xvEPA
United States
Fnvironmental Protection
Agency
Industrial Environmental Research EPA-600/7-80-015c
Laboratory January 1980
Research Triangle Park NC 27711
Experimental/
Engineering Support for
EPA's FBC Program:
Final Report
Volume III. Solid
Residue Study
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-80-015C
January 1980
Experimental/Engineering Support
for EPA's FBC Program:
Final Report -
Volume III. Solid Residue Study
by
C.C. Sun, C.H. Peterson, and D.L. Keairns
Westinghouse Research and Development Center
1310 Beulah Road
Pittsburgh, Pennsylvania 15235
Contract No. 68-02-2132
Program Element No. INE825
EPA Project Officer: D. Bruce Henschel
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
The Westinghouse R&D Center is carrying out a program to provide
experimental and engineering support for the development of fluidized-
bed combustion systems under contract to the Industrial Environmental
Research Laboratory (IERL), U.S. Environmental Protection Agency (EPA),
at Research Triangle Park, NC. The contract scope includes atmospheric
(AFBC) and pressurized (PFBC) fluidized-bed combustion processes as they
may be applied for steam generation, electric power generation, or
process heat. Specific tasks include work on calcium-based sulfur removal
systems (e.g., sorption kinetics, regeneration, attrition, modeling),
alternative sulfur sorbents, nitrogen oxide (NO ) emissions, particulate
X
emissions and control, trace element emissions and control, spent sorbent
and ash disposal, and systems evaluation (e.g., impact of new source
performance standards (NSPS) on FBC system design and cost).
This report contains the results of work defined and completed under
the spent sorbent and ash disposal task of the contract. Work on this
task was performed from January 1976 to January 1979 and is documented
in the following contract reports:
• "Disposal of Solid Residue from Fluidized-Bed Combustion:
Engineering and Laboratory Studies," EPA-600/7-78-049 (NTIS
PB 283-082), issued in March 1978, which presented the results
of work performed from January 1976 to January 1977
• The present report, which presents the results of extended
environmental impact tests, comparisons with reference materials,
and screening tests and analyses on the potential for processing
FBC residue for disposal or utilization. The report documents
work performed from January 1977 to January 1979 and subsequent
extensions from review of the draft through October 1979.
Since proper disposal of solid residue is of primary importance to
the commercialization of the FBC process, continuing effort is directed
iii
-------
toward solid residue studies under the current Westinghouse contract to
EPA (68-02-3110).
Work on the other tasks performed under this contract has also been
reported:
• Experimental/Engineering Support for EPA's FBC Program: Final
Report Volume 1, Sulfur Oxide Control, EPA-600/7-80-015a,
January 1980.
• Experimental/Engineering Support for EPA's FBC Program: Final
Report Volume II, Particulate, Nitrogen Oxide, and Trace Element
Control, EPA-600/7-80-015b, January 1980.
• Experimental/Engineering Support for EPA's FBC Program: Final
Report Volume IV, Engineering Studies, EPA-600/7-80-015d,
January 1980.
• Effect of S0? Emission Requirements on Fluidized-Bed Combustion
Systems: Preliminary Technical/Economic Assessment, EPA-600/7-78-163,
August 1978.
• Regeneration of Calcium-Based SO Sorbents for Fluidized-Bed
Combustion: Engineering Evaluation, EPA-600/7-78-039,
NTIS PB 281-317, March 1978.
• Alternatives to Calcium-Based SO. Sorbents for Fluidized-Bed
Combustion: Conceptual Evaluation, EPA-600/7-78-005, January 1978.
• Evaluation of Trace Element Release from Fluidized-Bed Combustion
Systems, EPA-600/7-78-050, NTIS PB 281-321, March 1978.
IV
-------
ABSTRACT
Partially reacted limestone or dolomite and coal ash from atmo-
spheric and pressurized fluidized-bed combustion systems must be dis-
posed of or utilized in an environmentally acceptable manner. A prelim-
inary understanding of the environmental impact of the disposal of the
solid residue and perspective on selected choices available for direct
disposal and residue processing for utilization or disposal have been
developed. Laboratory testing programs have been developed to determine
the chemical and physical characterization, leaching behavior, and
residual activity of FBC solid residues. Processing of FBC solid waste
(bed overflow and carry-over material) has been studied to Investigate
the potential for reducing environmental impact and to provide for
potential utilization. The impact of environmental legislation and reg-
ulations is assessed, with particular emphasis on the regulations to be
promulgated under the Resource Conservation and Recovery Act (RCRA,
1976). The environmental impact of the disposal of processed and unpro-
cessed FBC sorbent is projected and compared with natural gypsum and FGD
spent materials.
Environmental complications associated with the disposal of FBC
solid waste are not likely to limit the development of FBC processes,
and the solid residues are judged to be nonhazardous* materials under
the classifications of proposed RCRA regulations. Processing techniques
based on compaction of FBC solid waste may provide environmentally
improved or utilizable waste products.
*Nonhazardous is used in this report to identify a solid that does not
qualify as hazardous under RCRA. It does not imply an absence of
environmental impact.
-------
TABLE OF CONTENTS
Page
1 INTRODUCTION 1
2 SUMMARY AND CONCLUSIONS 4
Assessment of Environmental Impact 5
Spent Sorbent Processing 8
3 RECOMMENDATIONS 11
Assessment of Environmental Impact 11
Spent Sorbent Processing 12
4 ENVIRONMENTAL CONSTRAINTS 14
Regulations/Criteria 14
Test Program Criteria 19
Effect of Residue Processing 21
5 FBC SOLD RESIDUES 22
Residue Characteristics 22
Residues Tested for Environmental Impact 24
6 EXPERIMENTAL TEST PROGRAM: ENVIRONMENTAL IMPACT 30
Experimental Methods 30
Test Results 33
Performance Summary 96
7 CONVENTIONAL POWER PLANT RESIDUE: ENVIRONMENTAL IMPACT
TESTS 104
vii
-------
TABLES OF CONTENTS (Continued)
Page
FGD Residue Test Results 104
Comparison with FBC Residue 112
8 SPENT SORBENT PROCESSING 116
Review of Processing Options 116
Pressed Material 123
Environmental Impact of Processed Wastes 187
9 ENVIRONMENTAL ASSESSMENT 197
10 REFERENCES 201
APPENDIXES
)
A SIEVE ANALYSIS USING SINGLE SIEVES UNDER VACUUM 206
B ISOSTATIC PRESSING PROCEDURE 207
viii
-------
LIST OF TABLES
Page
1 Comparison of Environmental Concerns 7
2 Development of RCRA 3001 Regulation 16
3 Development of EPA Leach Test 17
4 Selected Water Quality Criteria 20
5 Projected Spent Sorbent Compositions from Three Basic FBC
Systems 23
6 Distribution of Spent Sorbent/Coal Ash in FBC Bed and
Carry-over 24
7 Summary of Composition of Spent Sorbent from FBC Systems 24
8 Process Conditions of Samples Studied for Their Environ-
mental Impact on Disposal 26
9 Summary of Processed FBC Specimens 29
10 Leachate Characteristics of B&W 19 Residues 35
11 Solid and Leachate Characteristics of B&W 19 Carry-over 37
12 Solid and Leachate Characteristics of PER Residues 39
13 Leachate Characteristics of PER Residues 40
14 Chemical Analysis of Battelle FBC Residues from the
1500-hr Corrosion/Erosion Test 44
15 Chemical Characteristics of Leachates from Battelle FBC
Residues 45
16 Identification of Spent MERC Bed Material and Carry-over 48
17 Summary of Leaching Results of MERC Samples by Continuous
Shake Method 49
ix
-------
LIST OF TABLES (Continued)
18 Solid and Leachate Characteristics of MERC-AFBC Residues:
Effect of Using CC^-Saturated Water as Leaching Medium 52
19 X-Ray Diffraction Identification of PFBC Residues 56
20 Leachate Characteristics of PFBC Residues 59
21 Solid and Leachate Characteristics of PFBC Residue from
Exxon 61
22 Solid and Leachate Characteristics of PFBC Residues with
Sorbent Regeneration 62
23 Particle Size Distribution of Exxon 67 Third Cyclone
Fines 63
24 Chemical Characteristics of Exxon 67 Third Cyclone Fines
and Their Leachate 64
25 Solid and Leachate Characteristics of Adiabatic PFBC
Baghouse Residue from Combustion Power CP-403 Run 70
26 Summary of Leaching Test Methods Applied to the Exxon 27
Sorbent/Ash Compacts and Unprocessed Sorbent/Ash and
Mixtures 74
27 Specific Conductance of Leachate from the Exxon 27
Sorbent/Ash Compacts 75
28 pH of Leachate from the Exxon 27 Sorbent/Ash Compacts 76
29 Calcium Concentration of Leachate from the Exxon 27
Sorbent/Ash Compacts 77
30 Sulfate Concentration of Leachate from the Exxon 27
Sorbent/Ash Compacts 78
31 Leachate Characteristics of Processed FBC Residue as a
Function of Leaching Cycle 82
32 Comparison of Leachate Characteristics with Water and
Acetate Leaching Media 84
33 Correlation between TDS and Specific Conductance 89
-------
LIST OF TABLES (Continued)
Page
34 Heat Release Property of FBC Waste 92
35 Preliminary Indications of Environmental Impact of FBC
Solid Waste Disposal 102
36 Summary of FGD Sludge Samples 105
37 Chemical Characteristics of FGD Sludge, Liquor, and
Leachate 110
38 Comparison of Leachate Characteristics of the FBC and FGD
Residues HA
39 Chemical Analyses of FBC Pilot Plant Residues 126
40 Weight Loss on Heating Ground B&W Spent Bed Material 128
41 Particle Size Analysis of As-Received -44 vim (-325 mesh)
B&W Run 19 Spent Bed Material Using an Alpine Air-Jet
Sieve 129
42 Particle Size Analysis of As-Received -44 ym B&W Run 19
Spent Bed Material Using a Coulter Counter 130
43 Test Matrix for Freeze-Thaw Tests on FBC Residues 138
44 Freeze-Thaw Resistance if Isostatically Pressed Cylinders
of FBC Spent Bed Material from B&W Plant: Test Specimens 141
45 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of FBC Spent Bed Material from B&W Pilot Plant: Control
Specimens 142
46 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of FBC Spent Bed Material from Exxon Pilot Plant: Test
Specimens 143
47 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of FBC Spent Bed Material from Exxon Pilot Plant: Con-
trol Specimens 144
xi
-------
LIST OF TABLES (Continued)
Page
48 Freeze-Thaw Resistance of Isostatlcally Pressed Cylinders
of FBC Spent Bed Material from PER Pilot Plant: Test
Specimens 145
49 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of FBC Spent Bed Material from PER Pilot Plant: Control
Specimens 146
50 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of Type I Portland Cement: Test Specimens 147
51 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of Portland Type I Cement: Control Specimens 148
52 Freeze-Thaw Resistance of Cast Cylinders of Portland
Type I Cement and River Sand: Test Specimens 149
53 Freeze-Thaw Resistance of Cast Cylinders of Portland
Type I Cement and River Sand: Control Specimens 150
54 Distribution of Cycle Times 158
55 Distribution of Cycle Temperatures 159
56 Mean Values for Cycle Parameters 160
57 Comparison of Selected Cycle Parameters with ASTM C666 160
58 Allocation of Chemical Components in FBC Residues Among
Normal Cement Constituents 162
59 Comparison of X-Ray Peaks from Selected Calcium Compounds 170
60 X-Ray Diffraction Peaks from Babcock & Wilcox Specimen
No. OBS0014C-3 174
61 X-Ray Diffraction Peaks from Exxon Specimen
No. OES0014C-3 176
62 X-Ray Diffraction Peaks from PER Specimen No. OPS0014C-3 178
63 X-Ray Diffraction Analysis of Selected Isostatically
Pressed Specimens of FBC Residues after 14 Days Curing in
Water 180
xii
-------
LIST OF TABLES (Continued)
Page
64 Weight Loss from Cured Isostatlcally Pressed Specimens
When Heated to 1000°C 181
65 Typical Analysis of Sintered Fly Ash 183
66 Bulk Density of Materials for Aggregate Tests 184
67 Water Absorption by Penn Virginia 1/2 in. x No. 8
Sintered Pelletized Fly Ash 186
68 Conrpressive Strength of Cylinders Made from Portland
Cement, FBC Spent Bed Materials, and Sintered Fly Ash 186
69 Constants in the Empirical Correlation of Cumulative
Calcium Leached versus Number of Extractions 190
70 Calculated Leach Rates for Selected FBC Residues 191
71 Comparison of the Environmental Impact of the
FBC and FCD Solid and Liquid Waste Disposal 198
xiii
-------
LIST OF FIGURES
Page
1 Leachate Quality as a Function of Intermittent Leaching 36
2 SEM and EDAX of (a) RS-PER Shakedown Bed Material,
Particle Surface, (b) RS-PER Shakedown Bed, Fractured
Surface, (c) RS-PER Carry-over 41
3 SEM Photomicrograph (a) and EDAX Spectrum (b) of the
White Precipitate Formed Readily in Air, from the
Leachate from PER Bed Material and Carry-over 42
4 Leaching Characteristics of Battelle FBC Bed and Carry-
over Material as a Function of 72-hr Carry-over Inter-
mittent Leaching 46
5 Leachate Characteristics of Spent MERC Bed Material, Ash,
and Fines as a Function of Intermittent Leaching 51
6 Leachate Characteristics from Codisposal MERC Bed and
Carry-over Materials 54
7 Leachate Quality of PFBC Residues as a Function of
Intermittent Leaching 60
8 Leaching Characteristics of Exxon 67 Third Cyclone
Particulates as a Function of Intermittent Leaching 65
9 Morphological Characteristics of CP-403 Ash 71
10 Leachate Characteristics of CPU-400 P403 Ash from
Combustion Power as a Function of Intermittent Leaching 72
11 Comparison of Leachate Characteristics of Processed
Exxon 27 Sorbetit/Ash Compact Crushed Powder from Such
Compact, Unprocessed Sorbent/Ash Mixture, and Natural
Gypsum 79
xiv
-------
LIST OF FIGURES (Continued)
Page
12 Comparison of Specific Conductance of Leachates from
Exxon 27 Sorbent/Ash Compact, Crushed Compact, Unpro-
cessed Exxon 27 Sorbent/Ash Mixture, Ash Gypsum 81
13 Correlation between TDS and Specific Conductance in the
FBC Leachate System 88
14 Correlation between TDS and Specific Conductance In FBC
and CAFB Leachates 88
15 Heat Release Property as a Function of Solid/Water Ratio 95
16 Heat Release Property of Spent Bed and Carry-over from
the Atmospheric FBC System 95
17 Heat Release Property of Spent Sorbent and Fly Ash from
the Atmospheric FBC System 97
18 Comparison of Heat Release Property of Spent Bed and
Carry-over from the FBC Process of Different Processing
Variations 97
19 Leachate Characteristics as a Function of Continuous
Leaching 99
20 Leachate Characteristics as a Function of Intermittent
Leaching 100
21 Comparison of FBC and FGD Residue Photomicrographs 106
22 SEM and EDAX of FBC and FGD Ash 108
23 Leachate Characteristics of Dried FGD Sludge as a
Function of Continuous Leaching 111
24 Leachate Characteristics of Dried FGD Sludge as a
Function of Intermittent Leaching 112
25 Comparison of Leachability of Processed and Unprocessed
FGD and FBC Waste with Natural Gypsum 113
26 Particle Size Distribution of Ground B&W Spent Bed
Material from Run 19 131
xv
-------
LIST OF FIGURES (Continued)
27 Mold Assembly for Preparation of Specimens by Isostatic
Pressing 132
28 Mold Assembly 133
29 Specimens Produced by Isostatic Pressing 133
30 Freeze-Thaw Chamber 135
31 Schematic of Freeze-Thaw Chamber 136
32 Sketch of Type I Portland Cement Control Specimen with
Embedded Thermocouple for Temperature Control 136
33 Freeze-Thaw Resistance of Isostatlcally Pressed Cylinders
of Portland Cement and River Sand 151
34 Freeze-Thaw Resistance of Isostatlcally Pressed Cylinders
of Portland Cement 151
35 Freeze-Thaw Resistance of Isostatlcally Pressed Cylinders
of Exxon Spent FBC Bed Material 152
36 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of B&W Spent FBC Bed Material 152
37 Freeze-Thaw Resistance of Isostatically Pressed Cylinders
of PER Spent FBC Bed Material 153
38 Failure Mode of Isostatically Pressed Cylinders of Spent
Bed Material in Axial Compression - Initial Specimens 154
39 Failure Mode of Isostatically Pressed Cylinders of Spent
Bed Material in Axial Compression - After 12 Cycles of
Freeze-Thaw Testing 155
40 Failure Mode of Isostatically Pressed Cylinders of Spent
Bed Material in Axial Compression - Outside Controls at
Age Corresponding to 12 Cycles of Freeze-Thaw 156
41 X-Ray Diffraction Tracing from Isostatlcally Pressed B&W
Spent Bed Material 165
42 X-Ray Diffraction Tracing from Isostatically Pressed PER
Spent Bed Material 165
xvi
-------
LIST OF FIGURES (Continued)
Page
43 X-Ray Diffraction Tracing from Isostatically Pressed
Exxon Spent Bed Material 166
44 X-Ray Diffraction Tracing from Isostatically Pressed
Type I Portland Cement and River Sand 166
45 X-Ray Diffraction Tracing from Quartz 167
46 X-Ray Diffraction Tracing from Calcium Carbonate 167
47 X-Ray Diffraction Tracing from Gypsum 168
48 X-Ray Diffraction Tracing from Plaster of Paris 168
49 X-Ray Diffraction Tracing from Calcined Gypsum 169
50 X-Ray Diffraction Tracing from Calcined Plaster of Paris 169
51 Leaching of Calcium from Exxon Spent Sorbent/Carry-over
Compacts 188
52 Leaching of Calcium from Exxon Spent Sorbent/Carry-over
Normalized Basis 188
xvii
-------
NOMENCLATURE
AFBC - atmospheric pressure fluidized-bed combustion
ANL - Argonne National Laboratory
B&W - Babcock and Wllcox Company
CAFB - chemically active fluidized bed
Ca/S - calcium-to-sulfur ratio
CBC - carbon burnup cell
CSO - Columbus Southern Ohio Company
DOE - Department of Energy
DWS - Drinking Water Standards
EDAX - energy dispersive analysis by X-ray
EMA - electron microprobe analysis
EP - extraction procedure
EPA - Environmental Protection Agency
EPRI - Slectric Power Research Institute
ESP - Electrostatic Precipitator
FBC - fluidlzed-bed combustion
FGD - flue gas desulfurization
GE - General Electric Corporation
IERL/RTP - Industrial and Environmental Research Laboratory, Research
Triangle Park, NC
LGE - Louisville Gas and Electric Company
MERC - Morgantown Energy Research Center
NIPDWR - National Interim Primary Drinking Water Regulations
PER - Pope, Evans and Robbins
PFBC - pressurized fluidized-bed combustion
RCRA - Resource Conservation and Recovery Act of 1976
SEM ~ scanning electron microscopy
SFA - sintered fly ash
xviii
-------
NOMENCLATURE (Continued)
SBM - spent bed material
TDS - total dissolved solids
TEP - toxic extraction procedure
TGA. - therraogravimetric analysis
TOG - total organic carbon
USPHS - U. S. Public Health Service
WHO - World Health Organization
xix
-------
ACKNOWLEDGMENT
We want to express our high regard for and acknowledge the contri-
bution of Mr. D. B. Henschel who served as the EPA project officer.
Mr. P. P. Turner and Mr. R. P. Hangebrauck, Industrial Environmental
Research Laboratory, EPA, are acknowledged for their continuing contri-
butions through discussions and support of the program.
We thank Mr. R. C. Hoke and Ms. S. Walther of Exxon Research and
Engineering; Mr. Robert Reed of Pope, Evans and Robbins; Mr. H. B. Lange
of Babcock and Wilcox; Mr. H. Stoner of Combustion Power; Messrs. J. S.
Wilson and R. Rice of Morgantown Energy Research Center; Messrs. A. M.
Hall and H. Carlton of Battelle - Columbus Laboratories and Mr. R. Stone
of Ralph Stone and Co. Inc. for their cooperation in supplying FBC resi-
dues. We also acknowledge the kind assistance by Mr. P.P. Leo of Aero-
space Corporation and Mr. R. P. Van'Ness of Louisville Gas and Electric
Co. and Mr. D. Henzel of Dravo Lime Co. in supplying FGD scrubber sludge
samples for this study.
In addition, we acknowledge the cooperative efforts and contribu-
tions by many Westinghouse personnel, in particular, the thoughtful
review and contributions to this report by Dr. R. A. Newby; the techni-
cal assistance in carrying out the laboratory experiments by Messrs.
J. T. McAdaras, R. Brinza and L. Thomas; and the contributions in sample
characterization by many members of the Analytical Chemistry, Physical
Metallurgy, Ceramics, and Materials Testing Departments within the
Westinghouse R&D Center.
xx
-------
1. INTRODUCTION
Fluldized-bed combustion (FBC) for electric power generation, when
compared with conventional technology, provides the potential for
improved thermal conversion efficiency, reduced costs, and reduced
environmental impact. The fluidized-bed combustion process, operated at
atmospheric (AFBC) or elevated pressure (PFBC), typically results in the
production of dry, partially utilized dolomite or limestone particles
from 0 to 6 mm in size. In addition, fine particles of sorbent and ash
are collected in the particulate-removal system. The sorbent material
may be either regenerated for recycling to the fluid-bed boiler for
repeated sulfur dioxide (802) removal or disposed of in its partially
sulfated form in a once-through system. The properties of these spent
sorbents (size distribution, composition, etc.) depend on the FBC oper-
ating and design parameters. The major compounds in the waste stone to
be disposed of are calcium sulfate (CaS04), calcium oxide (CaO), calcium
carbonate (CaC03>, and magnesium oxide (MgO), when dolomite is used; and
CaS04 and CaO or CaC03 when limestone is used. Trace elements arising
from impurities in the coal and sorbent will also be present.
The quantity of spent sorbent will depend on the sulfur content of
the fuel, the emission standard, the operating conditions, and the sor-
bent characteristics. The spent sorbent for disposal will generally
range from 0.01 to 0.5 kg sorbent/kg coal. Disposal of the coal ash is
also considered.
The environmental acceptability of FBC residue disposal is dictated
by the constraints of environmental laws and regulations. The three
primary environmental laws are the Resource Conservation and Recovery
Act (RCRA) of 1976 (the Solid Waste Disposal Act of 1965, as amended by
-------
P. L. 94-580, 1976),* the Federal Water Pollution Control Act of 1972
(Public Law 92-500, 1972, as amended by Clean Water Act, P. L. 95-217,
1977),2>3 and the Clean Air Act Amendments, 1977 (Clean Air Act, 1970 as
amended).^ The passage of RCRA closed the legislative loop of environ-
mental laws (air/water/solid) and created a new level of control over
solid waste disposal. The regulations to be promulgated under RCRA Sub-
title C and Subtitle D on hazardous and nonhazardous waste disposal will
have the greatest impact on FBC residue disposal.
An understanding of the environmental impact of the disposal or use
of the spent sorbent and fly ash is critical to the successful implemen-
tation of FBC processes. This investigation is designed to provide a
basis for projecting the environmental impact of disposal, interpreting
results from large-scale demonstration sites, screening utilization
options in view of the environmental impact, and developing optimal sys-
tem design and operating requirements to minimize the environmental
impact of the spent sorbent and coal ash. The scope of the program
conceived to achieve these objectives includes identification of FBC
spent sorbent and coal ash characteristics and quantities, development
of laboratory tests to quantify the environmental impact of disposal,
conducting environmental impact tests on actual FBC spent sorbent and
ash, and performing studies to investigate potential processing options
and their environmental impact.
An assessment of the potential environmental impact and a perspec-
tive on the potential for direct disposal and for processing for dispo-
sal or utilization are presented. Since proper disposal of solid resi-
due is of primary importance to the commercialization of the FBC
process, continuing effort is directed toward solid residue studies
under the Westinghouse current contract to EPA (68-02-3110).
-------
Two reports are being prepared that will present the results of FBC
solid residue leachate characteristic determination using the RCRA test
procedures and the results of an evaluation of methods for disposal of
FBC residues.
-------
2. SUMMARY AND CONCLUSIONS
The FBC solid residue studies reported here cover two major topics:
• Environmental impact. Experimental studies were carried
out to investigate the environmental impact of direct
disposal of dry FBC residues and of disposal after
processiag.
• Residue processing. Experimental engineering studies were
conducted to evaluate residue processing options for uti-
lization or for reducing environmental impact.
Results from these and related studies lead to the following
conclusions:
• FBC residue disposal will not be an obstacle to commer-
cialization of FBC systems.
• FBC residues tested would not constitute a hazardous
waste based on RCRA criteria.
• Potential concerns for the disposal of FBC residue are the
quantity; the chemical impact of calcium (Ca), sulfate
(804), dissolved solids (IDS), and pH; and the thermal
activity under specific operating conditions.
• Low-temperature processing of FBC residue can reduce its
environmental impact. Stable compacts have been formed
and tested, showing that fixation reduces both the rate of
leaching and the concentration of solutes.
• Low-temperature processing offers the potential for pro-
ducing material applicable for use as aggregate.
• Options ranging from direct disposal to fixation are
available for meeting specific applications, site require-
ments, and environmental impact constraints.
-------
ASSESSMENT OF ENVIRONMENTAL IMPACT
The work accomplished in this area Includes the following:
• Comprehensive testing encompassing three major areas:
- Residue characterization
- Leaching property
- Thermal activity
• Development and use of screening tests to investigate
leaching and thermal properties
• Testing of over 30 samples of FBC solids covering a wide
range of FBC process variations:
- FBC units: Exxon miniplant PFBC (once through and
regenerative)
Argonne National Laboratory (ANL) PFBC
Combustion Power PFBC, adiabatic
Pope, Evans and Robbins (PER) AFBC
Babcock and Wilcox (B&W) AFBC
Battelle - Columbus Laboratories AFBC
Morgantown Energy Research Center (MERC) AFBC
- FBC system: AFBC/PFBC/adiabatic; once through/
regenerative
- Residue source: spent bed material and carry-over (pri-
mary, secondary, and tertiary flue gas
particle collectors)
- Fuel: six sources of coal
- Sorbent: five limestones and four dolomites
- Operating conditions: T =» 600 to 1000°C
P - 101.3 to 1013 kPa (1 to 10 atra)
Ca/S = 0.5 to 4
• Environmental impact of both direct disposal of FBC rest-
due and disposal after further processing
• Conventional power plant residues (fixed and unfixed FGD
sludge) and natural gypsum tested to provide perspective
and comparison
-------
• Preliminary Investigation of leaching property as a func-
tion of leaching media (deionized water, CC^-saturated
water, and acidic acetate buffer)
• Review of legislative and regulatory constraints, in par-
ticular, The Resource Conservation and Recovery Act
(RCRA),1 as they are related to FBC residue disposal
• Assessment of the potential environmental impact of FBC
residue disposal and the identification of major concerns.
Table 1 summarizes the results from the comprehensive test program.
Conclusions based on these results are that:
• The potential concerns with FBC residue disposal are the
pH, the TDS, calcium and (804) in the leachate, its ther-
mal activity, and the quantity of solid to be disposed of
for some sites.
• The leachability and the thermal activity can be effec-
tively reduced by further processing.
• The trace elements and total organic carbon (TOC) in
leachates are extremely low in the leachate of both the
unprocessed and the processed FBC residue and are not
expected to be a problem.
• The leachate of the FBC carry-over has, on the average,
lower pH, calcium, and TDS, but higher trace elements than
does the leachate of bed material.
• Thermal activity is largely dependent on the amount of CaO
present in the residue and is a function of the operating
conditions, resulting in either calcined or uncalcined
sorbent in the form of CaO or CaCC>3, respectively.
• Thermal activity is higher in the AFBC than in the PFBC
residues tested, higher in bed material than in the carry-
over.
• The leachate of AFBC residue with limestone sorbent aver-
ages higher pH, calcium, and TDS than does that of PFBC
with dolomite sorbent.
-------
Table 1
COMPARISON OF ENVIRONMENTAL CONCERNS
Dwg. 1713B33
Process
Parameter
Hazardous Characteristics3
Leachate Characteristics
Trace Elements
Total Organic Carbon
pH
TDS
Ca
Mg
Sulfate
Sulfite
Sulfide
Thermal Activity
Physical Strength
Quantity
FBC
Unprocessed
no
X
X
X
X
d
yes
no
yes
Processed
no
c
c
c
c
no
no
yes
FGDb
Unprocessed
no
X
X
X
X
X
X
no
yes
yes
Oxidized
no
X
X
X
X
X
no
no
yes
Stabilized
no
X
X •
X
X
no
no
yes
Natural
Gypsum
Unprocessed
no
X
X
X
no
no
Not
applicable
X = Exceeds DWS where they exist.
Detailed report on test data will be issued later in 1980.
Limited to lime or limestone scrubber sludge.
cLeachability significantly reduced, degree of improvement depending on processing alternatives.
TJot a concern for the once-through FBC residue; further investigation needed for spent sorbent with sorbent
regeneration.
Associated with residue with a substantial amount of unutilized CaO, e.g. AFBC bed material.
-------
• Leaching property Is a function of the pH of the leaching
medium. Trace elements in leachate, in general, increase
with decreasing pH.
• Although an absolute comparison may not be possible, gen-
eral trends have shown that the physical, chemical, and
leaching properties of FBC residue are superior to the
nonstabilized FGD sludge with regard to disposal and are
comparable to the chemically and physically stabilized FGD
residue from the conventional power plant.
• On the basis of the proposed regulations and identifica-
tion methods for hazardous waste under RCRA Sec. 3001 and
the FBC residue characteristics reported here, we expect
FBC residue to be nonhazardous. Further work is being
carried out under Westinghouse1s current contract with EPA
and will be discussed in a separate report.
• In summary, based on the findings reported here and in
previous Westinghouse reports,^»6 we expect that disposal
of FBC residue will not be a problem In the commercializa-
tion of the FBC process.
SPENT SORBENT PROCESSING
We have reported processing alternatives previously^*" and have dem-
onstrated that stable, solid compacts could be produced from blends of
FBC spent sorbent and coal ash at ambient temperature and pressure. The
work reported here on spent sorbent processing presents a preliminary
engineering assessment of fixation into compacts as a means of protecting
the environment, data on the potential long-term stability of compacts,
and an assessment of the possibility of using the residues directly as
aggregate or concretes.
An assessment of the potential for reducing the enviromental impact
of FBC residues by processing them at low temperature into compacts was
-------
carried out on the basis of leachate results from unprocessed and pro-
cessed materials. Calcium leached was used to represent leaching char-
acteristics. Results show:
• Batch leaching behavior for 5-cm cubes made from blends of
spent bed material and carry-over can be predicted at
least to 1080 hours by an empirical relation of the form
LT - L0 (1 - e~b]-N ) + b2N ,
where
Lj = cumulative calcium leached, rag-moles
N = number of 72-hr extraction, and LQ, bj_, \>2 are
constants.
• For large masses the leaching rate/rn^ of exposed surface
is more meaningful than the rate/g. For such masses the
amount of material leached is constrained by the amount of
local precipitation. Leaching of major constituents
(Ca"1"*, S04=, OH~ ) from such masses includes an initial
transient that in most situations will probably not be an
environmental problem.
• Processing FBC residues to 5-cm cubes reduced the leaching
rate of calclum/g of sample by a factor of 5.
• The concentration of leachates will be unsaturated and
below 10 times the drinking water standard (10 x DWS) for
unprocessed and processed material, with the time for com-
plete solution of residue from a 200 MW plant requiring
>3000 years.
A. freeze-thaw test was selected to provide a screening test for sta-
bility. The current effort did not permit a comprehensive Investigation,
and, thus, only isostatlcally pressed compacts were used to provide an
-------
estimate of limiting performance capability. Specimens were made from
ground spent socbent and were subjected to 300 cycles of freeze-thaw
exposure at -18 to +4°C. Results show:
• Spent bed material from the FBC of bituminous coal can be
processed to large masses that are resistant to degrada-
tion by freeze-thaw cycling.
• Mixtures of spent bed material and carry-over may have to
be preslaked either separately or jointly in order to
achieve the same results indicated above.
• Further work is needed to identify the nature of the bond
in the processed spent bed material.
Utilization of FBC residue directly as an aggregate was investi-
gated. Results show:
• Spent bed material containing up to 39 wt % total CaO can
be used as fine aggregate In conjunction with sintered
fly ash to make concrete mortar with a corapressive
strength in the range of normal concrete.
• Briquetting of FBC residue has the potential for use us an
aggregate.
10
-------
3. RECOMMENDATIONS
The FBC residue studies were initiated in 1974 and subsequently
expanded into the current comprehensive program encompassing two major
areas:
• The environmental impact assessment of direct disposal and
disposal after further processing
• Residue processing for disposal and for utilization.
ASSESSMENT OF ENVIRONMENTAL IMPACT
An understanding of the environmental impact of residue disposal is
critical to the successful commercialization of the FBC process. Based
on the studies reported in this and previous Westinghouse reports,
further investigations are recommended:
• The impact of RCRA should be fully understood. An under-
standing of the impact of future regulations such as the
radioactivity and bioaccuraulation characteristics stated
in the Advance Notice in the Federal Register
(December 18, 1979) for hazardous waste should be devel-
oped. Standards and guidelines for hazardous and nonhaz-
ardous waste disposal facilities to be promulgated under
RCRA Sections 3004 and 4004, respectively, deserve special
attention because of their potential economic impact on
solid residue disposal.
• Laboratory-scale screening of residue disposal should con-
tinue in order to determine environmental impact as a
function of process variations and to fill gaps in the
data, in particular by using residues from those systems
from which representative samples were unavailable during
this study. Examples of these system variations are FBC
11
-------
with sorbent regeneration, adiabatic PFBC operation, and
residue from FBC using western coals. In addition, the
carry-over fines collected after the secondary cyclone
require further study since some trace elements have
appeared in the leachates from such materials at levels
exceeding the DWS.
• A methodology should be developed to evaluate the environ-
mental impact of disposal at specific sites based on the
site-specific hydrology, geology, climate, and soil compo-
sitions. Such methodology will simplify the site-
selection and disposal facility design tasks.
• Perspective should be provided on the role of leachate
attenuation by soil or other disposal media.
• The effect of acidic leaching medium is recommended.for
further investigation.
• The potential advantages and disadvantages of codisposal
of FBC waste, which is highly alkaline, with acidic waste,
such as coal-cleaning waste, should be assessed.
• A systematic engineering evaluation of the alternative
methods of handling and disposing of the FBC residue is
required to provide a basis for making the optimal eco-
nomic choice and for meeting the environmental criteria.
• The environmental impact of processed FBC residue should
be investigated further as improved and/or additional pro-
cessing or utilization options are developed.
SPENT SORBENT PROCESSING
An understanding of spent sorbent processing is important for
developing an understanding of the choices available for minimizing the
12
-------
environmental impact of disposal and for utilizing the residue to
achieve optimal resource recovery or economic benefit. Specific areas
require further study:
• Analysis of environmental impact reduction using low-
temperature compacts should be extended to improve the
ability to relate test results to disposal site
conditions.
• The long-term stability tests should be extended to
include additional FBC residue characteristics (e.g.,
fines, coal source, etc.) to incorporate other criteria
(e.g., autoclave expansion and sulfate resistance), and to
include alternative low-temperature processed materials
(e.g., compacts formed without pressing).
• The development work on utilization of residue as an
aggregate should be extended to assess the impact of resi-
due characteristics (e.g., fines concentration), to under-
stand the mechanism of bonding responsible in compacts for
the unusually high compressive strengths, and to implement
further tests that would permit commercial utilization of
FBC residue as an aggregate (e.g., abrasion resistance and
concrete cylinder tests).
13
-------
4. ENVIRONMENTAL CONSTRAINTS
The environmental acceptability of FBC residue disposal is dic-
trated by the constraints of environmental laws and regulations. This
section reviews the federal laws and regulations as well as the criteria
selected for this investigation.
REGULATIONS/CRITERIA
The federal laws having legislative power over the environmental
impact of solid x^aste disposal are:
• Resource Conservation and Recovery Act (RCRA), 1976
- Solid Waste Disposal Act, 1965
- Resource Recovery Act, 1970
• Clean Water Act, 19772
- Federal Water Pollution Control Act, 1972
- Safe Drinking Water Act, 1974
4
• Clean Air Act Amendments, 1977
- Clean Air Act, 1970
• Toxic Substances Control Act (TSCA), 1977
• Occupational Safety and Health Act (OSHA), 1970
• Marine Protection, Research and Sanctuaries Act, 1972.
Of the above, those that most affect solid waste disposal are RCRA
and the Federal Water Pollution Control Act of 1972 (Public Law 92-500,
1972, as amended by Clean Water Act, P.L. 95-217, 1977).2'3 Eventually,
disposal guidelines are to be promulgated by EPA under the authority
of the former.
The passage of RCRA closed the legislative loop of environmental
laws (air/water/solid) and created a new level of control over solid
14
-------
waste disposal. Of special concern are the regulations to be promulgated
ist€
10
7-9
under Subtitle C - Hazardous Waste Management and Subtitle D - State
and Regional Solid Waste Plans.'
9
On December 18, 1978, EPA issued in the Federal Register the pro-
posed rules under RCRA Section 3001 (identification and listing of
hazardous waste), Section 3002 (standards applicable to generators of
hazardous waste), and Section 3004 (standards applicable to owners and
operators of hazardous waste treatment, storage, or disposal facilities).
These proposals, together with those already published pursuant to Sec-
tion 3003 (standards applicable to transporters of hazardous waste,
April 28, 1978, 43 FR 18506-18512), Section 3006 (authorized state
hazardous waste programs, February 1, 1978, 43 FR 4336-4373), Sec-
tion 3008 (federal enforcement, August 4, 1978, 43 FR 34738-34747), and
Section 3010 (effective date, July 11, 1978, 43 FR 29908-29918) and
that of the Department of Transportation pursuant to the Hazardous Mate-
rials Transportation Act (May 25, 1978, 43 FR 22626-22634), along with
Section 3005 (permits for treatment, storage, or disposal of hazardous
waste), constitute the hazardous waste regulatory program under Sub-
title C of the Act.
According to the proposed rules under RCRA Section 3001, those
characteristics that identify a waste as hazardous are its ignitability,
corrosivity, reactivity, and toxicity. Also proposed in the same publi-
cation is the "Advance Notice for Proposed Rulemaking" which requests
information and comments on the additional characteristics being con-
sidered for identifying hazardous waste — radioactivity, genetic activ-
ity, bioaccumulation, and additional aspects of toxicity. A set of
tests was proposed for each of the above hazardous characteristics.
Among them the test for toxicity (extraction procedure, EP) received the
most attention and was most relevant to FBC residue disposal. Tables 2
and 3 summarize the development of this procedure.
15
-------
Table 2
DEVELOPMENT OF RCRA 3001 REGULATION
Draft
Regulations
Proposed
Regulations
Key Issues
Westinghouse Actions
March 1978
Toxic Extraction Procedure (TEP),
drafted for hazardous waste
identification
Sept. 1978
TEP significantly modified, to be
renamed extraction procedure, EP
TEP initiated on selected samples
• to assist RCRA Sec. 3001
development
• to provide initial indication
of "hazardous" or "non-
hazardous" nature of residues.
Nov. 1978
Dec. 18, 1978
Extraction Procedure (EP)
proposed - significantly dif-
ferent from the previous TEP
A structural integrity test
(SIT) specified for monolithic
block.
"Special Waste" Category created
(including utility waste):
subject to partial exemption
of hazardous waste regulations.
Advance notice of Proposed Rule-
making on radioactivity and
bioassay
EP initiated on selected FBC ref-
erence materials and raw sorbents,
-------
Table 3
DEVELOPMENT OF EPA LEACH TEST
f
Parameters
Sample Size
Sample Preparation
Leaching Medium
Titrating Agent
Maximum Titration
Final pH
Temperature
Extraction Time
Solid/Liquid Ratio
Agitator
Draft Date
March 1978
Not specified
Grinding to 3/8 in.
Deionized water with NaOH
or acetic acid added
IN NaOH or 1:1 acetic
acid
No Limit
4.9-5.2
Room temperature
2 x 24 hr = 48 hr
total
1:10 for each extraction,
plus original liquor
Not specified
Sept. 1978
>_ 100 g
Grinding to 3/8 in., or SIT
hammer test on monolithic block
Deionized water with 0.5N
acetic acid added
0.5N acetic acid
Maximum = 4 ml/g solid
4.9-5.2 or controlled by
maximum acid allowed
20-30°C
24 hr
Single extraction
1:20 plus original liquor
Not specified, but overhead
stirring suggested
Proposed in
Fed. Reg.
Dec. 18, 1978
>_ 100 g
Same as
Sept. 1978
Same as
Sept. 1978
Same as
Sept. 1978
Same as
Sept. 1978
Same as
Sept. 1978
20-40°C
Same as
Sept. 1978
Same as
Sept. 1978
Same as
Sept. 1978
-------
Also under the proposed rules of RCRA Sec. 3001, there are two
mechanisms by which to identify a waste as hazardous:
• Inclusion of the substance on the predetermined hazardous
waste list of processes and sources
• Failure of the substance to meet the proposed test cri-
teria for any of the identified hazardous characteristics.
FBC residue is not on the list. Should the identifying tests determine
it to be hazardous (and preliminary results indicated that it would not
be), in all likelihood it would be classified as a "special waste" and
be subjected to partial exemption of RCRA Subtitle C - hazardous waste
regulations. Should testing determine FBC residue to be nonhazardous,
(as is predicted), the disposal would be subject to the "Proposed
Classification Criteria for Solid Waste Disposal Facilities" under RCRA
id (
11
Sec. 4004 and the "Proposed Guidelines for Landfill Disposal of Solid
Waste" under RCRA Sec. 1008."
In addition to any controls that may be imposed upon FBC residue
in connection with RCRA, some requirements may also result from the
Federal Water Pollution Control Act (amended as the Clean Water Act,
2 3
1977). ' The primary environmental concern with solid waste disposal is
the potential ground and surface water contamination caused by leachate
runoff or seepage. The federal regulation that most nearly relates to
a limit on seepage water quality is the EPA's "Alternative Waste Manage-
12
ment Techniques for Best Practical Waste Treatment" under the authority
3
of the Federal Water Pollution Control Act of 1972 amended by the Clean
2
Water Act of 1977. These criteria state that the contaminant levels
in the groundwater under a disposal site shall be limited to the maximum
contaminant levels contained in the National Interim Primary Drinking
13
Water Regulations (NIPDWR) or to the existing concentration if the
latter is greater. If the groundwater is to be used for other than a
drinking water supply, the groundwater criteria should be established by
the Regional Administrator.
18
-------
Because of the wide variation in the characteristics of solid wastes
in general, weather, soils, topography, groundwater from site to site,
and nearby stream quality and flow characteristics, solid waste disposal
permits are currently being awarded on a site-specific basis. Eventually,
as a result of the RCRA, state regulations will apply, but these regula-
tions will not be enacted until federal standards are promulgated.
Depending on the actual site selected for disposal, the resultant water
would have to meet the water quality criteria for the specific water
14
use. Furthermore, the success of a land disposal application depends,
above all, on the design, construction, and operation of a specific dis-
posal site based on the geology, hydrology, and meteorology of that
particular site.
TEST PROGRAM CRITERIA
A series of laboratory leaching tests on FBC residue is reported
here. In order to have some criteria that could be used to provide a
preliminary indication of the potential acceptability of the observed
leachates from these laboratory studies, Westinghouse decided to compare
the observed leachate concentrations with available drinking water
standards/guidelines/criteria. These drinking water standards include
the NPIDWR, United States Public Health Service (USPHS)15 Drinking
Water Standards, and the World Health Organization (WHO) Potable Water
Standards. In addition, although the guidelines for power plant
effluents developed by EPA are not applicable to the disposal of dry
spent sorbent from the fluidized-bed combustion process, they are used
as additional references in this investigation. Table 4 lists the
selected water quality criteria for comparison of leachate.
One should note that the drinking water standards are used as the
criteria for comparison in this investigation only in an effort to put
data into perspective, in the absence of formal EPA guidelines; this
approach should not be construed as suggesting that the leachate must
necessarily meet drinking water standards. This use of the drinking
19
-------
Table 4
SELECTED WATER QUALITY CRITERIA
Substance
Ag
As
Ba
Ca
Cd
Cr
Cu
Fe
Hg
Mg
Mn
Ni
Pb
Se
Sn
Zn
so4
Cl
N03
F
pH
(pH unit)
TDS
Drinking Water Standards, mg/£
NIPDWR13
0.05
0.05
1.0
0.01
0.05
0.002
0.05
0.01
10 (as N)
1.4-2.4
USPHS
0.05
0.05
1.0
0.01
0.05
1.0
0.3
0.05
2.0
0.05
0.01
1.0
5.0
250
250
45
1.7
500
WHO16
Highest
Desirable
Level
0.05
75
0.01
0.05
0.1
0.001
30
0.05
0.1
0.01
5.0
200
200
45
1.7
7.0-8.5
500
Maximum
Permissible
Level
0.05
200
0.01
1.5
1.0
0.001
150
0.5
0.1
0.01
15
400
600
45
1.7
6.5-9.2
1500
Effluent
Guidelines for
Standards for
Steam Electric
Power
Generation, 17 mg/£
0.2
1.0
1.0
1.0
6.0-9.0
20
-------
water standards for the purposes of this study is conservative; it
assumes that no attenuation or dilution of the leached contaminants will
occur in the substrate beneath the disposal pile, or in the groundwater.
By comparison, according to the extraction procedure proposed under
Sec. 3001 of RCRA to determine whether a substance is to be considered
hazardous due to toxicity, a criterion of ten times the NPIDWR is used;
it assumes that the leachate will be diluted by a factor of 10 between
the disposal pile and the receptor (e.g., a well tapping the groundwater)
Another leachate concentration goal available is the Multimedia
Environmental Goals (MEG) being developed by EPA's Industrial Environ-
mental Research Laboratory in the Research Triangle Park, N.C.
(IERL-RTP). ' These independent, very conservative goals are, in
their simplest form, based upon relatively simple manipulation of avail-
able Threshold Limit Value (TLV) and Lethal Dose 50 (LD^) data.
EFFECT OF RESIDUE PROCESSING
While it may be technologically simpler and economically less
costly to dispose of FBC residues directly, there is reason to believe
that some on-site processing can reduce the environmental impact. The
objective of the processing may be to place the product in a suitable
disposal site or to use the product in road construction, in cement
block, in agriculture, or in other ways. In either case an assessment
of the product's conformity to environmental regulations will be
required.
21
-------
5. FBC SOLID RESIDUES
There are two major sources of solid residue from an FBC system:
spent bed material and carry-over. The former consists largely of spent
limestones or dolomite sorbent (calcium sulfate [CaSCyJ, calcium oxide
[CaO], calcium carbonate [CaC03], magnesium oxide [MgO]); and the latter
consists of finer particles of spent sorbent and coal ash carried out
from the bed and removed by a particle control device, such as a
cyclone.
RESIDUE CHARACTERISTICS
Residue characteristics were projected and compared with actual FBC
solids in the previous EPA report. The projected spent sorbent compo-
sitions for three basic fluidized-bed combustion concepts are summarized
in Table 5.
The size distribution of spent sorbent from the bed will be similar
to the sorbent feed size distribution. The spent sorbent fines appear-
ing in the fly ash will depend upon sorbent attrition rate, bed elutria-
tion rate, and fines recycle (if applied). Practically all the coal ash
will be elutriated from the combustor. The quantity of sorbent fines is
estimated to range from 0.25 to 1 times the coal ash content of the fly
ash for a nominal 10 percent ash coal.
Table 6 summarizes the approximate distribution ranges of spent
sorbent and coal ash in the different sources of FBC residue (bed, carry-
over) investigated in this study. Table 7 summarizes the composition
range of the spent calcium-based sorbent in these solids. The individual
samples investigated during this study and their chemical compositions
will be presented in greater detail in Section 6 of this report.
22
-------
Table 5
Dwo. J7D9C75
PROJECTED SPENT SORBENT COMPOSITIONS FROM THREE BASIC FBC SYSTEMS
14
Process
Atmospheric- Pressure
FBC
Once-through
100% load
Pressurized Boiler
Once-through
100% load
Adiabatic Combustor
Once-through
100% load
Pressurized Boiler
Once-through
100% load
Atmospheric Pressure
FBC
One-step regenera-
tion 100% load
Pressurized Boiler
Once-through
turndown to
minimum load
Sorbent
Limestone
Dolomite
Dolomite
Limestone
Limestone
Dolomite
CaS04
25
(43.7)
80
(64.1)
50
(46.7)
40
(60.6)
24e/12.8f
(43. 1/26. 11
60
(47.6)
Spent Sorbent Composition,
mole%(wt%)
CaS
0
(0)
0
(0)
0
(0)
0
(0)
06/1.2f
(0/L3)
0
(0)
CaO
75
(54)
20
(6.6)
50
(19.2)
60
(37.4)
766/86f
(54.6/70.1)
0
(0)
CaCOj
0
(0)
0
(0)
0
(0)
0
(0)
0
(0)
40
(23.4)
MgO
ob
(0)
1.19C
(28.1)
L19C
(32.7)
ob
(0)
f
(0)
1.19C
(27.8)
Balance
173d
(2.31
2.11d
(1.2)
2.11d
(1.4)
1.73d
(2)
1.73d
(2.3/2.5)
2.11d
(1.2)
Basis: Limestone 1359, Dolomite 1337 and Ohio-Pgh No. 8 coal Grams/mole calcium
MgO included with balance of components
cMoles MgO/mole Ca
Spent sorbent from combustor
Spent sorbent iron; regenerator
In addition to the unprocessed AFBC and PFBC residues described in
Table 8, we have also included processed FBC residues in the form of
solid compacts (processed by the sorbent/ash blending option) to deter-
mine their environmental impact. The processing of FBC residue to mini-
mize leachability by reducing the solid/water contacting surface is
discussed in Section 8 of this report.
23
-------
Table 6
DISTRIBUTION OF SPENT SORBENT/COAL ASH IN FBC BED AND CARRY-OVER
Residue Source
Bed Material
Carry-over
(from primary particulate
control device, e.g., cyclone)
Carry-over,
(from final particulate
control device, e.g.,
3rd-cyclone or fabric filter)
wt % Spent Sorbent
60-90
20-50
10-40
wt % Coal Ash
10-40
50-80
60-90
Table 7
SUMMARY OF COMPOSITION OF SPENT SORBENT FROM
FBC SYSTEMS
AFBC
PFBC
PFBC
PFBC
Process
Composition
(Ca-based
CaSO^
, Once-through 0
, Once-through 0
, Regenerative 0
, Adiabatic 0
.10-0.
.10-0.
.10-0.
.40-0.
50
70
40
70
molar
CaO
0.50-0
0-0
0.60-0
0-0
.90
.30
.90
.30
fraction)
CaCO
0-0
0.10-0
0-0
0-0
3
.10
.80
.10
.30
CaS
<0.
<0.
<0.
<0.
01
01
01
01
RESIDUES TESTED FOR ENVIRONMENTAL IMPACT
Without Processing
To assess the environmental impact of FBC solid waste disposal, char-
acterization, leaching, and activity tests were performed on carry-over
and spent bed materials obtained from experimental FBC units operated
24
-------
20 21
by Argonne National Laboratories (ANL), Exxon, Pope, Evans and
22 23
Robbins (PER), Morgantown Energy Research Center (MERC), Babock and
2A 25
Wilcox, Inc. (B&W), Battelle - Columbus Laboratories, and Combustion
O f
Power Company, all of which are conducting FBC development studies
under the Department of Energy (DOE), EPA, or the Electric Power Research
Institute (EPRI). The FBC residues investigated during this work include
spent bed material and carry-over from both the pressurized and atmos-
pheric units operating with either once-through injection of sorbent or
with sorbent regeneration.
Table 8 summarizes the samples studied under this contract and the
FBC process conditions under which these samples were produced. For the
purpose of clarification, spent bed material is sometimes referred to as
spent sorbent because the bed material consists predominantly of partially
utilized sorbent. The carry-over material is often referred to as fly
ash (or ash) and fines; the former is carry-over collected in the primary
particulate removal system (first- and second-stage cyclones), and the
latter refers to the fines collected from additional particulate filtering
devices, such as bag filters, additional cyclones, or stack-gas sampling
devices.
After Processing
As part of the effort to identify and evaluate processing methods
for reducing the environmental impact of FBC residues, we subjected
selected residues to leach testing after processing them into 5-cm cubes.
The residues are listed in Table 8 and the specimens tested are in
Table 9. The latter were all made from Exxon Run 27 spent bed material
and carry-over. Other materials used for comparison tests were natural
gypsum and unprocessed residues from Exxon Run 19.6.
25
-------
^, 0 Dwg. 2626C07
Table 8
PROCESS CONDITIONS OF SAMPLES STUDIED FOR THEIR ENVIRONMENTAL IMPACT ON DISPOSAL
Conditions
Coal
Sorbent
Run Length, hr
Pressure. kPa
Avg Bed Temp. . °C
Lower Bed Temp. ,°C
Gas Velocity, m/s
Expanded Bed Height, m
Settled Bed Height, m
Coal Feed Rate, kg/h
Sorbent Feed Rate, kg/h
Ca/S Molar Ratio
Excess Air
SO- Emission, ppm
N0x. ppm
CO. ppm
co2. %
o2.%
S Retention. %
Lb SOJM Btu
Lb NO^M Btu
Argonne
C2/C3
Arkwright
Tymochtee
dolomite
23
810
900-955
_d
1.61-1.67
-
-
-
-
1.1-1.5
-
80-375
135
-
-
3.0
82-%
-
-
Argonne
VAR-4
Arkwright
Tymochtee
dolomite
11
810
900
-
0.7
0.91
-
7.9
2.6
1.9
17
122
185
50
18
3.0
95
0.23
0.25
F
Argonne
REC-3
Arkwright
Tymochtee
dolomite
80.6
810
900
-
0.7
0.91
-
13.5
2.9
L5
17
450
120
64
16
3,2
79
0.85
0.16
'FBC Samples '
Argonne
CCS-10
Triangle
Tymochtee
dolomite
2.6
152
1100
-
1.24
0.46
-
-
-
-
Reducing
67.000
-
-
-
-
-65*
Regeneration
-
—
Argonne
LST-1
Arkwright
Limestone
2203
34.7
810
870
-
0.76
0.91
-
12.1
1.7
1.5
17
900
150
74
16
3.5
63
1.5
0.18
Argonne
LST-2
Arkwright
Dolomite
1337
25.5
810
870
-
0.76
0.91
-
11.7
3.6
1.8
17
160
135
50
15
3.1
93
0.28
0.17
Argonne
LST-3
Arkwright
Dolomite
1351
1&4
810
870
-
0.76
0.91
-
11.7
3. 1
1.4
17
270
185
30
16
2.9
89
0.47
0.23
Argonne
LST-4
Arkwright
Limestone
1336
17.5
810
870
-
0.76
0.91
-
11.6
1.5
1.4
17
990
95
35
17
3.0
58
1.7
0.12
Identification of experimental FBC units : see text.
Third combustion experiment in ten-cycle combustion/regeneration series of experiments.
Tenth regeneration experiment in ten-cycle combustion/regeneration series of experiments.
d
Dash indicates data unavailable.
-------
Table 8 (Continued)
Dwg. 2626C08
Conditions
Coal
Sorbent
Run Length, hr
Pressure, kPa
Avg Bed Temp. , °C
Lower Bed Temp. ,°C
Gas Velocity, m/s
Expanded Bed Height, m
Settled Bed Height, m
Coal Feed Rate, kg/h
Sortent Feed Rate, kg/h
Ca/S Molar Ratio
Excess Air
SO. Emission, ppm
NO^ ppm
CO. ppm
co2. *
Oy*
S Retention. %
Lb SO^M Btu
Lb NO_/M Btu
PFBC Samples3
Exxon
8.4
Arkwright
Grove limestone
1359
11
906-907
_c
877-908
1. 77-1. 83
-
0.66-L19
75-112
10. 3-15. 2
1.67
18-72
-
50-200
-
-
—
62
1.8
Exxon
19.6
Champion
Grove limestone
1359
7.5
930
880-888
-
2.01-2.04
-
1.58
113-163
-
2.5
15
500
104
-
11.7-12.3
2.5-3.0
68
1.0
0.14
Exxon
26
Champion
Grove limestone
1359
15.5
930
885-927
949
1. 9-2. 1
-
1. 12-2. 28
130
-
3.7
9. 5-11. 5
140-300
180-185
50
13
L 8-2. 15
81-91
0. 29-0. 59
0. 25-0. 28
Exxon
27
Champion
Pfizer
dolomite 1337
240
930
829-930
840-960
1. 7-2. 2
3-7
-
112-149
-
0-2.5
8-23
20-1290
70-210
30-110
1. 1-17
1. 5-3. 9
41-100
0.03-2.5
0. 12-0. 30
Exxon
30.2
Champion
Grove limestone
1359
8.5
920
929
945
2.5
-
-
137
-
3.7
17.2
137
-
45
15.1
3.1
89
-
Exxon 30. 4
(via RS)a
Champion
Grove
Limestone
6
920
835
833
1.9
-
2.26
120
-
3.7
16.1
894
70
67
14.4
2.8
-
-
Exxon
34
Champion
Pfizer
dolomite 1337
13.25
932
900
868
1.5
-
2.29
90
-
0.75
20.9
100-300
52
61
15.5
3.5
-
-
Exxon 43.2
Illinois No. 6
Pfizer 1337
Dolomite
7
870
845
850
1.7
5.7
2.1
79
-
0.75
42
720
120
250
14.0
6.1
-
-
Exxon 43. 3
Illinois No. 6
Pfizer 1337
Dolomite
6
885
940
950
1.77
3.13
1.9
93
-
0.75
25.2
870
90
240
12.3
4.2
63.1
2.39
0.178
N3
Identification of experimental FBC units: see text.
Sample obtained from Ralph Stone and Co. ^
C0ash indicates data unavailable.
-------
Table 8 (Continued)
Dwg. 2626C09
Conditions
Coal
Sorbent
Run Length, hr
Pressure. kPa
Avg Bed Temp. . °C
Lower Bed Temp. . °C
Gas Velocity, m/s
Expanded Bed Height, m
Settled Bed Height, m
Coal Feed rate, kg/hr
Sorbent Feed Rate
kg/hr
Cats Molar Ratio
Excess Air. %
SO,, Emission, ppm
NOX. ppm
CO, ppm
coz. %
0?*
S Retention. %
Lb SOyM Btu
Lb NO^M Btu
PFBC Samples a
Exxon 43. 5
(via RS)3
Illinois No. 6
Pfizer 1337
Dolomite
4
940
945
-
-
-
2.18
95
_
0.5
25
1330
75
200
14
4
—
-
—
Exxon 45
Combustor
Champion
Grove Limestone
1359
120
780
630-900
705-8«
L19-L40
2. 92-3. 59
-
79-85
0-6.4
0-1.3
Avg =0. 55
27. 4-39. 5
107-635
-
-
-
4.4-58
77
-
—
Exxon 45
Regenerator
_c
-
100
770
1010
-
0.61
2.3
-
-
Solid recircu-
lation rate =45
-
-
Avg =0.53 m*
-
-
-
-
—
-
—
Exxon 67
Illinois No. 6
Pfizer 1337
Dolomite
100
930
875-915
-
3.23
-
-
122
_
1.25
19.5
640
84
243
13.25
3.25
—
-
—
Comb. Power
P-403
Illinois No. 6
Kaiser
Dolomite
175
405
855
-
-
-
-
-
_
-
-
-
50.2
-
6.2
-
—
-
—
AFBC Samples a
B & W No. 19
Pittsburgh No. 8
Lowellville 14 Limestone
(Carbon Limestone Co.)
25
98.2
843
843
2.13
-
0.38
227
77
3
-
839
285
1000
14.1
2.7
—
•
—
Battelle
(Erosion/Corr.Run)
Illinois. No. 6
Grove
Limestone
1500
101
877
-
2.3
-
1.23
69
20.5
1.8
-
520
-
-
-
6.8
-75
0.9
"~
MERC
(3/9/77)
Arkwright
Greer
Limestone
-100
101
838
838
1.13
0.66
0.40
11.9
( premixed)
4.4
3.5
-
30±20
400
-300
-10
-8
-
-
—
PER
Sewickley
Grove
Limestone
-
10L3
816
-
2.7-4.6
-
0.3-0.9
272-363
0-182
-
-
-
—
-
-
3
—
-
—
PER
Shakedown
(viaRS)b
Sewickley
Greer
Limestone
6
101
982
-
-
-
0.61
191
_ .
3
23
2000
350
700
13
4.5
-
-
—
CD
Identification of experimental FBC units: see text.
Sample obtained from Ralph Stone and Co. 27
A dash indicates data unavailable.
-------
Table 9
Dug. 261ZC19
PROCESSED FBC SPECIMENS
Sample
I1C
I ID
I2C
120
I3C
no
I4C'
I4D1
II 1C
II ID
II 3C
II 3D
II 3£
II 3F
II4C
II 40
II -1C'
II 40'
I112C
III2D
Exxon 19.6
Bed Material
Exxon 19.6
Fly Ash
Exxon 27
Sorbent/Ash
Mixture
Gypsum
Description
Wt ratio fly ash/total solid =0.358
Wt ratio H_0/total sol id = 0.445
Air cured 7 days
Same as 1 1C
Wt ratio fly ash/total solid = 0.358
Wt ratio H20/total solid = 0.445
Air cured II days
Same as 1 1C
Wt ratio fly ash/total solid =0.358
Wt ratio HjO/total solid = 0. 445
Air cured 28days
Same as 1 3C
Wt ratio fly ash/total solid = 6'. MO
Wt ratio H20/total sol id =0.445
Air cured 60 days
Same as 1 4C'
Wt ratio fly ash/total solid = 0.100
Wt ratio HjO/total solid = 0.300
Air cured 7 days
Same as II ID
Wt ratio fly ash/total solid -0.100
Wt ratio HjO/total solid = 0.300
Air cured 28 days
Same as II 3C
Wt ratio fly ash /total solid =0.100
Wt ratio water/total solid = 0.445
Air cured 28 days
Same as II 3E
Wt ratio fly ash/total solid =0.100
Wt ratio water/total solid = 0.300
Air cured 60 days
Same as II 4C
Same as II 4C
Same as II 4C
Wt ratio fly ash /total solid =0.15
Wt ratio H.O/total solid = 0.3
Air cured 21 days
Same as III 2 C
Unprocessed (Spent limestone)
Unprocessed
Wt ratio Bed/ash = 80/20,
Unprocessed mixture
Natural
Leach ing Method
1242-hr continuous shake.
wt ratio solid/ water =1/10
15 x 72-hr intermittent shake,
wt ratio sol id/ water =1/3
1080-hr continuous shake.
wt ratio solid/water =1/10
15x 72-hr intermittent shake.
wt ratio sol id/ water =1/3
1080-hr continuous shake.
wt ratio sol id/ water =1/10
15 x 72-hr intermittent shake.
wt ratio sol id/ water = 1V3
15 x 72-hr intermittent shake.
compact crushed to powder.
wt ratio solid/water = 1/3
15 x 72 hr intermittent shake.
original cube iuncrushedl.
wt ratio sol id/ water =1/3
1224-hr continuous shake.
wt ratio solid/water = 1/10
15 x 72-hr intermittent shake.
wt ratio sol id/ water =1/3
1080-hr continuous shake,
wt ratio solid/water = 1/10
15 x 72-hr intermittent shake.
wt ratio sol id/ water =1/3
1080-hr continuous shake.
wt ratio solid/water = 1/10
15x 72-hr intermittent shake.
wt ratio sol id/ water = 1/3
1080-hr continuous shake,
wt ratio solid/ water = 1/10
15 x 72-hr intermittent shake.
wt ratjo solid/water = 1/3
15x72-hr intermittent shake.
compact crushed to powder.
wt ratio solid/water = 1/3
15 x 72 hr intermittent shake.
original cube < uncrushedl.
wt ratio sol id/ water = 1/3
HltO-hr continuous shake,
wt ratio sol id/ water = I/ 10
15 x 72-hr intermittent shake.
wt ratio solid/water = 1/3
15 x 97-hr intermittent shake.
wt ratio solid/ water = 1/3
Same a& above
Same as above
Same as above
29
-------
6. EXPERIMENTAL TEST PROGRAM: ENVIRONMENTAL IMPACT
EXPERIMENTAL METHODS
The environmental impact of any material disposed of is a function of
its physical and chemical properties as well as of the quantity involved.
Potential water pollution problems can be predicted from the chemical
characteristics of leachates, such as pH, specific ion concentrations,
trace element dissolution, and total dissolved solids (IDS). Disposal
of the solid wastes from the fluidized-bed coal combustion process may
also create air pollution, odor nuisance, and heat-release problems. To
assess the environmental impact of FBC solid waste disposal and the
suitability of waste material as landfill, we investigated the physical
and chemical characteristics of the residue and its leaching and heat-
release properties.
Characterization
Chemical, physical, and morphological characterization of the spent
bed and carry-over material was carried out. The methods employed
included optical microscopy, scanning electron microscopy (SEM), energy
dispersive analysis by X-ray (EDAX), electron microprobe analysis (EMA),
X-ray diffraction, thermogravimetric analysis (TGA), emission and atomic
absorption spectroscopy, and wet chemical methods.
Leaching Tests
At the time that this study was conducted, no standard EPA leaching
test existed for assessing the potential environmental contamination from
a solid waste. Since the time that these tests were conducted, an extrac-
tion procedure (EP) has been proposed by EPA under Sec. 3001 of RCRA
9
(Federal Register^ December 18, 1978) for determining whether a residue
is hazardous because of toxicity. The proposed EP specifies that the
residue sample be agitated in a container of buffered acetic acid solution
30
-------
for 24 hours, with up to 4 ml of 0.5N acetic acid per gram of solid
added during the shaking period in an effort to maintain the pH at 4.9
to 5.2. The residue would be considered toxic if any trace metal element
exceeded 10 times the NIPDWR. Westinghouse will conduct some tests in
the future that employ EP under their continuing contract with EPA.
Parallel to the EPA effort to develop the EP, ASTM committee 19.12
(subcommittee 19.1203) is also developing a standard leaching test for
solid waste materials. It proposes a 48-hour shake method using either
type IV reagent water (ASTM D-1193) or pH = 4.5 sodium (Na) acetate-
28
acetic acid buffer.
In the study covered in this report, leachates were induced by the
shake test developed by Westinghouse except where otherwise specified.
Samples of waste stones were mixed with deionized water in Erlenmeyer
flasks at room temperature. An automatic shaker at 70 excursions per
minute was used to agitate the mixtures. Among the parameters investi-
gated were sorbent/water loading, sample mixing time, and pH of the
leaching medium. The supernatants resulting from this operation were
filtered, and the filtrate was determined for pH, specific conductance,
TDS, calcium, magnesium, sulfide, sulfate (SO,), trace metal ion and anion
concentrations, and TOC. The solid samples before and after the leaching
operation were also analyzed for their chemical and physical character-
istics. Since CaSO, is a major constituent of the waste stone from the
fluidized-bed combustion process and leachates contained high calcuim
and sulfate concentrations, a naturally occurring gypsum was tested
under similar leaching conditions for comparison.
Two shake procedures have been employed. These are described below.
• Continuous shake test. It establishes equilibrium condi-
tions between the solid and its aqueous surrounding and
. provides the worst possible case with respect to contamina-
tion release. Westinghouse has used this method since
1975 as one of the screening tests for determining leaching
31
-------
properties of FBC spent solids. Typically, a 1:10 solid-
to-water ratio is used, and the mixture is shaken for up
to 400 hours.
• Intermittent shake test. A series of ten to fifteen cycles
of a 72-hour shake test was adopted as part of the leach-
ability study to provide leaching rate as a function of
time over a long period under worst case conditions.
Leachates are analyzed at the end of each interval, and a
fresh charge of deionized water is added for each 72-hour
leach cycle. Typically, a 1:3 solid-to-water ratio is
used.
Both shake tests are severer than conditions anticipated under
actual land disposal; results from the shake tests are expected to pro-
ject the worst case.
Activity Tests
No standard EPA activity test exists that can be applied to FBC
residues. Under Sec. 3001 of RCRA, EPA's Office of Solid Waste has pro-
posed some general tests to determine whether a waste is hazardous because
of reactivity. These tests concentrate on hazardous properties such as
explosiveness and chemical and mechanical instability, however, and do
7 29
not apply to residual lime. '
The activity of residual lime in spent FBC materials can be deter-
mined by its heat release property on contact with water, as the hydra-
30
tion reaction of CaO is extremely exothermic. Literature on lime
reactivity and slaking rate has been reviewed, including the ASTM C110
for slaking rate of quick lime (CaO), Murray's study of lime reactivity
as a function of porosity and shrinkage characteristics during calcina-
32
tion, and American Water Works' standard on lime for water treatment.
The heat release activity of FBC residue was measured calorimetrically
in this study. The temperature rise of a solid/water system containing
free CaO is a function of solid/water ratio. In our experimental effort
to establish a screening test for the residual activity in spent FBC
32
-------
solids produced under varying processing conditions, a solid/water pro-
portion of 3 g to 20 ml (which is in the bulk range specified by the
ASTM-C110 test and by Murray's work) was found empirically to provide
much better repeatability than that from a higher solid/water ratio that
would give greater magnitude of temperature rise but would lack repro-
ducibility, probably because of local heating. Higher solid/water ratios
were also used, however, because they provide higher sensitivity and
simulate rainfall onto the disposed solid.
Chromel-alumel thermocouples were used to monitor the temperature
rise in the stone/water system with an Omega cold junction compensator
and a millivolt recorder. The heat release tests were conducted on the
actual spent sorbent and on carry-over fines from the fluidized-bed
combustion units. Calcined and uncalcined limestone and dolomite sam-
ples were also tested for comparison.
TEST RESULTS
AFBC Residue Characterization and Leachate
Since we anticipate that the first generation of the commercial
fluidized-bed combustion process will be a once-through, atmospheric
system, we obtained several batches of spent AFBC bed and carry-over
materials for testing.
24
Babcock and Wilcox Residues
Under contract to EPRI, B&W has been operating a 3 ft by 3 ft AFBC
unit in Alliance, Ohio. Both the spent-bed and carry-over solids from
B&W run 19 were tested for the environmental impact of disposal.
X-ray diffraction which was used to identify the crystalline phases
in the solids showed that B&W 19 bed material consists of major quantities
of CaO and minor quantities of CaSO,. The carry-over consists of CaO
and silicon dioxide (SiO?) as major, CaSO, and hematite (aFe^O.,) as minor,
and calcite (CaCO~) and magnetite (Fe-0.) as trace species. These
results (at best semiquantitative) agree well with the AFBC composition
ranges (obtained by wet chemical analysis) shown in Table 5, Section 5.
33
-------
Results from the leaching tests are summarized in Table 10 for those chemi-
cal characteristics resulting from the leaching of the major species -
sulfated and nonsulfated calcined limestone. Lower initial values and
greater improvement of CaSO , pH, and specific conductance from the
carry-over leachate, in comparison with the spent bed leachate, can be
noted here, as would be expected.
The specific conductance is generally a good index for leachate
quality because the specific conductance of a leachate provides an
approximation of total dissolved solids (TDS). We must emphasize here,
however, that the correlation between specific conductance and TDS is not
absolute and depends on many factors, such as the molecular weight,
the valence, and the mobility of the dissolved species. As an approxi-
mation, a specific conductance of 1.5 millimho/cm is generally accepted
33
as equivalent to 1000 mg/£ of TDS. Figure 1 plots leachate specific
conductance as a function of total leachate volume and time. The dif-
ference between bed and carry-over leachate is clear. The Drinking
Water Standard (DWS), 500 mg/£ TDS, approximated as 0.75 millimho/cm, is
very much exceeded by the initial leachate of both the bed and carry-
over (2 to 9 millimho/cm) and gradually approached and passed by the
carry-over leachate after nine 72-hour leach cycles.
Because of the large number of residues to be studied during this
investigation, and because the carry-over solid and leachate generally
contain higher levels of trace elements (as will be illustrated for MERC
residue in a later section), the entire spectrum of solid and leachate
characterization was performed only on the carry-over material of B&W
run 19 residue. Results presented in Table 11 show that the trace ele-
ments meet the DWS and the TOC is below the detectable level, but calcium,
sulfate, pH, and TDS exceed the DWS. These results agree well with
those reported previously.
Heat release test results indicated higher residual lime activity
in the spent bed material than in the carry-over, as would be expected
with the higher CaO content in the spent bed. The heat release results
will be discussed in detail later under Thermal Activity Tests. Both
34
-------
Table 10
LEACHATE CHARACTERISTICS OF B&W 19 RESIDUES
Dxvq. 1 701 B
Residue
Bed
Carry-over
Chemical
Characteristics
H
Sp. Conductance
milli mho/cm
Ca. mg/X
S04, mg/1
PH
Sp. Conductance
millimho/cm
Ca, mq/l
S04, mqlt
Intermittent Shake.
no. of 72-hr intervals
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
12.0 12.2 12.1 12.2 12.2 12.2 12.1 12.2 12.2 12.0 12.2 12.4 12.2 12.0 12.0
8.48 8.72 8.34 8.7 8.22 8.78 8.33 9.00 7.75 8.64 8.69 8.05 7.66 7.04 7.95
1136 1404 1424 1384 1352 1400 1320 1480 1344 1396 1360 1084 1264 1012 996
1352 1094 1236 1273 1248 1298 1128 1325 1128 1298 1213 1353 1071 1175 1225
12.2 11.9 11.8 11.7 11.7 11.7 11.5 11.4 10.6 10.7 10.8 10.8 10.5 9.6 9.2
7.65 4.06 2.46 2.32 1.92 1.46 1.26 1.24 0.38 0.64 0.38 0.64 0.38 0.47 0.49
888 88 64 36 84 70 32 32 24 14 40 32 44 80 88
1325 122 0 9 11 15 15 25 50 75 88 75 74 175 216
Continuous
Shake, hr
100
12.1
9.36
1364
1175
12.0
8.1
1288
1273
200
12.2
9.24
1392
1209
11.9
6.47
1184
1248
-------
Curve b972H-6
= 4
Normalized Leachate Quantity, ml/gof Starting Solid
6 12 18 24 30 36 42
B & W 19 Bed
B&W 19 Fly Ash
2 4 6 8 10 12 14
n = Total Number of 72-hr Intermittent Leach
( 4
360 720
Total Leach Time = (72) (n) hr
1080
Figure 1 - Leachate Quality as a Function of
Intermittent Leaching
the AFBC bed and carry-over materials released more heat than did the
PFBC residue tested and reported in the previous report. This too, is
as expected, since the PFBC residue would be presumed to have a greater
fraction of its calcium content in the form of CaCO.,, which unlike CaO,
will not hydrate and release heat.
22 27
PER Residue '
Two PER residues produced from a 0.46-by-l.83-m AFBC unit funded by
the Department of Energy (DOE) were tested during this reporting period.
These were a spent bed material from a shakedown test and a carry-over
from a different run (origin not identified). Both residues were
27
obtained through Ralph Stone and Co. They had obtained the material
from PER and had also tested it. The shakedown bed residue was iden-
tified by X-ray diffraction to contain CaO, SiO , calcium hydroxide
36
-------
Table 11
D«q. 1709B93
SOLID AND LEACHATE CHARACTERISTICS OF B&W 19 CARRY-OVER
(200-hr Continuous Shake Test)
Substance
Al
Aj
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hq
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
SO*
S=
F
Cl
NO^Ias N
TOC
PH
SC. n mho/cm
Solid.
ppm
<1
200
5
30
9.44%
<3
5
50
30
0. 096%
140
10
200
20
<5
<33
<10
200
> 1,000
40
<33
3.36%
0.094%
Leachate.
mqlf,
<1
<0.01
< 0.003
0.01
<1
<0.01
<0.01
'/X.V&//,
<0.01
<0.01
<0.01
<0.05
<0.05
< 0.002
<10
<0.01
0.15
3
<0.05
<0.01
<0.01
< 0.003
1
<0.01
>10
<0.05
<0.01
<1
<1
m:y///
<10
<1
12
<10
<10
'/W///
'*&///
DWS*
mg/£
0.05
0.05
1.0
200
0.01
0.05
1.0
0.3
0.002
150
0.05
2.0
0.05
0.01
1.0
5.0
250
2.4
250
10
6. 5to9.2
-750
DWS - NIPDWR, USPHS, and WHO drinking water standards.
exceeds DWS.
37
-------
(Ca(OH) ) , a-Fe 0,., and CaSO, in decreasing concentrations. The
trace quantity of CaSO is not typical of spent bed material; one
would expect bed material to contain about 25 molar percent of CaSO.
(Table 5). The PER carry-over material consists of major quantities of
CaO and SiO~, minor quantities of CaSO, and a-Fe^O.,, and trace amounts
CaCO- and Fe_0,. Chemical analysis data (shown on Table 12 along with
leachate characteristics that will be discussed later) confirmed these
findings.
Figure 2 shows SEM photomicrographs and EDAX spectra of the PER bed
and carry-over. One can see the low sulfur concentration in the EDAX
spectrum of the PER shakedown bed material also. SEM and EDAX of the
carry-over show the material to have chemical characteristics one would
expect to find in bed material rather than in carry-over. We believed
that the material was collected from the primary cyclone without recycle
and is, therefore, very much like the bed material, although the actual
identity of the run was not known.
Tables 12 and 13 summarize the concentrations of major and trace
species in solids and leachates from the shake tests. Calcium ion fluc-
tuation in the leachate was caused by the ready formation of a white
precipitate during filtration and handling of the leachate in air. This
white precipitate was identified by X-ray diffraction and TG to be
CaCO_. Figure 3 shows a SEM and EDAX spectrum of the white crystalline
agreeing with the X-ray and TG results. Since the PER shakedown test
bed material consists mostly of CaO, its leachate forms carbonation
products most readily; other residues, with more typical (lower) CaO
contents (and higher CaSO^ contents) did not show this tendency to form
precipitates. Due to the very low CaSO, concentration in the PER shake-
down bed material, its leachate displayed much lower sulfate characteris-
tics than would the residue of a typical FBC system. Table 13 shows that
TOC and trace metal elements are sufficiently low in the PER leachate to
meet the existing DWS. Heat-release property will be reported in a
later section.
38
-------
Table 12
Dwn. 1709B91
SOLID AND LEACHATE CHARACTERISTICS OF PER RESIDUES
(200-Hr Continuous Shake Test)
Substance
Al
Aq
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hg
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
S04
TOC
PH
SC. M mho/cm
PER Solids. wt%3 PER Leachates. mg//
(1)
>10
0.003
0.01
0.0002
< 0.0003
27.52
< 0.003
< 0. 0005
0.003
0.005
> 10
3.93
0.02
0.002
0.1
0.005
0.001
>10
< 0.003
0.1
a ooi
0.007
a 01
26.4
(2)
>10
0.003
0.01
0.0001
< 0.0003
16.64
<0.003
< 0.0005
0.003
0.005
> 10
2.88
0.01
0.001
0.1
0.005
0.001
>10
< 0.003
0.1
a ooi
0.01
0.01
45.76
(1)
0.05
<0.01
< 0.002
<0.05
<1
<0.01
<0.01
'//KM
<0.01
<0.05
<0.05
<0.05
<0. 1
< 0.001
14.4
<0.05
0.05
>5
<0.05
5
<0.05
<0.05
<1
<0. 1
<10
<]Q
V,W,
'4,W
(2)
1.0
<0.01
< 0.002
0.8
4:1.0
<0.01
<0.01
Y/X.UA/S
<0.01
<0.05
<0.05
<0.05
<0. 1
< 0.001
14.4
<0.05
0.2
> 5
<0.05
<0.05
<0.05
< 0.003
>5
<0.05
>5
<0.05
<0.05
<1
<0.1
YAW/;
-------
Table 13
LEACHATE CHARACTERISTICS OF PER RESIDUES
Residue
Bed
Carry-over
Chemical
Characteristics
pH
Sp. Conductance
millimho/cm
Ca, mglt
504, mq/t
PH
So. Conductance
millimho/cm
Ca, mq/t
S04, mq/t
Intermittent Shake,
no. of 72- hr intervals
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
12.0 12.3 12.1 12.1 11.9 12.2 12.2 12.1 12.1 11.9 11.9 12.1 11.9 11.5 116
7.55 7.37 7.22 7.03 6.98 6.98 6.69 6.W 6.77 7.08 7.15 6.9 6.8 7.32 6.79
400 400 608 - 472 - 528 496 600 588 500 504 496
298 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10 < 10
12.2 12.4 12.2 12.2 12.1 12.3 12.2 12.2 12.2 12.1 12.1 12.2 12.0 12.0 12.0
7.78 7.55 7.39 7.49 7.48 7.35 7.35 7.3 7.02 6.72 6.46 5.45 5.68 3.7 3.15
976 960 952 1136 904 944 936 900 1008 856 876 704 736 448 424
1298 1338 1260 1138 1082 1260 1353 1128 1128 1150 850 704 778 778 590
Continuous
Shake, hr
100
12.2
"
7.%
244
100
12.1
8.52
1056
1623
200
12.0
7.85
720
< 10
12.1
8.06
1264
1410
-------
r
W
(a)
(b)
Figure 2 -
(c)
SEM and EDAX of (a) RS-PER Shakedown Bed Material,
Particle Surface, (b) RS-PER Shakedowi
Fractured Surface, (c) RS-PER Carry-over
41
-------
(a)
19SEC 16482 INT
v s:s eee H s : SBEV/CH
Q Hlffi|
EDAX
(b)
Figure 3 - SEM Photomicrograph (a) and EDAX Spectrum (b)
of the White Precipitate Formed Readily in
Air from the Leachate from PER Bed Material
and Carry-over
42
RM-71557
-------
Since the PER residues used in this study were of uncertain or
nontypical origin, we recommend further investigation when more represen-
tative samples are available.
25
Battelle Residue
In August and November of 1977, we obtained from the Battelle
Columbus Laboratories two batches of bed and carry-over (cyclone) mate-
rials withdrawn during a corrosion/erosion run (600 to 1000 hr into the
test) and at the end of the 1500-hr continuous run. The Battelle
corrosion/erosion experiment was conducted in a 24-in. diameter AFBC
unit at 877°C, using Illinois No. 6 coal and Grove limestone with a
calcium-to-sulfur ratio of Ca/S = 1.8. The objective of the DDE-funded
run was to conduct corrosion/erosion experiments on candidate heat-
exchanger and superheater tube materials exposed to high-temperature
(e.g., 593 to 870°C) gaseous corrosion and particulate erosion in the
bed of an atmospheric fluidized-bed coal combustor containing limestone
as a sulfur oxide (SO ) sorbent, to establish engineering durability data
X
as a function of materials properties.
Chemical analyses summarized in Table 14 showed similar compositions
for the two batches of samples withdrawn at different times into the
same run (one at 600 to 1000 hr into the test and one at the end of the
1500-hr run). Because of their similarity, only the solids withdrawn at
the end of the 1500-hr run were used in the leaching studies.
Table 15 summarizes the chemical characteristics of the 200-hr
leachates from the Battelle bed and carry-over materials. As we expected,
both leachates had much higher calcium, SO,, pH, and specific conductance
than do the DWS, because of the dissolution of the major species, CaO
and CaSO,, but did not exceed any of the trace elements where DWS existed.
Figure 4 presents the results from the intermittent leaching tests. The
difference between the bed and carry-over leachates is clearly demonstrated
by the much more rapid improvement in the leachate quality of the latter.
Results from the Battelle residues testing, in general, agree well with
those previously reported.
43
-------
Table 14
Dwci. 7709A9?
CHEMICAL ANALYSIS OF BATTELLE FBC RESIDUES
FROM THE 1500-HR CORROSION/EROSION TEST
Sample
Bed. 1500 hr
Bed, 1000 hr
Carry-over. 1500 hr
Carry-over, 600 hr
Wet Chemical Method
Ca
33.8
30.8
27.0
25.6
Mg
1.0
0.14
0.91
0.14
S~
0.05
0.02
0.4
0.02
S04
39.2
35.6
25.6
27.0
«
CaS04
55.5
50.4
36.3
38.2
CaO*
24.4
22.4
22.8
20.1
X-Ray Diffraction
CaS04
Major
Major
Major
Major
CaO
Minor
Minor
Minor
Minor
Ca(OH)2
Lo-minor
Lo-minor
Lo-minor
Lo-minor
Si02
Lo-minor
Lo-minor
Major
Major
aFe^j
-
—
Lo-minor
Lo-minor
-o
JS
'Estimated by chemical analysis data
-------
Table 15
CHEMICAL CHARACTERISTICS OF LEACHATES
FROM BATTELLE FBC RESIDUES
(200-hr Continuous Shake Test)
Dwg. I694B71
Substance
Al
Ag
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hg
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
F
Cl
N03(asN)
so4
TOC
PH
Sp. Cond.
(p-mho/cm)
Leachate Characteristics, ppm
Bed
<1
<0.01
<0.05
>4
<1
<0.01
<0.01
'///,W///S
<0.01
<0.05
0.05
<0.1
<0.02
< 0.002
<1
<0.01
0.26
80
<0.05
<0.01
<0.05
<0.01
<1
<0.05
>4
<1
<0.03
<1
<1
<1
12
<10
s/////' "///
^1460x%
///////////
<10
'//WL:W////,
y///m////,
Carry-over
<1
<0.01
<0.05
2
<1
<0.01
<0.01
'/////A®'/////.
<0.01
<0.05
0.04
<0.1
<0.02
<0.002
<1
<0.01
0.08
7.6
<0.05
0.05
<0.05
<0.01
<1
<0.05
5
<1
<0.03
<1
<1
<1
12
<10
'////// '"'////
yy/y%l™%%,
//////// /////
<10
'/////AtV/////,
y//////$w////,
Drinking Water
Standards, ppm
0.05
0.05
1.0
200
0.01
0.05
1.0
0.3
0.002
150
0.05
2.0
0.05
0.01
1.0
5.0
2.4
250
10
250
6.5to9.2
-750
Drinking Water Standards: NIPDWR: USPHSandWHO
Drinking Water Standards
Exceed the drinking water standards
45
-------
Curve 69^98-B
Normalized Leachate Quantity, ml/g of Starting Solid
6 12 18 24 30
2000
1000
O
I I
^ n n
01
2000
1000 -
E
o
o
O
2 4 6 8 10
n =Total Number of 72-hr Intermittent Leach ing
I i i |
0 360 720
Total Leach Time =(72) -n.hr
Figure 4 - Leachate Characteristics of Battelle FBC Bed (o)
and Carry-over Material (D) as a Function of
72-hr Carry-over Intermittent Leaching
46
-------
MERC Residue23
The MERC residue was produced from a spent sorbent production run
requested by Westinghouse as part of EPA's environmental testing program
on the MERC 18-in.-id AFBC unit. This combustor includes two cyclones
in series on the flue gas (cyclone 1 followed by cyclone 2), followed
by two bag filters in parallel. Batches of bed material (three samples),
cyclone ash (five samples), and bag-filter fines (two samples) were
collected at various times throughout the 100-hr continuous run.
Table 16 summarizes the compounds identified by X-ray diffraction.
Because of the nature of the X-ray diffraction method, we shall not
*
attempt to go beyond semiquantitative description of the results, major
minor, and trace. All of these materials (bed, cyclone, and bag filter)
were high in CaSO,. Of the unsulfated portion, the bed material contained
more CaO (or Ca(OH)? formed by CaO hydration in air), while the fines
contained more CaCO~. This explains the lower heat release by ash and
fines reported in detail in a later section. When the bed material was
separated physically, we found that the white (interior) particles were
mostly CaO and Ca(OH)2, the grey particles were unsulfated CaCO.,, and
the darker particles were rich in SiC- and Fe^O,.. The raw sorbent
(Greer limestone) and coal (Arkwright) were also analyzed. The Greer
limestone was found to consist of major amounts of CaCO_, minor Si02, and
dolomite (CaMg(CO-)2) with trace CaSO,. The crystalline species in the
coal were found to be SiO? and iron sulfide (FeS).
Table 17 summarizes the leaching results by the continuous shake
method of three batches of bed material collected at three different
times, two batches of cyclone ash from the two different cyclones and
two batches of bag-filter fines from two different filters. Agreeing
with the previous leaching results from other FBC spent materials,
high pH, TDS, calcium, and SO, resulted from the bed material leachate.
Material from cyclone 1 produced leachates similar to the bed material;
this result is not surprising, since this cyclone is the first of the
series of control devices on the unit and, thus, collects the coarsest
47
-------
Table 16
IDENTIFICATION OF SPENT MERC BED MATERIAL AND CARRY-OVER
Dwg. 1690B61
Sample
MERC Bed. 3/8
ii
ii
ii
ii
ii
MERC Bed, 3/9
M
MERC Cyclone 2
Carry-over
MERC Bag Filter
1&2 Fines
Greer Limestone
Arkwright Coal
Separation
General grind
Brown throughout
Thin brown 00
white ID
Brown OD
black center
Grey throughout
Black
General grind
Black
General grind
General grind
General grind
General grind
Chemical Composition
CaS04
Major
Major
Trace
Minor
Trace
Lo Minor
Major
Trace
Major
Major
Trace
CaC03
(Calcite)
Trace
Trace
Major
Trace
Trace
Lo Minor
Minor
Major
CaO
Trace
Trace
Trace
Lo Minor
Trace
Lo Minor
Trace
Ca(OH>2
Minor
Trace
Major
Trace
Si02
Trace
Hi Minor
Trace
Trace
Hi Minor
Major
Hi Minor
Major
Major
Major
Minor
Minor
Fe203
Major
Minor
Minor
Others
Possible trace of vaterite CaCO,
Trace vaterrte CaC03
MinorCaMg(C03)0
Major FeS
-C-
oo
-------
Table 17
SUMMARY OF LEACHING RESULTS OF MERC SAMPLES
BY CONTINUOUS SHAKE METHOD
Sample
Solid/
Water
Ratio
Leaching
Time,
hr
Leachate Characteristics
PH
Specific
Conductance,
millimho/
cm
mg/1
so4,
mg/1
MERC Bed, 3/9
MERC Bed, 3/9
MERC Bed, 3/8
MERC Bed, 3/8
MERC Bed, 3/8-9
MERC Bed, 3/8-9
MERC Cyclone 2
MERC Cyclone 2
MERC Cyclone 1
MERC Cyclone 1
MERC Bag Filter 1&2
MERC Bag Filter 1&2
MERC Bag Filter 2
MERC Bag Filter 2
Greer Limestone
Greer Limestone
Arkwright Coal
Arkwright Coal
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
1:10
220
400
220
400
200
400
220
400
200
400
220
400
200
400
220
400
220
400
12.0
11.9
11.9
11.9
12.0
11.9
11.8
11.6
12.0
11.9
10.8
10.6
11.6
11.1
8.0
8.0
2.8
2.8
8.93
7.61
8.38
7.80
10.1
8.48
4.78
2.98
9.14
8.94
3.44
2.39
3.2
2.36
0.2
0.2
2.02
1.83
1336
1188
1232
1208
1488
1384
8
0
1208
1244
220
48
48
40
16
16
96
104
1273
1690
1395
1370
1395
1285
282
374
1225
1020
830
488
276
514
30
37
960
1050
49
-------
carry-over, which is likely to be most like the bed material. Leach-
ates from cyclone 2 and the bag-filter fines had lower pH, IDS,
calcium, and SO,. In every case leachate quality improved between
the 200-hr and 400-hr continuous shake time, most likely because of car-
bonation of dissolved Ca(OH)~ by dissolved carbon dioxide (CCO. Leach-
ing results from the raw sorbent and coal are also reported in Table 17.
Greer limestone, which was practically insoluble, produced leachate of
very good quality, while Arkwright coal produced highly acidic (pH = 2.8)
leachate with a high SO, concentration.
Figure 5 summarizes the results from intermittent leaching for the
MERC materials. Leachate characteristics from two batches of bed mate-
rial collected at different times in the test were in good agreement in
the intermittent leach tests. There was little change over time or with
increasing total leachate volume passing through the bed material.
Leachates from the cyclone and bag-filter fines showed improvement with
total leachate volume and time.
In addition to those using deionized water, continuous shake leach-
ing tests were also conducted using C0~-saturated deionized water (pH = 4)
to simulate the leaching that might take place in the environment due
to exposure of the residue to'rainwater or surface water. Table 18 sum-
marizes the chemical characteristics, including major and trace metal
and anions, and the TOC for leachates induced with deionized water and
C0?-saturated water media. Several points are worth noting, based on
the test results summarized in Tables 17 and 18:
• The leachate of MERC bed material appeared to be similar
to leachates from all the other FBC residues examined
previously.
• Unlike all the previous FBC leachates tested, the leachate
from MERC carry-over material exceeded the drinking water
standards for two elements — barium (in one of the four
carry-over leachates indicated in Table 18) and chromium
(in all four carry-over leachates). Note, also, that the
DWS standards are used here for comparative reference only.
50
-------
Curve 691328-B
O
MERC Bed, 3/9
MERC Bed, 3/8
MERC Cyclone 2 Ash
•--MERC Bag Filter l&2Fines
Normalized Leachate Quantity, ml/g of Starting Solid
6 12 18 24 30 36 42
AAJU
E. 1000
ro"
O
'
2000
^
41000
CO
1
l£
a. 10
8
1
10
•° B 8
o °
• "E ^
£"I 4
2
n
i i i ' i i
fin • M ° " J "
_uBODBn"HH"UL)1Jn*_
DWS
Jk--o-- a- „ o- o- -Q - Q g fl 9 J — ? — 8 — g
i i i i i i
1 ° B
•n""lnRnl linn
q ~ 3 n 0 D H y B B • i
i ~~^ — ••-— •?-— ft1-^ — ^--S^. "f" ^p .^8_-I^ ^.
6 • o • n n
~""**<-~9— JUw^-- 1^ J. " " L
DWS Range
i i ' i i i '
-9
- \ • B " B u • — B — a — B — P — g -
- . Vx
~DWS. ^x
~approx'd ""^^JL
2 4 6 8 10 12 14
n = Total Ho. of 72 - h r Intermittent Leach ing
i | i
360 720 1080
Total Leach Time = (72) (n), hr
Figure 5 - Leachate Characteristics of Spent MERC Bed Material, Ash,
and Fines as a Function of Intermittent Leaching
51
-------
Table 18
SOLID AND LEACHATE CHARACTERISTICS OF MERC-AFBC RESIDUES:
EFFECT OF USING C02-SATURATED WATER AS LEACHING MEDIUM
(200-hr Continuous Shake Tests)
Dwg. 26I6C60
Substance
Al
As
g
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hq
Mq
Mn
Mo
Ma
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
Cl
F
Br
N02
NO (asN)
P04
S04
TOC
pH
SC
(yhos/cm)
Solid. ppma
(1)
Major
<1
1Q
<10
39*
<3
5
60
30
Major
>1000
190
30
>1000
40
<10
<33
Major
<10
500
>1000
40
«33
20%
(21
Major
<1
500
10
<10
18%
<3
15
100
100
Major
>1000
220
10
>1000
70
100
<33
Major
<10
1000
>1000
100
100
13%
(3)
Major
<1
500
10
<10
16%
<3
15
100
100
Major
>1000
280 _
10
>1000
70
130
<33
Major
r1000
80
100
12%
Leach ate, ppm
(1AI
<1
<0.01
<05
<1
<1
<0.01
<0.01
'Z/my'A
<0.01
<0.02
0.04
<0.1
<0.1
<0.002
<1
<01
0.04
4
<0.04
<0.01
_
(2AI
2
<0.01
«X05
<0.2
'///I////
<0.01
C0.01
16
x
<0.04
z
t
2
<0.01
<0.05
3
<1
<01
<01
'//ss>//,
<01
<02
Y//W//,
2
»4
2
<0.04
»d
2
209
2
<0,04
»d
-------
• The TOC was also higher in the leachate from MERC bag-
filter material (<30 ppm) than from that found in all the
other FBC leachates (<10 ppm).
• The pH and major species (calcium, SO,, and TDS) decreased
with decreasing particle size (in the order of bed,
cyclone, and bag-filter materials), in both the solids
and their leachates. This observation confirms the expec-
tation that the spent sorbent (the cause of the high pH,
calcium, SO,, and TDS levels) is present primarily in the
coarser fractions.
• Trace elements (e.g., B, Co, Cr, Cu, Mn, Mo, Ni, Pb, Sr,
V, Zn, Cl, Br) increased with decreasing particle size in
the solids and, in many cases, in their leachates as well.
That is, trace elements were more concentrated in the
carry-over materials and their leachates (bag-filter
catch > cyclone 2 > bed material).
• The effect of leaching with C02-saturated deionized water
was not pronounced, probably because of the high alkaline
content in the FBC residues; the mild acidic effect of the
C02 is quickly neutralized.
Leaching property was also investigated using a sodium acetate-
acetic acid buffer solution with pH = 4.5 ± 0.1. Results will be dis-
cussed in a later section, together with acid leaching of many other
FBC residues.
Earlier results using PFBC residue showed improvement of leachate
5 34
quality by codisposal of the spent bed material with the carry-over. '
Codisposal of MERC bed, cyclone, and bag-filter residue was also
investigated here. A mixture of 60:20:20 of bed:cyclone:bag-filter
materials was used in the two repeated leaching tests. This mix ratio
was selected as being reasonably representative of the ratio in which
the materials would be disposed of in a commercial AFBC facility. The
results of the two tests were different. Figure 6 shows these results
and compares them with leachate conductance from unmixed bed, cyclone,
53
-------
Curve 696880-'-
Normalized Leachate Quantity, nu/gof Starting Solid
6 12 18 24 30 36
10
6 -
•S 4
Bed Material
Cyclone Carry-over
Bag Filter Carry-over
— •— Bed/Carry-over Mix I"
Bed/Carry-over Mix II
\
\
y».
NWS. V0X.
approximated ^J
/ Bed/Carry-over Mix Solidified in Flask;
incompact Broken to Pieces;
/
Shake Cycle Continued
246 8 10 12
n = Total No. of 72-hr Intermittent Leaching
180 360 540 720
Total Leach Time, = (72) • (n) hr
Figure 6 - Leachate Characteristics from Codisposal
of MERC Bed and Carry-over Materials
and bag-filter solids. In the first test (MX I), leachate quality improved
rapidly as the solid mixture solidified at the bottom of the flask after
four 72-hr cycles. The cementlike compact was then broken, and leaching
cycles continued. This accounts for the discontinuity of the MX I
result on Figure 6. The experiment was repeated in the MX II test,
which did not solidify, but its leachate showed slow, steady improvement
over total leaching time and leachate volume. These experiments
demonstrated:
• The possibility of the residue hardening (cementlike set-
ting) when spent bed and carry-over materials are disposed
of together in a landfill
• The effect of the fissuring of the hardened residue on the
resultant leachate
54
-------
• The potential of reduced environmental impact from codis-
posal of spent bed material with the carry-over due to
the pozzolanic reactions.
Acid mine neutralization experiments were also conducted, using
mixtures of Arkwright coal and MERC residues. Results showed that spent
FBC sorbent can be an efficient neutralization agent for the acid mine
drainage because of the high free-lime content. Westinghouse is con-
tinuing work in this area under contract to EPA.
PFBC Residue Characterization and Leachate
Exxon Miniplant
Most of the FBC residues investigated and reported previously were
from PFBC once-through systems. The PFBC residues tested during this
reporting period include spent bed material, cyclone carry-over, and
21
stack fines from the Exxon miniplant at Linden, New Jersey. The mini-
plant consists of a 32-cm-id pressurized combustor; it also includes a
21.6-cm-id regenerator that can be used to regenerate the spent sorbent
for reuse, if desired. We also obtained some regenerator residue sam-
ples for testing. Table 19 summarizes the chemical identification by
X-ray diffraction of the Exxon residues investigated during this report-
ing period.
Exxon run 43 is a once-through PFBC run using dolomite operated at
different bed temperatures - for example, Exxon 43.2 at 845°C and Exxon
43.3 at 940°C. At the lower temperature, the calcium fraction of the
dolomite would not be expected to calcine, so only the magnesium frac-
tion would calcine; at the higher temperature, both fractions should
calcine. The mixed spent sorbent, therefore, is expected to contain
both the half-calcined material from the lower-temperature portion of
the test and the fully calcined dolomite from the higher temperature
test. Indeed, X-ray diffraction identified the presence of both CaC03
and CaO.
55
-------
Table 19
X-RAY DIFFRACTION IDENTIFICATION OF PFBC RESIDUES
DI..I. I7.JBO?
Sample
Exxon 43.2/43.3
Bed Mat 'I
Bed Mat'l
Bed Mat'l
Bed Mat'l
Bed Mat'l
Exxon 43. 3 Carry-over
Exxon 45. Regenerative
Comb. Bed Mat'l. RegVe
Comb. Bed Mat'l
Comb. Bed Mat'l
Exxon 45
Regenerative Bed Mat'l
Reg. Bed Mat'l
Reg. Bed Mat'l
Reg. Bed Mat'l
Reg. Bed Mat'l
Exxon 45 Carry-over
Exxon 30. 4 Bed
Hydrated(via RS)
Exxon 30. 4 Carry-over
(viaRS)
Exxon 43. 5 Carry-over
(viaRS)
Exxon 67 Fines
(3rd eye lone)
Separation
General grind
Black particles
Tan OD. white ID
Dark reddish OD
Grey ID
Brown OD. white ID
General grind
General grind
Brown OD. white ID
Brown OD, grey ID
Agglomerates
- general qrind
Agglomerates
- brown glassy phase
Granular-
- general grind
Granular-
- brown OD. white ID
Granular-
- slag phase
General grind
General grind
General grind
General grind
General grind
Chem Composition
CaS04
Major
Major
Minor
Major
Hi-Minor
Major
Major
Major
Major
Hi-Minor
Minor
Minor
Minor
Hi-Minor
CaS03
Calcite
Minor
Hi-minor
Trace
Hi-Minor
Trace
Minor
Trace
Trace
Major
CaO
Minor
Hi-Minor
Hi-Minor
Hi-Minor
Major
Major
Hi-Minor
Major
Possible
Ca(OH)2
Trace
Trace
Trace
Trace
Trace
Trace
Trace
MgO
Minor
Hi-minor
Trace
Si02
Major
Major
Trace
Major
Trace
Minor
Major
Major
Major
Major
a Fe203
Major
Minor
Trace
Minor
Minor
Minor
Minor
Fe3°4
Spinel
Trace
-
Others
Trace
Trace Ca.SiO-
Trace Ca3Si05
Major Ca.SiO-
Trace ALS'^Oj,
-------
Exxon 45 is a run with sorbent regeneration, using limestone as
the desulfurizing sorbent. The regenerator material is partially
agglomerated; the agglomerates might have been formed during shutdown,
when a temperature excursion occurred due to CaS oxidation. Because of
the small amount of CaS present, X-ray diffraction patterns did not
show CaS at all. Indeed, wet chemical analysis showed very low sulfide
(<0.1%) in the regenerator bed material. Exxon 45 is not considered
representative of the regenerative PFBC process. X-ray diffraction
identified the agglomerated portion of the bed as primarily CaO (with
little CaSO, or CaS), and with SiO? and Fe~0 as minor species.
The Exxon 67 run was a 100-hr shakedown run for the DOE-funded mate-
rial testing program on the EPA miniplant, in which boiler tube mate-
rials supplied by Westinghouse and gas turbine specimens supplied by the
General Electric Company (GE) were exposed to FBC environments. In this
run three stages of cyclones in series were used on the miniplant corn-
bus tor flue gas in order to remove enough particulates to give a meaning-
ful test of the gas turbine specimens. The solid residue Westinghouse
obtained for the testing reported here is fine material collected by
the third cyclone.* For comparison, the other Exxon carry-over material
discussed in this report (from runs other than 67) is material collected
by the second cyclone.*
In order to correlate the various EPA contractors' efforts in the
area of FBC solid waste disposal, we requested and received samples from
27
the Ralph Stone Company, which had carried out a FBC solid waste pro-
gram under contract to EPA. The samples obtained included the two PER
residues discussed earlier and three Exxon residues (Exxon 43.5 bed,
Exxon 30.4 bed, and Exxon 30.4 carry-over). Sample Exxon 30.4 bed had
been hydrated and air dried prior to being sent to us.
In general, PFBC once-through bed material consists of major amounts
of CaSO, , minor amounts of CaCO., (sometimes CaO, depending on whether
the temperature is high enough to calcine the calcium fraction), and
*These refer to third-stage and second-stage cyclones.
57
-------
minor-to-trace amounts of SiO™, aFe CL, and Fe.O^ spinel. Once-through
PFBC bed usually contains less CaO (and more CaCO_) than does AFBC bed
material, which suggests a possibly significant difference in residue
leaching/activity properties. Regenerative PFBC bed contains major
CaSO, and CaO, and minor-to-trace SiO,, and aFe«0«. PFBC carry-over has
major SiO? and minor CaSO, and aFe20-. The lighter particles consist
mostly of sulfated and unsulfated limestone or dolomite (CaSO,, CaCO~,
CaO, Ca(OH)2, MgO). The darker particles are high in SiO-, aFe^™,
Fe_0,, and other trace impurities.
Leaching results using continuous and intermittent shake tests are
summarized in Table 20. Leachate quality is similar for all four bed
materials. Calcium, SO,, TDS, and pH are major concerns. Dissolved
calcium exhibited fluctuation because of carbonation by C0~ in air;
white CaCO., precipitate was crystallized on the leachate surface on
exposure to air. This was especially true for the leachate of Exxon 45
regenerator bed agglomerate; like the one other residue that showed this
precipitation tendency (the PER AFBC material discussed previously), the
PFBC regenerator agglomerate contained CaO as a major species. The pH
improved with time for carry-over leachate but remained unchanged for
the bed material. Sulfate was relatively unchanged with time with the
exception of leachate from Exxon 45 regenerator agglomerates, which were
low in solid SO,.
4
Finally, the leachate from the carry-over is not only superior ini-
tially; it also improves faster with time or total leachate volume passing
from the sample. The results are not surprising in view of the lower
CaO content of the carry-over; also, the probable pozzolanic reactions
between the high SiO^ content in coal ash and the CaO present in sorbent
fines could serve to reduce leaching tendencies. The better leachate
quality from the carry-over material is also shown in Figure 7, which
plots specific conductance versus total leaching time or leachate
volume.
58
-------
Table 20
LEACHATE CHARACTERISTICS OF PFBC RESIDUES
Residue
Ex. 412/413
Bed
Ex 43.3
Carry-over
Ex. 43.5
Carry-over
Ex 30. 4 Bed.
Hyd rated
Ex 30.4
Carry-over
Ex 45
Combustor Bed
1 regenerative)
Ex 45
Region Bed
1 granular)
Ex 45
Regtor Bed
(agglomerated)
Ex 45
Carry-over
Ex 67
Carry-over
(fines<15|im>
Chemical
Characteristics
1 L 2 1 3
P" milllmho/uo u2 li2
Specif ic Conductance '" &M i01 7-»
Ca. mg// 1136 1304 1328
SO-^mg// 1352 1352 1366
PH
S£, millimho/cm
Ca. mg//
S04. mg//
PH
S.C.millimho/cm
Ca. mg//
SO., mg//
12 0 11. 3 10. 8
6. 73 2 98 2 41
888 672 664
1247 1352 1445
11.1 10.7 10.4
288 2.43 222
648 648 620
1529 1353 1311
pH 121 123 121
S.C.millimho/cm a 07 7.38 7.12
Ca. mg// 1048 1056 964
SO., mg// 894 1225 1163
pH 10.1 10.1 10.1
S.C millimho/cm 24 2. 24 2 16
Ca. mg// 736 6% 565
SO., mg// 1510 1380 1260
pH 122 123 12.2
S.C.millimho/cm 8.26 7.56 7.79
Ca. mg// 1024 1104 1120
S0<. mg/i 1325 1366 1445
S. mg// 83 36 36
pH 122 123 122
S.C.millimho/cm 8.59 7.85 7.91
Ca. mg// 1024 1128 1056
S04 mg// 1366 1298 1188
S. mg// 90 48 23
pH 122 123 122
S.C.millimho/cm 8.05 7.33 7.57
Ca. mg// 464 448
S04. mg// <10 <10 <10
S. mg// 23 19 14
pH 11.8 10.9 10.6
S.C.millimho/cm 3.93 262 238
Ca. mg// 628 672 656
SOf. mg// 1380 1547 1325
S. mg// 49 34 44
pH 8.5 8.3 ai
S.C.millimho/cm 19.7 7.72 3.58
Ca. mg// 424 428 516
Mg, mg// 5582 1538 358
SO. 20400 6025 2210
1 4
122
7.97
1288
1225
10.7
249
656
1410
10.2
218
588
1187
121
7.19
1112
950
9.9
222
632
1236
120
6.74
988
1285
14
120
7.79
984
1200
15
120
7.34
648
<10
11
10. 2
228
624
1298
18
ao
252
608
125
1583
1 5
122
7.43
1216
1187
10.6
243
MO
1510
10.0
210
600
1105
120
7.15
848
1040
9.9
2.21
668
1273
12.0
7.66
944
1116
15
120
7.69
908
1071
13
120
7.19
<10
13
10.1
223
648
1213
23
ao
223
616
53
1645
Intermittent Shake, no.
J 6
12.2
7.89
1260
1213
10.6
2.51
656
1366
10.1
1.9
552
1139
122
6.89
872
1150
9.7
2.16
636
1366
12. 2 "
7.46
1000
1030
12
12.2
7.60
968
1150
16
12.2
7.0
<10
14
10.1
2.06
592
1186
23
7.9
214
616
24
1353
7
122
7.48
11V2
1380
10.3
241
656
1278
9.7
1.83
548
1020
121
6.46
784
1030
9.2
2.10
664
1260
122
7.51
1000
1200
15
122
7.77
1016
1182
15
12 2
6.95
00
15
9.8
20
604
1225
20
7.9
208
584
24
1285
1 »
121
a ic
1336
1445
10.4
246
672
1445
9.2
1.41
500
923
120
6.2
776
1127
9.4~
1.99
608
1213
121
7.54
1000
1325
10
12.1
7.77
1064
1285
<10
122
7.04
<10
<10
9.3
1.89
576
1116
12
7.7
211
608
14
1285
of 72-hr Intervals
Li
122
7.47
1224
1150
10.3
237
656
1458
9.5
1.70
440
885
120
5.44
832
1150
8.9
205
560
1200
121
7.35
1140
1339
-------
16
14
Curve 697272-B
Normalized Leachate Quantity, ml/gof Starting Solid
6 12 18 24 30 36 42
10
1
o Exxon 43.2/43.3 Bed
• Exxon 43.3 Fly Ash
a Exxon 45 Comb. Bed
0 Exxon 45 Reg. Bed. Granular
Q Exxon 45 Reg. Bed, Agglomerated
* Exxon 45 Fly Ash
A Exxon 30.4 Bed
A Exxon 30.4 Fly Ash
• Exxon 43.5 Fly Ash
•DWS, approx'd
2 4 6 8 10 12 14
n = Total Number of 72-hr Intermittent Leach
360 720
Total Leach Time = (72) (n) hr
1080
Figure 1 - Leachate Quality of PFBC Residues as a
Function of Intermittent Leaching
Trace elements in PFBC leachates which had been investigated more
extensively than the AFBC leachate in the previous report are presented
here again for several of the Exxon PFBC residues, with and without sor-
bent regeneration, in Tables 21 and 22. Other solid and leachate char-
acteristics are also summarized, including major and minor species, pH,
specific conductance, and TOC in the leachate. As expected from the
previous findings, all trace metal elements were lower than the DWS
except for one, barium in the leachate of the agglomerated portion of
the Exxon 45 regenerator bed material. This is not of great concern
because agglomeration would not be expected in a typical run. The TOC
was low for all; calcium, SO,, pH, specific conductance were high, as
previously reported.
60
-------
Table 21
Dwa. 1709B89
SOLID AND LEACHATE CHARACTERISTICS OF PFBC RESIDUE FROM EXXON
(200-hr Continuous Shake Test)
Substance
Al
Aq
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hq
Ma
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
S0d
TOC
PH
SC.umho/cm
a
PFBC Solid, Wt*
(1)
>10
0.005
0.01
0.0001
< 0.0003
35.04
< 0.003
0.001
0.003
0.005
>10
1.15
0.01
0.001
0.1
0.01
0.005
>10
< 0.0003
0.1
0.001
0.01
0.01
38.72
(2)
> 10
0.003
0.02
0.0003
< 0.0003
5.76
< 0. 003
0.001
0.003
0.01
>10
0.67
0.01
0.002
0.1
0.01
0.005
>10
< 0.0003
0.2
0.003
0.01
0.01
30.56
(3)
>10
0.003
0.02
0.0003
< 0.0003
4.56
< 0.003
0.001
0.003
0.01
>10
3.02
0.01
0.002
0.1
0.01
0.005
>10
< 0.0003
0.2
0.003
0.01
a 01
40.64
a
PFBC Leachate.mg//
(1)
<0.05
<0.01
0.002
0.2
<1
<0.01
<0.01
^24^
<0.01
<0.05
<0.05
<0.05
<0. 1
< 0.001
14.4
<0.05
0.05
5
<0.05
<0.05
<0.05
< 0.003
0.5
<0.05
5
<0.05
<0.05
<1
<0. 1
w&//
<10
^/12. 03/
/I. 810 '
// /A
(2)
<0.05
<0.01
0.019
0.2
<1
<0.01
<0.01
W*//
<0.01
<0.05
<0.05
<0.05
<0. 1
< 0.001
<5
<0.05
0.1
<5
<0.05
<0.05
<0.05
< 0.003
<0.05
<0.05
>5
<0.05
0.1
<1
<0. 1
W*/s
<10
//* &/
'#•&/
(3)
<0.05
<0.01
0.013
0.5
<1
<0.01
<0.01
//JW/S
<0.01
<0.05
<0.05
<0.05
<0. 1
< 0.001
28.8
<0.05
0.2
>5
<0.05
<0.05
<0.05
< 0.003
<0.05
<0.05
>5
<0.05
0.05
<1
<0.1
YW,
<10
/A. <&'/
'//m
b
DWS,
mg//
0.05
0.05
1.0
200
0.01
0.05
1.0
0.3
0.002
150
0.05
2.0
0.05
0.01
1.0
5.0
250
6.5to9.2
-750
(1) Exxon 30.4 Bed. Hydilated and Dried
(2) Exxon 30.4Carry-over
(3) Exxon 43.5Carry-over
DWS: NIPDWR. USPHS. and WHO drinking water standards.
exceeds DWS
61
-------
Table 22
Dwg. 2624C72
SOLID AND LEACHATE CHARACTERISTICS OF PFBC RESIDUES
WITH SORBENT REGENERATION
(200-hr Continuous Shake Test)
Substance
Al
Ag
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hg
Mq
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
S04
s-
TOC
PH
SC.M mho/cm
a
Solid Analysis. Wt%
(1)
>10
< 0. 0001
<0.01
0.08
0.0003
< 0. 0033
3L2
< 0.0005
< 0. 003
0.005
0.006
>10
0.76
0.03
0.001
<0. 1
0.005
< 0. 001
<0.01
»10
0.001
0.1
0.005
0.003
27.45
0.07
(2)
»10
< 0. 0001
<0.01
0.05
... flJfflL ._
< 0. 0033
11.04
< 0.0005
< 0.003
0.005
0.006
>10
0.8
0.03
0.003
>1
0.005
0.005
< 0. 01 1
»10
0.001
0.5
0.01
0.03
14.5
0.1
(3)
»10 j
< 0. 0001
<0.01
0.08
0.0003
1 < 0.0033
32
< 0.0005
< 0. 003
0.005
0.003
h > 10
0.9
0.03
0.001
<0. 1
0.005
< 0.001
<0 Qi
»10
0.001
0.3
0.005
0.005
21.6
0.06
(4)
>10
< 0.0001
<0.01
0.01
< 0. 0001
< 0. 0033
51.36
< 0.0005
0.005
0.005
< 0.001
>10
1.3
0.03
< 0.001
<0. 1
0.005
< 0.001
<0 01
»10
0.001
0.1
0.003
0.001
1.8
0.06
a
Leachate Analysis, mqll
(1)
<0.05
<0.01
< 0.002
0.3
<1
<0.01
<0.01
YAfW./
<0.01
<0.05
<0.05
0.2
<0. 1
< 0.001
24
<0.05
0.1
<5
<0.05
<0.05
<0 05
< 0.003
0.5
<0.05
>5
<0.05
<0.05
<1
<0. 1
^35^
42
<10
YALrt,
'//W
(2)
0.5
<0.01
< 0.005
0.3
<1
<0.01
<0.01
^648//
<0.01
<0.05
<0.05
'< 0. 05
<0. 1 j
< 0.001
<5
<0.05 1
0.2
>5
<0. 05
<0.05 J
<0. 05
< 0.003
>5
<0.05
>5
<0.05
0.1
<1
<0. 1
'AW//
27.3
<10
^10.'4^
^2.620;/
'/ ////
( 3)
<0.05
<0.01
< 0.002
0.2
<1
<0.01
<0.01
y/x.?&\
<0.01
<0.05
<0.05
< 0. 05
<0. 1
< 0.001
<5
<0.05
0.05
<5
<0.05
<0.05
_5
<0.05
<0.05
<1
<0. 1
''V&'//
38
<10
; 12:02"^
KW/,
(4)
0.1
<0.01
0.002
2.0
'/£>&£
<0.01
<0.01
YAK//.
<0.01
<0.05
<0.05
<0.05
<0.1
< 0.001
2L6
<0.05
<0.05
<5
<0.05
<0.05
. 5
<0.05
<0.05
<1
<0.1
0
14
<10
/AI^V/
'4W7,
b
DWS.
mg/£
0.05
0.05
1.0
200
0.01
0.05
1.0
0.3
0.002
150
L T05 '
•
2.0
0.05
0.01
To
•
5.0
250
6.5to9.2
-750
(1) Exxon No. 45combustor bed
(2) Exxon No. 45carry-over
(3) Exxon No. 45 regenerator (granular)
(4) Exxon No. 45 regenerator (agglomerated)
DWS: NIPDWR, USPHS. and WHO drinking water standards.
£2 exceeds DWS
62
-------
The Exxon. 67 sample, which was collected from the third cyclone,
was much finer than the other batches of carry-over tested. These third-
cyclone fines represent the material that passed uncaptured through the
second cyclone; the Exxon carry-over for the other tests reported here
is material captured by the second cyclone. The particle size distribution
of the Run 67 fines is shown in Table 23. The SEM and EDAX spectrum
(Figure 8) shows that the physical appearance and elemental distribution
of Exxon 67 fines are similar to those of a previously tested PFBC stack
5 34
particulate sample obtained after the second cyclone of Exxon 34. '
Cenospheres are lacking in both ash samples. The similarity between
the third-cyclone catch and the post-second-cyclone flue gas particulate
sample is, of course, exactly what would be expected.
Both the continuous and the intermittent leach methods were employed
on the Exxon 67 sample; results are shown in Tables 20 and 24 and in Fig-
ure 8. The leaching property with deionized water was found to be differ-
ent from the other FBC solids tested previously in several areas:
• Magnesium leachability - The high magnesium concentration
in the leachate (Table 23), which decreased rapidly with
total leachate volume passing through the sample (Fig-
ure 8), contrasts with the previous results, which indi-
cated that magnesium did not leach out, even from spent
dolomite sorbent. After several intermittent shakes,
the magnesium concentration fell to the low levels
typical of previous results (Figure 8).
Table 23
PARTICLE SIZE DISTRIBUTION OF EXXON 67 THIRD CYCLONE FINES
Weight, %
Size less
than , ym
5
1.33
10
1.73
25
2.65
50
4.75
75
8.30
90
12.50
95
15.70
63
-------
Table 24
2618C86
CHEMICAL CHARACTERISTICS OF EXXON 67 THIRD
CYCLONE FINES AND THEIR LEACHATE
(200-hr Continuous Shake Test)
Substance
A£
Ag
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hq
K
Mq
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Ti
V
Zn
Zr
S03
S04
S^
F
Cl
Br
N02
NO^(asN)
P04
FreeC
TOC
PH
SC((j mho/cm)
Solid.ppm
25.4%
<1
300
500
15
<1
7.2%
3
10
200
30
5.53%
1. 16%
3.84%
330
30
0.59%
150
30
<33
15. 47.
<5
> 1000
200
500
500
'i'i 23%
<0.057o
<0.02%
0. 14%
0. 'M
1.2%
Leachate. mqll
Deionized Water.
pH-7.0
<1
<0.01
0.01
2
<1
<0.01
<0.01
^^^488 W///////.
<0.01
<0.04
0.03
<1
'/////////. bA'/////////.
0.0009
'//////////,\m '/////////,.
?/////////. §.-&'/////////,
2
<1
<0.02
<0.5
'///////M 0.03^%^
5
<0.2
<0.2
0.15
4
<1
<10
'/////////A ®K'//////f//A
<10
?////////,. br///////////,
10
<1
<1
<1
<1
<20
8.02
-mm^w//////A
Acetate Buffer,
PH - 4. 5
1
<0.01
0.002
1
<1
<0.01
<0.01
'////////,™ ''//////,
'//////// *.\'//////,
0.1
0.03
<1
0.3
0.0007
Y////////.W\bW/////,
mw// wm////.
0.2
<;i
<0.02
<0.5
< 0.001
> 10
<0.2
<0.2
<0.05
<4
<1
<10
w/////, ww///////
<10
<1
<1
<1
<10
<1
<1
f////////, umw,
WM^Y/M
DWS.
mg/l
0.05
0.05
1.0
200
0.01
0.05
1.0
0.3
0.002
150
0.05
2.0
0.05
0.01
1.0
5.0
250
2.4
250
10
6.5to9.2
-750
DWS - NIPDWR, USPHS and WHO Drinking Water Standards
Exceed DWS 64
-------
800
600
400
200
0
6000
5000
4000
3000
2000
1000
0
Curve 696108-
Normalized Leachate Quantity, ml/gof starting solid
6 12 18 24 30
20,000 -
15.000
10,000 -
5.000 -
DWS Range
c
S
8
I
7
f
20
15
10
A n i.
• Exxon 67 fines
• Natural gypsum_
i
j_
2 4 6 8 10
n = Total No. of Intermittent Leaching
i i i i
180 360 540
Total Leach Time = 72x In), hr
720
Figure 8 - Leachate Characteristics of Exxon 67 Third Cyclone
Particulates as a Function of Intermittent
Leaching
65
-------
• SO* concentration - The sulfate concentrations in the pre-
vious FBC leachates were similar to those in the natural
gypsum leachate and were dominated by CaSO, solubility,
but in the Exxon 67 leachate SO, was 5 to 15 times higher.
The high SO, decreased rapidly with the total leachate
volume passing through the sample and with total leaching
time. After four 72-hr cycles, the Exxon 67 leachate SO^
equilibrated to within the previous SO^ range, which
appeared to be controlled by CaSO^ solubility (Figure 8).
• Specific Conductance - The total dissolved solids as
estimated by specific conductance were also higher in the
Exxon 67 leachate than in the previous carry-over mate-
rials. Like calcium, magnesium, and SO^, the TDS settled
down much like natural gypsum leachate after four 72-hr
leach cycles (Figure 8).
• Calcium solubility - The calcium solubility seemed to be
suppressed initially when magnesium and SO. concentrations
were high but increased to be dominated by CaSO, solubil-
ity when magnesium was depleted after four cycles
(Figure 8).
• £H - The initial leachate pH from Exxon 67 fines was lower
than most FBC carry-over leachate pH (^9 to 10), possibly
because of the lower CaO content of this material.
• Trace elements - Unlike the previous findings, several
trace elements (manganese, selenium, fluorine, and iron)
exceeded the DWS, as shown on Table 24, with deionized water
leaching. This increase could be due to the lower leachate pH.
The higher magnesium, SO,, and TDS in the initial leachate suggested
20
the presence of a highly soluble species, MgSO^ or CaSO^-SMgSO^. ANL
has reported the presence of CaSO,•SMgSO, in some FBC sorbent. The
chemical analysis of the Exxon 67 carry-over solid also suggested the
presence of other sulfate in addition to CaSO, because the SO,-to-calcium
66
-------
molar ratio was greater than 1 (S0,/Ca = 1.3). X-ray diffraction, how-
ever, showed the presence of neither the magnesium salt suggested above
nor any other sulfate (in addition to CaSO ), so that if any sulfate
salts were present in addition to CaSO,, the concentrations were lower.
Leaching tests were also conducted on Exxon 67 residue using sodium
acetate/acetic acid buffer with pH = 4.4 and specific conductance
28
3.31 millimho/cm, as suggested by the ASTM proposed leaching method to
simulate an inhomogeneous landfill site where codisposal of municipal
and industrial waste often results in acidic leaching conditions. In
addition, at the time these tests were conducted, it was known that EPA
was considering an acetate/acetic acid shake test in order to determine
whether a waste should be considered hazardous under RCRA because of
toxicity.
Note from Table 24 that leaching with a lower pH medium had the
following effect on the resultant leachates:
• Aluminum, calcium, cadmium, magnesium, manganese, silicon,
SO,, and IDS increased with decreasing pH. One would
expect that the more acidic leaching medium might tend to
dissolve the alkaline calcium-containing residue, thus
exacerbating the leaching of components out of the residue.
• Arsenic, boron, selenium, vanadium, zinc, fluorine, chlo-
rine, and pH of the resultant leachate decreased with
decreasing pH.
Leaching with an acidic medium is discussed further in a later sec-
tion, "Leaching Medium Effect on Leachate."
The residual heat-release property of Exxon 67 fines (discussed in
a later section) falls into the range of the FBC carry-over materials
investigated previously.
The leaching property of the PFBC Exxon 67 third-cyclone particulate
indicated the need for further tests on FBC carry-over, particularly
fines of less than 15 vm in size. Previous AFBC materials from a MERC
67
-------
run have also shown that although the major species are lower in the
leachate of the carry-over material, trace elements increase both in the
solid and the leachate with decreasing particle size.
26
Combustion Power Residue
The CPU-400 Process Development Unit (PDU) was originally con-
structed for EPA at Combustion Power Co. to convert the heat energy of
municipal solid waste to electrical energy by using a fluid-bed combustor/
gas turbine cycle. The pressurized fluidized-bed combustor on the CPU-
400 unit is called an adiabatic combustor because no heat transfer sur-
face is immersed in the bed; rather, bed temperature is controlled by
using high levels of excess air (^300% excess) in the combustor. More
recently, the facility has been used under DOE funding to demonstrate
the viability of the direct coal-fired gas-turbine approach based on
CPU-400 technology. The P-403 run was a 175-hr test in the CPU-400 at
405.2 kPa (4 atm) pressure and 855°C. Illinois No. 6 coal and Kaiser
dolomite were used. Burgess No. 10 pigment (aluminum silicate) was fed
with the coal-dolomite mix at a rate of 0.4 percent by weight of the
coal feed as a corrosion inhibitor. Note that the P-403 sample tested
here is an aged sample (>2 years old); for this reason — as well as
because of the use of the pigment — the samples may not represent
typical residue from the adiabatic PFBC system.
The residue sample obtained from run P-403 was carry-over material
collected in a baghouse. Both physical and morphological characteristics
were investigated. The particle size ranged between <1 and 40 urn. Fig-
ure 9 shows a SEM photomicrograph of P-403 carry-over and its EDAX spectra
scanned over the entire area and three specific locations. The area-
scan EDAX analysis (Figure 9b) indicated the presence of calcium, silicon,
sulfur, magnesium, and potassium, in decreasing order.
The bright particle shown at Site A is found to consist of CaSO,,
SiCL, MgO, and Fe-O- (Figure 9c). The dark particle shown at Site B
contains mostly unsulfated CaCO., or CaO (Figure 9d). Another dark area
at Site C shows CaSO predominantly (Figure 9e).
68
-------
The leaching property of P-403 ash was investigated, employing
both the continuous and intermittent shake procedure. Table 25 and
Figure 10 summarize the results. Leachate of the P-403 ash from the
adiabatic combustor showed chemical characteristics similar to those
of the nonadiabatic PFBC ash from the standpoint of leaching of major
calcium species.
The P-403 leachate had low TOC. The trace metal elements among
other chemical characteristics in the solid and its leachate are sum-
marized in Table 25. The P-403 leachate met the drinking water standards
for those elements where standards exist. Although there is no drinking
water standard for boron, it should be noted that the P-403 ash leachate
had a boron concentration of more than 5 ppm. The EPA-recommended cri-
terion for boron for long-term irrigation on sensitive crops is 0.75 ppm.
The recommended maximum concentration for long-term irrigation for
tolerant plants is 2 ppm. For shorter periods of time, higher concentra-
tions are acceptable.
Processed FBC Residue Characterization and Leachate
As discussed in the previous Westinghouse report to EPA and Sec-
tion 8 of this report, Westinghouse is investigating the means for
reducing the environmental impact of FBC residue disposal. One option,
which has received the most attention, is the fixation of FBC spent
sorbent/carry-over mixtures at room temperature. The basis for this
effort is that, upon exposure to water, mixtures of FBC spent sorbent
and fly ash tend to set up as a cementlike mass. For more complete dis-
cussions on spent sorbent processing (i.e., fixation), refer to Section 8
of this report as well as to the previous Westinghouse report to EPA.*
As part of the laboratory leaching studies described in this chapter,
continuous and intermittent, shake tests were conducted on samples of
fixed material produced as described in the previous Westinghouse report
*Pages 123-135 of Reference 17 describe the preparation of these sam-
ples. Note, however, that the caption of Table 34, p. 128, should
refer to Exxon Run 27, not 43.
69
-------
Table 25
Dwa. 1709B92
SOLID AND LEACHATE CHARACTERISTICS OF ADIABATIC PFBC
BAGHOUSE RESIDUE FOR COMBUSTION POWER CP-403 RUN
Substance
Al
Aq
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hq
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
S04
S=
TOO
DH
SC. Mmho/cm
Solid.
ppm
Major
<1
500
<10
16%
<3
15
100
100
Major
, &7%
220
50
> 1.000
80
50
<33
Major
<10
100
> 1.000
60
150
18.5*
0.02%
Leach ate.
mqli,
2
<0.01
< 0.003
>5
<1
<0.01
<0.01
'///Mb//,
<0.01
<0.01
5
<0.05
4
<0.01
>10
<0.05
0.02
<1
<1
VAWM,
<5
<10
'/Alt//.
'/Vftk
a
DWS,
mglt
a 05
a 05
LO
200
0.01
0.05
1.0
0.3
0.002
150
0.05
2.0
0.05
0.01
1.0
5.0
250
6.5to9.2
-750
DWS - NIPDWR. USPHS. and WHO drinking water standards
0 exceeds DWS
70
-------
(a)
(d)
(c)
(e)
Figure 9 -- Morphological Characteristics of CP-P403 Ash: (a) Typical
SEM Photomicrograph, (b) Area-Scan EDAX Spectrum, (c) EDAX
on Site A, (d) EDAX on Site B, (e) EDAX on Site C
71
RM-73656
-------
Curve 691327-B
2000
Normalized Leachate Quantity, ml/g of Starting Material
6 12 18 24 30 36 42
en
O
1000
T
^ f\ S\ +* *+
• DWS
2000
^
CT>
c^lOOO
en
O O
DWS
i
O.
E
o
"c»
E
o
o
1U
8
6
4
2
n
i i i i i i i
- -
- -
- °^°^^a__ _
DWS.approx'd
( i i i i i i
24 6 8 10 12 14
n = Total No. of 72-hr Intermittent Leaching
360 720
Total Leaching Time = (72) (n), hr
1080
Figure 10 - Leachate Characteristics of CPU-400 P403 Ash from Combustion
Power as a Function of Intermittent Leaching
72
-------
to EPA. These shake tests were to identify how effectively this low-
temperature fixation reduced leaching properties. Among the fixation
parameters investigated were the sorbent/ash-mixing ratio, water content,
and air-curing time. Table 26 summarizes the samples studied. All are
5-cm cubes unless otherwise stated. Unprocessed spent bed material,
carry-over (ash), an 80/20 bed/carry-over (sorbent/ash) mixture and
gypsum were tested for comparison. The operating conditions of Exxon 27
runs from which the starting materials (FBC residues) were produced are
presented in Table 8 of this report. Note that in the work reported here
the only fly ash mixed with the spent bed material was carry-over from
the same PFBC miniplant run that produced the spent bed material used in
these tests. Work has also been carried out to investigate means for
increasing the strength of the fixed residue by using fly ash generated
in high-temperature conventional coal boilers; use of conventional boiler
»
fly ash was not considered in this study.
Tables 27 through 30 summarize the leaching results on the 5-cm-cube
compacts; all samples were leached as uncrushed cubes unless specified
otherwise. On the basis of limited data points, no definite correlation
was found between leaching property and compact processing parameters
(sorbent/ash ratio, water content, and curing time). The initial
leachates displayed some variation among different compacts, including
identically processed cubes from different batches (HAD and II4D').
After five repeated, 72-hr shakes, the leachate quality of all compacts
fell within the same range, which was superior to that from natural
gypsum. The leachate quality from the 1080-hr continuous shake of all
the compacts reached the level of gypsum leachate except that the compacts
had a slightly higher pH.
Figure 11 compares the chemical characteristics of the leachate
from processed sorbent/ash compact (I4D') prepared from Exxon 27 residues,
the leachate of crushed powder from a similar compact (I4C1), and the
leachate from an unprocessed bed/carry-over mixture (without being fixed:
cast/cured). Although the spent bed/carry-over ratio (80/20) of the
latter unprocessed mixture was different from that (64/36) of the pro-
cessed compacts I4D' and I4C', the comparison is meant to show general
73
-------
Table 26
SUMMARY OF LEACHING TEST METHODS APPLIED TO THE EXXON 27 SORBENT/
ASH COMPACTS AND UNPROCESSED SORBENT/ASH AND MIXTURES
(Solids from EPA's PFBC Miniplant at Exxon)
Sample
I1C
I ID
I2C
120
I3C
130
I4C'
140'
II 1C
II 10
II 3C
1130
II 3E
II 3F
II 4C
II 40
II 4C1
1140'
III2C
111 20
Exxon 19.6
Bed Material
Exxon 19.6
Fly Ash
Exxon 27
Soroent/Ash
Mixture
Gypsum
Description
Wt ratio riy ash/total solid =0.358
Wt ratio H20/total solid = 0.445
Air cured 7 days
Same as 1 1C
Wt ratio fly ash/total solid =0.358
Wt ra
-------
Table 27
SPECIFIC CONDUCTANCE OF LEACHATE FROM
THE EXXON 27 SOKBENT/ASH COMPACTS
Dug. J612CZO
Sample
I1C
110
I2C
I2D
I3C
130
I4C1 (crushed)
140'
II 1C
HID
II 3C
II 30
II 3E
II 3F
II 4C
II 40
II 4C1 (crushed)
II 40'
III2C
III 20
Unprocessed
Exxon 19.6 Bed
Unprocessed
Exxon 19. 6 Fly Ash
Unproc. Exxon 27
Sorbent/Ash
Mixture (80/201
Gypsum
Specific Conductance, millimho/cm
Intermittent Shake, no. of 72 hr Intervals
1
2.58
2.43
2.32
3.43
2.93
2.74
5.17
4.84
2.59
7.31
3.59
4.32
8.29
4.09
7.93
2.12
2
2.38
2.08
2.18
2.81
1 85
1.75
3.17
3.46
2.68
4.9
2.88
2.12
8.50
3
2.06
166
4
1.77
i 1.45
1.85 ' 1.60
2.26 2.35
1.43
1.43
1.81 ' 1.42
1.92
2.92
1.67
3.%
1.97
1.65
7.90
2.28 , 2.01
7.13
2.21
2.95
2.09
1.63
2.12
L47
3.32
1.39
1.36
7.86
1.88
2.66
2.08
5
1.52
1 57
1.75
2.11
1.34
1.25
1.74
4.44
1.20
2.62
1.05
1.21
7.17
1.64
2.27
1.96
6
1.42
1.14
1.66
2.13
1.24
1.06
1.41
1.71
1.12
2.62
1.09
1.19
6.52
1.70
2.27
2.08
7
1.12
1.08
1.43
2.01
1.21
1.20
1 21
1.63
l.'U
2.25
1.00
0.81
6.11
1.58
2.33
2.10
8
1.01
1.05
1.30
1.82
1.16
1.11
1.07
1.39
0.92
1.98
0.87
0.88
3.97
1.54
2.40
2.12
9
1.17
0.97
1.20
1.77
1 23
1.08
0.83
1.13
0.88
2.00
0.85
1.03
3.09
1.21
2.19
2.05
10
0.98
1.06
1.18
1.77
1.19
1.29
0.82
1.16
0.83
1.82
0.83
0.88
2.43
1.25
2.07
1.91
11
0.99
1.13
1.01
1.73
1.12
1.20
0.84
1.14
0.95
1.78
0.81
0.97
2.45
1.23
2.33
2.12
12
0.74
1.06
1.01
1.55
1.05
1.17
0.76
1.11
0.85
1.65
0.94
0.8
2.30
1.22
2.27
2.16
13
0.9
1.00
1.06
1.51
0.99
1.05
0.73
1.10
1.07
1.63
0.74
0.86.
2.21
1.10
2.23
2.12
14
0.85
0.91
0.93
1.63
1.15
1.00
0.72
1.10
0.90
1.50
0.62
0.93
1.82
0.75
1.87
2.08
15
0.82
0.93
0.89
1.62
1.14
0.77
0.72
1.05
0.86
1.50
0.57
0.85
1.92
0.83
1.95
1.96
1080 hr.
Confs
Shake
2.55
2.57
2.72
2.23
2.16
2.25
2.37
2.10
-------
Table 28
pH OF LEACHATE FROM THE EXXON 27 SORBENT/ASH COMPACTS
Dug. Z612C23
Sample
I1C
110
I2C
120
I3C
130
I4C'
(Crushed)
140'
II 1C
II 10
II 3C
II 30
II 3E
II 3F
II 4C
II 40
II 4C'
II 40'
III2C
III 20
Unprocessed
Exxon 19.6 Bed
Unprocessed
Exxon 19. 6 Fly Ash
Unproc. Exxon 27
Sorbent/Ash
Mixture (SO/ 20)
Gypsum
PH
Intermittent Shake, no. of 72 hr Intervals
1
10.3
10.2
10.2
11.3
11.2
11.5
12.1
12.0
11.4
12.2
11.9
11 9
12.3
11.9
12 1
7.7
2
10.8
11.0
11.0
11.3
10.7
11.4
11.7
11.7
11.4
11.8
11.5
11 5
12.1
10.5
12 0
7.6
3
10.9
11.0
10.8
11.1
10.7
11.5
11.4
11.6
11.2
11.6
11 S
11.5
12.2
10.4
11.4
7.8
4
10.7
10.9
10.9
11.1
10.7
11.2
11.4
11.4
11.1
11.7
11 3
11 2
12.1
10.1
11 1
7.6
5
10.6
10.7
10.5
10.9
10.7
10.9
11.4
10.9
11.1
11.5
11 1
10.6
12.2
9.7
10.4
7.6
6
10.4
10.6
11.5
10.9
10.4
10.9
10.5
10.5
11.1
11.5
11.1
11.0
12.2
9.6
10.3
7.6
7
10.2
10.2
9.5
10.7
1C. 3
11.0
10.5
10.6
11.0
11.2
11.0
10.9
11.9
9.8
10.3
7.5
8
10.1
10.2
9.4
10.7
10.2
11.4
10.8
10.8
10.9
10.5
10 7
10.9
11.7
10.1
10.6
7.3
9
9.9
9.7
9.1
10.7
9.9
10.9
10.5
10.4
10.8
10.7
10 6
10.9
11.6
10.4
10.7
7.4
10
9.8
9.6
9.0
10.7
9.7
10.9
10.5
10.2
10.7
10.7
10.5
10.0
11.4
10.1
10.7
7.9
11
9.1
9.5
8.8
10.6
9.9
10.9
10.4
10.3
9.9
10.7
10 2
10.0
11.1
10.2
10.8
7.6
12
9.6
9.2
8.8
10.4
9.8
10.6
10.2
10.2
10.3
9.8
10 0
10.3
10.6
9.9
10.4
7.4
13
9.9
8.9
8.9
10.3
9.6
10.5
9.9
9.9
10.2
9.8
10 0
9.8
10.4
9.9
10.3
7.5
14
9.9
8.7
8.5
10.1
9.4
10.2
9.8
9.8
9.4
10.2
9 7
9.8
10.2
9.7
9.9
7.5
15
10.0
8.6
8.5
10.1
9.4
9.5
9.5
9.7
9.2
10.2
9 1
9 3
10.3
9.6
9 8
7.3
1080 hr
Cent's
Shake
8.5
7.7
7.9
8.4
8.7
9.1
8.3
8.3
-------
Table 29
CALCIUM CONCENTRATION OF LEACHATE FROM
THE EXXON 27 SORBENT/ASH COMPACTS
Dug. 2612C22
Sample
I1C
I ID
I2C
120
I3C
130
I4C1
(Crushed)
140'
II 1C
HID
II 3C
II 3D
II 3E
II 3F
II 4C
II 40
II 4C' (Crushed!
II 40'
III2C
III 20
Unprocessed
Exxon 19. 6 Bed
Unprocessed
Exxon 19. 6 Fly Ash
Unproc. Exxon 27
Sorbent/Ash
Mixture (80/20)
Gypsum
Calcium, mg/1
Intermittent Shake, no. of 72-hr Intervals
1
572
552
548
648
604
640
896
868
600
1224
744
776
1184
736
1024
632
2
544
456
520
628
432
360
584
664
544
832
536
408
1336
568
1168
608
3
544
376
432
604
352
384
388
608
3.6
720
336
328
1168
548
712
624
4
400
316
336
568
328
258
328
464
300
680
260
260
1232
512
6%
616
5
344
304
360
536
280
272
376
448
200
632
204
260
1232
472
672
5%
6
320
248
344
512
264
184
2%
3%
204
5%
200
232
968
480
680
624
7
248
240
2%
5M
268
224
240
356
264
548
168
160
1008
424
680
616
8
224
240
288
440
240
212
220
328
160
488
168
188
800
400
672
608
9
2%
200
256
428
280
208
184
2%
152
508
168
208
744
292
632
616
10
208
200
248
416
248
256
176
276
152
448
156
184
700
356
656
608
11
212
208
224
412
240
248
168
268
180
440
152
204
680
320
648
624
12
136
200
224
392
240
208
152
256
152
416
180
168
652
296
620
632
13
168
196
192
372
216
208
148
260
200
420
144
200
644
272
632
624
14
164
176
184
384
232
184
136
236
176
416
128
192
528
172
528
616
15
168
176
176
400
256
160
144
240
168
404
104
160
600
200
588
608
1080 hr
Cont's
Shake
576
552
5%
560
560
576
544
556
-------
Table 30
SULFATE CONCENTRATION OF LEACHATE FROM
THE EXXON 27 SORBENT/ASH COMPACTS
Sample
1 1C
110
I2C
I2D
I3C
130
I4C'
(Crushed)
I4D'
II1C
HID
II 3C
II 3D
II 3£
II V
INC
II 4D
II 4C1 (Crushed!
II 40'
III2C
III 20
Unprocessed
Exxon 19.6 Bed
Unprocessed
Exxon 19. 6 Fly Ash
Unproc. Exxon 27
Sorbent/Ash
Mixture 180/20)
Gypsum
Sulfate. mg/l
Intermittent Shake, no. of 72 hr Intervals
1
1600
1493
1325
1353
1236
983
1000
1040
1083
1212
1175
1163
1410
1260
1460
1285
2
1260
1105
1175
1273
893
543
903
1150
1128
1071
703
751
1247
1083
1352
1187
3
1071
840
876
1260
745
682
633
1060
543
1225
575
620
1427
1150
1380
1380
4
903
683
737
1105
632
420
575
932
509
1236
611
503
1187
940
1285
1285
5
885
710
720
1225
710
611
704
894
488
1645
425
620
1492
1127
1445
1162
6
704
641
665
1020
611
403
633
813
438
1127
382
575
1272
893
1175
1000
7
555
565
627
1083
632
543
530
751
555
1116
376
399
1225
950
1690
1582
8
515
555
633
991
620
487
502
650
440
950
370
529
1690
950
1737
1475
9
690
418
549
885
650
470
414
620
368
1093
337
439
1667
730
1565
1475
10
400
489
487
950
632
543
358
536
343
962
382
399
1622
830
1582
1547
11
850
503
530
885
590
536
363
575
457
982
410
466
1366
565
1247
1272
12
418
488
503
830
530
674
334
543
348
960
477
399
1260
633
1175
1272
13
530
448
455
777
440
502
334
555
466
922
324
466
1410
530
1325
1225
14
575
408
425
840
548
522
316
650
406
830
270
414
1010
355
1010
1352
15
503
418
382
858
522
395
358
603
439
850
240
363
1175
438
1236
1366
IflMhr
Cent's
Shake
1692
1410
1445
1366
1260
1475
1548
1565
78
-------
Curve 689565-B
Normalized Leachate Quantity. mJ/g of Starting Solid
e
o
8
I
3
3
o
1
l/l
1
\
*
\
Gypsum
Unprocessed Sorbent/Ash
-Mixture, Exxon Run No. 27
Crushed Powder from Bed/Ash
Compact, I4C'
Bed/Ash Compact, I4D'
-DWS, approx'd
L
_L
_L
5 10 15
(360) (720) (1080)
n = Total No. of 72-hr Intermittent Leach
(t = Total Leach Time)
Figure 11 - Comparison of Leachate Characteristics of Processed
Exxon 27 Sorbent/Ash Compact, Crushed Powder from
Such Compact, Unprocessed Sorbent/Ash Mixture, and
Natural Gypsum
79
-------
trends. We note that the compact as cast and cured produced a leachate
of better quality, i.e., lower calcium, SO,, specific conductance, and
pH, than that induced from the crushed compact in powder form, which
in turn gave better leachate quality than did the unprocessed sorbent/
ash mixture from the same FBC run. These findings are extremely important.
The fact that the leachate of the crushed, fixed material was worse than
that of the uncrushed compact indicates that the decreased surface area
in the fixed (compact) FBC residue favorably affects the resultant leach-
ing property, and suggests that increased leachability might result from
fissuring a disposal pile of fixed FBC material. A mixture of unpro-
cessed spent bed/carry-over had higher leachate concentrations, however,
than did a crushed, fixed cube, indicating that processing has a favorable
effect that will not be entirely eliminated by crushing or fissuring.
Nevertheless, despite fixation/processing according to the procedures
tested in Section 8, the initial leachate calcium, SO, , and TDS are still
above DWS. Comparison of leaching results from processed and unprocessed
residues is presented in Section 8.
Figure 12, which one can extrapolate to long-term leachability, again
indicates that gypsum leached constantly, independently of the total
leach time and the volume of leachate passing through the sample. The
leachates from the sorbent/ash compact, the crushed compact powder, and
the unprocessed sorbent/ash mixture improved with time and total volume
and, therefore, are much less contaminating with long-term leachability
of CaSO, than is the natural gypsum.
Table 31 shows leachate characteristics of processed FBC residue as
a function of leaching cycle from a series of ten 72-hr intermittent
shake tests. The TOC was low throughout the tests. Figures 11 and 12,
and Tables 27 through 30 show improvement of major species and pH.
Table 31 illustrates a trend of decreasing trace element concentrations
(e.g., Al, B, Ca, Cu, Mo, Si) in leachates as a function of repeated
shake cycles. These concentrations are extremely low because the trace
element concentrations in the initial leachate (first 72-hr shake) are
below the DWS.
80
-------
Curve 689564-A
Normalized Leachate Quantity, m£/g of Starting Solid
—I
1
E
u
i
o
i
'E
u
c
S3
o
3
1
O
O
.y
I
1U.U
8.0
6.0
4.0
2n
.0
1.0
0.8
0.6
0.4
0.2
1 I l 1 i II
- o -
D
— ~~
-
O
'""""""""••^^o "Q~~^>^_^
^^***"^^**fc^ **"~ 1 .-.
ft* 2^_^ S^rT-iS™
~ A ^ TT^*^»
""*~~*^ ^^'CWla>OCL'
•"*->>>v.^*
^r^****^s-
~ n\A/c ———
Gypsum
— a — Unprocessed Sorbent/Ash Mixture
— —
—o— I 4C' Crushed Powder
— • — I4D' Original Compact
—
i i i i 1 i i i i 1 i
20
40
1000
4000
60 80 100
Total Leach Time, hr
Figure 12 - Comparison of Specific Conductance of Leachates from
Exxon 27 Sorbent/Ash Compact, Crushed Compact,
Unprocessed Exxon 27 Sorbent/Ash Mixture, and
Gypsum
Activity tests showed no heat release from these compacts on con-
tact with water, as one would expect, since the casting involved a wet
processing step in which CaO would have been hydrated.
Finally, results presented here clearly indicate that the sorbent/
ash compacting process improves the leaching property and reduces the
potential negative environmental impact of leachate contamination.
Leaching Medium Effect on Leachate
The leaching property of FBC residue was investigated as a function
of leaching medium. Three media with varying pH levels were investigated.
First, deionized water was used in the majority of the leaching tests
(except where specified otherwise). Second, studies using CCL-saturated
81
-------
Table 31
Own. 1709B90
LEACHATE CHARACTERISTICS OF PROCESSED FBC RESIDUE
AS A FUNCTION OF LEACHING CYCLE
(Ten 72-hr Intermittent Shake Tests of Compact II4D Prepared
by Sorbent/Ash Blending Using Exxon 27 Solid)
Substance
Al
Ag
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hg
Mg
Mn
Mo
Ni
Pb
Sb
Se
Si
Sn
Ti
V
Zn
Zr
$04
TOC
DH
SC. u mho/cm
Leachate of Processed FBC Compact, mg/,0 a
1
0.3
<0.01
30
<0.03
<0.01
<0.01
<1
<1
'///?&//
<10
///,\\. &/
wv%
i
0.3
<0.01
<0.05
0.5
<1
<0. 1
<0.01
'//^VA
<0.01
<0. 1
<0.01
0.01
0.02
< 0.002
19.2
<0.01
0.02
<0. 1
<0.03
<0.03
<0.01
>20
<0.03
<0.01
<0.01
<1
<1
YMV/s
<10
'/.ILK/,
^
3
0.3
<0.01
<0.05
0.2
<1
<0. 1
<0.01
//W//,
<0.01
<0. 1
<0.01
0.01
0.02
< 0.002
<10
<0.01
0.01
<0. 1
<0.03
<0.03
<0.01
>20
<0.03
<0.01
<0.01
<1
<1
\/,W//s
<10
'/1L W/,
'ftW/,
5
0.2
<0.01
<0.05
0.01
<<1
<0. 1
<0.01
Wll//
<0.01
<0.1
<0.01
0.005
0.005
< 0.002
<5
<0.01
0.003
<0.1
<0.03
<0.03
<0.01
>10
<0.03
<0.01
<0.01
<1
<1
'/&V/.
<10
'MO. 85^
'#&/,
1
0.1
<0.01
<0.05
0.1
<1
<0. 1
<0.01
//?&//,
<0.01
<0. 1
<0.01
0.005
< 0.005
< 0.002
14.4
<0.01
0.003
<0. 1
<0.03
<0.03
<0.01
>10
<0.03
<0.01
<0.01
<1
<1
'/J*V/
<10
//to. 10
<0.03
<0.01
<0.01
<1
<1
WW)
<10 1
>/10. 94-S
//1. 080>
'/////
b
DWS.
mg/f
0.05
0.05
1.0
200
0.01
0.05
1.0
0.3
0.002
150
0.05
2.0
0.05
0.01
1.0
5.0
250
6. 5 to 9. 2
-750
b
Number of 72-hr intervals
DWS - NIPDWR. USPHS. and WHO drinking water standards
exceeds DWS
82
-------
deionized water (pH = 4) were conducted to simulate surface water leach-
ing where dissolved C02 might be high. The preliminary results with
MERC residue discussed previously indicated that the effect of leaching
with C02~saturated deionized water was not pronounced, probably because
of the high alkaline content in the FBC residues. Third, leaching tests
were also conducted using a sodium acetate/acetic acid buffer with a
pH = 4.5 and a specific conductance of 3.3 millimho/cm, as suggested by
28
the proposed ASTM test, and as being considered by EPA under RCRA, to
simulate an inhomogeneous disposal site where codisposal of municipal
and industrial wastes often results in acidic leaching conditions.
Results from Exxon 67 fines were discussed previously.
Several residues, in addition to that of Exxon 67, were tested
using the buffered medium prepared by dissolving 4.9 g of glacial acetic
acid and 3.7 g of sodium acetate in 1 I of deionized water. The resultant
buffer solution has a pH = 4.5 ± 0.1 and a specific conductance of
3.3 millimho/cm. The 200-hr continuous shake tests using the acidic buffer
medium were carried out in the preliminary study presented here. The
final leachate pH is dictated by the solid residue, since no additional
acid was added to the original solid/acetate medium mixture.
Table 32 compares leachate characteristics of a variety of FBC
residues (spent limestone and dolomite) and natural gypsum, using
deionized water and a sodium acetate/acetic acid buffer solution as the
leaching media. The samples listed in Table 32 are described in Table 8.
Preliminary comparisons of the results indicate the following effects:
• The pH of the leachate resulting from the highly alkaline
residue, using an acidic leaching medium, was decreased
only slightly in comparison with deionized water. The one
major exception was Exxon 67 fines, whose pH of 8 with
deionized water fell to 4.7 with the buffer. This result
is not surprising, since the Exxon 67 fines are not as
strongly alkaline as the other residues.
83
-------
Table 32
Dwg. I702B01
COMPARISON OF LEACHATE CHARACTERISTICS
WITH WATER AND ACETATE LEACHING MEDIA
(200-hr Continuous Shake Tests)
Sample
Battell e Bed
Battelle Bed
Battelle Carry-over
Battelle Carry-over
B & W No. 19 Bed
B & W No. 19 Bed
B&WN0.19 Carry-over
B&WNo.l9Carry-over
Exxon 43.2/43. 3 Bed
Exxon 43. 2/43. 3 Bed
Exxon 43. 3 Carry-over
Exxon 43. 3 Carry-over
ANL C2/C3 Bed
ANL C2/C3 Bed
Exxon 67 Carry-over.
Fines < 15pm
Exxon 67 Carry-over,
Fines < 15 urn
Gypsum
Gypsum
Sodium Acetate/
Acetic Acid Buffer
Sorbent
Type
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Dolomite
Dolomite
Dolomite
Dolomite
Dolomite
Dolomite
Dolomite
Dolomite
-
-
—
Leach Medium
DionizedH20
Acetate buffer
Dionized H20
Acetate buffer
Dionized HzO
Acetate buffer
Dionized HpO
Acetate buffer
Dionized H^
Acetate buffer
Dionized H;*)
Acetate buffer
Dionized 1^0
Acetate buffer
DionizedHoO
£
Acetate buffer
Dionized HoO
Acetate buffer
—
pH
12.1
12.1
12.0
12.0
12.2
12.1
11.9
11.4
12.2
12. 1
11.3
9.2
11.9
10.0
8.0
4.7
7.4
46
45
Specific
Conductance.
millimho/cm
8.6
144
8.4
13.0
9.24
146
6.47
8.7
& 43
13.9
2.99
9.0
3.98
9.0
9.16
13.0
2.15
5.75
3.31
Ca.
mqll
1280
3040
1072
2640
1392
3064
1184
1832
1304
2848
672
1440
960
1844
488
536
615
976
-
Mq.
mg/X
<20
<20
<20
<20
<20
24
<20
38
<20
53
<20
581
<20
384
1672
2016
<20
48
-
S04.
mqlt
1083
812
1150
982
1200
1030
1248
200
1273
1339
1285
1763
1830
1188
6975
7575
1470
1445
—
84
-------
• Leachate calcium was increased using a sodium acetate/
acetic acid medium. This result is to be expected, also,
since the low pH of the medium may be expected to cause
additional quantities of the alkaline residue to dissolve
in order to raise the leachate pH.
• Magnesium that was found to be insoluble in deionized
water from spent dolomite sorbent was highly leachable
in the acetate leaching medium. Again, this is due to the
solubility of the alkaline residue in the acidic medium.
• The effect on leachability of SO, was not clear.
• Acidic acetate leaching resulted in increased specific
conductance and TDS, because of the higher ionic strength
and lower pH of the leaching medium. This effect was
much more pronounced with FBC residues than with natural
gypsum.
• Effects on trace elements were discussed earlier with
regard to acid leaching of Exxon 67 fines. The concentra-
tions of some trace metals in the leachate increased with
the acidic medium, some decreased. Westinghouse is cur-
rently conducting further studies under the continuing
contract to EPA.
Measurement of Total Dissolved Solids
The TDS in a leachate is a good index of leachate quality. Total
dissolved solid, which can be determined by the time-consuming evaporating
procedure, can also be estimated by multiplying the easily measured
specific conductance by an empirical factor. This factor may vary,
depending on the soluble components in the particular aqueous system
and the temperature of measurement. We have selected a constant tempera-
ture, 25°C, for the latter throughout our leaching studies. This sec-
tion summarizes our effort in determining the multiplying factor empiri-
cally for the FBC leachates.
85
-------
Initially five samples — Exxon 43.2/43.3 bed, Exxon 43.3/carry-over
(ash), ANL C2/C3 bed, B&W 19 bed and carry-over (ash) — were selected
for this investigation to represent a broad spectrum of the FBC spent
materials: limestone/dolomite, PFBC/AFBC, bed/carry-over (ash).
Leachate was induced by a 48-hr shake procedure. A portion of the orig-
inal leachate from each sample was diluted to provide solutions of 1/2,
1/4, and 1/8 fractions of the original concentrations. Specific conduc-
tance, pH, and TDS were determined for all 20 leachate solutions.
The procedure for determining TDS described in Standard Method for
33
Water and Wastewater was used to obtain TDS at the evaporating tem-
perature of 103°C. This was not the true TDS because the residue at
103°C contained physically occluded water, hydration and carbonation
products (Ca(OH) , CaSO, -j H2 CaSO^'y H2
-------
resulted from samples with a lower pH (9 to 11). The reasons are clear:
spent materials in the former case were rich in CaO, resulting in a higher
pH and in greater conductivity due to the presence of large numbers of
highly mobile and conductive hydroxide ions in the solutions. On the
other hand, spent materials that were high in CaCO~ and Si02 had lower
hydroxide concentrations in their leachate and, therefore, lower conduc-
tivity and pH.
To verify the postulated grouping, eight additional samples were
investigated. In addition to the FBC samples, these samples included
four spent bed residues from the chemically-active fluidized bed (CAFB)
process. These CAFB samples consisted of large quantities of CaO and
very little Si02, since they were residues from CAFB oil gasification
runs where coal ash was not present. The results from the additional
eight samples shown on Figure 14 confirmed the slopes measured on the
original five materials. Table 33 summarizes the correlation between
TDS and conductance for all the samples studied.
Similar conclusions can be drawn for the FBC and the CAFB data.
For spent materials with large amounts of CaO present (e.g., AFBC bed
and carry-over, and PFBC bed where significant CaO is present), leachates
generally have a pH of ^12, and a specific conductance of >7000 ymho/cm.
A multiplying factor of 0.37 should be applied to approximate TDS from
specific conductance in this case. On the other hand, a higher factor,
0.87, should be multiplied by the specific conductance to estimate the
TDS in leachates of spent materials with more CaCO- and Si02 (e.g., PFBC
carry-over and most PFBC bed material where CaCO,, is the major species).
These leachates generally have a lower pH (<11.5) and lower specific con-
ductance (2000 to 4000 umho/cm).
Note that the results presented here are empirical, based on typical
FBC leachates. The TDS obtained in this manner are only approximated
values. Note that a typical FBC leachate has a TDS of approximately
3000 mg/£ and the DWS for TDS is 500 mg/£.
87
-------
Curve 691637-A
3000
_ Slope =0.87 mg-cm-
limho- t
oo
co
o
CO
8 2000
!£
C3
S
.O
S
1000
4000 -
Slope =0.37 mg-cm
prnho- t
3000
o
1 2000
I
Sample 1: Exxon 43.2/43.3 Bed o
Sample 2: Exxon 43.3 Ash o
Sample 3: Argonne C2/C, Bed v
Sample 4: B&W 19 Bed A
Sample 5: B&W 19 Ash •
I
1000
Slope =0.87 mg-cm-pmho- t
Slope =0.37 mq-cm-pmho- t
Exxon 43.2/43.3 Bed
Exxon 43.3 Ash
ANLC2/C3Bed
B & W 19 Bed
B & W 19 Ash
Exxon 8.4 Bed
ANL LSI - 2 Bed
Exxon 34 Fines
Rs - PER Ash
CAFB - 10A Bed
CAFB - 10 Ash
CAFB -8 Fines
CAFB - 8 Bed
I I I I I I • I I I
0 2 4 6 8 10
Specific Conductance, millimho/cm
Figure 13 - Correlation between IDS and
Specific Conductance in the
FBC Leachate System
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Specific Conductance, millimho/cm
Figure 14 - Correlation between IDS
and Specific Conductance
in FBC and CAFB Leachates
-------
00
VO
Table 33
CORRELATION BETWEEN TDS AND SPECIFIC CONDUCTANCE
Sample
ANL LST-2 Bed
ANL C2/C3 Bed
Exxon 43.3 Ash
Exxon 34 Fines
Exxon 8.4 Bed
Exxon 43.2/43.3 Bed
B&W 19 Bed
B&W 19 Ash
RS-PER Ash
CAFB-10A Bed
CAFB-10 Ash
CAFB-8 Bed
CAFli-8 Fines
Description
PFBC bed, dolomite
PFBC bed, dolomite
PFBC ash, dolomite
PFBC fines
PFBC bed
PFBC bed
AFBC bed
AFBC ash
AFBC ash
CAFB-gasifier bed
CAFB-gasifier ash
CAFB-regenerator bed
CAFB-stack fines
PH
11.3
11.2
9.1
11.0
11.8
12.0
12.0
12.0
11.8
12.0
12.0
12.0
12.0
Sp.
Cond . ,
umho/cm
3670
3240
3770
2700
8940
8780
8980
7850
8060
9610
9700
11900
11900
Multiplying
Factor,
mg-cm-
pmho'1-)!."1
0.87
0.87
0.87
0.87
0.37
0.37
0.37
0.37
0.37
0.37
0.37
0.37
0.37
Calcu-
lated
TDS,
mgM
3193
2819
3280
2349
3309
3249
3323
2905
2980
3556
3589
4403
4403
-------
Thermal Activity Tests
.'-^
The activity of the residual lime in spent bed and carry-over mate-
rial was determined by the lime's heat release property on contact with
water, as the hydration reaction of CaO is extremely exothermic. In
an effort to select an appropriate activity test for the work reported
here, we have reviewed the literature on lime reactivity and slaking
31
rate tests. ASTM C110-76 provides a test for the slaking rate of
quicklime (CaO). In this test 76 g of quicklime is added to 380 ml of
distilled water in a modified Dewar flask covered with a rubber gasket
fitted with a mechanical stirrer. The temperature is read with a ther-
mometer at 30-second to 5-minute intervals, depending on the reactivity
of the quicklime, until a constant temperature is reached. The slaking
rate is determined by the following quantities: temperature rise at
30 seconds, total temperature rise, and active slaking time.
32
Murray studied lime reactivity as a function of porosity and
shrinkage characteristics during calcination and found that calcitic
quicklime of low shrinkage and high porosity was highly reactive. He
used a lime/water ratio of 1/7 by weight. Since preliminary slaking
tests indicated a wide range in slaking rates, an empirical compromise
point was selected as indicative of the rapidity of slaking. The tem-
perature rise in five seconds was selected, and the reactivity coefficient
was designated as AT,-. He readily acknowledges, however, that his test
was based against the extremely reactive limes, in which slaking was
actually completed in three to four seconds, so that a reading at five
seconds made them appear to be slower than they actually were.
American Water Works's standard on lime for water treatment employs
a lime slaking test with lime/water proportions of 100 g/400 ml, follow-
ing the test procedure of ASTM-C110.
The temperature rise of a solid/water system containing free CaO is
a function of solid/water ratio. In our experimental effort to establish
a screening test for the residual activity in spent FBC solids produced
under varying processing conditions, a solid to water proportion of 3 g
90
-------
to 20 ml (which is in the bulk range specified by the ASTM-C110 test and
by Murray's work) was found empirically to provide much better repeatabil-
ity than that from a higher solid/water ratio. A higher ratio would give
greater temperature rise but would lack reproducibility, most likely
because of local heating. The lower ratio was initially adopted as the
screening test for heat-release property because of its speed, the small
quantity of stone required, and the good reproducibility of results.
A higher ratio (small quantity of water added to larger quantity of
solid), however, was also used in some cases because it provides higher
sensitivity and simulates rainfall onto the disposed solid.
Chromel-alumel thermocouples were used to monitor the temperature
rise in the solid/water system with an Omega cold junction compensator
and a millivolt recorder. The heat release tests were conducted on the
actual spent sorbent, fly ash, and fines from the fluidized-bed combustion
units. Calcined and uncalcined limestone and dolomite samples were also
tested for comparison. Table 34 summarizes the maximum temperature rise
when 3 g of solid were added to 20 ml of deionized water in a Dewar
flask. The samples referenced are identified in Table 8.
Results showed that the spent solids from the once-through, pres-
surized FBC system gave off little heat spontaneously on contact with
water. This finding is not surprising, since the CaCO,. fraction of the
spent solids from PFBC may be largely uncalcined because of the high CO-
partial pressure that exists in PFBC units, unless the bed temperature
is high enough to cause calcination.
In cases where there was residual CaO present in the PFBC samples,
they might have been hydrated in air during storage, dead-burned during
the process, or coated with impermeable CaSO, so that little spontaneous
heat of hydration was detected. The only batch of the once-through PFBC
spent materials that released detectable heat was the spent sorbent from
Exxon 43.3. It had an average bed temperature of 940°C, which is high
enough to calcine the CaCO,. even at the elevated CO^ partial pressures
in PFBC. This sample showed a gradual temperature rise over 35 minutes
91
-------
Table 34
HEAT RELEASE PROPERTY OF FBC WASTE
Dwg. 2619C51
Process
PFBC, Once- thru
PFBC, Once- thru
PFBC. Once-thru
PFBC, Once-thru
PFBC. Once-thru
PFBC. Once-thru
PFBC. Once-thru
PFBC. Once-thru
PFBC, Once-thru
PFBC, Once-thru
PFBC, Once-thru
PFBC. Once-thru
PFBC, Once-thru
PFBC. Once-thru
PFBC, Once-thru
PFBC, Once-thru
PFBC. Once- thru
PFBC. Once-thru
PFBC. Once-thru
PFBC. Once- thru
PFBC. Once-thru
PFBC. Regenerative
PFBC, Regenerative
PFBC, Regenerative
PFBC, Regenerative
PFBC.. Regenerative
PFBC, Regenerative
PFBC, Adiabatic
FBC
Unit
ANL
ANL
ANL
ANL
ANL
ANL
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
ANL
ANL
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Exxon
miniplant
Combustion
Power
Run
C2/C3
VAR-4
LST-1
LST-2
LST-3
LST-4
8.4
27
19.6
30.2
43.2
43.3
43.2/43.3
30.4
19.6
21
26
30.4
43.5
26
67
REC-3
CCS- 10
45
45
45
45
P-403
Source
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Bed
Cyclone
ash
Cyclone
ash
Cyclone
ash
Cyclone
ash
Cyclone
ash
Fines
3rd cyclone
fines <15um
Comb, bed
Reg1 tor bed
Comb, bed
Reg. bed
(granular)
Req. bed
(agglomerated)
Cyclone
ash
Baghouse
fines
Solid/Water
3 g/ 20 ml
3g/20ml
3g/20ml
3 g/ 20 ml
3g/20ml
3g/20ml
3g/20ml
3g/20ml
3 g/ 20 ml
3 g/ 20 ml
3g/20ml
3g/20ml
3g/20ml
3g/20ml
3 g/ 20 ml
3 g/ 20 ml
3 g/ 20 ml
3 g/ 20 ml
3 g/ 20 ml
3 g/ 20 ml
3 g/20 ml
3 g/ 20 ml
3 g/20 ml
3 g/20 ml
3 g/20 ml
3g/20ml
3 g/20 ml
3 g/20 ml
ATmax
< 0.2°C
< 0.2°C
< 0.2°C
< 0.2°C
<0,2°C
<0.2°C
< 0.2°C
< 0.2°C
< 0.2°C
< 0.2°C
<0.2°C
1.4°C
<0.2°C
<0.2°C
<0.2°C
<0.2°C
<0.2°C
<0.2°C
<0.2°C
<0.2°C
1°C
<0.2°C
<0.2°C
0.5°C
2°C
<0.2°C
<0.2°C
<0.2°C
Remark
Over 35 min
Slow Rise
Sample obtained
via RS; previously
hydrated and dried
by RS
Sample obtained
via RS
Sample obtained
via RS
In 3 min
Very slow rise over
l.Shr
Over 2 hr
Old sample
stored > 1 yr
92
-------
Table 34 (Continued)
Dwg. 2619C52
Process
AFBC. Once-thru
AFBC. Once-thru
AFBC. Once-thru
AFBC, Once-thru
AFBC. Once-thru
AFBC, Once-thru
AFBC, Once-thru
AFBC, Once-thru
AFBC, Once-thru
AFBC, Once-thru
AFBC. Once-thru
AFBC. Once-thru
Gypsum. Iowa 114
Limestone, 1359
Dolomite.Tymochtee
Calcined Limestone
1359.+35-i8Mesh
FBC
Unit
PER
PER
PER
B&W
B&W
MERC
MERC
MERC
Battelle
Battelle
Battelle
Battelle
Run
Unidentified
Shakedown
Unidentified
19
19
& 3/9/77
«•> 3/9/77
£ 3/9/77
Erosion/
corr. run
Erosion/
corr. run
Erosion/
corr. run
Erosion/
corr. run
Source
Bed
Bed
Fly ash
Bed
Fly ash
Bed
Cyclone
ash
Baghouse
fines
Bed. 1000 hr
Bed. 1500 hr
Cyclone ash
600 hr
Cyclone ash
1500 hr
Solid/Water
3g/20ml
3g/20ml
3g/20ml
3g/20ml
3 g/ 20 ml
3g/20ml
3-g/20 ml
3 g/ 20 ml
3 g/ 20 ml
3 g/ 20 ml
3 g/ 20 ml
3g/20ml
3g/20ml
3 g/ 20 ml
3 g/ 20 ml
3g/20ml
ATrax
<0.2°C
10.1°C
1.8°C
11°C
1.0°C
7°C
0.8°C
1.2°C
Remark
Highly sulfated. from
a storage pile of unknown
history and age
Sample via RS
Sample via RS
In 30 min
In 15 min
< 0.2"C
3.5°C
3°C
2.5°C
<0.2°C
<0.2°C
<0.2°C
> 55°C
In Ihr
In 10 min
In 15 min
In Ihr
Varies with types of
limestone and calcination
temperature and time
93
-------
to reach a maximum of 1.4°C. These results seem to indicate that heat
release may not be a problem for the once-through PFBC process. One must
bear in mind, however, that the heat release property of spent sorbent
is a function of the processing conditions; e.g., temperature, stone
residence time, degree of sulfation, and degree of dead-burning.
We also observed moderate and slow temperature rise for the regen-
erative PFBC spent sorbent when contacting water. This rise is due to
the presence of CaO formed during the one-step regeneration process.
All the spent solids from the AFBC system showed significant heat
release characteristics, with the exception of an aged PER sample that
had an unidentified process and storage history; its result, therefore,
should be discarded. We judge that the heat release property of the
spent sorbent from the AFBC process would probably require special care
in handling and disposal.
The results presented in Table 34 also indicate that the spent bed
material had much higher thermal activity than did the carry-over. This
finding is easily understood in the light of the fact that the carry-over
material generally contains more coal ash and less CaO, which is
responsible for the heat release. Despite the fact that the AFBC bed
material may cause concern because of its thermal activity, the rate
and magnitude of heat it releases when contacting water is much less than
the calcined limestone release, which represents the upper limit case
(where neither sulfation nor high-temperature dead-burning of CaO nor
contamination of coal ash takes place).
We also investigated the heat release property using higher solid/
water ratios. Figure 15 compares the temperature rise as a function of
solid/water ratio for an AFBC spent sorbent and a calcined limestone.
A higher temperature rise and a faster response are observed when a
smaller amount of water is added to a larger amount of solid, as is
expected. Figure 16 shows the temperature rise profile when 4 ml water
are added to 16 g of AFBC spent solid. A lower solid/water ratio was
used for the calcined limestone because of the limestone's extremely
94
-------
Curve 690481-A
80
70
60
o
o
.- 50
ce.
to
a.
40
30
20
10
B & W 19 Bed, 16 g/8 ml and 16 g/4 ml
B&W19Bed. 16 g/16 ml
B&W 19 Bed, 3g/20ml
Calcined Limestone 1359 at 960°C. 15 g/ 20 ml
Calcined Limestone 1359 at 960°C, 5 g/ 20 ml
_L
10
20
Time, min
30
40
Figure 15 - Heat Release Property as a Function of Solid/Water Ratio
Calcined Limestone; 15 g/20 ml
MERC Bed Mat'l; 16 g/4 ml
B&W 19 Bed Mat'l; 16 g/4 ml
RS-PER Shakedown Bed Mat'l; 16g/4ml
PER Bed Mat'l (unidentified run); 16 g/4 ml
RS-PER Carry-over 16 g/4 ml Carry-over
B&W 19 Carry-over; MERC Carry-over
MERC Fines; 16 g/4 ml
Battelle Bed 16 g/4 ml
Battelle Carry-over, 16 g/4 ml
a.
Time, min
Figure 16 - Heat Release Property of Spent Bed and Carry-over
from the Atmospheric FBC System
95
-------
violent heat release characteristics. The same data are presented in
Figure 17, which emphasizes the difference among the initial heat release
characteristics of various solids by plotting the total time on a loga-
rithmic scale.
Finally, Figure 18 compares the heat release property of spent bed
material, fly ash, and fines from the AFBC, PFBC (once-through and
regenerative), and adiabatic systems and illustrates that:
• AFBC residue has higher thermal activity than does PFBC residue
• Spent bed material has higher thermal activity than does spent
carry-over.
Figure 18 shows once again the general trend indicated in Table 34,
that the heat release property decreases in the following order: calcined
limestone ->• AFBC bed ->• AFBC ash -*• PFBC bed, regenerative •> PFBC bed,
once-through -»- PFBC ash. Since P-403 ash that has been stored for over
one year is the only material available from the Combustion Power
adiabatic process and since the CaO content of this sample may have
hydrated during the storage period due to moisture in the air, the
heat release property of the adiabatic combustor system must be investi-
gated further.
PERFORMANCE SUMMARY
Leaching property was investigated for over 30 samples of FBC
residue, including spent limestone and dolomite sorbent and carry-over,
from pilot-scale fluidized-bed combustors and regenerators of both the
atmospheric and pressurized systems. Both the continuous and the inter-
34
mittent shake tests were employed. Figure 19 summarizes the results
from the continuous shake test, showing the chemical characteristics of
leachates from spent FBC bed (sorbent) and carry-over (ash) materials
caused predominantly by the dissolution of the major species, CaO and
CaSO,. The results formed two broad bands each for calcium, SO., pH,
4 H
and specific conductance in the FBC leachate. The average leachate
from the FBC ash was better than that from the spent bed material
because of the higher SiO and lower CaO present in the fines. Leachate
96
-------
70
60
^.50
S
o>
!»
20
10
Calcined Limestone. 15 g/20 ml
MERC Bed Mat'l; 16 g/4 ml
B &W 19 Bed Mat'l; 16g/4 ml
RS-PER Shakedown Bed Mat'l; 16 g/4 ml
PER Bed Mat'l (unidentified run); 16 a/4 ml
RS-PER Fly Ash; 16 g/4 ml
B&W 19 Fly Ash; MERC Fly Ash
MERC Fines-, 16gm/4ml
I 5
Time, mln
Figure 17 - Heat Release Property of Spent Sorbent and Fly
Ash from the Atmospheric FBC System
Curve 69M87-B
80
70
60
P. 50
r\
5" V\
I
1
tf 30
• Calcined Limestone, 15g/20ml
AFBC Bed Mat' I (MERC); 16 g/4 ml
AFBCBed(Battelle).16g/4ml
AFBC Carry-over I B&W 19.MERCJ; 16 g/4 ml
PFBC Bed Mat'l, Once-through (Exxon 43.3. BedT =
945°C); 16 g/8 ml
PFBC Bed Mat'l, RegVe (Exxon 45); 16g/4 ml
PFBC Bed Mat'l, Once-through (Exxon 43.2, BedT =
845°C); 16 g/4 ml
PFBC Carry-over; 16 g/4 ml
PFBC Adiabatic Combustor Fines (CP-POU, P-403);
8 g/ 2 ml
30
Time, min
Figure 18 - Comparison of Heat Release Property of Spent Bed and Carry-
over from the FBC Process of Different Processing Variations
97
-------
from most of the FBC carry-over, except for its higher pH, was similar,
to that from a natural gypsum. Leachate from spent bed material was
generally of poorer quality than that from natural gypsum, except in the
SO . In contrast with the constant leachability over time for the natu-
ral gypsum, the FBC leachate showed a slow but steady improvement with
continuous leaching time (over 400 hours) , most likely because of a
reaction with CO in air.
Figure 20 shows results from the 72-hour intermittent shake test,
again showing two bands of data, with carry-over leachate superior to
the spent bed leachate. The average leachate from the FBC carry-over
also improved faster with time, as well as with the total volume of
leachate passing through the residue, than did the spent bed material.
Magnesium was not presented in these figures because it was generally
found to be insoluble from the FBC residue, including spent dolomite
sorbent, with only one exception (Exxon 67 third-cyclone fines, dis-
cussed earlier). Sulfide is below the detectable level in leachate
from once-through residue but should be investigated further for the
regenerative FBC system.
Total organic carbon was found to be as low (<10 ppm) as that in a
natural gypsum leachate for all FBC residues (below or near the detection
level) except for one carry-over (30 ppm for MERC bag-filter fines).
Leachability of the trace metal ions was low for all the spent FBC mate-
rial, so that all except two batches of carry-over collected after the
second cyclone (MERC bag-filter fines and Exxon 67 third-cyclone fines)
met the DWS for trace elements for which a DWS exists. The low metal con-
centration in FBC leachates is understandable since the leachates are
highly alkaline, and the solubilities of the practically insoluble metal
hydroxides and carbonates control the trace metal ion concentrations.
Leaching property is a function of the pH of the leaching medium
and the resultant leachate. Initial results from sodium acetate/acetic
acid leaching (pH ^4.5) were noticeably different from the leachate
98
-------
Curve 6931 7'.-B
2000
1000 -
2000
CD
O
CO
^1000
DWS
1
o>
O
J
0
to
10
8
6
4
2
0
DWS. approx'd
100 200 300
Total Continuous Leach Time, hr
400
Figure 19 - Leachate Characteristics as a Function of Continuous Leaching
EZZZZZBed Material, LV\\N Carry-over, Gypsum
99
-------
Curve 693170-B
Normalized Leachate Quantity, ml/g of Starting Solid
6 12 18 24 30 36 42
2000
e 1000
ea
O
DWS:
2000
CD
E
f^< S S S S S ./././„/./..,/„'./.S.S./.S.S.S.S
o>
O
-
c c
O =
O ;=
•- '
5 10 15
n = Total no. of 72-hr Intermittent Leach
360 720
Total Leach Time = (72) x (n), hr
1080
Figure 20 - Leachate Characteristics as a Function of Intermittent
Leaching. V////A Bed Material, k\NS\SN Carry-over,
Gypsum
100
-------
induced with deionized water in both major species and trace elements.
Further investigation is required and is being carried out under our
continuing contract with EPA to determine the effect of the pH of the
leaching medium on the leaching property of FBC residue.
Lacking EPA's criteria on heat release property from solid waste
disposal, we evaluated the environmental acceptability of the FBC spent
solids on the basis of their potential residual heat release activity
on a relative scale. The results to date indicate that:
• Heat release would not be a problem for the once-through
PFBC process, if the bed temperature is low enough, because
of the large amount of CaCO present,
• The heat release property of the spent sorbent from the AFBC
process would probably cause environmental, occupational, and
handling concerns because of the large quantity of CaO present.
• The spent bed material has a higher activity than do the carry-
over fines from the same system.
• Spent sorbent from the regenerative system may contain CaO from
the one-step regeneration reaction, and its heat release property
depends on the composition of the spent sorbent.
Some spent sorbents, however, contained CaO but released very little
heat when contacting water. In such cases the CaO must have been
hydrated in air during storage, dead-burned during the process, or coated
with impermeable CaSO, so that no temperature rise was detected.
We must emphasize that the preliminary results presented here are
based on the limited number of available FBC residues investigated and
that the heat release property of spent sorbent is a function of the FBC
conditions - temperature, sorbent residence time, degree of sulfation,
and degree of dead-burning.
Further processing of FBC residue by a sorbent/ash compacting
process improves the leaching and heat release properties and reduces
the potential environmental impact through leachate contamination.
101
-------
Table 35
Dug. ?612C01
PRELIMINARY INDICATIONS OF ENVIRONMENTAL IMPACT OF FBC SOLID WASTE DISPOSAL*
Sample
Process
Sorbent Type
Heat Release tb'c)
(spontaneous
temp rise).
3 9/ 20 ml
Environmental Parameters
Trace
Metal
Total
Dissolved
Solids
Total
Organic
Carbon
TbT
PH
Calcium
Sulfate
S =
(b)
Mg
Bed Material
Pressurized FBC,
once-through
Limestone
< 0.2°C
< lOppm
Bed Material
Pressurized FBC,
once-through
Dolomite
<0.2°C
< 10 pom
Bed Material
Pressurized FBC,
regenerative
Dolofr.ite/
limestone
0-3 °C
< 10 ppm
o
NJ
Bed Material
Atmospheric FBC,
once-through
Limestone
5tol5°C
< 10 ppm
x
\ X A
Carry-over MiterUI
Atmospheric FBC.
once-through
Limestone
0-3°C
<30
(e)
ppm
Carry-over Material
Pressurized FBC,
once-through
Limestone.
<0.2°C
'dolomite
(d)
< 10 ppm
(d)
Mixtures of Bed and
Carry-over Material
(unprocessed)
Pressurized FBC,
once-through
Dolomite.
<0.2°C
limestone
< 10 ppm
Processed Compacts
from Bed Carry-over
Mixtures
Pressurized FBC,
once-through
Dolomite-
< 0.2°C
limestone
< 10 ppm
Gypsum
Natural
< 0.2°C
< 10 ppm
Do not meet either the drinking water or gypsum leachate criteria.
Pass gypsum leachate criteria but not drinking water standards.
Pass both drinking water and gypsum leachate criteria.
(a) Based on data from currently available FBC residues
(b) No existing criteria
(c) Subjected to the specified procedures
(d) All except two carry-over fines pass drinking water standards
(e) All except one < 10 ppm
-------
Table 35 summarizes preliminary indications of the environmental
impact of FBC solid waste disposal. Since we lacked definite disposal
criteria, we compared leachate characteristics with DWS, and with
leachate from natural gypsum, to put leachate quality into perspective,
103
-------
7. CONVENTIONAL POWER PLANT RESIDUE ENVIRONMENTAL
IMPACT TESTS
In order to provide another point of comparison, we repeated the
tests described previously on FBC residue on samples of sludge from flue
gas desulfurization (FGD) systems in conventional boilers.
FGD RESIDUE TEST RESULTS
After settling, a typical untreated FGD sludge using lime or lime-
stone sorbent contains 30 to 70 percent solid matter. The major consti-
tuents of the solid are CaSO~'l/2 H20, CaSO,*2 H20, CaCOy, coal ash that
consists of SiO , A1.0-, Fe20 • and trace elements. The exact composi-
tion varies, depending on many factors, including the type of coal, the
type of scrubber system, and boiler and scrubber operating conditions.
Six samples of FGD sludge from pilot- and commercial-scale SO-
scrubbing systems were tested during this investigation. These included:
untreated, ponded, oxidized, and stabilized lime or limestone scrubber
sludges. Table 36 summarizes the sample source, scrubber system, further
treatment, and X-ray identification of the sludges. All sludge samples
except one (the stabilized) were wet with supernatant liquors as received.
The liquors were separated by vacuum filtration and chemically analyzed.
The dewatered sludges were then dried (^95 to 105°C), and the sludge
powders underwent the leaching tests described previously for FBC
residues.
Figure 21 shows photomicrographs of three types of FGD residues
investigated on the left (untreated, oxidized, and stabilized) and three
sources of FBC residue on the right (bed, carry-over, and fines collected
from the final particulate control device). Microscopic examination of
the dried FGD sludges by SEM shown on the left half of Figure 21 reveals
104
-------
Table 36
SUMMARY OF FGD SLUDGE SAMPLES
Sample
Process Description
X-Ray Identification
Louisville Gas and Electric
35
Company (LGE)
Fresh, untreated, unponded;
line sludge With gmall amount
of MgO added
Major: CaS03«l/2 H20
Low minor: (Fe,Mg) Al_o, or
(Mg,Fe) Si04 spinel
Columbus Southern Ohio Company
(cso)36
Untreated lime sludge; 98% fly
ash removal
Major: CaS03«l/2 H20
Duquesne Light Company (DLC)
36
Untreated lime sludge; con-
taining ^50% fly ash
Major: Si02
Minor: CaS03«l/2
Low minor: Fe_00
TVA Shawnee, Pond E
37,38
Untreated, ponded limestone
sludge bottled in pond liquor
for 2.5 yr
Major: CaSO.j'1/2
Major: CaO>3
Trace: SiO_
TVA Shawnee — Oxidized
Sludge37'39
Lime sludge followed by
forced air oxidation to gypsum
Major: CaSO^'2 H
Duquesne Light Company
36
Stabilized Sludge
"Calcilox" stabilized lime
sludge containing ^50% fly ash;
stabilized and ponded for 3 yr
Major: Si02, amorphous
phase
Minor: CaC03> CaSO3«l/2
-------
:.| u<».! i-
untreated
cyclone
carryover
"<;.,!< I lux"
Figure 21 - Comparison of FBC and FGD Residue Photomicrographs
106
RM-78038
-------
the difference in the physical characteristics among the three FGD sam-
ples. The unprocessed sludge exhibits the small platelet crystallites
35—38
of CaSO -1/2 H70 that have been reported by the FGD investigators
j £
to be responsible for the difficulties with dewatering and settling, as
well as the thixotropic property of the untreated sludge. The ponded
sludge often has mixtures of the flaky platelets and bulkier crystals that
result from partial oxidation of sulfite to sulfate. On the other hand,
the oxidized TVA sludge shows large crystals of gypsum (CaSO -2H 0). The
potential environmental hazard (due to sulfite oxygen demand) has been
reduced, and dewatering and settling difficulties are greatly improved.
In fact, oxidation to gypsum has been recommended as one of the methods
39
by which to stabilize FGD sludge. Cenospheres from coal ash are also
present in the sludge samples and may also cause settling problems in
ponding. Photographs of the stabilized sludge shows a mixture of
cenospheres and a fluffy mass that appears frequently to be clustered
and to adhere to the cenospheres. The platelet crystallites are no
longer evident. Dravo has reported that the compressive strength of the
stabilized sludge increases as a function of curing (solid setting) time
40
in a manner similar to concrete.
The right side of Figure 21 shows that the FBC residues are granular
solids from the bed and much finer particles in the carry-over material.
The absence of the cenospheres in FBC ash, including the fines collected
in the third cyclone (note the different magnification in the third-
cyclone photo because of its finer size), is in vivid contrast to the
appearance of conventional coal ash. This difference is understandable,
as FBC is operated at a much lower temperature than are the conventional
boilers.
EDAX spectra (Figure 22) show that the platelet crystallites of the
FGD solid are high in calcium and sulfur (presumably CaSO «l/2 H20),
that the cenospheres are rich in silicon, aluminum, and iron (coal ash),
and that the FBC stack carry-over consists of intimate mixtures of coal
ash (produced under FBC conditions) and sorbent fines.
107
-------
SEM and EDAX of FBC and FGD Ash
(a) FBC Ash (Exxon)
(c) FGD Sludge (TVA)
(d) FGD Sludge, Platelet
(b) FBC Ash, Area Scan (e) FGD Sludge, Cenosphere
Figure 22 -- SEM and EDAX of FBC and FGD Ash
108
-------
Leaching properties were investigated using both the continuous and
the intermittent shake methods described in the previous section. These
samples resulted from processes using different coals/sorbents; absolute
comparison of one process versus another, therefore, may not be possible,
although one would hope that the general trends indicated would be
meaningful.
Figure 23 shows leachate characteristics of the dried sludge as a
function of continuous leaching time. One can see that the leachate
from the stabilized FGD sludge is very similar to leachate from gypsum.
On the average the untreated sludge leachate has higher levels of cal-
cium, magnesium, SO,, pH, and IDS. As in the case of FBC residue, the
leachate of FGD residue also exceeded the DWS in calcium, SO,, and IDS.
Unlike the FBC leachate, however, the pH of FGD leachate fell within
the DWS range.
Figure 24 shows the specific conductance in the leachate from the
intermittent shake test. The better initial leachate quality is again
seen in the case of the stabilized sludge. The specific conductance of
the initial leachate from the untreated, ponded, and oxidized sludge
was much higher but improved with total leaching time and total leachate
volume so that the specific conductance of the leachates of the other
sludges fell within the same range as that of the stabilized sludge and
natural gypsum after two or three shake cycles. The lower specific
conductance in the Columbus-Southern Ohio (CSO) leachate after two
72-hr leach cycles was due to the low solubility of CaSO-j, which was the
predominant species in the untreated fly-ash-free CSO sludge. Both FBC
and FGD leachates exceed the TDS for DWS.
One must keep in mind that the leachate characteristics presented
in Figures 23 and 24 are of vacuum-filtered and dried sludge. The
supernatant liquors of the sludges had much higher TDS and specific ion
concentrations, as seen in Table 37, which summarizes the chemical char-
acteristics of the dried solids and the original liquor, as well as the
109
-------
Table 37
CHEMICAL CHARACTERISTICS OF FGD SLUDGE, LIQUOR, AND LEACHATE
Bwg. 2618C35
a
Substances
Al
Ag
As
B
Ba
Be
Ca
Cd
Co
Cr
Cu
Fe
Hq
Mq
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
Zn
Zr
- S1-
S03
--••-$•--
Br
F
NOjIasNl
; No2
4
TQC
PH
Sp. Cond.
Sludge. ppma
Ion dry basisl
a 2 - 20*
< i
3ST50
30 to 300
10 to 1000
Oto 15
<10
10 to 30*
<3
Oto 10
ItQlKL
Ito60
a i to 10*
0. 1 to 3
0. 1 to 10*
10 to > 1000
Oto 20
0. 01 to 5*
Oto 30
Oto 200
<33
Ho 5
0. 5to20*
<10
100 to 1000
100 to > 1000
Oto 100
20 to 200
1 to 1000
0.1 to 15*
0. 05 to 50*
3 to 65*
Otol*
10 to 50
Liquor, ppm
Untreated
<2
<0.01
0.03
ItoZO
< 1
7
7
< 1
2 .
<3
<2
<10
y/ffstyiy//
///zany//.
/// &////
20
<0.01
7. stoas j
/2000103000/X
{///////A
Ponded11
<1
3
< 1
i
< i
'////
<10
uo
<1
<30
^.°
ows.
a 05
a 05
LO
200
401
a 05
a oo?
150
a 05
2-0
a os
0.01
LO
5.0
250
250
2.4
10
----- - _
-
6i5to9. 2 ]
lai Rangeot 3untreated sludge samples (LGE.CSO.OLCI
ID) 1 ponded sludge sample ITVAI
icI 1 oxidized sludge sample ITVA)
Idi C emically treated by Oravo's "Calcilox"
0 exceed drinking water standards INIPDWR. USPH Sand WHO)
110
-------
Curve 693172-B
2000
1000
o
.DWS
I
5» 3000
E 2000
° 1000
0
— -DWS
DWS Range
3 E
o •=
, approx'd
100 200 3001 400
Total Continuous Leach Time, hr
Figure 23 - Leachate Characteristics of Dried FGD Sludge as
a Function of Continuous Leaching for:
A LGE Untreated
V CSO Untreated
o DLC Untreated
0 TVA Ponded
O TVA Oxidized
• Calcilox - Stabilized
— Natural Gypsum
- - - DWS
111
-------
Curve I 95c»-b
Normalized Leachate Quantity, m,e/g Starting Solid
6 9 12 15 18 21 24
T
T
T
27
T
30
T~
o
c.
1
•o
_0
*^
I
--OWS, approx'd
j I
I
I
I
I
1
345678
n = Total No. of 72-hr Intermittent Leach
10
180 360
Total Leach Time = 72x (n). hr
LGE Untreated D DLC Untreated
CSO Untreated O TVA Ponded
Natural Gypsum
DWS
540
o TVA Oxidized
• Calcilox-Stabilized
720
Figure 24 - Leachate Characteristics of Dried FGD Sludge as
a Function of Intermittent Leaching
leachate following the 200-hr continuous shake test. The trace ele-
ments are lowest in the leachate of stabilized sludge. Although oxida-
tion to gypsum increased the crystal size and improved the sludge setting
39
property and shear stress, it did not appear to significantly affect
trace element concentrations in the leachate. As expected, in all cases
the original liquor was of poorer quality than the leachate from the
dried sludge.
COMPARISON WITH FBC RESIDUE
Figure 25 compares the leachate quality as indicated by the specific
conductance as a function of the intermittent shake cycle for the spent
112
-------
Curve 6956 91*-A
E
o
o
"i
= 10
E 8
-------
Table 38
Owg.2618C32
COMPARISON OF LEACHATE CHARACTERISTICS
OF THE FBC AND FGD RESIDUES
Substance
Al
Ag
As
8
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hq
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Si
Sn
Sr
Ti
V
zn
Zr
SO,
SO*
Cl
F
NOolas N)
foe
DH
IDS
Specific
Conductance.
millimho/cm
Original FGD
Liquor,3 mqlt
Oto20
<0.05
<0.05
> 5
<1
<0.02
<0.04
^>500^$S
^$0 to 0.2^$$$
<0.1
<0.05
<1
<0.3
< 0.002
^ 0 to > 1000SS
^0 to 20^5^
0.1 to 7.0
Oto> 100
<1
<0.05
<0.5
•-^0.001 to 0.5^:
Oto30
<1.0
Oto40
<2
<2
<3
<2
<10to40
^1000 to 7000$;
M»0 to 6MO\^
SSilQtoSO:^
Mo£l°0^1
<30
s>SJ6tolO>^^
^ 5000 to 1400$
5.0 to 17.0
Leachate, mg/f
FBCb
Oto> 2
<0.05
<0.05
Oto> 5
< 10)
<0.02
<0.04
>$;>500^^
<0.01
<0.1
<0.05©
<1
< 0.3 CD
< 0.002
<300
< 0.05(0
<5
0 to > 100
<0.1
<0.05
<0.5
< 0.01 (7)
Oto30
<1.0
0 to > 10
<2
<1
<3
<1
<10
^1000-2000^;
< 250(JJ
<2.4QL>
<10
<30
^9 to 12 WS
^2000 to 4000;
0.5 to 10.0
FGDC
<1
<0.05
^0 to 0.1^$:
> 1
<1
<0.02
<0.04
^>500$^$;
<0.01
<0.1
<0.05
<1
<0.3
< 0.002
^.0 to 500^;
^0 to 0.1^
<1
<30
<0.1
<0.05
<0.5
§JOto0.1^
Oto5
<1.0
Oto5
<2
<1
<3
<1
<10
^1000-3000$i
i^30 to 300^
^1 to 10 ^
<10
<30
6 to 9
2000 to 3000^
2.0 to 3.0
FGD(stabilized)d
<1
<0.01
0.05
1
<1
<0. 1
<0. 04
^^^>500^^^
<0.01
<0. 1
<0.05
<1
<0.03
< 0.002
16
<0.05
<1
6
<0.1
<0.05
<0. 5
0.006
3
<1.0
<2
<1
<3
<1
h <10
$N$$$^ 1400 $-$$$S$^
Z
^^s^5 >sss^m
<10
[ <30
8.0
$$$^$§ 2000 to 2500$^$^
2.35
Drinking Water6
Standards,
mqlt
0.05
0,05
1.0
200
0.01
0.05
1.0
0.3
0.002
150
0.05
2.0
0.05
0,01
1.0
5.0
250
Z5U
2.4
10
6. 5to 9.2
500
(a) Liquor obtained from filtration of 5 FGD sludge samples
(untreated, ponded, oxidized)
(b) The figures in the FBC column include the full range of data reported previously in this
section and in Ref. 17 for 30 unprocessed FBC residues, including both spent bed
material and carry-over
(cj Leachate from dried FGD sludge - untreated, ponded, oxidized (range of 5 samples I
(d) Leachate from dried FGD sludge - stabilized with Dravo's "Calcilox" (1 samplel
(e) NIPDWR. USPHS and WHO Drinking Water Standards
(J) Drinking Water Standards met by leachate from all 30 FBC residues tested except for
2 batches carry-over fines
gg Exceed Drinking Water Standards
114
-------
stabilization of the FGD sludge improve the leachate quality. We can
say that, on the average, the IDS in the leachate of FBC and FGD resi-
dues are within a range similar to that of natural gypsum. The liquor
of FGD sludge, however, contains much higher IDS, major species, and
trace elements, as shown in Table 38.
Since the leaching tests were conducted on the dried sludge fol-
lowing vacuum filtration, the actual initial leachate quality from the
FGD sludge would be worse than that shown here because of the presence
of sludge liquor even in the dewatered sludge (V30 percent solid).
Table 38 shows that the FGD liquor contains much higher dissolved major
and trace species than the leachate of the dried sludge solid, as
expected. The trace elements in FGD leachate are further reduced by
stabilization.
Table 38 shows that both the FBC and the stabilized FGD residues
have leachates that generally pass the DWS for trace metal elements.
Leachate from both FBC and FGD residues have high calcium, SO,, and TDS,
as does leachate from natural gypsum. The leachate of FBC residues has
high pH. In addition to the major species, some minor and trace element
concentrations in the leachate from the dried, nonstabilized FGD sludge
tested here exceeded the DWS, notably magnesium, chlorine, fluorine,
manganese, arsenic, and selenium, although further testing of additional
FGD samples would be necessary to confirm the point.
The comparison between the FBC and FGD residues presented in this
section has been limited to their chemical and leaching properties. The
physical properties of the FGD sludge are reported in the literature.
115
-------
8. SPENT SORBENT PROCESSING
REVIEW OF PROCESSING OPTIONS
The previous sections have dealt with the potential environmental
impact of FBC residues, emphasizing the response of unprocessed residues
to leaching with aqueous solutions. We have conducted other studies to
identify processes that could minimize such impact if required for spe-
cific applications or could result in further resource utilization.
These studies, described in this report, represent an extension of
prior work at Westinghouse, under contract to EPA, which is also
reviewed.
The new material presented extends the investigation into producing
solid compacts as an environmentally acceptable disposal method for FBC
residues. The major portion is concerned with obtaining further data on
the long-term stability of the compacts, and another section deals with
a method for direct disposal of FBC residues for utilization as fine
aggregate in concrete. Finally, we present an analysis of selected
leachate data with an interpretation applicable to a full-scale residue
disposal operation.
Perspective on Residue Processing
For the most part FBC residues will be produced at elevated temper-
ature (300-900°C). Even in the case of direct disposal, some on-site
processing equipment will be required to permit safe handling of these
residues. This equipment will include conveyors, coolers, and probably
enclosed storage silos. For our purposes processing means those opera-
tions beyond simple cooling and transport and will include, for example,
blending, grinding, and sieving. Other operations may also be involved,
depending on the particular process option under discussion.
116
-------
The specific questions that arise concerning the direct placement
of FBC residues in the environment include the amount and quality of
leachate produced when the residue contacts precipitation or ground-
water, the amount of heat released, and the amount of fugitive dust
emitted. The environmental question, therefore, can be formulated as
what method of disposal will offer the least environmental degradation
for the longest period of time and for the lowest cost?
Processing FBC residues is under investigation because it may pro-
vide alternatives to direct disposal that:
• Could minimize negative environmental impact, or
• Might result in recovering a useable resource.
Numerous processing options have been identified,^«" and these may be
grouped under two main headings:
• Processing for disposal
• Processing for utilization.
Although many of these are under investigation, the work at Westinghouse
has focused on low-temperature (below 100°C), simple, physical pro-
cesses. We have deferred work on high-temperature processes because we
anticipate that the additional energy requirement of such processes
would make them less attractive economically.
Another consideration in the Westinghouse development program was
the planning horizon. Some processes might be developed in 1 to
3 years, while others might require 10 to 20.
Because of the large tonnage of residues that can be produced in
large-scale FBC systems, processing for utilization offered the possi-
bility of minimizing environmental problems and enhancing system econom-
ics at the same time. On the other hand, to demonstrate the reliability
of physical and mechanical properties, conformity to (new) standards and
specifications, and market acceptance would have required a commitment
in time far beyond the scope of the previous contract.
117
-------
We have, therefore, given greater attention to processing for dis-
posal. The technical criteria applicable to this group of processes are
expected to be fewer and less restrictive than for the other group,
which results in a product for the commercial market.
Specific alternatives to direct disposal include landfill, road
base material, treatment of acidic wastes as acid mine drainage, agri-
culture, and coarse aggregate. In the case of direct disposal, as in
unlined ponds, the residues would remain as a potential source of leach-
ate indefinitely into the future and would represent a continuing poten-
tial for degradation of groundwater by permeation and of surface waters
by overflow. Membrane-lined ponds eliminate the former hazard but only
for the life of the membrane - some 20 years. Then, action would be
required to control potential leachates to specified standards. Clay-
lined ponds might appear to be longer lived, but such liners are subject
to cracking by earth movements and deterioration from chemical interac-
tions with soil. Further, suitable clays are not usually located near
the site of the pond.
In contrast, use of the alkaline FBC residues in treating acid mine
drainage would simply transfer the solids disposal problem to another
location even if this treatment were successful.
Investigation of agricultural uses is being investigated by the
Department of Agriculture and was considered outside the scope of the
existing contract.
Review of Previous Work
Westinghouse Feasibility Tests
Work reported in 19785 had been directed toward establishing the
feasibility of disposing of FBC residues as landfill, coarse aggregate,
or road base material. Such use implies minimal on-site processing
coupled with finality of disposal.
118
-------
Westinghouse demonstrated that stable solid compacts could be pro-
duced from blends of FBC spent bed material and conventional coal ash at
ambient temperature and pressure. By converting the granular residues
into larger masses, one would reduce the surface-to-volume ratio,
thereby reducing potential leachate production when the residues were
placed in the environment. If a given mass of residue consisting of N
particles of particle density po is processed to a single mass of den-
sity p, the total external surface is reduced by the ratio Lopo/Lp,
where L is a characteristic dimension. Since spheres and cubes have the
minimum surface-to-volurae ratio for curved and rectilinear shapes
respectively, L here refers to the length of either the diameter of a
sphere or the edge of a cube. Thus, if the density is essentially
unchanged by the size increase process, a 10-fold increase in particle
size is accompanied by a 10-fold decrease in surface area.
For porous particles, the internal surface of the pores can be many
times the external surface area. Two situations then arise. In one,
the pores permit flow into and possibly through the particles. This is
considered to be unusual. In the other and more likely case leaching
involves diffusion of solute from within the particle through stagnant
films and thence out through the pores to the bulk leaching medium sur-
rounding the particles. It is of interest, therefore, to demonstrate
whether internal pore surface in FBC residues contributes significantly
to leachate constituents, whether pore surface is reduced by the process
of forming the residues into larger masses, and whether the larger
masses are environmentally stable to:
• Cycles of hydration and drying
• Cycles of freezing and thawing
• Attack by chemical agents, such as sulfates and chlorides.
The third action is relevant to ocean disposal as well as to disposal on
land in sites where contact with natural brines is possible.
119
-------
Stability is an important property, for if the compacts should
break down into smaller fragments on aging, freezing and thawing,
hydrating and drying, crushing under load, or reacting with other sub-
stances, the reason for making them could be negated.
Feasibility tests were conducted in the laboratory using PFBC resi-
dues from the Exxon pilot plant to cast 5-cm cubes from a mix of ground
bed material, carry-over, and water. The result was stable, solid com-
pacts that had compressive strengths up to about 15 MPa (2175 psi).
Higher strengths (90 MPA or 13,000 psi) were obtained by isostatically
pressing Exxon spent sorbent ground to -125 utn and then water curing it
for 14 days. We concluded that the residues displayed sufficient poz-
zolanic activity to warrant further investigation of the processing
options identified above.
In parallel with this early laboratory work, we reviewed relevant
literature for information applicable to our studies. We were specifi-
cally interested in data on composition limits, processing methods, and
performance.
Literature Study
One of the areas of relevant technology we reviewed was the utili-
zation of other sulfate wastes. Smith, et al.,^1 in work sponsored by
the Federal Highway Administration, had tested sulfate wastes from FGD
scrubbers, hydrofluoric acid manufacture, titanium dioxide extraction,
steel pickling liquor neutralization, and acid mine drainage neutraliza-
tion. The purpose of their work was to determine whether treatment of
these wastes would result in materials useful for road construction.
Included in the tests were determinations of the effects of water con-
tent, lime type, sulfite vs. sulfate, lime/sulfate ratio, fly ash con-
tent, sulfate content, and Portland cement addition on the mechanical
properties of the processed wastes.
Useful indications were that the water content could be perhaps as
low as 18 percent, that calcitic lime is better than dolomitic lime,
120
-------
that sulfite and sulfate results were roughly equivalent, that an opti-
mum lime-sulfate ratio exists at about 1:1, and that a gypsum content of
up to 10 to 20 percent favored higher compressive strengths. X-ray
examinations were attempted but were inconclusive. One sulfite sludge
mix developed a compressive strength of about 11 MPa (1600 psi).
Strengths for all mixes increased on aging to 91 days. Permeability was
reduced for some of the mixtures to as low as 1 x 10~° cm/s. Freeze-
thaw resistance was poor - few cycles, low strength.
Since these studies basically dealt with lime/silica reactions, we
concluded that we could expect some similarity in our FBC work. For
example, water is an essential ingredient of concrete, but the evidence
is that it should be minimized to achieve maximum strength and environ-
mental stability. The figure of 18 percent noted above is less than the
theoretical for hydration of Portland cement (24 wt %). The lowest
value of water used in Westinghouse screening tests was 30 percent. The
Smith work suggested that future Westinghouse tests could aim for a
lower water content.
The gypsum content Smith found was much higher than that of Port-
land cement (5.1 wt % CaS04). This supported earlier Westinghouse
results that CaSO^ content greater than 3 percent was not detrimental to
compressive strength. Ideally, for a once-through process FBC residue
would be a mixture of coal ash and lime that had been sulfated 100 per-
cent. For a 3 percent sulfur, 10 percent ash coal, the CaSO^ content at
90 percent sulfur capture would be 51 percent of the solid residues.
This suggests that fixed FBC residues might have lower compressive
strengths when lime is minimized. These strengths might be acceptable
if the fixed residues are merely discarded to the environment, but
higher values might be sought through adjustment of the residual CaO
content if the fixed residues are to be utilized.
In other work, done by Dunstan^ through the Bureau of Reclama-
tion, substitution of 15 to 25 wt % of Portland cement by lignite and
subbituminous fly ashes yielded concretes with acceptable freeze-thaw
121
-------
resistance and compressive strengths but with inadequate sulfate resis-
tance. The effect of iron oxide content appeared to merit investigation.
With respect to test methods and commercial specifications, P. W.
Brown of the National Bureau of Standards'^ found that existing tests
for materials may represent obstacles to the use of FBC residues in the
construction industry in that the tests may be inadequate, biased, or
not representative of field conditions. Existing specifications may
also favor suppliers of conventional materials. These conclusions
helped shift the focus of the Westinghouse residue processing studies
from utilization to disposal.
Performance Criteria
Whatever the form into which FBC residues are processed, there will
be environmental constraints that must be met. This is true whether the
processed residues are deposited in the environment or are utilized in
some way in the environment. The effort to define adequate environ-
mental protection in legal and practicable terras is a continuing one,
and so the statements of performance criteria given here are subject to
change.
1. The processed residues should not release substances that can
migrate to and enter natural water supplies In concentrations
in excess of accepted water quality standards applicable to the
water use Involved (drinking water, agriculture, industry) at
the point of use.
2. The processed residues should conform to all standards or regu-
lations on fugitive dust, odor, or other considerations.
3. The processed residues should have sufficient physical and
chemical stability so that environmental actions such as wet-
ting, drying, freezing, thawing, or contact with other natural
substances in the environment will not impair their performance
under Item 1 above.
122
-------
Other performance criteria will be added to this list If the FBC resi-
dues are processed for utilization.
PRESSED MATERIAL
General
We have hypothesized that the binding forces in FBC compacts are
similar to those In normal Portland cement concrete. Others have estab-
lished that hydratlon of di- and tricalcium silicates is mainly respon-
sible for the corapressive strength developed In cement and concrete.
Corapressive strength also appears to correlate with other properties
such as splitting tensile strength and therefore represents an easily
measured parameter for screening the effects of several variables.
Hydration of other compounds, such as tricalcium alumlnate ^A) and
tetracalcium aluminum ferrite (C4AF), where C * CaO, A » A1203, and F
= Fe203, has been shown to result In compressive strengths much lower
than those of the hydrated calcium silicates, as discussed later in this
report.^
Objective
The feasibility of preparing solid compacts from FBC residues was
demonstrated in the previous work. The objective of the current work is
to obtain evidence that these compacts do have long-term environmental
stability. Several measures might have been investigated, but budget
constraints dictated that only one be selected at this time. We decided
to concentrate on freeze-thaw resistance because the compacts would more
certainly be exposed to temperature changes in an actual plant than to
sulfates and because freeze-thaw cycling was considered a severer test
than simple wetting and drying.
123
-------
Test Plan
Variables potentially affecting the stability of the compacts and
possibly also the composition of leachate producible from them
included:
• Type of sorbent - limestone, dolomite
• Source of sorbent - crystal size, porosity
• Chemical composition of sorbent
• Particle size distribution of sorbent
• Sorbent/fuel treat ratio in FBC process
• Sulfur capture achieved in FBC process
• Residence time of sorbent in FBC process
• Ash content of fuel
• Chemical composition of fuel ash
• Water content of compacts
• Curing time and curing temperature for the compacts
• Composition of the compacts - spent sorbent, fuel ash,
other materials.
Since the main FBC process is still under development, no one sorbent
has emerged as the one to be used on a commercial scale. The test plan,
therefore, was designed to cover a range of compositions in an attempt
to bracket future design values. One constraint adopted was process
simplicity. Rather than have a separate disposal process for spent sor-
bent and fuel ash, we decided to aim for a single process through which
all sorbent/fuel solid residues could be disposed of. This meant fixing
the ratio of feed rates to the disposal process for the two residues at
production proportions. Further discussion is given in a later section
of this report.
Solid state reactions are involved, and therefore the effect of
particle size distribution was included in the test plan. On the basis
of previous work, we selected two size ranges: -125 +44 ym and -44 um.
The latter reflects commercial practice in the cement industry. Because
124
-------
grinding energy increases rapidly as particle size desired decreases,
the former size range represents a compromise in reactivity as grinding
energy.
In most of the previous FBC residue processing work,-*- wet-
casting at ambient pressure and temperature were used. This approach is
similar to the casting of cement and results In the phenomena of setting
and hardening as in cement. We believe that similar phenomena would
occur naturally in an FBC residue disposal site operated to promote this
type of fixation processing. Since freeze-thaw cycling was expected to
be a severe test, we decided to select a method of preparing specimens
that would maximize the probability of their surviving the tests. Iso-
static pressing was chosen over direct casting because it led to speci-
mens with a much higher compressive strength. If these specimens
failed, then direct casting would also be questionable as a method for
making specimens with long-term stability; if they survived, then other
compaction methods more suited to commercial-scale operation, such as
briquetting or tabletting, would be investigated. The use of pressed
compacts might also give information on the leaching properties of con-
struction blocks or other similar items fabricated from FBC residues.
Test Materials
Spent bed material was obtained from the Exxon, the PER, and the
B&W pilot plants. Chemical analyses are reported in Table 39.
The PER stone was a high-silica, high-alumina lime in which the
calcium was about 50 mole % sulfated. The B&W stone was a low-silica,
low-iron, low-alumina lime that was about 20 mole % sulfated. The Exxon
stone was a dolomitlc lime with about 50 mole % of the calcium in the
bed material sulfated. Since all of these specimens were exposed to
elevated temperatures and since the loss on ignition for the pilot plant
materials was no greater than 4.23 percent, we judged the residual car-
bonate content (as CaC03> to be negligible. In more definitive tests
125
-------
Table 39
CHEMICAL ANALYSES OF FBC PILOT PLANT RESIDUES
Source
Material
Duquesne
Fly Asha
PER
Bed
Exxon
Run 43.3
Bed
Run 43.3
Carry-over
B&W
Run 19
Bed
Composition, wt %
CaO
MgO
Si02
A12°3
Fe2°3
S03
LOI
Other
Composition
CaO
MgO
Si02
A1203
Fe203
803
0.4
1.0
44.9
19.1
9.6
—
14.0
11.0
100.0
, moles/100
0.0071
0.0248
0.7472
0.1873
0.0601
—
28.84
2.13
27.20
10.26
5.39
19.90
1.04
5.27
100.00
g
0.5142
0.0528
0.4526
0.1006
0.0338
0.2486
38.64
13.25
8.60
5.70
3.95
29.00
0.32
0.44
100.00
0.6890
0.3287
0.1431
0.0559
0.0247
0.3635
15.62
7.70
29.40
11.20
8.80
16.38
4.23
6.67
100.00
0.2785
0.1910
0.4893
0.1098
0.0501
0.2046
67.01
1.25
3.90
5.55
2.00
19.09
0.51
0.69
100.00
1.1949
0.0310
0.0649
0.0544
0.0125
0.2384
Molar ratios
803 /CaO
CaO/Si02
Net CaO/
Si02b
—
0.0095
0.0435
1.1361
0.5868
0.5276
4.8148
2.2746
0.7346
0.5692
0.1510
0.1995
18.4114
17.738
aFly ash from a conventional boiler plant at the Elrama, PA station of
Duquesne Light Company.
CaO - CaO less 803.
126
-------
the amount of residual carbon, rather than carbonate, would be deter-
mined specifically because carbon laydown on particle surfaces might
affect reactivity of the bed material more than the presence of inert
carbonates.
Preparation of Test Materials
The FBC residues were ground to -125 and -44 pm (-120 and -325
mesh) by outside contractors. The Duquesne fly ash was not ground
because typically it is 95 percent through 149 um (100 mesh). Upon
return to Westinghouse, each size fraction was blended for uniformity.
Quantities available were as follows (kilograms):
Particle Size Range, urn
Sorbent Source
PER
B&W
Exxon
-1255 +44 |
88
56
80
-44
52
36
47
A check of the particle size distribution of the -44 urn (-325 mesh)
fraction from the B&W stone showed only 24 percent through 44 pm
(325 mesh). Since particle size was identified as an important vari-
able, we judged it necessary to explain this discrepancy. One possibil-
ity was water absorption, which would lead to swelling of the particles
and possibly also agglomeration.
Therraogravimetrlc analysis showed an 8.65 percent drop in weight
when the stone was heated in nitrogen to 1000°C. Slight slope changes
occurred at 390°C and at 660°C. About half of the total weight loss was
attributed to water; the balance may be C02« Data on weight loss at 700
and 1000°C for four samples of ground sorbent are in Table 40. Heating
the sorbent to 700"C should drive off the surface moisture and water of
hydration of gypsum and Ca(OH)2. We concluded that a significant amount
of water was present and, therefore, ground B&W spent bed material
tended to pick up water on exposure to ambient air.
127
-------
Table 40
WEIGHT LOSS ON HEATING GROUND B&W SPENT BED MATERIAL
Size Range, ym
-125 + 44
-44
Location in Drum
Bottom
Middle
Bottom
Middle
% Loss @ 700°C
4.50
2.14
7.38
7.28
% Loss @ 1000°C
6.13
2.57
8.65
9.85
Next, the particle size distribution of the -44 ym (-325 mesh)
fraction of the ground B&W stone was checked by means of an Alpine Model
200 Air-Jet Sieve. The procedure is given in Appendix A. One sample
was from the material as received from the outside grinders and the
other from a -63 ym (-230 mesh) fraction sieved by Westinghouse from
this material. The results are in Table 41. Another sample of the
as-received -44 ym material was dried at 600°C for 30 minutes and then
examined by Coulter Counter. The medium used was Isoton and 30 percent
glycerine. These results are in Table 42 and are plotted along with the
data of Table 41 in Figure 26.
Since the drying process was accompanied by a shift in particle
size distribution to smaller sizes, we concluded that the finely ground
materials were at least hygroscopic, most likely because of the free CaO
present. The effect is to Increase particle sizes. The question
remained, however, of whether the material had actually been ground to
100 percent through -44 urn (-325 mesh). Rather than complicate the test
procedure with a drying step, we decided to screen the nominal -44 ym
(-325 mesh) fraction to obtain an actual -63 ym (-230 mesh) fraction for
use in the freeze-thaw tests. The quantity prepared was 3000 g; its
size distribution is given in Table 41.
128
-------
Table 41
PARTICLE SIZE ANALYSIS OF AS-RECEIVED -44 ym (-325 mesh) B&W
RUN 19 SPENT BED MATERIAL USING AN ALPINE AIR-JET SIEVE
Sieve Size, ym
Weight % Retained on Indicated Sieve
As Received
-63 um (-230 mesh) Fraction
150 0.0 0.0
105 4.5 0.0
74 30.0 0.0
63 46.0 0.0
45 59.5 1.0
32 74.0 6.0
20 80.2 56.4
10 98.8 94.3
5 100.0 99.4
Preparation of Test Specimens
Previous work on direct casting of 5-cm cubes from blends of Exxon
residues had indicated that 15 weight % carry-over on total solids was
near the optimum amount, as reflected in corapresslve strength. Initial
attempts to extend this work by use of isostatic pressing of dry mixes
failed. Details of this procedure are given in Appendix B. The dry mix
is placed in a mold, subjected to hydrostatic pressure of 138 MPa
(20,000 psig) for 1 minute, removed from the mold, cured in water for
24 hours, cut into desired lengths, and then placed under water for com-
pletion of curing.
Since carry-over was not available for all three pilot plant resi-
dues, these initial tests were conducted with blends of spent bed mate-
rial and Duquesne fly ash. All specimens, however, fell apart when
cured in water. One specimen prepared with Wyoming lignite fly ash also
crumbled in water.
129
-------
Table 42
PARTICLE SIZE ANALYSIS OF AS-RECEIVED -44 pm B&W RUN 19
SPENT BED MATERIAL USING A COULTER COUNTER
Diameter, pm
2.52
3.17
4.00
5.04
6.35
8.00
10.08
12.7
16.0
20.2
25.4
32.0
40.3
50.8
64.0
80.6
101.6
128
181
203
Weight % greater than
Indicated Diameter
100.0
97.1
94.9
92.1
88.7
84.0
79.0
73.8
67.8
61.2
57.1
52.3
46.8
39.5
28.4
22.3
11.9
5.4
2.5
0.0
Note: Sample dried at 600°C for 30 minutes.
130
-------
400
200
100
60
20
oAs Received -44 urn Product: Air Sieve
a- As Received-63pm FractionrAir Sieve
°As Received-44um Product. Dried
at 600°C for 30 Minutes: Coulter
Counter
0.1
1 10 20 40 60 80 90 99
Weight* Less Than Indicated Size
Figure 26 - Particle Size Distribution of Ground B&W
Spent Bed Material from Run 19
We interpreted this as indicating the need for preslaking the resi-
dues. One method was to slake the bed materials and carry-over speci-
mens separately, dry them at 100°C, and set them aside. Since this step
represented an undesirable complication of the contemplated disposal
process, we decided to test the materials separately, beginning with the
bed material. Because of constraints of time, funding, and material
available, a sequential test plan was evolved, the first stage of which
was freeze-thaw tests on isostatically pressed ground spent bed mate-
rial. Subsequent tests were to be devised on the basis of the results
from this stage.
Because of the number of test specimens required, we developed a
larger mold for use in isostatic pressing. Figure 27 is a diagram show-
ing details of the mold assembly used to prepare the specimens, while
Figure 28 is a photograph of the mold. The cylindrical specimens
131
-------
End Cap with d/4") *
Drilled and Tapped
Dwg. 6419M1
0.158
* J_
3.81
(1.50")
-*-
2.54
(1.00")
T
J
3.81
i
^-^
Jb^F*
i~~v* >f»
-4-
-$-
o.
"V"
-4-
A
T
1*- 3.81 --
i
^^
4 Rows, 90° Apart, 2.54
of 0.125 Dia. Holes ~^—
^^-on 1" Spacing ]
Reinforced Smooth Bore
Flexible PVC — __
Tubing 3. 49 O.D. and
36.83 2. 54 Nominal ID
(14.50")
^ —
\
_ iinn
Sample to Be Pressed
3.810. D.
Steel lube *
1
tna uap ___^
<
m
1
1
X
^V
V^
s>
X
^
\
^
^
^
JS
\:
x
k\
X
X
X
X
v
^
^
1
T
IT
L 5*
"S/S/Sf/YSSM
!. ^
fe.^
j'1-)'1'-/
.'.'V-?-1'
1 '' * '\ '•/
j S -> I
''. •''*>','>'
•'vv' -'-.'.*'
>• 'V; * ;;
' '' r ' . '
••>< .V- •-'''.'
>v' f-Jy.
;'..^'.\jv
•-;?.;f'
•".i-"/"-/-'
** . * ' / •'. *
"__,—,!« '.•
,; ^. . "*'. •
•\ * * , *;.
• if ' .'*.(. ' • .
• : '. ', - '-. •
"J1 '*''•'.
- / JT'.C,'
/.I .'*',-
v" '.•;'•!'•'!
1 * i *» .
•', '.• '•"-.''• •
',/•.' - V.'
(^555^5555?
1 ^
BQBI
1
1
\
\N
->
\^
X
N
^
\
V
X
X
X
X
^
X
\
\
X
\
X
\
^
Mil
\ \
UULSVI UUOIV
rJ
0.3175
^0-Ring
X
i
J_ (1M,
\ \
0.158 0.3175
Rubber Gasket 0- Ring
Note: Dimensions in Centimeters except as noted.
Figure 27 - Mold Assembly for Preparation of Specimens by
Isostatic Pressing (Drawing not to Scale)
132
-------
Figure 28 - Mold Assembly
B&W
Exxon
PER
Figure 29 - Specimens Produced by Isostatic Pressing
133
RM-73234
-------
produced were about 1.8 cm In diameter and 25 to 50 mm long. Figure 29
shows typical specimens. From these were cut the actual test pieces,
which were about 3.7 cm long (1.5 in). All of the test specimens pres-
sed from ground bed material as well as the control specimens were suc-
cessfully cured in water for 14 days.
Test Apparatus
Figures 30 and 31 show details of the apparatus used to conduct the
freeze-thaw tests. The chamber was contained in a cubical cabinet about
60 cm on an edge. Electric heating and dry-ice cooling were used to
change the temperature of the circulating air.
Six thermocouples were used to check on the uniformity of the air
temperature within the chamber, and two were mounted on a Portland
cement control specimen. Figure 32 shows the Portland cement specimen
with an embedded thermocouple used as the temperature control. The
external thermocouple was used to show whether there was a significant
difference in temperature between the interior and the exterior of the
control specimen. The mode of control was on-off, actuated by the
embedded thermocouple. In practice there was no more than a 2°C (4°F)
temperature difference between the temperature shown by the embedded
thermocouple and the external one.
Test Procedure - General - The test procedure was patterned after ASTM
45
C666, Resistance of Concrete to Rapid Freezing and Thawing. Specimens
were frozen and thawed in air, which is Procedure B modified in the
thawing cycle by using air instead of water immersion to achieve a
faster response and, hence, shorter cycle times.
ASTM C-666 involves observing any change in the fundamental trans-
verse frequency of vibration of rectangular prisms on exposure to
freeze-thaw conditions. This procedure has the advantage of being a
nondestructive test, so one specimen can be used for the entire period
of the test. Its disadvantage is that the final results are in terms of
a relative dynamic modulus of elasticity, which does not permit predic-
tion of service life under field conditions. Because the specimens to
134
-------
Dwg. 6428A85
o
- Indicates Thermocouple
Placement
Dry Ice Chamber
Test Specimens
1.8cm Dx 3.7 cmL
Temperature Control
Thermocouple in Portland
Cement Control Specimen
Thermometer
Figure 30 - Freeze/Thaw Chamber
135
-------
Dwq. 6428A86
Control Box
Dry Ice Chamber and
Circulating Fan
G
Heating Unit and
Circulating Fan
^Insulated Chest
Support Rack for
Test Cylinders
Figure 31 - Schematic of Freeze/Thaw Chamber
Dwg. 6428A84
^- Type I Portland
/ Cement Specimen
Copper Constantan T/C —
Implanted during
Casting
1.9
i
1cm
V^^-^__ —^
i
H
X
1
.81cm
- V
Copper Constantan T/C
Taped to Side of
Specimen
cm
Figure 32 - Sketch of Type I Portland Cement Control Specimen
with Embedded Thermocouple for Temperature Control
136
-------
be used were small, we decided to use multiple specimens coupled with
the direct measurement of compressive strength after various numbers of
freeze-thaw cycles. We carried out this procedure by having multiple
specimens in the freeze-thaw chamber and removing them in sets after
predetermined times of exposure.
Test Matrix
Initially, we planned to obtain data on the effect of spent sorbent
source, particle size range, fly ash content, lime content (as reflected
in the composition of the spent bed material), and forming method (cast-
ing, isostatic pressing, sintering, melting). We would use replicates
to estimate experimental random error. The test matrix included three
spent bed materials, two mix compositions, two particle size ranges,
four cycle times, and three replicates. The number of test samples was
therefore 3x2x2x4x3, or 144. As noted above, mixes with fly ash
were unsuccessful, and time constraints dictated further reduction of
the matrix, so only one particle size range was tested, reducing the
number of primary test samples of 36.
Table 43 shows the actual test matrix with identifying code numbers
for the specimens having the following meaning, using BS0014-3A as an
illustration. The first letter denotes the source of the stone:
B Babcock and Wilcox
E Exxon
P Pope, Evans and Robbins.
The second letter denotes the particle size. Two ranges were contem-
plated: S for small, -63 um (-230 U.S. sieve), L for large, -125
+ 63 urn (-120 +230 U.S. sieve). We tested only the small size. The
next two digits denote the weight % fly ash in the mix (zero in all
cases tested), and the next two denote the number of days the specimens
were cured in water prior to freeze-thaw testing. The next letter
denotes N^, the number of freeze-thaw cycles to which the specimen was
137
-------
Table 43
TEST MATRIX FOR FREEZE-THAW TESTS ON FBC RESIDUES
Source of Spent Bed Material
Number of Cycles
N!
N2
N3
N4
Exxon
ES0014A-1/3
ES0014B-1/3
ES0014C-1/3
ES0014D-1/3
B&W
BS0014A-1/3
BS0014B-1/3
BS0014C-1/3
BS0014D-1/3
PER
PS0014A-1/3
PS0014B-1/3
PS0014C-1/3
PS0014D-1/3
exposed, and the final digit denotes the number of the replicate. Since
the freeze-thaw resistance could not be predicted, the values for A
through D were to be selected according to the observed results. We
expected that a curve of compressive strength versus time (number of
cycles) would be obtained, and it was desirable to distribute the data
points selectively along the curve.
Various control specimens were used. To isolate the effect of
freeze-thaw cycling, specimens of all three test materials were carried
along during the test but outside of the freeze-thaw chamber. Thus, if
the outside specimens increased in compressive strength with time, this
result would be interpreted as being similar to the normal strength
development curve for Portland cement. If freeze-thaw exposure had no
effect, then the final values of strength for inside samples would be
about the same as those for the outside samples. The initial points on
the strength curves would be identical for inside and outside specimens,
so only 3 materials x 3 times x 3 replicates, or 27 of these control
specimens were needed.
138
-------
In addition, isostatlcally pressed samples of Type I Portland
cement were used as reference Indicators. Again, both inside the out-
side specimens were used with three points on each strength curve;
9 inside specimens and 6 outside specimens were provided.
As a further exploration of the effect of freeze-thaw cycling on
normal materials, specimens were prepared from Portland cement/river
sand mixes, using the normal weight ratio of 2.75 parts of sand to
1 part cement. For this effort the number of specimens was the same as
for the spent bed materials, 12 inside and 9 outside. The total number
of specimens then was 63 primary and control spent bed specimens,
15 primary and control Portland cement specimens, and 21 primary and
control cement/sand specimens, or 99 in all.
Test Procedure - Specific - The large number of specimens meant that
specimen preparation was spread over a period of time. Enough mix was
prepared each time for all replicates required for each planned exposure
time. For example, specimen ES0014A, which was to be exposed for N^
cycles was pressed and cured for 24 hours under water. From it were cut
three smaller cylinders to serve as the three replicates for that point
on the strength-time curve for Exxon material. These cylinders were
cured for thirteen more days in water and then placed in the freeze-thaw
chamber. After N^ cycles of freezing and thawing, they were removed and
tested for axial compressive strength. The first set of measurements
was made at N^ * 0 cycles, i.e., with no freeze-thaw exposure.
The specimens were observed daily for signs of deterioration
(chalking, spalling, cracking, etc.). Because we expected a rapid
increase in compressive strength within three days, the second test time
(N2> was set at about 24 hours (12 cycles). The third (N3) was initi-
ally set at twice this number, but since no visible deterioration had
occurred it was delayed to 36 cycles. The last test was set at a con-
venient time after 300 cycles except for one group of B&W samples as
noted below. Since the cycles of testing were started at different
139
-------
times, the overall number of cycles of freeze-thaw was 344. Within this
time interval the total number of cycles for each material was a conve-
nient arbitrary number greater than 300.
Test Results
Tables 44 through 53 contain the freeze-thaw data on isostatically
pressed specimens of spent sorbents from the three FBC pilot plants.
The data are plotted in Figures 33 to 37. All the specimens but one
survived over 300 cycles of freeze-thaw exposure between the limits -18
and +4°C. One of the B&W samples showed surface spalling at 172 cycles
so we determined its compressive strength at that point. Since it was
low (22.5 MPa) we checked the other two replicates, but these showed
increased strength (79.6 and 73.0 MPa).
In all cases compressive strength increased with exposure to about
50 cycles, after which further increases in strength were smaller.
Exposure to freeze-thaw conditions resulted in somewhat higher compres-
sive strengths for Portland cement/river sand, Exxon, and PER specimens
and somewhat lower strengths for plain Portland cement and the B&W spec-
imens in comparison to the outside controls which were not exposed to
freeze-thaw cycling.
The Portland cement specimen achieved the greatest strengths (90 to
100 MPa, average of three values). The Portland cement/river sand mix-
tures were in the range of 73 to 83 MPa. Specimens prepared from pilot
plant sorbents were in the range of 60 to 83 MPa. These values are
above the strength of normal concrete (about 42 MPa).
Since there was no evidence of deterioration in the specimens when
the testing was terminated after more than 300 cycles, the technique of
compressing the ground spent sorbent and curing it in water appears to
offer a way to minimize the environmental impact of the sorbent.
140
-------
Table 44
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS OF
FBC SPENT BED MATERIAL FROM B&W PILOT PLANT: TEST SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compressive Strength
MPa
Psi
Number of
Cycles
BS0014A 1 9/15/77 10/12/77
f\ tt t*
3
Average after 0 cycles
BS0014B 1 9/15/77 10/13/77
e\ tt K
3
Average after 12 cycles
BS0014C 1 9/15/77 10/25/77
2
3
Average after 36 cycles
BS0014D 1 9/15/77 11/15/77
2 " 11/8/77
O •• ti
Average after 172 cycles
aSurface spalling observed on Sample 1
59.2
45.5
19.9
41.5
21.1
46.1
53.2
40.1
44.1
36.4
71.9
50.8
22.5
79.6
73.0
58.4
but not on
8583
6602
2898
6028
3060
6682
7723
5822
6402
5278
10430
7370
3260
11550
10590
8467
2 and 3.
0
12
36
172a
141
-------
Table 45
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS
OF FBC SPENT BED MATERIAL FROM B&W PILOT
PLANT: CONTROL SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compressive Strength
MPa
Psi
Number of
Cycles
OBS0014B 1
2
3
Average after
OBS0014C 1
2
3
Average after
OBS0014D 1
2
3
Average after
9/15/77 10/13/77
,.
11
12 cycles
9/15/77 10/24/77
it M
.. i.
36 cycles
9/15/77 12/6/77
II M
II tt
317 cycles
68.2
48.8
57.6
58.2
34.1
54.5
53.2
47.3
79.4
79.4
67.0
75.3
9895
7080
8351
8442
4941
7897
7713
6850
11520
11510
9720
10917
12a
36
317
aThis number means these outside controls were tested after a lapse of
time corresponding to that for 12 freeze-thaw cycles, so that the
effect of aging alone could be compared with the effect of aging and
freeze-thaw cycling.
142
-------
Table 46
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS OF
FBC SPENT BED MATERIAL FROM EXXON PILOT
PLANT: TEST SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Corapressive Strength
MPa
Psi
Number of
Cycles
ES0014A 1
2
3
Average after
ES0014B 1
2
3
Average after
ES0014C 1
2
3
Average after
ES0014D 1
2
3
Average after
9/27/77 10/12/77
ii i*
it ti
0 cycles
9/27/77 10/13/77
II It
tl *t
12 cycles
9/27/77 10/24/77
•i ti
it (i
36 cycles
9/27/77 12/7/77
ii ii
it ii
317 cycles
25.4
36.5
26.6
29.5
21.0
41.3
16.4
26.2
18.1
35.3
22.0
25.1
76.5
106.9
67.7
83.7
3683
5299
3865
4282
3046
5986
2375
3802
2626
5118
3196
3647
11090
15504
9820
12140
0
12
36
317
143
-------
Table 47
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS
OF FBC SPENT BED MATERIAL FROM EXXON PILOT
PLANT: CONTROL SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compressive Strength
MPa
Psl
Number of
Cycles
OES0014B 1
2
3
Average after
OES0014C 1
2
3
Average after
OES0014D 1
2
3
Average after
9/15/77 10/13/77
ft if
,.
12 cycles
9/27/77 10/25/77
it *t
it it
36 cycles
9/27/77 12/6/77
ft u
it >•
317 cycles
29.1
35.2
25.1
29.8
48.0
42.2
33.5
41.2
66.0
43.0
61.4
56.8
4224
5100
3640
4321
6964
6118
4854
5979
9570
6230
8910
8237
12a
36
317
aThis number means these outside controls were tested after a lapse of
time corresponding to that for 12 freeze-thaw cycles, so that the
effect of aging alone could be compared with the effect of aging and
freeze-thaw cycling.
144
-------
Table 48
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS OF
FBC SPENT BED MATERIAL FROM PER PILOT
PLANT: TEST SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compressive Strength
MPa
Psl
Number of
Cycles
PS0014A 1 9/27/77 10/12/77
2
•\ « «
Average after 0 cycles
PS0014B 1 9/27/77 10/13/77
f\ i* it
3
Average after 12 cycles
PS0014C 1 9/27/77 10/24/77
2
o *• »•
Average after 36 cycles
PS0014D 1 9/27/77 12/7/77
O •* 11
3
Average after 317 cycles
29.8
25.7
35.7
30.4
20.4
37.9
30.7
29.7
54.4
37.8
57.7
49.9
80.0
84.5
82.9
82.5
4328
3729
5172
4410
2958
5494
4452
4301
7886
5479
8369
7245
11600
12260
12020
11960
0
12
36
317
145
-------
Table 49
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS
OF FBC SPENT BED MATERIAL FROM PER PILOT
PLANT: CONTROL SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compressive Strength
MPa
Psi
Number of
Cycles
OPS0014B 1 9/27/77 10/13/77
f\ it it
o •» ••
Average after 12 cycles
OPS0014C I 9/27/77 10/27/77
2 *• *•
3
Average after 36 cycles
OPS0014D 1 9/27/77 12/6/77
rt ti 11
3
Average after 317 cycles
27.7
38.2
23.3
29.7
52.0
59.3
28.2
46.5
66.2
52.8
78.9
66.0
4023
5539
3377
4313
7536
8598
4086
6740
9600
7660
11440
9567
12a
36
317
aThis number means these outside controls were tested after a lapse of
time corresponding to that for 12 freeze-thaw cycles, so that the
effect of aging alone could be compared with the effect of aging and
freeze-thaw cycling.
146
-------
Table 50
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS OF
TYPE I PORTLAND CEMENT: TEST SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Corapressive Strength
MPa
Psi
Number of
Cycles
IP0014A 1
2
3
Average
IP0014B 1
2
3
Average
IP0014C 1
2
3
Average
10/3/77 10/24/77 106.9
52.9
82.3
80.7
10/3/77 10/25/77 122.5
83.3
48.4
84.7
10/3/77 12/12/77 77.6
75.8
126.9
93.4
15504
7670
11931
11702
17762
12075
7019
12285
11250
11000
18400
13550
0
12
342
147
-------
Table 51
FREEZE-THAW RESISTANCE OF ISOSTATICALLY PRESSED CYLINDERS OF
PORTLAND TYPE I CEMENT: CONTROL SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compressive Strength
MPa
Psi
Number of
Cycles
OP0014B 1
2
3
Average after
OP0014C 1
2
3
Average after
10/3/77 10/25/77
ii ii
M 11
12 cycles
10/3/77 12/12/77
11 tf
n »*
343 cycles
110.9
20.4
48.7
60.0
104.4
65.5
135.5
101.8
16086 12a
2959
7064
8703
15150
9500
19650
14767 342
aThis number means these outside controls were tested after a lapse of
time corresponding to that for 12 freeze-thaw cycles, so that the
effect of aging alone could be compared with the effect of aging and
freeze-thaw cycling.
148
-------
Table 52
FREEZE-THAW RESISTANCE OF CAST CYLINDERS OF PORTLAND TYPE I
CEMENT AND RIVER SAND: TEST SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compress ive Strength
MPa
Psl
Number of
Cycles
IPRS14A 1
2
3
Average after
IPRS14B 1
2
3
Average after
IPRS14C 1
2
3
Average after
IPRS14D 1
2
3
Average after
7/7/77 10/12/77
ii *•
ii ii
0 cycles
7/7/77 10/13/77
tt tt
II M
12 cycles
7/7/77 10/24/77
ii ii
it it
36 cycles
7/7/77 12/7/77
..
•t ii
317 cycles
50.1
25.5
30.3
35.3
59.2
47.2
44.9
50.4
67.8
67.2
49.6
61.6
93.4
92.4
60.9
82.2
7263
3699
4390
5117
8581
6846
6511
7313
9842
9753
7193
8929
13540
13400
8840
11927
0
12
36
317
149
-------
Table 53
FREEZE-THAW RESISTANCE OF CAST CYLINDERS OF PORTLAND TYPE I
CEMENT AND RIVER SAND: CONTROL SPECIMENS
Sample
Designation
Date
Pressed
Date
Tested
Compressive Strength
MPa
Psi
Number of
Cycles
OPRS14B 1
2
3
Average after
OPRS14C 1
2
3
Average after
OPRS14D 1
2
3
Average after
7/7/77 10/13/77
« ii
it ti
12 cycles
7/7/77 10/25/77
ii (•
it ti
36 cycles
7/7/77 12/6/77
it ii
ti n
317 cycles
32.5
52.4
30.5
38.5
17.6
35.8
47.6
33.7
71.9
79.5
74.0
75.1
4715
7599
4425
5580
2556
5194
6909
4886
10430
11530
10730
10897
12a
36
317
aThis number means these outside controls were tested after a lapse of
time corresponding to that for 12 freeze-thaw cycles, so that the
effect of aging alone could be compared with the effect of aging and
freeze-thaw cycling.
150
-------
Curve 694001-A
100 i-
CD
03
CL
o
o
0
Exposed to Freeze-Thaw
100 200
Number of Freeze-Thaw Cycles
Figure 33 - Freeze-Thaw Resistance of Isostatically Pressed
Cylinders of Portland Cement and River Sand
-D
50
O
O
0
0
utside Controls
Exposed to Freeze-Thaw
100 200
Number of Freeze/Thaw Cycles
300
Figure 34 - Freeze-Thaw Resistance of Isostatically Pressed
Cylinders of Portland Cement
151
-------
Curve 693999-A
100
s.
50
o>
o.
E
o
o
Exposed to Freeze- Thaw
100
200
300
Figure 35 - Freeze-Thaw Resistance of Isostatically Pressed
Cylinders of Exxon Spent FBC Bed Material
100
S.
CD
K 50
o
o
0
0
Outside Controls
Exposed to Freeze- Thaw
100 200
Number of Freeze/Thaw Cycles
300
Figure 36 - Freeze-Thaw Resistance of Isostatically Pressed
Cylinders of B&W Spent FBC Bed Material
152
-------
Curve 694002-A
100 r-
OJ
>
E
o
O
Exposed to Freeze-Thaw
Outside Controls
0
1
0
100 200
Number of Freeze/Thaw Cycles
300
Figure 37 - Freeze-Thaw Resistance of Isostatically Pressed
Cylinders of PER Spent FBC Bed Material
These results have prepared the way for future leaching tests on
similarly prepared compacts in which the effect of freeze-thaw cycling
or leachability will be determined.
Figure 38 shows the mode of failure in compression for the A sam-
ples, corresponding to zero freeze-thaw exposure. Of these, the spent
bed specimens show cleaner fracture than do the control specimens made
with Portland cement and river sand (IPRS14A). The B samples, Fig-
ures 39 and 40, after 12 freeze-thaw cycles, however, show similar frac-
tures for spent bed specimens and the cement/sand control specimen both
for exposed and unexposed specimens.
153
-------
P.SOO/V-/?
ME TRIG 1
krtrrf
Figure 38 - Failure Mode of Isostatically Pressed Cylinders
of Spent Bed Material in Axial Compression -
Initial Specimens
154
RM-73161
-------
- B
- &
psoo
METRIC 1
Figure 39 - Failure Mode of Isostatlcally Pressed Cylinders
of Spent Bed Material in Axial Compression -
After 12 Cycles of Freeze-Thaw Testing
155
RM-73162
-------
METRIC 11
mm
Figure 40 - Failure Mode of Isostatically Pressed Cylinders
of Spent Bed Material in Axial Compression -
Outside Controls at Age Corresponding to 12
Cycles of Freeze-Thaw
156
RM-73163
-------
Tables 54 and 55 show the distributions of cycle times and cycle
temperatures during the freezing and thawing portions of the cycles.
Cycle temperature, it may be recalled, is the temperature inside one of
the Portland cement controls in the freeze-thaw chamber. From these
were calculated means and standard deviations for the low point of the
freezing portion and the high point of the thawing portion, as recorded
in Table 56. Both mean temperatures were slightly higher than in the
ASTM C666 procedure. Deviations from the planned temperature limits of
-17.8°C and +4.4°C (0° and +40°F) resulted from the simplicity of the
equipment plus incidental mechanical difficulties. On two the freezing
cycle continued until the temperature had dropped to about -50 to -60°C.
The specimens were thus subjected to even more severe stress from
freezing than had been planned. The main degradation, however, if any,
probably results from cycling through the phase change for water at
0°C.
Table 57 further compares cycle parameters used with those of ASTM.
Except for their cooling rates, the parameters are comparable to those
for the minimum cycle time of two hours. The maximum cooling rate in
the ASTM procedure can be estimated only from the maximum heating rate,
since the ASTM does not specify it. The estimated value of 0.4°C/min
applicable to the actual tests should be regarded as a minimum maximum.
If a lower heating rate Is used, less time is available for freezing,
and hence the cooling rate must be faster.
A review of the temperature charts showed the temperature changed
continuously on both cooling and heating. Only on close examination was
it possible to say that a plateau on a time-temperature plot occurred at
0°C, corresponding to the phase change for water. It appears this
plateau was of no more than three minutes' duration and was
distinguishable only in the early cycles. A simple model of Q\, the
time required to remove the heat of fusion of water, is given by:
a,\
9X - r
(cs + acw)(dt/d9)
157
-------
Table 54
DISTRIBUTION OF CYCLE TIMES
Interval
mln
Class Mark,
mln
Number of Cycles In Interval
Freezing
thawing
10-20
20-30
30-40
40-50
50-60
60-70
70-80
80-90
90-100
100-110
110-120
120-130
130-140
140-150
170-180
220-230
290-300
310-320
Other8
Total
15
25
35
45
55
65
75
85
95
105
115
125
135
145
175
225
295
315
11
174
84
30
14
11
5
3
1
1
2
1
1
1
1
1
341
3
344
1
6
78
127
86
24
11
6
1
1
341
3
344
aRecorder out of chart paper.
158
-------
Table 55
DISTRIBUTION OF CYCLE TEMPERATURES
Interval,
min
Class Mark,
°F
Number of Cycles in Interval
Freezing
Thawing
-14.5 to -9.5
-9.5 to -4.5
-4.5 to +0.5
6.5 to 5.5
5.5 to 10.5
10.5 to 15.5
15.5 to 20.5
20.5 to 25.5
25.5 to 30.5
30.5 to 35.5
35.5 to 40.5
40.5 to 45.5
45.5 to 50.5
50.5 to 55.5
55.5 to 60.5
-64.5 to -59.5
-89.5 to -84.5
Other4
Total
-12.0
-7.0
-2.0
+3.0
8.0
13.0
18.0
23.0
28.0
33.0
38.0
43.0
48.0
53.0
58.0
-62.0
-87.0
1
9
50
155
88
29
7
1
1
341
3
344
4
8
68
150
84
24
3
341
3
344
aRecorder out of chart paper.
159
-------
Table 55
MEAN VALUES FOR CYCLE PARAMETERS
Mean \
Freezing
Temperature, °C -15
°F 3
Cycle Time, rain 36
Thawing
Temperature, °C 6
°F 43.
Cycle Time, rain 78.
7alue Standard Deviation
.58 4.37
.95 7.87
.4 27.8
5 2.8
7 5.0
5 14.6
ASTM
C666
-17.8 ± 1.7
0 ± 3
4.4 ± 1.7
40 ± 3
30-60 minimum
Table 56
COMPARISON OF SELECTED CYCLE PARAMETERS WITH ASTM C666
ASTM C666
Actual
Cycle Time
Total (freezing & thawing)
Heating time/total time
Rates, °C (°F)/min
Heating - mean
maximum
Cooling - mean
maximum
2-4 hr
20% minimum
1.57 (2.83)
Not specified
(est. 0.4 (0.8))
114.9 ± 42.4 minutes
68%
0.28 (0.51)
0.57 (1.02)
0.61 (1.10)
1.23 (2.22)
160
-------
where
Xf - heat of fusion
cg, cw » specific heat of solid and water respectively
a « weight fraction of water in the compact and
dt/d6 - time rate of change of temperature.
For 10 percent water and a cooling rate of 0.5°C/min, about 40 min would
be required to remove the heat of fusion. This time indicates that one
or more of the following apply:
• The actual water content is no more than about 1 percent.
• The heat of fusion is equilibrated rapidly throughout the
compact so that it is removed continuously as sensible
heat.
• The total water is distributed throughout the compact so
only a very small fraction of it affects the thermocouple
tip.
Discussion of Results
The freeze-thaw results on isostatically pressed spent sorbent spe-
cimens support the view that spent solids from the fluidized-bed combus-
tion of fossil fuels can be processed to environmentally stable com-
pacts. The reason for the high compressive strengths demonstrated, how-
ever, was not evident. In particular, demonstrating the presence of
calcium silicates, as in normal Portland cement, would add credibility
to the hypothesis that the compacts do have long-term stability. As a
first step, we examined the chemical compositions of the test materials,
as shown in Table 58.
Here, the oxide compositions for spent bed materials shown in
Table 39 were allocated to normal Portland cement constraints, trical-
cium silicate, (CaO)3'81025 dicalcium silicate, (CaO>2'Si02» tricalcium
aluminate, (GaO^-A^Os? tetracalcium aluminum ferrite, (CaO)4«Al203
and CaSO^ We do not contend that these species exist in the
161
-------
Table 58
ALLOCATION OF CHEMICAL COMPONENTS IN FBC RESIDUES AMONG
NORMAL CEMENT CONSTITUENTS
Constituent W
Type I
Portland Cement
t % moles/100 g
C3S 46.2 0.202
C2S 26.2 0.152
PER,
moles/100 g
0.059
0.044
Exxon Bed,
moles/100 g
0.072
0.054
C3A 11.3 0.042
C4AF
CS'
Other
Excess S102
Excess CaO
8.2 0.017
3.1 0.022
5.0
—
—
—
0.249
0.350
—
—
0.364
0.017
—
B&W,
moles/100 g
0.037
0.028
0.042
0.012
0.238
0.616
8 - 8102
S' - S03
bed materials either as such or in the proportions calculated, but
rather that these allocations offer some insight into the nature of the
compacts.
We first assumed that all the 803 content was tied up as CS1 (See
Table 58 for coding) for all three residues. Next we found that this
left insufficient CaO to form either C3A from all of the alumina in both
the PER and the Exxon spent bed materials or C4AF from all of the Fe203
in these residues. Since the average composition of normal Portland
cement as shown in Table 58 * corresponds to 0.202 moles £38 per
0.152 moles C2S, the remaining CaO in these two residues was allocated
in this proportion, leaving excess free silica as shown. For the B&W
residue, the high CaO content permitted showing all of the Fe203 as
162
-------
C4AF, and the balance of the A1203 as 03^. The C3S and C2S contents
were again shown on the mole ratio as above, an amount sufficient to
account for all of the silica present, leaving an excess of 0.616 moles
CaO per 100 g bed material.
The reason for allocating CaO preferentially to at least C$S is
twofold. First, microscopic studies by others^ on hydration of these
individual components show that C$S and C3A start to hydrate immedi-
ately, whereas hydration of C2S requires several days. Hydration
appears to result in hydrolysis, releasing Ca(OH)2, some of which may
crystallize and some of which may participate In other reactions.
Second, Bogue and Lerch^' concluded that the early strength Port-
land cement was due to hydration of 038, while the contribution to
strength of €28 was significant only after 28 days, although in one year
its strength was about equal to that of €38. The contribution of C3A
remained less than 10 percent of that of the €28 or €38.
In comparison to Portland cement, the above allocations suggest
that the spent bed materials may contain no more than one-third to one-
sixth of the 038 and 028 contents of cement. We conclude that these
substances are probably not present in amounts sufficient to account for
the compressive strengths of the isostatically pressed specimens.
Supplementary Data
X-Ray Inspections
Having demonstrated that isostatically pressed specimens of ground
spent material could survive over 300 cycles of freezing and thawing
after curing for 14 days under water, further support for the conclusion
that these specimens were environmentally stable was sought through X-
ray inspections. The objective was to determine whether constituents
similar to those in normal concrete, especially hydrated calcium sili-
cates, could be found.
163
-------
For these Inspections samples were taken from the specimens used in
the freeze-thaw tests. All specimens had been isostatically pressed
dry, cured under water for 14 days, and then aged in ambient air for
various lengths of time. The B&W specimen was aged in air for 68 days,
the Exxon and the PER specimens for 56 days, and the Portland cement/
river sand reference specimen for 138 days. Before being scanned, all
specimens were ground manually in a mortar and pestle to a fine powder
estimated to be -44 urn.
Figures 41 to 50 are X-ray diffraction tracings obtained with a
Norelco Diffractoraeter of samples from these ground specimens. Excita-
tion was produced by a current of 35 mA at 40 kV using a copper target
with a monochromator. The divergence slit was set at 1° and the receiv-
ing slit at 0.2°. The scanning rate was 2 deg/mln; the chart speed
1 in/min; the counting rate 500/s.
From these tracings crystal lattice springs can be calculated by
the relation d = 0.770255 sin 9, where 9 is one-half the goniometer
angle and d is the lattice spacing in Angstroms. The height of the
tracings is a measure of the relative intensity of the X-rays exiting
from the sample from the various angles of incidence. The ordinate is
an arbitrary intensity scale. The instrument is usually set so the
highest peak falls near 100 in this scale. If a peak has an intensity
greater than 100, it is truncated on the chart, as shown, for example,
in Figure 49 for calcined gypsum at a d-spacing of 3.50.
Table 59 shows a comparison of the 20 sharpest peaks and their
associated d-spacings as actually observed for calcined gypsum (Fig-
ure 49), plaster of paris (Figure 48), gypsum (Figure 47), and silica
(Figure 45). The process of concluding that a particular molecular
species is present in the sample being analyzed involves the following
logic:
1. If a compound is present, its strongest traces ought to be pre-
sent, and, if absent, these traces should be absent.
164
-------
Ul
T" I"
W
";..*
a
i
t a. t ^.
: GrrCyr*1
T
i....... ^,f^.-nrr^
t.*i ^f; »( i i • x*v< «••• i -'i i. "j'jij'T J^ : ji! ! •' f
.(... | • i...U»*j.1..._ j^.^.y^y ^^i; ^^.jj•v-^rV''vlu:V V*fV^'**«»rrt''V'^
L- V-K : vK1 Vv 'i *''\ : ' :''v -: •'••''' : •'• --• - -•'>
._i_4_4__,.._L_^4_.L_l_i
, ITT 1 r .IT.ITI I
Ti .
»*«.' ; »•«*
• r
I
Figure 41 - X-ray Diffraction Tracing from Isostatically
Pressed B&W Spent Sorbent
IiT«;
«M>
J.
t
-r
T"
;t~r
T
*
^L ; ^ -».. t; «i «.
« •" i" S> S •<
•!"n~--T
Q
o ;
$
_..!.! ... ! !...)....:
#»
*'
(j.
6-
Q
p
C*
T *
•W^'>*fV<
Figure 42 - X-ray Diffraction Tracing from Isostatically
Pressed PER Spent Sorbent
-------
;••"•••* •:-* J » ^ • •£ V i • 2 '
v» NI j 7, -i w » •* ^ •»- *
k,^^'
T ^--T--1-
**»«.•-
-z i • i : v 4 - •*• "^ ^
A*
. j Cr\
f-'V*
lt*\\ !
!'- !f
: \l -'. •>.
Ct,
I • I -i MI ji vr
—r— »"l f .••••—r(--l--il-;-- -i"! L j
I/ * I* ' A l5/'J J -'J>
svv^'Arf^vr K 'i r>
IT
> • i
^ p.. |, ..j...
! ft i* f^ -w^-M ^/M/V^'M1
^ I h !•* \fcj' VV *V*% "• / ;'/'P • 1
: J.I .^ iu T^* 'i*/ i* T» .; ;..'!* r ...;...!....;
tJ-±
IV
rir
G, .
Figure 43- X-ray Diffraction Tracing from Isbstatically
Pressed Exxon Spent Sorbent
IT""
:1
-i '«. *
+F
A
^LLd
TTTTTT
' " «r ' t : ?
!_.,.
£
i'i
ll.:Si..ii'
i ; ; ^u,11:
]i,iV*r%ff' .*". j
\
_r_.r ..
* • *• ^ %• -*
•» n J5 — —
Q
51-^
r
•<'
i i S.
•o.
;* '*(;rr~
...j.. . - .
I
"*"i
• |--r«
ulL.
i . I :.. I .1
__'...I.
/
Figure 44 - X-ray Diffraction Tracing from Isostatically
Pressed Type I Portland Cement and River Sand
-------
+
\f
ill:
&
T-"! ' •-•! •-!-••!••
— A » «* . O
• <
.: I
i
V
•>WV- S1
^
.,,
irt:
I -
.v»
-IT
1
j... :
I :
:E
-t
_
—
«*•«
I1!
-«••
£
i
—
1
"
-
i
V-'
A
,i>
Figure
•^•!i!-,. N
i.
.. ..
~
j
--
,1
i-
]
"
ni
- i
•I
'
_„
I -
i
-TJ--
"*T
!
'....
.-
w<
!
•
V
.. ,
i
I.
i
.^
:l\\l
45-
• X-ray D
•<•%.' £ $
' -\ •
-•}•-
1
*/M^*-
U
.li.
i
f
L.
;!i
/
•;ilni,
...I..
-f-
i
!
J^,
.. U
^
i
i
I :
'•L
tffraction 1
1 f— J-- r-~| • ,-
.... , -- . M
f
i .
i
i
I
t"
W
Klillll
^
......
|
Illlnli
i
.,
.!,„
Cracing from Quartz
i'lU -
.. ......
*s\*
j !
x»^
.1...
J:
V|
..;;.
: >
j •
'!
i -
^'
,in
i
|
^
in
^
ill
1 R
i .
j
*
n
X-
•1!
1 "i
..J«
i
1
f
....
'
*
•1
Trl"^
. . . .
t
'i:
•w
....i...
A^
1
.»
T-T-
"-r
i .
(
y^
1
J..'.
=
; •!'
—
i- 1
1
';'
^
I ' i
....
#
:l
<*>
i
•
I •
"*|
.#
1
li I
.,,1., 1
-
. J
1
1
1
Figure 46 - X-ray Diffraction Tracing from Calcium Carbonate
-------
00
Figure 47 - X-ray Diffraction Tracing from Calcined Gypsum
s
*'
1 ;
' !
1
1
i
' ; ,
1,
)
1.
•-
«
! ' "
5~
h
>
?
.--
1
---
|
i:
*
-
*
i
.!
•-;
I
\
•
1
...
1
i
""!
...j
.
1
:l
*
i
j
h «
r •:
!
....:...
v*
r.-
—
\J
;!
•yrr-
->
....
*'.
f
$
r
1
-
A
r
i
i*
•il
|
:
t'
1
!
1
;
;. 1
;
J
jt
i
j
1
••
J
<
...
I
"it
1
i !
' U
" 1 ;
i
1
1
1
i
i
;""
;.
;' j
1
. ...
^
^
•v
J
i
•'•• -<
«<
j
- *»
\w
n i <
;
j
1 '
i
: •
-Vr
b . .
»
>.t*
•^''
,•
1
i
^*
I
"i
••\
-•
r!:;
.
r
*(*• M'
k
.:.,
\ 3
'
«*
w\ /JhJ
Lt'f:-
; .
;
:
i :
^•Ijf^it.'lf.
v, Afl "f tf+.
I
....,....J .-
: |
.
..
$
L'r"
i
i
A^
-TTT
f
i
r r-s»-
Figure 48 - X-ray Diffraction Tracing from Calcined Plaster of Paris
-------
VO
,,: I : I,
-ffl
fl
•;-:^4^^R
o ,»
-+ V-
:>
J
*
t^
^-T-r-'-j--:-^--^ T •-•- .i rp-v-r-,--
\J
iitttt
_j_
Figure 49 - X-ray Diffraction Tracing from Gypsum
-i'
: || i :j-y-T--T—p-T-^ ! —| -|~p|-j-r-,--, ^ : j-|fr*r—|-^T]|-i—f-r~|~T-| i | ; [~r-
t
•-
K
V 'I
IV
"T~JTt
Illllllllllllll
•i
..Lit
TjT
-A-U
*
L_
J....
y^
Figure 50 - X-ray Diffraction Tracing from Plaster of Paris
-------
Table 59
COMPARISON OF X-RAY PEAKS FROM SELECTED CALCIUM COMPOUNDS
o
D, A CaS04 CaS04'l/2 1^0
8.93
7.56
4.87
4.57
4.31
4.27
4.25
4.04
3.88 9
3.86
3.80
3.50 100+
3.46 100+
3.44 100+
3.34
3.32 5
3.13 5
3.06
3.04
3.02 17
3.00 100+
2.87
2.86 63
2.85 11
2.80 9
2.79 61
2.68
2.60
2.50
2.49
2.48 13
2.45
2.34 9
2.33 42 9
2.32 9
2.29
2.27 11
2.24
2.22
2.21 41 7
CaS04'2H20 CaC03 S102
100+ 5
5
5
5
100+
100+
5
24
54
100+
100+
100+
11
45
8
14
13
43
16
30
11
73
45
70
30
36
170
-------
Table 59 (Continued)
0
D, A CaS04 CaS04'l/2 H20
2.18 19
2.14 29
2.12
2.10
2.09 16
2.07
1.996 10
1.992
1.979
1.932
1.928
1.916
1.910
1.909 15
1.901
1.894
1.879
1.868 34
1.850 9
1.847 50
1.816
1.812
1.779
1.751 10
1.748 30 10
1.732 12
1.694 25
1.671
1.668 18
1.665 18
1.657
1.649 23
1.633
1.627
1.622
1.609
1.594 5
1.564 7
1.540
1.527 7
CaS04«2H20 CaC03 S102
45
36
29 15
20
12
33
16
18
44
43
41
48
19
100+
29
26
41
21
12
8
8
24
19
94
1.491 11
171
-------
2. If the same trace appears for more than one compound, It can be
used to identify only one of them.
3. Traces are considered to match references on standards if the
0
d-spacings are within ±0.02 A.
Thus, if silica in the form of quartz is present, one would expect
to find intense traces at d-spacings of 4.25, 3.34, and 1.816 Angstroms.
If any of these are absent, quartz is probably not present. Strong
traces at d-spacings of 2.45, 2.27, and 1.540 would serve as further
confirmation.
Calcium carbonate has only one intense peak - that at 3.04 Angstrom
spacing - and so the presence of this molecule depends more heavily on
the presence of relatively weak traces - those at 2.29, 1.916, 1.910,
and 1.894 Angstroms.
Calcium sulfate can be identified by the strong peak at 3.50
Angstroms and the weaker peaks at 2.86, 2.33, and 2.21 Angstroms.
Gypsum has three strong peaks: 7.56, 4.27, and 3.06 Angstroms.
The hemihydrate (plaster of paris) also has three strong peaks: 3.46,
3.44, and 3.00 Angstroms.
Since these characteristic peaks differ by at least 0.02 Angstroms,
one would expect that the identification of these five molecular species
in a given specimen would be easy; in practice, it was not. First, the
intensity of peak is proportional to the concentration of the molecular
species. Second, compounds containing similar elements or structures
may show peaks at the same d-spacing but at different intensities.
Third, the presence of other species may cause the peaks to be displaced
somewhat from their normal location on the tracings, corresponding to a
modification of the magnitude of the normal d-spacing.
Table 60 is a compilation of all peaks on the X-ray tracing for the
B&W specimen for which the relative intensity above background noise was
at least 5 percent. The expectation was that the sample would contain
172
-------
at least silica, anhydrite, and possibly hydrated CaSO^. Of the 35
peaks with a relative intensity of 5 or greater, 10 could be assigned to
gypsum. Anhydrite could account for another 10 peaks, while silica
could account for three more. The hemihydrate of CaS04 and limestone
could not be positively identified from the three peaks that matched
their characteristic tracing. This left unidentified nine peaks with
relative intensities in the range of 5 to 9 percent. Thus, even if
hydrated silicates are present, the intensity of their traces approaches
that of background noise.
Tables 61 and 62 are compilations of all the peaks on the X-ray
tracings for the Exxon and PER specimens from the freeze-thaw tests for
which the relative Intensity above background noise was at least 5 per-
cent. The Exxon specimen appears to contain gypsum, anhydrite, silica,
and CaC03, while the hemihydrate of CaS04 *-s probably absent. The same
result is obtained for the PER specimen. Nine peaks in the Exxon speci-
men and 21 in the PER specimen remain unidentified.
Table 63 summarizes the conclusions on the constituents expected in
the specimens. In only the cement mortar control was it possible to say
a calcium silicate was present.
The foregoing detail was presented to stress the uncertainties
encountered in attempting to identify molecular species in mixtures of
compounds when these compounds have similar d-spacings and therefore
show similar peaks. Because a compound may be present in minute
amounts, even characteristic peaks may show up with such low intensities
as to be Indistinguishable from background noise. Natural minerals can
be used to produce reference tracings, but differences will appear if
samples of a given mineral are taken from different geographic loca-
tions. This discrepancy is attributed to the presence of different
contaminants.
We reviewed the possibility of matching the peaks via computer.
Our X-ray Department had previously attempted this technique but found
173
-------
Table 60
X-RAY DIFFRACTION PEAKS FROM BABCOCK & WILCOX SPECIMEN NO. OBS0014C-3
Relative
Intensity
29
o
d, A
Relative
Intensity above
Background
Quartz
S102
Anhydrite
CaS04
Plaster
of Paris
CaS04'l/2H20
Gypsum
CaS04'2H20
Limestone
CaC03
62.5
44.0
31.3
30.0
27.1
24.0
17.0
19.1
22.7
13.0
14.8
14.0
17.0
13.9
12.8
25.0
12.5
25.4
26.6
31.4
20.8
29.1
31.1
40.8
33.3
11.6
43.3
38.7
48.6
33.2
35.6
50.1
9.0
35.5
3.50
3.35
2.85
4.27
3.07
2.87
2.21
2.69
7.62
2.09
2.32
1.872
2.70
2.52
1.819
9.82
2.53
53
35
23
21
19
16
13
12
11
11
10
10
9
9
9
8
7
3.34
4.25
2.24 ?
2.12 ?
1.816
3.50
3.32 ?
2.86
3.13
2.21
2.09
2.33
1.868
3.46
2.85
2.21
2.32
4.27
3.06
2.87
2.22
2.68
7.56 ?
2.09
1.879
1.812
2.85
3.04 ?
2.09
2.50 ?
-------
Table 60 (Cont'd)
in
Relative
Intensity
29
o
d, A
Relative
Intensity above
Background
Quartz
S102
Anhydrite
CaS04
Plaster
of Paris
CaS04-l/2H20
Gypsum
CaS04'2H20
Limestone
CaC03
11.9
10.0
10.0
10.0
15.0
14.5
15.0
10.5
15.0
13.8
13.5
13.2
12.8
12.8
11.0
10.5
8.1
7.8
36.2
52.2
55.7
55.8
20.5
22.9
23.2
41.4
15.9
23.5
24.1
24.2
27.7
32.0
34.2
36.0
39.5
45.4
2.48
1.751
1.649
1.646
4.33
3.88
3.83
2.18
5.57
3.78
3.69
3.67
3.22
2.79
2.62
2.49
2.28
1.996
7
7
7
7
6
6
6
6
5
5
5
5
5
5
5
5
5
5
2.45 ?
1.665?
(1.657?)
2.48
1.748
1.649
3.88
2.18
2.80
1.751
1.668 ?
1.665 ?
2.27
2.79
2.27
1.622 ? (1.633 ?)
4.31
3.86
3.80
1.996
2.79
2.60
2.49
(1.992 ?)
2.29
-------
cr>
Table 61
X-RAY DIFFRACTION PEAKS FROM EXXON SPECIMEN NO. OES0014C-3
Relative
Intensity
70.9
43.0
43.8
40.8
33.0
31.5
35.5
25.0
23.0
22.8
20.0
18.0
15.0
15.1
19.0
14.1
13.2
14.0
11.8
11.0
17.0
10.9
10.3
16.1
13.2
26
25.5
29.2
20.7
29.4
31.4
31.2
11.6
42.9
43.0
33.4
43.4
40.9
48.7
38.7
23.4
37.8
48.4
36.0
50.4
55.7
18.6
47.5
52.3
17.9
26.6
0
d, A
3.49
3.06
4.29
3.04
2.85
2.86
7.62
2.11
2.10
2.68
2.08
2.20
1.868
2.32
3.80
2.38
1.879
2.49
1.809
1.649
4.77
1.912
1.748
4.95
3.35
Relative
Intensity above
Background Si02 CaSO^
64 3.50
36
34
34
26
25 2.86
22
22 2.12
20 2.09
17
CaS04'l/2H20 CaS04'2H20
3.06
4.27
2.85
7.56
2.09
2.68
17 2.09 2.09, 2.07
15 2.21, 2.18 2.21 2.22
12 1.868
11 2.33 2.33, 2.32
10 3.80
CaC03
4.31 ?
3.04
2.85
2.10, 2.09
2.09
10
10 1.879
9 2.49
9 1.812 ?
9 1.649
8
8 1.748 1.748
8
7
7 3.34
1.910
-------
Table 61 (Cont'd)
Relative
Intensity 29
13.5 32.1
10.0 47.4
18.5 9.9
19.9 10.4
20.0 10.7
13.7 28.7
13.0 29.8
22.2 9.3
13.6 15.8
13.2 16.0
14.7 23.1
14.8 23.3
10.6 34.5
Relative
o Intensity above
d, A Background S102 CaSO^ C
2.79 7
1.916 7
8.93 6
8.50 6
8.26 6
3.11 6 3.13 ?
3.00 6
9.50 5
5.60 5
5.53 5
3.85 5
3.81 . 5
2.60 5
aS04'l/2H20 CaS04«2H20 CaC03
2.79 2.79
1.916
8.93
3.00 3.00
3.86
3.80
2.60
-------
00
Table 62
X-RAY DIFFRACTION PEAKS FROM PER SPECIMEN NO. OPS0014C-3
Relative
Intensity
60.0
47.1
48.2
38.8
33.7
35.1
25.0
25.0
23.8
25.0
21.8
14.0
14.1
16.0
13.4
18.9
12.1
10.3
16.0
12.4
10.0
9.8
15.9
15.0
15.2
29
20.8
29.2
26.7
11.7
31.2
25.5
31.1
33.4
31.4
11.6
23.8
40.7
43.4
34.2
40.9
18.1
47.9
50.3
18.0
36.6
37.4
38.7
13.1
17.9
19.8
o
d, A
4.27
3.06
3.34
7.56
2.86
3.49
2.87
2.68
2.85
7.62
3.80
2.22
2.08
2.62
2.20
4.90
1.897
1.812
4.92
2.45
2.40
2.32
6.75
4.95
4.48
Relative
Intensity above
Background S102 CaSO^
51 4.25 ?
40
38 3.34 3.32
26
26 2.86
25 3.50
17 2,86
17
16
12
CaS04'l/2 2H20 CaS04«2H20
4.27
3.06
?
7.56
2.85 2.87
2.85 2.87
2.68
2.85
12 3.80
11 2.22
11 2.09 2.09, 2.07
10 2.60
10 2.21, 2.18 2.21 2.22
9
8
8
7
7 2.45
7
7 2.33
6
6
6
1.812
2.32, 2.33
CaC03
3.04 ?
2.85
2.85
2.85
2.09
4.87
1.894
-------
Table 62 (Cont'd)
VO
Relative
Intensity
14.8
19.4
14.4
12.8
13.0
13.8
8.0
10.9
8.2
8.3
26.1
22.8
13.7
14.0
13.8
14.4
13.8
12.2
12.0
8.2
9.8
9.8
9.8
8.0
7.8
8.3
26
22.5
23.3
24.1
28.1
32.2
35.7
45.6
48.5
55.8
56.7
8.0
9.2
16.0
18.6
19.1
20.4
21.7
27.7
32.6
39.5
48.8
50.8
51.0
51.5
52.3
55.2
o
d, A
3.95
3.81
3.69
3.17
2.78
2.51
1.988
1.875
1.646
1.622
11.04
9.60
5.53
4.77
4.64
4.53
4.09
3.22
2.74
2.28
1.865
1.796
1.789
1.773
1.748
1.622
Relative
Intensity above
Background S102 CaS04 C
6
6
6
6
6 2.80
6
6
6
6 1.649
6
5
5
5
5
5
5
5
5
5
5 2.27
5 1.868
5
5
5
5 1.748
5 1.665
aS04«l/2 2H20 CaS04«2H20 CaC03
3.80
2.79 2.80, 2.79
2.50
1.992
1.879
1.622
2.27 2.29
1.748
1.665
-------
Table 63
X-RAY DIFFRACTION ANALYSIS OF SELECTED ISOSTATICALLY PRESSED SPECIMENS OF FBC
RESIDUES AFTER 14 DAYS CURING IN WATER
oo
o
Constituents
Ca(OH)2
CaS04
CaS04*2H20
CaS04'l/2H20
MgO
Si02
CaS103
CaC03
Sample Number
Cement Mortar
OPRS14C-3
Minor
Trace
Trace
Major
Trace
Babcock & Wilcox
OBS0014C-3
Trace
Major
Major
NCb
Minor
ND
NC
Exxon
OES0014C-3
NDa
Major
Major
NC
Minor
Trace
ND
Trace
Pope, Evans & Robbins
OPS0014C-3
Trace
Minor
Major
NC
Minor
ND
Trace
aND
Not detected
Not confirmed
-------
that the results were variable. If the tolerance on the d-spacing was
set too wide, the number of compounds containing a particular peak was
increased; if set too narrow, a possible identification would be missed.
We concluded that some other technique was needed. One such possibil-
ity, although not attempted here, is a microscopic technique that might
take into account crystal form and index of refraction. The latter
would depend on the preparation of thin sections. The overall objective
of this line of investigation would be to make more definite predictions
about the long-term environmental stability of the solid compacts.
Weight Loss on Heating
Two specimens were spot checked by TGA for their weight loss when
heated to 1000°C as another means of determining whether hydrated sili-
cates could be present.
Table 64
WEIGHT LOSS FROM CURED ISOSTATICALLY PRESSED SPECIMENS
OF F3C RESIDUES WHEN HEATED TO 1000°C
Sample
IP0014C-3
ESOOUC-3
% Loss <§ 700°C
9.93
12.00
% Loss (3 1000°C
10.82
13.70
The weight losses to 700° interpreted as water losses are only
about half the theoretical values for hydration of Portland cement
(about 24 wt %). It is interesting that corapressive strengths well
above those of normal cement can be obtained with so little water.
Discussion
The low water-content of the cured, isostatically pressed specimens
does not support the conclusion that the high corapressive strengths
obtained are due to the presence of hydrates typical of Portland cement.
181
-------
The only hydrate identified, gypsum, is reported in the literature to
have a compressive strength in the range 6.2-1512 MPa (900-2200 psi).
An alternative explanation is that the strength is due to intermolecular
forces operative because of the fine particle size of the powders prior
to being pressed and to the high pressure used to form the specimens.
This explanation would be consistent with the freeze-thaw results
in that any ice crystals formed during the freezing portion of the cycle
would have unoccupied pore volume into which to expand. If the pores
were saturated with water, the specimens would probably have deterio-
rated in the freeze-thaw testing. Hence, formation of compacts by dry
pressing of FBC residues ground to at least 100 percent through 63 ym
with subsequent water curing for 14 days offers promise of a way to
reduce the potential for leachate production from these residues that at
the same time confers long-term freeze-thaw resistance on the compacts.
Aggregate
Process
Two of the processes conceived for disposing of FBC residue lead
to its utilization as aggregate. The compacts described in the section
under Pressed Material were investigated with the view of depositing
them in the environment, but they might also be usable as coarse aggre-
gate. Another possibility is to take the granular material from the FBC
process and use it directly as fine aggregate in concrete. We therefore
devised tests to explore the technical feasibility of this option.
Concrete is basically a physical mixture of three types of solids.
Most of the volume of this mixture consists of fragments of rock of size
distributions appropriate to the application. This is termed coarse
aggregate. The aggregate is held together by cement, which develops its
binding strength when water is added to it through the formation of
hydrated silicates. To minimize the quantity of cement required in the
mixture, the voids between the coarse aggregate pieces are filled with
182
-------
fine aggregate, which typically is a sand. Sand also is a source of
additional silica which can react with Ca(OH)2 released on hydrolysis of
silicates in the cement to form gels.
Test Materials
Sintered pelletized fly ash (SFA), 1/2 in. x No. 8, from the Penn
Virginia Materials Corporation of Eastlake, Ohio was used as the coarse
aggregate. Chemical analysis of typical Penn Virginia SFA is in
Table 65. Chemical analysis for FBC residues were shown in Table 40.
Bulk density measurements of spent sorbents and various fly ashes
are in Table 66. Although the as-is material was used in the tests
described below, the bulk densities of two fractions are included for
future reference.
Table 65
TYPICAL ANALYSIS OF PENN VIRGINIA SINTERED FLY ASH
Constituent
Si02
A1203
F6203
CaO
MgO
803
C
LOI
Available alkalis
Weight %, Dry Basis
53.4
27.8
15.2
2.52
0.48
0.04
2.25
1.91
0.13
Note
1. Moisture content of as-is sample: 5.52%
183
-------
Table 66
BULK DENSITY OF MATERIALS FOR AGGREGATE TESTS, g/cc
Material
FBC Residues
B&W-Run 19
PER
Exxon-Run 43-3
As is
1.084
1.513
1.587
-125 + 44 urn
1.236
1.448
1.511
-44 )am
0.996
1.252
1.245
Fly Ashes from Conventional Boilers
Duquesne 1.099
Wyoming 1.413
Penn Virginia SFA 1.450
As a preliminary, water absorption measurements were made on 100-g
random samples of Penn Virginia 1/2 in. x No. 8 SFA. The SFA appeared
porous and therefore potentially capable of absorbing a significant por-
tion of the water added to the mix, thereby interfering with the desired
hydration reactions. Three immersion times of duplicate samples were
investigated. The first 100 g samples were taken directly from the
208 fc (55 gal) drum of the as-received Penn Virginia fly ash and dried
to constant weight at 105°C. The weight loss was then calculated as
percent moisture. Two duplicate samples, labelled A and B, were
immersed in water for 24 and 144 hours, respectively. The samples were
blotted dry with paper towels, weighed, and then oven dried to constant
weight at 105°C. The results are presented In Table 66. The samples
appear to show negligible changes In water content when exposed for
24 hours and no change thereafter. One gram of aggregate takes on about
0.017 cc additional water in 24 hours. This may be a significant change
when compared to the total pore volume of the dry aggregate In investi-
gation of freeze-thaw resistance but is considered unimportant in the
explanation of behavior in hydration reactions.
184
-------
Performance Characteristics
Proceeding, then, with the main tests, we found that mixes of B&W,
PER, and Exxon spent bed materials, 1/2 in. by No. 8 SFA, Type I
Portland cement and water developed enough heat to produce steam and
premature set of 7.6 cm by 15.2 cm (3 in. by 6 in.) cylindrical speci-
mens. Slaking the spent bed materials prior to blending eliminated the
problems caused by heat generation. A 500 g sample of each was there-
fore hydrated with 1000 ml of water, mixed thoroughly, and allowed to
cool to room temperature. The excess water was decanted and saved for
addition later.
The specimens were made using a modified ASTM C-192 procedure for
mortar composition, substituting 1/2 in. by No. 8 SFA and sorbent for
white Ottawa sand.
The final mortar compositions consisted of 500 g of Portland Type I
cement, the slaked stone prepared as described above, 500 g of 1/2 in.
by No. 8 SFA, and the previously decanted water of hydration. One spe-
cimen was prepared with this composition for each of the three FBC spent
bed materials. In the case of the B&W material, an additional specimen,
BA 8527, was prepared using a lower bed material/SFA ratio (150 g of bed
material/850 g SFA). The casting procedure was as per ASTM C192 using
7.6 cm by 15 cm (3 in. by 6.2 in.) cardboard cylinder molds. A control
cylinder of Type I Portland cement and white Ottawa sand was also cast.
All of the specimens were moist cured for a 14-day period except for BA
8527, which was moist cured for 27 days. The specimens were then tested
for axial compressive strength, with the results as shown in Table 67.
The data show that the Exxon and PER spent bed materials, after preslak-
ing, can be used as fine aggregate in conjunction with SFA to make
Portland cement mortars that develop compressive strengths equal to nor-
mal concrete at the same curing age. Both the B&W specimens were sub-
stantially lower in strength (less than half). One possible reason is
the higher CaO content, which may have resulted in excessive Ca(OH)2
185
-------
Table 67
WATER ABSORPTION BY PENN VIRGINIA 1/2 IN. X NO. 8
SINTERED PELLETIZED FLY ASH
Immersion Time, hr
Weight %
Sample A
0 3.0
24 4.7
144 4.7
Moisture
Sample B
3.4
5.2
5.3
Notes:
1. Samples were random 100 g portions from 208 I (55 gal)
drum.
2. Samples were blotted dry with paper towels after immersion
and then oven dried to constant weight at 105°C.
Table 68
COMPRESSIVE STRENGTH OF CYLINDERS MADE FROM PORTLAND CEMENT,
FBC SPENT BED MATERIALS AND SINTERED FLY ASH
Specimen
Designation
BA 5014
BA 8527
EA 5014
PA 5014
PWS 14 (2)
Spent Sorbent
B&W Run 19
B&W Run 19
Exxon Run 43
PER
None
Axial Compressive
MPa
10.1
10.7
22.9
22.1
23.8
Strength, 14 days
Psi
1460
1550
3320
3210
3450
Notes:
1. Cylinder size 7.62 cm D by 15.24 cm H (3 in. by 6 in.).
2. Portland cement/white sand control.
186
-------
content in the final cylinders. The BA 8527 specimens, however, which
had less than one third as much bed material as did BA 501A, showed only
a slight increase in strength relative to BA 5014. Further tests are
clearly needed to identify the reason for this behavior.
ENVIRONMENTAL IMPACT OF PROCESSED WASTE
Empirical Correlation for Laboratory Results
Further perspective on the effect of fixation on potential environ-
mental impact is provided by analysis of selected data from Section 6
covering the leaching behavior of raw and processed residues from Run 27
of the Exxon miniplant. The data on intermittent leaching could be used
to calculate the average rates of extraction of a particular solute, as
calcium, for each time period. We chose rather to calculate the amount
of calcium leached in terms of mg-moles/extraction, cumulate these
values, and plot these totals against the number of the extraction
period. These data would simulate repeated rainfalls on exposed resi-
dues, either fixed or as produced, if we assume a constant rainfall/
solid weight ratio and 100 percent efficiency of contacting.
The data are presented on a normalized basis in Figure 52, obtained
by dividing each of the ordinate values of Figure 51 by the weight of
the specimen involved. Figure 52 shows the substantial reduction in
leaching obtained by fixation [Samples I1D and HAD1 versus spent bed
material (SBM)]. It also shows that crushing the cubes results in mate-
rial from which more calcium can be leached than from the raw residue.
We noted that these curves developed constant slopes after about 10
extractions, so empirical equations of the following form were fitted to
the data:
LQ(1 - e
-biN
187
-------
Curve 722629-A
Curve 722702-A
00
00
21
o>
E
c.
E
3
^
n
O
o>
>
E
0
_r
160
140
120
100
80
60
An
HU
20
0
i i i i i i i i i i i i
Legend
Sample Weight. Description Symbol
- 1 1 D 182 5-cm Cube o
II 4 D1 206 5-cm Cube °
II 4 C' 110 Crushed Cube a
- SBMa 33 Raw Residue ?
aSpent Bed Material
Note: Extraction Cond itions Were 3 g Water/ g
Solids for 72 Hours Each
t/
S
/*
Correlating Curve ^ ^
.?^
Experimental Data^^^/
/ _o^-°"
/ -^°^o— °"°~
— /& ^^"fi^^^""
X o^°"*8"
" ^^>^^^^
^Y i i i i t i i i i i t
1 '
-
-
11 *y
r -
~
I 10
II 40'
SBM ~
-
, ,
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
N, Number of Extractions
o
CTV
1.2
ii 0. 8
ns
o
o>
_>
5 0.6
^
E
^
o
-8 0.4
o 0.2
Sample
I1D
II4D1
II4C1
Legend
Description
5-cm Cube
5-cm Cube
Crushed Cube
Raw Residue
Spent-Bed Material
SBM
a
Symbol
o
a
A
v
Note: Extraction conditions were 3g water/g
solids for 72 hours each extraction
Correlating Curve
Experimental Data
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
N. Number of Extractions
Figure 51 - Leaching of Calcium from Exxon
Spent Sorbent/Carry-over
Compacts
Figure 52 - Leaching of Calcium from Exxon
Spent Sorbent/Carry-over -
Normalized Basis
-------
where LT - cumulative calcium leached after N extractions, mg-moles, and
LQ, t>i and b2 are constants. Differentiating this yields an expression
for the leaching rate, mg-moles per extraction:
Table 69 presents numerical values for the constants for four
cases. Figures 51 through 54 show that the experimental data are repre-
sented well by the empirical equations. We used a piece-wise, least-
squares approach to the curve fitting which can probably be improved if
desired.
Calculated Leach Rates
For comparisons we calculated rates at two different times, the
initial rate and the rate at 1080 hours (15 extractions) (Table 70).
Although the leach water was always three times the sample weight, the
sample size was not constant, so rates were normalized by dividing by
the initial sample weight to obtain specific leach rates in terms of
mg-moles calcium/g of sample-hr.
With respect to initial rates, the raw spent bed material had the
highest specific calcium loss rate. Forming the material into cubes
reduced this rate by a factor of about 3. A longer curing time (60 vs.
7 days) was associated with a further decrease in specific leach rate.
Crushing the cube to powder resulted in an increase in leach rate, but
even then only to about 80 percent of that of the unprocessed material.
In all four cases specific leach rates attenuated with time, the
greatest reduction occurring for II4D' , which had been cured the longest
(60 days). To evaluate the long-term results further, the nature of the
test material should be invoked:
1. The spent bed material (SBM) was tested as received. The orig-
inal sorbent was 2380 by 707 ym (8-25 mesh).
189
-------
Table 69
CONSTANTS IN THE EMPIRICAL CORRELATION OF CUMULATIVE CALCIUM LEACHED
VERSUS NUMBER OF EXTRACTIONS
\D
O
Sample
Number
Description
Carry over
Total Solids
Water
Total Solids
Curing
Time, days
Form
Constants
LO* -bit 1 b2
I1D
II4D'
II4C'
Spent
Bed
Material
0.358
0.100
0.100
0.000
0.445
0.300
0.300
0.000
7
60
60
•"•
Cube
Cube
Powdered Cube
Raw Powder
27.954
27.371
26.637
11.746
0.29713
0.30930
0.22848
0.29062
2.266
1.934
7.113
1.458
*LQ and 02 have the dimensions mg-raoles Ca/extractlon.
is dimensionless.
-------
2. The other three specimens were prepared from SBM that had been
ground to -125 jam.
3. Specimen II4C' was crushed to powder before being tested.
If the average particle size for SBM Is taken as about 1200 pm, then the
surface area per gram of material for SBM should be about 40 times that
for the 5-cm cubes, yet the leach rate was only 3-to-5 times that for
the cubes. This suggests that grinding the SBM may make the material
Table 70
CALCULATED LEACH RATES FOR SELECTED FBC RESIDUES
Quantity
Sample Identification
I1D
II4D'
II4C'
Spent Bed
Material
Sample Weight, g 182 206 110 33
Leach Water, g/extraction 546 618 330 100
Sample Composition, wt %
Spent bed material
Carry-over
Water
Total Calcium, as Ca
44.4
24.8
30.8
100.0
11.6
Leach Rates, yg moles Ca/g sample-hr
Initial 0.807
After 1080 hours 0.180
Leach Rates, mg moles Ca/m^-hr
Initial 11.75
After 1080 hours 2.62
69.2
7.7
23.1
100.0
15.7
0.701
0.136
11.56
2.24
69.2
7.7
23.1
100.0
15.7
1.667
0.923
100.0
100.0
21.8
2.050
0.632
191
-------
more susceptible to leaching even in fixed form. This conclusion is
further supported if one noted that the total calcium in the SUM was
actually 30 to 100 percent higher than in the other three specimens.
Specimen II4C" shows a long-term leach rate 50 percent higher than
that of the raw SBM. This difference might be due to a combination of
the initial grinding effect noted above plus the regrinding to powder.
The particle size distribution was not measured, but if it averaged
800 um this could account for the higher leach rate long terra compared
to SBM. It would not explain the lower initial rate, however, but sub-
sequent work might show that the difference in leach rates between spec-
imens such as II4C" and SBM are within experimental variance.
Leaching Model
For an actual residue disposal operation, the residues may be pro-
cessed to a monolithic mass. To show the advantage of such a choice,
one may visualize a mass 10 m high by 100 m long by 100 m wide, which
might be accumulated in one year from operation of a 200 MWe plant. We
would further assume, as a first approximation, that leaching is mainly
a surface phenomenon. We also assume no fissuring of the monolith.
Leach rate at time t is then expressible as:
where
LX" is the leach rate in mg-moles calciura/hr, ,
A is the total surface exposed to leaching, m^, and
k is a proportionality constant, mg-moles calcium/m^-hr.
For a rectangular mass,
A - 2(LW + WH + LH)
192
-------
where
L « length of the mass, m
W » width, m, and
H » height, m.
Two of these dimensions can be related to the third:
L - ajW
H - a2W
so that,
A • 2W^(a^ -f a2 + a^a2)
If, instead of as a monolith, the residue was deposited as N particles
of uniform size, say a cube of side E, the number of particles would be
N _ LWH m W^
u Ci
and the total surface area would be
,_2 W3 , W3
S • 6E —T- a1a2 - fca^ g-
If one compares,
i » 3ala2 W
If &i - 1, and &2 * 0>1 to conform to the mass visualized above,
S 1 W
A " "4 E *
193
-------
So if E < W/4, the surface area will be larger for the partlculate
deposit than for the monolith, and the leach rate will be larger. If
one assumes E = 5 cm, then S is 500 times A, and the leach rate for 5-cm
cubes is 500 times that of the monolith. This illustrates the rationale
for fixation of residues to monoliths.
Leach rates for granular materials appear to be more meaningful
when expressed as quantity of solute leached per gram of material being
leached per unit time. For masses the more meaningful rate is the quan-
tity of solute leached per unit of surface area per unit time.
Further calculations can be made to bring out the effect of certain
constraints, if we assume these operate independently. For the actual
deposit we would think that the area leached by rainfall would be the
top and perhaps the (four) sides, but not the bottom. We would neglect
contributions due to permeability. The surface area for the monolith
visualized above would be 14,000 in-*. Table 71 shows the leach rate
after 1080 hours to be 2.24 mg-raoles calcium/m^-hr, yielding a rate of
4.3 kg/hr, expressed as CaSO^. Presumably this rate would attenuate
with time for two reasons: the surface areas would decrease, and, since
the block is really not all soluble calcium, diffusional resistances
would increase. If, however, we assumed no decrease in leaching rate
after 1080 hours and that the mass was all CaS04, complete solution of
the mass would require nearly 3500 years. For 30 years' operation, if
only the solution rate were involved, the total mass produced would
require 105,000 years for complete solution. A. more exact estimate
would take into account the fact that the area exposed to leaching is
decreasing.
An average rainfall of 1.5 ra/year, however, is another constraint
since this corresponds to a flow of only 1.71 m3/hr (452 gal/hr). With
maximum efficiency of contacting, the concentrations of CaS04 in the two
leachates would then be 2510 and 3000 ppm, respectively. The solution
rate of the unfixed residue would be limited by rainfall to 5.13 mg
CaS04/hr.
194
-------
Still further, the actual calcium content will probably be consid-
erably less than that for saturated CaS04. The laboratory leach data at
1080 hours suggest a concentration of perhaps only 100 mg Ca/fc, which is
equivalent to 340 mg CaS04/£, well below 10 x DWS or 250 mgA. With
modest improvement this amount would be within the DWS itself. Without
fixation the concentration could be six times as great, although still
unsaturated and less than 10 x DWS. The corresponding solution rates
with the available water would thus be 0.58 and 3.49 Mg CaS04/hr. The
latter is thus set by the quantity of water available and the efficiency
of contacting. The former is not governing since the effective con-
straint is the area available for leaching, and the rate is the
0.0043 Mg/hr calculated above. In contrast, the leaching rate for the
unprocessed spent sorbent of 0.632 ug moles Ca/g-hr shown in Table 70 is
equivalent to 11.2 Mg CaS04/hr, which is 2600 times that for monoliths
and would result in complete solution in less than two years.
Casting small masses, as for example 5-cm cubes, is not very help-
ful: the time for complete solution of the 10 x 100 x 100 m quantity is
about 4.5 years.
Further Constraints
The above calculations are, however, constrained further by the
efficiency of contacting and, for the granular material, also by the
quantity of rainfall available. If we'assume that a saturated solution
of CaS04 is formed (3000 ppm), complete solution of the above mass would
require 4.3 x 10' m^ of water. At an average rainfall of 1.5 m/yr in
the United States, the mass would last nearly 3000 years, whether or not
the residue is processed to monoliths. For 30 years' operation the
total mass produced would last 90,000 years. Placing residues from
large FBC plants in the environment, therefore, means making a source of
substances such as CaS04 and Ca(OH>2 available for leaching by environ-
mental waters for a very long time.
195
-------
The overall conclusion is that, considering the leachate production
as above, if the FBC residues are placed in a landfill, the effect will
be the potential generation of an average of 1.71 m^/hr of a saturated
solution of CaS04. The actual amount is expected to be very much less.
In any case, it is not yet clear that this will represent an unaccept-
able environmental impact since, in general, there will be dilution from
natural waters. Further, fixation of the residues into large masses
offers the potential for reducing the rate of leaching to negligible
levels.
196
-------
9. ENVIRONMENTAL ASSESSMENT
Results from the laboratory-scale test data on the currently avail-
able spent bed and carry-over materials suggest that it is technically
feasible to dispose of the FBC residues directly without polluting the
environment. Site selection, design, and management of such disposal,
based on the site-specific hydrology, geology, climate, and soil compo-
sition, are critically important to success. Processing the spent FBC
material will reduce its surface area and permeability and improve the
heat release and leaching properties. Process choices are available to
the disposal management task, with the preferred selection dependent on
environmental criteria and the economics of the acceptable options.
Pending the implementation of EPA criteria with which to assess the
environmental acceptability of the disposal of FBC residues, the chemi-
cal, physical, and leaching properties of the spent FBC material are
compared with DWS, with the leaching properties of natural gypsum, and
with the leaching properties of residues from conventional coal-burning
power plants with currently commercialized FGD processes. A preliminary
comparison of the environmental impact of the disposal of FBC solid wastes
and FGD sludge residues from varying processing systems is presented in
Table 71 based on the up-to-date results from parallel environmental
testing programs. Since the samples tested resulted from the use of dif-
ferent coal and sorbents, an absolute comparison may be impossible,
although the available data indicate the trends are meaningful.
Table 71 shows that the•'physical, chemical, and leaching properties
of FBC residue are superior to the nonstabilized FGD sludge with regard
to disposal and are comparable to the chemically and physically
stabilized FGD residue.
197
-------
Table 71
g. 2618C33
COMPARISON OF THE ENVIRONMENTAL IMPACT OF THE FBC AND FGD
SOLID AND LIQUID WASTE DISPOSAL**
Process
Sample
Environmental Parameters
Sorbent Type
Heat Release*'0'
(spontaneous
temp rise I
3 9/20 ml
Trace
Metal
TDS
Pressurized FBC,
Once-through
Bed material.
leachate
Limestone
<0.2°C
Pressurized FBC.
Once-through
Bed material
leachate
Dolomite
< 0.2°C
Pressurized FBC.
Regenerative
Bed material
leachate
Dolomite
limestone
0-3°C
Atmospheric FBC.
Once-through
Bed material
leachate
Limestone
5 to 15°C
Atmospheric FBC.
Once-through
Carry-over
material, leachate
Limestone
0-3°C
Pressurized FBC,
Once-through
Carry-over Limestone,
material, leachate
<30ppm
Pressurized FBC.
Once-through
Processed compacts : Dolomite
from bed carry-over
mixtures, leachate
<0.2°C
<30ppm
FGD
LiQUorof
untreated sludge ..'"'limestone
< 30 pom
FGD
Liquor of Lime
oxidized sludge
<30ppm
FGD
«U«C
OOppm
FGD
Leachate of dried
oxidized sludge
<0.2°C
<30ppm
FGD
Leachale of
dry stabilized
sludge
<0.2«C
< 30ppm
Gypsum
Leachate
Natural
<0.2°C
<30ppm
ra Do not meet either the drinking water or gypsum leachate criteria
^ Piss gypsum leachate criteria but not Drinking Water Standards
Q Pass both drinking water and gypsum leachate criteria
la) Based on data from currently available FBC and FGD residues
Ibl No existing criteria
(c) Subjected to the specified procedures, i.e.. 3 g sol id added to 20 ml H?0
Idl All except two carry-over fines pass Drinking Water Standards (total of 30 samples tested)
198
-------
The work on spent sorbent processing provides a basis for choosing
among alternatives to direct disposal that can result in reduced environ-
mental impact through disposal or utilization. This work includes studies
on the long-term stability of low-temperature processed compacts, on the
possibility of using the residues directly as aggregate in concrete (which
would result in resource recovery as well as minimizing environmental
impact), and on a preliminary engineering assessment of fixation.
Several approaches to investigating long-term stability were possible,
but the most relevant test was judged to be the freeze-thaw cycle. If
the compacts broke down into smaller fragments on aging, freezing, hydrat-
ing, or reacting with other substances, the whole point of making them
would be negated. In the processing studies we decided to pursue the
option of disposing of residues in production proportions only. This
meant that only one process would be required rather than one for bed
material and one for carry-over.
The variables included spent sorbent source, particle size range,
carry-over content, spent bed material content, and forming method. To
meet the constraints of time and the quantity of test material available,
we devised a sequential experimental plan. We selected isostatic pressing
as the initial forming method in the expectation that compacts produced
in this manner would be more likely to survive freeze-thaw testing.
Failure, however, would have indicated that the concept of compact forma-
tion was probably unfeasible. Specimens made from ground spent sorbent
survived for 300 cycles of freeze-thaw exposure at -18 to +40°C (0° to
40°F).
An analysis of leaching concentration and rate data over 1080 hours
on unprocessed and processed compacts demonstrates the potential for
significantly reducing environmental impact. The potential for reducing
direct effluent concentrations to the DWS level is indicated on the
basis of the preliminary analysis.
199
-------
An alternative disposal method tested was utilization directly by
replacing some of the sand in normal concrete with fine aggregate. Por-
tions of the three pilot plant spent sorbents were slaked in water,
blended with sintered fly ash and Portland cement, and cast into 7.6 cm
x 15.2 cm cylinders. After being cured for 14 days, the cylinders demon-
strated axial compressive strengths in the range 10 to 24 MPa - the low
end of the range for normal concrete.
The environmental impact and spent sorbent processing test results
are encouraging. A comparison of the physical, chemical, and leaching
properties indicates that the disposal of FBC residue would probably not
cause more negative environmental effects than the residue from a conven-
tional coal-burning power plant with FGD systems. On the basis of the
findings reported here and in previous Westinghouse reports, ' we expect
that disposal of FBC residue will not be an obstacle in the commercializa-
tion of the FBC process.
The spent sorbent processing studies show that alternatives are
available that can offer economic, resource, and environmental advantages
for the disposal of FBC residues.
200
-------
10. REFERENCES
1. Resource Conservation and Recovery Act, Public Law 94-580; 1976.
2. Clean Water Act, Public Law 95-217; 1977.
3. Federal Water Pollution Control Act, Public Law 92-500; 1972.
4. Clean Air Act, Public Law 95-95; 1970; as amended 1977.
5. Sun, C. C., C. H. Peterson, R. A. Newby, W. G. Vaux, and D. L. Keairns,
Disposal of Solid Residue from Fluidized-Bed Combustion: Engineering
and Laboratory Studies. Report to EPA, Westinghouse Research and
Development Center, Pittsburgh, PA, March 1978, EPA-600/7-78-049,
PB 283-082/6 ST.
6. Keairns, D. L., D. H. Archer, J. R. Hamm, S. A. Jansson,
B. W. Lancaster, E. P. O'Neill, C. H. Peterson, C. C. Sun,
E. F. Sverdrup, E. J. Vidt, and W.-C. Yang, Fluidized Bed Combustion
Process Evaluation, Phase II - Pressurized Fluidized Bed Coal Combus-
tion Development. Report to EPA, Westinghouse Research Laboratories,
Pittsburgh, PA, September 1975, EPA-650/2-75-027c, NTIS PB 246-116.
7. Corson, A., D. Friedman, and D. Viviani, Hazardous Waste Management
Division, EPA-Office of Solid Waste Management. Private
Communication Environmental Protection Agency, 1978.
8. Environmental Protection Agency, Hazardous Waste Guidelines and
Regulations - Criteria, Identification, and Listing of Hazardous
Waste, Draft, March 1978 and September 1978.
9. Environmental Protection Agency, Hazardous Waste: Proposed Guide-
lines and Regulations and Proposal on Identification and Listing,
Federal Register, December 18, 1978.
201
-------
10. Environmental Protection Agency, Solid Waste Disposal Facilities -
Proposed Classification Criteria, Federal Register, February 6, 1978.
11. Environmental Protection Agency, Landfill Disposal of Solid Waste -
Proposed Guidelines, Federal Register, March 26, 1979.
12. Federal Register, 41 (29); February 11, 1976.
13. Environmental Protection Agency, National Interim Primary Drinking
Water Regulations, Federal Register, 40 FR 59565; December 24, 1975;
Environmental Reporter: 81; February 13, 1976.
14. Water Quality Criteria, Ecological Research Series; March 1973,
EPA-R3-73-033.
15. U. S. Drinking Water Standards 1962, U. S. Public Health Service
Publication 956; 1962.
16. Interim Standards for Drinking Water, 2nd Edition, Geneva: World
Health Organization, 1971.
17. EPA Effluent Guidelines and Standards for Steam Electric Power
Generation, Federal Register, 40-FR 23987; June 4, 1975; Environ-
mental Reporter, S-259; July 11, 1975.
18. Hangebrauck, R. P., Status of IERL-RTP Program to Develop Environ-
mental Assessment Methodology for Fossil Fuel Processes, working
document; February 1977.
19. Cleland, J. G., and G. L. Kingsbury, Multimedia Environmental Goals
for Environmental Assessment, Vol. I, November 1977,
EPA-600/7-77-136a.
20. Jonke, A. A., et al., Annual Report on a Development Program in Pres-
surized Fluidized-Bed Combustion, Argonne National Laboratories,
Argonne, IL, July 1976, ANL/ES-CEN-1016.
21. Studies of the Pressurized Fluidized-Bed Coal Combustion Process,
Office of Research and Development. Report to EPA, Exxon Research
and Engineering Co., Linden, NJ, September 1977, EPA-600/7-77-107,
NTIS PB 272-722.
202
-------
22. Multicell Fluidized Bed Boiler Design Construction and Test Program.
Interim Report No. 1, Pope, Evans and Robbins, Inc.; August 1974,
PER-570-74.
23. Wilson, J. S., and R. Rice, EPA-sponsored FBC run, Morgantown
Energy Research Center, Morgantown, WV; March 31, 1977.
24. SO- Absorption in Fluidized-Bed Combustion of Coal - Effect of Lime-
stone Particle Size. Report to EPRI, Babcock and Wilcox Company,
Alliance, OH, March 31, 1977, RP719-1.
25. Hall, A. M., Testing, Identification, and Evaluation of Commercial
and "Advanced Experimental" Materials and Coatings under Design
Conditions Simulating Fuel Power Cycle Combinations, Task II.
Monthly Technical Progress Report No. 12 to ERDA, Battelle - Columbus
Laboratories, Columbus, OH; June 6, 1977, FE-2325-12.
26. Energy Conversion from Coal Utilizing CPU-400 Technology, Combus-
tion Power Company, Contract No. E(49-18)-1536, March 1976.
27. Stone, R., and R. L. Kahle, Environmental Assessment of Solid
Residues from Fluidized-Bed Fuel Processing: Final report
to EPA, Ralph Stone and Co., Inc., Los Angeles, CA, June 1978,
EPA-600/7-78-107.
28. Proposed Test Methods for Leaching of Waste Materials, ASTM D19 1203,
June 1978.
29. Strategy for the Implementation of the Resource Conservation and
Recovery Act of 1976 (draft), U. S. Environmental Protection Agency;
December 5, 1977.
30. Boynton, B.S., Chemistry and Technology of Lime and Limestone.
New York: Interscience Publishers; 1966.
31. Physical Testing of Quick Lime, Hydrated Lime and Limestone, ASTM
C110-76. Annual Book of ASTM Standards, Part 13; 68-85; 1976.
32. Murray, J. B., et al., Shrinkages of High-Calcium Limestone during
Burning, J. Am. Ceram. Soc., 37 (7): 323-28; 1974.
203
-------
33. Standard Methods for the Examination of Water and Waste Water,
13th Edition. Washington, D.C.: American Public Health Association;
1974.
34. Sun, C. C., C. H. Peterson, and D. L. Keairns, Environmental Impact
of the Disposal of Processed and Unprocessed FBC Bed Material and
Carry-over, Proceedings of the Fifth International Conference on
Fluidized-Bed Combustion, Washington, B.C., December 12-14, 1977.
McLean, VA: The Mitre Corporation, 1978.
35. Van'Ness, R. P., Louisville Gas and Electric Co., Private
Communication, 1977.
36. Henzel, D., Dravo Lime Co., Private Communication, March 1978.
37. Leo, P. P., Aerospace Corporation, Private Communication, 1977.
38. Disposal of By-Products from Non-Regenerable Flue Gas Desulfuriza-
tion Systems. Second Progress Report to EPA, Aerospace Corporation,
May 1977, EPA-600/7-77-052.
39. Sludge Oxidation in Limestone FGD Scrubbers, EPA-IERL, Research
Triangle Park, NC, June 1977, NTIS PB 268-525.
40. Selmeczi, J. G., D. H. Marlin, and D. W. Kestner, Stabilization of
Sludge Slurries, Dravo Corporation, Pittsburgh, PA. U.S.
Patent 3,920,795, November 18, 1975.
41. Smith, L. M., et al., Technology for Using Sulfate Waste in High-
way Construction, Federal Highway Administration, Gilette Research
Institute, Rockville, MD., December 1975, FHWA-RD-76-31, NTIS
PB 254-815.
42. Dunstan, E. R. Jr, Performance of Lignite and Subbituminous
Fly Ash in Concrete; A Progress Report, Bureau of Reclamation,
Denver, CO, January 1976, REC-ERC-76-1, NTIS PB 253-010.
43. Brown, P. W., et al., Energy Conservation through the Facilitation
of Increased Blended Cement Use, Interim report to ERDA, Institute
for Applied Technology, National Sureau of Standards, Washington,
DC, July 1-Dec. 1, 1975, NBSIR 76-1008, NTIS PB 251-218.
204
-------
44. Bogue, R. H., The Chemistry of Portland Cement, 2nd Edition,
New York; Reinhold Publishing Corporation, 1955; p. 32.
45. ASTM Annual Book of Standards, Part 10, Concrete and Mineral Aggre-
gates, Philadelphia: American Society of Testing and Materials;
1973.
46. Bogue, R. H., Op. cit., pp. 608-610.
47. Bogue, R. H., <>£• £±L» p. 672.
205
-------
APPENDIX A
SIEVE ANALYSIS USING SINGLE SIEVES UNDER VACUUM
A single sieve (150 ym opening size) preweighed to 0.1 g was placed
in an Alpine Model 200 Air-Jet Sieve (manufactured by Alpine American
Corp., Natick, MA). The sample (20 g) was placed on the sieve and the
whole was covered by a transparent cover. The equipment was started and
the sieving under vacuum allowed to proceed to 3 min. The vacuum was at
least 10 in. water pressure (below atmosphere). The amount of powder
remaining on the sieve was obtained by subtracting the weight of the
empty sieve from the gross weight. The above procedure was repeated
with sieves of decreasing opening size until an analysis had been obtained
for a 32 pm screen. For smaller opening sizes, the special adaptors
supplied with the machine had to be used, as did the special etched
nickel sieves. For the 20 and 10 ym opening sizes, the sample weight
was approximately 2 g weighed to an accuracy of 0.001 g, and, after cover-
ing the assembly with a transparent cover, the sieving time under vacuum
was 9 min. For the 5 ym opening size, the sample weight used was approxi-
mately 1 g weighed to an accuracy of 0.001 g, and the sieving time under
vacuum was 9 min. In all these cases the vacuum was at least 25 in. of
water below atmosphere. With all of the smaller sieves and several of
the larger it was found necessary to tap the cover often with the small
rubber mallet supplied. The weights of powder remaining on each screen
were calculated and the results plotted as "cumulative weight percent
finer than" versus "equivalent spherical diameter (microns)."
206
-------
APPENDIX B
ISOSTATIC PRESSING PROCEDURE
1. Insert the blank end cap into one end of a clean, dry mold and
seal with waterproof vinyl tape.
2. Place the mold assembly on the vibratory table and secure in
place. Set vibration to a low level and slowly add the sorbent
material to be pressed. Filling is accomplished by layering
and rodding until each layer offers resistance to the tamping
rod. Layering and tamping ensures better composition and
release of entrapped air from the sorbent.
3. Insert the end cap with the pressure vent open. This procedure
releases air being compressed by the end cap during insertion.
4. Tighten the pressure relief screw and tape as in 1.
5. Place the prepared mold in the isostatic pressure chamber, set
the pressure to 138 MPag (20,000 psig), and hold for 60 seconds.
(Refer to Autoclave Engineering operation manual).
6. Release the pressure gradually, remove the mold assembly from
the pressure chamber, and wipe dry.'
7. To remove the pressed specimen from the mold assembly, remove
the pressure relief screw from the end cap, untape and remove
the end cap. Carefully remove the pressed specimen.
8. After the pressed specimen has been removed from the mold assem-
bly, place it into a water bath at 25°C for 24 hours before cut-
ting it to the desired length with a diamond cut-off wheel.
9. Return the specimens to the water bath for 12 additional days
for completion of the curing cycle.
207
-------
Notes:
1. The isostatic mold assemblies must be clean and free of
moisture. Water will react with the sorbent and cause early
curing before pressing.
2. Air must be vented from the mold through the hole in the end
cap while the filled mold assembly is capped. This ensures that
no positive pressure is exerted on the inside of the mold before
pressing. Failure of the-mold will result.
3. Both end caps must be wrapped with waterproof vinyl tape. This
prevents the isostatic pressing fluid from entering the mold
assembly during pressing and contaminating the isostatic press
with spent sorbent.
208
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-80-015C
2.
3. RECIPIENT'S ACCESSION NO.
TITLE ANDSUBTITLE Experimental/Engineering Support for
EPA's FBC Program: Final Report
Volume 3. Solid Residue Study
. REPORT DATE
January 1980
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
C.C.Sun, C.H.Peterson, and D. L.Keairns
. PERFORMING ORGANIZATION NAME AND ADDRESS
Westinghouse Research and Development Center
1310 Beulah Road
Pittsburgh, Pennsylvania 15235
10. PROGRAM ELEMENT NO.
INE825
11. CONTRACT/GRANT NO.
68-02-2132
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 1/77 - 12/78
14. SPONSORING AGENCY CODE
EPA/600/13
is.SUPPLEMENTARY NOTES IERL-RTP project officer is D. Bruce Henschel, Mail Drop 61,
919/541-2825. EPA-600/7-78-049 also relates to this work.
16. ABSTRACT
The report gives results of experimental studies of the leaching properties
and thermal activity of solid residues from the fluidized-bed combustion (FBC) of
coal. Means for processing the residues, to reduce the environmental impact of
their disposal, are also studied. Previous leaching and thermal activity tests were
expanded to include residues from additional experimental FBC units, including both
atmospheric and pressurized systems. Leaching tests were conducted on both un-
treated residues and residue samples processed into a cement-like material.
Results indicate that the major potential contaminants in the leachate from the FBC
residues are the high pH, total dissolved solids, and sulfate levels , all of which are
above drinking water regulations. Heat release when initially exposed to water, due
to calcium oxide hydration, may also create a handling problem. Trace metals in
the leachate do not generally exceed drinking water regulations. Processing of the
residues caused some reduction in leachate contaminant levels. Additional tests on
processing and utilizing FBC residues are also reported.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
D.IDENTIFIERS/OPEN ENDED TERMS
COSATi Field/Group
Pollution Metals
Combustion Waste Treatment
Fluidized Bed Processing
Coal
Residues
Leaching
Pollution Control
Stationary Sources
Thermal Activity
Trace Metals
13B
2 IB
13H,07A
21D
07D
07B
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
229
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
209
------- |