaust Gas Flow Rates
Exhaust gas flow rate measurements were taken at the inlet and outlet test sites
as a check on fan flow. Table 23 summarizes exhaust gas conditions during this
129
-------
TABLE 22,
EXCAVATION AREA S02
DATA
Date
Average.
Highest,
2nd
Minimum,
(1990)
Time (24-h)
ppm
PDffl
highest, pom
ppm
Comments
6/26
0700-0800
323.2
797.9
434.3
30.3
Tar moving
0800-0900
363.6
424.2
399.0
313.1
Char evaporation
0900-1000
373.7
489.9
464.6
262.6
1000-1100
429.3
868.6
424.2
308.1
1100-1200
363.6
686.8
686.8
222.2
1200-1300
171.7
222.2
222.2
136.4
1300-1400
111.1
136.4
136.4
90.9
Foaming
1400-1500
80,8
90.9
90.9
70.7
1500-1600
65.7
70.7
70.7
60.6
1600-1700
60.6
75.8
70.7
53.5
1700-1800
65.7
101.0
93.9
50.5
1800-1900
70.7
96.0
93.9
50.5
1900-2000
68.7
85.9
80.8
55.6
2000-21Q0
55.6
63.6
60.6
50.5
2100-2200
58.6
68.7
68.7
50.5
2200-2300
85.9
114.1
111.1
63.6
Tar seepage
2300-2400
90.9
101.0
99.0
75.8
6/27
0000-0100
70.7
75.8
75.8
63.6
Tar seepage
0100-0200
65.7
65.7
65.7
60.6
0200-0300
60.6
60.6
60,6
60.6
0300-0400
58.6
60.6
60.6
55.6
0400-0500
55.6
55.6
55.6
55.6
0500-0600
53.5
53.5
53.5
53.5
0700-0800
59.6
82.2
64.7
54.4
Tar seepage
0800-0900
51.4
61.6
56.5
36.0
0900-1000
131.5
246.5
188.0
66.8
1000-1100
69.8
102.7
85.3
51.3
1100-1200
100.7
184.9
184.9
51.3
1200-1300
49.3
71.9
71.9
25.7
1300-1400
25.7
30.8
28,8
23.6
1400-1500
20.5
23.6
23.6
18.5
(continued)
130
-------
TABLE 22 (continued)
Date
(1990)
Time (24-h)
Average,
ppm
Highest,
ppm
2nd
highest, ppm
Minimum,
ppm
Conments
6/28
1000-1100
57.7
164.0
142.1
27.8
Moved char
1100-1200
54.7
129.2
84.5
34.8
Moved tar
1200-1300
27.8
37.8
34.8
22.9
1300-1400
19.9
22.9
22.9
17.9
6/Z9
1100-1200
63.9
268.8
266.7
<20.3
Char moved
1200-1300
142.0
302.2
238.3
76.1
1300-1400
65.9
74.0
60.9
53.8
1400-1500
69.0
187.6
182.3
43.6
1500-1600
233.2
365.1
339.7
142.0
1600-1700
116.6
172.4
121.7
81.1
1700-1800
76.1
91.3
81.1
63.9
1800-1900
50.7
63.9
63.9
38.5
1900-2000
33.5
38.5
38.5
23.3
2000-2100
20.2
25.4
25.4
<20.3
7/2
0700-0800
53.3
150.3
129.0
<19.4
0800-0900
158.1
232.8
169.7
116.4
Moved tar
0900-1000
145.5
201.8
194.0
106.7
Moved and mixed
waste
1000-1100
97.0
140.6
129.0
58.2
1100-1200
48.5
61.1
61.1
38.8
1200-1300
22.3
24.2
24.2
19.4
1300-1500
Below detection limit
1500-1600
46.6
158.1
130.9
<19.4
1600-1700
32.0
43.6
38.8
24.2
1700-1800
24.2
29.1
27.1
19.4
1800-
7/16 0900
Below detect
:ion limit
7/16
0900-1000
89.5
119.3
102.4
49.7
Hoved tar
1000-1100
104.4
114.3
109.3
99.4
1100-1200
32.8
42.7
39.8
22.9
1200-
7/17 0700
Below detection limit
(continued)
131
-------
TABLE 22 (continued)
Date
Average,
Highest,
2nd
Minimum,
(1990)
Time (24-h)
ppm
pom
highest, ppn
ppm
Comments
7/17
0700-0800
98.9
128.6
123.6
84.1
0800-0900
69.2
101.9
82.1
39.6
0900-1000
47.5
64.3
59.3
34.6
1000-1100
64.3
74.2
72.2
54.4
1100-1200
47.5
69.2
62.3
29.7
1200-1300
29.7
39.6
39.6
24.7
1300-1400
Below detection limit
132
-------
TABLE 23. SUMMARY OF EXHAUST GAS CONDITIONS
Time
f 24-h)
Flow
rate
Temperature,
•F
Moisture
volume, %
Run No.
Date
(1990)
acfma
dscfm^
Stack outlet
M-l
6/7
1540
1391
1276
95
3.0
M-2b
6/15
0754
845
802
81
3.0
M-2
6/18
0840
1094
1040
80
3.0
M-3
6/21
1306
1017
958
83
3.0
M-4
6/25
1242
778
702
105
3.0
M-5
6/27
1252
1332
1181
111
3.0
M-6
6/28
1210
1309
1183
102
3.0
M-7
6/29
1737
1385
1265
96
3.0
M-8
7/2
1559
1084
979
103
3.0
M-9
7/3
1419
621
560
103
3.0
M-10
7/9
1313
1267
1167
95
3.0
M-ll
7/11
0914
458
415
104
3.0
M-12
7/12
1433
1460
1307
103
3.0
M-13
7/13
1507
1557
1392
103
3.0
M-14
7/16
1500
1561
1414
98
3.0
M-15
7/17
1440
1521
1362
103
3.0
Scrubber
inlet
M-l
6/7
1530
1174
1092
93
2.0
M-2
6/18
0815
940
920
70
2.0
M-3
6/21
1258
766
730
81
2.0
M-4
6/26
1224
810
745
99
2.0
M-5
6/27
1245
1195
1079
106
2.0
M-6
6/28
1158
1182
1093
95
2.0
M-7
6/29
1745
1292
1219
84
2.0
M-8
7/2
1616
683
633
95
2.0
M-9
7/3
1428
581
537
95
2.0
M-10
7/9
1320
1099
1030
91
2.0
(continued)
133
-------
TABLE 23 (continued)
Flow rate
Run No.
Date
(1990)
Time
(24-h)
acfma
dscfm^
Temperature,
"F
Moisture
volume, %
M-ll
7/11
0905
359
339
86
2.0
M-12
7/12
1420
1356
1225
103
2.0
M-13
7/13
1520
1494
1353
101
2.0
M-14
7/16
1450
1584
1455
96
2.0
M-15
7/17
1436
1485
1348
100
2.0
a Flue gas flow rate at actual temperature and pressure.
k Flue gas flow rate In dry standard cubic feet per minute; 68* F,
29.92 in.Hg., and zero percent moisture.
c Moisture based on stack temperature and Method 6 results.
134
-------
program. When flows decreased, the system was checked and the problems were
corrected. The flow through the system affected the concentration levels in the enclo-
sure. It obviously took longer for the concentration levels in the enclosure to decrease
when there was a decrease in exhaust flow.
4.2 Perimeter Air Monitoring
As part of EPA's commitment to the community as embodied in the McColl
Community Response Contingency Plan, four perimeter air emission monitors were
situated around the site to continuously monitor for S02 and THC. See Figure 66 for
perimeter monitoring locations. There were no exceedances of the established work
stoppage levels of 0.5 ppm S02 for 5 minutes or 70 ppm THC for 30 seconds during
the trial excavation.
4.2.1 Sulfur Dioxide
The highest concentrations of S02 for the month of May occurred on May 25 at
Station No. 1 during the 0300 hour at 0.08 ppm. During June, the highest concentra-
tion was recorded on June 25 at 0.09 ppm at Station No. 4 during the 1100 hour.
During July, the highest concentration was recorded at 0.01 ppm, which was recorded
numerous times during the activities. The arithmetic mean for S02 during the trial
excavation was less than 0.01 ppm.
4.2.2 Total Hydrocarbons
The highest concentration of THC for May occurred on May 27 at Station 1 at
3.2 ppm during the 0400 hour. For June, the highest concentration was recorded on
June 12 at Station 4 at 21.9 ppm during the 1900 hour. In July, the highest concentra-
tion was recorded on July 11 at Station 2 at 9.8 ppm during the 1800 hours. The
arithmetic mean for May, June, and July were 0.15 ppm, 0.75 ppm, and 0.93 ppm
respectively. It is important to note that the highest monthly reading occurred when
excavation activities at the site were not occurring. This indicated that a portion of the
THC emissions recorded during the trial excavation were not related to the McColl
excavation site, but may have resulted from seeps. Seeps are common at the McColl
site, and some were close in relationship to the monitor stations.
4.3 Air Speciation Data
As part of the objectives of the trial excavation, it was desirable to undertake a
sampling and analytical study to evaluate the organic and inorganic air pollutant
species that were emitted during the excavation activities at McColl. A sampling and
analytical plan was prepared as part of the McColl project requirements. Subcontract
arrangements were made with Pacific Analytical, Inc. (PAI) to perform gas chromato-
graphic analysis using flame photometric detection to analyze air samples for trace
135
-------
Rosecrans Avenue
a
MET
TOWER
' » EQUIP.
*J DECON
AREA
Sumps
K-1
Excavation
Area
W
t-3
Sumps
Los CcyoiM
Celt Court#
LL
• Perimeter monitoring sites
Figure 66. MeCoIl site in Fullertort, California.
136
-------
quantities of sulfur-containing compounds. PAI was also requested to perform gas
chromatography-mass spectrometry (GC/MS) analysis on the same air samples to
determine a broad spectrum of organic species that also might be emitted from
McColl wastes. The intent of the use of the air speciation data was to help assist the
EPA and the CHS in evaluating the potential health risks for future planning for the
cleanup of McColl.
Air samples were taken at designated times during excavation activities. The air
samples were collected in Tedlar bags on each designated day and delivered by
special courier to the PAI laboratories. The samples were analyzed within 24 hours
after receipt as prescribed in the analytical protocol.
Unfortunately, during a subsequent EPA quality assurance audit, it was deter-
mined that the PAI air speciation results did not meet the data quality requirements
and were not valid. The results are therefore not discussed in this report and cannot
be used as originally intended. However, the results will be maintained in a "RAW
DATA FILE" by EPA Region 9. The summary of the EPA quality assurance audit,
which was conducted by the Research Triangle Institute (RTI), is presented in
Appendix H.
4.4 Foam Efficiency Evaluation
4.4.1 Field Use of Foam During Excavation
Vapor-suppressing foams had been used successfully on other similar waste
and were therefore selected to test their effectiveness to reduce air emissions during
this excavation work. The earlier reported foam effectiveness values were based on
measurements of emissions from stationary samples of waste (i.e., static conditions)
with and without foam application. No reported data were available on the ability of
the foam to control emissions during actual excavation operations (i.e., under dynamic
conditions).
During this work, temporary foam was sprayed manually on freshly excavated
waste material or initially on stored material. Stabilized foam was then sprayed on all
waste surface areas at the end of each work day. The overall qualitative assessment
of the foam vapor suppressants used during this trial was that they were not as effec-
tive as expected. This assessment was based on visual observation of the foam
which disintegrated and did not adhere well to the raw wastes nor form a cohesive
film. The foam appeared to react with the highly acidic waste and at times turned from
greenish yellow to deep red. More over, total hydrocarbon and sulfur dioxide con-
centrations in the enclosure exhaust control system stream were higher than expected
(see Section 4.1); this is due mainly to lack of control by the foam. When stabilized
foam was placed on the waste at the end of a period of activity, air concentrations
slowly decreased. However, this decrease was in part due to no fresh waste being
137
-------
excavated and exposed, and in part to a constant flow of ambient ventilation air
sweeping across the enclosure, which had the effect of reducing concentration to an
equilibrium level. In an effort to increase the stabilized foam's effectiveness, the
concentration of stabilizer was increased June 21. The intent was to double the stabi-
lizer concentration. Analytical data from 3M indicated the concentration increased
from 9.6 to 10.5 percent. The increase in the foaming strength resulted in increasing
the foam's effectiveness, but did not cure the problems cited above.
The determination of foam efficiency, as applied under field conditions, is very
difficult since many factors affect the emissions at any given time. These factors
include the excavation activity (how much waste is being excavated or disturbed), the
flux rate of the particular waste, the enclosure air exhaust rate, the ambient
temperature, and any covering such as foam placed on the waste. In addition, the
hydrocarbons emitted by diesel engines increasd the THC measured in the enclosure
exhaust, and thus decreased any effect due to foam application. The following esti-
mates of foam efficiency application (based on selected increments of activity) were
determined by measurements made at the exhaust air system inlet monitoring location.
(See previous plots of concentration vs. time and also Appendix B).
° On 6/11, tar was leveled off in bins and total hydrocarbon (THC) concen-
tration was 50 ppm. Stabilized foam was applied at 1246 to all mud and
tar and THC decreased to 30 ppm by 1530 indicating a foam reduction
efficiency of 40 percent. S02 was fairly low and constant all day.
o On 6/12, tar excavation proceeded for about 25 minutes without foam
due to a communications problem. At 1433, temporary foam was ap-
plied, resulting in a decrease in S02 levels of about 40 percent (from
1000 ppm to 600 ppm) and a decrease in THC levels of about 40
percent (from about 250 ppm to 150 ppm).
o On 6/13, stabilized foam was applied to all mud and tar storage areas at
0945 to 1005. S02 decreased from 140 to 25 ppm by noon indicating an
efficiency of about 80 percent. THC decreased from 100 to 40 ppm over
the same time period (a 60 percent control).
In addition to assessing the effectiveness of stabilized foam under field condi-
tions, the effectiveness of temporary foam applied under conditions of waste excava-
tion and movement was desired. Operating periods have been identified in which
mud, tar, and char were either excavated or moved within the enclosure. The S02
and THC concentrations within the enclosure during these periods, as measured at
the inlet to the wet scrubber, have been summarized in Tables 24, 25, and 26, re-
spectively. "Hie S02/THC activity periods shown in the tables correspond to the se-
lected operating periods, but include allowances for system time lags between the
start/finish of an operation and the time at which changes in pollutant concentrations
138
-------
TABLE 24. EFFECT OF FOAM USE DURING MUD EXCAVATION AND MOVEMENT
SO, concentration Inside enclosure THC concentration Inside enclosure
S0,/THCa
Date activity Start.
(1990) period ppm
(time)0
Peak,
ppn
(time)0
Change,
ppm
Change
rate.
ppm/min
Start.
PP"
(time)0
Peak.
ppm
(tlme!°
Change.
ppm
Change
rate.
ppm/mtn
(S)1ng!e- or
(O)ouble-
strength foam
Hud excavation without foam
6/8 1021-1045 0.1
(1020)
0.5
(1045)
0,4
0.0
12
(1020)
21
(1045)
9
0.4
NAd
Hud excavation with foam
6/9 1033-1037 0
(1030)
0.1
(1035)
0.1
0.0
6.3
(1030)
6.3
(1035)
0
0.0
S
Mud movement without foam
Hi 1003-1150 145
M 1346-1505 29
(tooo)
(1350)
145
29
(1000)
(1350)
0
0
0.0
0.0
155
133
(1000)
(1350)
164
151
(1100)
(1435)
9
18
o o
»
NS
NA
Average
0
0.0
14
0.3
a
(missions activity period corresponding to mud excavation and movement, taking Into consideration time lags In the system,
b Based on S-mlnute average emission data.
Ł
Start of corresponding 5-mlnute average period.
^ Hot applicable.
-------
TABLE 25. EFFECT OF FOAM USE DURING TAR EXCAVATION AND MOVEMENT
SO, concentration Inside enclosure THC concentration Inside enclosure
SO,/IHCa
Change
Change
(S)lngle-
Date activity
Start.
c
Peak,
c
Change,
rate.
Start,
c
Peak.
Change,
rate.
(0)oub1i
(1990) period
ppm
(time)
ppm
(time)
ppm
ppm/mln
ppm
(time)
ppm
(time)
ppm
ppm/mln
strength
Tar excavation without
foam
6/9 1345-1414
0.1
(1345)
27
(1410)
27
1.1
13
(1345)
35
(1410)
22
0.9
d
NA
6/12 1408-1433
0.3
(1405)
1000
(1450)
1000
22.2
23
(1405)
306
(1445)
221
ZJ.
na
Average
514
11.7
153
4.0
Tar excavation with foam
6/9 1*14-1449
27
(1410)
136
(1445)
109
3.1
35
(1410)
92
(1445)
57
1.6
s
6/13 1 EOS -17 00
152
(1605)
342
(1700)
190
3.5
104
(1605)
155
(1700)
51
0.9
5
6/14 1325-1535
119
(1325)
564
(1520)
445
3.9
136
(1325)
464
(1515)
328
3.0
S
6/15 1035-1115
62
(1035)
142
(1110)
JSO
Jul
226
(1035)
226
(1050)
M
5
Average
206
3.2
109
1.4
Tar movement without foam
7/16 0845-0950
0
(0845)
41
(0950)
41
0.6
16
(0845)
117
(0945)
101
1.7
NA
7/18 1351-1445
7
(1350)
18
(1430)
11
0.3
117
(1350)
176
(1425)
59
1.7
NA
7/17 0725-0748
5
(0725)
85
(0745)
80
4.0
95
(0725)
200
(0745)
105
5.3
NA
7/17 1025-1045
70
(1025)
73
(1030)
_1
0.6
278
(1030)
278
(1030)
0
0.0
NA
Average
34
1.4
66
2.2
Tar movement with foam
6/14 1115-1200
123
(1115)
183
(1150)
60
1.7
149
(1115)
193
(1155)
44
1.1
S
7/17 0759-0815
82
(0800)
82
(0800)
0
0.0
231
(0800)
231
(0800)
0
0.0
S
7/17 0927-1030
31
(0925)
73
(1030)
42
0.6
166
(0925)
278
(1025)
112
1.9
S
?/!» 1045-1J15
67
11045)
73
(1055)
M
263
(1045)
267
(1050)
4
M
S
Average - single
27
0.7
40
1.0
strength
6/29 1545-1615
195
(1545)
195
(1514)
0
0.0
249
(1545)
249
(1545)
0
0.0
0
7/2 0700-0730
3.5
(1700)
19
(1725)
Ji
tJ
45
(1700)
78
(1725)
_33
1.3
0
Average - double
8
0.3
19
0.6
strength
4 Emissions activity period corresponding to char excavation and movement, taking Into consideration time lags In the system.
Based on 5-mlnut# average emission data.
C Start of corresponding 3-mlnute average period.
^ Not applicable.
-------
TABLE 26. EFFECT OF FOAM USE DURING CHAR EXCAVATION AND MOVEMENT
SO, concentration ins id* enclosure
THC concentration Inside enclosure.
SO,/THC
Change
U"
Change
(S)lngle- or
Date activity
Start,
c
Peak,
c
Change,
rate,
Start,
c
Peak,
e
Change,
rate.
(D)ouble-
(19901 period
ppm
(time)
ppm
(time)
ppm
ppm/mln
ppm
(tine)
ppm
(time)
ppm
ppm/mtn
strength foam
Char excavation with
foam
•/IS 1115-1315
142
(1115)
755
(U45)
613
20.4
270
(1115)
341
(1130)
71
4.7
S
6/26 ono-om
35
(0710)
462
(0740)
427
14.2
83
(0710)
142
(0730)
59
3.0
S
8/26 0908-1003
316
(0905)
594
(0945)
m
_U
126
(0905)
345
(0955)
m
Id
S
Average
439
13.9
116
4.0
Char movement with foam
6/26 0908-1003
316
(0905)
594
(0945)
278
7.0
126
(0905)
345
(0955)
219
4.4
S
6/28 0935-1100
23
(0930)
61
(1055)
J#
M
92
(0930)
145
(1040)
J2
JLI
S
Average - single
168
3.9
136
2.6
ltrength
6/29 1123-1215
13
(1120)
162
(1200)
149
3.7
59
(1120)
139
diss)
80
2.3
D
6/29 1441-1545
47
(1435J
214
(1520)
16?
3.7
91
(1435)
250
(1540)
159
2.4
0
111 0730-0830
19
(0725)
155
(0815)
136
2.7
78
(0725)
143
(0815)
65
1.3
0
7/2 0917-1000
117
(0915)
163
(0940)
Jfi
U
125
(0915)
159
(0955)
Ji
M
0
Average - double
strength
125
3.0
85
Based on 5-mlnute average emission data.
1.;
Emissions activity period corresponding to tar excavation and movement, taking Into consideration time lags In the system.
Start of corresponding 5-mlnute average period.
-------
are measured at the scrubber. The five-minute data sets used for this section is
presented in Table 27. The 5-minute data sets show the general trends of the
enclosure gas concentrations. Foam use refers to the temporary foam applied during
these operations; stabilized foam was only applied at the completion of a day's activi-
ties.
Foam Use During Mud Excavation and Movement
The data in Table 24 show S02 and THC concentration changes during mud
excavation with and without foam application and during mud movement without foam
application; no data are available for mud movement with foam application. The table
shows that no significant S02 emissions were observed for either mud excavation or
movement.
Small increases were recorded in THC concentrations during these operations.
These were likely due both to THC emissions from operating equipment diesel engines
and to emissions from mud waste. Because of the limited number of comparison
periods and the low emission levels recorded for excavation with and without foam, no
substantial foam control effectiveness conclusions can be drawn.
Foam Use During Tar Excavation and Movement
Table 25 summarizes the available data for tar excavation and movement both
with and without foam use. During tar movement periods, both single- and double-
strength foams were applied; these periods are grouped together in the table. The
concentration change rate data in the table show considerable variability, especially for
S02. While limited with respect to variability and number, the comparison periods do
suggest trends in foam control effectiveness based on the average concentration
change rates calculated. For tar excavation, use of single-strength foam resulted in a
73 percent reduction in the average S02 change rate and a 65 percent reduction in
the average THC change rate. Other factors being equal, the concentration change
rate is directly proportional to the waste emission rate. Use of single-strength foam
during tar movement operations resulted in a 50 percent reduction in the average S02
change rate and a 55 percent reduction in the average THC change rate.
Increasing the foam concentration to double strength resulted in a 79 percent
reduction in average S02 change rate and a 73 percent reduction in the THC change
rate. No data are available for tar excavation with double-strength foam.
Foam Use During Char Excavation and Movement
Because of the high emissions potential expected and observed for char exca-
vation and movement, these operations were always conducted with foam being ap-
plied. As a result, no data are available for char operations without foam and, hence,
142
-------
TABLE 27. FIVE-MINUTE DATA AVERAGES
Time
SO,
SO,
THC
THC
Date
(24-h 5
Inlet
Outlet
Inlet
Outlet
06/08/90
10
20
0.1
0.4
12.3
4.0
Mud excavation without foam (1021
06/08/90
10
25
0.2
0.4
14.2
4.6
- 1037)
06/08/90
10
30
0.2
0.4
15.9
3.8
06/08/90
10:35
0.3
0.4
17.6
5.6
06/09/90
10:30
0.0
0.0
6.3
3.5
Mud excavation with foam (1033 -
06/09/90
10
:35
0.1
0.0
6.3
3.8
1037)
06/09/90
13
:45
0.1
0.0
13.3
4.6
Tar excavation without foam (1345
06/09/90
13
:50
0.2
0.0
16.0
4.5
- 1414)
06/09/90
15
:55
1.0
0.0
19.0
4.8
06/09/90
14
:00
4.3
0.0
23.9
5.9
06/09/90
14
:05
14.9
0.0
29.5
6.2
06/09/90
14
:10
27.0
0.0
34.6
7.5
06/09/90
14
15
41.6
0.0
40.0
8.1
Tar excavation with foam (1414 -
06/09/90
14
.20
68.8
0.0
46.6
10.9
1449)
0B/09/90
14
.25
75.2
0.0
49.9
11.8
06/09/90
14
•30
74.9
0.0
49.7
12.1
06/09/90
14
.35
76.2
0.0
61.1
14.1
06/09/90
14
40
124.1
0.0
81.7
18.8
06/09/90
14
45
135.9
0.0
91.6
23.0
06/12/90
14
05
0.3
0.2
23.0
11.2
Tar excavation without foam (1408
06/12/90
14
10
0.4
0.2
27.5
11.9
1433)
06/12/90
14
15
5.8
0.2
42.2
13.7
06/12/90
14
20
18.5
0.2
46.7
15.3
06/12/90
14
25
16.3
0.3
55.2
16.1
06/12/90
14
30
262.4
0.2
106.0
21.3
Tar excavation with foam (1433 -
06/12/90
14
35
630.5
0.3
172.0
32.7
1450)
06/12/90
14
40
633.8
0.3
96.4
36.8
06/12/90
14
45
641.5
0.2
305.6
49.0
06/12/90
14
50
1000
0.2
253.3
50.0
06/13/90
16:05
152.1
0.5
104.2
34.3
Tar excavation with foam (1606 -
06/13/90
16
10
167.0
0.5
113,9
35.8
1700)
06/13/90
16:15
178.6
0.6
117.9
37.3
06/13/90
16:20
201.5
0.6
120.3
39.4
06/13/90
16:25
210.9
0.6
125.9
40.9
06/13/90
16:30
206.8
0.7
126.0
41.6
06/13/90
16:35
215.7
0.8
131.6
44.3
06/13/90
16:40
230.2
0.7
138.2
46.7
06/13/90
16:45
291.4
1.0
150.9
49.5
06/13/90
16:50
327.0
1.7
150.9
51.3
06/13/90
16:
55
333.3
3.1
150.8
51.9
06/13/90
17:00
341.7
5.0
154.7
52.9
06/14/90
11:10
107.2
0,0
139.5
69.6
Tar movement with foam (1115 -
06/14/90
11:
15
123.4
0.0
148.9
70.4
1200)
06/14/90
11:
20
132.6
0.0
153.8
73.2
06/14/90
11:
25
141.5
0.0
159.3
75.3
06/14/90
11:
30
149.4
0.0
168.1
75.7
06/14/90
11:
35
164.8
0.0
175.6
76.4
06/14/90
11:
40
177.1
0.0
179.7
79.5
06/14/90
11:
45
182.9
0.0
180.1
80.6
06/14/90
11:
SO
183.4
0.0
186.1
80.7
06/14/90
11:
55
182.9
0.0
192.8
83.3
(continued)
143
-------
TABLE 27 (continued)
Date
Time
(24-h)
SO,
Inlet
so,
Outlet
THC
Inlet
THC
Outlet
06/14/90
13
25
118.6
0.0
135.9
76.1
06/14/90
13
30
123.4
0.0
143.1
74.9
06/14/90
13
35
129.7
0.0
154.5
76.5
06/14/90
13
40
130.1
0.0
163.7
79.6
06/14/90
14
35
12?.?
0.1
168.9
82.5
06/14/90
13
50
124.?
0.0
172.3
86.5
06/14/90
13
55
122.4
0.1
175.8
88.5
06/14/90
14
00
122.?
0.1
172.6
91.2
06/14/90
14
05
120.5
0.0
175.1
93.3
06/14/90
14
10
114.0
0.0
173.1
93.8
06/14/90
14
15
111.6
0.0
173.6
92.6
06/14/90
14
20
110.6
0.1
175.0
95.4
06/14/90
14
25
109.2
0.0
174.2
95.2
06/14/90
14
30
112.1
0.1
179.4
94.8
06/14/90
14
35
104.6
0.1
179.0
94.8
06/14/90
14
40
123.9
0.1
150.7
96.6
06/14/90
14
45
147.5
0.1
210.8
101.0
06/14/90
14
50
143.4
0.1
207.3
131.6
06/14/90
14
55
168.5
0.2
217.9
99.5
06/14/90
15
00
238.6
0.1
246.1
101.8
06/14/90
15
05
333.3
0.1
270.8
108.7
06/14/90
15
10
329.2
0.1
276.1
109.8
06/14/90
15
15
331.4
0.0
464.0
143.6
06/14/90
15
20
564.4
0.1
324.6
139.0
06/14/90
15
25
335.5
0.1
283.2
131.6
06/14/90
15
30
319,3
0.1
272.6
130.4
06/14/90
15
35
306.6
0.1
272.6
128.1
06/15/90
10
35
62.1
0.2
225.8
94.6
06/15/90
10
40
62.7
0.1
226.7
92.4
06/15/90
10
45
62.8
0.2
225.8
91.8
06/15/90
10
50
62.8
0.1
225.8
89.6
06/15/90
10
55
62.3
0.1
225.0
86.8
06/15/90
11
00
63.7
0.2
231.1
89.0
06/15/90
11
05
58.9
0.1
264.7
90.7
06/15/90
11
10
142.4
0.1
270.0
92.4
06/15/90
11
15
295,5
0.1
292.0
95.7
06/15/90
11-
20
608.8
0.1
312.3
99.6
06/15/90
11-
25
612.1
0.2
332.6
105.2
06/15/90
11
30
714.8
0.2
341.4
112.9
06/15/90
11-
35
713.8
0.2
327.3
120.2
06/15/90
11-
40
732.2
0.1
335.2
125.7
06/15/90
11:
45
754.8
0.1
340.5
133.0
06/15/90
11:
50
744.2
0.1
337.0
136.9
06/15/90
11
55
739.9
0.2
332.6
140.2
06/15/90
12-
00
731.7
0.1
325.5
141.9
06/15/90
12,
05
721,1
0.1
322.0
143.6
06/15/90
12.
10
717,7
0.1
319.4
143.0
06/15/90
12
15
711.4
0.1
319.4
144.7
06/15/90
12:
20
Hiss
Miss
Miss
Miss
06/15/90
12:
25
Miss
Miss
Miss
Miss
06/15/90
12.
30
Bad
Bad
Bad
Bad
Tar excavation with foam (1325 -
1535)
Tar excavation with foam (1035 -
1115)
Char excavation with foam (1115 -
1315)
(continued)
144
-------
TABLE 27 (continued)
Time
SO,
SO,
THC
THC
Date
(24-h)
Inlet
Outlet
Inlet
Outlet
06/15/90
12
35
696.5
0.1
314.1
151.3
06/15/90
12
40
681.1
0.1
308.8
149.7
06/15/90
12
45
670.0
0.1
306.1
149.7
06/15/90
12
50
659.9
0.1
306.1
151.3
06/15/90
12
55
655.0
0.1
304.4
151.3
06/15/90
13
00
649.7
0.1
303.5
150.2
06/15/90
13
05
643.0
0.1
301.7
148.6
06/15/90
13
10
634.8
0.1
299.1
147.4
06/15/90
13
15
631.4
0.1
297.3
150.2
06/26/90
07
10
34.8
38.2
82.6
30.6
06/26/90
07
15
35.0
38.4
94.8
32.3
06/26/90
07
20
43.8
38.3
105.7
35.4
06/26/90
07
25
64.9
Bad
120.7
38.9
06/26/90
07
30
73.2
Bad
141.8
43.8
06/26/90
07
35
229.4
Bad
138.5
46.0
06/26/90
07
40
462.2
Bad
139.0
47.0
06/29/90
09
05
316.7
Bad
125.6
37.8
06/26/90
09
10
309.0
Bad
125.7
37.6
06/26/90
09
15
306.1
Bad
130.4
38.2
06/26/90
09
20
321.0
Bad
139.8
39.8
06/26/90
09
25
350.9
Bad
157.3
44.4
06/26/90
09
30
424.6
Bad
232.3
51.0
06/26/90
09
35
474.8
Bad
263.7
58.9
06/26/90
09
40
530.2
Bad
298.3
66.2
06/26/90
09
45
593.2
Bad
326.5
72.0
06/26/90
09
50
569.2
0.5
326.5
75.3
06/26/90
09
55
576.5
0.5
345.3
76.7
06/26/90
10
00
550.9
0.6
321.3
74.2
6/28/90
09
30
22.5
0.2
91.7
46.0
6/28/90
09
35
26.1
0.1
99.5
48.3
6/28/90
09
40
30.1
0.1
108.2
50.0
6/28/90
09
45
32.1
0.0
116.3
53.2
6/28/90
09
50
36.8
0.0
135.3
60.4
6/28/90
09
55
39.2
0.0
126.6
60,7
6/28/90
10
00
40.8
0.0
127.6
62.1
6/28/90
10
05
42.0
0.0
128.1
62.8
6/28/90
10
10
43.2
0.0
128.6
62.1
6/28/90
10
15
44.1
0.0
128.9
61.5
6/28/90
10
20
44.5
0.0
124.7
60.8
6/28/90
10
25
42.2
0.0
124.4
61.3
6/28/90
10
30
11.2
0.0
139.2
64.9
6/28/90
10
35
70.1
0.0
141.3
66.6
6/28/90
10
40
79.3
0.2
144.6
67.5
6/28/90
10
45
82.4
0.1
143.8
65.6
6/28/90
10
50
81.9
0.1
144.1
64.1
6/28/90
10
55
80.7
0.3
143.9
62.3
6/28/90
11
00
72.3
0.2
145.9
61.7
6/29/90
11
20
13.3
0.1
58.8
35.6
6/29/90
11
25
14.3
0.1
63.6
37.2
6/29/90
11
30
22.7
0.3
70.6
38.6
6/29/90
11
35
26.9
0.2
78.8
39.5
6/29/90
11
40
35.0
0.1
103.1
43.9
6/29/90
11
45
82.2
0.1
110.8
47.4
6/29/90
11
50
102.7
0.1
119.7
49.8
Char excavation with foam (1115
1315) (continued)
Char excavation with foam (0710
0744)
Char excavation with foam (0908
1003)
Char movement with foam (0908 -
1003)
Char movement with foam (0930
1100)
Char movement with foam (1123
1215) single strength
(continued)
145
-------
TABLE 27 (continued)
Time
SO,
so,
THC
THC
Date
(24-h)
Inlet
Outlet
Inlet
Outlet
6/29/90
11
:55
151.6
0.6
139.3
53.0
6/29/90
12
:00
162.4
0.1
138.7
56.3
6/29/90
12
: 05
151.1
0.1
126.9
56.8
6/23/90
12
:10
134.0
0.0
125.5
55.8
6/29/90
12
: 15
126.8
0.0
126.7
55.3
6/29/90
14
35
47.2
0.7
91.4
49.0
6/29/90
14
40
49,9
0.7
101.9
49.9
6/29/90
14
45
79.8
0.7
117.8
52.0
6/29/90
14
50
117.6
0,7
133.1
55.5
6/29/90
14
55
145.3
0.7
152.1
59.3
6/29/90
15
00
165.6
0.7
164.4
64.3
6/29/90
15
05
171.1
0.6
169.8
68.1
6/29/90
15
10
179.8
0.6
170.1
69.6
6/29/90
15
15
199.8
0.7
189.6
71.9
6/29/90
15
20
213.8
0.7
240.0
73.1
6/29/90
15
25
211.8
0,7
233.4
74.2
6/29/90
15
30
213,5
0.8
239.6
76.3
6/29/90
15
35
212.8
0.7
249.1
78.6
6/29/90
15
40
205.6
0.8
250.1
80.5
6/29/90
15
45
194.7
0.7
24S.1
82.2
6/29/90
15
50
192.8
0.8
257.4
82.8
6/29/90
15
55
187.0
0.8
247.0
83.4
6/29/90
16
00
157.9
0.7
226.0
80.1
6/29/90
16
05
138.8
0.8
260.0
77.0
6/29/90
16
10
132.8
0.8
181.8
74.4
6/29/90
16
15
129.4
0.8
182.6
73.8
7/02/90
07
00
3.5
0.0
44.5
29.3
7/02/90
07
05
3.6
0.0
47.7
29.8
7/02/90
07
10
4.0
0.0
55.8
31.8
7/02/90
07
15
6.1
0.0
61.5
32.7
7/02/90
07
20
10.9
0.1
71.7
34.1
7/02/90
07
25
18.5
0.0
77.9
36.5
7/02/90
07
30
21.4
0.0
84.1
37.5
7/02/90
07
35
25.7
0.0
90.7
39.8
7/02/90
07
40
35.5
0.2
97.3
42.6
7/02/90
07
45
23.3
0.3
102.1
44.9
7/02/90
07
50
74,2
0.3
109.8
47.4
7/02/90
07
55
95,4
0.2
118.5
50.1
7/02/90
08
35
133.0
0.0
130.8
65.4
7/02/90
OS
40
131.6
0.3
130.8
66.5
7/02/90
08
45
Bad
Bad
Bad
Bad
7/02/90
OS
50
127.7
0.3
136.3
69.0
7/02/90
08
55
124.8
0.4
130.9
69.3
7/02/90
09
00
123.9
0.6
128.2
70.0
7/02/90
09
05
122.2
0.5
127.8
89.6
7/02/90
09
10
118.3
0.4
128.1
69.9
7/02/90
09
15
116.6
0.4
124.5
70.8
7/02/90
09
20
125.8
0.5
133.2
72.6
7/02/90
09
25
147.7
0.4
142.7
76.2
7/02/30
09
30
155.0
0.5
145.7
78.3
7/02/90
09
35
160.3
0.1
149.6
80.4
7/02/90
09
40
162.9
0.1
152.1
82.8
7/02/90
09
45
159.1
0.4
152.6
85.2
7/02/90
09
50
153.3
0.1
154,8
87.0
7/02/90
09
55
147.5
0.0
159.4
88,0
Char movement with foam (1123 -
1215} single strength (continued)
Char movement with foam (1441
1545) single strength
Tar movement with foam (1545
1615) single strength
Tar movement with foam (0700
0730) single strength
Char movement with foam (0730
0830) single strength
Char movement with foam (0917
1000) single strength (6.75)
(continued)
146
-------
TABLE 27 (continued)
Time
SO,
SO,
THC
THC
Date
(24—h)
Inlet
Outlet
Inlet
Outlet
7/02/90
10:00
144.6
0.2
154.9
90.1
7/02/90
10:05
139.5
0.2
153.3
91.2
7/02/90
10
: 10
133.5
0.0
154.6
89.3
7/02/10
10
: 15
122.4
0.0
152.8
91.6
7/02/90
10:20
117.8
0.2
150.1
92.6
7/02/90
10
:25
117.1
0.1
152.8
92.2
7/02/90
10:30
115.0
0.0
162.4
89.9
7/02/90
10:35
111.1
0,0
161.9
90.6
7/02/90
10:40
106.5
0.0
161.1
90.6
7/02/90
10:45
102.2
0.0
160.7
89.8
7/02/90
10
:50
97.6
0.0
161.4
90.7
7/02/90
10:55
93.7
0.0
162.6
89.1
7/02/90
11
:00
88.9
0.0
163.9
89.5
7/02/90
11
:05
83.4
0.4
161.2
90.4
7/02/90
11
:10
74.5
0.0
149.5
88.2
7/02/90
11
:15
70.9
0.0
149.8
86.7
7/02/90
11
:20
66.0
0.0
148.7
85.1
7/02/90
11
:25
62.4
0.2
145.6
82.4
7/02/90
11
:30
59.5
0.0
144.1
79.5
7/02/90
11
:35
Miss
Hiss
Mi ss
Miss
7/02/90
11
:40
Hiss
Mi ss
Mi ss
Miss
7/02/90
11
:45
Miss
Hiss
Mi ss
Miss
7/02/90
11
:50
Hiss
Miss
Mi ss
Miss
7/02/90
11
:55
Bad
Bad
Bad
Bad
7/02/90
13:40
27.5
0.4
95.3
51.8
7/02/90
13
:45
20.0
0.5
127.7
51.6
7/02/90
13
:50
29.4
0.6
133.3
40.8
7/02/90
13
:55
28.2
0.2
132.0
58.0
7/02/90
14
:00
27.0
0.1
132.2
58.2
7/02/90
14
:05
26.3
0.5
133.7
58.1
7/02/90
14
:10
25.3
0.5
137.6
58.7
7/02/90
14
.15
24.3
0.5
140.0
58.9
7/02/90
14
:20
23.9
0.5
142.1
58.7
7/02/90
14:25
21.9
0.5
146.9
59.5
7/02/90
14:30
20.7
0.5
148.8
59.6
7/02/90
14:35
19.8
0.5
150.8
59.9
7/02/90
14
40
18.8
0.5
144.0
59.7
7/02/90
14
45
18.1
0.3
149.4
59.0
7/02/90
14
50
16.9
0.0
137.6
60.0
7/02/90
14
55
16.4
0.0
138.6
62.5
7/02/90
15:00
16.1
0.0
141.7
61.2
7/02/90
15:05
15.9
0.0
143.4
61.0
7/16/90
08
40
0.0
0.5
13.1
28.7
7/1S/90
OS
45
0.0
0.6
16.0
30.2
7/16/SO
08
50
0.0
0.4
18.8
31.4
7/16/90
08
55
0.2
0.4
25.1
31.1
7/16/90
09
00
2.5
0,4
43.6
34,1
7/16/90
09
05
11.7
0.4
64.2
42.6
7/1B/90
09
10
28.7
0.4
70.2
46.2
7/16/90
09
15
37.3
0.4
75.8
48.8
7/16/90
09
20
25.6
0.3
90.7
51.2
7/16/90
09
25
19.0
0.3
88.0
51.2
Mud movement without foam (1003 -
1150)
Mud movement without foam (1346 -
1505)
Tar movement without foam (0842 -
0950)
(continued)
147
-------
TABLE 27 (continued)
Date
Time
(24—h)
SO,
Inlet
SO,
Outlet
THC
Inlet
THC
Outlet
7/16/90
09
:30
30.0
0.4
83.
8
49.5
7/16/90
09
:35
27.1
0.5
81.8
48.5
7/16/90
09
:40
32.0
0.5
96.5
52.4
7/16/90
09
:45
39.4
0.6
117
.4
57.8
7/16/90
09
: 50
40.8
0.5
115
.0
62.1
07/16/90
13
50
7.3
0.4
117
4
88.9
07/16/90
13
55
8.2
0.3
121
.6
91.2
07/16/90
14
00
9.8
0.3
143
4
97.4
07/16/90
14
05
11.6
0.3
146
9
99.8
07/16/90
14
10
12.0
0.3
149
9
101.8
07/16/90
14
15
13.3
0.3
163
7
105.6
07/16/90
14
20
16.0
0.3
168
9
106.4
07/16/90
14
25
17.8
0.3
175
5
106.4
07/16/90
14
30
18.2
0.3
172
9
106.4
07/16/90
14
35
16.3
0.3
166
7
106.1
07/16/90
14
40
15.8
0.3
158
4
103.1
07/16/90
14
.45
14.9
0.2
139
7
102.7
07/17/90
07
20
1.7
0.0
74.5
Bad
07/17/90
07
25
4.8
0.0
94.9
Bad
07/17/90
07
30
38.3
0.0
144
5
Bad
07/17/90
07
35
49,3
0.0
172
0
Bad
07/17/90
07
40
79.2
0.0
188
1
Bad
07/17/90
07
45
85.3
0.0
100
0
Bad
07/17/90
07
50
76.9
0.0
207
6
Bad
07/17/90
07
55
74.5
0.0
219
2
Bad
07/17/90
08
00
81.7
0.0
230
8
Bad
07/17/90
08
05
72.4
0.0
228
2
Bad
07/17/90
08
10
67.5
0.0
224
6
Bad
07/17/90
08
15
62.3
0.0
217
4
Bad
07/17/90
08
20
58.5
0.0
210
3
Bad
07/17/90
08
25
55.0
0.0
203
1
Bad
07/17/90
08
30
52.2
0.0
197
7
Bad
07/17/90
08
35
49.4
0.0
193
3
Bad
07/17/90
08
40
47.0
0.0
188
8
Bad
07/17/90
08
45
44.8
0.0
185
2
Bad
07/17/90
08
50
42.8
0.0
180
7
Bad
07/17/90
08
55
40.2
0.0
176
3
Bad
07/17/90
09
00
38.8
0.0
175
4
Bad
07/17/90
09
05
36.9
0.0
170
9
Bad
07/17/90
09
10
34.7
0.0
170
0
Bad
07/17/90
09
15
33.7
0.0
168
2
Bad
07/17/90
09
20
32.3
0.0
165
5
Bad
07/17/90
09
25
31.4
0.0
165
5
Bad
07/17/90
09
30
38.6
0.0
202
2
Bad
07/17/90
09
35
42.8
0.0
219
2
Bad
07/17/90
09
40
40.7
0.0
233
5
Bad
07/17/90
09
45
42.3
0.0
235
2
Bad
07/17/90
09
50
45.7
0.0
244
3
Bad
07/17/90
09
55
50.4
0.0
256.
8
Bad
07/17/90
10
00
53.5
0.0
257.
7
Bad
Tar movement without foam (0842 -
0950)
Tar movement without foam (1351 -
1445)
Tar movement without foam (0725 -
0745)
Tar movement with foam (0759 -
0815)
Tar movement with foam (0927 -
1030)
(continued)
148
-------
TABLE 27 (continued)
Date
Time
(24—h)
SO,
Inlet
SO,
Outlet
THC
Inlet
THC
Outlet
07/17/90
10:05
58.1
0.0
268.4
Bad
Tar movement with foam (0927 -
07/17/90
10:10
56.7
0.0
259.5
Bad
1030) (continued)
07/17/90
10:15
59.6
0.0
265.7
Bad
07/17/90
10:20
70.7
0.0
277.4
Bad
07/17/90
10:25
69.7
0.0
278.3
lad
Tar movement without foam (1030
07/17/90
10:30
73.0
0.0
278.3
Bad
1045)
07/17/90
10:35
70.3
0.0
271.1
Bad
07/17/90
10:40
68.1
0.0
265.7
Bad
Tar movement with foam (1045 -
07/17/90
10:45
66.7
0.0
263.1
Bad
1115)
07/17/90
10:50
70.4
0.0
266.6
Bad
07/17/90
10:55
72.5
0.0
264.8
Bad
07/17/90
11:00
70.0
0.0
257.7
Bad
07/17/90
11:05
66.4
0.0
246.1
Bad
07/17/90
11:10
64.5
0.0
241.6
Bad
07/17/90
11:15
63.3
0.1
237.1
Bad
149
-------
no foam control effectiveness levels can be established. The data in Table 26 do
show, however, that foam-controlled average S02 and THC concentration change
rates were higher for char excavation than for tar excavation.
With respect to char movement, average S02 concentration change rates were
23 percent lower with double-strength foam than with single-strength foam. Average
THC change rates were 35 percent lower with double-strength foam than with single-
strength foam. Foam-controlled average change rates for both S02 and THC were
higher for char movement than for tar movement.
While the average pollutant concentration change rates just discussed are use-
ful in considering the effects of foam application on waste excavation and movement
operations, both the average and maximum emission rates should be considered in
the design of final remediation systems.
4.4.2 Static Flux Chamber Measurements For Foam Effectiveness Evaluations
As part of the evaluation of vapor-supressing foams as described previously in
Section 3.4.2, flux chamber tests were performed on the two foam types. Since the
flux chamber tests are unique methods for determining the evaluation of emissions
from a variety of waste materials, It was decided to subcontract the flux chamber tests
to a recognized expert in this type of testing. The flux chamber tests were completed
according to the prescribed plan. The flux chamber Tedlar bag samples were
submitted to PAI on June 13, 14, and 15, 1990. The contractor shipped the other flux
chamber stainless steel canister samples to Radian Laboratories in Austin, Texas, on
June 15, 1990.
The samples were analyzed and the air speciation results were reported by PAI
and Radian. The flux chamber tedlar bag sample results from PAI did not meet the
data quality requirements and were found to be invalid as determined by an EPA quali-
ty assurance audit. The stainless steel canister sample results were also audited and
found invalid. These data are not of sufficient quality to include in this report, but the
data will be maintained in the "RAW DATA FILE" previously mentioned in Section 4.3.
4.5 Tar Processing
4.5.1 Use of Portland Cement and Fly Ash as Stabilization Agents
Portland cement and pozzolanic fly ash were originally selected as stabilization
agents to be mixed with tar because of the anticipated viscous, runny nature of this
waste material. As discussed in Subsection 3.2, tar material, as excavated, was in the
form of blocks and chunks that were pliable and plastic, but not viscous or mobile at
enclosure temperatures. Nevertheless, the excavated tar material will require further
size reduction to be used as feed to most thermal treatment devices.
150
-------
Material Processing Rates
Data related to tar processing rates arid total material processing rates for the
10 pug mill tests are summarized in Table 28. As stated previously, all but Runs 4 and
10 were conducted at low idle paddle speed on the pug mill. Most of Run 4 was also
conducted at low idle speed, except during the last minute, when the speed was
increased to about 3/4 of maximum speed. All of Run 10 was conducted at maximum
paddle speed.
TABLE 28. TAR AND TOTAL MATERIAL PROCESSING RATES
Feed quantities, lb
Elapsed
Tar
Total material
Run
Ho.
Tar
Cement
Water
Fly ash
Total
time,
min
Hill
stopcaaes
processing
rate, tons/h
processing
rate, tons/h
1
294
1390
0
0
1684
15
1
0.59
3.37
2
294
2085
351
0
2730
17
0
0.52
4.82
3
294
2085
585
D
2964
17
3
0.52
5.23
4
588
1390
273
0
2251
9
2
1.96
7.50
5
294
1390
312
76S
2762
12
3
0.74
6.91
E
294
695
312
1531?
2833
9
1
0.98
9.44
7
588
695
156
7615
2205
5
0
3.53
13.23
8
588
2085
312
(}
2985
6
0
2.94
14.93
9
588
0
234
3064
3886
14
1
1.26
8.33
10
294
2085
312
0
2691
4
0
2.21
20.18
The data in the table indicate that other factors in addition to paddle speed had
a significant effect on the total material processing rate. As expected, the highest total
material processing rate of 20.2 tons/hour was observed during Run 10. This rate
was more than five times higher than the rate for Run 2, which had nearly identical
conditions except for paddle speed. However, Run 2 was the first run conducted with
tar, and more than the usual amount of time was allowed for the start of tar feeding
and for clearing of material from the mill at the end of the run. A more representative
total material processing rate at low idle speed, but with no stoppages, is given by the
results of Runs 7 and 8 (i.e., an average of 14.1 tons/hour). These data indicate that
increasing the paddle speed from low idle to maximum resulted in a 43 percent in-
crease in the total material processing rate.
It is clear from the data in Table 28 that mill stoppages also had a significant
effect on processing rates. The results from Runs 3 through 9 show that the highest
rates were associated with runs with no stoppages, while the lowest rates were associ-
ated with runs with two or more stoppages.
151
-------
Paddle speed and mill stoppages had similar effects on tar processing rates
during the 10 runs. In addition, the effect of the amount of tar charged to the feed
hopper must be considered in evaluating these data. The highest tar processing rates
of Runs 7 and 8 correspond to low paddle speed, no stoppages, and high tar feed
concentrations. The tar processing rate for Run 10 was lower than those for Runs 7
and 8 (even though the paddle speed was higher) because only half as much tar was
initially charged to the feed hopper. It is also interesting to note that the amounts of
time required for processing the higher tar feed concentrations (in Runs 4, 7, 8, and 9)
were not significantly greater than those for lower tar feed concentrations. This result
indicates that the pug mill operation was not tar-feed-rate-limited during these runs.
Thus, if the tar feed concentration in Run 10 had been doubled, it is reasonable to
expect that the tar processing rate would also have doubled to near 4 tons/hour.
The tar and total material processing rates observed during tar treatment tests
correspond to batch operation of the pug mill. Processing rates are characteristically
low with batch operations, since time must be allowed near the end of the run for the
mill inventory to empty (to avoid contamination with the next batch). In addition, no
product material is generated during the first minute or more of pug mill operation,
since material must travel from the feed hopper to the head of the mill and then fill the
mill to the point where material overflows from the tail-end dam. If continuous opera-
tion were maintained, these allowances for mill inventory purging and startup delays
would be eliminated and the overall processing rate would increase. The ease with
which the pug mill processed the tar and other material during these treatment tests
suggests that continuous operation of the pug mill would have been possible. It is es-
timated that use of continuous operation would have increased processing rates by a
factor of at least 2.
The pug mill sustained no physical damage as a result of the tar treatment
tests. A small layer of cement built up on the paddles and internal surfaces of the mill,
and had to be chiseled off after decontamination. This buildup could be eliminated in
future operations by washing the mill with water at the completion of each processing
period.
Product Characteristics
Another important area of results for the tar treatment tests concerns the physi-
cal characteristics of the treated tar. Table 29 summarizes the characteristics re-
corded for the product material from the 10 pug mill tests: size distribution, flowability,
and angle of repose. Size distribution is important, since it was desired to produce
material that was less than 2 inches in size to facilitate feeding the material to a ther-
mal treatment system. The degree to which product material is free-flowing will dictate
the ease with which it can be transported from the pretreatment area to the final treat-
ment area. The angle of repose will determine the height to which a given quantity of
material can be stacked and the amount of ground space required for storage. The
152
-------
table also summarizes the important feed ratios, namely tar-to-cement-plus-ash
P7(C+A)1, water-to-cement-pl js-ash [W/(C+A)], and ash-to-cement-plus-ash
[A/(C+A)].
TABLE 29. TAR TREATMENT PRODUCT CHARACTERISTICS
Feed ratios
Product size distribution,
wt X
Free
Angle of
Run
No.
T/(C+A)
w/(OA)
A/(C+A)
-2 in.
-1 in.
-0.5 in.
-0.25 in.
flowing,
r/N
repose,
degree
1
0.21
0.00
0.00
100
-
95
80
Y
30
2
0.14
0.17
0.00
100
-
95
80
Y
45
3
0.14
0.28
0.00
95
-
-
-
H
10-20
4
0.42
0.20
0.00
90-85
-
-
-
Y
-
5
0,14
0.14
0.36
95
-
90
80
Y
30
6
0.13
0.14
0.69
95
-
90
80
Y
30
7
0.40
0.11
0.52
95
90
-
85
Y
45
8
0.28
0.15
0.00
99
95
-
90
Y
45
9
0.19
0.08
1.00
95
90
-
85
Y
45
10
0.14
0.15
0.00
99
95
-
90
Y
45
The table shows that the product size distribution did not change greatly during
the 10 tests despite the change in feed ratios indicated. Essentially, all the product
material was less than 2 inches; in size for Runs 1, 2, 8, and 10. During these runs,
the T/(C+A) ratio ranged from 0.14 to 0.28; the W/(C+A) ratio ranged from 0 to 0.17;
and the A/(C+A) ratio was 0 to 1.0. The most significant difference among these
products was the lower angle of repose observed during Run 1 with no water added.
The only run which resulted in a product that was not free flowing was Run 3,
conducted at a W/(C+A) ratio of 0.28. This product was described as a "watery
paste-like mass" with a repose angle of 10 to 20 degrees. All other runs, conducted at
W/(C+A) ratios of 0.2 or less produced product material that was free flowing. Al-
though this material was dry to the touch, it would cake if squeezed by hand; when
dropped, the caked material would break apart. It was also observed that the product
material would set overnight into a more rigid mass. This indicates that the cementi-
tious reactions expected when mixing water with cement and pozzolanic fly ash were
slow and required approximately 16 hours to produce noticeable setting.
These results suggest that a tradeoff may apply during final remediation when
adding water to cement and ash during tar treatment. Adding water reduces dusting
in the mill and makes the product more stackable, thereby requiring less ground
space for storage. However, adding water may also cause the product material to set
after several hours or days, which could make it more difficult to move the material
from a stockpile or storage hopper.
153
-------
The use of ash in place of cement appeared to increase the amount of > 2-inch
material slightly (from 0 to 1 percent with cement alone to about 5 percent with ash or
ash/cement mixtures). Again, this indicates a tradeoff may occur in the final design of
such a system, since ash is generally less costly than cement but additional costs will
be incurred to recycle the > 2-inch material.
It is of interest to note that the product size distribution did not change marked-
ly during the final run when the paddle speed was increased to its maximum. This in-
dicates that acceptable product material could be obtained at the highest throughput
rate for the pug mill as configured. Use of continuous operation of the pug mill in-
stead of batch operation would not be expected to change this result.
The greatest amount of > 2-inch product material was produced during Run 4
when the paddle speed was increased at the end of the run. Visual observation
showed that this change caused more agitation of materials in the mill and more
"jumping" on the part of tar lumps in the middle and tail end of the mill. As a result,
oversized lumps bumped along the top of the paddles and exited the mill over the
dam without being further reduced in size. This behavior was not observed during
Run 10, also conducted at high paddle speed. In Run 10, tar lumps were broken up
at the head of the mill as they fell off the feeder belt. These results indicate that the
paddles should be maintained near a constant speed during tar processing
operations. If paddle speed increases are required, they should be made after the
inventory in the mill has been largely cleared. After the paddle speed change has
been made, tar processing can resume.
Physical inspection of the tar pellets processed through the pug mill showed
that the exterior of the pellet was coated with a thin (1-millimeter to 1/4-inch) layer of
cement and ash (if present). The middle of the tar pellet did not contain a significant
amount of cement or ash. A processed pellet that was broken into two pieces is
shown in Figure 67. These conditions suggest that large tar lumps from the feed
hopper were broken into smaller lumps, or pellets, by the shearing action of the pad-
dles in the pug mill. These smaller pellets, in turn, were coated by cement or ash as
they passed through the mill. The amount of water added to the mill may have been a
factor in determining the thickness of the cement/ash coating that built up on the tar
pellets.
Analytical Results
The proximate and ultimate analyses reported in Subsection 4.7 indicate that
the volatiles, fixed carbon, and heating value all declined between raw tar and treated
tar. (The treated tar sample came from the Run 10 product material.) These declines
would be expected, since the water and cement added to the tar in the pug mill act to
reduce these values and increase the moisture and ash contents, all other things
being equal. The data in Table 30 show that the as-received treated tar proximate
154
-------
Figure 67. Treated tar pellet broken into two pieces.
TABLE 30. PROXIMATE ANALYSIS COMPARISONS AMONG RAM AMD TREATED TAR
Proximate
analysis
Raw tar
Treated tara
Projected
As received
MAFb
As received
MAF
treated
tar
Moisture, %
Volatiles, %
Fixed carbon, %
Ash, %
11.6
69.9
16.9
1.6
80.5
19.5
8.1
14.0
2.0
75.9
87.5
12.5
12.9
7.7
1.9
77.6
Heating value,
Btu/lb
9160
10,553
2200
13,750
1000
a From Tar Treatment Run No. 10
b Moisture- and ash-free basis.
analysis agrees well with the analysis projected from the raw tar analysis and the feed
quantities specified in Table 29.
The volatile and fixed carbon contents of raw and treated tar can also be calcu-
lated on a moisture-and-ash-free (MAF) basis to eliminate the effects of the added
water and cement, as shown in Table 30. These data surprisingly indicate that the
volatile content of the tar increased somewhat between raw tar and treated tar, while
the fixed carbon content decreased. This result is reinforced by the increase in MAF
heating value from 10,553 Btu/lb in raw tar to 13,750 Btu/lb in treated tar. An
analogous trend was seen in the hydrogen/carbon (H/C) ratio from the ultimate analy-
ses; the H/C ratio increased from 0.08 in raw tar to 0.50 in treated tar. These results
suggest that no devolatilization of the tar took place in the pug mill during tar treat-
ment.
155
-------
4.5.2 Use of Cement and Fly Ash for Emission Control During Tar Treatment
A secondary effect of using portland cement and fly ash during tar treatment in
the pug mill was the substantial reduction of S02 emissions below expected levels.
The emissions data in Table 31 show that the S02 concentrations in the enclosure air
rose by a maximum of about 5 ppm during the tar treatment tests and, in most cases,
remained essentially unchanged. This was surprising in view of the shearing action of
the pug mill paddles, which had the effect of exposing new tar surface area for devola-
tilization, including S02. Such low emission increases were in contrast to periods
during the same operating days in which the digging of tar out of storage bins alone
resulted in S02 emission increases of more than 40 ppm. Much less new tar surface
area was exposed during the latter operation than the former.
Based on the appearance of the treated tar pellets, it is speculated that the
cement and/or fly ash in the pug mill effectively sealed exposed tar surfaces before
any significant devolatilization could occur. The data in the table indicate there was no
significant difference between cement, fly ash, and cement/ash mixtures in limiting S02
emissions during tar treatment.
TABLE 31. EMISSION CHANSES DURING TAR TREATMENT TESTS
Preliminary THC Preliminary SO, emissions,
Feed ratios Elapsed emissions, ppm ppm
Run
No.
T/tC+A)
w/(c+a)
A/EC+A)
time,
min
Beginning
End
Change
Beginning
End
Change
1
0,21
0.00
0.00
15
20
90
70
<10
<10
0
2
0.14
0.1?
0.00
17
145
196
51
20
15
-5
3
0.14
0.28
0.00
17
142
164
22
<10
<10
0
4
0.42
0.20
0.00
9
168
198
30
<10
<10
0
5
0.14
0.14
0,36
12
240
320
80
<10
15
5+
6
0.13
0.14
0.69
9
300
420
120
15
15
0
7
0.40
0.11
0.52
5
148
260
112
6
10
4
8
0.28
0.15
0,00
6
260
340
80
7
6
-1
9
0.19
0.08
1.00
14
240
300
60
5
4
-1
10
0.14
0.15
0,00
4
77
110
33
1
1
0
Emissions of THC showed an increase during tar treatment operations ranging
from about 20 to 120 ppm. However, another major source of THC emissions during
pug mill operation was the mill's diesel engine. Based on the devolatilization behavior
discussed previously, it is suspected that the THC emission increases observed during
treatment tests were a result of the engine operation. Attempts to correlate THC
emission increases with pug mill operating duration (as measured by elapsed times)
did not produce any significant relationships.
156
-------
4.5.3 Potential Use of Pug Mil! and Solidification Agents for Char and Mud Processing
Although no treatment tests were conducted with char and mud in the pug mill,
the ease with which the mill processed the tar suggests that the same treatment ap-
proach could be used for char and mud. Excavated char and mud lumps would likely
require less power to process in a pug mill, since they will be easier to break up than
the plastic-like tar lumps. The presence of cement and/or fly ash in the mill should act
to limit S02 and THC emissions during processing, as was observed during tar treat-
ment. However, because of the high emission flux rates observed for char, it is not
certain that char treatment emissions will be controlled to the same degree as reported
for tar.
4.6 Equipment-Related Operational Factors
The equipment-related operational problems encountered during excavation and
processing of the McColl wastes are summarized in Table 32 according to the date on
which they occurred, the nature and cause of the problems, and the action taken to
remedy or address it during the trial excavation. The table is a synopsis of the opera-
tional factors discussed in the Appendix A operations log summaries. A total of 12
separate operational factors are listed (two occurred more than once). All of these
either caused work to be stopped or caused significant slowdowns in work progress.
Review of the nature of the factors listed in the table shows that they can be
grouped into three general categories: 1) mechanical equipment failures, 2) waste
emissions-related problems, and 3) engine emissions-related problems. Each of these
categories will be discussed in this section with respect to the action taken during the
trial excavation and the remedies recommended for full-scale remediation, assuming
the same general approach will be employed for excavation and treatment of waste
material.
4.6.1 Mechanical Equipment Failures
The general approach recommended for problems related to mechanical equip-
ment failures is to conduct frequent inspection of operating equipment, follow a regular
maintenance program, and maintain an adequate supply of spare parts. The inspec-
tion and regular maintenance activities will act to spot and address mechanical prob-
lems before they impact work progress. An adequate supply of spare parts will allow
timely repair of those mechanical systems that will inevitably fail during an operation of
this type. These approaches would likely have eliminated the trial excavation problems
related to the missing check valve, foam concentrate feed line pinching, and leaks in
Level A PPE suits.
The foam stabilizer became contaminated from the use of the foam mixing
trailer on another training assignment before being used at McColl without flushing of
157
-------
TABLE 32. OPERATIONAL FACTORS AFFECTING TRIAL EXCAVATION
Date
{1990)
Operational problem
Cause
Action taken
6/8 fowl leak at mixing trailer,
6/8 Foam nozzles and hose plugged
with stabilized foam.
6/12 High SO, and THC levels inside
enclosure caused work stop-
page.
6/13 Lou visibility inside enclo-
sure caused work stoppage.
6/15 Low ventilation airflow rate
through wet scrubber.
6/21 Poor-quality stabilized foam.
6/22 Level A suits failed pressure
test.
6/22 Observation camera would not
tilt up and down.
6/26 High SO, levels Inside enclo-
sure caused work stoppage.
6/2? High tswperatures Inside en-
closure and related heat
stress concerns caused work
stoppage.
6/28 Steering and braking control
problems with loader/backhoe.
6/28 Mud layer on floor of enclo-
sure caused traction diffi-
culties for wheel-mounted
loader.
Check valve in feed line missing.
Contaminated batch of foam stabilizer.
High waste emission rates. Low foam vapor
suppression effectiveness.
Oiesel engine particulate emissions from
trackhoe and loader/backhoe obscured visi-
bility. Oust from excavation operations.
High pressure drop across scrubber caused
by solids buildup in scrubber solution.
Feed line from foam concentrate storage
tank pinched, shutting off supply of foam
to mix system.
Suits had not been used in years and had
developed leaks in storage.
Several teeth on plastic gear inside tilt
unit were sheared off.
High waste emission rates. Low foam vapor
suppression effectiveness.
High ambient temperatures, solar gain,
exhaust heat from diesel engines, and
Level A PPE.
Attack of steering and braking system by
dirt, dust, and waste emissions.
Breakdown of foam and accumulation of sta-
bilized-foam purge water.
Installed missing check valve.
Drained contaminated stabilizer from trailer tank.
Replaced with new stabilizer.
Stopped work. Applied stabilizer foam. Waited for
ventilation system to reduce pollutant concentrations
overnight.
Stopped work. Malted for ventilation system to clear
enclosure air overnight.
Blew down scrubber solution and replaced with fresh
solution. Initiated twice/day checking of particulate
filters upstream of scrubber and changing of filters
when loaded.
Feed line tied off and shielded to prevent further
pinching.
Patched leaks in old suits and ordered new suits.
Tilt unit replaced with new one.
Stopped work. Applied stabilized foam. Waited for
ventilation system to reduce pollutant concentrations
overnight.
Stopped work. Waited until the following day to re-
gime work.
Steam cleaning of loader/backhoe steering and braking
systems corrected control problems.
Replaced wheel-mounted loader with track-mounted Bob-
cat.
(continued)
-------
TABLE 32 (continued)
Date
(1990)
Operational problem
Cause
Action taken
7/11 Low ventilation airflow rate
through wet scrubber.
7/25 Uhlte powdery buildup on alu-
minuit I-beams of enclosure
structure.
High pressure drop across scrubber. Cause
of high pressure drop not Identified.
Uhlte powder speculated to be aluminum
sulfate, caused by the attack of sulfuric
acid {formed by SO, and moisture) on alu-
minum surfaces.
Scrubber liquid recirculation rate lowered to reduce
liquid hold-up In tower, thereby reducing pressure
drop arid increasing airflow to normal levels.
None required during trial excavation. Beams manually
cleaned of deposits offslte after demobilization using
conrnercial phosphoric acid-based cleaner.
-------
the material in the tanks. This points out the need for fresh, high-quality materials to
be used for the full-scale remediation and the requirement that all equipment be ade-
quately cleaned and flushed before coming on site. It is likely that a new, dedicated
foam mix trailer would be used for full-scale remediation if continued use of foam for
vapor suppression were called for.
Solids buildup in the wet scrubber solution was addressed during the trial exca-
vation by more frequent inspection and changing of the dust filters upstream of the
scrubber. The filters used during this program were common household furnace fil-
ters. For full-scale remediation, an industrial-type particulate collection system is rec-
ommended, should be upstream of other ventilation air pollution control equipment.
This system would likely consist of a wet venturi-type scrubber or an electrostatic pre-
cipitator (ESP). Alternatively, a wet scrubber or ESP designed for both particulate and
S02 removal could be employed.
The problems observed with steering and braking of the backhoe/loader would
be mitigated by the emission reduction procedures discussed later in this section.
However, it may also be desirable to enclose sensitive systems to isolate them from
dirt, dust, and other emissions. Enclosure of the observation camera and zoom lens
during the trial excavation was successful in avoiding operating problems with these
sensitive devices.
Finally, the traction difficulties encountered by the wheel-mounted loader on the
muddy floor of the enclosure were due to the chemical breakdown of temporary and
stabilized foam by the char and tar wastes and the accumulation of purge water from
stabilized foam applications. At the completion of stabilized-foam applications, foam
and water had to be purged from the delivery lines to prevent set-up of foam in the
system; purging was not required after temporary foam applications. The result of the
foam breakdown and purge water accumulation was a layer of mud and foam on the
floor that was 6 to 12 inches deep. Besides making traction difficult for the loader, the
mud also prevented the free movement of tar and waste bins about the enclosure
(due to sinking) and made personnel footing quite uncertain.
For the trial excavation, the problem was addressed by substituting a track-
mounted Bobcat for the wheel-mounted loader. Because of the Bobcat's smaller
bucket size, this change reduced the waste-moving productivity of operating person-
nel. In addition, personnel took more care in directing the stabilized-foam purge water
into 55-gallon drums rather than onto the enclosure floor.
If foam application is retained for a full-scale remediation, it may be necessary
to devise a drainage system around waste-handling areas to drain off accumulated
water. In addition, portable blowdown tanks should be located near foaming opera-
tions to catch purge water and to remove it periodically from the enclosure. Depend-
160
-------
ing on the success of these systems, track-mounted equipment may be required for
material-handling operations.
4.6.2 Waste Emission-Related Problems
The high emission levels of S02 and THC measured for the tar and char waste
materials during the trial excavation caused work stoppages. These were due to
health and safety concerns, interfered with equipment steering and braking systems,
and the formation of aluminum sulfate on the enclosure support I-beams. As
discussed else-where in this report, the emission flux rates from char and tar wastes
were higher than expected during the trial excavation and the vapor suppression ability
of the foam used was lower than expected. Since the ventilation air flow rate was
fixed, this system was not able to provide enough fresh air to keep pollutant
concentrations below design levels.
To address these concerns in the full-scale remediation, a more effective vapor
control system would be desirable. Alternative formulations for foams should be inves-
tigated, especially those which contain chemical bases and have the potential to
chemically bond with the surface of the acidic McCoIl waste. Improvements to foam
application techniques such as leveling of waste surfaces to minimize foam sloughing
may be possible. Alternatively, other vapor-suppression systems should be evaluated,
including the use of a lime or limestone slurry such as has been applied in coal mines
for dust suppression.
However, even with improvements, the vapor-suppression system cannot be
expected to provide complete control of waste emissions due to the dynamic condi-
tions of waste excavation and movement To maintain pollutant concentrations inside
the enclosure below Immediately Dangerous to Life and Health (IDLH) or Permissible
Exposure Limit (PEL) levels, a larger air-ventilation system will in all likelihood be called
for. This will require a larger fan, air pollution control devices (APCDs), and associated
ducting. By generating a higher airflow rate, the larger ventilation system would
provide for more frequent turnover of the air inside the enclosure, and hence lower
pollutant concentrations.
4.6.3 Engine Exhaust-Related Problems
During the trial excavation, diesel engines were operated on the trackhoe, back-
hoe/loader, Bobcat, and pug mill. The emissions from these engines directly contribu-
ted to a work stoppage due to heat stress, to a work stoppage due to low visibility,
and to work stoppages related to high THC levels. The exhaust gases from diesel
engines add heat, particulate matter, and hydrocarbon species to the enclosure air
(S02 contributions were likely small in light of the low amount of sulfur in diesel fuel).
161
-------
One approach for the full-scale remediation would be to use electric engines in
place of diesel engines. The pug mill could have been equipped with an electric
engine for the trial excavation except that the electrical demand requirements would
have exceeded the supply available on site. Further work shojld be conducted on the
size of the pug mill required for full-scale remediation and associated power require-
ments. In addition, it may be possible to use an electrically-powered gantry crane sys-
tem inside the enclosure for movement of material and for excavation of some or all of
the waste materials.
If diesel engines on some operating equipment cannot be feasibly eliminated for
the full-scale remediation, a system for directly venting engine exhaust to the APCDs
should be investigated. It may be possible to suspend movable ducting from the
enclosure ceiling that could be connected to engine exhausts and directly transport
exhaust gases to the APCD system without their entering the enclosure air. This
would be easiest to accomplish on equipment that did not move about greatly within
the enclosure (e.g., pug mill or trackhoe). For more mobile equipment, it might prove
feasible to direct exhaust gases through a filter, carbon canister, and water cooler
system mounted directly on the machine. This approach would likely require frequent
changing of the filter media, carbon, and water to maintain effectiveness.
4.7 Waste Characterization
Samples of raw mud, raw tar, raw char, and treated tar were taken during this
project for subsequent analyses. The treated tar sample was from Run 10. Previous
samples from the L-4 sump were taken from borings made at this site and, where
appropriate, these earlier analytical data are presented for comparison.
4.7.1 Proximate and Ultimate Analyses
Table 33 presents the results of the proximate and ultimate analyses performed
on the four sets of samples. In all of these analyses, some volatile matter was driven
off of the sample when it was heated to 105 "C to determine the moisture fraction.
Also, the moisture content of these samples may be higher than undisturbed waste
due to tiie foam vapor suppressants sprayed on the waste, which contained large
amounts of water. Results for these analyses are included in Appendix F.
These data show the mud fraction of this waste (the upper layer) consisted
largely of inorganic, noncombustible material with an ash content of 82.9 percent and
a heating value of less than 500 Btu/lb. The raw tar sample contained a high percent-
age of combustible material and had a heating value of over 9000 Btu/lb, an ash con-
tent of less than 2 percent, and a high sulfur content of 10.6 percent. The treated tar
sample contained cement dust and fly ash (low-sulfur, high-ash components), and the
addition of this material decreased all of the combustible parameters and increased
162
-------
the ash value. Raw char has a fairly high ash level of about 55 percent, a sulfur level
of 4.5 percent, and a heating value of 5200 Btu/lb.
TABLE 33. TRIAL EXCAVATION WASTE SAMPLE ANALYSIS - PROXIMATE AND ULTIMATE
Parameter
Raw mud
Raw tar
Treated tar
Raw char
Proximate
Moisture, %"
13. J!
11.6
8.1
21.2
Volatiles, %
3.7
69.9
14.0
20.1
Fixed carbon, %
0.2
16.9
2.0
4.0
Ash, %
82.9
1.6
75.9
54.7
Btu/1b
<500
9160
2200
5200
Ultimate
Carbon, %
1.1
51.1
2.4
8.6
Hydrogen, %
2.5
5.4
2.1
3.7
(1.0),:
(4.1)c
(2 -1 )cc
(1.3)c
Sulfur, %
0.8
10.6
3.6
4.5
Nitrogen, %
0.1
0.1
0.1
0.1
Oxygen, %d
12.5
31.1
15.8
28.3
(0.8)<:
(20.8)c
(8.6)c
(9.5)c
a Moisture includes some volatiles lost at 105*C.
k Fixed carbon determined by difference in proximate analysis.
c Excludes hydrogen and oxygen contained in water.
^ Oxygen determined by difference in ultimate analysis.
Previous data on the proximate and ultimate analyses of waste from the L-4
sump were obtained from core samples at various depths but were not specified as
mud, tar, or char (GP11989). Table 34 presents these data, which are shown here for
comparison purposes. Though not exactly comparable, the bore sample at 15 to 16
feet deep is probably in the tar layer and, except for the ash and fixed carbon, com-
pares approximately with the samples from the trial excavation. The deeper samples
are probably in the char layer and again compare with the values obtained during this
study except for the sulfur contents, which were much higher in the bore samples.
163
-------
TABLE 34. BORE SAMPLES - PROXIMATE AND ULTIMATE ANALYSES
(AS-RECEIVED BASIS)
Depth, ft
15-16 18-19 24-25
Proximate
Moisture, %
10.7
14.0
11.5
Volatiles, %
66.0
31.8
32.5
Fixed carbon, %
8.7
4.9
2.2
Ash, %
14.7
49.4
53.8
Btu/lb
9057
3503
1179
Ultimate
Carbon, %
44.5
19.5
8.6
Organic hydrogen, %
5.9
2.2
1.8
Sulfur, %
14.5
10.4
10.4
Nitrogen, %
0.2
0.2
0.2
Oxygen, %
7.4
4.4
13.7
4.7.2 Selected Chemical Analyses
A common indicator for petroleum waste is the concentration of benzene,
toluene, ethylbenzene, and xylene (BTEX). Samples of the McColl waste were
analyzed for these compounds, as presented in Table 35. These data show that the
tar fraction of this waste contains the highest levels of these compounds and that the
mud layer contains only a relatively small portion of these compounds.
TABLE 35. BTEX CONTENT OF HASTE
iml
Compound
Mud
Tar
Char
Benzene
<0.7
240
97
Toluene
1.5
580
150
Ethyl benzene
0.9
140
35
Total xylenes
8.6
910
220
Previous analytical data on bore hole samples from the L-4 sump did not yield
values that were this high. The maximum values measured were 11, 24, 8, and
41 ppm for benzene, toluene, ethylbenzene, and xylene respectively (CH2M Hill 1989;
Table D-6).
4.7.3 Leach Procedure Tests
Toxicity characteristics of the excavated waste and the treated tar were deter-
mined by the Toxicity Characteristic Leaching Procedure (Federal Register 1986) and
164
-------
the California wet test (Caiif. Admin. Code 1985). In the EPA TCLP test an acetic acid
based leaching solution is mixed with waste which has been crushed to pass through
a 9.5 mm screen at a 20:1 ratio. The leaching is performed in a zero head space
extractor for 18 hour?. The California wet test uses a 0.2 M sodium citrate extraction
solution mixed with waste that is crushed to pass through a 2 mm screen at a 10:1
ratio. Leaching is carried out over 48 hours. Analytical data for these tests, including
blank values and matrix spike recovery data, are presented in Appendix F.
Metals
The TCLP and California wet test results for designated metals are presented in
Table 36. These data show fairly low levels of these metals and all of the TCLP results
are well below regulatory levels. Some chromium was evident in all the samples
except mud. Lead and barium were also present in the treated tar, possibly from the
cement/fly ash additive.
A more extensive list of metals is analyzed in the California wet test. A higher
detection limit, due to dilutions required by the high sodium levels in the extract, was
evident in these tests. Chromium, vanadium, and zinc were found in all of these
sample extracts. Antimony, copper, lead, and molybdenum were also found in the
treated tar sample extract.
Organic Compounds
Designated volatile organic compounds were analyzed as specified in the EPA-
TCLP procedure. Semivolatile compounds were also analyzed by the EPA-TCLP and
in the California wet test extracts. The California wet test procedure does not lend
itself to volatile organic compound analyses since no zero head space extraction
apparatus is used.
Benzene and 2-butanone (MEK) were detected in the volatile organic com-
pound analyses in all sample extracts except mud as shown in Table 37. The ben-
zene concentration in the treated tar sample was below the regulatory level of 500
Mg/liter and was found at the 1000 pg/liter level in tar and char extracts. Matrix spike
recoveries for these compounds were generally good and within prescribed limits.
Semivolatile compounds were not detected in any extracts from the EPA-TCLP
and California wet test except for pyridine found in the char California wet test extract
at a concentration of 1100 Mg/iter (Table 38).
165
-------
TABLE 36. SAMPLE EXTRACTION TESTS FOR METALS
(ing/1 iter)
Metal
Mud
Tar
Treated tar
Char
Regulatory 1
TCLP analysis for metals
Arsenic
<0.01
<0.01
<0.01
<0.01
5.0
Barium
<0.2
<0.2
0.6
<0.02
100
Cadmi urn
<0.005
<0.005
<0.005
<0.005
1.0
Chromium
<0.01
0.09
0.15
0.23
5.0
Lead
<0.05
<0.05
0.12
<0.05
5.0
Mercury
<0.0002
<0.0002
<0.0002
<0.0002
0.2
Selenium
<0.005
<0.005
<0.005
<0.005
1.0
Silver
<0.01
<0.01
<0.01
<0.01
5.0
California
wet test for
metals
Antimony
<0.6
<0.6
0.9
<0.6
15
Arsenic
<0.25
<0.25
<0.25
<0.25
5,0
Barium
<2
<2
<2
<2
100
Beryl 1i urn
<0.05
<0.05
<0.05
<0.05
0.75
Cadmium
<0.05
<0.05
<0.05
<0.05
1.0
Chromium
0.2
0.3
1.1
0.6
560
Cobalt
<0.5
<0.5
<0.5
<0.5
80
Copper
<0.25
<0.25
0.44
<0.25
25
Lead
<0.5
<0.5
0.8
<0.5
5.0
Molybdenum
<0.2
<0.2
0.3
<0.2
350
Mercury
<0.002
<0.002
<0.002
<0.002
0.2
Nickel
<0.4
<0.4
<0.4
<0.4
20
Selenium
<0.125
<0.125
<0.125
<0.125
1.0
Silver
<0.1
<0.1
<0.1
<0.1
5
Thallium
<0.25
<0.25
<0.25
<0.25
7.0
Vanadium
0.9
1
0.9
1.4
24
Zinc
1
1
0.2
1.6
250
166
-------
TABLE 37, TCLP ORGANIC COMPOUNDS pig/liter)
Parameter Mud Tar Treated tar Char Regulatory limits
Volatile organic compounds
Vinyl chloride
<13
<100
<10
<50
200
1.1-Dichloroethene
<5
<50
<5
<25
700
Chloroform
<5
<50
<5
<25
6,000
1,2-Dichloroethane
<5
<50
<5
<25
500
2-Butanone
<10
350
29
310
200,000
Carbon tetrachloride
<5
<50
<5
<25
500
Trichloroethene
<5
<50
<5
<25
500
Benzene
<50
<5
<25
100,000
Semi volatile organic
compounds
1,4-Dichlorobenzene
<10
<100
<100
<100
7,500
Hexachloroethane
<10
<100
<100
<100
3,000
Nitrobenzene
<10
<100
<100
<100
2,000
Hexachlorobutadiene
<10
<100
<100
<100
500
2,4,6-Trichlorophenol
<10
<100
<100
<100
2,000
2,4,5-Trichlorophenol
<50
<500
<500
<500
400,000
2»4-Dinitrotoluene
<10
<100
<100
<100
130
Hexachlorobenzene
<10
<100
<100
<100
130
Pentachlorophenol
<50
<500
<500
<500
100,000
Total Methyl phenol
<10
<100
<100
<100
Pyridine
<50
<500
<500
<500
5,000
TABLE 38. CALIFORNIA WET TEST SEMIVOLATILE ORGANIC COMPOUNDS
Oig/liter)
Parameter Mud Tar Treated tar Char
1,4-Dichlorobenzene
<10
<10
<10
<100
Hexachloroethane
<10
<10
<10
<100
Nitrobenzene
<10
<10
<10
<100
Hexachlorobutadiene
<10
<10
<10
<100
2,4,6-Trichlorophenol
<10
<10
<10
<100
2,4,5-Trichlorophenol
<50
<50
<50
<500
2,4-Dinitrotoluene
<10
<10
<10
<100
Hexachlorobenzene
<10
<10
<10
<100
Phentachlorophenol*
<50
<50
<50
<500
Total methyl phenol
<10
<10
<10
<100
Pyridine
<50
<50
<50
1100
a Regulatory limit is 1700 /ig/liter. Other compounds are not
listed in California Regulations,
167
-------
SECTION 5
QUALITY ASSURANCE SUMMARY
This section summarizes the information used to check the quality of the data
obtained during this study. Two audit reports will be prepared by Research Triangle
Institute to assess data quality. The first, documenting audits of onsite activities, is
included in Appendix H. The second report, describing the audit of data quality, will
be included in the raw data file.
5.1 Enclosure Exhaust Air Monitoring for S02 and Hydrocarbons
As described in Section 3, a continuous emission monitoring (CEM) system was
installed to measure S02 and total hydrocarbons before and after the exhaust air
control system. This system was subject to stringent quality assurance procedures.
PEI Associates was responsible for these data, the quality assurance plans, and inter-
nal quality assurance review of the sampling procedure and data reduction.
5.1.1 CEM System Audits
An EPA audit was conducted during the beginning stages of this test program
(June 5-8) prior to any excavation. The audit was performed by EPA quality
assurance staff and Research Triangle Institute (RTI), a contractor for EPA. Auditors
observed CEM sampling procedures relative to the guidelines set forth in the Quality
Assurance Project Plan (QAPP) prepared for this project and included in Appendix I.
The following additions to the QAPP were recommended by the audit team and imple-
mented for this project:
1) CEM data were recorded in bound log books. The information on the
original CEM field data sheets was copied into the log book, and these
logs were kept in a file box in the CEM trailer.
2) Strip charts were left on the recorders for a 1-week period before being
sent back to PEI and copied. This reduced the amount of paper han-
dling and improved organization.
168
-------
3) A file system was set up for ali CEM-related test data. The file was used
to store weekly strip charts, log books, floppy disks, ODESSA computer
data, Method 6 data, and velocity data.
4) Settings of the zero and span potentiometers on the CEM's were
recorded daily to document CEM stability.
The CEM sampling systems were also checked with audit gases supplied by
RTI. Audit gas values were compared with values generated from the strip charts and
the ODESSA data logging system. This audit validated the CEM sampling system
operation, responses from chart recorders, and data generated by the ODESSA.
Results of the audit are presented in Table 39. Percentage differences were calculated
based on each audit gas concentration value. These data show that the measured
values were within 5.5 percent of the audit values at concentrations about 10 ppm and
within 27.4 percent at the 1.75-ppm level.
5.1.2 CEM Quality Assurance Objective
The steps taken to ensu'e quality data during this test program included:
o Initial leak checks were made of the sample system.
° Initial calibrations were injected directly into the instruments and com-
pared with calibrations throughout the sample system. Daily calibrations
were performed by direct injection to the analyzer. Periodic sample-
system injections were made throughout the program.
° Sample flow rates were checked to ensure excess flow to each analyzer.
° External (RTI) audits were performed to ensure accuracy in sampling and
analyses.
° The gases used were EPA protocol calibration gases (±1%) for S02
instruments and Master Gas Certified (± 2%) gases for the THC instru-
ments. Certificates are in Appendix B.
Daily 4-point and span/zero checks were performed throughout the
program.
° Performance of manual SO« tests by EPA Method 6 for comparison with
CEM data.
169
-------
TABLE 39. RTI CEM SAMPLE SYSTEM AUDIT
JUNE 5, 1990
Audit,
Computer,
Chart
Chart,
Di fference
Difference,
Difference,
PWTI
ppm
divisions
ppm
comp-audi t
X*
Chart-Aud
X
SO, Inlet CEM
40.86
38.8
82
38.6
-2.06
-5
-2.26
-5.5
30.44
29.4
62
29.1
-1.04
-3.4
-1.34
-4.4
20.7
20.3
42
19.64
-0.4
-1.9
-1.06
-5.1
10.37
10.2
21.2
9.8
-0.17
-1.6
-0.57
-5.5
1.75
1.95
3.8
1.52
0.2
11.4
-0.23
-13.1
0
0.2
0.2
-0.2
0.2
—
-0.2
SO, Outlet CEH
40.86
43
91
42.4
2.14
5.2
1.54
3.8
30.44
32.1
68
31.6
1.66
5.5
1.16
3.8
2C.7
21.2
44.5
20.6
0.5
1.6
-0.1
-0.5
10.37
10.6
22.5
10.3
0.23
2.2
-0.07
-0.7
1.75
1.6
3.3
1.27
-0.15
-8.5
-0.48
-27.4
0
0
0
-0.3
0
-0.3
—
THC Inlet CEH
98
98.8
48
99.3
0.8
0.8
1.3
1.3
25.2
24.5
12
25
-0.7
-2.8
-0.2
-0.8
0
0
0.2
0.7
0
—
0.7
—
THC Outlet CEM
98
95.7
93.5
96.7
-2.3
-2,3
-1.3
-1.3
25.2
24.7
24.3
25
-0.5
-1.9
-0.2
-0.8
0
0.1
0
-0.2
0.1
—
-0.2
—
* .. Audit DDm - chart Dom
Percent difference = ™
170
-------
The objectives for critical measurements for the CEM system are presented in
Table 40. The Method Detection Limit (MDL) is based on the instrument's ability to
quantitate concentrations at an accuracy of 2 percent full scale. The S02 analyzer had
a full-scale value of 50 ppm, which corresponds to an MDL of 1 ppm. Any S02 values
less than 1 ppm should be interpreted as being below the detection abilities of the
analyzer. Precision was defined as the percentage difference between duplicate cali-
bration gas injections. It should be noted that discussions about the use of a daily
afternoon span check, as a daily precision check, were held in the initial stages of the
project. The span and zero afternoon checks were a good indication of morning to
afternoon drift, but not of instrument precision. Precision was defined as the instru-
ment's ability to quantitate a gas concentration as accurately as possible between
duplicate injections of the calibration gas. The data that more accurately show instru-
ment precision were taken from actual duplicate injections of the span gas. Table 41
presents the precision data which show that precision was always within the objective
of 4 percent. The difference (drift) between morning and afternoon calibrations was
also calculated.
Analyzer calibration error is the difference between the gas analyzer response
and the known concentration of the calibration gas. Calibration error values are used
to determine the instrument accuracy. Strip chart data were reduced daily by calculat-
ing the analyzer calibration error by linear regression equations. Calibration error data,
along with drift data for example? days are presented in Tables 42 and 43 for S02 and
THC, respectively. Calibration errors based on the span gas injections only, were all
under 5 percent of span.
Drift was defined as the difference in the initial morning calibration of the instru-
ment and the following morning calibration. Drift data are also presented in Tables 42
and 43. Calculations of daily (24-hour) drift were performed with the strip chart data.
A drift greater than 5 percent was considered excessive and this only occurred twice
during the study. A complete track of calibration and drift information was maintained
in the field log books.
Completeness is the amount of on-line sampling data collected over each
24-hour day. The goal of 80 percent completeness was met during most of the pro-
gram.
Response time is the amount of time necessary for the instrument to display 95
percent of a change in gas concentration on the data recorder. Response time tests
were conducted at the beginning and end of this program. Results are presented in
Table 44. Injections were made into the front end of the sample system to represent
actual sampling conditions. Responses show good sample flow and quick response
by all instruments.
171
-------
TABLE 40. QA OBJECTIVES FOR PRECISION, ANALYZER CALIBRATION ERROR, COMPLETENESS, AND
METHOD DETECTION LIMIT
Critical measurement
Matrix type
Reference
Method
HDL,
ppm
Preci-jj
si on, X
Sens!- c
tlvity, X
Analyzer call-4
brat ion error, X
Drift, X6
Complete-
ness, X
Inlet - SO,
(Scale 0-50 ppm)
Process exhaust
air
EPA Method 6C
1
4
2
5
S
80
Outlet - SO,
(Scale 0-50 ppm)
Control device
exhaust air
EPA Method 6C
1
4
2
5
5
80
Inlet - THC
(Scale 0-200 ppm)
Process exhaust
air
EPA Method Z5A
4
4
2
5
5
80
outlet - THC
(Scale 0-100 ppm)
Control device
exhaust air
EPA Method 25A
2
4
2
5
5
80
* MDl ¦ Method Detection Limit.
b
*8 percent difference of duplicate calibration gas injection. Four percent of full scale was used for this program.
*sj c
fvj Recommended manufacturers instrunent sensitivity. Two percent of full scale Mas used.
^ The difference between the gas analyier response and the known concentration of the calibration gas. Five percent of full scale was used.
e The difference in the Initial calibration of the Instrument and the following calibration at the lero and span values, five percent of
full scale was used.
Calibrations, changing filters, and system maintenance required 5 hours per day. The 80 percent completeness fs based on monitoring
emissions 19 hours out of a 24-hour day. Calibrations were conducted when minimal emissions were anticipated.
Note: The precision, analyier calibration error, and drift percentages are slightly higher than the EPA Reference Method guidelines for
source analyzers. The instruments were on the toner detection ranges, which wake it very difficult to meet the source instrument
guidelines.
-------
TABLE 41. CEM PRECISION DATA
CEM
Date
Time
Concentration, ppm*
Injection 1 Duplicate
Precision15
S02 Inlet
6/6
1046
45.3
45.8
1.1
S02 Outlet
6/6
1100
45.4
46.1
1.5
THC Inlet
6/6
1025
180.3
180.3
0.0
THC Outlet
6/6
1010
91.3
91.6
0.3
S02 Inlet
6/8
0645
45.3
46.0
1.5
S02 Outlet
6/8
0710
45.4
45.7
0.7
S02 Inlet
7/13
1315
43.9
44.1
0.5
S02 Outlet
7/13
1330
44.9
44.9
0.0
THC Inlet
7/13
1245
182.2
182.2
0.0
THC Outlet
7/13
1300
90.4
90.4
0.0
3 Concentration in parts per million, calculated from the strip
chart.
b Percent difference of duplicate gas injections.
Precision = —Ł1-^2— x 100
(C, + C2)/2
where C, and C2 are the observed values.
173
-------
TABLE 42.
S02 CALIBRATION SUMMARY
Correlation
Calibration error,b
Date
Span drift8
Zero drift coefficient
% of span
Inlet
6/9
0.0
-0.2
0.9999
0.02
6/12
-3.0
0,0
0.9999
0.02
6/14
-5.0
-0.2
0.9999
-0.41
6/17
-2.0
-0.5
0.9996
-0.20
6/21
-6.8
+3.0
0.9999
0.62
6/25
+0.5
+0.3
0.9999
-0.20
6/26
+1.0
-0.5
0.9999
0.00
6/29
-0.5
+0.2
0.9999
-0.90
7/9
+2.0
-0.5
0.9999
0.10
7/11
+4.7
0.0
0.9999
-0.08
7/13
-1.0
0,0
0.9999
0.25
7/16
-2.0
-0.1
0.9999
0.41
7/17
-1.0
0.0
0.9999
-0.41
Outlet
6/9
+0.7
-0.5
0.9999
0.65
6/12
-1.0
0.0
0.9999
0.23
6/14
-3.0
-0.3
0.9999
-0.21
6/17
+3,5
-0.3
0.9999
-0.21
6/21
+1.0
+0.2
0.9999
-0.42
6/25
+2.0
+0.5
0.9999
-0.40
6/26
+4.5
+0.7
0.9999
0.00
6/29
-2.0
-0.5
0.9999
-0.20
7/2
-0.5
0.0
0.9999
0.20
7/9
+2.0
-0.2
0.9999
-0.60
7/11
-1.0
-0.2
0.9999
0.20
7/12
-0.5
+0.6
0.9999
0.00
7/13
-1.0
+0.6
0.9999
0.00
7/16
+0.5
+2.5
0.9992
-1.30
7/17
+4.0
0.0
0.9999
-0.44
Drift « Final % of full scale - Initial % of full scale
,in1,Iini. . m , ,t • , nt'rrir .-Calibration qas conc. - conc. indicated,
Analyzer calibration error - [ful1.spa„ concentrationJ * 100
174
-------
TABLE 43. THC CALIBRATION SUMMARY
Correlation Calibration error,6
Date Span drift3 Zero drift3 coefficient % of span
Inlet
6/9
0.0
0.0
0.9999
0.20
6/12
-3.0
-0.2
0.9999
0.19
6/14
+4.3
0.0
0.9999
0.29
6/17
-1.0
0.0
0.9999
-0.19
6/22
-0.5
0.0
0.9999
0.05
6/25
-2.3
0.0
0.9999
0.00
6/26
-0.5
+0.3
0.9999
-0.10
6/28
-2.0
0.0
0.9999
0.00
6/29
+0.7
0.0
0.9999
-0.10
7/9
+3.5
0.0
0.9999
-0.20
7/11
+4.5
+0.2
0.9999
0.20
7/13
-1.0
+0.1
0.9998
0.50
7/15
+9.0
0.0
-
-
7/16
+0.5
+0.2
0.9999
0.25
7/17
+1.0
-0.1
0.9999
0.35
Outlet
6/9
-0.3
+0.2
0.9999
0.10
6/12
+2.2
0.0
0.9999
0.00
6/14
+0.8
0.0
0.9999
0.00
6/17
0.0
0.0
0.9999
0.10
6/21
+2.5
0.0
0.9999
-0.20
6/22
1.5
0.0
0.9999
-0.10
6/25
-1.5
0.0
0.9999
0.10
6/26
-1.5
0.0
0.9999
0.10
6/28
0.0
0.0
0.9999
0.10
6/29
0.0
0.0
0.9999
0.40
7/9
+1.0
+1.5
0.9998
0.50
7/11
-0.3
0.0
0.9999
-0.10
7/13
-1.5
-0.2
0.9999
-0.10
7/15
-4.0
0.0
-
-
7/16
+2.5
0.0
0.9999
-0.15
7/17
+1.0
0.0
0.9998
0.38
Drift « Final % of full scale - Initial % of full scale
• linlirTnr ril!hl_nt. ni.,_ni. (-Calibration aas conc. - conc. indicated, ,nn
Analyzer calibration error - [ fu,,.^concentration 1 * 100
175
-------
TABLE 44. OEM RESPONSE TIME RESULTS
CEM
Date
Time (24-h)
Response, seconds®
S02
Inlet
6/6
1045
<120
S02
Outlet
6/6
1100
<120
THC
Inlet
6/6
1015
<15
THC
Out!et
6/6
1025
<15
so2
Inlet
7/13
0610
<120
so2
Outlet
7/13
0625
<120
THC
Inlet
7/13
0640
<20
THC
Outlet
7/13
0650
<20
8 Calculations of response time were based on calibration gas injections
through the sample system (sample line, conditioning systems, and sample
pumps).
Comparison of strip chart data and ODESSA data were made during a 1-hour
sampling period each day. These hourly comparisons were used to validate the
ODESSA computer data results. Examples of these data for S02 and THC are pre-
sented in Tables 45 and 46, respectively.
5.1.3 Method 6 and Velocity QA Procedures
Sampling was conducted periodically according to EPA Method 6 (a manual
method for measuring S02 concentrations) as an independent check on CEM S02
values. Testing was performed at the inlet and outlet test sites. Results of the Method
6 sampling are presented in Table 47 along with the corresponding CEM values. The
outlet levels were too low to be measured by this manual technique, and the inlet
values agreed within 12.5 percent of the average at higher values and within 30.4 per-
cent at the single check at lower values. Considering the basic experimental
inaccuracies in the manual test method and the difficulty in comparing the exact time
intervals for the CEM and manual tests, these comparisons are reasonable.
Velocity measurements were also conducted at the inlet and outlet sites. These
data were presented in Subsection 3.5.1. The following steps were taken to ensure
quality data from the Method 6 and velocity sampling:
® Calibration of field sampling equipment. Table 48 summarizes calibration
of equipment used for this test series.
° Train configuration and calculation checks.
176
-------
TABLE 45. COMPARISON OF STRIP CHART HOURLY DATA KITH ODESSA HOURLY
DATA FOR S02 INLET
Inlet concentration, ppm Outlet concentration, ppm
Date
Time
Strip chart*
ODESSA1*
Strip chart3
0DESSAb
6/9
1100-1200
0.0 (< 1.0)c
0.1 (< 1.0)c
0.0 (< 1.0)C
0.1 (< 1.0)c
6/12
1000-1100
0.5
0.5
0.2
0.1
6/14
1100-1200
149.0
146.0
0.0
0.0
6/19
1600-1700
30.7
30.9
n.o
n o
6/21
1500-1600
12.3
11.6
0.1
0.1
6/26
1300-1400
153.5
147.7
1.5
2.0
6/28
0800-0900
18.9
18.4
0.1
0.0
7/2
1000-1100
119.1
117.0
0.2
0.1
7/9
1000-1100
0.0
0.0
0.0
0.0
7/11
0700-0800
0.0
< 1.0d
0.0
0.0
7/13
0700-0800
0.2
0.3
0.3
0.0
7/16
1600-1700
8.8
7.3
0.3
0.3
7/18
0900-1000
1.9
0.0
0.0
0.0
a Strip chart data are reduced by linear regression equations by using calibra-
tion gas values versus responses on strip charts,
k ODESSA data are calculated by set concentration levels given to the zero and
full-scale points.
c Method detection limit is 1 ppm, or 2% of full scale (full scale = 50 ppm).
d Scale was 0 to 400 ppm, MDL < 8.0 ppm.
-------
TABLE 46. COMPARISON OF STRIP CHART HOURLY DATA WITH ODESSA HOURLY
DATA FOR THC
Inlet concentration, ppm Outlet concentration, ppm
Date
Time
Strip chart"
ODESSA*
Strip chart3
ODESSA^
6/9
1100-1200
10.0C
9.0C
4.4C
4.0d
6/12
1000-1100
32.7
34.4
13.7
15.2
6/14
1100-1200
162.0
163.0
75.0
75.0
6/19
1600-1700
75.8
78.4
21.5
25.9
6/21
1500-1600
78.8
78.0
31.1
31.2
6/26
1300-1400
183.4
180.8
47.2
43.8
6/28
0800-0900
74.4
75.1
43.3
40.8
7/2
1000-1100
161.1
157.4
86.2
90.7
7/9
1000-1100
31.6
32.5
28.0
29.4
7/11
0700-0800
152.0
157.5
69.0
59.4
7/13
0700-0800
120.4
120.1
105.2
96.2
7/16
1600-1700
128.1
131.3
91.3
91.4
7/18
0900-1000
58.8
58.3
-
-
7/18
1000-1100
-
-
61.8
55.9
8 Strip chart data are reduced by linear regression equations by using calibra-
tion gas values versus responses on strip charts.
b ODESSA data are calculated by set concentration levels given to the zero and
full-scale points.
c Method detection limit is 4 ppm, or 2% of full scale (full-scale - 200 ppm).
-------
TABLE 47. SUMMARY OF S02 EMISSION DATA COMPARISON
Method 6
CEM
a
Run
Location
Date
Time
ppm
ppm
Difference, %
S0S-1
Stack outlet
6/13/90
1557-1617
.,.3"
-------
TABLE 48. FIELD EQUIPMENT CALIBRATION
Equipment
ID
No.
Calibrated
against
A11owable
error
Actual
error
Within
allowable
1imits
Comments
Meter box
VB-2
Bubble meter
(Y ± 0.05
post-test)
0.03
Yes
Pi tot tube
242
505
Geometric speci-
fications
a
a
a
a
OK
OK
Visually inspected on
site
Digital
indicator
VB-2
Millivolt signals
±0.5%
0.23%
Yes
Maximum reading from 10
test points
Stack ther-
mocouple
432
101
ASTM-3F
+1.5%
0.20%
0.20%
Yes
Yes
Maximum difference
Maximum difference
Balance
198
Type S weights
±0.5 g
0.1 g
Yes
Barometer
411
NBS-traceable
barometer
±0.10 in.Hg
(±0.20 in.Hg,
post-test)
0.05
Yes
Dry gas
thermocouple
VB-2
ASTM-3F
±5*F
2*F
Yes
Inlet
a See Appendix B.
-------
° On-site quality assurance checks such as sampling train, pitot tube, and
sample line leak checks,
° Use of designated analytical equipment and sampling reagents.
o Use of designated analytical procedures.
° Internal and external audits to ensure accuracy in analysis. Table 49
summarizes the results of the laboratory audit.
TABLE 49. SUMMARY OF LABORATORY S02 AUDIT RESULTS
True value,
Obtained
Difference,
Date
Type
mg/dscm
value, mg/dscm
%
6/21
Lot
0881-3246
190.7
189.2
-0.8
7/2
Lot
0881-4205
2402.2
2424.6
+0.9
Sampling equipment, reagents, and analytical procedures for this test series
followed and met all necessary guidelines set forth for accurate test results. All^samp-
ling equipment was calibrated within limits described for EPA Methods 2 and 6." A
sulfur dioxide analytical audit was conducted to ensure the quality of the laboratory
data. The results of the audit were within the acceptable limits. The blank results for
the hydrogen peroxide sampling reagent were higher than normal, but they did not
affect results.
5.2 Air Speciation Measurements
Air speciation samples were taken by PEI field staff and analyzed within
24 hours by Performance Analytical, Inc. (PAI). This laboratory was certified for air
analysis by the State of California, was close enough to the site to allow prompt deli-
very of the samples, and had previous experience with air sample analyses. This
phase of the study was performed under the direction of PEI Associates. Analytical
data were reviewed by PEI and by Research Triangle Institute (RTI), who also per-
formed a laboratory audit of the PAI facilities.
The air speciation results obtained from these samples did not meet the quality
control limits and were therefore? not included in this study.
* 40 CFR 60, Appendix A, Reference Method 2 and 6.
181
-------
Analytical data were considered unsatisfactory because of poor recoveries and
reported nondetectable values. The tetrahydrothiophene results were within accept-
able limits, but other compounds such as toluene and xylene were not acceptable.
Hydrogen sulfide data were also reported inaccurately and acetone was reported
above the blank values in audit samples, as determined by RTI, The analytical data
received from PAI are part of the raw data file for this study, but as previously stated,
are not included in this report.
5.3 Waste Characterization
The TCLP and California wet tests were conducted by IT Analytical Services in
Export, Pennsylvania. This laboratory is certified under EPA's Contract Laboratory
Program and adheres to the quality assurance procedures required therein. The IT
project manager reviewed all data and transmitted it to PEI for use in this study.
Proximate and ultimate analyses of the waste samples were run by Broeman
Laboratories in Cincinnati, Ohio, using ASTM standard methods for fuels. This labora-
tory had been previously audited by PEI for these types of analyses.
Data quality assurance for the analyses of raw mud, raw tar, raw char, and
treated tar included the analyses of blank samples, duplicate analyses, matrix spike,
and surrogate spike recoveries. Duplicate analyses were run on one set of metals and
one set of semivolatile analyses, and matrix spike recovery samples were run on all
EPA TCLP tests. The analytical methods used are referenced as follows;
Method Reference:
Toxicity Characteristic Federal Register, Vol 55, No. 61, Thursday,
Leaching Procedure March 29, 1986, Appendix II.
(TCLP)
Acid Digestion of Aqueous
Samples and Extracts for
Total Metals for Analysis
Method 3010, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Ed., 1986
by FLAA or ICP-Spectroscopy
Acid Digestion of Aqueous
Samples and Extracts for Total
Metals for Analysis by
Furnace Atomic Absorption
Spectrometry
Method 3020, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Ed., 1986.
Inductively Coupled Plasma
Method
Method 6010. Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Ed., 1986.
182
-------
Arsenic, Furnace Method
Method 7060, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Ed., 1989.
Selenium (Furnace Method)
Mercury, Manual Coid Vapor
Thallium; AA Furnace Technique
CAM-WET Extraction
Gas Chromatography/Mass
Spectrometry for Volatile
Organics Including BTEX
Gas Chromatography/Mass
Spectrometry for Semivolatile
Organics: Capillary Column
Technique
The range of method detection limits established for each analyses type is
summarized in Tables 50 and 51. None of the listed compounds was detected in the
blank samples.
For the TCLP metals analyses, the analytical and matrix spike recovery percent-
ages were generally within the ranges specified in the methods. The exception was
the element selenium, which exhibited poor recovery for the majority of samples anal-
yzed. One possible explanation for this was the presence of sulfates in the samples,
which was a major interference in the selenium analysis causing low response in the
analytical instrumentation used (Atomic Absorption). Table 52 presents example
matrix spike recovery data for these analyses.
The TCLP analysis for volatile organic compounds exhibited matrix surrogate
spike recoveries generally within the limits specified by the methods. Table 53
presents example data for these analyses.
Method 7740, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Ed., 1986.
Method 7470, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Ed., 1986
Method 7841, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Ed., 1986.
California Administrative Code. Title 22, Social
Security Revision Record for Register 85,
No. 2-B, January 12, 1985.
Method 8240, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Edition,
1986.
Method 8270, Test Methods for Evaluating
Solid Waste. USEPA SW-846, 3rd Edition,
1986.
183
-------
TABLE 50. HASTE CHARACTERIZATION METHOD DETECTION LIMITS, METALS
(mg/L)
Parameter TCLP Analysis CAMWET Analysis
Metals
Arsenic
0.01
0.25
Barium
0.2
2
Cadmium
0.005
0.05
Chromium
0.01
0.1
Lead
0.05
0.5
Mercury
0.0002
0.002
Selenium
0.005
0.005
Silver
0.01
0.1
Antimony
-
0.6
Beryl1i um
-
0.05
Cobalt
-
0.5
Copper
-
0.25
Molybdenum
-
0.2
Nickel
-
0.4
Thallium
-
0.25
Vanadium
-
0.5
Zinc
-
0.2
184
-------
TABLE 51. HASTE CHARACTERISTICS METHOD DETECTION LIMITS, ORGANICS
(M9/L)
Parameters TCLP CAMWET
Volatile orqanics
Vinyl chloride
10-50
Not
applicable
1,1-Dichloroethene
5-25
Not
applicable
Chloroform
5-25
Not
applicable
1,2-Dichloroethane
5-25
Not
applicable
2-Butanone
10-50
Not
applicable
Carbon tetrachloride
5-25
Not
applicable
Trichloroethene
5-25
Not
applicable
Benzene
5-25
Not
applicable
Tetrachloroethene
5-25
Not
applicable
Chlorobenzene
5-25
Not
applicable
>mivolatile oraanics
1,4-Dichlorobenzene
10-100
10
Hexachloroethane
10-100
10
Nitrobenzene
10-100
10
Hexachlorobutadi ene
10-100
10
2,4,6-Trichlorophenol
10-100
10
2,4,5-Trichlorophenol
50-500
50
2,4-Dinitrotoluene
10-100
10
Hexachlorobenzene
10-100
10
Pentachlorophenol
50-500
50
Total methyl phenol
10-100
10
Pyridine
50-500
50
185
-------
TABLE 52. SAMPLE MATRIX SPIKE
RECOVERY DATA, METALS
Matrix spike
Parameter
percent recovery
Arsenic
92
Barium
93
Cadmi urn
92
Chromium
86
Lead
90
Mercury
114
Selenium
0 to 60
Silver
106
TABLE 53. EXAMPLE MATRIX AND SURROGATE SPIKE DATA, ORGANICS
Surrogate spike recovery, X
Matrix spike
Toluene 1.2-dichloro- 4-Sromof1uoro
Parameter
percent recovery
d8 ethane tJ4 benzene
Vinyl chloride
98
92-102 105-106 99-107
1.1-Dichloroethene
92
Chloroform
98
1,2-Dichloroethane
98
2-Butanone
102
Carbon tetrachloride
88
Trichloroethene
98
Benzene
104
Tetrachloroethene
88
Chlorobenzene
100
The TCLP semivolatile organic analyses data exhibited matrix and surrogate
recoveries generally within the limits specified by the methods. The exception was
phenolic compounds which exhibited poor recoveries consistently for each waste type.
One possible explanation is mentioned in SW-846 Method 8270 as being poor, or er-
ratic, chromatograph behavior for these type of compounds. Erratic chromatography
coupled with potential sample matrix effects would tend to result in poor recoveries.
5.4 Flux Chamber Measurements
Analyses of the samples collected for the flux chamber tests were performed by
two companies under the direction of Dr. C. E. Schmidt, Environmental Consultant,
who performed these tests. Radian Corporation in Austin, Texas, performed analyses
on canister samples for VOC, and PAI performed analyses on Tedlar bag samples for
S02. The Radian data were the responsibility of the project manager at Radian and
were issued to Dr. Schmidt and PEI for review and use. The PAI S02 results were re-
viewed by the PAI manager and were also submitted to Dr. Schmidt and PEI
186
-------
Associates. The quality control concerns previously mentioned in Section 5.2 regard-
ing the PAI analyses also affected the S02 data used for these tests and they are not
included in this report.
The canister analyses performed by Radian utilized Varian 3700 and HP 5880
gas chromatographs equipped with multiple detectors, including a flame ionization de-
tector (FID), a photoionization detector (PID), and a halogen-specific detector (HSD).
A multipoint calibration of the FID was performed during the analysis period.
Three concentrations of a hexane standard plus an ultrahigh-purity nitrogen blank
were analyzed to establish a multipoint slope response factor based on carbon. The
detector response was considered acceptable if the correlation coefficient for the re-
sulting calibration curve was >0.9950. A weekly response factor check was per-
formed to assess any potential drift. The acceptance criteria for the single-point check
was ±20 percent of the multipoint slope.
The analysis of compounds performed on the canister samples was subject to
the following detection limits:
Alkanes 0.2 to 1 ppb
Alkenes 0.2 to 1 ppb
Alkynes 0.2 to 1 ppb
Aromatics 0.2 to 1 ppb
Halogenated compounds 0.2 to 1 ppb
Ketones and esters 1 to 5 ppb
Aldehydes and esters 3 to 15 ppb
Sulfur and nitrogen compounds 3 to 15 ppb
Based on a review of the canister sample analyses by RTI, the data did not
meet Agency quality control requirements and are therefore not included in this report.
187
-------
SECTION 6
TRIAL EXCAVATION STUDY COSTS
This section summarizes the on-site costs associated with the various aspects
of this trial excavation. These cost values may also be expressed on a dollars-per-
cubic-yard basis by dividing by 137; the amount of waste excavated. Because this
study represents an experimental trial excavation, the costs per unit of waste ex-
cavated are high and could be greatly reduced when the economy of excavating on a
large scale is considered.
6.1 Enclosure and Air Exhaust Controls
Table 54 presents the costs of erecting and using the temporary enclosure
described in Subsection 2.1. These costs consist of fixed costs of $29,087 for erect-
ing and eventually dismantling the building, purchasing the PVC covering, and a
monthly rental of $11,463. A supervisor and five-person team erected the building on
a smoothly graded site in 5.5 days. A supervisor and a three-person team accom-
plished the dismantling in 3 days. A remote-control video camera with zoom lens and
a 19-inch monitor were also rented for $2500 per month; these costs are added to the
enclosure cost because the video system was necessary to monitor progress and to
assess any safety problems. The addition of these items brings the total cost for the
enclosure system to $70,976 for a 3-month period [$29,087 + (3 x $13,963)].
The major components of the air exhaust control system (consisting of the
scrubber and carbon adsorber) were leased for this study; therefore, most of the costs
are incurred on a monthly basis. The fixed costs include the ductwork system and
fan, the initial installation, and the freight {as summarized in Table 55).
These data show that initial costs of $20,775 were incurred to get the basic
control equipment and ancillary equipment installed at the site. Installation of electrical
lines and switchgear for this system was the single highest cost. A monthly charge of
$9820 was incurred during operation, of which the scrubber rental at $5650 per month
was the major cost item. The total cost for 2 months of operation amounts to the
fixed initial cost plus 2 months of operation, for a total of $40,415 [$20,775 + 2
($9,820)].
188
-------
TABLE 54. COSTS ASSOCIATED WITH TEMPORARY ENCLOSURE
Item Cost, $
Fixed costs
Site clearing 3,000
Erection (equip, and labor) 5,000
FVC enclosure covering 9,600
Dismantling and return 5,300
Decontamination of framework 6,187
Total fixed costs 29,087
Rental costs
Rental of 60-ft x 160-ft enclosure 11,463/mo
Video camera with 2:00m lens and 19-inch
monitor 2.500/mo
Total rental cost 13,963/mo
a Site was fairly level and free of large objects.
TABLE 55. COSTS FOR AIR CONTROL SYSTEM
Item Cost, $
Fixed Costs
Fan 2,200
Ductwork 3,500
Drum pump 680
Installation 2,300
Freight 2,095
Electrical lines and switchgear 10,000
Total 20,775
S/month
Operating costs
Scrubber rental 5,650
Carbon adsorber and housing 1,600
Carbon consumption 600
Caustic consumption 270
Spent caustic disposal 400
Maintenance labor 800
Electricity 500
Total 9,820
a 2-month minimum rental.
k Based on one carbon charge per month at S600.
c Average cost per month.
d Based on 2 days/month and $50/hour.
189
-------
6.2 Foam Costs
Costs to apply foam consisted of the foam reagents (foamer and stabilizer); the
rental of the foam generator (mixer and pump); labor to apply the foam; and miscel-
laneous charges for water, hose-line rental, and fuel. These charges are Itemized in
Table 56. The major costs are for foam reagents and labor, which account for
$38,741 and $45,800, respectively, of the total of $89,591.
TABLE 56. FOAM APPLICATIONS COSTS
Item Costs, $
Foamer (FX-9162) 12,635
Stabilizer (FX-9161) 26,106
Equipment rental (2 months) 3,050
Application labor® 45,800
Miscellaneous supplies** 2,000
Total 89,591
a Includes personnel safety equipment.
b Hoses, fuel, compressed nitrogen gas.
6.3 Tar Processing Costs
The costs associated with the tar processing operation include the mobilization/
demobilization of the pug milt, pug mill rental, operating labor, personnel safety equip-
ment, and delivered costs of portland cement and fly ash. The charges incurred for
the week of tar treatment during the trial excavation are summarized in Table 57. As
shown, the major costs are for field labor, safety equipment, and pug mill rental. If the
total incurred costs are divided by the amount of tar processed during the week, the
per-ton costs for the waste treatment experiments amount to almost $7,400.
However, this is not a realistic estimate of the per-ton costs for waste treatment
that would apply to a full-scale remediation since the costs for labor, safety equipment,
and pug mill rental would remain essentially unchanged while the amount of waste
processed would increase dramatically. As an example, based on a 40 ton/hour
waste processing rate with a waste/total material ratio of 0.25, processing costs of
$340/ton of waste are projected for a 7 hour/day, 5 day/week operation.
190
-------
TABLE 57. TAR PROCESSING COSTS
Item
Cost, $
Pug mill mobilization/demobilization
1,406
4,022
Pug mill rental
Field labor
Manager (3.5 days at $528/day)
2,198
Equipment operator (3.5 days at
$380/day)
1,330
2,016
Two helpers (3.5 days at $576/day) 2,016
Safety officer (3.5 days At S600/dav)2.100 7,644
Subtotal
Personnel safety equipment:
Level B at $180/person/day
Portland cement (6.95 tons at $115/ton)
Fly ash (3.06 tors at $113/ton)
Total costs
3,150
799
346
17,367
6.4 Excavation Costs
Costs to excavate the overburden and waste consist of excavation equipment
charges, field labor, and other equipment (e.g., personnal safety equipment). In
addition, equipment mobilization and demobilization charges and equipment decon-
tamination charges are incurred regardless of the amount of time spent on site. Table
58 summarizes the daily charges for excavation labor and equipment incurred during
this study. Charges for the pug mill are not included here because they are part of
the waste processing cost already presented. The data in Table 58 show a daily
charge of $4584 for a six-person team equipped with Level B personal protection
($2564 + $940 + $1080).
6.5 Air Monitoring
Costs for air monitoring include the rental cost for instruments, recorders, and
data loggers; operating labor; and fixed costs for calibration gases and other ex-
pendable supplies. These costs are summarized in Table 59 and amount to a variable
cost of $46,580/month plus a fixed cost of $7000.
For 2 months of monitoring, these costs amount to $100,160 [(2 x $46,580) +
$7,000].
191
-------
TABLE 58. EXCAVATION LABOR AND EQUIPMENT COSTS
Item
Cost, $
Field Labor
Manager
628/day
Two equipment operators
760
Two helpers
576
Safety officer
600
Subtotal
2564
Excavation Equipment
Trackhoe
464/day
Backhoe
234
Trench shield
242
Subtotal
940
Personal Safety Equipment
Level B at $180/person/day
Level A at $36Q/person/day
Mobilization and demobilization costs
Trackhoe and backhoe
250
Decontamination costs for field equipment
-5000
TABLE 59. AIR MONITORING COSTS
Item S/month
Enclosure inlet and outlet monitors
8,000
Shelter, chart records, data loggers, computer
Perimeter monitors
1,550
4,880
Shelters, chart records, data logger
Labor
2,150
30.000
Subtotal
46,580
Fixed costs
3,000
Calibration gases and supplies
4,000
Electrical hookups and data transmission wiring
a 1-1/3 people for enclosure monitors and 1 person for perimeter
monitors at 80 hours/week each or 600 hours/month at $50/hour.
192
-------
6.6 Cost Summary
The total costs expended for the field work of this trial excavation are sum-
marized in Table 60. Item 1 consists of the initial site preparations and erection of the
enclosure, enclosure and video camera rental, and dismantling. These costs totaled
$70,976 over the 3-month minimim rental period for the enclosure. The air emission
control costs in Item 2 include both erection and dismantling charges and rental over a
2-month period. These costs amounted to $40,415. A large amount of foam vapor
suppressant reagents were used in an attempt to control the air emissions released
during excavation. The reagent cost and application labor costs in addition to equip-
ment charges totaled $89,591 for this item. Item 4 includes the equipment, labor, and
supplies, incurred at a rate of $4,584 per day over an 18-day period. The tar process-
ing runs were conducted at a total cost of $17,367 including labor, additives, and
equipment as shown in Item 5.
Continuous air monitoring at the air emission control system inlet and outlet and
at the four perimeter sites costs $100,160. This cost includes all equipment, labor,
and supplies over a 2-month period.
TABLE 60. SUMMARY OF TRIAL EXCAVATION FIELD COSTS
Item
Cost, $
1. Temporary enclosure (3 month minimum)
2. Air emission controls (2 months)
3. Foam vapor suppressants (as used)
4. Excavation (based on 18 days)
5. Tar processing (10 test runs)
6. Air monitoring (2 months)
Total
70,976
40,415
89,591
82,512
17,367
100,160
401,021
193
-------
REFERENCES
Arniella, E., et al. Solidifying Traps Hazardous Waste, Chemical Engineering.
February 1990.
California Administrative Code, Hazardous Waste Regulations. Article 10,1985.
CH2M Hill, Supplemental Reevaluation of Alternatives, Draft Report, McColl Site,
Fullerton, California. Prepared for U.S. Environmental Protection Agency, Region IX.
February 1, 1989.
Geotechnical Professionals, Inc. Letter to Bhuphen Gandhi (PRC Environmental Man-
agement, Inc.): Summary of Waste Sampling-Hollow-Stem Boring. October 31,1989.
PRC Environmental Management, Inc. Demonstration of a Circulating Bed Combustor
at the McColl Waste Disposal Superfund Site in Fullerton, California. Prepared for U.S.
Environmental Protection Agency, Office of Research and Development, RREL.
Cincinnati, Ohio. October 9, 1989.
Radian Corp. McColl Phase II. Summary of the Nature and Extent of the Problem at
the McColl Site, Fullerton, California. DCN 83-211-078-01-04. March 1983.
Radian 1983. McColl Phase II: Emission Control Test, Final Report, Radian Corpora-
tion. DAN 83-211-081-04, June 1983.
Schmidt, C. E. The Effectiveness of Foam Products for Controlling the Contaminant
Emissions From the Waste at the McColl Site in Fullerton, California. Prepared for
U.S. Environmental Protection Agency. November 20, 1989.
Schmidt, C. E. and J. K. Meyer-Schmidt. Assessment, Monitoring and Modeling from
a Superfund Site Remedial Action. Presented at the Air Pollution Control Assoc.
Annual Meeting. Paper No. 85-66.2, 1985.
Schmidt D. E., R. Stephens, G. Turl. 1987. Case Study: Control and Monitoring of
Air Contaminants During Site Mitigation, Presented at the 79th Annual APCA Meeting,
June 1987.
U.S. Environmental Protection Agency. Measurement of Gaseous Emission Rates
From Land Surfaces Using an Emission Isolation Flux Chamber. User's Guide. U.S.
EPA, Las Vegas, Nevada. February 1986.
194
-------
TECHNICAL REPORT DATA
/Please read Insmictwni on the reverse before comph r
1. REPORT NO. [2.
EPA/540/R-92/015 |
3.
4. TITLE ANDSUBTITLE
Technology Evaluation Report
SITE Program Demonstration of a Trial Excavation
at the McColl Superfund Site
5. REPORT OATE
September 1992
6. PERFORMING ORGANIZATION COOE
7. AUTHORIS)
IT Corporation
8- PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
IT Corporation
11499 Chester Road
Cincinnati, OH 45246
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-4284
12. SPONSORING AGENCY NAME AND ADDRESS
Risk Reduction Engineering Laboratory--Cin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE OF REPORT AND PERIOO COVERED
Final Ranort
14. SPONSORINgBvGENCY COOE
EPA/600/14
15, SUPPLEMENTARY NOTES
Project Manager: S. Jackson Hubbard (513) 569-7507
16. ABSTRACT
.. excavation of approximately 137 cubic yards of waste was performed at
the McColl Superfund Site in Fullerton, CA, to better determine the nature of this
waste, any treatment needed to improve its handling characteristics, and the extent
of air emissions that might occur during excavation. This type of information is
necessary to plan full-scale remediation of this highly acidic petroleum refinery
waste buried at this site. The trial excavation was conducted within a temporary
enclosure with air exhausted from the enclosure through a sodium hydroxide-based wet
scrubber and activated-carbon bed adsorber to reduce air emissions of sulfur dioxide
and organic compounds. Foam was used in an attempt to suppress atmospheric releases
from the raw waste during excavation, storage, and processing. The air exhaust was
monitored for total hydrocarbons and sulfur dioxide before and after the air emission
control system. In addition, total hydrocarbons and sulfur dioxide were monitored
along the site perimeter to determine potential impact of air emissions on the nearby
community.
This report descibes the equipment used to measure and control emissions, the
excavation progress, and resulting emissions before and after the air control system
An assessment of the foam vapor suppressants is also provided as well as information
on the costs incurred for the field portion of this study.
17. KEY WORDS AND DOCUMENT ANALYSIS
». DESCRIPTORS
b.IDENTIFIERS/OPEN ENDEO TERMS
c. COSATl Field/Group
Air Pollution
Excavation
Petroleum
Hydrocarbons
Sulfur Dioxide
Petroleum Wastes
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
mmims!Tl',s*'pon>
21. NO. OF PAGES
20. SECURITY class (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220—1 (R«v. 4-77) previous edition
E
------- |