Characterization of Mercury Enriched Coal Combustion Residues from Electric Utilities Using Enhanced Sorbents for Mercury Control


&EPA
United States
Environmental Protection
Agency	
EPA-600/R-06/008
 February 2006
       Characterization of
       Mercury-Enriched Coal
       Combustion Residues
       from Electric Utilities
       Using Enhanced Sorbents
       for Mercury Control

-------
                                                EPA-600/R-06/008
                                                  February 2006
  Characterization of Mercury-Enriched
Coal Combustion Residues from Electric
  Utilities Using Enhanced Sorbents for
                  Mercury Control

                            by

             F. Sanchez1, R. Keeney2, D. Kosson1, and R. Delapp1
         1Vanderbilt University                 2ARCADIS G&M, Inc.
Department of Civil and Environmental Engineering     4915 Prospectus Drive, Suite F
         Nashville, TN 37235                   Durham, NC 27713
              Contract No. EP-C-04-023, Work Assignment 1-31
                 EPA Project Officer: Ms. Susan Thorneloe
                Office of Research and Development (ORD)
           National Risk Management Research Laboratory (NRMRL)
             Air Pollution  Prevention and Control Division (APPCD)
                 Research Triangle Park, North Carolina.
                  U.S. Environmental Protection Agency
                  Office of Research and Development
                      Washingtion, DC 20460

-------
Characterization of Coal Combustion Residues
Abstract
This report evaluates changes that may occur to coal-fired power plant air pollution control residues from the use
of activated carbon and other enhanced sorbents for reducing air emissions of mercury and evaluates the poten-
tial for captured pollutants leaching during the disposal or use of these residues. Leaching of mercury, arsenic,
and selenium during land disposal or beneficial use of coal combustion residues (CCRs) is the environmental
impact pathway evaluated in this report. Coal combustion residues refer collectively to fly ash and other air
pollution control solid residues generated during the combustion of coal collected through the associated air
pollution control system. This research is part of an on-going effort by U.S. Environmental Protection Agency
(EPA) to use a holistic approach to account for the fate of mercury and other metals in coal throughout the life-
cycle stages of CCR management.

Air pollution control residues were obtained from coal combustion electric utility facilities with a representative
range of facility configurations (including air pollution controls) and coal types combusted. Each of the residues
sampled has been analyzed for selected physical properties, and for total content and leaching characteristics.
Results of laboratory leaching tests were used to develop estimates of constituent release under field manage-
ment scenarios. Laboratory leaching test results also were compared to field observations of leaching.

This report focuses on facilities that use injected sorbents for mercury control. It includes four facilities with
activated carbon injection (ACI) and two facilities using brominated ACL Fly ash has been obtained from each
facility with and without operation of the sorbent injection technology for mercury control. Each fly ash sampled
was evaluated in the laboratory for leaching as a function of pH and liquid-to-solid ratio. Mercury, arsenic and
selenium were the primary constituents of interest; results for these elements are presented here.

-------
                                                Characterization of Coal Combustion Residues
Foreword
   The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's
land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to
formulate and implement actions leading to a compatible balance between human activities and the
ability of natural systems to support and nurture life. To meet this mandate, EPA's research program is
providing data and technical support for solving environmental problems today and building a science
knowledge base necessary to manage our ecological resources wisely, understand how pollutants
affect our health, and prevent or reduce environmental risks in the future.

   The National Risk Management Research Laboratory (NRMRL) is  the Agency's center for
investigation of technological and management approaches for preventing and reducing  risks from
pollution that threaten human  health  and the environment. The focus of the Laboratory's research
program is on methods and their cost-effectiveness for prevention and control of pollution to air, land,
water, and subsurface resources; protection of water quality in public water systems; remediation of
contaminated sites, sediments and ground water; prevention and control of indoor air pollution; and
restoration of ecosystems. NRMRL collaborates with both public and private sector partners to foster
technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRLs research
provides solutions to environmental problems by: developing and promoting technologies that protect
and improve the environment; advancing  scientific and engineering information to support regulatory
and  policy  decisions; and providing the technical support and information transfer to ensure
implementation of environmental regulations and strategies at the national, state,  and community levels.

   This publication has been produced as part of the Laboratory's strategic long-term research plan.
It is published and made available by EPA's Office of Research and Development to assist the user
community and to link researchers with their clients.
                                       Sally Gutierrez, Director
                                       National Risk Management Research Laboratory
                                           in

-------
Characterization of Coal Combustion Residues
EPA Review Notice
The U.S. Environmental Protection Agency through its Office of Research and Development funded and man-
aged the research described here under Contract EP-C-04-023 to ARCADIS Geraghty & Miller, Inc. It has been
subjected to Agency review and has been approved for publication as an EPA document. Mention of trade names
or commercial products does not constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Information Service, Springfield, Vir-
ginia 22161.
                                             IV

-------
                                                    Characterization of Coal Combustion Residues
Table of Contents
Section                                                                                   Page
Abstract	ii
Foreword	iii
EPA Review Notice	iv
List of Figures	vii
List of Tables	viii
Glossary of Terms	ix
Acknowledgments	xi
Executive Summary	xii

1  Introduction	1
    1.1 Regulatory Context	2
       1.1.1 Waste Management	2
       1.1.2 Air Pollution Control	3
    1.2 Configurations of U.S. Coal Fired Power Plants and Multi-pollutant Control Technologies	3
       1.2.1 Current Air Pollution Control Technologies	5
       1.2.2 Enhancement of Controls for Mercury Removal: Sorbent Injection	6
       1.2.3 Mercury Control by Conventional PAC Injection	8
       1.2.4 Mercury Control by Halogenated PAC Injection	8
    1.3 Coal Combustion Residues	9
    1.4 Residue Management Practices	9
       1.4.1 Beneficial Use 	10
       1.4.2 Land Disposal	11
    1.5 Leaching Protocol	11
2  Materials and Methods	15
   2.1 CCR Materials for Evaluation	15
       2.1.1 Reference Fly Ash	15
       2.1.2 Facilities Using Injection of Standard Activated Carbon	15
               .2.1 Brayton Point	15
               .2.2 Pleasant Prairie	19
               .2.3 Salem Harbor	20
               .2.4 Facility C	20
       2.1.3 Facilities Using Injection of Brominated Activated Carbon	20
            2. .3.1 St. Clair	20
            2. .3.2 Facility L	20
    2.2 Leaching Assessment Protocols	21
       2.2.1 Alkalinity, Solubility and Release as a Function of pH(SR002.1)	21
       2.2.2 Solubility and Release as a Function of LS Ratio (SR003.1)	21
    2.3 Analytical Methods 	21
       2.3.1 Surface Area and Pore Size  Distribution	21
       2.3.2 pHand Conductivity	22
       2.3.3 Moisture Content	22
       2.3.4 Carbon Content: Organic Carbon/Elemental Carbon Analyzer	22

-------
Characterization of Coal Combustion Residues
Table of Contents (concluded)
Section                                                                                Page
       2.3.5 Mercury (CVAA, Method 3052, and Method 7473)	22
       2.3.6 Other Metals (ICP-MS, Method 3052 and Method 6020)	22
            2.3.6.1ICP-MS Analysis at Vanderbilt	23
            2.3.6.2 Severn Trent Laboratories, Inc. (STL)	23
       2.3.7 X-Ray Fluorescence (XRF)	23
       2.3.8 MDL and ML for Analytical Results	23
   2.4 Quality Assurance Validation	24
       2.4.1 Homogenization of Individual CCR Samples and Aliquots for Analyses	24
       2.4.2 Leaching Test Methods and Analytical QA/QC	24
       2.4.3 Laboratory Mass Balance Verification for Leaching Test Methods	24
       2.4.4 Improving QA/QC Efficiency	27
   2.5 Interpretation and Presentation of Laboratory Leaching Data	27
   2.6 Long-Term Release Assessment	29
3  Results and Discussion	35
   3.1 Leaching Characteristics from Field Observations of CCR Landfills and Impoundments	35
   3.2 Quality Assurance for Laboratory Leaching Tests	39
       3.2.1 Mass Balance using EPAReference Fly Ash	39
       3.2.2 Analytical Quality Control/Quality Assurance	39
   3.3 Laboratory Test Results	40
       3.1.1 Mercury Results	43
       3.3.2 Arsenic Results	45
       3.3.3 Selenium Results	47
   3.4 Long-Term Release Assessment	48
       3.3.4 Long-Term Release Estimates for Mercury	49
       3.3.4 Long-Term Release Estimates for Arsenic	50
       3.3.4 Long-term Release Estimates for Selenium 	50
4  Conclusions and Recommendations 	55
   4.1 Assessment of CCRs Without and With Activated Carbon Injection	55
   4.2 Implementation of Leaching Test Methods	56
5  References	57

Appendix A U.S. EPA Science Advisory Board Consultation Summary	59
Appendix B DOE NETL Full-Scale Test Site Flow Diagrams	73
Appendix C Quality Assurance Project Plan	79
Appendix D Brayton Point Fly Ashes	116
Appendix E Pleasant Prairie Fly Ashes	131
Appendix F Salem  Harbor Fly Ashes	146
                                              VI

-------
                                                   Characterization of Coal Combustion Residues
List of Figures
Figure                                                                                  Page
1   Coal-Fired Boiler with Sorbent Injection and Spray Cooling	7
2   Flow Diagram for Power Plant with a Hot ESP, Carbon Injection, and aCOHPAC	7
3   Life-Cycle Evaluation of Coal Combustion Residues	9
4   Uses of CCRs Based on 2003 Industry Statistics	10
5   Mixing Fly Ash Priorto Obtaining Aliquots for Laboratory Analyses	24
6   Flow Diagram for Mass Balance and Quality Control on Laboratory Leaching Procedures	26
7   Example of Extract Concentrations as a Function of pH from SR002.1 	28
8   Leachate pH Distribution: Scenario of Disposal in a Combustion Waste Landfill	30
9   LS Ratio Distribution: Scenario of Disposal in a Combustion Waste Landfill	31
10  Example of Variability of Hg Concentrations as a Function of pH from SR002.1	31
11  Example of Regression Fits and Corresponding Equations for Solubility as a Function of pH	32
12  Example of Cumulative Probability Distribution for Release of Selenium from Brayton Point CCR ....32
13  Example of Comparison of 100 Yr Cumulative Release Estimates for Arsenic	33
14  Range of pH Observed in Field Leachate at Landfills and Impoundments Used for
    Disposal of CCRs	36
15  Arsenic Concentrations Observed in Field Leachate at Landfills and Impoundments Used for
    Disposal of CCRs	37
16  Selenium Concentrations Observed in Field Leachate at Landfills and Impoundments Used
    or Disposal of CCRs 	38
17  STL Versus Vanderbilt Analytical Results for Arsenic and Selenium from Sr002.1	41
18  Example Results from SR002.1	42
19  Example Results from SR002.1 and SR003.1 for Selenium in Fly Ashes from Salem Harbor
    with Enhanced Hg Control	43
20  Comparison of Total Mercury Content in Baseline Cases and with Sorbent Injection for
    CCRs from Different Facilities	44
21  Ranges of Laboratory and Field Leachate Mercury Concentrations Compared with the
    Drinking Water Maximum Contaminant Level	45
22  Comparison of Total Arsenic Content in Baseline Cases and with Sorbent Injection for
    CCRs from Different Facilities	46
23  Ranges of Laboratory and Field Leachate Arsenic  Concentrations Compared with the
    Drinking Water Maximum Contaminant Level	47
24  Comparison of Total Selenium Content in Baseline Cases and with Sorbent Injection for
    CCRs from Different Facilities	48
25  Ranges of Laboratory and Field Leachate Selenium Concentrations Compared with the
    Drinking Water Maximum Contaminant Level	49
26  Example Regression Curves of Experimental Data of Arsenic Solubility as a Function
    of pH for Brayton Point	49
27  Example 100-Year Arsenic Release Estimates for Brayton Point as a Function of the
    Cumulative Probability for the Scenario of Disposal in a Combustion Waste Landfill	50
28  Upper Bound of 100 yr Mercury Release Estimates for Landfill Scenario Without and
    With Activated Carbon Injection	51
29  Upper Bound of 100 yr Arsenic Release Estimates for Landfill Scenario Without and
    With Activated Carbon Injection	52
30  Upper Bound of 100 yr Selenium Release Estimates for Landfill Scenario Without and
    With Activated Carbon Injection	53
                                              vii

-------
Characterization of Coal Combustion Residues
List of Tables
Table                                                                                    Page

1   List of CCRs collected to date for evaluation of under this research program	4
2   General Characteristics of Coals Burned in U.S. Power Plants	5
3   Projected Coal-Fired Capacity by APC Configuration	6
4   Beneficial Uses of CCRs	11
5   Characteristics of Coal Combusted and Facilities Sampled and Reported Here	16
6   Fly Ashes from Brayton Point, Pleasant Prairie, Salem Harbor, and Facility C: Elemental
    Composition (by x-ray fluorescence) and Other Characteristics	16
7   Fly Ashes from St. Clair and Facility L: Elemental Composition (by x-ray fluorescence)
    and Other Characteristics	17
8   CCRs from Facilities with Electrostatic Precipitators: Total Content of Mercury, Arsenic,
    Cadmium, Lead and Selenium	18
9   Detection Limits and Quality Control Information for ICP-MS Analysis for As, Pb, and
    SeatVanderbilt	23
10  Total Content Analysis Results for Reference Fly Ash after Mixing	25
11  Composition of Combustion Waste Landfill Leachate	28
12  Distribution of pH and Concentrations of Arsenic and Selenium from Field CCR
    Management Facilities from the U.S. EPADatabase	35
13  Distribution of pH and Concentrations of Arsenic, Selenium, and Mercury from Field
    CCR Management Facilities	39
14  Leaching (Method SR002.1) and Mass Balance Results for the EPA-Reference Fly Ash	40
                                              Vlll

-------
                                                   Characterization of Coal Combustion Residues
Glossary of Terms
Term         Definition
ACAA     American Coal Ash Association
ACI       activated carbon injection
APC       air pollution control
APPCD    Air Pollution Prevention and Control Division
ASTM     American Society for Testing and Materials
BET       Brunauer, Emmett, and Teller (method for quantification of surface area)
B-PAC     Brominated Powdered Activated Carbon (product name from Sorbent Technologies Corp,
           Twinsburg, OH)
BPB       Brayton Point baseline
BPT       Brayton Point test (with ACI)
BML      below method limit
CAIR      Clean Air Interstate Rule
CAMR     Clean Air Mercury Rule
CCRs      coal combustion residues
CCV       continuous  calibration verification
COHPAC   Compact Hybrid Particulate Collector
CS-ESP    cold-side electrostatic precipitator
CVAA     cold vapor atomic adsorption
DAFs      dilution and attenuation factors
DOE       U.S. Department of Energy
DI         deionized (i.e., deionized water)
DRC       dynamic reaction chamber
EPA       U.S. Environmental Protection Agency
EPRI      Electric Power Research Institute
ESP       electrostatic precipitator
FF         fabric filter (baghouse)
FGD       flue gas desulfurization
FID       flame ionization detector
GAB       Facility C baseline
GAT       Facility C test (with ACI)
HC1       hydrogen chloride
Hg°       elemental mercury
Hg2+       oxidized or ionic mercury
HS-ESP    hot-side electrostatic precipitator
ICP-MS    inductively coupled plasma-mass spectroscopy
ICV       internal calibration verification
JAB       St. Clair facility baseline
JAT       St. Clair facility test (with B-PAC)
kd         CCR linear partition coefficient
                                              ix

-------
Characterization of Coal Combustion Residues
Glossary  of Terms (concluded)
Term         Definition
LOI       loss on ignition
LS        liquid-to-solid (LS ratio)
MC       mechanical collector
MCL      maximum concentration limit (for drinking water)
MDL      method detection limit
ML       minimum quantification limit
NETL     National Energy Technology Laboratory
NOX      nitrogen oxides
NSPS     New Source Performance Standard
OC/EC    organic carbon/elemental carbon
ORD      EPA's Office of Research and Development
OSW      EPA's Office of Solid Waste
PAC      powdered activated carbon
PJFF      pulse jet fabric filter
PM       particulate matter
PPB       Pleasant Prairie baseline
PPT       Pleasant Prairie test (with ACI)
PRB      sub-bituminous coal mined in Wyoming's Powder River Basin
QA/QC    quality assurance/quality control
RCRA    Resource Conservation and Recovery Act
RFA      reference fly ash
PS        particulate scrubber
SAB      EPA's Science Advisory Board
SCA      specific collection area
SCR      selective catalytic reduction
SDA      spray dryer absorber
SGLP     synthetic ground leaching procedure
SHE      Salem Harbor baseline
SHT      Salem Harbor test (with ACI)
SNCR    selective non-catalytic reduction
SOX       oxides of sulfur
SPLS      synthetic precipitation leaching procedure
S/S       stabilization/solidification
SWDA    Solid Waste Disposal Act
TCLP     toxicity characteristic leaching procedure
TOXECON toxic emissions control
WS       wet scrubber
XRF      X-ray fluorescence

-------
                                                  Characterization of Coal Combustion Residues
Acknowledgments
Extensive input on the research program design was provided by G. Helms, U.S. EPA, Office of Solid Waste
(Washington, DC). Additional technical support was provided by B. Ghorishi, formerly of ARCADIS, and J.
Kilgroe, formerly of U.S. EPA, National Risk Management Research Laboratory.

Laboratory testing described herein was carried out by ARCADIS with technical support from Vanderbilt Uni-
versity. R. Delapp was responsible for the chemical analyses carried out at Vanderbilt University. Technical
assistance also was provided by A. Garrabrants and V. Laohom.

K. Ladwig and EPRI are gratefully acknowledged for assistance in obtaining coal combustion residue samples
and providing information from the EPRI database on coal combustion residues.

C. Senior also is gratefully acknowledged for her assistance in obtaining coal combustion residue samples and
related background information for this project.
                                              XI

-------
Characterization of Coal Combustion Residues
Executive Summary
This report evaluates changes that may occur to coal-fired power plant air pollution control residues from the use
of activated carbon and other enhanced sorbents for reducing air emissions of mercury and evaluates the poten-
tial for captured pollutants leaching during the disposal or use of these residues. Leaching of mercury, arsenic,
and selenium during land disposal or beneficial use of coal combustion residues (CCRs) is the environmental
impact pathway evaluated in this report. Coal combustion residues refer collectively to fly ash and other air
pollution control solid residues generated during the combustion of coal collected through the associated air
pollution control system. This research is part of an on-going effort by U.S. Environmental Protection Agency
(EPA) to use a holistic approach to account for the fate of mercury and other metals in coal throughout the life-
cycle stages of CCR management.

The specific objectives of the research reported here are to:
1. Evaluate the potential for leaching to groundwater of mercury, arsenic, and selenium removed from coal-
fired power plant air emissions by air pollution control technology and, as a result, are contained in CCRs;
2. Provide the foundation for assessing the impact of enhanced mercury and multi-pollutant control technol-
ogy on leaching of mercury and other constituents of potential concern from CCRs during the lifecycle of
CCR management, including storage, beneficial use, and disposal; and
3. Perform these assessments using the most appropriate evaluation methods currently available. This in-
volved use of a laboratory leach testing approach developed by Kosson, et al. (2002), which considers the
effects of varying environmental conditions on waste constituent leaching. Effective use of this approach
required technology transfer to the U.S. EPANational Risk Management Laboratory, and development of
a quality assurance/quality control (QA/QC) framework to help evaluate and verify test results.

This testing approach was chosen for use because it evaluates leaching over a range of values for two key
variables (pH and liquid: solid ratio) that both vary in the environment and affect the rate of constituent release
from waste. The range of values used in the laboratory testing encompasses the range of values expected to be
found in the environment for these parameters. Because the effect of these variables on leaching is evaluated in
the laboratory, prediction of leaching from the waste in the field is expected to be done with much greater
reliability.
xii

-------
Characterization of Coal Combustion Residues
Application of this leach testing approach to mercury leaching involved an extensive QA/QC program. This
included conducting a mass balance of mercury and other metals using a reference fly ash to ensure that unac-
ceptable losses of mercury in lab equipment or glassware or due to volatilization did not occur. If mercury was
not found to be leaching, it was critical to ensure that this result was real and not due to mercury losses during
handing or processing of the samples.

In addition, results from laboratory leaching evaluation were compared to field leachate concentrations from
CCRmanagement facilities available from a U.S. EPA database and an Electric Power Research Institute (EPRI)
database to determine if laboratory testing results reasonably represented field observations.

Constituent release estimates projected to occur over a 100 year period were developed to evaluate the potential
cumulative impacts of different CCRmanagement scenarios. Disposal in a combustion waste landfill was exam-
ined using Monte Carlo simulation based on reported field conditions and for three hypothetical default landfill
scenarios.

Summary of Conclusions

Assessment ofCCRs with and without Use of Activated Carbon Injection and Brominated Carbon Injection.

Analysis has been completed for CCRs from four coal combustion facilities using powdered activated carbon
injection and from two facilities using brominated powdered activated carbon injection to control mercury emis-
sions. For each facility, the evaluation included assessments of CCRs generated both with and without use of the
activated carbon injection. None of these facilities had scrubbers as part of their air pollution control technology.
The following conclusions are drawn for this class of facilities:
• Application of activated carbon injection substantially increased the total mercury content in the resulting
CCRs for five of the six facilities evaluated. Substantially increased arsenic and selenium content in the
CCRs was observed at the one facility that employed compact hybrid particulate collector (COHPAC1)
fabric filter particulate control technology. This may have resulted from additional arsenic and selenium
adsorption onto the CCR while retained in the fabric filters. Significant increase in the selenium content of
one additional facility was noted.
• Mercury is strongly retained by the CCR and unlikely to be leached at levels of environmental concern.
Leaching that did occur did not depend on total mercury content in the CCR, leaching pH, or liquid to solid
ratio, and mercury concentrations in laboratory extracts appeared to be controlled by non-linear adsorp-
tion equilibrium. Laboratory extract concentrations ranged from less than the minimul detection level
(0.01|ig/L)to0.2|ig/L.
• Arsenic and selenium may be leached at levels of potential concern from CCRs generated at some facili-
ties both with and without enhanced mercury control technology. Further evaluation of leaching of arsenic
and selenium from CCRs that considers site specific conditions is warranted.
• Leachate concentrations and the potential release of mercury, arsenic and selenium do not correlate with
total content. For many cases, leachate concentrations observed are a function of final pH over the range of
field conditions, and the observed leaching behavior implies that solubility in the leachate or aqueous
extract controls observed liquid concentration rather than linear adsorption equilibrium. For these cases,
use of linear partition coefficients (K^ in modeling leaching phenomena does not reflect the underlying
processes. In addition, for many cases, the amount of mercury, arsenic, and selenium estimated to be
released over a 100 year interval is a small fraction (< 0.1% - 5%) of the total content. For selenium,
release from less than 5% up to the total content of selenium can be anticipated over the 100 year period.
1 For the COHPAC air pollution control configuration, combustion gasses pass through an electrostatic precipitator; then
activated carbon is injected into the gas stream before it passes through a fabric filter for particulate collection.
Xlll

-------
Characterization of Coal Combustion Residues
      Therefore, it is not recommended to base landfill management decisions on total content of constituents in
      CCRs since total content does not consistently relate to quantity released.
    •  Results of this assessment also suggest management conditions (e.g., by control of infiltration and pH)
      that may result in reduction releases of arsenic and selenium by as much as two orders of magnitude in
      comparison to upper bound estimated releases.
    •  Use of the leaching framework facilitated understanding the variations in anticipated leaching behavior
      under the anticipated field landfill disposal conditions, including expected ranges of constituent concen-
      trations in leachate and cumulative release over a defined time  interval. In  addition, insights into the
      mechanisms controlling constituent leaching were obtained. This depth of understanding would not have
      been possible using leaching tests focused on a single extraction condition (e.g., toxicity characteristic
      leaching procedure, synthetic precipitation leaching procedure, or synthetic  groundwater leaching proce-
      dure).
    •  This study provides baseline data which allows using a reduced set of laboratory testing conditions as a
      screening leaching assessment for CCRs from coal combustion facilities employing similar air pollution
      control technology. For mercury, extraction only at the material's natural pH at a liquid-to-solid ratio (LS)
      of 10 is adequate. For arsenic,  extraction at four conditions is warranted to  define the range of expected
      leachate concentrations and release: (i) pH 5.5-6.0 at LS=10, (ii) pH 7.5-8.5 at LS=10, (iii) pH 12.0-12.5
      at LS=10 and (iv) natural pH at LS=2. For selenium, either the total content or the same conditions as
      recommended for arsenic can  be used. At least duplicate extractions should be used. Results from this
      more limited testing can be evaluated in comparison with the results presented in this report to determine
      if more extensive evaluation is warranted.

Implementation of Leaching Test Methods

The leaching assessment approach published by Kosson et al. (2002) and implemented in this report was se-
lected because, after internal EPA review (Office of Research and Development, Office of Solid Waste) and
consultation with the Environmental Engineering committee of the EPA Science Advisory  Board, it was consid-
ered the only available, peer reviewed, and published approach that allowed consideration of the range of poten-
tial field management scenarios expected for CCRs and provided a fundamental foundation for extrapolation of
laboratory testing to field scenarios. Additional development and validation of the leaching assessment approach
through this project provides the following conclusions:
    •  Laboratory leaching test results were consistent with observations of ranges of field leachate pH and
      mercury, arsenic, and selenium concentrations. Thus, the leaching test methods employed in this study
      provide an appropriate basis for evaluating leaching  under the range  of anticipated field management
      scenarios.
    •  Leaching test methods SR002.1 (Solubility and Release as a Function of pH) and SR003.1 (Solubility and
      Release as a Function of LS ratio) have been successfully implemented at the EPA National Risk Manage-
      ment Research Laboratory. The use of these methods is now considered near routine methodology for the
      laboratory.
    •  QA/QC methodology conforming with EPA Category 3 requirements has  been developed and demon-
      strated for the leaching test methods SR002.1 and SR003.1.
    •  Further efficiency in implementation of the QA/QC methodology may be obtained, based on the results
      from testing the initial set of CCRs, by reducing the number of replicates and control analyses required
      under the initial QA/QC plan. These improved project efficiencies are being implemented for evaluation
      of additional CCRs under this project.
    •  A mass balance around the laboratory leaching test procedures has been completed for mercury and se-
      lected metals of potential concern. These results indicate that recoveries were  between 60% and 91% for
      mercury during the leaching tests and subsequent analytical procedures, which is within the uncertainty
      resulting from heterogeneity within the CCR. Additional mass balance verification may be warranted if
      future samples have significantly different characteristics that may result in greater volatility of the con-
      stituents of interest than in the  reference sample evaluated.
                                                 xiv

-------
                                                     Characterization of Coal Combustion Residues
This is the first of a series of reports that will address the potential for leaching of constituents of potential
concern from CCRs. Subsequent reports will address:
    •  CCRs from coal-fired power plants that use SO2 scrubbers as a part of their air pollution control technol-
     ogy
    •  CCRs from coal-fired power plants that use air pollution control technologies other than evaluated in
     earlier reports necessary to span the range of coal-types and air pollution configurations.
    •  Assessment of leaching for constituents of potential concern under additional management scenarios,
     including impoundments and beneficial use.
    •  Broader correlation of CCR leaching characteristics to coal type, combustion facility characteristics and
     geochemical speciation within CCRs supported by information and analysis on additional trace elements
     and primary constituents.
                                                xv

-------
Characterization of Coal Combustion Residues
                                           xvi

-------
Characterization of Coal Combustion Residues
1. Introduction
Congress has directed EPA to document that the Clean Air
Act regulations are not trading one environmental burden
for another. The Air Pollution Prevention and Control Di-
vision (APPCD) of EPA's Office of Research and Devel-
opment (ORD) is conducting the research to help identify
any potential leaching with unacceptable impacts result-
ing from land disposal or beneficial use of mercury-en-
riched CCRs. The research results presented in this report
are part of that effort.

The focus of this report is to present an evaluation of
changes to air pollution control residues that may result
from the use of sorbent injection as enhanced mercury con-
trol technology at coal-fired power plants and to document
the potential for transfer of pollutants from the resulting
residues to water resources or other environmental systems
(e.g., soils, sediments). The residues studied for this report
were fly ashes generated using activated carbon and other
enhanced sorbents for reducing air emissions of mercury
from the power plants and fly ashes from the same power
plants without use of sorbent injection. The potential for
leaching of mercury and other constituents of potential
concern during land disposal or beneficial use of the coal
combustion residues (CCRs) is the more narrow focus of
this assessment. This research is part of an on-going effort
by EPA to use a holistic approach to account for the fate of
mercury and other metals in coal throughout the life-cycle
stages of CCR management.

CCRs include bottom ash, boiler slag, fly ash, scrubber
sludge, and other miscellaneous solids generated during
the combustion of coal. The boiler slag and bottom ash are
not of interest in this study because enhanced mercury
emission controls will not change their composition. Fly
ash characteristics are expected to change from implemen-
tation of enhanced mercury emission controls. Fly ash in-
cludes unburned and inorganic materials in coal that do
not burn, such as oxides of silicon, aluminum, iron, and
calcium. Fly ash is light enough to be entrained in the flue
gas stream and captured in the air pollution control equip-
ment. Scrubber sludge is the by-product of the SO2 wet
scrubbing process resulting from neutralization of acid
gases. Air pollution control can concentrate or partition
metals to fly ash and scrubber sludge.

When coal is burned in an electric utility boiler, the result-
ing high combustion temperatures vaporize the Hg in the
coal to form gaseous elemental mercury (Hg°). Subsequent
cooling of the combustion gases and interaction of the gas-
eous Hg° with other combustion products result in a por-
tion of the Hg being converted to gaseous oxidized forms
of mercury (Hg2+) and particle-bound mercury. The spe-
cific chemical form—known as the speciation—has a
strong impact on the capture of mercury and other metals
by boiler air pollution control (APC) equipment.

Sorbents, typically finely ground powdered activated car-
bon, can be used to capture mercury. The sorbent is typi-
cally injected upstream of the particulate control device,
where both the sorbent and adsorbed mercury are collected.
Depending upon the type of sorbent, gas conditioning, and
other air pollution control technology in use, there may be
changes occurring to the fly ash that may affect the stabil-
ity and mobility of mercury and other metals in the CCRs.

In response to increasingly stricter requirements to reduce
air emissions of mercury and other pollutants from coal-
fired power plants, additional advanced air pollution con-
trol technology is being introduced. Activated carbon in-
jection (ACI) or brominated powdered activated carbon
injection (B-PAC) are two technologies being considered
for widespread use. This research is evaluating changes to
air pollution control residues as a result of these technolo-
gies, and the impacts of land disposal or commercial use
of the residues.

The specific objectives of the research reported here are
to:
1 Evaluate the potential for leaching to groundwater of
mercury, arsenic, and selenium removed from coal-
1

-------
Characterization of Coal Combustion Residues
    fired power plant air emissions by air pollution con-
    trol technology and, as a result, contained in CCRs;
  2 Provide the foundation for assessing the impact of en-
    hanced mercury and multi-pollutant control technol-
    ogy on leaching of mercury and other constituents of
    potential concern from CCRs during the lifecycle of
    CCR management, including storage, beneficial use
    and disposal; and
  3 Perform these assessments using the most appropriate
    evaluation methods currently available. This has in-
    volved use of a laboratory leach testing approach de-
    veloped by Kosson, et al. (2002), which considers the
    effects of varying environmental conditions on waste
    constituent leaching. Effective use of this approach
    required technology transfer to the U.S. EPA National
    Risk Management Laboratory and development of a
    quality assurance/quality control (QA/QC) framework
    to help evaluate and verify test results.
The approach to satisfying these objectives was to obtain
samples of air pollution control residues from a variety of
coal-fired power plants, each under operating conditions
with and without use of enhanced mercury control tech-
nology.

This report focuses on facilities that employ electrostatic
precipitators (ESPs) but do not have SO2 scrubbers for air
pollution control.  Enhanced mercury control technology
consisted of activated carbon injection for  four facilities
(Brayton Point, Pleasant Prairie, Salem Harbor, and Facil-
ity  C1) and injection of brominated activated carbon for
two facilities (St. Clair and Facility L1). This is the first of
a series of reports that will address the potential for cross-
media transfer of constituents of potential  concern from
CCRs. Subsequent reports will address:
  • CCRs from coal-fired power plants that use scrubbers
    as a part of their air pollution control technology (re-
    port 2);
  • CCRs from coal-fired power plants that use air pollu-
    tion control technologies other than evaluated in ear-
    lier reports necessary to span the range of coal-types
    and air pollution configurations (report 3);
  • Assessment of leaching for constituents of potential
    concern under additional management  scenarios, in-
    cluding impoundments and beneficial use on the land
    (report 4); and,
  • Broader correlation of CCR leaching characteristics
    to coal type, combustion facility characteristics and
    geochemical speciation within CCRs supported by in-
    formation and analysis on additional trace elements
    and primary constituents (report 4).

Table 1 provides a summary of facilities sampled to date
and indicates which of the reports will present an evalua-
tion of which samples.

Sampled CCRs were subjected to multiple leaching condi-
tions according to the designated leaching assessment ap-
proach. Leaching conditions included batch equilibrium2
extractions at acidic, neutral, and alkaline conditions at a
liquid-to-solid ratio  (LS)  of 10 mL/g, and LS ratio from
0.5 to 10 mL/g using distilled water as the  leachant. The
results of this testing are being used to evaluate the likely
range of leaching characteristics under a variety of CCR
management scenarios. Results of the laboratory leaching
tests were compared to the range of observed constituent
concentrations in field leachates reported in a U.S.  EPA
database (EPA, 2000) and an Electric Power Research In-
stitute (EPRI) database (EPRI, 2005). A probabilistic as-
sessment approach considered leaching estimates over the
observed range of field pH and LS from the U.S. EPA da-
tabase to develop  100 yr release estimates for constituents
of interest from the CCRs tested.

As part of this research program, a QA/QC plan consistent
with EPA requirements was developed for the leaching
assessment approach. The QA/QC methodology included
verification of acceptable mercury retention during labo-
ratory testing through evaluation of amass balance around
testing procedures. Laboratory testing for leaching assess-
ment was  carried out at the EPA National Risk Manage-
ment Laboratory (Research Triangle Park, North Carolina)
with technical assistance from Vanderbilt University.

1.1. Regulatory Context

1.1.1. Waste Management
Management of coal combustion residues is subject to the
Resource Conservation and Recovery Act (RCRA), which
is  the federal law regulating both solid and hazardous
wastes. Subtitle C under the RCRA pertains to hazardous
waste;  other solid, non-hazardous wastes fall under  Sub-
title D. Subtitle C wastes are federally regulated while  Sub-
title D wastes are regulated primarily at the state level. The
1 These facilities have requested to remain anonymous.
2
2 In the context of leaching tests, the term "equilibrium" is used
to indicate that the test method result is a reasonable approxima-
tion of chemical equilibrium conditions even though thermody-
namic equilibrium may not be approached for all constituents.

-------
Characterization of Coal Combustion Residues
original version of the RCRA did not specify whether CCRs
were Subtitle C or D wastes. In 1980, the Solid Waste Dis-
posal Act (SWDA) amendments to RCRA conditionally
excluded CCRs from Subtitle C regulation pending comple-
tion of a study of CCR hazards. Since that time, CCRs
have been regulated at the state level under Subtitle D.

The SWDA amendments to RCRA required EPA to pre-
pare a report to Congress identifying CCR hazards and rec-
ommending a regulatory approach for CCRs. In this report
(EPA, 1988), EPA recommended that CCRs generated by
electric utilities continue to be regulated under Subtitle D
(See 58 FR 42466, August 9, 1993).

Other residues generated at coal-fired electric utilities were
not included in this 1993 decision. EPA conducted a fol-
low-up study specifically aimed at low-volume, co-man-
aged wastes3 and issued another Report to Congress (EPA,
1999) with a similar recommendation. In April 2000, EPA
issued a regulatory determination exempting these wastes
from hazardous waste regulations (See 65 FR 32214, May
22, 2000). However, concern was expressed over the use
of CCR as backfill for mining operations, and it was de-
cided that this practice be regulated under a federal Sub-
title D rule. It was also decided by EPA that federal regula-
tions under Subtitle D are needed for CCR when they are
disposed in surface impoundments and landfills. Currently,
the Agency is in the process of developing these regula-
tions. The results presented in this report, and subsequent
reports, will help provide the information needed to iden-
tify the release potential of mercury and other metals that
have been removed from stack gases into air pollution con-
trol residues, over a range of plausible management op-
tions. These data will help identify those conditions that
will either reduce or enhance releases to the land so that
the effects of different management conditions can be fac-
tored into any controls developed under the regulations.

1.1.2. Air Pollution Control
On March 10, 2005, EPA announced the Clean Air Inter-
state Rule (CAIR; FR25162, May 2005), which is expected
to increase the use of wet scrubbers and selective catalytic
reduction (SCR) units to help reduce sulfur dioxide and
nitrogen oxides from coal-fired power plants. On March
15, 2005, EPA announced the Clean Air Mercury Rule
(CAMR; FR 28606, May 2005) for reducing mercury emis-
sions through the use of a cap and trade program. Power
plants are the largest remaining source of anthropogenic
3 Co-managed wastes are low-volume wastes that are co-man-
aged with the high-volume CCRs.
mercury emissions in the county. When fully implemented,
a reduction of 70% is projected to occur (from 48 tons to
15 tons annually).

The Clean Air Mercury Rule establishes "standards of per-
formance" that limit mercury emissions from new—
through new source performance standards (NSPS)—and
existing (through emission guidelines) coal-fired power
plants by creating a market-based cap-and-trade program
that will reduce mercury emissions in two phases. The first
phase caps national annual mercury emissions at 38 tons
through co-benefit reductions achieved by controlling sul-
fur dioxide (SO2) and nitrogen oxides (NOX) emissions
under CAIR. In the second phase, due in 2018, coal-fired
power plants will be subject to a second cap, which will
reduce emissions to 15 tons per year upon full implemen-
tation.

Congress has directed EPA to document that the Clean Air
Act regulations are not trading one environmental burden
for another. The Air Pollution Prevention and Control Di-
vision (APPCD) of EPA's Office of Research and Devel-
opment (ORD) is conducting the current research to help
identify any potential pollutant transfers resulting from land
disposal or beneficial use of mercury-enriched CCRs. The
research results presented in this report are part of that ef-
fort.

In response to the evolving implementation of advanced
air pollution control technology for coal-fired power plants,
this research is directed towards understanding changes in
CCR characteristics that may increase environmental bur-
dens from land disposal of CCRs or impact CCR usage in
commercial applications.

1.2. Configurations of U.S. Coal Fired
Power Plants and Multi-pollutant Con-
trol Technologies
The approximately 450 coal-fired electricity generating
facilities in the United States uses a range of coals and
plant configurations. The coal type burned and facility
design characteristics affect the effectiveness of various
mercury control methods that are or could be used at these
plants. The U.S. coal-fired power plants typically burn one
of three types of fuel: (1) bituminous coal (also referred to
as "high rank" coal), (2) subbituminous coal, and (3) and
lignite (subbituminous coal and lignite are referred to as
"low rank" coals). Some of the characteristics of interest
related to the possible environmental impacts of burning
these different coal types are given in Table 2 (EPA, 2005).

-------
Characterization of Coal Combustion Residues
Table 1. List of CCRs collected to date for evaluation of under this research program. [Some facilities are identified by
code letter only (e.g., "Facility C") to preserve the confidentiality of the CCR source.]
     Facility
Coal Type
                                         APC Configuration
                                            CCR/Ash Type
                                              (CCR pHa)
Report 1: ACI and B-PAC (this report)

Brayton Point     low-sulfur bituminous

Salem Harbor     low-sulfur bituminous

Pleasant  Prairie   PRBd subbituminous
Facility C

St. Clair

Facility l_s<
low-sulfur bituminous

PRB subbituminous/ low-sulfur
butuminous blend (85:15)

low-sulfur bituminous
CS-ESPb with and without ACI

CS-ESP with and without ACI; with SNCRc/urea

CS-ESP with and without ACI

HS-ESP6 with and without ACI; with COHPAC'

CS-ESP with and without B-PAC

HS-ESP with and without B-PAC; with SOFA1
ports "on" for NOxj control.
                                                                   Class F
                                                                  (12.2,9.5)
                                                                   Class F
                                                                 (11.7, 10.3)
                                                                   Class C
                                                                 (11.2, 11.9)
                                                                   Class F
                                                                  (11.1,8.4)
                                                                  Class C/F
                                                                 (12.1, 12.2)
                                                                   Class F
                                                                  (5.8, 6.0)
Report 2: Facilities with scrubbers

Facility Ak        low-sulfur bituminous

Facility B         low-sulfur bituminous

Facility Hm        high-sulfur bituminous

Facility I"         high-sulfur bituminous

Facility Kp        medium-sulfur bituminous
                   fabric filter, limestone wet scrubber with and
                   without SNCR/urea
                   CS-ESP, magnesium-enhanced lime scrubber;
                   with SCR'/ammonia

                   CS-ESP, limestone wet scrubber

                   CS-ESP, limestone wet scrubber with forced
                   oxidation; SCR not in use when sample taken
                   CS-ESP, magnesium-enhanced lime wet
                   scrubber, natural oxidation
                                                                            Class F
                                                                           (10.3, 10.5)
                                                                            Class F
                                                                           (10.3,9.5)
                                                                            Class F
                                                                              (8.5)
                                                                            Class F
                                                                             (NT0)
                                                                            Class F
                                                                              (9.2)
Report 3: Miscellaneous configurations

Facility E (Unit 1)  medium-sulfur bituminous

Facility E (Unit 2)  medium-sulfur bituminous

Facility E (Unit 3)  low-sulfur bituminous

Facility E (Unit 4)q  low-sulfur bituminous

Facility F         low-sulfur bituminous

Facility G         low-sulfur bituminous
                   CS-ESP SCR operating

                   CS-ESP SCR off

                   CS-ESP SCR operating

                   HS-ESP SCR operating

                   CS-ESP

                   CS-ESP, SNCR operating
                                                                            Class F
                                                                              (4.8)
                                                                            Class F
                                                                              (4.3)
                                                                            Class F
                                                                              (4.8)
                                                                            Class F
                                                                             (NT0)
                                                                            Class F
                                                                              (4.2)
                                                                            Class F
                                                                              (4.3)
a The pH of the CCR (with and without ACI or B-PAC injection, as applicable) when mixed in distilled water at a ratio of 1 g
  CCR per 10 mL water.
b CS-ESP = cold-side electrostatic precipitator.
c SNCR = selective non-catalytic reduction.
d PRB = Powder River Basin.
e HS-ESP = hot-side electrostatic precipitator.
f COHPAC = compact hybrid paniculate collector.
                                                                                            continued

-------
Characterization of Coal Combustion Residues
Table 1 (concluded). List of CCRs collected to date for evaluation of under this research program. [Some facilities are
identified by code letter only (e.g., "Facility C") to preserve the confidentiality of the CCR source.]

9 This facility has HS-ESP for paniculate control. The fly ash undergoes pneumatic control. Collected fly ash with and without
use of B-PAC.
h Two samples were collected from this facility. The only difference between the two samples (Run #1 and Run #2) was the
Run #1 sample was allowed to accumulate in the hopper for 4 hours and the Run #2 sample for 30 minutes prior to
collection. Not enough fly ash was collected with 30 minutes of accumulation to evaluate leaching potential. Therefore, the
sample that was allowed to accumulate in hopper from HS-ESP for about 4 hours prior to collection was used for leaching
evaluation. A concern is that mercury may have partially desorped from fly ash prior to collection because of the high
temperature in the collection hopper. Total mercury analyses were used to evaluate the change in mercury content for fly ash
with and without brominated carbon injection after being collected over 4 hours and 30 minutes.
1 SOFA = separated overfire air.
i NOX = oxides of nitrogen.
k CCR samples obtained when SNCR was in use (during summer months) and not in use.
1 SCR = selective catalytic reduction.
m For Facility H, sludge is first collected in the absorber at 12-15% solids, then goes through cyclone to achieve 50% solids, and
finally is dewatered using a belt press to >90% solids. While on belt, gypsum is "sprayed" to remove excess soluble salts. For
this facility, samples have been obtained of (1) prepared gypsum (which is used for wall board production) and (2) fly ash from
the CS-ESP.
n Facility I has a 500 MW tangential-fired boiler. Samples from this facility include (1) fly ash when the SCR was operating, (2) fly
ash when SCR was not operating, and (3) raw FGD sludge when the SCR was not operating. Scrubber sludge from this
facility is used in making gypsum for producing wallboard. The samples from this facility were gypsum and FGD sludge. There
was not enough sludge to test for leaching, so the pH was not tested.
0 NT = not tested.
p Facility K is an 800 MW facility with two 400 MW units (tangential fired). APC includes CS-ESP, magnesium-enhanced lime
wet scrubber with natural oxidation. There is no SCR. Samples received from Facility K, are (1) partially dewatered FGD
sludge, (2) fly ash sample, and (3) fly ash stabilized sludge.
q Fly ash found to have low mercury and selenium content and, therefore, was not included in the leaching evaluation.
Table 2. General Characteristics of Coals Burned in U.S. Power Plants (EPA, 2005).
Coal
Bituminous
Subbituminous
Lignite
Mercury
ppm (dry)
Range Avg
0.036-0.279 0.113
0.025-0.136 0.017
0.080-0.127 0.107
Chlorine
ppm (dry)
Range Avg
48-2730 1033
51-1143 158
133-233 188
Sulfur
%(dry)
Range Avg
0.55-4.10 1.69
0.22-1.16 0.50
0.8-1.42 1.30
Ash
%(dry)
Range Avg
5.4-27.3 1 1 .1
4.7-26.7 8.0
12.2-24.6 19.4
HHVa
BTU/lb (dry)
Range Avg
8646-14014 13203
8606-13168 12005
9487-10702 10028
1 HHV = higher heating value.
1.2.1. Current Air Pollution Control Tech-
nologies
The current combined capacity of U.S. coal-fired power
plants is just over 300 GW and includes a wide range of
combinations of installed air pollution control (APC) de-
vices.

Table 3 shows the current and projected coal-fired capac-
ity by APC configuration. Several of the air pollution con-
trol devices described here will remove some mercury (co-
benefit control) from stack gases as they perform their main
function. Current APC devices are designed primarily to
control particulates, oxides of sulfur (SOX), and NOX.

Post-combustion particulate matter controls used at coal-
fired utility boilers in the United States can include ESPs,
fabric filters (FF), particulate scrubbers (PS), or mechani-
cal collectors (MC). Post-combustion SO2 controls can
consist of a wet scrubber (WS), spray dryer adsorber (SDA),
or duct injection. Post-combustion NOX controls can in-
volve SCR or selective noncatalytic reduction (SNCR).

In response to current and proposed NOX and SO2 control
requirements, additional NOX control and flue gas desulfu-
rization (FGD) systems are expected to be installed and
more widely used in the future. Over half of the U.S. coal-
fired capacity is projected to be equipped with SCR and/or
FGD technology by 2020.

The mercury capture efficiency of existing ESPs and FFs
appears to heavily depend on the partitioning of mercury
between the particulate and vapor phases and the distribu-

-------
Characterization of Coal Combustion Residues
Table 3. Projected Coal-Fired Capacity by APC Configuration (EPA, 2005).
APC Configuration3
Cold-Side ESP
Cold-Side ESP + Wet Scrubber
Cold-Side ESP + Wet Scrubber + ACI
Cold-Side ESP + Dry Scrubber
Cold-Side ESP + SCR
Cold-Side ESP + SCR + Wet Scrubber
Cold-Side ESP + SCR + Dry Scrubber
Cold-Side ESP + SNCR
Cold-Side ESP + SNCR + Wet Scrubber
Fabric Filter
Fabric Filter + Dry Scrubber
Fabric Filter + Wet Scrubber
Fabric Filter + Dry Scrubber +ACI
Fabric Filter + SCR
Fabric Filter + SCR + Dry Scrubber
Fabric Filter + SCR + Wet Scrubber
Fabric Filter + SNCR
Fabric Filter + SNCR + Dry Scrubber
Fabric Filter + SNCR + Wet Scrubber
Hot-Side ESP
Hot-Side ESP + Wet Scrubber
Hot-Side ESP + Dry Scrubber
Hot-Side ESP + SCR
Hot-Side ESP + SCR + Wet Scrubber
Hot-Side ESP + SNCR
Hot-Side ESP + SNCR + Wet Scrubber
Total Existing Units
Current Capacity13
(MW)
111,616
41 ,745
—
2,515
45,984
27,775
...
7,019
317
11,969
8,832
4,960
...
2,210
2,002
805
267
559
932
18,929
8,724
...
5,952
688
684
474
304,955
2010 Capacity0
(MW)
75,732
34,570
379
3,161
35,312
62,663
11,979
4,576
2,830
10,885
8,037
4,960
195
2,950
2,601
805
267
557
932
1 1 ,763
10,509
538
3,233
6,864
1,490
474
298,263
2020 Capacity0
(MW)
48,915
33,117
379
5,403
22,528
98,138
13,153
2,534
6,088
7,646
9,163
4,960
195
1,330
4,422
2,363
345
557
1,108
10,160
10,398
538
1,847
9,912
1,334
627
297,161
New Builds of Coal Steam Units
Fabric Filter + SCR + Wet Scrubber
221
17,292
Total All Units
304,955
298,484
314,453
a Integrated gasification combined cycle units are not included in this list.
b Current capacity includes some SCR and FGD units projected to be built in 2005 and 2006.
c 2010 and 2020 is capacity projected for final CAIR rule; Integrated Planning Model projects some coal retirements and some
new coal in 2010 and 2020.
tion of mercury species (e.g., elemental or oxidized) in the
vapor phase. In general, ESPs and FFs are quite efficient
at removing mercury in the particulate phase; however, the
overall mercury removal efficiency in these devices may
be low if most of the mercury entering the device is in the
vapor phase (MTI, 2001). Many factors contribute to this
range of performance. Differences in mercury contents of
U.S. coals result in a range of mercury concentrations in
the flue gas from the boiler. In general, it is easier to achieve
higher mercury percent removal with higher mercury inlet
6
concentrations (MTI, 2001). The addition of NOX controls
may improve the mercury capture efficiency of particulate
collection devices for some cases.

1.2.2. Enhancement of Controls for Mercury
Removal: Sorbent Injection
Unlike the technologies described earlier, where mercury
removal was incidental and achieved as a co-benefit with
removal of other pollutants, controls are under develop-
ment that target mercury removal by injecting sorbent

-------
Characterization of Coal Combustion Residues
materials into the gas stream of coal-fired boilers. Injec-
tion of dry sorbents, such as powdered activated carbon
(PAC), has been used for control of mercury emissions from
waste combustors and has been tested at numerous utility
units in the United States. However, sorbent injection ex-
perience on waste combustors may not be directly trans-
ferable to coal-fired electric utility boilers due to differ-
ences in facility sizes and mercury content and speciation
in the combustion gases.
Figure 1 presents a coal-fired boiler with sorbent injection
and spray cooling. Figure 2 presents a power plant with a
hot-side ESP (HS-ESP), carbon injection, and a compact
hybrid particle collector. Dry sorbent is typically injected
into the ductwork upstream of a particulate matter (PM)
control device—normally either an ESP or FF. Usually the
sorbent is pneumatically injected as a powder, and the in-
jection location is determined by the existing plant con-
figuration. Another approach, designed to segregate col-
Figure provided by ADA
Environ mental Solutions, Inc.
ESP- Electrostatic Preciptator
FF-Fabric Filters
CEM-Continuous Emission Monitor
Figure 1. Coal-Fired Boiler with Sorbent Injection and Spray Cooling (Senior et al., 2003a).

r
r~i

Boiler
House

Figure provided by AD A
Environmental Solutions. I no
WF -wallfired
A/H -ar heater
COHPAC - Compact
Hybrid Particulate
C oHector
Figure 2. Flow Diagram for Power Plant with a Hot ESP, Carbon Injection, and a COHPAC
(Senior et al., 2003a).

-------
Characterization of Coal Combustion Residues
lected fly ash from collected sorbent, would be to retrofit a
pulse-jet FF (PJFF) downstream of an existing ESP and
inject the sorbent between the ESP and the PJFF. This type
of particulate removal configuration is called a Compact
Hybrid Particle Collector (COHPAC) by its manufacturer
and, when combined with sorbent injection, is called Toxic
Emission Control (TOXECON). The TOXECON configu-
ration can be useful because it avoids commingling the
larger flyash stream with mercury recovered on the inj ected
sorbent. Implementation of sorbent injection for mercury
control will likely entail either:
  •  Injection of powdered  sorbent upstream of the exist-
    ing PM control device (ESP or FF); or
  •  Injection of powdered sorbent downstream of the ex-
    isting ESP and upstream of a retrofit PJFF, the
    TOXECON option; or
  •  Injection of powdered sorbent between ESP fields
    (TOXECON-II approach).

In general, factors that affect the performance of sorbent
technology for mercury methods include:
  •  Injection concentration of the  sorbent measured in lb/
    MMacf;4
  •  Flue gas conditions, including temperature and con-
    centrations of hydrogen chloride (HC1) and sulfur tri-
    oxide (SO3);
  •  The air pollution control configuration;
  •  The characteristics of the sorbent; and
  •  The method of injecting the sorbent.

1.2.3. Mercury Control by Conventional PAC
Injection
The most widely tested sorbent for mercury control at util-
ity boilers is PAC.

In general, the efficacy of mercury capture using standard
PAC increases with the amount of oxidized or ionic mer-
cury (Hg2+) in flue gas relative to elemental mercury (Hg°),5
the number of active sites in the PAC,6 and lower tempera-
4 Sorbent injection concentration is expressed in Ib/MMacf (i.e.,
pounds of sorbent used for each million actual cubic feet of gas).
For a 500 MW boiler, a sorbent rate of 1.0 Ib/MMacf will corre-
spond to approximately 120 Ib/hour of sorbent.

5 Standard PAC binds mercury via physical (i.e., weak) bonds,
which are formed more easily with Hg2+. There have been re-
sults that show a similar removal for both elemental and oxi-
dized mercury. However, the results do not account for surface
catalyzed oxidation of Hg° followed by sorption on the carbon
(EPA, 2005).
8
ture. The amount of Hg2+ in flue gas is usually directly
influenced by the amount of chlorine present in the flue
gas, with higher chlorine content enhancing Hg2+ forma-
tion. Based on these factors, standard PAC injection ap-
pears to be generally effective for mercury capture on low-
sulfur bituminous coal applications, but less effective for
the following applications:
  • Low-rank coals with ESP (current capacity of greater
   than 150 GW; the capacity with this configuration is
   not expected to increase significantly  in the future).
   Lower chlorine and higher calcium contents in coal
   lead to lower levels of chlorine in flue  gas, which re-
   sults in reduced oxidation of mercury and, therefore,
   lower Hg2+ in flue gas;
  • Low-rank coals with SDA and FF (current capacity of
   greater than 10 GW. These number of facilities with
   this configuration is expected to increase significantly
   in the future). Similar effect as above, except lime re-
   agent from the SDA scavenges even more chlorine from
   flue gas;
  • High-sulfur coal (current capacity with wet FGD of
   approximately 100 GW. The number of facilities with
   this configuration is likely to increase to more than
    150 GW capacity by 2015). Relatively high levels of
   SO3 compete for active sites on PAC, which reduces
   the number of sites available for mercury. Generally,
   plants will  use wet FGD and, in  many cases,  SCR;
   PAC injection may be  needed as  a trim application;
   and
  • Hot-side ESPs (current capacity of approximately 30
   GW. The number of facilities with this configuration
   is not likely to increase.). Weak (physical) bonds get
   ruptured at higher temperatures resulting in lower sorp-
   tion capacity.

1.2.4. Mercury Control by Halogenated PAC
Injection
Some situations, as described above, may not have adequate
chlorine present in the flue gas for good mercury capture
by standard PAC. Pre-halogenated PAC sorbents have been
developed to overcome some of the limitations associated
with PAC injection for mercury control in power plant ap-
plications (Nelson et al., 2004; Nelson, 2004). Two halo-
genated PAC sorbents have been tested extensively  in the
field. They are  Sorbent Technologies  Corp. brominated-
PAC (B-PAC)  and Norit America's halogenated PAC
(DARCO HG-LH, formerly known as E-3).
6 These are collection of atoms/radicals such as oxygen, chlorine,
and hydroxyls that provide binding sites.

-------
Characterization of Coal Combustion Residues
Halogenated PACs offer several potential benefits. Rela-
tive to standard PAC, halogenated PAC use:
• may expand the usefulness of sorbent injection to many
situations where standard PAC may not be as effec-
tive;
• may avoid the need for installation of downstream FF,
thereby improving cost-effectiveness of mercury cap-
ture;
• would, in general, be at lower injection rates, which
potentially will lead to fewer plant impacts and a lower
carbon content in the captured fly ash;
• may result in somewhat better performance with low-
sulfur (including low-rank) coals because of less com-
petition from SO3; and,
• may be a relatively inexpensive and attractive control
technology option for technology transfer to develop-
ing countries as it does not involve the capital inten-
sive FF installation.

Performance of a halogenated sorbent such as B-PAC ap-
pears to be relatively consistent regardless of coal type and
appears to be mostly determined by whether or not the cap-
ture is in-flight—as in upstream of a cold-side ESP (CS-
ESP)—or on a fabric filter.

1.3. Coal Combustion Residues
Fossil fuel combustion (burning of coal, natural gas, or
oil) is the primary source of energy in the United States—
providing approximately 67% of the total demand in 1997.
Coal-fired utilities provide more than 50% of all electric
power generated using fossil fuels (EPA, 1999). In 1994
there were approximately 1,250 separate coal-fired boilers
in operation at 45 0 different utilities throughout the United
States (EPA, 1999). These boilers used approximately 900
million tons of coal and produced approximately 105 mil-
lion tons of high-volume coal combustion residues—fly
ash, bottom ash, boiler slag, and FGD wastes (EPA, 1999).
Regulations that require the reduction of mercury air emis-
sions from coal-fired power plants will result in changes
to coal combustion residues including increasing the con-
centration of mercury and other trace metals (Figure 3; EPA,
2002).

CCRs result from unburned carbon and inorganic materi-
als in coal that do not burn, such as oxides of silicon, alu-
minum, iron, and calcium. Air pollution control can con-
centrate or partition metals in fly ash and scrubber sludge.
Bottom ash and boiler slag are not affected by air pollution
control technology, and therefore, these materials are not
examined in this report. Bottom ash is the unburned mate-
rial that is too heavy to be entrained in the flue gas stream
implement CAIR
orCAMR
Lower
Concentration of
Hgin Flue Gas

Increased
Concentration of
Mg and other
metals In CCRs
Greater Potential
forHg Releases?

CAIR: Clean Air Interstate Rule
CAMS: Clear Air Mercury Rule
,,
Commercial
Applications
Greater Potential
for Hg Releases?
Figure 3. Life-Cycle Evaluation of Coal Combustion
Residues (EPA, 2002).

and drops out in the furnace. Boiler slag, unburned carbon
or inorganic material in coal that does not burn, falls to the
bottom of the furnace and melts.

Fly ash and scrubber sludge are the two types of CCRs of
interest in this report. Fly ash is the unburned material from
coal combustion that is light enough to be entrained in the
flue gas stream, carried out of the process, and collected as
a dry material in the air pollution control equipment. Sev-
enty million tons of fly ash were produced in 2003.

FGD wastes (or scrubber sludge) result from a SO2 wet
scrubbing process and generally contain 5% to 10% sol-
ids. The quantity of FGD material produced depends on
the sulfur content of the coal and the amount of coal being
combusted. Thirty million tons of FGD wastes were gen-
erated in 2003.

The properties of fly ash and scrubber residues from many
facilities are likely to change as a result of enhanced air
pollution controls for reducing mercury stack emissions.
Changes in CCR properties will include increased content
of mercury and other co-collected metals (e.g., arsenic,
selenium) and the presence of injected sorbent or other
chemical modifiers to improve mercury removal. In sev-
eral prevalent APC configurations, the sorbent will be com-
mingled with either fly ash or other residue streams, modi-
fying both chemical and physical properties of the CCR.

1.4. Residue Management Practices
CCRs can be disposed in landfills or surface impoundments
or used in commercial applications to produce concrete
and gypsum wallboard, among other products. The major
pathway of concern for release from land disposal and some
beneficial use applications is leaching. Research on the
impact of CCR disposal on the environment has been con-
ducted by many researchers and has been summarized by
9

-------
Characterization of Coal Combustion Residues
the EPA (1988, 1999). However, most of the existing CCR
data are for CCRs prior to implementation of mercury or
multi-pollutant controls.

1.4.1. Beneficial Use
In the United States, approximately 31% of all CCRs pro-
duced are reused in commercial applications or other ben-
eficial uses. Thirty-two percent of fly ash is used in com-
mercial applications such as making concrete/grout, struc-
tural fill, and highway construction (ACAA, 2000;
Thorneloe, 2003). Six million tons of the scrubber sludge
(or 26%) was used in making wall board (ACAA, 2000;
Thorneloe, 2003). In Europe, use of CCRs for commercial
applications/beneficial uses is much higher (over 50%).
Table 4 ACCA, 2003) and Figure 4 present the primary
commercial uses of CCRs, and a breakdown of U.S. pro-
duction and usage by CCR type. The primary commercial
applications or commercial uses of CCRs are shown.

Some of the beneficial uses may have the potential to re-
lease mercury from the CCRs, particularly in high-tem-
perature processes. In cement manufacturing, for example,
CCRs are inputs to the cement kiln. It is expected that vir-
tually all mercury will be volatilized from CCRs in this
application. Even where mercury can be captured by the
controls on cement kilns, approximately two-thirds of ce-
ment kiln dust captured by the control devices is reintro-
duced into the kiln. Therefore, a significant fraction of the
mercury in CCRs introduced into cement kilns may be
emitted to the air at the cement plant. Some mercury may
also be revolatilized when CCRs are used as a filler for
asphalt or when FGD material is used in wallboard manu-
facturing. A separate report will present the results from a
study conducted to evaluate the thermal stability of mer-
cury and other metals during application of these high-
temperature processes.

The fate of mercury and other metals is a potential concern
when CCRs are used on the land (mine reclamation, build-
ing highways, soil amendments, agriculture and in making
concrete, cement) or to make products that are subsequently
disposed (e.g., disposal of wall board in unlined landfill).

For several commercial uses, it appears less likely that
mercury in CCRs will be reintroduced into the environ-
ment, at least during the lifetime of the product. For ex-
ample, mercury appears unlikely to be volatilized from
confined uses such as concrete, flowable fill, or structural
fill. The potential for leaching of mercury in these applica-
tions also seems limited, in part due to the relative imper-
meability of concrete and flowable fill; however, special
applications such as those involving continuous immer-
2
o
tfl
o:
o
o
ro
3
o
Concrete Structural Wallboard Cement Waste Mining Road Misc./ Blasting Soil Snow Aggre
Fill S/S Rsclaimatinn Base/ Other Grit/' Modification/ and Ice
Pavement Roofing Stabilization Control
Granules

CCR Applications

Figure 4. Uses of CCRs Based on 2003 Industry Statistics (ACAA, 2003).
Asphalt Agriculture
10

-------
                                                             Characterization of Coal Combustion Residues
Table 4. Beneficial Uses of OCRs (ACAA, 2003).
        CCPa Categories
Fly Ash
                      FGD
Bottom     FGD     Material    Boiler
 Ash     Gypsum     Wet      Slagb
                   Scrubbers
   FGD
  Material    FGD
   Dry      Other"
Scrubbers"
 FBC
Ashb'c
CCP Production Category Totalsd    70,150,00018,100,00011,900,00017,350,0001,836,235   1,444,273  167,345  796,718
CCP Production Total                                                                               121,744,571
CCP Use Category Totals6         27,136,524   8,247,273  8,299,060   484,412 1,756,004     197,509       0  263,623
All CCP Used Total                                                                                 46,384,405
CCP Use by Application'
Concrete/Concrete Products/Grout
Cement/Raw Feed for Clinker
Flow/able Fill
Structural Fills/Embankments
Road Base/Sub-base/Pavement
Soil Modification/Stabilization
Mineral Filler in Asphalt
Snow and Ice Control
Blasting Grit/Roofing Granules
Mining Applications
Wai I board
Waste Stabilization/Solidification
Agriculture
Aggregate
Miscellaneous/Other
CCP Category Use Totals
Application Use to Production Rate
Overall CCP Utilization Rate

12,265,169
3,024,930
136,618
5,496,948
493,487
515,552
52,608
1,928
0
683,925
0
3,919,898
12,140
137,171
396,150
27,136,524
38.68%


298,181
493,763
20,327
2,443,206
1,138,101
67,998
0
683,556
42,604
1,184,927
0
30,508
3,534
512,769
1,327,797
8,247,273
45.57%


65,693
420,043
0
0
0
0
0
0
0
0
7,780,906
0
32,518
0
0
8,299,060
69.74%


0
0
0
224,100
0
704
0
0
0
259,608
0
0
0
0
0
484,412
2.79%


15,907
15,766
0
1 1 ,074
29,800
0
31,402
102,700
1,455,140
59,800
0
0
0
31,600
2,815
1,756,004
95.63%


34,284
2,469
9,184
12,141
0
114
0
0
0
130,723
0
0
2,295
6,299
0
197,509
13.68%


0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.00%


0
0
0
0
0
188,708
0
0
0
1 1 ,049
0
49,217
0
0
14,649
263,623
33.09%
38.10%
a The American Coal Ash Association uses CCP (coal combustion products) to refer to CCRs.
b As submitted based on 60% coal burn.
c FBC =  fluidized-bed combustion.
d CCP Production totals for Fly Ash, Bottom Ash, FGD Gypsum, and Wet FGD are extrapolated estimates rounded off to the
 nearest 50,000 tons.
e CCP Used totals for Fly Ash, Bottom Ash, FGD Gypsum, and Wet FGD are per extrapolation calculations (not Rounded off).
f CCP Uses by application for Fly Ash, Bottom Ash, FGD Gypsum, and Wet FGD are calculated per proportioning the CCP Used
 Category Totals by the same percentage as each of the individual types' raw data contributions to the as-cubmitted raw data
 submittal total (not rounded off).
sion in water may produce different results. The impact of
advanced mercury emissions control technology (e.g., ACI)
on beneficial use applications is uncertain. There is con-
cern that the presence of increased concentrations of mer-
cury, certain other metals, or high carbon content may re-
duce the suitability of CCRs for use in some applications
(e.g., in Portland cement concrete).


1.4.2. Land Disposal
There are approximately 600 land-based CCR waste dis-
posal units (landfills or surface impoundments) being used
by the 450 coal-fired power plants in the United States
(EPA, 1999). About 70% of the 122 million tons of CCRs
generated annually are land disposed. Landfills may be
located  either on-site or off-site while surface impound-
                       ments are almost always located on-site with the combus-
                       tion operations. Although the distribution of units is about
                       equal between landfills and surface impoundments, there
                       is a trend toward increased use of landfills as the primary
                       disposal method.


                       1.5. Leaching Protocol
                       One of the major challenges facing this research was iden-
                       tification of an appropriate test protocol for evaluating the
                       leaching potential of CCRs that may have increased levels
                       of several metals, particularly mercury. The goal of this
                       research is to develop the most accurate estimates of likely
                       constituent leaching when CCRs are land disposed. These
                       estimates of leaching need to be appropriate for assessing
                       at a national level the likely impacts through leaching of
                                                                         11

-------
Characterization of Coal Combustion Residues
pollutants from CCRs that is a consequence of installing
enhanced mercury and/or multi-pollutant controls. To
achieve this goal requires that U.S. EPA evaluate leaching
potential for CCRs as-managed (to the degree this is known)
and that the leach testing results can be appropriately ex-
trapolated to a national assessment. A large part of the ap-
proach to achieving this has been to identify and evaluate
CCR samples collected from the most prevalent combina-
tions of power plant design (with a focus on air pollution
control configurations) and coal type used. U.S. EPA and
EPRI have also examined and collected data on the actual
disposal conditions for CCRs because these conditions will
affect leaching and will also vary  over time. When dis-
posed, CCRs are typically monofilled or  disposed  with
other CCRs. However, CCR composition can change over
time, due  to changes in the source of coal or coal  type
burned or due to installation of additional  pollution  con-
trol equipment, so the conditions of leaching created by
the CCRs will also change overtime.

Many leaching tests have been developed by regulatory
agencies, researchers, or third-party technical standards
organizations and are described in the published literature.
States and others have expressed concern with the variety
of leaching protocols in use, the lack of correlation of test
results with field conditions and actual leaching, and lack
of comparability of available data because of incomplete
reporting of test conditions. There is also limited or no
quality assurance (QA) information for many of these tests.
Leaching tests such as the toxicity characteristic leaching
procedure (TCLP),7 which reflects  municipal solid waste
co-disposal conditions; the synthetic precipitation leach-
ing procedure  (SPLP); or any number of deionized-water
based tests may be inappropriate or are at least not optimal
for evaluating the leaching potential of CCRs as they are
7 TCLP was not included as part of this study for two reasons.
First, EPA previously made a waste status determination under
RCRA that coal combustion residues are non-hazardous (65 FR
32214, May 22,2000). Therefore, use of TCLP was not required
as indicated under the RCRA toxicity characteristic regulation
for determination of whether or not CCRs were hazardous. Sec-
ond, TCLP was developed to simulate co-disposal of industrial
waste with municipal solid waste as a mismanagement scenario
and to reflect conditions specific to this scenario. However, the
vast majority of CCRs are not being managed through co-dis-
posal with municipal solid waste, and the test  conditions for
TCLP are different from the actual management practices for
most CCRs. In seeking a tailored, "best-estimate" of CCRleach-
ing, the leaching framework provides the flexibility to  consider
the  effects of actual management conditions on these wastes,
and so will be more accurate in this case.
12
actually managed. These tests either presume a set of pre-
vailing landfill conditions that may or may not exist at CCR
disposal sites (e.g., TCLP), try to account for an environ-
mental factor considered to be important in leaching (e.g.,
SPLP), or presume that the  waste  tested will define  the
disposal conditions—such as deionized (DI) water tests.
Most existing leaching tests are empirical in that results
are presented simply as the contaminant concentrations
leached when using the test and presented without mea-
suring or reporting values for factors that may affect waste
leaching or that provide insight into the chemistry that is
occurring in leaching. Most tests are performed as a single
batch test and so do not consider the effect of variations in
conditions on waste constituent leaching.8

In searching for a leach testing approach that will produce
the most reliable results for this waste and that can be used
to predict leaching nationally, EPA sought an approach that
(1) considers the range of known CCR chemistry and man-
agement conditions (including re-use) and (2) permits de-
velopment of data that are comparable across U.S. coal
and CCR types. Because the data resulting from this re-
search will be used to support regulations, careful scrutiny
of the data is expected. Therefore, the use of a published,
peer-reviewed protocol is also considered to be an essen-
tial element of this work.

EPA's Office of Research and Development (ORD)  has
worked closely with EPA's Office of Solid Waste (OSW)
to identify an appropriate leaching protocol for evaluating
CCRs. The  protocol that has been adopted is the "Inte-
grated Framework for Evaluating Leaching in Waste Man-
agement and Utilization of Secondary Materials" (Kosson
et al, 2002) and referred to here as  the "Leaching Frame-
work," or Framework. The Leaching Framework consists
of a tiered approach to leaching assessment. The general
approach under the Leaching Framework is to use labora-
tory testing to measure intrinsic leaching characteristics of
a material (i.e., liquid-solid equilibrium partitioning as a
function of pH and LS ratio,  mass transfer rates) and then
use this information in conjunction with mass transfer
models to estimate constituent release by leaching under
specific management scenarios (e.g., landfilling). Unlike
8 Many factors are known or may reasonably be expected to
affect waste constituent leaching. The solubility of many metal
salts is well known to vary with pH; adsorption of metals to the
waste matrix varies with pH; redox conditions may determine
which metal salts are present in wastes; temperature may affect
reaction rates; water infiltration can affect the leaching rate and
also affect leaching chemistry and equilibrium.

-------
                                                             Characterization of Coal Combustion Residues
other laboratory leaching tests, under this approach, labo-
ratory testing is not intended to directly simulate or mimic
field conditions. Development work to-date on the Frame-
work has focused on assessing metals leaching, and it in-
cludes equilibrium batch testing (over a range of pH and
LS ratio values),  diffusion-controlled mass transfer,  and
percolation-controlled (column) laboratory test methods in
conjunction with mass transfer models to estimate release
for specific management scenarios based on testing results
from a common set of leaching conditions. EPA's OSW
and ORD believe that this approach successfully addresses
the concerns identified above because it seeks to consider
the effect of key disposal conditions on constituent leach-
ing and to understand the leaching  chemistry of wastes
tested.

The following attributes of the Leaching Framework were
considered as part of the selection process:
  • It will permit development of data that are comparable
    across U.S. coal and CCR types;
  • It will permit comparison with existing laboratory and
    field leaching data on CCRs;
  • It was published in the peer-reviewed scientific litera-
    ture;
  • On consultation with EPA's OSW, it was recommended
    as the appropriate protocol based on review of the range
    of available test methods and assessment approaches;
    and
  • On consultation with the Environmental Engineering
    Committee of the Science Advisory Board (June 2003),
    the Committee considered the Leaching Framework
    to be responsive to earlier SAB criticisms of EPA's
    approach to leaching evaluation and to be broadly ap-
    plicable and appropriate for this study. The complete
    summary of the SAB consultation is provided as Ap-
    pendix A.

For this study, the primary leaching tests used from the
Leaching  Framework were Solubility and Release as a
Function of pH (SR002.1) and Solubility and Release as a
Function of the Liquid-Solid Ratio (LS) (SR003.1).9 These
tests represent equilibrium-based leaching characterization.
9 LS refers to liquid to solid ratio (mL water/g CCR or L water/
kg CCR) occurring during laboratory leaching tests or under
field conditions. SR002.1 is carried out at LS=10 with several
parallel batch extractions over a range of pH, while SR003.1 is
carried out using several parallel batch extractions with deion-
ized water at LS= 0.5,1,2, 5 and 10. Under field conditions, LS
refers  to the cumulative amount of water passing through the
total mass of CCR subject to leaching.
The range of pH and LS ratio used in the leaching tests
includes the range of conditions (pH and LS ratio) observed
for current CCR management practices. Results of these
tests provide insights into the physical-chemical mecha-
nisms controlling constituent leaching. When used in con-
junction with mass transfer and geochemical speciation
modeling, the results can provide conservative but realis-
tic estimates of constituent leaching under a variety of en-
vironmental conditions (pH, redox, salinity, carbonation)
and management scenarios.

Laboratory testing for leaching assessment was carried out
at the U.S. EPA National  Risk Management Laboratory
(Research Triangle Park, NC) with technical assistance
from Vanderbilt University.
                                                                                                         13

-------
Characterization of Coal Combustion Residues
14

-------
Characterization of Coal Combustion Residues
2. Materials and Methods
2.1. CCR Materials for Evaluation
The CCR materials tested in this study include a reference
fly ash and fly ashes collected by ADA-Environmental
Solutions from designated coal combustion facilities un-
der contract for the Department of Energy's National En-
ergy Technology Laboratory (NETL) field evaluation pro-
gram of sorbent injection upstream of existing particulate
control devices. This program represents the first time that
PAC has been injected on a large scale for a period of sev-
eral weeks as enhanced mercury control technology. All
six of the facilities evaluated in this report burn either low-
sulfur bituminous coal (Brayton Point, Salem Harbor, Fa-
cility C, Facility L), sub-bituminous coal (Pleasant Prai-
rie) or a sub-bituminous/low-sulfur bituminous coal blend
(St. Clair) and have particulate control devices only (no
SO2 scrubbers). This facility configuration is representa-
tive of 75% of the coal-fired utilities in the U.S. The same
commercial sorbent (Norit Americas FGD Carbon )10 was
used for all of the tests using ACI. This sorbent has a sur-
face area of approximately 600 m2/g and a mass-mean di-
ameter of 18 |im. The tests using B-PAC used sorbent ob-
tained from Sorbent Technologies Corp., with a surface
area of 700 to 1070 m2/g and a mass-mean diameter of 19
|im. Samples of fly ash were collected from each facility
under conditions with the enhanced mercury control tech-
nology turned off and in use.

The facilities and associated CCRs reported here are de-
scribed below. Appendix B provides a schematic flow dia-
gram for each facility. Table 5 provides characteristics of
the low-sulfur bituminous coal combusted at Brayton Point,
Salem Harbor, Facility C and Facility L, the sub-bitumi-
nous coal combusted at Pleasant Prairie and the sub-bitu-
minous/low-sulfur bituminous coal blend combusted at St.
Clair. Elemental composition by x-ray fluorescence and
additional characteristics of the fly ashes from baseline
10 DARCO FGD carbon is currently sold under the trade name
DARCO-HG.
testing and testing with enhanced mercury control are pro-
vided in Table 6 and Table 7. For samples from Salem
Harbor, the loss on ignition (LOI) is more than twice the
total carbon content because of a relatively high fraction
of uncombusted particulate in the CCR. Total content analy-
ses for mercury, arsenic, cadmium, lead and selenium re-
sults are provided in Table 8.

2.1.1. Reference Fly Ash
The reference fly ash was obtained from the EPA, National
Risk Management Research Laboratory (Research Triangle
Park, NC). X-Ray Fluorescence (XRF) analysis shown in
Table 6 is typical of a Class F fly ash from an eastern bitu-
minous coal. This fly ash was selected for this program
because it was available in large quantities (approximately
two 55-gallon drums) and it contained low mercury levels.
The large quantity allows for inter-laboratory comparisons
at a later date. The low mercury content was important to
test the laboratories' ability to close the mercury mass-bal-
ance around the leaching and thermal desorption studies
in the limit case of very low mercury content.

2.1.2. Facilities Using Injection of Standard
Activated Carbon

2.1.2.1. Bray ton Point
Brayton Point Station (Somerset, MA) is operated by
PG&E National Energy Group. This facility is composed
of four fossil fuel fired units designated as Units 1, 2, 3,
and 4. The test unit selected, unit 1, has atangentially fired
boiler rated at 245 MW. Brayton Point Unit 1 was chosen
for this evaluation because of its combination of firing low-
sulfur bituminous coal with a cold-side ESP. This configu-
ration represents a wide range of coal-fired power plants
located in the eastern U.S. (Senior et al., 2003a).

The primary particulate control equipment consists of two
CS-ESPs in series, with an EPRICON flue gas condition-
ing system that provides SO3 for fly ash resistivity control.
15

-------
Characterization of Coal Combustion Residues
Table 5. Characteristics of Coal Combusted and Facilities Sampled and Reported Here (Senior et al., 2003a, Senior et
al., 2003b, Senior et al., 2004).
Parameter
Coal
Sulfur, wt%
Ash, wt%
Moisture, wt
HHVd, Btu/lb
Hg, ug/g
Cl, ug/g
As, ug/g
Cd, ug/g
Pb, ug/g
Se, ug/g
Particulate
Control Device
Sorbent
Injection Point
SOX and NOX
Control
Sampling
Location
Brayton
Point
Low-sulfur
Bituminous
(2002)a
0.7
10.8
4.7
12,780
0.044
1475
5.68
0.055
8.9
3
2 CS-ESPs
in series
Between the
2 ESPs
NAe
Ash Hopper
Row C
Pleasant
Prairie
PRBb Sub-
bituminous
(2002)
0.3
5.1
30.7
8,385
0.109
8.1
NT
NT
NT
NT
CS-ESP
Before ESP
NA
ESP Hopper 1
and 2 Composite
Salem
Harbor
Low-sulfur
Bituminous
(2002)
0.67
6.48
9.05
12,420
0.0617
64.3
2.4
0.14
3.8
4.8
CS-ESP
Before ESP
SNCR
ESP Hopper
A
Facility C
Low-sulfur
Bituminous
1.24
14.78
6.85
1 1 ,902
0.136
169
NT
NT
NT
NT
HS-ESP +
COHPAC
Between HS-ESP
and COHPAC
NA
B-Side Hopper
St. Clair
PRB Subbitu-
minous/Low-sulfur
Bituminous
(85:15) Blend
NTC
NT
NT
NT
NT
NT
NT
NT
NT
NT
CS-ESP
Before South
Side ESP
NA
North and South
side Hoppers
Facility L
Low-sulfur
bituminous
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
HS-ESP
Before B-Side
ESP
Separated Over-
fire Air Ports
A- and B-side
Hoppers
a Year over which coal sampled to obtain average values.
b PRB = Powder River Basin.
c NT = not tested.
d HHV = higher heating value.
e NA = not applicable.
Table 6. Fly Ashes from Brayton Point, Pleasant Prairie, Salem Harbor, and Facility C: Elemental Composition (by x-ray
fluorescence) and Other Characteristics.
Elemc
Al
As
Ba
Br
Ca
Cl
Cr
Cu
Fe
I
K
Mg
Mn
Reference
mt Fly
(Average%)
14.700
0.010
0.110
BML
0.860
0.026
0.017
0.018
5.110
BML
2.460
0.637
0.015
Brayton Point
(Average%)a
Baseline
13.430
BMLb
0.010
0.005
6.080
0.030
0.022
0.022
4.650
BML
1.853
0.800
0.041
with AC I
12.400
BML
0.095
0.065
2.030
0.440
0.018
0.020
2.500
0.014
1.500
0.641
0.020
Pleasant Prairie
(Average%)
Baseline
10.050
BML
0.695
BML
18.430
BML
0.016
0.022
4.310
BML
0.371
2.810
0.057
with AC I
10.220
BML
0.647
BML
16.640
.0.045
0.013
0.022
4.280
BML
0.455
2.460
0.020
Salem Harbor
(Average%)
Baseline
9.523
BML
0.091
BML
1.298
0.101
0.011
0.007
4.870
BML
1.250
1.785
0.045
with ACI
7.623
BML
0.099
BML
0.803
0.203
0.012
0.008
3.630
BML
0.977
0.420
0.021
Facility C
(Average%)
Baseline
12.25
BML
0.206
0.0025
2.07
0.0373
0.0177
0.0247
7.43
BML
1.84
1.679
0.0196
with ACI
8.96
BML
0.148
0.0097
1.92
0.0790
0.0134
0.0179
5.90
BML
1.34
0.586
0.0179
                                                                                               continued
16

-------
                                                           Characterization of Coal Combustion Residues
Table 6 (concluded). Fly Ashes from Brayton Point, Pleasant Prairie, Salem Harbor, and Facility C: Elemental Composition
(by x-ray fluorescence) and Other Characteristics.
Reference
Element Fly

Na
Ni
Pb
Pxc
Se
Si
Sr
Sd
Sxe
Ti
V
Zn
Zr
(Average%)
0.346
0.011
0.008
0.087
BML
26.400
0.089
BML
0.174
0.897
0.031
0.023
0.050
Brayton Point
(Average%)a
Baseline
0.511
0.015
BML
0.161
0.005
23.080
0.124
BML
0.351
1.015
0.043
0.021
0.031
with AC I
0.242
0.016
0.010
0.042
0.020
23.240
0.083
BML
0.582
0.100
0..32
0.011
0.031
Pleasant Prairie
(Average%)
Baseline
1.660
0.006
BML
0.056
BML
16.600
0.369
BML
0.635
0.964
0.030
0.009
0.035
with AC I
1.310
0.006
BML
0.508
BML
16.250
0.341
BML
0.971
0.943
0.033
0.010
0.035
Salem Harbor
(Average%)
Baseline
0.270
0.009
0.005
0.086
0.005
21.898
0.042
BML
0.335
0.453
0.029
0.013
0.019
with ACI
0.293
0.009
BML
0.057
0.005
23.468
0.032
BML
0.761
0.407
0.030
0.013
0.019
Facility C
(Average%)
Baseline
0.374
0.0173
0.0066
0.303
0.0157
17.48
0.143
BML
0.544
0.709
0.0411
0.0139
0.0245
with ACI
0.287
0.0149
0.0037
0.184
0.0487
12.92
0.104
BML
1.18
0.574
0.0322
0.0105
0.0202
Physical Parameters
Total
lULdl ,-. -,/-,
Carbon °76
Surface
Area' 1 .36
(m2/g)
LOI (wt%) 0.85


2.3

6.5

5.5


13

92

12


0.25

1.8

0.60


3.6

23

3.5


7.8

28

21


11

36

25


10.9

14.10

18.0


24.44

36.55

36.26
a Unless otherwise noted.
b BML = below method limit (As<0.009%, l<0.006%, Pb<0.003%, Se<0.003%).
c Px = phosphorus in oxidized form such as phosphate.
d S = sulfur in elemental form.
e Sx = sulfur in oxidized form such as sulfate.
f Brunauer, Emmett, and Teller method for quantifying surface area.
Table 7. Fly Ashes from St. Clair and Facility L: Elemental Composition (by x-ray fluorescence) and Other Characteristics.
Reference
Element Fly

Al
As
Ba
Br
Ca
Cl
Cr
Cu
Fe
I
K
(Average%)
14.700
0.010
0.110
BML
0.860
0.026
0.017
0.018
5.110
BML
2.460
St. Clair
(Average%)a
Baseline
10.63
BMLb
1.20
BML
12.06
0.0156
0.0116
0.0170
5.35
BML
0.794
with BPAC
10.16
BML
1.01
0.0962
11.35
0.0412
0.0109
0.0148
5.52
0.014
0.768
Facility C
(Average%)
Baseline
13.19
BML
0.0652
BML
0.328
0.0389
0.0147
0.0100
2.39
BML
2.27
with BPAC
13.15
BML
0.0632
0.0061
0.319
0.0339
0.0147
0.0093
2.36
BML
2.22
                                                                      continued
                                                                                                      17

-------
Characterization of Coal Combustion Residues
Table 7 (concluded). Fly Ashes from St. Clair and Facility L: Elemental Composition (by x-ray fluorescence) and Other
Characteristics.
Reference
Element Fly

Mg
Mn
Na
Ni
Pb
Pxc
Se
Si
Sr
Sd
Se
X
Ti
V
Zn
Zr
(Average%)
0.637
0.015
0.346
0.011
0.008
0.087
BML
26.400
0.089
BML
0.174
0.897
0.031
0.023
0.050
St. Clair
(Average%)a
Baseline
3.07
0.0350
4.66
0.0074
0.0045
0.219
BML
16.65
0.565
BML
1.23
0.759
0.0292
0.0104
0.0274
with BPAC
2.92
0.0314
4.09
0.0064
0.0040
0.169
0.0024
17.00
0.517
BML
1.03
0.713
0.0257
0.0077
0.0283
Facility C
(Average%)
Baseline
0.584
0.0088
0.134
0.0109
0.0046
0.0262
BML
24.75
0.0322
BML
BML
0.882
0.0232
0.0075
0.0284
with BPAC
0.580
0.0085
0.132
0.0109
0.0046
0.0240
BML
24.8
0.0322
BML
BML
0.878
0.0233
0.0065
0.0277
Physical Parameters
Total
Carbon
Surface
Area'
(m2/g)
LOI (wt%)
0.76

1.36

0.85
0.16

2.50

0.41
2.65

24.86

3.19
5.56

8.23

12.28
5.92

27.01

12.38
               a Unless otherwise noted.
               b BML = below method limit (As<0.009%, l<0.006%, Pb<0.003%,
               Se<0.003%).
               c Px = phosphorus in oxidized form such as phosphate.
               d S =  sulfur in elemental form.
               e Sx = sulfur in oxidized form such as sulfate.
               f Brunauer, Emmett, and Teller method for quantifying surface area.
Table 8. CCRs from Facilities with Electrostatic Precipitators: Total Content of Mercury,
Selenium. [All analyses are according to EPA Method 3052, except for mercury (thermal),
(EPA, 1998b)].
Arsenic, Cadmium, Lead and
which is by EPA Method 7473
Sample ID
Brayton Point Baseline
Brayton Point with ACI
Pleasant Prairie Baseline
Pleasant Prairie with ACI
Salem Harbor Baseline
Salem Harbor with ACI
Facility C Baseline
Facility C with ACI
Mercury
(ng/g)
650.6+6.8
1529.6+1.1
157.7+0.2
1180+1.2
528.5+5.3
411.5+12.6
15.8+0.9
1150.7+14
Mercury
(thermal)
(ng/g)
582.2+2.1
1414.1+43.7
146.9+3.9
1176.8+16.4
573.8+8.7
454.0+12.1
10.5+0.7
1090.1+24.1
Arsenic
(|jg/g)
80.5+1.9
27.9+2.1
21.3+0.3
24.0+0.8
25.9+0.0
26.0+0.0
93.6+5.5
506.3+28.7
Cadmium
(|jg/g)
BMLa
BML
BML
BML
NTb
NT
NT
NT
Lead
(|jg/g)
117.3+4.9
82.9+2.3
41.6+0.8
47.0+0.3
24.9+1.4
24.0+0.0
55.8+0.7
114.4+5.8
Selenium
(|jg/g)
51 .4+1 .7
151.9+6.2
BML
BML
41.9+0.1
44.0+0.0
BML
206.3+0.9
                                                                                             continued
18

-------
Characterization of Coal Combustion Residues
Table 8 (concluded). CCRs from Facilities with Electrostatic Precipitators: Total Content of Mercury, Arsenic, Cadmium,
Lead, and Selenium. [All analyses are according to EPA Method 3052, except for mercury (thermal), which is by EPA
Method 7473 (EPA, 1998b)].
Sample ID
St. Clair Baseline
St. Clair with B-PAC
Facility L (Run 1) Baseline0
Facility L (Run 1) with B-PAC0
Facility L (Run 2) Baseline01
Facility L (Run 2) with B-PACd
MDL
Minimum Quantification Limit
Mercury
(ng/g)
110.9+5.8
1163.0+8.9
13.0+0.2
37.7+1.3
20.3+0.14
71 .4+0.03
0.2 ng/g
0.72 ng/g
ivici uui y
(thermal)
(ng/g)
NT
NT
NT
NT
NT
NT
0.145 ng/g
1 -0 ng/g
Arsenic
(|jg/g)
43.4+2.6
40.8+1.1
20.0+1.1
18.7+0.7
44.4+1.1
44.3+1.4
1.12
4.0
Cadmium
(|jg/g)
1.4+0.1
1.3+0.1
0.4+0.0
0.3+0.0
0.6+0.1
0.9+0.2
1.0
10.0
Lead
(Mg/g)
46.3+17.9
34.9+1 .7
44.8+0.7
42.2+0.3
60.2+3.8
63.0+2.8
0.18
0.6
Selenium
(Mg/g)
10.7+0.1
12.6+0.9
4.1+0.1
4.3+0.2
3.0+0.3
4.3+0.0
0.72
4.0
a BML = below method limit.
b NT = not tested.
0 Pneumatic controls were turned off for 4 hr to collect fly ash.
d Pneumatic controls were turned off for 30 min to collect fly ash. Not tested for leaching.
The EPRICON system is not used continuously, but on an
as-needed basis. The first ESP ("Old ESP") in this particu-
lar configuration was designed and manufactured by
Koppers. The Koppers ESP has a weighted wire design
and a specific collection area (SCA) of 156 ft2/1000 acfm.
The second ESP ("New ESP") in the series configuration
was designed and manufactured by Research-Cottrell. The
second ESP has a rigid electrode design and an SCA of
403 ft2/1000 acfm. Total SCA for the unit is 559 ft2/1000
acfm. The precipitator inlet gas temperature is nominally
280 °F at full load (Senior et al., 2003a).

Hopper ash is combined between both precipitators in the
dry ash-pull system. The ash is processed by an on-site
Separation Technology Inc. (STI) carbon separation sys-
tem, to reduce the carbon content. This processed ash is
sold as base for concrete and the remainder of the higher
carbon ash is land disposed (Senior et al., 2003a).

The injection rate of the PAC was 20 Ib of sorbent used for
each million actual cubic feet of gas (Ib/MMacf) at the
time when the CCR with ACI in use was collected from
this facility.

The baseline and post-control ashes used for this study were
collected as composite samples from the C-row ash hop-
pers of the new ESP before processing for carbon separa-
tion.11 The baseline ash was collected on 6 June 2002. The
post-control fly ash was collected on 21 July 2002. Both
11 Ash for this study was collected before processing for carbon
separation because not all facilities do this processing.
fly ashes were stored in covered five gallon buckets in the
onsite trailer at ambient temperatures.

2.1.2.2. Pleasant Prairie
Wisconsin Electric Power Company, a subsidiary of Wis-
consin Energy, owns and operates Pleasant Prairie Power
Plant located near Kenosha, WI. The plant has two 600
MW balanced-draft coal-fired boilers designated Units 1
and 2. Unit 2 is the test unit. This site was of key interest
because it was the only plant in the NETL program that
burns a variety of Powder River Basin (PRB) low sulfur,
sub-bituminous coals. In addition, this facility has the abil-
ity to isolate one ESP chamber (1/4 of the unit) (Starns et
al., 2002).

The primary particulate control equipment consists of CS-
ESPs of weighted wire design with a Wahlco gas condi-
tioning system that provides SO3 for fly ash resistivity con-
trol. The precipitators were designed and built by Research-
Cottrell. The design flue gas flow was 2,610,000 acfm.
The precipitator inlet gas temperature is nominally 280 °F
at full load (Starns et al., 2002).

Precipitator #2 is comprised of four electrostatic precipita-
tors that are arranged piggyback style and designated 2-1,
2-2,2-3, and 2-4. Each of the four precipitators is two cham-
bers wide and four mechanical fields deep with eight elec-
trical fields in the direction of gas flow. The SCA is 468
ft2/kacfin (Starns et al., 2002).

Hopper ash is combined from all four precipitators in the
dry ash-pull system and sold as base for concrete (Starns
etal.,2002).
19

-------
Characterization of Coal Combustion Residues
The PAC injection rate was 10 Ib/MMacf at the time when
the CCR with ACI in use was collected from this facility.

The baseline ash was collected as a composite sample from
ash hoppers 7-1 and 7-2 of ESP 2-4. The post-control ash
was collected as a grab sample from ash hopper 7-2 of
ESP 2-4 (see Appendix B for flow diagram). The baseline
ash was collected on 11 September 2001, and the post-
control fly ash was collected on 13 November 2001. Both
fly ashes were stored in covered five gallon buckets in the
onsite trailer at ambient temperatures.

2.1.2.3. Salem Harbor
PG&E National Energy Group owns and operates Salem
Harbor Station located in Salem, MA. There are four fossil
fuel fired units at the facility designated as Units 1, 2, 3,
and 4. Units 1-3 fire a low sulfur, bituminous coal and use
oil for startup. Unit 4 fires #6 fuel oil. Unit 1, the test unit,
is a B&W single-wall-fired unit with twelve DB Riley CCV-
90 burners. It is rated at 88 gross MW. Salem Harbor Unit
1 was chosen for this evaluation because of its combina-
tion of firing low-sulfur bituminous coal with urea-based
SNCR, high LOI, and a CS-ESP. The opportunity to quan-
tify the impact of SNCR on mercury removal and sorbent
effectiveness is unique in this program. In addition, test
results from prior mercury tests have indicated 87% to 94%
mercury removal efficiency on this unit without sorbent
injection (Senior et al., 2003a). However, fly ash from this
facility has a relatively high percentage of total carbon
without carbon injection (7.8%, see Table 6), which likely
serves as a sorbent for mercury.

The particulate control equipment consists of a two-cham-
ber CS-ESP (chambers designated 1-1 and 1-2), which pro-
vides two separate gas flow paths from the outlet of the
tubular air heaters to the ID fan inlets. This Environmental
Elements ESP has a rigid electrode design and a SCA of
474 ft2/1000 acfrn. The precipitator inlet gas temperature
is  nominally 255 °F at full load. Typical LOI or carbon
content of the Unit 1 ash is about 25%. This ash is landfilled.

The PAC injection rate was 10 Ib/MMacf at the time when
the CCR with ACI in use was collected from this facility.

The baseline and post-control ashes used for this study were
collected as grab samples from the first ash hopper (hop-
per A) of row 1-1 of the ESP. The baseline ash was col-
lected on 6 June 2002, and the post-control fly ash was
collected on 7  July 2002. Both fly ashes were stored in
covered five gallon buckets in an onsite trailer at ambient
temperatures.

20
2.1.2.4. Facility C
This plant has four 270 MW balanced draft coal-fired boil-
ers designated as Units 1-4. All of these units fire a variety
of low-sulfur, washed, Eastern bituminous coals. Unit #3
was used for the ACI studies.

All of the units at this plant employ HS-ESP as the pri-
mary particulate control equipment. The HS-ESP of unit
#3 is followed by COHPAC. The COHPAC system is a
pulse-jet cleaned baghouse designed to treat flue gas vol-
umes of 1,070,000 acfm at 290 °F. The COHPAC baghouse
consist of two sides, with the A-side being the control and
the B-side being the side where activated carbon was in-
jected after the HS-ESP but before the COHPAC. An ESP
followed by COHPAC and combined with sorbent injec-
tion is referred to as the TOXECON configuration.

The  injection rate of the PAC was  1.5 Ib/MMacf at the
time when the CCR with ACI in use was collected from
this facility.

2.1.3. Facilities Using Injection ofBromi-
nated Activated Carbon

2.1.3.1. St. Clair
Detroit Energy St. Clair Power Plant Unit #1  is a 160 MW
boiler that typically burns a 85:15 blend of PRB and bitu-
minous coals.12 The flue gas from the boiler splits  and is
directed into two parallel CS-ESPs (designated the "South
ESP" and the "North ESP", each treating half of the flue
gas). The flue gas is then recombined before exiting the
stack. During testing, B-PAC was injected upstream of the
South ESP. The unit has no NOX or SO2 controls.

The  injection rate of the B-PAC was 5 Ib/MMacf at the
time when the CCR with B-PAC in use was collected from
this facility.

2.1.3.2. Facility L
This facility is configured similarly to St. Clair except that
it used one HS-ESP with two compartments rather than
two CS-ESPs, and it uses  separated overfired air (SOfA)
ports for NOX control. As a result, the fly ash collection
temperature is between 300 and 450 °F. Samples were col-
lected from hoppers which were evacuated under negative
12 The unit sometimes switches to 100% PRB on the weekends.
However, during our flue gas/fly ash sampling, the unit was
burning the PRB/bituminous blend.

-------
                                                            Characterization of Coal Combustion Residues
pressure. The pneumatic hopper controls were turned off
to allow enough samples to collect for the leaching evalu-
ation. The controls were off for about 4 hr. There is con-
cern that because of the high temperature within the fly
ash collection hoppers, some mercury may have desorbed
prior to sampling. Therefore, the  samples  obtained for
evaluation may have a lower metal content. Because of the
concern about mercury desorbing from the fly ash, addi-
tional fly ash was collected by turning off the pneumatic
transfer for 30 min (2 weeks after the original samples were
collected). Total metal content determinations were com-
pleted for all samples, which includes with and without B-
PAC for fly ash collected after accumulation in the hopper
for 4 hr (first sampling) and 30 min (second sampling).
The leaching evaluation was conducted only on the samples
collected over 4 hr intervals since this provided adequate
sample size (5 gallons).

2.2. Leaching Assessment Protocols
Laboratory testing for this study focused on leaching as a
function of pH and LS ratio as defined by the  Leaching
Framework. This is considered Tier  2 testing (equilibrium-
based) for detailed characterization, which was selected to
establish baseline CCR characteristics. Mass transfer rate
testing (Tier 3, detailed characterization) may be carried
out in the future for specific cases where results from equi-
librium-based characterization indicate a need for detailed
assessment.

2.2.1. Alkalinity, Solubility and Release as a
Function ofpH (SR002.1)
Alkalinity, solubility, and release as a function of pH were
determined according to method SR002.1 (Kosson et al.,
2002). This protocol consists of 11  parallel extractions of
particle size  reduced material at a liquid-to-solid ratio of
10 mL extractant/g dry sample. Particle size  reduction fa-
cilitates achieving equilibrium, but minimal size reduction
was required for the samples evaluated in this study. Each
extraction condition was carried out in triplicate using 40
g of material for each material evaluated. In addition, three
method blanks were included, consisting of the deionized
water, nitric  acid and potassium hydroxide  used for ex-
tractions. Typical particle size of the tested materials was
less than 300 |im. An acid or base addition schedule  is
formulated based on initial screening for eleven extracts
with final solution pH values between 3 and 12, through
addition of aliquots of nitric acid or potassium hydroxide
as needed. The exact schedule was  adjusted  based on the
nature of the material; however, the range of pH values
included the  natural pH of the matrix that may extend the
pH domain (e.g., for very alkaline or acidic materials). The
final LS ratio is 10 mL extractant/g dry sample, which in-
cludes DI water, the added acid or base, and the amount of
moisture that is inherent to the waste matrix as determined
by moisture content analysis. The eleven extractions were
tumbled in an end-over-end fashion at 28±2 rpm for 24 hr
followed by filtration separation of the solid phase from
the extract using a 0.45 |im polypropylene filter. Each ex-
tract then was  analyzed for constituents of interest. The
acid and base neutralization behavior of the materials was
evaluated by plotting the pH of each extract as a function
of equivalents of acid or base added per gram of dry solid.
Concentration  of constituents of interest for each extract
was plotted as a function of extract  final pH to provide
liquid-solid partitioning equilibrium as a function of pH.

2.2.2. Solubility and Release as a Function
of LS Ratio  (SR003.1)
Solubility and release as a function of LS ratio was deter-
mined according to method SR003.1 (Kosson et al., 2002).
This protocol consists of five parallel batch extractions over
a range of LS  ratios (i.e., 10, 5, 2, 1, and 0.5 mL/g dry
material), using DI water as the extractant with aliquots of
material that has been particle size reduced. Typical par-
ticle size of the material tested was less than 300 |im. Be-
tween 40 and 200 g of material, based on the desired LS
ratio,  were used for each extraction. All extractions were
conducted at room temperature (20 ±2 °C) in leak-proof
vessels that were tumbled in an  end-over-end fashion at
28±2 rpm for 24 hr. Following gross separation of the solid
and liquid phases by centrifugation or settling, leachate
pH and conductivity measurements were taken, and the
phases were separated by pressure filtration using 0.45 |im
polypropylene  filter membrane. The five leachates were
collected and preserved, as appropriate, for chemical analy-
sis. Each extraction condition was carried out in triplicate
and a method blank consisting of the DI water used for
extraction was  included.

2.3.  Analytical  Methods
2.3.1. Surface Area and Pore Size Distribu-
tion
A Quantachrome Autosorb IC-MS was used to perform 5-
point Brunauer, Emmett, and Teller (BET) method surface
area, pore volume, and pore size distribution analysis on
each as-received and size-reduced CCR. A 200 mg sample
was degassed under vacuum at 200 °C for at least 1 hr in
the sample preparation manifold prior to analysis with N2
as the analysis gas. Standard materials with known surface
area were routinely run as a QC check.
                                                 21

-------
Characterization of Coal Combustion Residues
2.3.2. pH and Conductivity
Conductivity and pH were measured for all aqueous ex-
tracts using an Accumet 925 pH/ion meter. The pH of the
leachates was measured using a combined pH electrode
accurate to 0.1 pH units. A 3-point calibration was per-
formed using pH buffer solutions at pH 4.0, 7.0, and 10.0.
Conductivity of the leachates was measured using a stan-
dard conductivity probe. The conductivity probe was cali-
brated using appropriate standard conductivity solutions
for the conductivity range of concern. Conductivity meters
typically are accurate to ± 1 % and have a precision of ± 1 %.

2.3.3. Moisture Content
Moisture content of the "as received" CCRs, was deter-
mined using American Society for Testing and Materials
(ASTM) method D 2216-92. This method supercedes the
one indicated in the published version of the leaching pro-
cedure.

2.3.4. Carbon Content: Organic Carbon/
Elemental Carbon Analyzer
Organic carbon (OC) and elemental carbon (EC) content
of each CCR tested was measured using a Sunset Lab ther-
mal-optical EC/OC analyzer using NIOSH Method 5040.
The sample, collected on quartz fiber filters, was heated
under a completely oxygen-free helium atmosphere in a
quartz oven in four increasing temperature steps (375,540,
670, and 870  °C)  at 60 s ramp times for the first three
temperatures and a 90 s ramp time for the final tempera-
ture. This removed all organic carbon on the filter. As the
organic compounds were vaporized, they were immedi-
ately oxidized to carbon dioxide in an oxidizer oven that
followed the sample oven. The flow of helium containing
the produced  carbon dioxide then  went to a quartz
methanator oven where the carbon dioxide was reduced to
methane, which was then detected by a flame ionization
detector (FID). After the sample oven was cooled to  525
°C, the pure helium eluent was switched to an oxygen/
helium mixture in the sample oven. At that time, the sample
oven temperature was stepped up to 850 °C. During this
phase, both the original elemental carbon and the residual
carbon produced by the pyrolysis of organic compounds
during the first phase were oxidized to carbon dioxide by
the presence of oxygen in the eluent. The carbon dioxide
was then converted to methane and detected by the FID.
After all carbon had been oxidized from the sample, a
known volume and concentration of methane was injected
into the sample oven, so each sample was calibrated to a
known quantity of carbon. This also provided a means of
checking the operation of the instrument.
The calibration range for these analyses was from 10 to
200 |ig/cm2 of carbon using a sucrose solution as the stan-
dard. The detection limit of this instrument is approximately
100 ng/cm2 with a linear dynamic range from 100 ng/cm2
to 1 g/cm2.

2.3.5. Mercury (CVAA, Method 3052, and
Method 7473)
Liquid samples were preserved for mercury analysis by
additions of nitric acid and potassium permanganate and
then prepared prior to analysis according to the following
method. For each 87 mL of sample, 3 mL of concentrated
nitric acid and 5 mL of 5 wt% aqueous potassium perman-
ganate solution were added prior to storage. Immediately
before cold vapor atomic absorption (CVAA) analysis, 5
mL of hydroxylamine were added to clear the sample, and
the sample was then digested according to ASTM Method
D6784-02 (Ontario Hydro) as described for the permanga-
nate fraction. On completion of the digestion, the sample
was analyzed for mercury by CVAA. Samples with known
additions of mercury for matrix analytical spikes also were
digested as described above prior to CVAA analysis.

Sample preparation of the solids and filters was carried
out by HF/HNO3 microwave digestion according to Method
3052 followed by CVAA analysis as indicated above. No
additional preservation or digestion was carried out prior
to CVAA analysis.

Mercury analysis of each digest, extract, and leachate was
carried out by CVAA according to EPA SW846 Method
7470A "Mercury in  Liquid Waste (Manual Cold Vapor
Technique)." A Perkin Elmer FIMS 100  Flow Injection
Mercury System  was used for this  analysis. The instru-
ment was calibrated with known  standards ranging from
0.025 to 1 |lg/L mercury.

Solids also were  analyzed by Method 7473 "Mercury in
Solids and Solutions by Thermal Decomposition, Amal-
gamation,  and Atomic Absorption Spectrophotometry"
(EPA, 1998b). A Nippon MD-1 mercury system was used
for this analysis. The instrument was calibrated with known
standards ranging from 1 to 20 ng of mercury. The method
detection limit for mercury in solids is 0.145 mg/kg.

2.3.6. Other Metals (ICP-MS, Method 3052
and Method 6020)
Inductively coupled plasma-mass spectroscopy (ICP-MS)
analyses for other elements of interest were carried out by
Vanderbilt and STL laboratories. These two laboratories
22

-------
Characterization of Coal Combustion Residues
were used to provide inter-laboratory comparison for se-
lected analyses.

Liquid samples for ICP-MS analysis were preserved by
addition of 3 mL of concentrated nitric acid (trace metal
grade) per 97 mL of sample. Solid samples were digested
by EPA Method 3052 prior to ICP-MS analysis. Known
quantities of arsenic, selenium, cadmium, and lead were
also added to sample aliquots for analytical matrix spikes.

2.3.6.1. ICP-MS Analysis at Vanderbilt
ICP-MS analyses carried out at Vanderbilt University (De-
partment of Civil and Environmental Engineering) were
completed using a Perkin Elmer model ELAN DRC II in
both standard and dynamic reaction chamber (DRC) modes.
Standard analysis mode was used for Pb, and DRC mode
with 0.6 mL/min of methane as the reaction gas was used
for As and Se. Nine-point standard curves were used for
an analytical range between approximately 0.1 and 500
|ig/L and completed daily. Analytical blanks and analyti-
cal check standards at approximately 50 |ig/L were run
every 10 samples and required to be within 10% of the
specified value. Samples for analysis were diluted gravi-
metrically to within the targeted analytical range using 1 %
v/v Optima grade nitric acid (Fisher Scientific). Typically,
analysis for As, Pb, and Se required 10:1 dilution. Twenty
microliters of a 10 mg/L internal standard consisting of
indium (In) (for As and Se) and holmium (Ho) (for Pb)
was added to 10 mL of sample aliquot prior to analysis.
Analytical matrix spikes were completed for As, Pb, and
Se on one of each of the three replicate extracts from
SR002.1. For each analytical matrix spike, 20 |iL of a 10
mg/L standard solution was added to 10 mL of sample ali-
quot (effective concentration addition of 200 |ig/L). Table
9 provides the element analyzed, analytical mode, corre-
sponding internal standard, method detection limit (MDL),
and minimum level of quantification (ML).

2.3.6.2. Severn Trent Laboratories, Inc. (STL)
STL (Savannah, GA) was selected as a commercial labo-
ratory to carry out some of the ICP-MS analyses for this
project. Analyses for As, Cd, Se, and Pb were performed
Table 9. Detection Limits and Quality Control Information
for ICP-MS Analysis for As, Pb, and Se at Vanderbilt.
Element
As
Pb
Se
Mode
DRC
Standard
DRC
Internal
Standard
20 ug/L In
20 ug/L Ho
20 ug/L In
MDL
(ug/L)
0.64
0.31
0.52
ML
(ug/L)
3.0
1.0
2.0
on an Agilent ICP-MS with octopole reaction system
(ORS). Mixed calibration standards were prepared for each
metal at five levels ranging from 0.5 |lg/L to 100 |lg/L.

2.3.7. X-Ray Fluorescence (XRF)
XRF analysis was performed on each CCR to provide ad-
ditional information on its total elemental composition. For
each CCR, two pellets were prepared as follows. Three
grams of material was weighed and mixed with 1.5 mL
(100 mg dry solids) of liquid binder to give a 32 mm diam-
eter pellet weighing 3150 mg with a material-to-diluent
ratio of 0.05. For high carbon content samples, 3.0 ml (100
mg dry solids) of liquid binder was used to give a 32 mm
diameter pellet weighting 3 3 00 mg with a material to diluent
ratio of 0.1. XRF intensities were collected on each side of
each pellet using Philips SuperQ data collection software
and evaluated using Omega Data System's UniQuant 4
XRF "standardless" data analysis software. The UQ/Flyash
calibration was used to analyze the samples. The pellets
were evaluated as oxides. Known flyash Standard Refer-
ence Materials (SRMs) were also run to assess the accu-
racy of the analysis. This information is useful in supple-
menting CVAA and ICP results.

2.3.8. MDL and ML for Analytical Results
The MDL is defined by 40 CFR Part 136, Appendix B,
July 1, 1995, Revision 1.11 as "the minimum concentra-
tion of a substance that can be measured and reported with
99% confidence that the analyte concentration is greater
than zero and is determined from analysis of a sample in a
given matrix containing the analyte."

The MDL was determined statistically from data gener-
ated by the analysis of seven or more aliquots of a spiked
reagent matrix and verified by the analysis of calibration
standards near the calculated MDL according to EPA
(2003). The MDL then was determined by multiplying the
standard deviation of the replicate measurements by the
appropriate Students t value for a 99% confidence level
(two tailed) and n-\ (six) degrees of freedom and also
multiplying by the minimum dilution factor required for
matrix preservation and analysis.

The ML is defined by 40 CFR Part 136, 1994 as "the low-
est level at which the entire analytical system must give a
recognizable signal and acceptable calibration point for the
analyte." According to EPA (2003), the ML is intended to
be the nearest integer value (i.e., 1, 2 or 5 x 10«, where n is
an integer) to 10 times the standard deviation observed for
determination of the MDL. This value is also multiplied
by the minimum dilution factor required for preservation

-------
Characterization of Coal Combustion Residues
and analysis of the sample matrix to obtain the ML re-
ported here.

Mercury, as measured by CVAA, required modification of
the calculation of the MDL and ML because very consis-
tent replication resulted in calculation of a MDL lower than
the instrument detection limit. For this case, the standard
deviation of seven replicate analyses  of 0.025  |ig/L was
0.00069. Therefore, the MDL was set equal to the instru-
ment detection limit of 0.001 |lg/L times the minimum di-
lution factor from sample preparation (3.59) to result in an
MDL of 0.0036 |ig/L. The ML was set to 10 times the
instrument detection limit and rounded to the nearest inte-
ger value as above. The resulting ML was 0.01 |ig/L.

2.4.  Quality Assurance Validation

2.4.1. Homogenization of Individual OCR
Samples andAliquots for Analyses
To ensure sample homogeneity the fly ashes were mixed
using aMorse single can tumbler model 1-305 (Figure 5).
This tumbler is designed to provide aggressive corner-over-
corner mixing at 23 RPM. Because the sample is tumbled
at an angle it yields superior mixing to a conventional tum-
bler. Briefly, each fly ash was mixed by filling a 5 gal bucket
to the halfway mark and tumbled for 1 hr. The bucket was
then inverted and tumbled for another hour.

At the beginning of this program a series of test were con-
ducted to ensure that the samples were being adequately
Figure 5. Mixing Fly Ash Prior to Obtaining Aliquots for
Laboratory Analyses.
24
mixed. The reference fly ash was mixed as outlined above
and 3 sub-samples taken from the top, middle and bottom
respectively and XRF pellets prepared. The XRF results
showed that the concentrations of 28 elements including
calcium and  silicon were consistent from  sub-sample to
sub-sample (Table 10)

2.4.2. Leaching Test Methods and Analytical
QA/QC
One of the objectives of this project was to establish a QA/
QC framework for the leaching assessment approach de-
veloped by Kosson et al. (2002). The developed QA/QC
framework incorporates the use of blanks, spiked samples,
and replicates, and Appendix C provides the complete
Quality Assurance Project Plan. For each designated leach-
ing test condition, triplicate leaching test extractions were
obtained (i.e., three separate aliquots of CCR were each
extracted at the designated test condition). The three types
of method blanks were the  deionized water case, the most
concentrated nitric acid addition case, and the most con-
centrated potassium hydroxide addition case. Each method
blank was carried through the entire protocol, including
tumbling and filtration, except an aliquot of CCR was not
added.

During analysis for mercury and elemental species by ICP-
MS, analytical spikes for the constituents of interest were
carried out for one replicate of each test case to assess ana-
lytical recoveries over the  complete range  of pH and liq-
uid matrix conditions. Using a standard obtained from a
source different from the calibration standards, multipoint
calibration curves using at least 7 standards and an initial
calibration verification (ICV) were completed daily or af-
ter every 50  samples, whichever was more frequent. In
addition, instrument blanks  and continuing calibration veri-
fication  (CCV) standards  were  analyzed after every 10
analytical samples and required to be within 10 percent of
the expected value. CCV standards and instrument blanks
also were run at the end of each batch of samples.

For both ICP-MS and CVAA analyses, each sample was
analyzed along with a matrix spike, which is an aliquot of
the sample plus a known spike concentration of the ele-
ment of interest.  The "spike recovery" should be within
80-120% of the expected value.

2.4.3. Laboratory Mass Balance Verification
for Leaching Test Methods
Mass balance analysis around the SR002.1  Solubility and
Release  as a  Function of pH leaching test procedure was
used to demonstrate retention of Hg, As, Se, Cd, and Pb

-------
                                                             Characterization of Coal Combustion Residues
Table 10. Total Content Analysis Results for Reference Fly Ash after Mixing (mercury analysis by Method 3052 followed
by analysis with CVAA).
                                                                                Standard
    Element          #1             #2             #3            ,0?1"        Deviation       %RSD
Si
Al
Fe
K
Ti
Ca
Mg
Na
S b
PC
X
Ba
Sr
Zr
V
Zn
Er
Cu
Cr
Y
Mn
Ga
As
Rb
Co
Ni
Pb
Sc
Hg
26.41
14.62
5.31
2.43
0.871
0.834
0.638
0.351
0.172
0.0834
0.107
0.0846
0.0509
0.0243
0.0219
0.0224
0.0173
0.0144
0.0162
0.0125
0.0073
0.007
0.008
0.0064
0.0057
0.0059
0.0033
51 ng/g
26.41
14.64
5.29
2.44
0.871
0.847
0.645
0.352
0.164
0.0854
0.106
0.0846
0.0486
0.0236
0.0243
0.0248
0.0173
0.0122
0.0153
0.0099
0.0073
0.0072
0.0079
0.0072
0.007
0.0044
0.0028
61 ng/g
26.43
14.52
5.38
2.46
0.877
0.835
0.645
0.34
0.172
0.0881
0.112
0.085
0.052
0.0234
0.025
0.024
0.0205
0.0122
0.0154
0.0108
0.0057
0.0072
0.0091
0.009
0.008
0.007
0.0025
66 ng/g
26.4167
14.5933
5.3267
2.4433
0.8730
0.8387
0.6427
0.3477
0.1693
0.0856
0.1083
0.0847
0.0505
0.0238
0.0237
0.0237
0.0184
0.0129
0.0156
0.0111
0.0068
0.0071
0.0083
0.0075
0.0069
0.0058
0.0029
59.3333 ng/g
0.011547
0.064291
0.047258
0.015275
0.003464
0.007234
0.004041
0.006658
0.004619
0.002359
0.003215
0.000231
0.001735
0.000473
0.001626
0.001222
0.001848
0.00127
0.000493
0.00132
0.000924
0.000115
0.000666
0.001332
0.001153
0.001305
0.000404
7.637626 ng/g
0.04
0.44
0.89
0.63
0.40
0.86
0.63
1.92
2.73
2.75
2.97
0.27
3.44
1.99
6.85
5.15
10.06
9.82
3.16
11.93
13.65
1.62
7.99
17.68
16.71
22.63
14.10
12.87
a Unless otherwise noted.
b Sulfur in oxidized form such as sulfate.
c Phosphorus in oxidized form such as phosphate.
during testing through the analysis of the EPA reference
fly ash. Six extraction conditions reflecting six different
extraction pHs were completed in triplicate. Figure 6 pro-
vides a flow diagram of the approach used to carry out a
mass balance analysis. This flow indicates the steps used
for completing the necessary analysis of one leaching test
condition (pH, LS ratio, and CCR) to assess the mass bal-
ance for mercury and other species of interest. The steps
indicated in solid lines were already incorporated in the
leaching assessment approach by Kosson et al. (2002). The
steps indicated in dotted lines were added during this project
to complete mass balance evaluation.

After the samples for each evaluated leaching condition
had been tumbled for the appropriate amount of time, they
were  each filtered and then divided into three fractions:
liquid, solid, and filter. In addition, glass containers used
in the procedure were rinsed with nitric acid after use, and
                                                  25

-------



87% Leachate
3% HNO3
5% KMnO4



Liquid





97% Leachate
3% HNO3
One pH Sample at One
LS Ratio of One CCR
,
1
.....A....
r 	 j Solid • — - I Fi
i •. 	 ; 1 1 	
	 I 	 A 	 	 }
Thermal Di9est bV D|9e
1 Analysis for 	 Method 	 " Mne,
Hgby ...3°?2B... ...30
          Clear Solution
              with
         Hydroxylamine
           (5% of total
            volume)
                        Spike
                       with 0.2
                       ppb Hg

a
.2
g



ICP
Method
 7473
                                                             CVAA
                                ICP
                                           Spike with 2
                                          ppm of As, Se,
                                             Pb, Cd
  Solid Lines - Standard Procedure
  Dashed Lines - Additional Analyses
       for Mass Balance
          Spike with
          0.2 ppb of
             Hg
                                                                   T
          Digest by
         OH method
          for KMnCv,
: Spike with 0.2 ppb;
I   As, Se, Pb, Cd  •
I       level       :
                     CVAA
           ..A..
            ICP
Spike with 0.2 ppb
 Hg and 2 ppm of
 As, Se, Pb, Cd
                                                                                                                                   o
                                                                                                                                   a
                                                                                                                                   5
                                                                                                                                   a
                                                                                                                                   ID
                                                                                                                                   N'
                                                                                                                                   5*
                                                                                                                                   5'
                                                                                                                                   3
                                                                                                                                   a
                                                                                                                                   o
                                                                                                                                   o
                                                                                                                                   0
                                                                                                                                   3-
                                                                             0

                                                                             DO
                                                                             (D
                                                                             M
                                                                             a
                                                                             c
             T
                                                                  CVAA
ICP

CVAA
                                                          ICP
Figure 6. Flow Diagram for Mass Balance and Quality Control on Laboratory Leaching Procedures.

-------
Characterization of Coal Combustion Residues
the resulting rinse was analyzed to verify that there were
no significant constituent losses to container walls. Each
fraction was independently analyzed to evaluate the mass
balance. Aliquots of each liquid sample were preserved
separately for mercury analysis and for analysis by ICP-
MS as described earlier. For the solid, an aliquot was
sampled for each case and digested according to EPA
Method 3052 (EPA, 1996). For the filter, the entire filter
with any retained solids was digested according Method
3052. Each digest and liquid sample was then analyzed in
triplicate for mercury using CVAA as described earlier.
Each digest and liquid sample also was analyzed in tripli-
cate for As, Se, Cd, and Pb by ICP-MS. In addition to each
liquid or digest sample analyzed, matrix spikes were used
to determine matrix effects on the analytes of interest (Hg,
As, Se, Cd, and Pb). For mercury spikes, the spiked solu-
tion then was digested by ASTM method D6784-02 prior
to analysis by CVAA.

The mass balance recovery was calculated for each test
case according to
%R =
m,
~ + mw
X 100
m
where %R is the percent recovery and mL, ms, mp, and mT
are the masses of the species of interest in the liquid phase,
solid phase, filter, and the total content initially in the
sample (based on independent analysis of the "as received"
EPA reference fly ash), respectively.

2.4.4. Improving QA/QC Efficiency
Throughout the study, the approach to QA/QC was regu-
larly reviewed for opportunities to increase evaluation ef-
ficiency without unacceptably degrading precision or ac-
curacy in results. Based on evaluation of results from
the first several facilities, the number of replicates for
Method SR002.1 (solubility as a function of pH) and
Method SR003 . 1 (solubility as a function of LS ratio) were
reduced from three to two. Study results have shown that
the precision between duplicate analyses is acceptable and
that a triplicate set does not significantly increase the qual-
ity of the data set. This finding follows from (i) the data
sets generated by Method SR002. 1 and SR003 . 1 must pro-
vide both consistency between replicate extractions and
analyses and internal consistency between results at dif-
ferent pH and LS ratio and (ii) precision is controlled pri-
marily by the degree of homogeneity of the CCR under
evaluation and representative sub-sampling, rather than by
the intrinsic variability of the leaching test methods. Re-
ducing the number of replicates has greatly improved labo-
ratory efficiency without compromising data quality.
2.5. Interpretation and Presentation of
Laboratory Leaching Data
Complete laboratory leaching results for Brayton Point,
Pleasant Prairie, Salem Harbor, Facility C, and St. Clair
are presented in Appendices D through H, respectively. For
each facility, a common format is used for presenting re-
sults. First, a titration curve of pH as a function of
milliequivalents of acid or base added is presented, with
acid additions considered positive (+) and base additions
considered negative (-). The titration figure is then followed
by a curve of pH as a function of LS ratio. The pH curves
are then followed by a series of figures for each species of
interest (i.e., mercury, arsenic, and selenium). The results
from Solubility and Release as a Function of pH (SR002.1)
are presented first, followed by the results from Solubility
and Release as a Function of LS ratio (SR003.1).

For Solubility and Release as a Function of pH (SR002.1),
results for the baseline case are presented side by side with
the results from the case with enhanced mercury control.
Results are presented as extract concentrations as a func-
tion of pH. Total content of the species of interest is pro-
vided above the first figure for that species. The natural
pH13 of the system is indicated as a vertical line to the av-
erage pH and a horizontal line to the j-axis indicating the
corresponding extract concentration. Included with each
figure are horizontal lines at the drinking water maximum
concentration level (MCL) and ML and MDL analytical
limits to provide a frame of reference for the results. Also
included with each figure is the 5 and 95 percentile for pH
(vertical lines) and for constituent concentration (horizon-
tal lines) from field observations of leachate from landfills
for combustion residues (Table 11; EPA, 200014; EPRI
2005), forming a rectangular box that encloses the corre-
sponding domain of field leachate observations. An anno-
tated example of the results is provided as Figure 7. Fig-
ures with corresponding analytical recoveries are provided
below the concentration results.

For Solubility and Release as a Function of LS ratio
(SR003.1), results are presented as extract concentrations
13 "Natural pH" of a material refers to the equilibrium pH when
the material is placed in deionized water at a ratio of f 0 g CCR
per fOO mL of water.

14 The EPA data represent six ash landfills for which data were
available. These data were not collected as nationally represen-
tative although they do portray the range of pH values also found
intheEPRfdata.
27

-------
Characterization of Coal Combustion Residues
                               5th %tile for pH
          Concentration
          at natural pH
        SR2-BPT-0001 -A
  SR2=SR002.1
                    Replicate A
         Sample name



\
_l
s
-------
                                                              Characterization of Coal Combustion Residues
ary material, relatively weakly adsorbed, and the presence
of complexing and/or competing species in solution are at
a constant concentration. For this case, leaching test re-
sults will indicate a constant concentration as a function of
pH at a fixed LS ratio and linearly increasing concentra-
tion as LS ratio decreases at constant pH. This case is rep-
resented mathematically as a linear equilibrium partition-
ing function, where the critical constant of proportionality
is the partitioning coefficient, Kd. Linear partitioning and
use of Rvalues is a common approach for mathematically
modeling contaminant transport at low contaminant con-
centrations in soils. It is a valid and useful  approach when
the necessary conditions (discussed above) are fulfilled.15

For mercury adsorbed on activated carbon, a complex com-
bination of adsorption mechanisms is indicated. During
laboratory leaching tests, mercury concentrations in the
leaching test extracts are relatively constant over the pH
range and LS ratio of interest and independent of total
mercury content in the CCR. In addition, the total mercury
content in the CCR is very low. These results indicate ad-
sorption phenomena where, in the adsorbed state, interac-
tions between adsorbed mercury species are stronger (ther-
modynamically) than the interactions between the adsorbed
mercury species and carbon surface.16 This observation has
been supported by the observation of mercury dimmer for-
mation during sorption (Munro et al., 2001) and the occur-
rence of chemisorption as the dominant adsorption mecha-
nism at temperatures above  75  °C  (consistent with condi-
tions in air pollution control devices; Vidic, 2002). In other
studies, this phenomenon has been observed as the forma-
tion  of molecular clusters on the  adsorbent surface
(Ruthven, 1984; Duong,  1998;Rudzinskietal, 1997). For
this case, use of a Kd approach would underestimate re-
lease because desorption is best represented as a constant
aqueous concentration until depletion occurs.

A third case is when the constituent of interest is present in
the waste or secondary material (e.g., CCR) as  a primary
15 Often specific Rvalues are a function of pH because of com-
petition for adsorption sites by hydrogen ions. However, a single
Kd or range of Kd values are often used in contaminant fate and
transport models without specific relationship between pH and
Kd.

16 For this case, the first mercury molecule is adsorbed more
weakly than subsequent mercury molecules because the adsorbed
mercury-mercury interaction is stronger than the adsorbed mer-
cury-carbon surface interaction.
or trace constituent in either an amorphous or crystalline
solid phase and there may be complexing or chelating co-
constituents in the aqueous phase. Observed aqueous con-
centrations are a non-linear function of pH and LS ratio
and reflect aqueous saturation with respect to the species
of interest under the given conditions (pH, co-constituents).
For these cases, an approximation of field conditions can
be made empirically based on  laboratory testing and ob-
served saturation over the relevant domain (as applied in
this report), or geochemical speciation modeling coupled
with mass transfer modeling  can be used to assess release
under specific field scenarios (the subject of a future re-
port). Use of a Kd approach would not be appropriate for
these cases because constituent concentrations will remain
relatively constant at a given pH until the controlling solid
phase is depleted and control is shifted to a new solid phase
or mechanism.

2.6.  Long-Term  Release Assessment
Long-term constituent release  estimates were developed
to evaluate the potential cumulative impacts of different
CCR management scenarios. A scenario of disposal in a
combustion waste landfill and three default scenarios were
examined. These scenarios were selected to provide upper
bounding estimates of release considering:
  • the range of field observations (pH and LS ratio15) for
    analogous impoundments or landfills of combustion
    wastes,
  • constituent release if occurring at the material's natu-
    ral pH, and,
  • constituent release if occurring at extreme acidic or
    alkaline pH.

A 100-year time interval was selected as a convenient ref-
erence period because it is beyond a lifetime but within a
comprehendible period. Cumulative release estimates are
provided on the basis of mass of constituent released per
17 For field scenarios, LS is directly a function of time (f), infil-
tration rate (inf), landfill depth (Hfla), and fill density (r) accord-
ing to
            L
inf (cm /yr) • t (year)
                  . Alter-
natively, LS can be related to pore volumes of water passing
through the CCRs (where Aftu is  the fill area) according to

                 Pore Volum e(L)
T S   	  =
          -
                                 m
                                                                                                          29
-------
Characterization of Coal Combustion Residues
mass of CCR disposed (|ig X/kg CCR).18 These estimates
are not intended to be absolute predictions of release, but
rather, an initial assessment of whether further evaluation
is warranted. These estimates for the monofill disposal sce-
nario assume local equilibrium, which is a conservative
assumption (i.e., estimated release is greater than actual
expected release). A more refined assessment can be made
using results from column leaching tests or diffusion con-
trolled leaching tests that will allow consideration of re-
lease kinetics in developing field release estimates. The
mass of constituent (e.g., As) that would be released if all
of the leachate percolating through the landfill for the given
scenario were at the MCL is provided as a reference value.
The  estimates presented here are only for constituent re-
lease from the waste and do not account for any dilution or
attenuation that would occur in the vadose zone or ground-
water or for the impact of a landfill liner.

For the scenario of disposal in a combustion waste land-
fill, a historical data set of typical leachate generated from
this landfill type was taken from the comprehensive data-
base of landfill leachate characteristics developed by the
EPA's Office of Solid Waste (EPA, 2000). Cumulative re-
lease estimates were  developed according to the method-
ology developed by Sanchez and Kosson (2005), and val-
ues of leachate pH were used to derive the probability dis-
tribution function of the field pH. Annual leachate genera-
tion quantities observed for industrial co-disposal landfills
were used to derive the probability distribution function
LS ratio that may be expected to contact the fill over the
estimated time period of 100 years. For each data set (field
pH and LS ratio), different distribution functions were used
to fit the data, and the one providing the best data fit based
on the chi-square test was selected.  The resulting field pH
probability distribution then was truncated and normalized
to the pH range of the field data (Figure 8). The distribu-
tion  for field pH was the result of over 158 sample obser-
vations from coal combustion residue disposal facilities at
six sites. The probability distribution for the LS ratio was
the result of over 41  sample observations from Industrial
D landfill facilities at 17 sites. (Figure 9).

For each CCR tested, results from SR002.1 (Alkalinity,
Solubility and Release as a Function of pH) were used to
develop an empirical functional relationship between so-
lution pH and expected concentration for mercury, arsenic,
and  selenium. Laboratory results of mercury  concentra-
                                 5   6   7   8   9   10  11   12  13  14
PH
n Field d
Fitted c
Simula


pH min
pH - 5%
pH - 50%
pH - 95%
pH max
Field data
5.40
5.80
7.7
12.09
12.80
ata
distribution (Logistic, (1))
ted (2)
Fitted
distribution
4.74
6.00
7.69
11.62
+infinity
Simulated
4.92
5.97
7.63
10.63
12.50
18 These release estimates can be converted to the amount re-
leased per unit area according to
M, [mg/m2] = M, [mg/kg] • p [kg/m3]
30
•Hfla[m].
Figure 8. Leachate pH Distribution: Scenario of Disposal
in a Combustion Waste Landfill.

tions typically showed a high degree of variability between
measured and non-detected values in the laboratory leach-
ing test extracts (Figure 10). This was likely due to micro-
scale sample heterogeneity with respect to carbon distri-
bution. However, the values of the measured mercury con-
centrations for a specific CCR typically did not vary sig-
nificantly as a function of pH. Therefore, as an upper bound-
ing approximation for each specific CCR, the expected
mercury  concentration over the expected field pH range
was set to the maximum observed extract concentration
over the anticipated field pH range for that CCR. As a re-
sult of this approach, all expected release of mercury should
be viewed as less than or equal to the indicated value at the
indicated percentile.

For arsenic and selenium,  a polynomial function was re-
gressed to the results from SR002.1 (Alkalinity, Solubility
and Release as a Function of pH) with each CCR case to
provide the expected leachate concentration as a function
of solution pH (Figure 11). The regression fits and corre-
sponding equations for solubility as a function of pH are
provided in the appendices for each case  examined (i.e.,
for each constituent in each CCR tested).
-------
                                                            Characterization of Coal Combustion Residues
               0.5      1      1.5
                    LS ratio [L/kg year]
                           2.5
LS 1 }
n Field dt
	 Fitted c
Simula

fear
LS min
LS - 5%
LS - 50%
LS - 95%
LS max
Field data
1 .OE-05
5.5E-04
0.06
1.50
2.50
3ta
istribution (Gamma, (1))
ed (2)
Fitted
distribution
3.3E-04
4.9E-04
0.08
1.07
+infinity
Simulated
3.3E-04
4.4E-04
0.08
1
1.99

Figure 9. LS Ratio Distribution: Scenario of Disposal in a
Combustion Waste Landfill.
MCL
1
U
0) 0.1
I
Om


n nm


n nn n

=_ . — .
«\ /so. 
-------
Characterization of Coal Combustion Residues
100000
10000 -
1000 -
u 100 -
S 10-
2 1-
0.1 -
0.01 -
n nni -

^_



,_.0r___n
	 "~—--^i-u!

A
f
J

100000
10000 -
1000 -
95% 3 100-
1 10-
5% c/>
< 1 -
0.1 -
0.01 -
n nn-i
& . . 	



&S
^_^__^aa


^

95%
5%
1 b%' • • -9b% U-UU1 ' 5%' ' ' '95%
2 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH
A)

n SR2-BPB-0001-A
0 SR2-BPB-0001-B
A SR2-BPB-0001-C
Fit PI irvp

B)
n SR2-BPT-0001-A
0 SR2-BPT-0001-B
A SR2-BPT-0001-C


Material log As (ng/L) pH range of R2 Number of
validity points
BPB 0.0004 pH5 -0.0135 pH4 0.1634 pH3 3-14 0.77 27
-0.8130 pH2 1.1609PH 2.7085
BPT 0.0005 pH5 -0.0207 pH4 0.3035 pH3 3-12.5 0.98 33
-2.0113 pH2 5.4552 PH -2.7126


Figure 11. Example of Regression Fits and Corresponding Equations for Solubility as a Function of pH.
    Mt = Cumulative
    release of selemium
    over 100 years
    (Mg Se/kg OCR)
         BPB = Baseline case
Selenium
mnnnn -,
mnnn -
2
"o) 1000 -
-i— »
^ 100 -
i_
03
>, -in -
o
° y
/ n A
/ «




1
^

)



^
AftU
!i

5%20 40 5C

6*
k6A^6*
\
\


°A60 80 "^C
                                                  Percentile
                                            nBPB
                                                                                BPT = Case with
                                                                                enhanced mercury
                                                                                control
    Figure 12. Example of Cumulative Probability Distribution for Release of Selenium from Brayton Point CCR.
32
-------
                                                            Characterization of Coal Combustion Residues
    100000
     10000
CD
.CU

£

.0

0)


CD
s_
CD
 O
 o
1 00 yr cumulative release for
^^^ . . default case of leaching at pH 3
RPR A re ^ n i r* s
80500 ng/kg " / (in Mg As/kg CCR and as % of
10Qn7 ,/ total As content in CCR)
:
\
!
: 	
Tot
	
	
al cont
/
20



95 ng/kg 2439 ^g/kg
9R% 3.0%











105^/kg 83 ^g/kg




ent Combustion Default-pH3 Default-pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 100
As
pe
fiel
. K/IPI
suming the 95th
•centile LS from
d data
' -\ nnn i ,^/i^^
MULLS95% ~ IWWW ny^y
MCLIC; ..-125 ua/ka
Cumul
release
years i
atMCL
for give
alive mass
id over 1 00
release occurs
concentration
>nLS
      1000 -=	
       100
       10
                                        Based on Monte Carlo simulation
                                        using field pH and pH dependent
                                        leaching results


Figure 13. Example of Comparison of 100 yr Cumulative Release Estimates for Arsenic.
                                                                                                       33
-------
Characterization of Coal Combustion Residues
34
-------
                                                           Characterization of Coal Combustion Residues
3. Results and  Discussion
3.1.  Leaching  Characteristics from
Field Observations of CCR Landfills
and Impoundments
In response to concerns raised by U.S. EPA Science Advi-
sory Board regarding leaching tests, observations of pH,
and concentrations of mercury, arsenic, and selenium from
field CCR management facilities were evaluated for com-
parison with laboratory results. Landfills and impound-
ments were the types of CCR management facilities con-
sidered. Information was available from a U.S. EPA data-
base (EPA, 2000) and an EPRI database (EPRI, 2005).

The U.S. EPA database (EPA, 2000) compiled by OSW
included data on six CCR monofills. Data included mea-
surements of pH,  mercury, arsenic, selenium and other
constituents as self reported by facilities to Office of Solid
Waste (OSW).19 Data in this  database was not coded to
allow association of different parameters (e.g., pH, mer-
cury, arsenic, selenium) from the same field sample. There-
fore, data from this database was evaluated based only on
the distribution of measurements for each parameter for
the class of CCR monofills (Table 12). Mercury data were
carefully reviewed, including re-evaluating the primary
Table 12. Distribution of pH and Concentrations of Arsenic
and Selenium from Field CCR Management Facilities from
the U.S. EPA Database (EPA, 2000).
Parameter
median
5th %
95th %
PH
7.7
5.8
12.09
As
(M9/L)
8.15
2.0
140.0
Se
(M9/L)
19.5
2.4
170.0
source data, and were not considered to be reliable and,
therefore, not included.20

The EPRI database (EPRI, 2005) included measurements
of samples obtained from CCRlandfills and impoundments.
Samples were from leachate collection points, lysimeters,
and pore water. For some facilities, multiple observations
were obtained from different locations within the facility
and over several years. Results from a recent sampling and
analysis program were observations from a range of facili-
ties (considered representative of management practices,
combustion facility configurations, and coal types) but with
only one or two samples per facility. Data included mea-
surements of pH,  mercury, arsenic,  selenium, and other
constituents. Data on selenium were more limited than data
on pH and arsenic. Data on mercury were limited to the
recent sampling and analysis program from multiple fa-
cilities. Data was coded to allow association of different
parameters. Only CCR management facilities that receive
residues from utilities that do not include scrubbers as part
of the air pollution control technology are considered here.

Information on pH from field observations is presented in
Figure 14. For landfills, the range of data in the U.S. EPA
database (5^-95^ percentile) was consistent with the data
reported in the EPRI database. In addition, the range of pH
reported in the EPRI database for individual facilities with
multiple observations was similar to the range reported for
multiple facilities  with limited numbers of observations.
Therefore, it is considered reasonable to use the pH range
of 5.8 to  12.09  (5^-95* percentile) reported in the U.S.
EPA database as the basis for extrapolating from labora-
tory leaching test results to field estimates of leaching from
landfills.
19 As noted previously, data from six coal combustion ash land-
fills were collected based on their availability and are not neces-
sarily representative of all coal combustion ash landfills.
20 Inconsistent methodologies were used, and most values were
either qualified results (e.g., estimated values) or below detec-
tion limits (with relatively high corresponding detection limits
reported).
                                                35
-------
Characterization of Coal Combustion Residues
c
Landfills
AM AA
ex )x x ] i see i ) )xc x xc tc x t »] x »))x )
_l_l__l__LI_ _l_l__l_ _|l_ _[_!_ J 	 l_
_j_j j_^_ _j_j j_ _|j_ _j_ j_ _j j_
1 1 1

) 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Field pH

* Facility 1
(subbit.,
western)
• Facility 2
(subbit.,
Wyoming)
A Facility 3
(subbit.,
Wyoming)
+ Multiple sites
(single
observations)
-*- EPA data (5th-
95th percentile)

Impoundments














>H8SK XXXXX

A/WWWWVAAA

>8CXX X

CD •
A ^.m. AJk AA tmttttM

• • M
-------
                                                             Characterization of Coal Combustion Residues
base. However, the uppermost bound of arsenic concen-
tration in the EPRI database exceeds the U.S. EPA data
range illustrated. In addition, the range of arsenic concen-
trations reported for individual facilities with multiple ob-
servations was similar to the range reported for multiple
facilities with limited numbers of observations. The great-
est arsenic concentrations are reported in the pH range be-
tween 7 and 10.
For impoundments, the upper range of arsenic concentra-
tions is substantially greater than reported for landfills. A
significant number of reported arsenic concentrations are
between 1,000 and 10,000 |Jg/L for impoundments, whereas
all reported concentrations are less than 1,000 |Jg/L for
landfills. Greater observed concentrations may be from
leaching of arsenic naturally associated with pyrite in coal
mill tailings co-disposed with CCRs rather than from the
-innnn -i

-i nnn

"•~r- -i nn
D :
<-i n

•1

n 1
c
Landfills


* ^ +
r^£ f '^•AA ^"^^^ ~I
•* ++*Vi+v ^
l+,++*rf* • • *•*
"* *~


) 1 2 3 4 5 6 7 8 9 10 11 12 13 1
PH










4


^ Facility 1 (subbit
western)
• Facility 2 (subbit.,
Wyoming)
A Facility 3 (subbit.,
Wyoming)
+ Multiple sites
(single
observations)


Impoundments
1 nnnn
-I nnn
T -inn -
"3> |
(0
<1 n -
-I
0 1
X A
yX. A »
' AX L * A *
A 1 >
AA^^T > ^
A A^ A ^ »*
A>£+/*
/ A A • *, X **+
AA 3, XA+^K^|+ °
-*- -H-
A D _L +
A / + *


0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
PH

n Facility 4 (subbit.,
Wyoming)
» Facility 5 (bit.,
Kentucky)
A Facility 6 (bit.,
eastern w/pyrite)
• Facility 7 (bit.,
Kentucky, western)
x Facility 8 (subbit.,
western)
A Facility 9 (bit.,
Kentucky, eastern)
Facility 10 (bit.
w/pyrite)
+ Multiple sites (single
observations)
Figure 15. Arsenic Concentrations Observed in Field Leachate at Landfills and Impoundments Used for Disposal of
CCRs—from EPRI database. Also indicated by dashed lines is the range (5th-95th percentile) of pH and arsenic values
for CCR landfills reported in the EPA database (EPA, 2000). Facilities 6 and 10 have co-disposal of pyrite from mill rejects
with CCRs. Data does not include facilities with scrubbers. Primary data from EPRI (EPRI, 2005).
                                                                                                        37
-------
Characterization of Coal Combustion Residues
CCRs. Alternatively, the significantly lower pH in the py-
rite co-disposal impoundments (Figure 14) may be caus-
ing the higher As concentrations (see Figure 13, leaching
at pH 3). EPRI has recommended alternative management
practices for coal mill tailings containing pyrite, and this
practice is diminishing (EPRI, 1999).
Information on selenium from field observations is pre-
sented in Figure  16. For landfills, the EPRI database in-
cludes a wider range (greater than and less than) of re-
ported concentrations than the U.S. EPA database. For
impoundments, the reported range of selenium concentra-
tions is within the same range as reported for landfills. For
-innnn -.

1000 :
•— r- -i nn -
^) I
o>
W-1 n -

-i _

0-1 _
c
Landfills
-H-
, + + +&{L

+ + . cP5
; 7: Pp
+_ Q nrf 0 f
%++•*•
m~r
._T____..__(___^[___,
+
,
) 1 2 3 4 5 6 7 8 9 10 11 12 13 1
PH









4


* Facility 1 (subbit.,
western)

Facility 2 (subbit.,
Wyoming)
A Facility 3 (subbit
Wyoming)
+ Multiple sites
(single
observations)


Impoundments

IUUU :
•— - -i nn
^) :
0)
c/5 -in
1 _
0 1 -

»
+ X «K *
* +
+•
4X»#
* fJ-4 *
+
+ +
++
+

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
PH


n Facility 4 (subbit.,
Wyoming)
» Facility 5 (bit.,
Kentucky)
x Facility 8 (subbit.,
western)
Facility 10 (bit.
w/pyrite)
+ Multiple sites
(single
observations)

Figure 16. Selenium Concentrations Observed in Field Leachate at Landfills and Impoundments Used for Disposal of
CCRs—from EPRI database. Also indicated by dashed lines is the range (5th-95th percentile) of pH and arsenic values
for CCR landfills reported in the EPA database (EPA, 2000). Facilities 6 and 10 have co-disposal of pyrite from mill rejects
with CCRs. Data does not include facilities with scrubbers. Primary data from EPRI (EPRI, 2005).
38
-------
                                                            Characterization of Coal Combustion Residues
both landfills and impoundments, the range of data reported
for a single facility (Facility 2) is fairly wide.

To balance the assessment of the EPRI data in comparison
with the EPA data and laboratory leaching test results, the
following data reduction steps were taken. For the facili-
ties with more than three observations, the mean value of
the observations from the individual facility was taken to
be representative of that facility. The mean value then was
included with the data of single observations from mul-
tiple facilities. The MDL was used in the data set when the
data were reported as less  than the MDL. The resulting
data set then was evaluated to obtain distribution statistics
for the EPRI data evaluated (Table 13). The median,  5th
and 95th percentiles for arsenic, selenium, and mercury then
were used to for comparison with the EPA data set and
laboratory results, as described later in this report.

3.2.  Quality Assurance for Laboratory
Leaching Tests

3.2.1. Mass Balance using EPA Reference
Fly Ash
The results of the Reference fly ash analysis are provided
in Table 14. These results show that the mass balance was
closed reasonably well for mercury, arsenic, cadmium, lead,
and selenium with a majority of the mass for each analyte
remaining in the solid. The mass balance closure is well
within the expected range, especially considering the mea-
surement of very small changes in the analyte mass in the
solid phase relative to the total content present. These re-
sults also indicate that large losses of mercury do not oc-
cur as a consequence of the leaching test methods and sub-
sequent analysis. However, additional mass balance veri-
fication may be required for implementation when testing
materials with more volatile components or for validation
of laboratories newly implementing the procedures.

3.2.2. Analytical Quality Control/Quality
Assurance
Implementation of the developed QA/QC plan facilitated
analysis of data quality and identification of testing uncer-
tainties. The coefficient of variation for  calibration  stan-
dards and continuing calibration standards and blanks was
within 5% for metals analysis by inductively coupled
plasma-mass  spectroscopy (ICP-MS) (Vanderbilt). Spike
recoveries for metals had a mean of 101% with a coeffi-
cient of variation within 5%. Typical inter-laboratory com-
parisons for arsenic and selenium analyses are presented
in Figure 17. Good agreement generally was obtained be-
tween the two laboratories except for cases of arsenic analy-
ses when the  concentration was less than 100 |Jg/L. For
this condition, the Vanderbilt analysis typically resulted in
greater values than the commercial laboratory. This result
was most likely from the differences in analytical ICP-MS
technology, where dynamic reaction chamber (DRC) mea-
surements (Vanderbilt) are considered more sensitive and
less susceptible to interferences. For all of these cases, the
spike recoveries for Vanderbilt analyses were within ±10%
of the expected value, with most cases within ±5% of the
expected value. However, the analytical results, including
repeating analysis when necessary, demonstrated the im-
portance of including a matrix spike to verify recovery for
each test condition. In contrast to the analytical uncertainty,
the mean (for different test conditions, i.e., pH values) co-
efficient of variation for replicate tests on each fly ash type
(resulting from variation in the subsamples of the solids
tested) varied between 5% and 25%.  Thus, the primary
source of uncertainty in the leaching test results is a conse-
quence of sampling, homogenization, and inherent hetero-
geneity of the primary material to be tested.
Table 13. Distribution of pH  and Concentrations of Arsenic, Selenium, and Mercury from Field CCR Management
Facilities—from the EPRI database (EPRI, 2005) for landfills and impoundments (including impoundments co-disposing
mill rejects with CCRs).
                                  Landfills
                     Impoundments
Parameter
Average
Median
Min
5th%
95th%
Max
As
(ng/L)
48.1
21.2
2.2
3.0
179.1
238.0
Se
(ng/L)
265.9
57.0
0.3
1.7
1733.0
1760.0
Hg
(ng/L)
0.0179
0.0102
0.0021
0.0021
0.0498
0.0606
As
(ng/L)
381.4
55.0
4.0
4.2
852.8
5223.0
Se
(ug/L)
50.6
18.5
0.2
0.6
278.6
315.0
Hg
(ug/L)
0.0019
0.0014
0.0002
0.0003
0.0056
0.0059
                                                                                                       39
-------
Characterization of Coal Combustion Residues
Table 14. Leaching (Method SR002.1) and Mass Balance Results for the EPA-Reference Fly Ash.
Metal
Hg (ng/g)
Pb (ug/g)
Se (ug/g)
Cd (ug/g)
As (ug/g)
«a
Received
59+8
84+2.5
2+0.3
1+0
87+2.6
Exposed
Solid
48+2
87+1.8
3+0.3
1+0
89+1.3
Leachate
solution
0.369+
0.011
BDa
BD
BD
BD
Leaching at pH of
Metal
Hg (ng/g)
Pb (ug/g)
Se (ug/g)
Cd (ug/g)
As (ug/g)
Received
59+8
84+2.5
2+0.3
1+0
87+2.6
Exposed
Solid
39+0.5
89+3.4
2+0.1
1+0
91+2.6
Leachate
solution
0.124+
0.215
BD
BD
BD
BD
Leaching at pH of
Metal
Hg (ng/g)
Pb (ug/g)
Se (ug/g)
Cd (ug/g)
As (ug/g)
Received
59+8
84+2.5
2+0.3
1+0
87+2.6
Exposed
Solid
44+2
87+2.6
2+0.4
1+0
92+0.6
Leachate
solution
0.115+
0.199
BD
BD
BD
BD
Filter
0.279+
0.233
8+1 1 .2
BD
BD
2+0.7
Recovery
%
76-91
94-132
1 1 7-1 94
100
99-1 1 0
4.0 (position 3)
Filter
0.045+
0.015
0+0.2
BD
BD
1+0.1
Recovery
%
60-75
99-114
83-124
100
106-112
10.0 (position 5)
Filter
0.019+
0.008
20+28
BD
BD
4+3.2
Recovery
%
70-82
88-169
70-141
100
103-118
Exposed
Solid
50+11
93+13.5
2+0.2
1+0
94+4.1
Leachate
solution
0.106+
0.184
BD
BD
BD
BD
Leaching at Natural
Exposed
Solid
46+8
89+6.3
3+0.3
1+0
90+3.4
Leachate
solution
0.122+
0.212
BD
BD
BD
BD
Leaching at pH of
Exposed
Solid
42+0.4
84+1 .5
1+0.1
1+0
86+0.8
Leachate
solution
BD
BD
BD
BD
BD
Filter
0.155+
0.083
2+2.2
BD
BD
1+0.6
Recovery
%
76-92
84-136
78-129
100
103-115
pH (position 4)
Filter
0.030+
0.003
2+1.2
BD
BD
2+1.5
Recovery
%
75-81
97-121
117-194
100
97-115
12 (position 6)
Filter
0.031 +
0.013
6+6.6
BD
BD
4+1.2
Recovery
%
63-82
95-120
39-59
100
98-121
1 BD = below detection limit.
3.3. Laboratory Test  Results
The  constituents of interest in this evaluation, based on
input from EPA-OSW and EPA-OAQPS, are mercury (Hg),
arsenic (As), cadmium (Cd), lead (Pb), and selenium (Se).
Initial screening indicated low content and leaching con-
centrations below levels of concern for lead and cadmium.
As a result, although complete data have been developed
for lead and cadmium, the results are not provided in this
report. Complete data also have been developed for other
constituents to facilitate evaluation  of geochemical spe-
ciation of constituents of concern and to provide more thor-
ough evaluation of leaching under alternative management
scenarios in the future, if warranted. Screening of leaching
results against drinking water maximum contaminant lev-
els (MCLs) indicates that antimony (Sb) may be a concern
for some cases and is being considered for inclusion in
future research. Complete results forpHtitration, mercury,
arsenic and selenium for each CCR reported here are pre-
sented in Appendices D through I.
40
For each CCR evaluated, results of the leaching tests pro-
vide the following information:
  • Leachate concentrations for the constituents of inter-
    est as a function of pH over the range of reported field
    management conditions (from test method SR002.1;
    example results provided in Figure 18 and Figure 19 A).
  • pH titration curves (from test method SR002.1). This
    information is useful in characterizing the CCR and
    assessing how it will respond to environmental stresses
    and material aging (e.g., carbon dioxide uptake, acid
    precipitation, mixing with other materials).
  • Leachate concentrations for the constituents of inter-
    est, pH and electrical conductivity as a function of LS
    ratio when contacted with distilled water (from test
    method SR003.1; example results are in Figure 19B).
    This information provides  insight into the  initial
    leachate concentrations expected during land disposal
    and the effects of pH and ionic strength at low LS ra-
    tios. Often these concentrations can be either greater
-------
                                                             Characterization of Coal Combustion Residues
                   Baseline Fly Ash
                                                            Fly Ash with Enhanced Hg Control
„ moo
2 100
o
O
_i
w m
1 _
|As
E
E
I/

A
j
^
A
ft
o3
X
A
9
[
i
^
X

H Rep A
0> RepB
1 RepC
                                                           10000
                                                            1000
                                                         o
                                                         c
                                                         o
                                                         O
                    10      100      1000    10000
                      VU Cone.
                                                                    10       100     1000     10000
                                                                      VU Cone.
    10000
     1000
  o
  c
  o
  O
  B)
100
            Se
                                         Rep A
                                                     10000
                                                  -r  1000
o
o
O
                    10       100      1000
                      VU Cone. [u.g/L]
                                       10000
                  10      100      1000
                    VU Cone. [^g/L]
10000
Figure 17. STL Versus Vanderbilt Analytical Results for Arsenic and Selenium from Sr002.1. Baseline Fly Ash and Fly
Ash with Enhanced Hg Control from Brayton Point Are Shown.
    than or less than concentrations observed at higher LS
    ratios (i.e., LS=10 mL/g as used in SR002.1) because
    of ionic strength and co-constituent concentration ef-
    fects.

The MCL is used as a reference threshold for the constitu-
ent of interest. However, releases identified here are esti-
mates of concentrations potentially leaching from landfills.
Any assessment of the environmental impact of these re-
leases needs to consider the dilution and attenuation of these
constituents in ground water and the plausibility of drink-
ing well-water contamination resulting from the release.
Dilution and attenuation factors for metals (DAFs) have
been estimated to be potentially as low as 2 to 10  on a
national basis or as high as 8,000 at a particular site with
                                                hydrogeology that indicates low transport potential.21
                                                Therefore, comparison with thresholds greater than the
                                                MCL and developed for specific scenarios may be appro-
                                                priate. The following comparisons are included for each
                                                CCR in Appendices D through I:
                                                  • Laboratory leachate concentrations as a function of pH
                                                    for each CCR are compared to (i) the constituent MCL;
                                                 21 See 60 FR 66372, Dec. 21, 1995, for a discussion of model
                                                 parameters leading to low DAFs, particularly the assumption of
                                                 a continuous source landfill. Implied DAFs for the metals of
                                                 interest here can be found at 60 FR 66432-66438 in Table C-2.
                                                 Site specific high-end DAFs are discussed at 65 FR 55703, Sep-
                                                 tember 14, 2000.
                                                                                                  41
-------
Characterization of Coal Combustion Residues
                  Baseline Fly Ash

           Total Hg content: 650.6+6.8 ng/g
IU
MCL
1 -
_i
TO 0.1 -
T
0 035
Orn
.Ul
Onm
/

	 ML
— - MDL


i B



i — . — .
 A ^
I 4





n n D



— . — _ — _ _
o> i
fe 8 1095°^
PH




DDD


•
, - —
y /w\/\
21Z21
I
•

                                              95%
                                              5%
                                                       "TO
                                                       D)
                                                            Fly Ash with AC I

                                                    Total Hg content: 1529.6+1.1 ng/g
IU
MCL
1
0.1
Orn
.U I
.UUD
Dnm
/
-
I
a °° a
z
•— 	 	
€\ IPQ ^
2 4 5%


	 n...n.n.n 	 	

________

6 8 9'51095%1




- 	
CXI
2 1
                                                                                 PH
                                                                                                    95%
ML
MDL
1
D
0
A
SR2-BPT-0001
SR2-BPT-0001
SR2-BPT-0001
-A
- B
-C
  CO
  <
      10000T
      MCL1
       6.67
             Total As content: 80.5+1.9 |jg/g
                                                     Total As content: 27.9+2.1 |jg/g
       0.001
   	ML
   — - MDL
L/VJ
DO
DO
10
7
1
.1
D1
11

cA
l ©
! AO

_ _ 	 . ,
|
5°,

& a

_ _ — _ — _ —

•1 ' ' 95%
A
S

i - —
1
.12.2
                                              95%
                                              5%
2    4     6    8    10   12    14
               PH          :
                    DSR2-BPB-0001 -A
                    OSR2-BPB-0001 -B
                    ASR2-BPB-0001 -C
1 \J\J\J\J
1000
100
=r MCLio
=d I U
TO 4.8
•S
« 0.1
0.01
Onm
.UU I \
e
£
-d
- ^S
: B
i - - - - 5^
_ . _ .

5°
I 4

*
- - ^- -jgSSA
-- — - — - —

/o '95 95%
6 8 .10 1

a

- —

2 1
                                            	ML
                                           —  - MDL
                                                                                 pH
                                                             a SR2-BPT-0001 - A
                                                             OSR2-BPT-0001 -B

                                                             ASR2-BPT-0001 -C
                                                                                                   95%
                                                                                                   5%
Figure 18. Example Results from SR002.1. Brayton Point fly ashes—mercury (top) and arsenic (bottom) release as a
function of pH for the baseline fly ash and the fly ash with enhanced Hg control. 5th and 95th percentiles of mercury and
arsenic concentrations observed in typical CCR monofill leachate are shown for comparison. Replicate A for mercury
results likely reflects sample heterogeneity (i.e., more activated carbon in sub-sample). ML=method limit (for quantification);
MDL=method detection limit; less than MDL reported at Vz MDL.
42
-------
                                                             Characterization of Coal Combustion Residues
  CO
1497
1000 -
100
10 -
1 -
0.1
0.01 -
0 001


im jV


=
. 5%





' ' ' 10.3
sza




QCO/
   	ML
 — - MDL
A. SR002.1
                                               95%
                                                5%
                                                         CO
4000 -
3000
7000 -
mnn
MCL
:
1
t
1
:


i




A
B









(




\

                             8    10
                            PH
14
                    DSR2-SHT-0001 -A
                    o SR2-SHT-0001 - B
                    A. SR2-SHT-0001 - C
          	ML
          — - MDL
         B. SR003.1
468
LS ratio [mL/g]
                                                  10    12
                               nSRS-SHT-0001 -A
                               oSRS-SHT-0001 -B
                               ASR3-SHT-0001 -C
Figure 19. Example Results from SR002.1 and SR003.1 for Selenium in Fly Ashes from Salem Harbor with Enhanced
Hg Control.
    (ii) the observed field leachate concentrations (5th and
    95th percentiles of reported concentrations) over the
    observed pH range for field leachates based on the U. S.
    EPA database (5th and 95th percentiles of pH), forming
    a "box" on the results figures (Figure 18); and, (iii)
    results for CCRs from the same combustion facility
    with and without the air pollution control technology
    specifically being evaluated (e.g., with  and without
    activated carbon injection);
  • Laboratory leachate concentrations and pH as a func-
    tion of LS  ratio for each CCR are compared to the
    leachate concentrations as a function of pH at LS=10
    to evaluate whether expected initial leachate concen-
    trations under land disposal conditions will be the same,
    less than, or greater than the concentrations used in
    comparison to field data and for cumulative release
    estimates. Figure 19 illustrates a case where initial se-
    lenium concentrations in leachate at low LS ratio (Fig-
    ure  19B, SR003.1) are expected to be greater than in-
    dicated by the evaluation of concentration as a func-
    tion of pH at LS=10 (Figure  19A, SR002.1).

3.3.1. Mercury Results
A comparison of total content and range of laboratory ex-
tract mercury concentrations as a function of pH and LS
ratio for CCRs from different facilities is provided in Fig-
ures 20  and 21, respectively. Total content, especially for
mercury, has exhibited considerable variability for reported
values from the same facility, most likely resulting from
         sample heterogeneity and variations in operating condi-
         tions. Values reported here are those measured as part of
         this study. For each facility, the baseline case and the case
         with enhanced air pollution control treatment—either ac-
         tivated carbon injection or brominated activated carbon in-
         jection (for the St. Clair facility and Facility L)—are com-
         pared. Also, note that Facility C uses COHPAC air pollu-
         tion control configuration. For each case in Figure 21, the
         range of laboratory  extract concentrations was based on
         the CCR's natural pH at LS=10 from SR002.1, and the
         minimum and maximum concentrations observed over 5.8
         < pH <  12.09 in results from testing over the range of pH
         (SR002.1) and LS ratio (SR003.1).  For most cases, the
         minimum value indicated is the MDL (0.004 |Jg/L). As
         indicated previously, this pH range is based on the 5th and
         95th percentiles of pH in field leachate samples from CCR
         landfills reported in the EPA database. The MCL is included
         to provide a reference basis, but consideration must be given
         to appropriate dilution and attenuation factors when mak-
         ing determinations for specific cases. Also included in Fig-
         ure 21 are the ranges of mercury concentrations observed
         in field leachates for landfills from the U.S. EPA database
         and for landfills and  surface impoundments from the EPRI
         database. For field observations, the symbol with error bars
         represents the median (SO^percentile), 5th and 95th percen-
         tiles of applicable observations in the respective database.
         The full  range of values was not included to avoid bias
         from outlier data points.
                                                                                                        43
-------
Characterization of Coal Combustion Residues
          10 q
           1 v
   en
   en
   I
         0.1 v
        0.01
                 BPB  BPT  PPB  PPT  SHB  SHT  GAB  GAT  JAB  JAT  FLB   FIT  FLB2  FLT2

                   Brayton    Pleasant     Salem     Facility C     St. Clair     Facility L    Facility L
                   Point      Prairie       Harbor                            (Run#1)    (Run #2)
Figure 20. Comparison of Total Mercury Content in Baseline Cases and with Sorbent Injection for CCRs from Different
Facilities. (Facility code suffixes B = baseline and T = treated with sorbent injection; for example, PPB = Pleasant Prairie
baseline, and PPT = Pleasant Prairie treated)
Considering the results provided in Appendices D through
I, and comparisons in Figure 20 and Figure 21, the follow-
ing observations for mercury are made:
  • Use of sorbent injection increased the total mercury
    content for the fly ash to ca. 1-2 mg/kg except for Sa-
    lem Harbor. This value may represent the maximum
    practical capacity for the sorbent entrained with the
    fly ash. The total content of mercury in fly ash from
    Salem Harbor may be relatively unchanged or slightly
    lower because of the high content of uncombusted car-
    bon (LOI = 21 wt%) for the baseline case, which acts
    a sorbent similar to activated carbon; in this case, in-
   jection of activated carbon serves to dilute the total
    mercury content in the CCR.
  • For Facility L, accumulation  of the  fly ash for sam-
    pling for 4 hours (Run #1) resulted in loss of mercury
    from the fly ash when compared to  fly ash accumu-
    lated for 30 minutes (Run #2), most likely by volatil-
    ization at the elevated temperatures  within the accu-
    mulation hopper. Fly ash obtained from Run #1 was
    used for leaching evaluation  because  of the limited
    sample quantity available from Run #2.
44
Although the use of ACI substantially increases the
total mercury content in the CCRs, the range of labo-
ratory leaching extract concentrations in the baseline
cases and cases with sorbent injection are either un-
changed  or the  maximum leaching concentration is
reduced as a consequence of activated carbon injec-
tion. The exceptions are Facility C and Facility L,
which have an increased maximum extract concentra-
tion for the case with sorbent injection.
The expected range of mercury leachate concentrations
based on these results is  from < 0.004 (below MDL)
to 0.2 |Jg/L over the range of pH conditions expected
in coal ash landfill leachate.
The  range of mercury concentrations observed from
laboratory extracts is consistent with the range reported
for field leachates from landfills in the EPRI database.
Reliable data on mercury concentrations in leachates
from landfills was not available in the EPA database.
A lower range of field concentrations is reported for
impoundments in the EPRI database, possibly result-
ing  from a combination  of dilution or volatilization
occurring during management in impoundments.
-------
                                                             Characterization of Coal Combustion Residues
            10
              1  - -
            0.1  - -    -r
   en
           0.01  - -
         0.001  - -
        0.0001
4 > Natural pH            ( >                     j [
..    I         i    I                                ::
<  min*     *    *           ******
                                   * Indicates 
-------
Characterization of Coal Combustion Residues
         1000
     en
     _*:
     ~O)
          100 --
           10
                   BPB  BPT  PPB  PPT  SHB  SHT  GAB  GAT  JAB  JAT   FLB   FLT  FLB2 FLT2

                                                     Facility C    St. Clair
Brayton
Point
Pleasant
Prairie
Salem
Harbor
Facility L    Facility L
Run #1     Run #2
Figure 22. Comparison of Total Arsenic Content in Baseline Cases and with Sorbent Injection for CCRs from Different
Facilities. (Facility code suffixes B = baseline and T = treated with sorbent injection; for example, PPB = Pleasant Prairie
baseline, and PPT = Pleasant Prairie treated)
    Use of ACI resulted in a substantial decrease in total
    arsenic content in CCR for Brayton Point.
    There was not a consistent pattern with respect to the
    effect of ACI on the range of laboratory extract con-
    centrations. For Salem Harbor and slightly for Pleas-
    ant Prairie facilities, the cases with ACI had an increase
    in the upper bound of extract concentrations compared
    to the same facility without ACI. For Facility C and
    the Brayton Point and St. Clair facilities, a correspond-
    ing decrease was observed.
    Very low extract concentrations were observed for the
    St. Clair facility without and with B-PAC, even though
    the total arsenic content was comparable to several of
    the other cases. Conversely, relatively  high extract
    concentrations  were observed for Facility L without
    and with  B-PAC, even though the total  arsenic con-
    centration was low compared to the other cases. Thus,
    the presence of other constituents in the  CCRs or the
    formation conditions appears to have a  strong influ-
    ence on the release of arsenic.
    The range of arsenic concentrations observed in the
    laboratory extracts is consistent with the range of val-
    ues reported for field leachates from landfills and im-
                                       poundments. For some cases, both laboratory (Salem
                                       Harbor, Facility C, Facility L) and field concentrations
                                       exceeded the MCL by more than a factor of 10. The ex-
                                       pected range of arsenic concentrations under field condi-
                                       tions is less than 10 |Jg/L to approximately 1000 |Jg/L.
                                       Arsenic leachate concentrations typically are strongly
                                       a function of pH over the entire pH range examined
                                       and within the pH range observed for field conditions
                                       (for example, see Figure 18). For some cases (for ex-
                                       ample, see St. Clair, Appendix H), measured concen-
                                       trations of arsenic are strongly a function of LS ratio
                                       at the material's natural pH, with much greater con-
                                       centrations observed at low LS ratio. Therefore, test-
                                       ing at a single  extraction  final pH or LS ratio would
                                       not provide sufficient information to characterize the
                                       range of expected leachate concentrations under field
                                       conditions. Furthermore, for some of the CCRs a shift
                                       from the CCR's natural pH within the range of antici-
                                       pated conditions (e.g., Facility L, Brayton Point with
                                       ACI, Salem Harbor baseline, Facility C baseline) can
                                       result substantial increases in leachate concentrations.
                                       Therefore, co-disposal of these CCRs with other ma-
                                       terials should be carefully evaluated.
46
-------
                                                            Characterization of Coal Combustion Residues
         10000
           1000 --
            100 --
      C/3
     <
            0.1
                                                           max
                                               •          I Natui
             10	
              1 --
                              Natural pH 95th%

                              min
--T	-sef»%-
                                                            5"%)
                                                                                                  -MCL
                    BPB BPT PPB PPT SHB SHT GAB GAT JAB JAT  LAB  LAT  EPA  LF  IMP
                     Brayton  Pleasant    Salem   Facility C  St. Clair   Facility L       EPRI
                     Point     Prairie      Harbor

Figure 23. Ranges of Laboratory and Field Leachate Arsenic Concentrations Compared with the Drinking Water Maximum
Contaminant Level (MCL). [For laboratory data, symbol represents the concentrations at the natural pH of the CCR
tested, and the error bars represent the minimum and maximum concentrations within the relevant field pH range of 5.8
to 12.09, inclusive. For field data, symbol and error bars represent the 5th, 50th and 95th percentiles of reported  values
(EPA = EPA database; LF = landfills from EPRI database; IMP = impoundments from EPRI database). Reliable data for
mercury was not available in the EPA database.]
  • For several cases (Brayton Point, Salem Harbor, Fa-
    cility C without ACI, Facility L), arsenic concentra-
    tions in laboratory extracts appear to be controlled by
    solid phase solubility, whereas adsorption processes
    appear to play a more important role  for other cases
    (Pleasant Prairie, Facility C with ACI, St. Clair).

3.3.3. Selenium Results
A comparison of total content and of the range of labora-
tory leach  test extract selenium concentrations as a func-
tion of pH and LS ratio for CCRs from different facilities
is provided in Figures 24 and 25, respectively. The approach
used and comparisons made in Figure 25 are the same as
for mercury in Figure 21.

Considering the results provided in Appendices D through
I, and comparisons in Figures 24 and 25, the following
observations for selenium are made:
  • For two cases (Brayton Point, Facility C), use of ACI
    resulted in a substantial increase in the total selenium
    content of the CCR in comparison to the same case
    without ACI. For Facility C, this is likely a direct con-
                             sequence of the COHPAC configuration when ACI is
                             in use. For the other cases, the change in total sele-
                             nium content resulting from application of ACI or B-
                             PAC was minor but increased in all cases.
                             The range of selenium concentration in laboratory leach
                             test extracts is not correlated with total selenium con-
                             tent in the CCRs. For example, Brayton Point with ACI
                             had much greater total selenium content than the other
                             cases except Facility C with ACI, but it had only the
                             fifth highest selenium concentration under the labora-
                             tory leaching conditions. Conversely, Facility C
                             baseline had one of the lowest selenium total content
                             (less than MDL), but it had second greatest selenium
                             concentration under the laboratory leaching conditions.
                             The range of selenium concentrations observed in labo-
                             ratory leach test extracts for Facility C are much greater
                             than the concentrations observed for other cases and
                             for field conditions. This is a COHPAC facility, and
                             field leachate composition data for CCRs from this type
                             of facility was not available in the EPA or EPRI data-
                             bases. For all other facilities, the range of concentra-
                             tions observed from laboratory testing is consistent with
                                                                          47
-------
Characterization of Coal Combustion Residues
         1000
    en
    _*:
    ch
    ,§
    CD
          100 v
           10 v
                                     Indicates < MDL
                   BPB  BPT  PPB  PPT  SHB  SHT  GAB GAT  JAB   JAT  FLB  FLT  FLB2  FLT2
                    Brayton    Pleasant
                    Point      Prairie
Salem     Facility C    St. Clair    Facility L    Facility L
Harbor                          (Run#1)    (Run #2)
Figure 24. Comparison of Total Selenium Content in Baseline Cases and with Sorbent Injection for CCRs from Different
Facilities. (Facility code suffixes B = baseline and T = treated with sorbent injection; for example, PPB = Pleasant Prairie
baseline, and PPT = Pleasant Prairie treated)
    the range reported in the EPRI database for landfills.
    The concentration range reported in the EPA database
    for CCR landfills has a much lower upper bound than
    reported in the EPRI database.
    The concentration range for laboratory extracts and
    field observations exceeded the MCL for all cases ex-
    cept Facility L. For 5 out of 12 of the cases used for
    laboratory evaluation and for some field observations,
    the MCL is exceeded by more than a factor of 10.
    Selenium concentrations in laboratory leach test ex-
    tracts typically are strongly a function of pH over the
    entire pH range examined and within the pH range
    observed for field conditions (for example, see Brayton
    Point, Salem Harbor, Facility C). For some cases (for
    example, see Figure 19 or Brayton Point, Salem Har-
    bor, St. Clair in Appendices D, F, and H, respectively),
    measured concentrations of selenium are strongly a
    function of LS ratio at the material's natural pH, with
    much greater concentrations observed at low LS ratio.
    Therefore, testing at a single extraction final pH or LS
    ratio would not provide sufficient information to char-
    acterize the range of expected leachate concentrations
    under field conditions.
             • For several cases (Brayton Point, Salem Harbor, Fa-
               cility C, Facility L), selenium concentrations in labo-
               ratory extracts appears to be controlled by solid phase
               solubility, whereas adsorption processes appear to play
               a more important role for other cases (Pleasant Prairie,
               St. Clair).

            3.4. Long-Term  Release Assessment
            Cumulative release estimates for CCRs from each facility,
            both for the baseline case and the case with enhanced mer-
            cury recovery, are presented in Appendices D through I.
            One hundred year release estimates of mercury, arsenic
            and selenium are presented. One example of long-term re-
            lease assessment results for arsenic and selenium is pro-
            vided in Figures 26 and 27. For each case, first the polyno-
            mial curve fits for solubility as a function of pH are pre-
            sented along with the corresponding data from laboratory
            leaching test results (SR002.1) and the 5th and 95th percen-
            tile of pH and constituent concentration from the U.S. EPA
            database. Next, the cumulative probability distribution for
            cumulative constituent release is provided from the Monte
            Carlo simulation for both the baseline and test cases.  Fi-
            nally, a bar chart, comparing total content of the constitu-
48
-------
                                                           Characterization of Coal Combustion Residues
        100000
     O)
     CD
     co
         10000 --
          1000 --
           100 --
            10 --
             1 --
            0.1
                                                        -r max
H
                                                          Natural pH
                                                                                             95th%
                                                           	iMCL
                   BPB BPT PPB PPT SHB SHT GAB  GAT JAB  JAT  LAB  LAT EPA  LF  IMP
                    Brayton
                    Point
Pleasant
Prairie
          Salem
          Harbor
Facility C   St. Clair  Facility L
EPRI
Figure 25.  Ranges of Laboratory and Field Leachate Selenium Concentrations Compared with the Drinking Water
Maximum Contaminant Level (MCL). [For laboratory data, symbol represents the concentrations at the natural pH of the
CCR tested, and the error bars represent the minimum and maximum concentrations within the relevant field pH range
of 5.8 to 12.09, inclusive. For field data, symbol and  error bars represent the 5th, 50th and 95th percentiles of reported
values (EPA = EPA database; LF = landfills from EPRI database; IMP = impoundments from EPRI database). Reliable
data for mercury was not available in the EPA database.]
ent evaluated, estimated cumulative release over 100 years,
and percent of total content released is provided for the
baseline and test cases. Similar results are not provided for
mercury because of the simplification used for the assess-
ment based on results and underlying mechanism (see sec-
tion 2.5.1).

3.4.1. Long-term Release Estimates for
Mercury
A comparison of the long-term (100 yr) mercury release
estimates from the Monte Carlo simulation for each case is
presented in Figure 28A on a mass basis (micrograms of
Hg released per kilogram of CCR) and Figure 28B as a
percent of total mercury released. Figure 28A also includes
the total mercury content for each case.

Considering the results provided in Appendices D through
I, and comparisons in Figure 28, the following observa-
tions for mercury are made:
  • The estimated mass of mercury released over the as-
    sessment period does not correlate with the total mer-
                                           n  SR2-BPT-0001-A
                                           o  SR2-BPT-0001-B
                                           A  SR2-BPT-0001-C
                                           	Fit curve
                        Figure 26. Example Regression Curves of Experimental
                        Data of Arsenic Solubility as a Function of pH for Brayton
                        Point.
                                                                       49
-------
Characterization of Coal Combustion Residues
     10000
  75  1000
  CO
  0)
  o
  o
n
o

1
1 .




'4

nc
nu
DD .t
n A A
A
A
A
t 	

!l ^ A
^AA^


]
i



           0  5% 20
405°% 60
Percentile
nBPB ABPT


Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
oro
ng/kg
0.2
0.9
152
2095
468
4693
%
0.0003
0.0011
0.2
2.6
0.6
5.8
or i
ng/kg
0.1
0.1
22
338
90
10157
%
0.0003
0.0005
0.0772
1.2
0.3
36.4
Figure 27. Example 100-Year Arsenic Release Estimates
for Brayton Point as a Function of the Cumulative Probability
for the Scenario of Disposal in a Combustion Waste Landfill.
(Mt refers to the cumulative release over the 100-year
interval.)
    cury content of the CCR. This is a consequence of the
    relatively constant leaching test extract concentrations
    independent of the total mercury content in the CCR.
    For all cases, the median  expected release over 100
    years is less than or equal to 1 |Jg/kg, with the 5th and
    95th percentiles less than or equal to 0.005 and 15 |Jg/
    kg, respectively.
  • The percentage  of total mercury estimated to be re-
    leased over 100 years ranges from a very small per-
    centage (less than 0.002%) to less than 5% for most
    cases. From less than a very small percentage (less than
    0.03%) to less than 80% of the total mercury may be
    released from cases Facility C baseline and Facility L.
    The higher percentages for these three cases reflects
    the lower total mercury content present in the CCR.

3.4.2. Long-Term Release Estimates for
Arsenic
A comparison of the long-term (100 yr) arsenic release
estimates from the Monte Carlo simulation for each case is
presented in Figure 29A on a mass basis (micrograms As
released per kilogram CCR) and Figure 29B as a percent
of total arsenic released. Figure 29A also includes the total
arsenic content for each case and MCLLS95% for reference.
MCLLS95o/0 is the amount of arsenic that would be released
(1,000 |Jg/kg) if the leachate concentration was equal to
the MCL for arsenic (10 |Jg/L) for the entire  100 year pe-
riod and if the infiltration rate was at the 95th percentile of
reference cases for landfills in the U.S. EPA database. For
the purposes of this study, values that exceed this thresh-
old may warrant further examination as to whether or not
additional management controls should be considered.

Considering the results provided in Appendices D through
I, and comparisons in Figure 29, the following  observa-
tions for arsenic are made:
  • The estimated mass of arsenic released over the as-
    sessment period does not correlate with the total ar-
    senic content of the CCR. For all cases except Salem
   Harbor,  Facility C, and Facility L, less than 0.1% to
    5% of the total arsenic content is anticipated to be re-
   leased.
  • Salem Harbor, Facility C baseline, and Facility L are
   cases where up to a very high percentage (more than
    30%) of the total arsenic content may be released un-
   der some management conditions.
  • The cases of Salem Harbor, Facility C, and Facility L
   are examples of where more detailed release evalua-
   tion is warranted, considering site specific management
   practices, infiltration rates, and dilution and attenua-
   tion factors.

3.4.3. Long-term Release Estimates for
Selenium
A comparison of the long-term (100 yr) selenium release
estimates from the Monte Carlo simulation for each case is
presented in Figure 30A on a mass basis (micrograms Se
released per kilogram CCR) and Figure 30B as a percent
of total arsenic released. The presentation in Figure 30 is
analogous to the presentation used for arsenic release esti-
mates in Figure 29 and discussed previously.

Considering the results provided in Appendices D through
I, and comparisons in Figure 30, the following  observa-
tions for selenium are made:
  • For all  cases except Brayton Point, from 40% up to
   the total content of selenium in the CCR is anticipated
   to be released at the 95th percentile, with between 3%
   and 20% for the median  case  (except Facility C
   baseline, where the median case is 100% of the total).
   For Brayton Point, from 1% to 30% of the total con-
50
-------
                                                            Characterization of Coal Combustion Residues
10000 -
1000 -
100 -
10 -
'oi :
en :
0.01 -
0.001 -
0.0001 -
A)
100 -
10 -

1 -
IF 0-1 -
£ •
•5 :
°S? 0.01 -
" 0.001 -
0.0001 -
0.00001 -

• * * * * *
* *


4

95lh% +
"|
(50lh%
4

5th %
4
»



» 4



»



)
4



> 4



> ^



* 4



> <




»


»
^


.A.
<

(

•^
>

i "

r

\

-*-
Median
Total Content








BPB BPT PPB PPT SHB SHT GAB GAT JAB JAT LAB LAT
Brayton Pleasant Salem Facility C St. Clair
Point Prairie Harbor

95th%

4




»
50th%
4
•5th %

4

»

4

1
4


4
>



»
i




1



>


4


4

»



> -r

4



•



BPB BPT PPB PPT SHB SHT GAB GAT JAB JAT
Brayton Pleasant Salem Facility C St. Clair
B) Point Prairie Harbor
Facility L
(




i
4





>






9 Median

LAB LAT








Facility L
Figure 28. Upper Bound of 100 yr Mercury Release Estimates for Landfill Scenario Without and with Activated Carbon
Injection. (A) mass released in ug of mercury released per kg of CCR and total content in ug of mercury per kg of CCR,
(B) percent of total mercury content released. Symbol with error bars represents 5th, 50th and 95th percentiles from Monte
Carlo simulation. (Facility code suffixes B = baseline and T = treated with sorbent injection; for example, PPB = Pleasant
Prairie baseline, and PPT = Pleasant Prairie treated)
                                                                                                       51
-------
Characterization of Coal Combustion Residues
O)
_*:
~O)
IUUUUUU 1
100000 i
10000 -

1000 -

100 -

10 -

\ .
1
0.1 -
0.01 -

n nni -
+
- * <





4




95lh%


50lh%-
(
(



5lh%



I '





• Median
^ Total Content

r 4


' <







•




t
I '





<





>

»
j




•
4
(





1






1
4







> ^
I <





»>
»






»






                                                                                        MCL     =1000|ig/kg
                                                                                            LS 95%      ^a a
    A)
                  BPB   BPT  PPB  PPT  SHB  SHT  GAB  GAT  JAB  JAT  LAB   LAT

                                                    Facility C    St. Clair   Facility L
Brayton    Pleasant
Point      Prairie
Salem
Harbor
          100 ^
 ro
,o

            1 -
          0.1 ^
         0.01 :
        0.001 : •    -1-
       0.0001 :
      0.00001
    B)
                   __95lh%
                     50lh%
                    Median
                  BPB   BPT  PPB  PPT  SHB  SHT  GAB  GAT  JAB  JAT   LAB  LAT
Brayton     Pleasant     Salem
Point       Prairie       Harbor
                                                    Facility C    St. Clair    Facility L
Figure 29. Upper Bound of 100 yr Arsenic Release Estimates for Landfill Scenario Without and with Activated Carbon
Injection. (A) mass released in ug of arsenic released per kg of CCR and total content in ug of arsenic per kg of CCR, (B)
percent of total arsenic content released. Symbol with error bars represents 5th, 50th and 95th percentiles from Monte
Carlo simulation. (Facility code suffixes B = baseline and T = treated with sorbent injection; for example, PPB = Pleasant
Prairie baseline, and PPT = Pleasant Prairie treated)

52
-------
                                                             Characterization of Coal Combustion Residues
  en
  CO
        1000000
         100000 --


          10000 --
     A)
1000 --


 100


  10 -f
   1


 0.1
                      50th%4 •
                                <>    ,
                       5th%
                                                                            O    II
                   •  Median
                   •  Total Content
                    BPB   BPT  PPB   PPT  SHB SHT  GAB GAT  JAB   JAT  LAB   LAT
Brayton    Pleasant
Point      Prairie
                                 Salem     Facility C   St. Clair    Facility L
                                 Harbor
  ro
 .o
 si
  CD
 CO
  CO
      B)


10-



1 -
0.1 -
0.01 -
nn-i -





4



•

95th%


50th%
I
4

5th%
Median


4



I





•
4









' 4









I



















4









>
4









1 4









I
4






•


• 4









»





                    BPB  BPT  PPB  PPT  SHB  SHT  GAB  GAT  JAB  JAT  LAB  LAT
          Brayton
          Point
          Pleasant
          Prairie
Salem
Harbor
                                                      Facility C    St. Clair    Facility L
Figure 30. Upper Bound of 100 yr Selenium Release Estimates for Landfill Scenario Without and with Activated Carbon
Injection. (A) mass released in ug of selenium released per kg of CCR and total content in ug of selenium per kg of CCR,
(B) percent of total selenium content released. Symbol with error bars represents 5th,  50th and 95th percentiles from
Monte Carlo simulation. (Facility code suffixes B = baseline and T = treated with sorbent injection; for example, PPB =
Pleasant Prairie baseline, and PPT = Pleasant Prairie treated)

                                                                                                        53
-------
Characterization of Coal Combustion Residues
    tent is anticipated to be released for more than half of        can be substantially reduced for each CCR case, either
    the anticipated conditions.                                 through control of pH or infiltration.
    Low fractional releases of selenium (less than 0.1%,      • All cases are examples of where more detailed release
    except for Facility C baseline) at the 5th percentile sug-        evaluation is warranted, considering site specific man-
    gest management scenarios where anticipated release        agement practices, infiltration rates, and dilution and
                                                            attenuation factors.
54
-------
                                                           Characterization of Coal Combustion Residues
4. Conclusions  and Recommendations
4.1.  Assessment of CCRs Without
and With Activated Carbon  Injection
Analysis has been completed for CCRs from six coal com-
bustion facilities that control mercury emissions by sor-
bent injection; four using powdered activated carbon in-
jection and two using brominated powdered activated car-
bon injection.  For each facility, the evaluation included
assessments of CCRs generated both with and without use
of the activated carbon injection. None of these facilities
had scrubbers as part of their air pollution control technol-
ogy. The following conclusions are drawn for this class of
facilities:
  •  Application of activated carbon injection substantially
    increased the total mercury content in the resulting
    CCRs for five of the six facilities evaluated. Substan-
    tially increased arsenic and selenium content in the
    CCRs was observed at the  one facility that employed
    COHPAC fabric filter particulate control technology.
    This may have resulted from additional arsenic and
    selenium adsorption onto the CCR while retained in
    the fabric filters. Significant increase in the selenium
    content of one additional facility was noted.
  •  Mercury is strongly retained by the CCR and unlikely
    to be  leached at levels of environmental concern.
    Leaching that did occur did not depend on total mer-
    cury content in the CCR, leaching pH, nor liquid to
    solid ratio, and mercury concentrations in laboratory
    extracts appeared to be controlled by non-linear ad-
    sorption equilibrium. Laboratory extract concentrations
    ranged between less than the MDL (0.01 |Jg/L) and
    0.2 Mg/L.
  •  Arsenic and selenium may be leached at levels of po-
    tential concern  from CCRs generated at some facili-
    ties both with and without  enhanced mercury control
    technology. Further evaluation of leaching of arsenic
    and selenium from CCRs that considers site specific
    conditions is warranted.
  •  Leachate concentrations and the potential release of
    mercury, arsenic and selenium do not correlate with
total content. For many cases, leachate concentrations
observed are a function of final pH over the range of
field conditions, and the observed leaching behavior
implies that solubility in the leachate or aqueous ex-
tract controls observed liquid concentration rather than
linear adsorption equilibrium. For these cases, use of
linear partition coefficients (Kd) in modeling leaching
phenomena does not reflect the underlying processes.
In addition, for many cases, the amount of mercury,
arsenic and/or selenium estimated to be released over
a 100 year interval is a small fraction (less than 0.1%
to 5%) of the total content. For selenium, release from
less than 5% up to the total content of selenium can be
anticipated over the 100 year period. Therefore, it is
not recommended to base landfill management  deci-
sions  on total content of constituents in CCRs since
total content does not consistently  relate to quantity
released.
Results of this assessment also suggest management
conditions (e.g., through control of infiltration and pH)
that may result  in reduction releases of arsenic and
selenium by  as much as two orders of magnitude in
comparison to upper bound estimated releases.
Use of the Leaching Framework facilitated understand-
ing the variations in anticipated leaching behavior un-
der the anticipated field landfill disposal conditions,
including expected ranges of constituent concentrations
in leachate and cumulative release over a defined time
interval. In addition, insights into the mechanisms con-
trolling constituent leaching were obtained. This depth
of understanding would not have  been possible using
leaching tests focused on a single extraction  condition
(e.g.,TCLP, SPLP,orSGLP).
This study provides baseline data which allows using
a reduced set of laboratory testing conditions as a
screening leaching  assessment for CCRs from coal
combustion facilities employing similar air pollution
control technology. For mercury, extraction only at the
material's natural pH at LS=10 is adequate.  For ar-
senic, extraction at four conditions is warranted to de-
                                             55
-------
Characterization of Coal Combustion Residues
    fine the range of expected leachate concentrations and
    release: (i) pH 5.5-6.0 at LS=10, (ii) pH 7.5-8.5 at
    LS=10, (iii) pH 12.0-12.5 at LS=10 and (iv) natural
    pH at LS=2. For selenium, either the total content or
    the same conditions as recommended for arsenic can
    be used. At least duplicate extractions should be used.
    Results from this more limited testing can be evalu-
    ated in comparison with the results presented in this
    report to determine if more  extensive evaluation is
    warranted.

4.2. Implementation of Leaching Test
Methods
The leaching assessment approach published by Kosson et
al. (2002) and implemented in this report was selected be-
cause after internal EPA review (Office of Research and
Development, Office of Solid Waste) and consultation with
the Environmental Engineering committee of the EPA Sci-
ence Advisory Board, it was considered the only available
peer reviewed and published approach that allowed con-
sideration of the range of potential field management sce-
narios expected for CCRs  and that provided a fundamen-
tal foundation for extrapolation of laboratory testing to field
scenarios. Additional development and validation of the
leaching assessment approach through this proj ect provides
the following conclusions:
  • Laboratory leaching test results were consistent with
    observed ranges of field leachate pH and with  mer-
    cury, arsenic, and selenium concentrations. Thus, the
    leaching test methods  employed in this study provide
an appropriate basis for evaluating leaching under the
range of anticipated field management scenarios.
Leaching test methods SR002.1 (Solubility and Re-
lease  as a Function of pH) and SR003.1  (Solubility
and Release as a Function of LS ratio) have been suc-
cessfully implemented at the EPANational Risk Man-
agement Research Laboratory. The use of these meth-
ods is now considered near routine methodology for
the laboratory.
QA/QC methodology conforming with EPA Tier 3 re-
quirements has been developed and demonstrated for
the leaching test methods SR002.1 and SR003.1.
Further efficiency in  implementation of the QA/QC
methodology may be obtained, based on  the results
from testing  the initial set of CCRs, by reducing the
number of replicates and control analyses required
under the initial QA/QC plan.  These improved project
efficiencies are being implemented for evaluation of
additional CCRs under this project.
A mass balance around the laboratory leaching test pro-
cedures has been completed for mercury and selected
metals of potential concern. These results indicate that
recoveries were between 60% and 91% for mercury
during the leaching tests and subsequent analytical
procedures, which is within the uncertainty resulting
from heterogeneity within the CCR. Additional mass
balance verification may be warranted if future samples
have significantly different characteristics that may
result in greater volatility of the constituents of inter-
est than in the reference sample evaluated.
56
-------
                                                                Characterization of Coal Combustion Residues
5.  References
ACAA (American Coal Ash Association), 2000. 1999 Coal Combustion Product (CCP) Production and Use (Short Tons),
available at http://www.acaa-usa.org/whatsnew/ccp_surveys.htm (accessed March 2001).

ACAA (American Coal Ash Association), 2003. 2003 Coal Combustion Product (CCP) Production and Use Survey, ACAA,
Alexandria, VA, available at http://www.acaa-usa.org/PDF/2003_CCP_Survey(10-l-04).pdf (accessed January 2006).

Clean Air Mercury Rule: 70 FR 28606; May 18, 2005.

Clean Air Interstate Rule: 70 FR 25162; May 12, 2005.

Duong D. Do., 1998. Adsorption Analysis: Equilibria and Kinetics. Imperial College Press, London, 892 p.

EPA, 1988. Report to Congress-Wastes from the Combustion of Coal by Electric Utility Power Plants. EPA/530/SW-88/002, U.S.
EPA, Office of Solid Waste and Emergency Response, Washington, DC.

EPA, 1996. Method 3052 "Microwave Assisted Acid Digestion of Siliceous and Organically Based Matrices," Test Methods for
Evaluating Solid Waste, Physical/Chemical Methods (SW-846), available at http://www.epa.gov/sw-846/main.htm (accessed
January 2006).

EPA, 1998. Method 7473 "Mercury in Solids and Solutions by Thermal Decomposition, Amalgamation, and Atomic Absorption
Spectrophotometry," Test Methods for Evaluating Solid Waste,  Physical/Chemical Methods (SW-846), available at
http://www.epa.gov/sw-846/main.htm (accessed January 2006).

EPA, 1999. Report to Congress-Wastes from the Combustion of Fossil Fuels: Volume 2-Methods, Findings and
Recommendations. EPA/530/R-99/010. U.S. EPA, Office of Solid Waste and Emergency Response, Washington, DC.

EPA, 2000. Characterization and evaluation of landfill leachate. EPA, Draft Report. 68-W6-0068, September 2000.

EPA, 2002. Characterization and Management  of Residues from Coal-Fired Power Plants, Interim Report. EPA/600/R-02/083,
December.

EPA, 2003. Technical Support Document for the Assessment of Detection and Quantitation Approaches. EPA/82 l/R-03/005. U.S
EPA, Office of Water, Washington, DC, available at (http://epa.gov/waterscience/methods/det/index.html)  [Superceded by
Revised Assessment of Detection and Quantitation Approaches,  EPA/82 l/B-004/005, available at
 http://www.epa.gov/waterscience/methods/det/rad/rad.pdf (accessed January 2006)].

EPA, 2005. Control of Mercury Emissions from Coal Fired Electric Utility Boilers: An Update, U.S. EPA, Office of Research and
Development, Research Triangle Park, NC, Feb 18, available at
http://www.epa.gov/ttn/atw/utility/ord_whtpaper_hgcontroltech_oar-2002-0056-6141.pdf accessed January 2006.

EPRI, 1999. Guidance for Co-management of Mill Rejects at Coal-Fired Power Plants. Report TR-108994.

EPRI, 2005. Personal communication of EPRI  leaching database (as of June 2005) summary information from K. Ladwig to D.
Kosson.

                                                                                                             57
-------
Characterization of Coal Combustion Residues
Kosson, D.S., van der Sloot, H.A., Sanchez, K, and Garrabrants, A.C., 2002. An Integrated Framework for Evaluating Leaching
in Waste management and Utilization of Secondary Materials, Environmental Engineering Science 19:3, 159-204.

MTI (McDermott Technology, Inc.), 2001. Mercury Emissions Predictions, available at
http://www.mtiresearch.con^aecdp/mercury.html#Coal%20Analyses%20and%20Mercury%20Emissions%20Predictions.
accessed November 2002.

Munro, L.J., Johnson, K.J., and Jordan, K.D., 2001. An Interatomic Potential for Mercury Dimmer, J.  of Chemical Physics,
114:13,  5545-5551.

Nelson,  S., 2004. "Advanced Utility Sorbent Field Testing Program," DOE/NETL Mercury Control Technology R&D Program
Review, Pittsburgh, PA, July 14-15, 2004.

Nelson,  S., Landreth, R., Zhou, Q., Miller, J., 2004. "Accumulated Power-Plant Mercury-Removal Experience with Brominated
PAC Injection", Joint EPRI DOE EPA Combined Utility Air Pollution Control Symposium, The Mega Symposium, Washington,
D.C., August 30-September 2.

Rudzinski, W., WA. Steele, and G. Zgrablich, 1997. Equilibria and Dynamics of Gas Adsorption on Heterogeneous Solid
Surfaces. Elsevier, Amsterdam.

Ruthven, D.M., 1984. Principles of Adsorption and Adsorption Processes. Wiley, New York, 433 p.

SAB (EPA Science Advisory Board, Environmental Engineering Committee), 2003. TCLP Consultation Summary, presented at
the SAB Environmental Engineering Committee consultation with EPA, Washington, D.C., June 17-18.

Sanchez, F., Kosson, D.S., 2005. Probabilistic Approach for Estimating the Release of Contaminants under Field Management
Scenarios, Waste Management, 25:5, 643-472.

Senior, C, Bustard, C.J., Baldrey, K., Starns, T. Durham, M., 2003a. "Characterization of Fly Ash From Full-Scale Demonstration
of Sorbent Injection for Mercury Control on Coal-Fired Power Plants," presented at the Combined Power Plant Air Pollutant
Control  Mega Symposium, Washington D.C., May 19-22.

Senior, C., Bustard, C.J., Durham, M., Starns, T, Baldrey, K, 2003b. "Characterization of Fly Ash From Full-Scale
Demonstration of Sorbent Injection for Mercury Control on Coal-Fired Power Plants," presented at the Air Quality IV
Conference and Exhibition, Washington D.C, September 22-21.

Senior, C., Bustard, C.J., Durham, M., Baldrey, K., Michaud, D., 2004.  Characterization of Fly Ash From Full-Scale
Demonstration of Sorbent Injection For Mercury Control on Coal-Fired Power Plants, Fuel Processing Technology, 85:6-7, 601-
612.

Starns, T, Bustard, J., Durham, M., Lindsey, C., Martin, C., Schlager, R., Donnelly, B.,  Sjostrom, S., Harrington, P.,
Haythornthwaite, S., Johnson, R., Morris, E., Change, R., Renninger, S., 2002. "Full-Scale Test of Mercury Control with Sorbent
Injection and an ESP at Wisconsin Electric's Pleasant Prairie Power Plant," presented at the Air & Waste Management
Association 95th Annual Conference and Exhibition, Baltimore, MD, June 23-27.

Thorneloe, S., 2003. Presentation to EPA Science Advisory Board (Environmental Engineering Committee), Washington, D.C.,
June 17th.

Vidic, R.D., 2002. Combined Theoretical and Experimental Investigation of Mechanisms and Kinetics of Vapor-Phase Mercury
uptake by Carbonacoues Surfaces. Final Report, Grant No. DE-FG26-98FT40119 to U.S. DOE, National Energy Technology
Laboratory.
58
-------
                                                Characterization of Coal Combustion Residues
                                  Appendix A
   U.S. EPA Science Advisory Board  Consultation  Summary
This summary was prepared at the close of the June 2003 U.S. EPA OSW and ORD consultation with the Science
Advisory Board, Environmental Engineering Committee Review Panel. These comments do not represent formal con-
sensus of the panel, and no consensus recommendations to the U.S. EPA were prepared. These comments do present
panel members views, with informal consensus on many points.
                                                                                  59
-------
Characterization of Coal Combustion Residues
                     TCLP CONSULTATION SUMMARY
                    Environmental Engineering Committee
                          Science Advisory Board
                    U.S. Environmantal Protection Agency
                              Washington, DC

                               June 18, 2003
                                 FOCUS
                  Alternatives to TCLP test for use in waste and
                  site situations where TCLP test is not
                  required by regulation
                  Focus Areas: contaminated site remediation;
                  waste material reuse; waste delisting
60
-------
                         Characterization of Coal Combustion Residues
             OVERVIEW
   Five specific consultation issues
   Two general consultation issues
   Key findings and recommendations
  SPECIFIC CONSULTATION ISSUE 1

Laboratory testing conditions should, to the
degree possible, anticipate the plausible
range of field conditions affecting waste
leaching in disposal and reuse situations.
These conditions will be most realistically
represented by a distribution of values for
factors affecting leaching,  and testing should
reflect this range of values to the degree
possible
                                                         61
-------
Characterization of Coal Combustion Residues
                                 COMMENTS
                     SPECIFIC CONSULTATION ISSUE 1

                  Agree with statement; comments that follow represent
                  consensus of committee
                  Statement should be related to some contextual use of
                  leaching test
                  Could expand probabilistic approach to include distributions
                  for field property parameters
                  Range of conditions considered depends on the intended
                  use of the information; need context
                  Need to define what the target problems are. What are we
                  trying to fix? Might be short list.
                  EPA needs to define better what the objectives are for the
                  broader leaching framework
                                  COMMENTS
                      SPECIFIC CONSULTATION ISSUE 1

                  Unclear what is the cost of making no change.
                  Unclear to what extent overly conservative
                  classification affects beneficial reuse
                  Also need to consider waste material properties,
                  e.g., physical form,  presence of oil, etc.
                  Need to consider organics as well as metals
                  Perhaps can group waste materials, consider
                  categories
62
-------
                             Characterization of Coal Combustion Residues
     SPECIFIC CONSULTATION ISSUE 2
   Conditions present at the end of a test (rather
   than initial test conditions) should be the
   basis for comparison with field conditions.
                 COMMENTS
     SPECIFIC CONSULTATION ISSUE 2

Statement indicates application is to dissolution of solids,
and to assessment of max aqueous phase cone of released
species for purposes other than waste classification
To the extent that the test aims to achieve equil conditions,
end measurement is appropriate
Issue motivated by theTCLP test, where final solution pH is
not measured.
Conditions in a reactor at equilibrium or at the end of a fixed
period of time are more relevant to the leaching measured in
the reactor at the time of sampling than the initial condition.
                                                              63
-------
Characterization of Coal Combustion Residues
                      SPECIFIC CONSULTATION ISSUE 3

                 For assessing metals leaching, pH is the strongest
                 predictor of leaching potential in most cases.
                 Other important factors include infiltration rate,
                 liquid/solid ratio, redox environment, effect of
                 common ions and ionic strength, effects of
                 external factors (co-disposed waste, biological
                 activity,  etc.), and exposure to  ambient air. The
                 relative  importance of these factors is likely to vary
                 for different wastes.
                                   COMMENTS
                       SPECIFIC CONSULTATION ISSUE 3

                 Redox condition (Eh), organic matter, aging-after-
                 disposal are important factors not in current tests
                 Microbes important, but not in current tests;
                 biotransformation can render solid phase metals soluble
                 Inclusion of microbes difficult for standard tests
                 pH important; not clear it is "strongest" predictor
                 Depends on constituent; pure metals, organics
                 influenced by different factors
                 R&D needed to be able to rank parameters
                 Again, need to define objectives better
64
-------
                                 Characterization of Coal Combustion Residues
       SPECIFIC CONSULTATION ISSUE 4

The development of multiple leaching tests, or a
flexible testing framework is required.  Selection of
a suitable leaching test should be made based on a
number of factors: anticipated use of test results,
waste characterization, the  range of plausible
disposal or reuse conditions, and previously
available information  on the subject waste or similar
wastes.  ...
                     COMMENTS
       SPECIFIC CONSULTATION ISSUE 4

  Framework of Kosson et al. is flexible tiered approach that encompasses
 equilibrium and kinetics and includes a suite of tests to address both, and
 allows for site-specific and generic release estimates using mass transfer
 modeling
  Framework of Kosson et al. is broad and potentially applicable to broad range
 of wastes and disposal scenarios
  Framework is open ended; it is a huge step beyond a single leach test; the
 manner in which it will be implemented by decisionmakers needs to be clarified
  Establishment of the framework for implementation will be resource intensive;
 EPA needs to justify the value of the information for decisionmaking, as
 balanced against other waste regulation needs.
  Need systematic approach for applying framework
  Need well-defined objectives for framework in order to develop step-by-step
 guidance for use
                                                                       65
-------
Characterization of Coal Combustion Residues
                         SPECIFIC CONSULTATION ISSUE 5
                     Modeling may also play an important role in
                     relating laboratory and field conditions to one
                     another, and in using leach test results to
                     assess the leaching potential of waste.
                                        COMMENTS
                          SPECIFIC CONSULTATION ISSUE 5

                   Concerned about use of deterministic models for prediction of leaching
                   potential; probabilistic modeling will be more appropriate in some cases, but
                   is resource intensive
                   Concern about incorporating modeling into leaching test protocol;
                   connecting model to field difficult
                   Modeling of leaching test may be useful for better understanding leach
                   mechanisms, and connection of test with field
                   For certain wastes, coupling of leach tests with a model should be
                   considered to predict solubilization over time, especially for organics
                   (Multiple equil. states may exist)
                   Usefulness of modeling depends on question to be answered; goals for
                   leach eval. need to be defined
66
-------
                             Characterization of Coal Combustion Residues
      GENERAL CONSULTATION ISSUE 1
 EPA requests SAB reaction to current research,
 and the potential to apply it to improve particular
 programs, specifically programs that do not now
 require the use of TCLP.
                  COMMENTS
      GENERAL CONSULTATION ISSUE 1

Capability to address organics, oily wastes, long-term
reliability need to be incorporated
Framework of Kosson et al. is broadly applicable; more
development work yet needed (guidance for specific
applications, database for field conditions and waste types,
data quality criteria, data interpretation/decisionmaking)
Framework of Kosson et al. is responsive to the 1999 SAB
commentary, but to this point is limited to inorganics
Current research proceeding without clear definition of
problem to be addressed by alternatives to TCLP
EPA should invest in identifying areas where alternative to
TCLP is vitally needed
                                                              67
-------
Characterization of Coal Combustion Residues
                                   COMMENTS
                      GENERAL CONSULTATION ISSUE 1

                 Leaching-related research inside/outside EPA
                 could be exploited more
                                   COMMENTS
                      GENERAL CONSULTATION ISSUE 1

                 Tiered structure of framework: enables tradeoffs in value of
                 information
                 EPA should prioritize R&D efforts based on assessment of
                 the problem most in need of alternatives to the TCLP, e.g.,
                 - If going to do evaluation of problems driving TCLP
                   alternatives, try to ascertain value of making a change,
                   i.e., economic analysis of problem
                 - Evaluate waste generation and management trends an
                   projections as well as current situation
                 - Cost-benefit analysis may be difficult; try to assess
                   opportunity cost of not pursuing alt. to TCLP
68
-------
                             Characterization of Coal Combustion Residues
       GENERAL CONSULTATION ISSUE 2
EPA requests SAB reaction to the direction for long-
term research work to further develop fundamental
understanding of leaching that would improve the
predictive capability of test suites or testing
frameworks.
                   COMMENTS
       GENERAL CONSULTATION ISSUE 2

  Goals for long-term research not well defined
  Increased fundamental knowledge will yield long-term
  advancement in assessment of leaching
  Funding priority for leaching research clearly is low.
  Long-term ORD research should be better coordinated with
  efforts inside/outside EPA, including DOD, FHWA, DOE
  Long-term ORD research is responsive to 1999 SAB
  commentary in science factors under study, but is focused on
  inorganics only and will benefit from clearer objectives
                                                              69
-------
Characterization of Coal Combustion Residues
                               COMMENTS
                    GENERAL CONSULTATION ISSUE 2

               Problem definition has two components
               - determine waste categories and field situations
                  most in need of TCLP alternatives
               - determine research priorities for the most
                  important waste/field situations
                               COMMENTS
                    GENERAL CONSULTATION ISSUE 2

               Organics, manufacturing process wastes, end-of-
               life product wastes need to be considered
               Industry/government/academic research
               consortium on leashing issues would be useful
               Industry may be willing to co-fund leaching
               evaluation R&D
               EPA should investigate collaborative efforts with
               European, Canadian, and Japanese researchers
70
-------
                            Characterization of Coal Combustion Residues
  KEY FINDINGS AND RECOMMENDATIONS

Alternatives to TCLP for evaluation of leach potential are
needed for some waste and site situations
Not clear if there is large or small number of waste and site
situations for which alternative approach is needed
Framework of Kosson et al. is broadly applicable; more
development work yet needed (guidance for specific
applications, database for field conditions and waste types, data
quality criteria, data interpretation/decisionmaking)
Framework of Kosson et al. is responsive to the 1999 SAB
commentary, but to this point is limited to inorganics
  KEY FINDINGS AND RECOMMENDATIONS

Current research needs clear definition of problem
to be addressed by alternative to TCLP
EPA should invest in identifying areas where
alternative to TCLP is vitally needed
The 1999 SAB commentary focused on science-
based issues in leaching: EPA has been
responsive within resource limitations.
Organic waste constituents need to be considered,
and a broader framework should include
assessment or organic constituent leaching
                                                            71
-------
Characterization of Coal Combustion Residues
                 KEY FINDINGS AND RECOMMENDATIONS

               Research and development should focus on most
               applicable waste/site situations, and possible
               beneficial reuse scenarios
               Given limited R&D resources, EPA should
               prioritize research efforts and leverage DOD,
               DOE, FHWA interest in leaching through cross-
               govt coordination, as well as industrial and
               international collaboration
               EPA intra-agency efforts should be more closely
               linked
72
-------
                           Characterization of Coal Combustion Residues
                 Appendix B
DOE NETL Full-Scale Test Site Flow Diagrams
                                                     73
-------
Characterization of Coal Combustion Residues
Brayton Point Unit 1
  •  Carbon injected upstream of second ESP (Research Cottrell). Only !/> of the unit was treated, or carbon was injected
    into one of the two new ESPs (Research Cottrell ESPs).
  •  Hopper IDs also shown. Samples from C-row are from the first row of hoppers in the second ESP.
                                                                             Gas Flow
          Hg
                                                                               HgS-CEM


                                                                             Air Preheater
                                                                   Erst ESP
                                                                   (Koppers)
                                                               Hg S-CEM
                                                            Sorbcnt Injection
                                        -Second ESP
                                         (Research-Cottrell)
                                    -Hg S-CEM
                                                     Hg S-CEM: Hg Semi Continuous Emission Monitors
74
-------
                                                   Characterization of Coal Combustion Residues
Bird's eye view of second ESP.
Samples taken from C-raw hoppers.
 North
   New
   Precips
    Old
    Precips
                             East
                                                                 West



                                   i
                                 it. -ir.iL.1ii

                           jssp!p*rfi'"" ""°m3ai
                               -_
                               1O
                                '
                           —ft—-
                           	_<.l V _„
                             T*
                                   .^ N
                    **
                    x   \     4   1
                   /•      x -^ .   -J f
                                                                       ">•
                                                                          *

        4r^
                                                                          X    N
                                                                   3

                                                    —-*	\    |—-^
                     Gas Flow

I!

                                                           ;
                                                              ^   ^
                                                                     ja
                                                                                         75
-------
Characterization of Coal Combustion Residues
Pleasant  Prairie Unit 2
Carbon injected upstream of cold-side ESP. Only % of the unit was treated. Test ESP was ESP 2-4.
76
-------
                                                     Characterization of Coal Combustion Residues
Salem Harbor Unit 1
Carbon injected upstream of cold-side ESP. Row-A hoppers were the front hoppers.

                                               Carbon
                                               Injection Point
     Boiler
Long Ai
Heater

=
r Short Air
Heater
II



V
                                 Steam Coils
Facility C
Carbon injected upstream of Unit 3B COHPAC baghouse (in between hot-side ESP and baghouse)
        WF
      BoBer
     Grade Level
                                                        HgJF Mercuy Analyzer
                                                                     Location
                •jftftftftfSSS
                                        Activated Carbon Injection
                                                                                          77
-------
Characterization of Coal Combustion Residues
St. Clair
       Total flow from
       boiler/econimizer
Facility L
                                       North-Side CS-ESP (ESP B)
                         T
                                        South-Side CS-ESP (ESP A)
                  B-PAC  injection
                   B-PAC injection
Total Flow to
    Stack
      Total flow from
      boiler/econimizer
                                          B-Side HS-ESP (ESP B)
                                        A-side HS-ESP (ESP A)
   To B Stack
                                                                                  To A Stack
78
-------
                     Characterization of Coal Combustion Residues
          Appendix C
Quality Assurance Project Plan
                                                79
-------
Characterization of Coal Combustion Residues
   U.S.  EPA/APPCD
                                   QAPP  FDR THE
                                   CHARACTERIZATION
                                   DF COAL COMBUSTION
                                   RESIDUES  (WA 4-D4)

                                   CATEGORY HI/APPLIED
                                   RESEARCH
        ARCAD1S
80
-------
                                                   Characterization of Coal Combustion Residues
ARCADIS
Susan Thorneloe
EPA WA Manager
QAPP for the
Characterization of Coal
Combustion Residues
(WA 4-4)

Category Ill/Applied
Research
Shirley Wasson
EPA QA Representative
                                                                Prepared for:
                                                                USEPA/APPCD
Rob Keeney
ARCADIS WA Leader
Laura Nessley
ARCADIS Designated QA Officer
Prepared by:
ARCADIS G&M, Inc.
4915 Prospectus Drive
Suite F
Durham
North Carolina 27713
Tel 919 544 4535
Fax 919 5445690
                                                                Our Ref :
                                                                RN9902013.0037


                                                                Date:
                                                                11 January 2006
                                                                                          81
-------
Characterization of Coal Combustion Residues
   ARCADIS                                                               TABLE PF CONTENTS
   1 .a  PROJECT DESCRIPTION AND  OBJECTIVES

        1.1   Purpose

        1.2   Project Objectives


   Z.D  PROJECT ORGANIZATION


   3.D  EXPERIMENTAL APPROACH

        3.1   Task I: Characterization of CCRs

        3.2   Task II:  Chemical Stability of Target Metals

        3.3   Task III: Thermal Stability of Target Metals

        3.4   Task IV: Biological Transformation and Volatilization of
                   Organo-Mercury


   A.O  SAMPLING PROCEDURES

        4.1   Sample Custody Procedures

        4.2   CCR and SRM Samples

             4.2.1  Physical and Chemical Characterization Samples

             4.2.2 Leaching Study Samples

             4.2.3 Fixed-Bed Reactor Samples

        4.3   Leachate Collection

             4.3.1  Tier 1 Screening Tests

             4.3.2 Tier 2 Solubility and Release as a Function of pH and
             LS Ratio

             4.3.3 Tier 3 Mass Transfer Rate

        4.4   Fixed-Bed, TPD Reactor Sampling

             4.4.1   Thermal Desorption Test Plan for High
                   Temperature CCR Commercial  Processes


   5.D  TESTING AND MEASUREMENT PROTOCOLS
82
-------
                                                    Characterization of Coal Combustion Residues
ARCADIS                                                              TABLE OF CONTENTS
     5.1   Physical Characterization

          5.1.1  Surface Area and Pore Size Distribution

          5.1.2  Density Measurements

          5.1.3  pH and Conductivity

          5.1.4  Moisture Content

     5.2   Chemical Characterization

          5.2.1  Carbon Content (TGA)

          5.2.2  Mercury (CVAA)

          5.2.3  Other Metals (ICP)

          5.2.4  Anion Analysis by 1C

          5.2.5  X-Ray Fluorescence (XRF) an Neutron Activation
          Analysis


6.a  QA./QC CHECKS

     6.1   Data Quality Indicator Goals

     6.2   QC Sample Types


7.a  DATA REDUCTION, VALIDATION, AND REPORTING


B.D  ASSESSMENTS


9.Q  REFERENCES
                                                                                             83
-------
Characterization of Coal Combustion Residues
    ARCADIS                                                               TABLE OF CONTENTS
   TABLES

         Table 3-1 Summary of testing under Task II to be performed on
                    theSRM

         Table 3-2 Summary of testing under Task II to be performed for
                    detailed characterization of CCRs

         Table 3-3 Summary of testing under Task II to be performed for
                    screening evaluation of CCRs

         Table 4-1 SRM certified values

         Table 6-1 Data quality indicator goals

         Table 8-1 PEA Parameters and ranges


    FIGURES

         Figure 2-1- Project organizational chart
                                                  in
84
-------
                                                          Characterization of Coal Combustion Residues
ARCADIS
                                                                                QAPP FOR THE
                                                                                CHARACTERIZATION
                                                                                OF COAL
                                                                                COMBUSTION
1 .D  PROJECT  DESCRIPTION AND OBJECTIVES

1 .1   PURPOSE

In December 2000, EPA determined that regulations are needed to control the risks of
mercury (Hg) emissions from coal-fired power plants.  A number of Hg control options
are currently being evaluated through bench-scale and full-scale demonstrations. For
each of the technologies that appear to have commercial application, the resulting
residues are to be evaluated to determine any potential cross-media impacts through
either waste management of these residues or use in commercial applications. Coal
combustion residues (CCRs) include bottom ashes, fly ashes, and scrubber sludges
from flue gas desulfurization (FGD) systems. The questions to be addressed through
this research include:

•   What are the changes to CCRs resulting from application of control technology at
    coal-fired power plants including changes in pH, metals content, and other
    parameters that may influence environmental release?

•   For CCRs that are land disposed, the questions to be addressed include:

           o   Will any of these changes result in an increase in the potential for
               leaching of Hg and other metals such as As, Se, Pb, and Cd leach
               from disposal of CCRs in impoundments, monofills, and minefills?

           o   What is the fate of Hg and other metals from CCRs that are land
               disposed?

           o   Is there a potential for organo-mercury being formed when anaerobic
               decomposition conditions exists?

•   For CCRs that are used in commercial applications, the questions to be addressed
    include:

           o   Will any of the changes to CCRs, from application of control
               technologies at coal-fired power plants, impact their use in
               commercial applications?

           o   What is the fate of Hg and other metals in CCRs when used in
               commercial applications?
                                                                                                       85
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                                    QAPP  FDR THE
                                                                                    CHARACTERIZATION
                                                                                    DF COAL
                                                                                    COMBUSTION
              o  What is the extent of Hg, As, Pb, Se and Cd release during high
                 temperature manufacturing processes used to produce cement clinkers,
                 asphalt, and wallboard?

              o  Are Hg and other pollutants such as As, Se, Pb and Cd present in
                 CCRs that are used in commercial applications such as highway
                 construction subject to conditions that would result in their release to
                 the environment?

   EPA's Air Pollution Prevention and Control Division (APPCD) through an on-site
   laboratory support contract with ARCADIS is to conduct a comprehensive study on the
   fate of mercury (Hg), arsenic (As), selenium (Se), lead (Pb) and cadmium (Cd) in
   CCRs. This research will be conducted in four tasks. Task I will focus on the
   characterization of different CCRs and the impact of Hg control technologies on these
   characteristics. Task II will focus on evaluating the potential for leaching of these
   toxic metals from CCRs that are generated with and without implementation of Hg
   control technologies under a range of management scenarios. Task HI will focus on
   the release of these toxic metals during high and low temperature utilization of CCRs
   in commercial processes.  Task IV will study the potential formation and volatilization
   of organo-mercury and inorganic mercury during simulated anaerobic decomposition
   processes. The scope of this QAPP covers Task I through Task III. Task IV will be
   addressed in a separate document.

   1 .2  PROJECT OBJECTIVES

   EPA's Office of Solid Waste (OSW) has been asked to provide general guidance on
   appropriate testing to evaluate the release potential of Hg and four other metallic
   contaminants (As, Se, Pb, and Cd) from CCRs via leaching, run-off, and volatilization
   when disposed hi landfills and incorporated into  commercial products using high/low
   temperature commercial processes. This evaluation in projected disposal and reuse
   situations (different waste management scenarios; see Section 1.1)  will both help
   assess the likely suitability of new or modified wastes for reuse, and ensure that Hg,
   As, Se, Pb, and Cd removed from stack emissions are not subsequently released to the
   environment in significant amounts as a result of CCR reuse or disposal practices.

   The primary objective of this project is to generate a comprehensive database that will
   enable EPA/OSW to (1) evaluate changes in CCRs resulting from the implementation
   of different Hg control technologies (see Section 3.1), and (2) assess environmental
   releases of these toxic metals during CCR management practices including land
                                                                                  Category
86
-------
                                                          Characterization of Coal Combustion Residues
ARCADIS
QAPP FDR THE
CHARACTERIZATION
OF COAL
COMBUSTION
disposal and commercial applications. OSW will be using the results to determine
needs in regard to future policies for managing CCRs whose characteristics are
changing as a result of the MACT under development for coal fired power plants.
OAR will be using the data to determine the potential for cross-media impacts and
potential changes to disposal and reuse practices which impact the economics of
potential regulations for coal-fired power plants. The data will also be used to address
questions raised by Congress and others regarding establishing the net benefit of
potential requirements for reducing emissions from coal-fired power plants.

Data on the chemical stability of these metals (leaching tests) will be generated using
the EPA/OSW recommended methods (EPA, 2002b) developed by Dr. David Kosson
and Dr. Florence Sanchez of Vanderbilt University titled An Integrated Framework for
Evaluating Leaching in Waste Management and Utilization of Secondary Materials
(Kosson et al., 2002a). The ability of these EPA/OSW methods to assess leaching of
the metals of interest will be further demonstrated with the use of a NIST standard
reference material (SRM) with certified amounts of trace metals. Data on the thermal
stability of these toxic metals during CCRs commercial applications will also be
determined by implementing temperature program desorption  techniques (see Section
4.4). The time/temperature profiles experienced by CCRs in their commercial
applications (highway construction, cement, asphalt, and wallboard manufacturing) as
determined based on the study conducted by RTI International and presented in the
publication titled Characterization and management of Residues from Coal-Fired
Power Plants/Interim Report, prepared for APPCD/NRMRL/EPA (EPA, 2002a).
Using this comprehensive database, EPA/OSW will determine the feasibility of the
application of the above methods to CCRs and they will assess the environmental
impacts of different types of CCRs' waste management practices.

A secondary objective of this project is to modify and develop a QA/QC framework
for the proposed leaching assessment approach developed by Kosson et al. The
reference fly ash may be an appropriate candidate for a method QC sample. These
activities will be carried out in cooperation with Drs. Kosson and Sanchez during
implementation of the proposed methods (see Task II, Section 3.2).

2.a  PROJECT ORGANIZATION

The organizational chart for this project is shown in Figure 2-1. The roles and
responsibilities of the project personnel are discussed in the following paragraphs. In
addition, contact information is also provided.
                                                                               Category
                                                                                                        87
-------
Characterization of Coal Combustion Residues
   ARCADIS                                                                     QAPP FOR THE
                                                                                  CHARACTERIZATION
                                                                                  OF COAL
                                                                                  COMBUSTION


   EPA Work Assignment Manager. Susan Thorneloe: The EPA WA Manager is
   responsible for communicating the scope of work, data quality objectives and
   deliverables required for this work assignment. The EPA WA Manager is also
   responsible for providing ARCADIS with the various types of CCRs to be
   characterized.

   Phone:(919)541-2709

   E-mail: thorneloe.susan@epamail.epa.gov

   EPA QA Representative. Shirley Wasson: The EPA QA Representative will be
   responsible for reviewing and approving this QAPP. This proj ect has been assigned a
   QA category ni and may be audited by EPA QA. Ms. Wasson is responsible for
   coordinating any EPA audits.

   Phone (919) 541-5510

   E-mail: wasson.shirlev@epamail.epa.gov

   ARCADIS Work Assignment Leader. Robert Keenev: The ARCADIS WA Leader is
   responsible for preparing project deliverables and managing the work assignment.  He
   will ensure the project meets scheduled milestones and stays within budgetary
   constraints agreed upon by EPA.  The WA Leader is also responsible for
   communicating any delays in scheduling or changes in cost to the EPA WA Manager
   as soon as possible.

   Phone (919) 541-3284

   E-mail: rkeenev@arcadis-us.com

   ARCADIS Inorganic Laboratory Manager, Robert Keenev. In addition to being the
   work assignment leader, Robert Keeney is also responsible for the operation of EPA's
   in-house Inorganic Laboratory. Mr. Keeney will review and validate all analytical data
   reports and ensure that the leaching studies are performed properly. He will also
   operate the TGA and surface area analyzers. For the leaching studies and mercury and
   metals analyses, Mr. Keeney will be supported by one chemist: Gene Gallagher and
   one technician: John Foley.
88
-------
                                                        Characterization of Coal Combustion Residues
ARCADIS                                                                    gApp  FDR  THE
                                                                              CHARACTERIZATION
                                                                              DF COAL
                                                                              COMBUSTION

Mr. Gallagher will perform HF extractions of solid CCR and SRM samples and also be
responsible for mercury analysis of samples by CVAA.  John Foley will perform the
leaching test. Mr. Keeney and Mr. Gallagher will submit the remaining HF digestates
to the subcontract analytical laboratory, STL-Savannah for ICP/MS analysis of the
other target metals. Mr. Keeney will also be responsible for assisting Drs. Kosson and
Sanchez in the development of appropriate QA/QC procedures for the leaching
assessment methods.

Phone (919) 541-3284

E-mail:  rkeenev@arcadis-us.com

STL-Savannah Analytical Manager. Angie Weimerskirk:  Ms. Weimerskirk will
review and validate the ICP/MS results and report them to Mr. Keeney.

Phone (912) 354-7858

E-mail:  aweimerskirk@stl-inc.com

ARCADIS Thermal Desorption Task Manager (Task HI), Behrooz Ghorishi: The
ARCADIS Thermal Desorption Task Manager is responsible for preparing task El
deliverables and managing task HI. He will ensure the task meets scheduled
milestones. The Thermal Desoption Task Manager is also responsible for
communicating any delays in scheduling to the ARCADIS Work Assignment Leader.
EPA WA Manager as soon as possible.  Dr. Ghorishi will be assisted by Jarek
Karwowski.  Mr. Karwowski will perform the thermal desorption test and submit the
samples to the laboratory.

ARCADIS Designated QA Officer. Laura Nessley: The ARCADIS QA Manager,
Laura Nessley, has been assigned QA responsibilities for this work assignment. Ms.
Nessley will be responsible for reviewing this QAPP  prior to submission to EPA QA
for review. Ms. Nessley will also ensure the QAPP is implemented by project
personnel by performing internal assessments. All QA/QC related problems will be
reported directly to the ARCADIS WAL, Robert Keeney.

Phone: (919) 544-2260 ext. 258

E-mail:  lnessley@arcadis-us.com
                                                                                                    89
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                                    QAPP FDR THE
                                                                                    CHARACTERIZATION
                                                                                    OF COAL.
                                                                                    COMBUSTION
   Vanderbilt University. Methods Development. Professors David Kosson and Florence
   Sanchez: Dr. Kosson in cooperation with Dr. Florence Sanchez developed the
   leachability methods being evaluated on this project. He will be available to consult
   regarding method optimization and development of QA/QC procedures. Dr. Kosson
   and Dr. Sanchez will be on-site in the early stages of the project to assist in setting up
   the procedures. Phones:  (615)322-1064; (615)322-5135
   E-mail: David.Kosson@vanderbilt.edu
          Florence.Sanchez@vandcrbilt.edu
                                                                                  Category
                        Figure 2-1- Project Organizational Chart
   3-D  EXPERIMENTAL APPROACH
   3.1   TASK
                 CHARACTERIZATION OF CCRs
   This task will focus on physical and chemical characterization of as-received CCRs.
   CCRs from different power plants and various types of control technologies will be
   selected and provided to ARCADIS by the EPA WA Manager. These CCRs will
   include fly ashes and scrubber sludges from DOE test facilities. These full-scale
   facilities are being used to test two different Hg control technologies: activated carbon
   injection as an adsorbent and addition of agents to wet scrubbers that maintain Hg in an
   oxidized state to facilitate removal by aqueous scrubbing. Physical characteristics of
   CCRs to be determined include specific surface area, moisture content and density.
   Chemical characterization will include total carbon analysis, pH of extract, total
90
-------
                                                           Characterization of Coal Combustion Residues
ARCADIS
                                                                                  CJAPP FOR  THE
                                                                                  CHARACTERIZATION
                                                                                  OF COAL.
                                                                                  COMBUSTION
concentration of target metals (Hg, As, Se, Pb, and Cd) and principal constituents (i.e.,            v
Fe, Cl", SO42", COa2")1. The types of analyses and instrumentations used to perform this
characterization are further described in Section 5.0. Identical characterizations will be
performed on the reference fly ash. Task I results will reveal the effect of implementing
the two Hg control technologies on the final characteristics of CCRs. This information
will help in decision-making regarding different waste management scenarios.

3.Z   TASK  II:  CHEMICAL STABILITY QF TARGET METALS

This task will investigate the fate of Hg, As, Se, Cd, and Pb during CCR management
practice of land disposal. Using the recently proposed test methods developed by
Kosson et al in coordination with EPA's Office of Solid Waste, leaching studies will
first be conducted on a reference fly ash. The reference fly ash is a high quantity fly
ash that has been characterized by ICP/MS  and CVAA analyses. The ICP/MS and
CVAA analyses will be checked using the NIST SRM 1633b. NIST SRM 1633B  is a
bituminous coal fly ash that is fully described in Section 4.2.2. The results obtained
from the reference fly ash leaching studies will be critical in evaluating the
performance of the method.  Using a known standard in place of the CCR material,
will also allow optimization of the proposed test methods. The quality control
procedures regarding the reference fly ash tests are described in section 6.

A summary of testing that will be carried out on the reference fly ash is presented in
Table 3-1 along with the number of replicates, the material mass required and the
number of extracts that will be generated. Detailed descriptions of the methods listed
in Table 3-1  can be found in the document titled An Integrated Framework for
Evaluating Leaching in Waste management and Utilization of Secondary materials
(Kosson et al., 2002a).
1 This will provide valuable insights concerning the characteristic behavior of the CCRs.
                                                                                                        91
-------
Characterization of Coal Combustion Residues
    ARCADIS
                                                                                     C?APP FDR THE
                                                                                     CHARACTERIZATION
                                                                                     OF COAL
                                                                                     COMBUSTION
                                                                                    Category
   TABLE 3-1.  SUMMARY OF TESTING UNDER TASK II TO BE PERFORMED ON THE
   REFERENCE  FLY ASH
Sample type
Level of testing

Tests

Baseline : Detailed pH001.1 (pH TOration Pretest)
Fly Ash" characterization j























| Moisture Content
I
-•




















AV002.1
Material
particle size

<2mm

<2mm

<2mm
Availability at pH 7.5 with EDTA I
(3 target points) |
|
SR002.1
Alkalinity, Solubility and Release
as a Function of pH
(11 pHs tested from 2 to 12)

SR003.1
Solubility and Release as a
Function of LS ratio
(LS=10, 5, 2, 1,0.5mL/g)
LSIOmUg
LSSmUg
LSZmUg
IS 1 mug
LS 0.5 mug

MTO02.1
Mass Transfer Rate in Granular
Materials
(10 extracts in 30 days)
- Optimum moisture content
- Leach test



<2mm


<2mm









<2mm
<2mm



Number of
replicates

2

3

3


3


3









2
S

Mass
material/
aliquot*
(9)

8

8

a


40




40
40
50
100
200



500
500

Mass
material/
test
replicate
(9)

8

8

24


440


430









500
500

Total mass material required (g
Total mass
of material
required
(9)

16

24

72


1320


1290







2500




5222
I
Number of
analytical
samples

NA

NA

3


33


15







30






    " Baseline fly ash is a fly ash that will be used as a reference material. Total elemental content will be determined by NAA/XRF analysis. ;

    After the proposed test methods have been successfully demonstrated on the reference
    fly ash, leaching studies will be conducted on high priority CCRs that will allow
    estimating constituent release by leaching for a range of conditions that are likely to
    occur during management practices. A separate test plan for the leaching experiments
    under this task is provided by its developers (Drs. Kosson and Sanchez). This test plan,
    titled "Draft (Revision #2), Sampling and Characterization Plan for Coal Combustion
    Residues from Facilities with Enhanced Mercury Emissions Reduction Technology"
    (Kosson et al., 2002b) together with this QAPP will cover all the issues regarding Task
    II. Two levels of testing will be performed. The first level will provide detailed

                                                     8
92
-------
                                                          Characterization of Coal Combustion Residues
ARCADIS
                                                                                         FDR THE
                                                                                 CHARACTERIZATION
                                                                                 OF COAL.
                                                                                 COMBUSTION
characterization of representative samples of CCRs that reflect each dominant CCR
chemistry with respect to mercury release. This will define the behavior of the general
class of CCR chemistry. This detailed characterization would establish a baseline for
comparison of subsequent test results. A summary of testing that will be carried out on
each dominant CCR chemistry is presented in Table 3-2 along with the number of
replicates, the material mass required and the number of extracts that will be generated.

The second level will provide screening evaluation of additional samples anticipated to
be representative of each dominant CCR chemistry. The second level screening will be
used to determine if the CCR being tested exhibits the same leaching behavior as the
general class of CCRs, which is assumed to have the same dominant chemistry. If the
leaching behavior is found to be significantly different than anticipated, then more
complete characterization can be completed. A summary of testing that will be carried
out for the screening level is presented in Table 3-3 along with the number of
replicates, the material mass required and the number of extracts that will be generated.

Residues collected before and after application of enhanced Hg control technologies
will be examined to evaluate the effect of the enhanced systems on the leaching
behavior of CCRs.

Estimates of the extent of release of the metals of concern during management
scenarios that include percolation through the CCRs or infiltration flow around the
CCRs (e.g., when compacted to low permeability or otherwise expected to behave as a
monolithic material) will be determined.  These data will be used to determine the risk
of land disposal of the different CCRs. Mass balances for each metal will be
determined using the chemical characterization data obtained in Task I. Utilization of
mass balance as a QA/QC tool is described in section 6. Details of this QA/QC
procedure are outlined hi section 6. In addition to  testing of the CCRs as generated,
CCRs as used in commercial products will be examined. Only commercial uses for
which there is a potential for release of Hg during  leaching will  be considered. One
commercial use of CCRs that may be of concern for Hg leaching is cement-based
materials (i.e., concrete/grout, waste stabilization,  road base/subbase). A generic
cement-based product made from samples representative of the  major coal fly ash
categories will be examined. A second commercial use of CCRs that may be of
concern is incorporation in gypsum board.  In this case leaching of Hg after disposal is
of concern. This task will consider the potential for Hg leaching after disposal from a
representative gypsum board product.
                                                                                Category
                                                                                                        93
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                         C^APP FDR THE
                                                                         CHARACTERIZATION
                                                                         or COAL
                                                                         COMBUSTION
   TABLE 3-2. SUMMARY OF TESTING UNDER TASK II TO BE PERFORMED FDR DETAILED
   CHARACTERIZATION OF CCRB
                                                                        Category
Sample type

Sample reflecting dominant
CCR chemistry
























!* Sample size required for each
," Optional
Level of testing

Detailed
characterization























Tests

pHOOI.1 (pH Trtratton Pretatt)

Moisture Content

AV002.1"
Availability at pH 7.S with EOTA
(3 target points)

SROC21
Alkalinity, Solubility and Release
as a Function of pH
(11 pHs tested from 2 to 12)

SR0031
Solubility and Release as a
Function of LS ratio
(LS=10, 5, 2, 1,0.5 mL/g)
LSIOmUg
LS5mUg
LS2mUQ
LS1mL/s
LSO.SmUg

MT002.1
Mass Transfer Rate in Granular
Materials
(10 extracts in 30 days)
-Optimum moisture content
- Leach tost



Material
particle size

< 2mm

< 2mm

< 2mm


< 2mm


< 2mm







<2mm






condttkxi within the test method. For example, one replicate of SRI
'

Number of
replicates

2

3

2


2


2







2




Mass Mass
material/ ^material/
aliquot* test
(g) Implicate
(g)

8

•

1


40




40
40
SO
100-
200



500
500

Total mass material


102.1 requires eleven!

e

a

24


440


430












required (g

Total mass
of material
required
(g)

16

24

46


L~ M0


MO







2000




3821

sample aliquots.
	 	 J_ '
Number of)
analytical
samples

NA

NA

2


22


10







20








                                           10
94
-------
                                                           Characterization of Coal Combustion Residues
ARCADIS
                                                                                         FOR THE
                                                                                 CHARACTERIZATION
                                                                                 OF COAL.
                                                                                 COMBUSTION
TABLE 3-3. SUMMARY OF TESTING UNDER TASK II TO BE PERFORMED FOR SCREENING    Category
EVALUATION OF CCRS
Sample type

Additional samples anticipated
to be representative of each
dominant OCR chemistry

















Level of testing

Scrawling









Tests


Moisture Content

SR002.1-A
(3 pHs © LS lOmL/g: ack»c,
neutral, alkali)

SRQ03.1-A
(LS - 10, 0.5 mL/g)
LSIOmljS
LSO.SmUg
:
MT0021-A






Material
particle size


<2mm

<2mm


< 1mm




<2mm
(4 extractions In 5 days)
- Optmum moisture content
- Leach test








Number of
replicates


3

3


a






2
3

Mass
material/
aliquot'
(g)


a

40




40
200



500
500

Mass
material/
test
replicate
(g)


8

120


240






h 5001
500

Total mass material required (g
! ]
Total mass
of material
required
(g)


24

360


720




2500




	 2§£

' Sample size required for each condition within the test method. For example, one replicate of SR002.1-A requires three sample aliquots.
Number of
analytical
samples


MA

3


•




12
	 _





3.3   TASK III: THERMAL STABILITY OF TARGET METALS

This task will investigate the potential release of the target metals during their
commercial applications such as cement, wallboard and asphalt manufacturing. The
potential long-term evaporation of these metals during low temperature applications
such as structural fills, highway construction, snow/ice control, and soil amendment
will also be determined Representative samples of CCRs will be tested in an in-house,
bench-scale, fixed-bed reactor system called the Thermal Program Desorption (TPD)
system. CCRs will be exposed to conditions that are experienced in actual situations.
The effluent of the fixed-bed reactor will be sampled and analyzed for inorganic Hg
(speciated, elemental and ionic), As, Se, Pb, and Cd using established methods such as
EPA Method 29 for metals and Ontario Hydro Method for mercury. Mass balances for
the metals will be determined using the chemical characterization data obtained in Task
I.  The reference fly ash will also be tested in the reactor and the results will be used as
a quality control check (see section 6).
                                                11
                                                                                                        95
-------
Characterization of Coal Combustion Residues
    ARCADIS
                                                                                   QAPP  FOR THE
                                                                                   CHARACTERIZATION
                                                                                   OF COAL.
                                                                                   COMBUSTION
    3.4  TASK IV: BIOLOGICAL. TRANSFORMATION AND VOLATILIZATION OF DRGANO-
    MERCURY

    This task will investigate the potential formation and release of organo-mercury during
    anaerobic decompositions. This will be simulated in a bioreactor and the effluent will
    be sampled for speciated inorganic and organic mercury. The details of these tests will
    be described in a separate QAPP. Due to the health and safety related concerns
    regarding organo-mercury compounds, prior to preparation of the QAPP, an extensive
    literature survey will be performed to select the most appropriate methods for sampling
    and analysis of these types of compounds.

    4.D  SAMPLING PROCEDURES

    The following subsections describe the sampling procedures to be used for each task.
    Whenever possible, standard methods will be followed. In some cases, draft methods
    may be evaluated and implemented.  Each method to be used will be cited and any
    deviations from the methods will be documented.

    4. 1   SAMPLE CUSTODY  PROCEDURES

    The following types of samples will be generated during these tests:

    1- "As-received" CCR samples before and after application of Hg control technologies,
    SRM and reference fly ash samples (solid samples) and treated CCR samples as used
    in commercial applications
    2- Post -leaching and post-thermal desorption CCR, reference fly ash samples  and
    treated CCR samples (solid samples)
    3- Leachate samples (liquid samples)
    4- Method 29 and Ontario Hydro Tram samples (liquid samples)

    Each sample generated will be  analyzed in-house but chain-of-custody procedures .will
    be required. CCRs will be logged as they are received by the ARCADIS WAL, Mr.
    Robert Keeney. Information regarding where each CCR originated and any other
    descriptive information available will be recorded in a dedicated laboratory notebook
    by Mr. Keeney. A 200 g grab sample will be taken from each "as-received" CCR and
    processed for physical and chemical characterization.  All samples will be properly
    contained and identified with a unique sample ID and sample label.  Sample labels at a
    minimum will contain the sample ID, date sampled, and initials of the analyst
                                                                                  Category
                                                   12
96
-------
                                                           Characterization of Coal Combustion Residues
ARCADIS
QAPP FOR THE
CHARACTERIZATION
OF COAL
COMBUSTION
responsible for preparing the sample. Chain-of-custody forms will be generated for all
samples prior to transfer for analysis.

Handling of CCR samples for the leaching tests (Task II) is described in detail by the
leaching procedure provided by its developers (Kosson et al., 2002a). For Task HI, a
lOg grab sample will be taken from each "as received" CCR and will be subjected to
the TPD procedures (see section 4.2.3)

4.2   CCR, AND REFERENCE FLY ASH SAMPLES

As mentioned, the focus of this program is to obtain information on the effect of Hg
control technologies on the stability of Hg, As, Se, Cd, and Pb in CCRs. Currently, two
different Hg control technologies are being tested in full-scale facilities. The focus of
the first technology is on powdered activated carbon (PAC) injection upstream of
participate matter control devices. The PAC  injection technology affects the properties
of fly ashes collected in electrostatic precipitators and baghouses. These facilities are
listed in the Vanderbilt test plan "Draft (Revision #2), Sampling and Characterization
Plan for Coal Combustion Residues from Facilities with Enhanced Mercury Emissions
Reduction Technology" (Kosson et al., 2002b). The second technology focuses on Hg
capture by wet scrubbers. Chemical modifications are being implemented in wet
scrubbers to enhance the Hg capture. The scrubber sludge from these facilities will be
impacted by this control technology. The scrubber sludge samples from these facilities
will be included in this test program.

The Hg control testing facilities will be identified and then" test reports will be obtained
and amended to this QAPP. The test reports  will include information on the
history/origin of each CCR sample, facility process description, CCR type, sampling
location, sampling time and method, coal type, operating condition, and sample storage
condition. Section 4.1 describes the sampling custody procedure.

4.2.1   Physical and Chemical Characterization Samples

"As received" CCR will be well mixed prior to taking samples for physical
characterization. Mixing of the sub-samples collected at the site will be done using a
riffle splitter. To ensure a good homogeneity of the final composite sample that will be
used for the study, the first two composite samples exiting the splitter will be
reintroduced at the top of the splitter. This procedure should be repeated at least 6
times. At the end, the two resulting homogeneous composite samples will be combined
in the same bucket and stored until laboratory testing. A 200 g representative sample
                                                                                Category III
                                                13
                                                                                                         97
-------
Characterization of Coal Combustion Residues
    ARCADIS
QAPP  FDR  THE
CHARACTERIZATION
OF COAL,
CDMBUBTIDN
    will be taken from the "as received" CCR and subjected to physical characterization
    measurements. Samples will also be taken of any CCRs that undergo size-reduction
    techniques (if size reduction is needed for testing purposes). The reference fly ash
    samples will be processed in the same manner as the CCRs. They will be tracked by
    lot number and will not require size-reduction.

    4.2.2  Leaching Study Samples

    CCRs used for leaching studies may undergo size reduction to acquire an adequate
    sample for testing. The size reduction method is outlined in the leaching test methods
    (Kosson et al., 2002a). If "as-received" CCRs are altered in any way prior to leaching
    studies, a representative sample will be submitted for physical and chemical
    characterization. SRM samples will not require size reduction. The NIST 1633B SRM
    is a bituminous coal fly ash that has been sieved through a nominal sieve opening of 90
    [an and blended to assure homogeneity. The certified values for the constituent
    elements are given in Table 4-1. The reference fly ash will also be certified using
    ICP/MS and CVAA.

    4.2.3  Fixed-Bed Reactor Samples

    Reactor samples will be essentially the same as the CCR samples used in the leaching
    studies.  Amount of material may be reduced, but physical and chemical characteristics
    will not be affected.
                                                                                  Category
    TABLE 4-1. SRM CERTIFIED VALUES
Element
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Concentration (mg/kg)
136.2 ±2.6
709 + 27
0.784 ± O.OO6
198.2 ±4.7
11 2. 8 ±2. 6
68. 2 ± 1.1
131. 8± 1.7
0.141 ±0.019
120.6 ±1.8
10.26±0.17
                                                   14
98
-------
                                                           Characterization of Coal Combustion Residues
 ARCADIS                                                                        qjAFP FOR THE
                                                                                  CHARAtTERIZATION
                                                                                  OF COAL
                                                                                  COMBUSTION
Strontium
Thorium
Uranium
Vanadium
1041 ± 14
25.7 ± 1.3
8.79 ±0.36
295.713.6
                                                                                 Category
4.3  LEACHATE COLLECTION

The proposed test method described in the publication titled An Integrated Framework
for Evaluating Leaching in Waste Management and Utilization of Secondary Materials
(Kosson, et. al., 2002a) will be used to conduct leaching studies. There are three tiers
to this test method:

Tier 1) Screening based assessment (availability)
Tier 2) Equilibrium-based assessment over a range of pH and Liquid/solid (LS) ratios
Tier 3) Mass transfer based assessment

The Tier 1 screening test provides an indication of the maximum potential for release
under the limits of anticipated environmental conditions expressed on a mg
contaminant leached per kg waste basis. Tier 2 defines the release potential as a
function of liquid-to-solid (LS) ratio and pH. Tier 3 uses information on LS
equilibrium in conjunction with mass transfer rate information. As mentioned
previously, prior to testing CCR, a reference fly ash will be used to demonstrate the
effectiveness of the proposed test methods.  Procedures for each tier are discussed in
the following subsections.

If needed, prior to tier testing, the "as-received" CCR will be size reduced using the
procedure PS001.1 Particle Size Reduction (Kosson et al., 2002a) to minimize mass
transfer rate limitation through larger particles. The pH will be then tested using the
method pHOOl.O pH Titration Pretest (Kosson et al., 2002a).

4.3.1 Tier 1 Screening Tests

Test Method AV002.1 Availability at pH 7.5 with EDTA (Kosson et al., 2002a) will
be used to perform the screening test.  This method measures availability in relation to
the release of anions at an endpoint pH of 7.5±0.5 and cations under enhanced liquid-
phase solubility due to complexation with the chelating agent. Constituent availability
is determined by a single challenge of an aliquot of the reference fly ash or size
                                                 15
                                                                                                         99
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                                             FDR THE
                                                                                     CHARACTERIZATION
                                                                                     OF COAL.
                                                                                     COMBUSTION
   reduced CCR material to dilute acid or base in DI water with the chelating agent,
   ethylenediamine-tetraacetic acid (EDTA). Extracts are tumbled end-over-end at 28±2
   rpm at room temperature for a contact time of 24 hours. At the end of the 24-hour
   period, the leachate pH value of the extraction is measured. The retained extract is
   filtered through a 0.45 jam polypropylene filtration membrane and the sample is stored
   at 4°C until analysis.

   The results from this test are used to determine the maximum quantity, or the fraction
   of the total constituent content, of inorganic constituents (Hg, As, Se, Pb, and Cd) in a
   solid matrix that potentially can be released from the solid material in the presence of a
   strong chelating agent. The chelated availability, or mobile fraction, can be considered
   (1) the thermodynamic driving force for mass transport through the solid material, or
   (2) the potential long-term constituent release. Also, a mass balance based on the total
   constituent concentration provides the fraction of a constituent that may be chemically
   bound, or immobile in geologically stable mineral phases.

   4.3.2 Tier 2 Solubility and Release as a Function of pH and LS  Ratio

   Test Method SR002.1 Alkalinity, Solubility and Release as a Function of pH
   (Kosson et al., 2002a) is the method to be used for Tier 2 pH Screening.  The protocol
   consists of 11 parallel extractions  of particle size reduced material at a liquid-to-solid
   ratio of 10 mL extractant per gram of dry sample. An acid or base addition schedule is
   formulated for 11 extracts with final solution pH values between 3 and 12, through
   addition of aliquots of HNO3  or KOH as needed. The exact pH schedule is adjusted
   based on the nature of the CCR; however, the range of pH values must include the
   natural pH of the matrix, which may extend the pH domain. The extraction schedule
   and the range of tested pHs are outlined in the developers' leaching test plan, "Draft
   (Revision #2), Sampling and Characterization Plan for Coal Combustion Residues
   from Facilities with Enhanced Mercury Emissions Reduction Technology" (Kosson et
   al., 2002b).

   If necessary, the material being evaluated is particle size reduced to <0.3 mm by
   sieving to remove any large pebbles present. A mortar  and pestle may be used to break
   up clumps of material. A 40 g dry sample of the reference fly ash or size reduced CCR
   is used for these tests. Using the schedule, equivalents of acid or base are added to a
   combination of deionized water and the reference fly ash or particle size reduced CCR.
   The final liquid-to-solid (LS) ratio is 10 mL extractant per gram of sample, which
   includes DI water, the added acid  or base, and the amount of moisture that is inherent
   to the waste matrix as determined by moisture content analysis. The 11 extractions are
                                                                                    Category III
                                                   16
100
-------
                                                           Characterization of Coal Combustion Residues
ARCADIS
                                                                                  QAPP FOR  THE
                                                                                  CHARACTERIZATION
                                                                                  OF COAL
                                                                                  COMBUSTION
tumbled in an end-over-end fashion at 28 rpm for a contact time of 24 hrs. Following
gross separation of the solid and liquid phases by settling, leachate pH measurements
are recorded and the phases are separated by pressure filtration through 0.45 urn
polypropylene filtration membranes.  Analytical samples of the leachates are collected
and preserved as appropriate for chemical analysis. For metal analysis, leachates are
preserved by acidification with HNOs to a pH <2 and stored at 4 °C until analysis. For
anion analysis, leachates are stored at 4°C until analysis.

Test method SR003.1 Solubility and Release as a Function of LS Ratio (Kosson et
al., 2002a) is the method to be used for Tier 2 LS ratio screening. The protocol consist
of five parallel batch extractions over a range of LS ratios (0.5,1,2, 5, and 10 mL/g
dry material) using the particle size reduced CCR and DI water as the extractant.
Extractions are conducted at room temperature in leak-proof vessels that are tumbled at
28±2 rpm for 24 hours. Solid and liquid phases are separated by settling and pH and
conductivity measurements are taken. The liquid is further separated by pressure
filtration using a 0.45 [tin polypropylene filter membrane.  Leachates are collected for
each of the 5 LS ratios and preserved as appropriate for chemical analysis. For metal
analysis, leachates are preserved by acidification with HNO3 to a pH <2 and stored at 4
°C until analysis. For anion analysis, leachates are stored at 4°C until analysis. The
range of tested LS ratios is outlined in the leaching test plan, "Draft (Revision #2),
Sampling and Characterization Plan for Coal Combustion Residues from Facilities
with Enhanced Mercury Emissions Reduction Technology" (Kosson et al., 2002b).

4.3.3 Tier 3 Mass Transfer Rate

Test method MT002.1 Mass Transfer Rates in Granular Materials (Kosson et al.,
2002a) is the method to be used for Tier 3 testing. This protocol involves continuous
water-saturation of the reference fly ash or CCR material.  Compacted granular
material is contacted with DI water using a liquid to surface area ratio of 10  mL DI
water for every cm2 of exposed solid surface.  Fresh DI water is exchanged with
leached material at specific intervals (2, 5 and 8 hours, 1, 2,4, and 8 days). The pH,
and conductivity for each time interval is also recorded. Each leachate sample is
prepared for chemical analysis by pressure filtration through a 0.45 (im polypropylene
filtration membrane and preserved as appropriate for chemical analysis. For metal
analysis, leachates are preserved by acidification with HNO3 to a pH <2 and stored at 4
°C until analysis.  For anion analysis, leachates are stored at 4°C until  analysis. In some
cases, this test will be extended up to a cumulative leaching tune of ca. 30 days to
provide more information about long-term material behavior.
                                                                                 Category
                                                17
                                                                                                         101
-------
Characterization of Coal Combustion Residues
   ARCADIS
CpAPP FDR THE
CHARACTERIZATION
DF COAL
COMBUSTION
   Test method MTOO1.1 Mass Transfer in Monolithic Materials will be used on the
   treated CCR samples when applicable instead of the MT002.1 protocol (Kosson et al.,
   2002a).
                                                                                  Category
   •4.4   FIXED-BED, TPD REACTOR SAMPLING

   Data on the thermal stability of toxic metals on CCRs during their commercial
   applications will be obtained using simulated conditions in a fixed-bed reactor called
   the Thermal Program Desorption (TPD) system. The time/temperature profiles
   experienced by CCRs in their commercial applications, type of other materials that
   CCRs are mixed with, and the composition of the flue gas that is in contact with CCRs
   have been determined by a separate study conducted by RTI International (EPA,
   2002a). The information from this report has been used to create simulated conditions
   in our in-house, fixed-bed TPD reactor. This reactor is described in detail in the
   Mercury Facility Manual. CCRs (mixed with other compounds) will be placed in the
   fixed bed reactor (about 10 grams). The simulated flue gas will be passed through the
   fixed-bed that is exposed to a specific time/temperature profile. Detailed description of
   the simulated flue gas preparation and temperature control of the fixed-bed reactor is
   also described in the Mercury Facility Manual. This manual also addresses the QA/QC
   issues regarding the operation of this reactor. The flue gas effluent of the fixed-bed
   reactor will be sampled for the duration of each test (the time profile) using two
   standard methods. EPA method 29 (EPA, 1996c) will be used to sample for As, Se, Pb,
   and Cd. Ontario Hydro Method (ASTM, 2002) will be used to sample for elemental
   and ionic mercury (inorganic forms of mercury). Since the effluents of the fixed-bed
   reactor is free of any entrained particulate matter, only the impinger portions of
   Method 29 and Ontario Hydro will be used (filter box will be eliminated).  After each
   test, the exposed CCR will be recovered and analyzed for Hg, As, Se, Cd, and Pb. The
   pre- and post-test analysis results of the CCRs together with the Method 29 and
   Ontario Hydro results will be used to determine mass balances for these five metals.
   This  experiment will also be performed on a sample of the reference fly ash material as
   a QC check.

   4.4.1    Thermal Desorption Test Plan for High Temperature CCR Commercial
           Processes

   A- Cement (Clinker) manufacturing:

   hi this process, fly ash is used as feed to a cement kiln. Fly ash typically represents a
   maximum of about 5% (weight) of the typical raw mix to the kiln. Other inputs include
                                                   18
102
-------
                                                          Characterization of Coal Combustion Residues
ARCADIS
                                                                                QAPP FOR THE
                                                                                CHARACTERIZATION
                                                                                OF COAL.
                                                                                CDMBUSTION
limestone (90%), iron ore (3%), and sand (2%), but fly ash can displace other
components of cement kiln like limestone and iron ore, so its percentage can go higher.
Cement kiln residence time is about one hour, temperature is about 1500°C, and the
gas environment is natural gas combustion flue gas. The assumption is that all Hg is
vaporized. Thus, the focus of this section will be on As, Cd, Se, and Pb. The test plan is
as follow:
    1-  Mix 5g fly ash, with 5g CaCO3 (limestone), 0.2g iron oxide (Fe203) and 0.5g
       sand and place the mixture in the fixed-bed reactor.

    2-  Inlet flue gas consists of 14% CO2, 3% O2,5.3% H2O, and 50 ppm NOX. The
       total flow rate will be 400 cc/min.

    3-  Heat the fixed-bed reactor to its maximum temperature (1000°C). With the
       current system, we will not be able to exceed 1000°C. If no evaporation of
       these metals is observed at this temperature, we will modify the system to
       achieve a desorption temperature of 1500°C.

    4-  Attach a mini-impinger, EPA Method 29 to the outlet of the fixed-bed reactor
       and sample the whole effluent of the fixed-bed reactor for As, Cd, Pb, and Se
       for one hour (residence time of cement kilns).

    5-  Recover the solid residue and the Method 29 train and submit for metal
       analysis.

    6-  Close metals' mass balances across all phases (pre-test fixed bed solid,
       exposed solid, and reactor effluent).

B- Wallboard Manufacturing

This process will be simulated using FGD waste (and not fly ash). Wallboard
manufacturing consists of two steps: calcining and drying. FGD waste experiences a
higher temperature during calcining, thus only this step will be simulated. The
calcining kettle temperature is about 310-370°C; but FGD waste temperature never
exceeds 170°C. FGD waste is not mixed with any other compounds, its residence time
in the kettle is about 1 hour and it is in contact with natural gas combustion flue gas.
The test plan will be as follow:
                                                19
                                                                                                      103
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                                    QAPP FOR THE
                                                                                    CHARACTERIZATION
                                                                                    OF COAL.
                                                                                    COMBUSTION
       1-  Place lOg of FGD waste in the fixed-bed reactor.

       2-  Inlet flue gas consists of 14% CO2,3% O2, 5.3% H2O, and 50 ppm NOX. The
           total flow rate will be 400 cc/min.

       3-  Heat the fixed bed reactor to 170°C.

       4-  Attach a mini-impinger Method 29 train to the outlet of the fixed-bed reactor
           and sample for As, Cd, Pb, and Se for one hour (residence time of calcining
           kettle); recover the solid residue and the Method 29 train and submit for As,
           Se, Pb, and Cd analysis.

       5-  Repeat steps 1 thru 3 (new batch of sample) and attach a mini-impinger
           Ontario Hydro train to the outlet of the reactor and sample for one hour.
           Recover solid residue and the train and  submit for Hg analysis. Ontario Hydro
           sampling will reveal the amount of total Hg, ionic Hg (inorganic Hg), and
           elemental Hg evolved from the calcining process simulation.

       6-  Close metals' mass balances across all phases (pre-test fixed bed solid,
           exposed solid, and reactor effluent).

    C- Asphalt manufacturing

    Asphalt manufacturing consists of two steps: a very short residence time mixing
    process (about one minute) and a long residence time storage process (several hours).
    The storage process occurs at temperatures of about 5°C higher than the mixing
    process. Thus, the most important step (in terms of thermal desorption) is the storage
    step. Hot mix asphalt (HMA) is 95% stone, sand, or gravel bound together with asphalt
    cement (crude oil); fly ash makes up approximately 5% of this mixture replacing
    natural fillers such as hydrated lime or stone. This mixture is hi contact with natural gas
    combustion flue gas. Storage temperatures usually range from 130-150°C for binder
    grade PG46-28 and  160-170°C for binder grade PG82-22. The test plan for asphalt
    manufacturing simulation will be as follow:

        1 -  Place 1 g of fly ash and 9 g sand in the fixed-bed reactor.

       2-  Inlet flue gas consists of 14% CO2, 3%  O2,5.3% H2O, and 50 ppm NOX. The
           total flow rate will be 400 cc/min.
                                                    20
104
-------
                                                          Characterization of Coal Combustion Residues
ARCADIS
                                                                                CPAPP FDR THE
                                                                                CHARACTERIZATION
                                                                                OF COAL
                                                                                COMBUSTION
   3-  Heat the fixed bed reactor to 170°C.

   4-  Attach a mini-impinger Method 29 train to the outlet of the fixed-bed reactor
       and sample for As, Cd, Pb, and Se for three hours (assumed residence time of
       asphalt storage); recover the solid residue and the Method 29 train and submit
       for As, Se, Pb, and Cd analysis.

   5-  Repeat steps 1 thru 3 (new batch of sample) and attach a mini-impinger
       Ontario Hydro train to the outlet of the reactor and sample for three hours.
       Recover solid residue and the train and submit for Hg analysis. Ontario Hydro
       sampling will reveal the amount of total Hg, ionic Hg (inorganic Hg), and
       elemental Hg evolved from the asphalt storage process simulation.

   6-  Close metals' mass balances across all phases (pre-test fixed bed solid,
       exposed solid, and reactor effluent).
                                                                               Category
5.D  TESTING AND MEASUREMENT PROTOCOLS

Whenever possible, standard methods will be used to perform required measurements.
Standard methods are cited in each applicable section. Where standard methods are
not available, operating procedures will be written to describe activities.  In situations
where method development is ongoing, activities and method changes will be
thoroughly documented in dedicated laboratory notebooks.

5.1   PHYSICAL CHARACTERIZATION

5.1.1 Surface Area and Pore Size Distribution

A Micromeretics ASAP 2400 Surface Area Analyzer will be used to perform
Brunauer, Emmett, and Teller (BET) method surface area, pore volume, and pore size
distribution analysis on each as-received and size reduced CCR. A 200 mg sample is
degassed at 200 C for at least one hour in the sample preparation manifold. Samples
are then moved to the analysis manifold, which has a known volume. Total gas
volume in the analysis manifold and sample tube is calculated from the pressure
change after release of an N2 and He mixed gas from the analysis manifold known
volume. Report forms are automatically generated after each completed  analysis. The
instrument uses successive dosings of N2 while measuring pressure. The surface area
                                                 21
                                                                                                      105
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                                     QAPF FDR THE
                                                                                     CHARACTERIZATION
                                                                                     OF COAL
                                                                                     COMBUSTION
   analyzer is calibrated annually by a Micromeretics service representative. Standards of
   known surface area are run with each batch of samples as a QC check. Detailed
   instructions for the operation of this instrument are included in the Mercury Facility
   Manual.
                                                                                   Category
   5.1.2 Density Measurements

   The Micromeretics Accupyc 1330, a helium pyncnometer, will be used to determine
   CCR density. This is a fully automatic gas displacement pycnometer, which measures
   the volume of solids. Instrument operation is based on the ideal gas law. By
   measuring the pressure change of He in a calibrated volume, the pycnometer
   determines sample volume from which density can be derived automatically if sample
   weight is know. Samples must be free of moisture to obtain true sample weight and to
   avoid the distorting effect of water vapor on volume measurement. The cell chamber
   should be kept closed at all times except when actually inserting or removing a sample.
   The size of the cell and expansion chamber is determined by calibration, which is
   performed every 10 runs or immediately prior to analysis after two weeks without
   operation.  The density measuring capacity of the Accupyc is calibrated for each run by
   analyzing two steel balls of known density.  Detailed  instructions for the operation of
   this instrument are included in the Mercury Facility Manual.

   5.1.3 pH and Conductivity

   pH and conductivity will be measured on all aqueous extracts. Conductivity is a
   measure of the ability of an aqueous solution to carry an electric current. This ability is
   dependent upon the presence of ions; on their total concentration, mobility, and
   variance; and on the temperature of the measurement.

   pH of the leachates will be measured using a combined pH electrode. A 2-point
   calibration will be done using pH buffer solutions. The pH meter will be accurate and
   reproducible to 0.1 pH units with a range of 0 to 14.

   Conductivity of the leachates will be measured using  a standard conductivity  probe.
   The conductivity probe will be calibrated using appropriate standard conductivity
   solutions for the conductivity range of concern. Conductivity meters are typically
   accurate to ±1% and have a precision of ±1%. The procedure to measure pH and
   conductivity will be as follow:
                                                   22
106
-------
                                                          Characterization of Coal Combustion Residues
ARCADIS
                                                                                 QAPP  FDR THE
                                                                                 CHARACTERIZATION
                                                                                 OF COAL.
                                                                                 COMBUSTION
Following a gross separation of the solid and liquid phases by centrifugation or
settling, a minimum volume of the supernatant to measure the solution pH and
conductivity will be taken and poured in a test tube. The remaining liquid will be
separated by pressure filtration and filtrates will be accordingly preserved and stored
for subsequent chemical analysis.
                                                                               Category
5.1.4  Moisture Content

Moisture content of the "as received" CCR, the reference fly ash and SRM samples
will be determined using ASTM D 2216-92 (ASTM, 1992). This procedure supercedes
the method indicated in the leaching procedure (Kosson et al., 2002a). This method,
however, is not applicable to the materials containing gypsum (calcium sulfate
dihydrate or other compounds having significant amounts of hydrated water), since this
material slowly dehydrates at the standard drying temperature (110°C). This slow
dehydration results in the formation of another compound (calcium sulfate
hemihydrate) which is not normally present in natural material. ASTM method C 22-
83 will be used to determine the moisture content of materials containing gypsum
(ASTM,  1983).

5.2   CHEMICAL CHARACTERIZATION

5.2.1 Carbon Content (TGA)

A Micromeretics Thermogravimetric Analyzer (TGA) is used to determine weight loss
due to thermal decomposition. Through knowledge of chemical components in the
sample, composition can be determined based on weight loss, decomposition
stoichiometry, and decomposition temperature. This method is used to estimate carbon
content of the CCR material. Carbon content corresponds to the weight loss in the
temperature range of 400-500°C, when TGA is operated in O2 or air. The TGA
requires temperature and weight calibration.  Temperature is calibrated using curie
point transistors of nickel and iron.  Weight is calibrated with a 100 mg standard
weight. Values are entered into the TGA whenever configuration is changed or at least
annually. Calcium oxalate monohydrate is analyzed to establish accuracy and is run at
least every 20 runs.

It should be noted that an alternative method for the analysis of carbon, nitrogen, and
sulfur is also available. This method uses a combustion technique followed by mass
spectroscopy. The instrument used in this analysis is a Carlo-Erba NA1500 Series II
                                                 23
                                                                                                       107
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                                    C3APP FOR THE
                                                                                    CHARACTERIZATION
                                                                                    DF COAL
                                                                                    COMBUSTION
   elemental analyzer. Similar instrumentation is also available from LECO Inc.
   (www.leco.com).

   5.2.2 Mercury (CVAA)

   Mercury analysis of each extract and leachate will be carried out by Cold Vapor
   Atomic Absorption Spectrometry (CVAA) according to EPA SW846 Method 7470A
   Mercury in Liquid Waste (Manual Cold Vapor Technique) (EPA, 1998). Samples
   are treated with potassium permanganate to reduce possible sulfide interferences. A
   Perkin Elmer FIMS 100 Flow Injection Mercury System is the instrument to be used
   for this analysis.  The instrument is calibrated with known standards ranging from 0.25
   to 10 jag/L mercury. The detection limit for mercury in aqueous samples is 0.05 ug/L.

   5.2.3 Other Metals (ICP)

   Analysis for arsenic (As), selenium (Se), cadmium (Cd) and lead (Pb) as well as
   principal constituents such as iron (Fe), calcium (Ca), phosphorus (P) and sulfur (S)
   will be performed on a ICP-MS using Method 3052 (EPA, 1996a).  Metals and
   estimated instrument detection limits are listed in the method.  The ICP will be profiled
   and calibrated for the target compounds and specific instrument detection limits will be
   determined. Mixed calibration standards will be prepared at least 5 levels.  Each target
   compound will also be analyzed separately to determine possible spectral interference
   or the presence of impurities. Two types of blanks will be run with each batch of
   samples. A calibration blank is used to establish the analytical curve and the method
   blank is used to identify possible contamination from varying amounts of the acids
   used in the sample processing. Additional daily QC checks include an Initial
   Calibration Verification (ICV) and a Continuing Calibration Verification (CCV). The
   1CV is prepared by combining target elements from a standard source different than
   that of the calibration standard and at a concentration within the linear working range
   of the instrument. The CCV is prepared in the same acid matrix using the same
   standards used for calibration at a concentration near the mid-point of the calibration
   curve.  A calibration blank and a CCV or ICV are analyzed after every tenth sample
   and at the end of each batch of samples. The CCV and ICV results must verify that the
   instrument is within 10% of the initial calibration with an RSD < 5% from replicate
   integrations. Procedures to incorporate the analysis of a MS/MSD for these CCR
   samples will be evaluated.

   These analyses will be  performed at two different ICP-MS facilities. The first facility
   is Severn Trent Laboratories in Savannah, Ga. This laboratory uses a Agilent ICP-MS
                                                                                  Category
                                                   24
108
-------
                                                           Characterization of Coal Combustion Residues
ARCADIS
                                                                                 PAPP FDR THE
                                                                                 CHARACTERIZATION
                                                                                 OF COAL
                                                                                 COMBUSTION
with octopole reaction system (ORS). The second facility is Vanderbilit University
(Department of Civil and Environmental Engineering). This laboratory uses a Perkin
Elmer model ELAN DRCII.  Standard analysis mode is used for PB and DRC mode is
used for analysis of As and Se.

5.2.4  Anions Analysis by 1C

Aqueous concentrations of anions (chloride, sulfate, sulfides, carbonate and nitrate)
will be determined using ion chromatography (1C). Standard methods (i.e., USEPA
guideline SW-846) will be used.

5.2.5  X-Ray Fluorescence (XRF) and Neutron Activation Analysis (NAA)

For the five target metals, XRF analysis will be performed on each CCR to provide
additional information on the CCR material. This information will be useful in
supplementing and/or validating CVAA and ICP results and calculating mass balances.
XRF is capable of detection limits in the ug range.  If levels are in the ng range, XRF
analysis will not be useful. Considering the high detection limit of the XRF, this
method will be used only as a second validation method or a "referee" method. Details
of XRF analysis are included in the Mercury Facility Manual.

Neutron activation analysis (NAA) is an established analytical technique with
elemental analysis applications. This method will be considered in this test program.
NAA is different than AA or inductively coupled plasma mass spectrometry (ICP-MS)
because it is based on nuclear instead of electronic properties. Neutron activation  .
analysis is a sensitive multielement analytical method for the accurate and precise
determination of elemental concentrations in unknown materials. Sensitivities are
sufficient to measure certain elements at the nanogram level and below, although the
method is well suited for the determination of major and minor elemental components
as well. The method is based on the detection and measurement of characteristic
gamma rays emitted from radioactive isotopes produced in the sample upon irradiation
with neutrons. Depending on the source of the neutrons, their energies and the
treatment of the samples, the technique takes on several differing forms. It is generally
referred to as INAA (instrumental neutron activation analysis) for the purely
instrumental version of the technique. RNAA (radiochemical neutron activation
analysis) is the acronym used if radiochemistry is used to separate the isotope of
interest before counting. FNAA (fast neutron activation analysis) is the form of the
technique if higher energy neutrons, usually from an accelerator based neutron
generator, are used.
                                                                                Category
                                                25
                                                                                                        109
-------
Characterization of Coal Combustion Residues
   ARCADIS
       FDR THE
CHARACTERIZATION
or COAL
COMBUSTION
                                                                                Category
   6.D  QA/QC  CHECKS
   6.1   DATA DUALITY INDICATOR GOALS
   Data quality indicator goals for critical measurements in terms of accuracy, precision
   and completeness are shown in Table 6-1.
    TABLE 6-1.  DATA QUALITY INDICATOR GOALS
Measuremen
t
As, Se, Pb,
Cd, Fe, P, S,
and Ca
Concentratio
n
Hg
Concentratio
n
Anions,
Sulfate,
Carbonates,
Chlorides
PH
Carbon
Content
Surface Area
Density
Moisture



Method

ICP/6010B



CVAA/7470A
IC/SW-846



Electrode
TGA

BET
Pycnometer
ASTM
D2216-92
ASTM C22-
83
Accuracy

10%



10%
1O%



2%
10%

5%
2%
N/A



Precision

10%



10%
10%



2%
1O%

5%
2%
1O%



Completenes
s
>90%



>90%
>9O%



100%
>90%

>90%
100%
N/A



    N/A: Not Applicable (see Appendix B)

    Accuracy will be determined by calculating the percent bias from a known standard.
    Precision will be calculated as relative percent difference (RPD) between duplicate
    values and relative standard deviation (RSD) for parameters that have more than two
                                                  26
110
-------
                                                           Characterization of Coal Combustion Residues
ARCADIS
                                                                                  CpAPP FOR THE
                                                                                  CHARACTERIZATION
                                                                                  DF COAL
                                                                                  COMBUSTION
replicates. Completeness is defined as the percentage of measurements that meet DQI
goals of the total number measurements taken.

Mass balance calculations will also be used as a data quality indicator. Different mass
balance recovery methods will be examined. The reference fly ash  sample will be
used to develop and validate an appropriate mass balance recovery method. Mass
balance will be determined by using the metals concentrations determined by analysis
of the "as-received" reference fly ash as the total. Results from successive leaching
samples and analysis of any solid residues will be combined to determine recoveries.

One approach that will be considered is the use of either total digestion (Method 3052)
or Neutron Activation Analysis (NAA) for the analysis of solid residues.

The mass balance recovery will be only performed on 3 pH points and one low LS
ratio. Uncertainty analysis will be considered for each mass balance. The selection of
the target pH values will be dependent on the natural pH of the material. If the natural
pH is <5, then natural pH, 7 and 9 will be selected as the target pH values. If the
natural pH ranges between 5 and 9, then 5, 7 and 9 will be selected as the target pH
values, and if the natural pH is >9, then 5, 7 and natural pH will selected as the target
pH values, hi addition, an extraction at the natural pH of the material and an LS ratio of
ImL/g will be carried out. At least 4 replicates per extract will be run. In the case
where the mass balance will be performed using total digestion or NAA, at least 3
representative samples per residue will be analyzed.

6.2  DC SAMPLE TYPES

Types of QC samples used in this project will include blanks, spiked samples,
replicates, and mass balance tests on the reference fly ash and the SRM. For physical
characterization testing, duplicate samples  of the CCR, reference fly ash and SRM will
be processed through each analysis. Duplicates must agree within ±10% to be
considered acceptable. For the leaching studies, an objective of this project is to
determine the appropriate types of QC samples to incorporate in the proposed leaching
methods. This will be accomplished by subjecting the reference fly ash to the leaching
procedure and determining the metals' mass balances by analyzing the leaching
solution  and the post-leachate solids. Initially, mass balances of 70-130% will be
considered as an acceptable QC of the leaching procedure. Further statistical analysis
on available data will be performed to narrow down the range of acceptable mass
balances. This method development will be thoroughly documented in a dedicated
laboratory notebook.  Leaching of the reference fly ash samples may also be used as
                                                                                 Category
                                                 27
                                                                                                         111
-------
Characterization of Coal Combustion Residues
    ARCADIS
GJAPP FOR THE
CHARACTERIZATION
DF COAL
COMBUSTION
   method controls during testing of CCR samples. For the fixed-bed reactor testing, one
   in every five tests will be run in duplicate. Duplicate results from the reactor testing
   are expected to agree within 20% to be considered valid. Identical to the leaching
   procedure, the use of the reference fly ash as a baseline QC sample will also be
   implemented during TPD tests (initial mass balances of 70-130%). Required QC
   samples for metals and mercury sampling trains are detailed in EPA Method 29 (EPA,
   1996c) and the Ontario Hydro Methods (ASTM, 2002). QC samples required for ICP,
   CVAA, 1C analysis are detailed in Methods 3052, 7470A, and SW-864 respectively.

   7.D   DATA REDUCTION, VALIDATION, AND REPORTING

   Chemical (ICP, CVAA, TGA, XRF, 1C, NAA) and physical (surface area, pore size
   distribution and density) characterization data are reduced and reports are generated
   automatically by the instrument software. The primary analyst will review 100% of
   the report for completeness and to ensure that quality control checks meet established
   criteria. If QC checks do not meet acceptance criteria, sample analysis must be
   repeated.  A secondary review will be performed by the Inorganic Laboratory Manager
   to validate the analytical report. If appropriate, certain chemical characterization data
   will be compared to the XRF and NAA analyses, In addition, the designated QA
   Officer will review at least 10% of the raw data for completeness. Analytical data will
   be summarized in periodic reports to the ARCADIS WAL. The procedure for
   reduction, validation and reporting of the leaching experiments Task II) are outlined in
   Appendix A. ARCADIS WAL is responsible for the implementation of these
   procedures. Data reduction and interpretation for the TPD experiments (Task IE) is
   included in the Mercury Facility Manual. ARCADIS WAL is responsible for
   implementing those procedures. ARCADIS WAL is also responsible for ensuring that
   mass balance criteria have been met. Mass balance for task n will be calculated as:

    [(metal in leachate + metal in residue)/metal in "as received" CCR]* 100

   Mass balance for Task ffl will be calculated as:

    [(metal in off gas + metal in thermally exposed sample)/metal in "as received"
    CCR]* 100

    Hg control benefit calculation for  each metal or other parameters (such as carbon
    content, surface area, and so on) is:
                                                                                 Category
                                                  28
112
-------
                                                          Characterization of Coal Combustion Residues
ARCADIS
                                                                                 C^APP FOR THE
                                                                                 CHARACTERIZATION
                                                                                 OF COAL
                                                                                 COMBUSTION
[(content in CCR, leachate, or residue)w Hgcontroi/(content in CCR, leachate, or
residue)w/oHgControi]*100

Values greater than 100% indicate an increase in that parameter due to implementation
of the Hg control technology. In summary, the ARCADIS WAL checks to ensure that
the duplicates meet the precision limits and the QC-checks for each run are within
documented data quality indicators. Progress of the research conducted under this
QAPP is reported in written monthly reports by ARCADIS WAL. Weekly verbal and
written progress reports  are submitted to the EPA WAM. Efforts will be made to
publish and present results. QA/QC activities will be mentioned in any published
materials. A data quality report will be provided in the final report of this investigation.

B.D  ASSESSMENTS

Assessments and audits  are an integral part of a quality system. This project is
assigned a QA Category in and, while desirable, doest not require planned technical
systems and performance evaluation audits. EPA will determine external or third-party
audit activities.  Internal assessments will be performed by project personnel to ensure
acquired data meets data quality indicator goals established in Section 6.  The
ARCADIS Designated QA Officer will perform at least one internal technical systems
audit (TSA) to ensure that this QAPP is implemented and methods are performed
according to the documented procedures. This audit will occur during the early stages
of the project to ensure any necessary corrective actions are implemented before large
amounts of data are collected.

There are currently not planned performance evaluation audits but Table 8-1 lists the
measurement parameters and expected ranges should EPA determine a PEA should be
provided.
                                                                                Category
B-l.  PEA PARAMETERS AND RANGES
Analyte or
Measurement
As, Se, Pb and Cd
Hg
PH
Method
ICP/3052
CVAA/7470A
Electrode
Expected Range
1-100 ug/mL
0.25 to 10ug/L
0-14
                                                29
                                                                                                       113
-------
Characterization of Coal Combustion Residues
   ARCADIS
                                                                                QAPP FDR THE
                                                                                CHARACTERIZATION
                                                                                DF COAL.
                                                                                COMBUSTION
   In addition to the TSA, the ARCADIS Designated QA Officer will perform an internal    Cate9°rV
   data quality audit on at least 10% of the reported data. Reported results will be verified
   by performing calculations using raw data and information recorded in laboratory
   notebooks.
   9.Q  REFERENCES

   ASTM, 1983.  Method C 22-83, "Standard Specification for Gypsum"

   ASTM, 1992.  Method D 2216-92, "Standard Test Method for Laboratory
   Determination of Water (Moisture) Content of Soil and Rock"

   ASTM, 2002.  Method D 6784-02, "Standard Test Method for Elemental, Oxidized,
   Particle-Bound, and Total Mercury in Flue Gas Generated from Coal-Fired Stationary
   Sources (Ontario-Hydro Method)"

   EPA, 1996a. Method 3052 "Microwave Assisted Acid Digestion of Siliceous and
   Organically Based Matrices". Test Methods for Evaluating Solid Waste,
   Physical/Chemical Methods (SW-846).

   EPA, 1996b. Method 6010 "Inductively Coupled Plasma-Atomic Emission
   Spectrometry". Test Methods for Evaluating Solid Waste, Physical/Chemical Methods
   (SW-846).

   EPA, 1996c . Method 29 "Determination of Metals Emissions from Stationary
   Sources." Code of Federal Regulations, Title 40, Part 60, Appendix A, April 1996.

   EPA, 1998. Method 7470A "Mercury in Liquid Waste (Manual Cold-Vapor
   Technique)". Test Methods for Evaluating Solid Waste, Physical/Chemical Methods
   (SW-846).

   EPA, 2002a. Characterization and Management of Residues from Coal-Fired Power
   Plants, Interim Report. EPA-600/R-02, 083, December 2002.

   EPA, 2002b. Reply to comments on EPA/OSW's Proposed Approach to
   Environmental Assessment of CCRs. Discussed March 5,2002.
                                                3O
114
-------
                                                        Characterization of Coal Combustion Residues
ARCAD1S                                                                   QAPP FOR THE
                                                                            CHARACTERIZATION
                                                                            OF COAL
                                                                            COMBUSTION

Kosson, D.S., van der Sloot, H.A., Sanchez, F. and Garrabrants, A.C., 2002a. An        Category 111
Integrated Framework for Evaluating Leaching in Waste management and Utilization
of Secondary materials. Environmental Engineering Science 19(3): 159-204.

Kosson, D.S., and Sanchez, F., 2002b. Draft (Revision #2), Sampling and
Characterization Plan for Coal Combustion Residues from Facilities with Enhanced
Mercury Emissions Reduction Technology. May 2002.
                                             31
                                                                                                  115
-------
Characterization of Coal Combustion Residues
                           Appendix D
                     Brayton Point Fly Ashes
116
-------
                                                                Characterization of Coal Combustion Residues
List of Figures
Figure                                                                                                 Page

D-l  pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	118
D-2  pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	118
D-3  Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	119
D-4  Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
     the Fly Ash with Enhanced Hg Control	120
D-5  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	121
D-6  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	122
D-7  Selenium Release (top) and Spike Recoveries (bottom)  as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	123
D-8  Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	124
D-9  Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH	125
D-10 Regression Curves of Experimental Data of Selenium Solubility as a Function of pH	126
D-ll 100-Year Arsenic Release Estimates as aFunction of the Cumulative Probability forthe Scenario of Disposal
     in a Combustion Waste Landfill	127
D-12 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
     in a Combustion Waste Landfill	127
D-13 100-Year Arsenic Release Estimates from A) Baseline Fly AshandB) Fly Ash with ACI	128
D-14 100-year Selenium Release Estimates from A) Baseline Fly AshandB) Fly Ash with ACI	129
                                                                                                            117
-------
Characterization of Coal Combustion Residues
pH Titration Curves
14 1
12.2
12
10 -
I 8
Q.
6 -
4 -
2
-1



Baseline Fly Ash
&

















El





I





* 01
^W





1
:
8
<
\





&













2 -10 -8-6-4-20 24 6 8
meq Acid/g dry
DSR2-BPB-0001 -A
OSR2-BPB-0001 -B
ASR2-BPB-0001 -C

14 -,
12
9.5 10
I 8-
Q.
6
4
2



Fly Ash with ACI







D






D







H
A
. . .5






i
B
0





0*
a





6

2 -1.5 -1 -0.5 0 0.5 1 1.5
meq Acid/g dry
n SR2-BPT-0001 - A
0 SR2-BPT-0001 - B
A SR2-BPT-0001 - C
Figure D-1. pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
pH as a  Function of LS Ratio
                Baseline Fly Ash
Fly Ash with ACI
-1O _
m

6
A
9
" S r

-
:
~
:
i





a











|
t




>
k
]




1?
m
I 0
0. °
A
9 -

-

L
-
~-



\




6
D











\





3



                24     6     8     10    12
                     LS ratio [mL/g]

                    n SR3-BPB-0001  - A
                    o SR3-BPB-0001  - B
                    A SR3-BPB-0001  - C
2     4    6     8    10    12
      LS ratio [mL/g]

    nSRS-BPT-0001 -A
    oSRS-BPT-0001 -B
    ASR3-BPT-0001 -C
Figure D-2. pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
118
-------
                                                           Characterization of Coal Combustion Residues
Mercury Release as a Function of pH
Baseline Fly Ash
Hg total content*: 650.6+6.8 ng/g
in *• n
MCL
1
U
"S> « A
Z: °-1
^ 0.035
.01
Onn-i




— . — .
«aw

D Q D



Fly Ash with ACI
Hg total content*: 1529.6+1.1 ng/g
MCL
1
IJ
nrn "Bi 0.1
95% .^
D)
I
k 0 01
0.006
" ~*o/_
, 	 T, n nm
.UU I I I I !'• x.xx i
2 4 5°7§ 8 1095°l212-214
PH
	 ML 	 ML
— - MDL
nSR2-BPB-0001 - A __ . MDL
OSR2-BPB-0001 - B
ASR2-BPB-0001 - C
*Total content as determined by digestion using method 3052.
1 50 •* cn
140

^ 130
•^ i?n

0) 100
3 90 -
Q.
w 80
D)
1 70 -
60

|
-
n
tr
i
i
:
;
i



T D n





140
• — • i^n
"7"" -ion
n n  110
oj inn
v on
Q.
w on
D) U°
I yn
i i en
2 4 6 8 10 12 14
PH
n SR2-BPB-0001 - A





	 — .
I I I I I I
2 4 5%




_ . _
- 1
, i ^ i ^J1'0! P

Q5C

H~— 5%

6 8 9-51095%12 14
PH
DSR2-BPT-0001 -A
OSR2-BPT-0001 - B
ASR2-BPT-0001 - C









t



n n











n u




? 4
D




u







LLlJ









] n








m




6 8 10 12 14
PH
SR2-BPT-0001 - A
Figure D-3. Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of mercury concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
                                                                                                    119
-------
Characterization of Coal Combustion Residues
Mercury Release as a Function of LS Ratio
Baseline Fly Ash Fly Ash with ACI
MCL 0 02 Mni n no
n ni*i -
IT
^ 001 -
D)
0 00*1
n
	 ML (
— - MDL
-i en
IDU
140
' — • i^n

>< 120
^
92 110
o
o mn
0> IUU
s~ nn
O> 90 -
.*:
'o_ 80
(/5
0) 70
I
60
en
(

-

L^1



s

n

a



, _ —



i



'.,,
n m ^
IT
™ n m
D)
n nn^
n


[
T,'


]
t



D



• - ^~

[

i

]

',.,

) 2 4 6 8 10 12 0 2 4 6 8 10 12
	 ML
LS ratio [mL/g] LS ratio [mL/g]
n SR3-BPB-0001 - A
0 SR3-BPB-0001 - B
A SR3-BPB-0001 - C
n SR3-BPT-0001 - A
0 SR3-BPT-0001 - B
A SR3-BPT-0001 - C
ie.n








;





















n ,













r









]






140
-i-an
^ -ion

!? 110 -
> mn
> 100
0 90
(D 3U
80
(D uu
-^ 70
•Q. /u
w en
°> en
I 50
>in
4U
•?n
-
'-
:_
-
I
:
=
:
_
:
:
:























, n
















	 c











] 	








D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
n SR3-BPB-0001 - A
n SR3-BPT-0001 - A
Figure D-4. Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
120
-------
                                                         Characterization of Coal Combustion Residues
Arsenic Release as a  Function  of pH
                 Baseline Fly Ash
            As total content*: 80.5+1.9 |jg/g
                Fly Ash with ACI
          As total content*: 27.9+2.1 |jg/g

1000

100

U MCL10 -
"Bj 6.67
" 1 -
< 0.1 -
0.01 -
0.001
t
\

=
nA
^Q
= AO
-

i_ _ _ _
2
=
-





n
A^ D

_ _ _ _ _ _ _



A

g



—n-S-

_ _






95%

5%



5% ' 95% 12 2
2 4 6 8 10 12 14
1 \J\J\J\J
1000 -
100 -
MCLHn

g> 4.8
— 1 -
< 0.1 -
0.01 -
n nm -

;
1 Bffl


!
5°,
2 4

A

A. tiff^^

/o 95 95%
6 8 10 1

a



2 1
                                                                                                95%
                                                                                                 5%
  ..... ML
 —  - MDL
                           PH
                   DSR2-BPB-0001 -A
                   OSR2-BPB-0001 -B
                   ASR2-BPB-0001 -C
	ML
— - MDL
                          PH
                  DSR2-BPT-0001 -A
                  OSR2-BPT-0001 -B
                  ASR2-BPT-0001 -C
*Total content as determined by digestion using method 3052.
140

"
•> nn
> nu
m mn
°J Qn
'Q_
w RO
< yn
cn
/
;
:
:
~
: n


:
:













[








]
D







D








D
Q-



I^IU
14D
" i^n
"C 1?0 -
^> nn
> I IU
CD 100
°J Qn
& 80-
< 70
en


:
:

D
:






u
D


















o-ir_







n
]












2 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH
n SR2-BPB-0001 - A
D SR2-BPT-0001 - A
Figure D-5. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of arsenic concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
                                                                                                 121
-------
Characterization of Coal Combustion Residues
Arsenic Release as a Function of LS Ratio
Baseline Fly Ash Fly Ash with ACI
4D An
-IK
on
„ OK
_l Z0
2. 20
.— *: ^-U
(/5
<1 ^
MCL
m
r;
0
C
	 ML
— - MDL
15D
140
" i^n
^ *M
>, -i on
9?
> -i-in
o
£ mn
o>
-^ Qn
Q.
w 80
(/5 OU
< 70
fin
i





: A f
: y 2

^^ i™ i •





^
i

^ T i^




o
S


• i ™i ^™








p i^—






5

^ i—^






i

- 1 ^—i •
oc -
an :
' — ' "y^
_j ^0 :
^. ?o :
(/5 :
<1 K
MCL 10 ;
5-
n •
) 2 4 6 8 10 12 0
LS ratio [mL/g] 	 ML
— - MDL
n SR3-BPB-0001 - A
oSRS-BPB-0001 -B
A SR3-BPB-0001 - C

I
\
-
\
~- n C
I
|
:
I




i








D

















c








i




150
140
g 130
>* 190
§ 110 -
£ 100 -
o>
^ 90 -
! 80-
< 70-
en


D


& &


^H • •








• • ^H




A
Q


— ^^m








— ^^m














a
;-—]•—•:

2 4 6 8 10 12
LS ratio [mL/g]
n SR3-BPT-0001 - A
oSRS-BPT-0001 -B
A SR3-BPT-0001 - C



I

:
;







•]









D

















[








3




D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratjo [mL/g]
n SR3-BPB-0001 - A
n SR3-BPT-0001 - A
Figure D-6. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
122
-------
                                                         Characterization of Coal Combustion Residues
Selenium Release as a  Function of pH

                 Baseline Fly Ash

            Se total content*: 51.4+1.7 |jg/g
      Fly Ash with ACI

Se total content*: 51.4+1.7 |jg/g
1000 -
100
MCL 57
=T 10
— 1 -
W 0.1-
0.01 -
n nm
t

• Sfos.




> A 5%
Z 4

*^ I—I i— i
i



6 8 1095°1
&





"12.2
2 1
95%

507



4
                                                      (U
                                                     CO
1000
164.3
100
MCL
10
9
i ^
0.1-
0.01
n nni


Sy *

— . — . ,

CO/
« a
^p

	 ' 	

9.5 qiso/n
*


- —


                                                                                              95%
                                                                                               5%
2 4 ^/U6 8 10^12 14
	 ML PH
— - MDL
Total content
150 -
140
^ 13° "
^ 120
> no
» 100 -
^ 90
Q.
% 80-
w 70
60 -
n SR2-BPB-0001 -A
0 SR2-BPB-0001 - B
ASR2-BPB-0001 -C

as determined by digestion using method 3C
;
:
i
;
; B
;
:
;
;













:








] n








n








Duc




2 4 6 8 10 12
PH
n SR2-BPB-0001 - A
1
	 ML
— - MDL
)52.
150
140
^ 130
21
^ 120
I 11°
2 100
3 90
Q.
w 80
(U
w 70
60
4

RO/ 9.5 QRO/
2 4 b/°6 8 10ybl2 14
PH
n SR2-BPT-0001 - A
0 SR2-BPT-0001 - B
A SR2-BPT-0001 - C




-




r




: C








n n







n








	 c
°°\








D








cP





I 4 6 8 10 12
PH
14
n SR2-BPT-0001 - A
Figure D-7. Selenium Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
Fly Ash with Enhanced Hg Control. 5th and 95th percentiles of selenium concentrations observed in typical combustion
waste landfill leachate are shown for comparison.
                                                                                                 123
-------
Characterization of Coal Combustion Residues
Selenium Release as a Function of
Baseline Fly Ash
ROD
cnn
/inn
4UU
_i
^ onn
.-V oUU -
, -ion
O>
> -i -in
8 11° "
tu -inn
CD
-2 an
._ 90
w 80
0) OU
W 70
Rn




: fi
;













D
u













Q





l
	 * 	
) 2 4 6 8 10 1
LS ratio [mUg]
LS Ratio
600
500
400
u
1 300
O)
W 200
100
MCL
0
	 ML
— - MDL
n SR3-BPB-0001 - A
0 SR3-BPB-0001 - B
A SR3-BPB-0001 - C
Fly Ash with ACI

: 6
_ D
~






g








6
D












<
I






5



0 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-BPT-0001 - A
o SR3-BPT-0001 - B
A SR3-BPT-0001 - C
icn

:
:
;
1
n c

1
1





3








D
















I








|]




140
^F 1^0-
sl I<:3U
^ 120 -
O)
> 110
8
P mn
O)
-^ Qn
Q.
w 80
O) OU
W 70
Rn





	 [|
















D
















C








]





D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
nSRS-BPB-0001 -A
n SR3-BPT-0001 - A

Figure D-8. Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
124
-------
                                                        Characterization of Coal Combustion Residues
Arsenic Solubility
                 Baseline Fly Ash
10000 -

1000 -
A r\r\
100 -

10 -

1 -
0.1 -
0.01 -
n nm

r




"^SL
^S^^_







>. 4





>— -S^
^^CH!
*i




i i i i
0
6 8 10 1:

A
Jl
n
f
A
, .f&
^9
i» T




95%
I 1



95%



CO/
Ovo




4
2

4 6 8 10 12
PH
n SR2-BPB-0001-A
0 SR2-BPB-0001-B
A SR2-BPB-0001-C
Fit n irxjp

14

Fly Ash with ACI
10000 -
1000 -
100 -
10 -
1 -
0.1 -
0.01 -
n nm -
/
:
i
:
!
.a ...
i N
;
!
:
I
|
' COy
I 4

--^H--#**

6 8 10 1

•ffl-
6

95%
2 1

KO/
Ovo
4
                                                                             PH
                                                                      n  SR2-BPT-0001-A
                                                                      o  SR2-BPT-0001-B
                                                                      A  SR2-BPT-0001-C
                                                                     	Fit curve
Material
BPB
BPT

0.0004 PH5
-0.8130 pH2
0.0005 pH5
-2.0113 pH2
log As (ng/L)
-0.0135 pH4
1.1609 PH
-0.0207 pH4
5.4552 PH
pH range of
validity
0.1 634 pH3 3-14
2.7085
0.3035 pH3 3-12.5
-2.7126
R2 Number of
points
0.77 27
0.98 33
Figure D-9. Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH.
                                                                                               125
-------
Characterization of Coal Combustion Residues
Selenium Solubility
Baseline Fly Ash Fly Ash with ACI
mnnnn -mnnnn
10000 -
1000-
U 100-
1 10-
-------
                                                    Characterization of Coal Combustion Residues
100-Year Arsenic Release  Estimates

    10000
 -5  1000
  0
 O
 O
n
n

1
1 .




L
' 4


LJ
n A i

j A
A
k 	
DD[
nn° A
AAA*
A


]
L



          0  5% 20    405°% 60    80 "" 100

                     Percentile

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
BF
ng/kg
0.2
0.9
152
2095
468
4693
3B
0.0003
0.0011
0.2
2.6
0.6
5.8
BF
0.1
0.1
22
338
90
10157
DT
%
0.0003
0.0005
0.0772
1.2
0.3
36.4
               n BPB
A BPT
Figure D-11.100-Year Arsenic Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
in a Combustion Waste Landfill.
100-Year Selenium Release Estimates
 O
 O
i uuuuu -
mono -
3
D mnn -
100 -
•• 10 -

OA
.1 H
(
!
!
:
I
\t
:
)


AA^1
fl«*
A ™
Q
!i

5%20 40 5C
^i





o/o ' ' 95°/
60 80






'o
1C






30

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
BF
ng/kg
1.5
7.6
1031
16681
3573
45169
3B
%
0.003
0.015
2.0
32.5
7.0
87.9
Bl
ng/kg
5.9
10.0
1661
23931
6038
151900
DT
%
0.004
0.007
1.1
15.8
4.0
100.0
                     Percentile
               nBPB
A BPT
Figure D-12. 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of
Disposal in a Combustion Waste Landfill.

                                                                                         127
-------
Characterization of Coal Combustion Residues
100-Year Arsenic  Release Estimates
                                   BPB -Arsenic
80500 |^g/kg
100000 100%
'o) 1 nnnn
1
8i
m mnn -
OJ
.0
c
0)
(/> H flfl
CO
CO
8 10-
1 -
A\



: 	
Tol
n

1
al cont
1
20



95 ^g/kg 2439 ^g/kg
9fi% 3.0%











105^/kg 83 ng/kg
0.1% 010/0


ent Combustion Default-pH3 Default-pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence ^2
                                                                            — MCLLS95%=1000|jg/kg


                                                                            -MCLLS125 = 125ug/kg
                                   BPT - Arsenic
H? 10000 -
^
8i
!J5 -innn
0-year arsenic rele
->• 0 C
DOC
i
1 .

275
• 	

• 	

300 |ig
100%



'kg
19
SSSjj/kg



1.2%





81 ng/
7.1%



kg
999 i^g/kg
3.6%





60 ^g/kg
0.2%






Total content Combustion Default-pH3 Default-pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 9.5



                                                                                       =1000|jg/kg
    B)

Figure D-13.100-Year Arsenic Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI. Release estimates
for percolation controlled scenario are compared to release estimate based on total content. The amount of the arsenic that
would be released if the release concentration was at the MCL is also shown for comparison (LSdefaultscenario = 12.5 L/kg and
LS95%=100L/kg).

128
-------
                                                          Characterization of Coal Combustion Residues
100-Year  Selenium  Release Estimates
                                     BPB - Selenium
       100000
        10000
    -£   1000
    I    100
     CO
     CO

    I     10
                51400







100%



16*






381 u9
32.5%



'kg
2667 ug/kg 3226 ug/kg
5.2% 6.3%
















71 3 ug/kg
1.4%













    A)
                Total content    Combustion   Default-pH3    Default-pH      Default-
                           Waste Landfill -                   12.5       Natural pH
                           95% confidence                               12.2
                                                       MCLLS95% = 5000 ug/kg


                                                       MCLLS125 = 625 ug/kg
                                    BPT- Selenium
      1000000
       100000
        10000
         1000
     TO
     a>
     >,
    6
    o
          100
           10
               151900 i ia/ka

— •




100%
	

23931 ,,a/ka 43479 ug/kg

	 .




15.8%
	


	 JC




41-pQL
0.9%


Kg 	 .




28.6%
	

2054 u9/kg
	




14% 	







—

                                                                               _MCL    = 5000 ug/kg
                                                                                  MCLS1  = 625 ug/kg
               Total content
    B)
 Combustion
Waste Landfill -
95% confidence
Default-pH 3
Default-pH
   12.5
 Default-
Natural pH
   9.5
Figure D-14.100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI. Release estimates
for percolation controlled scenario are compared to release estimate based on total content. The amount of the selenium
that would be released if the release concentration was at the MCL is also shown for comparison (LSdefaultscenario = 12.5 L/kg
andl_S95%=100L/kg).
                                                                                                  129
-------
Characterization of Coal Combustion Residues
Comments
Figure D-3:
  •  The fly ash from the test case had greater total Hg
    content than the fly ash from the baseline case  (by
    about 2.5  times).
  •  Hg release is low (but poor replication) for both baseline
    and test cases.

Figure D-5:
  •  The fly ash from the test case had lower total As con-
    tent than the fly ash from the baseline case (by about a
    factor of 3).
  •  The laboratory measurements fit within the 5 -95 % con-
    fidence intervals of the field observations.
  •  As release is most frequently worse in baseline case
    than test  case and exceeds MCL for many possible
    conditions.

Figure D-7:
  •  The fly ash from the test case had greater total Se
    content than the fly ash from the baseline case  (by
    about a factor of 3).
  •  The laboratory measurements fit within the 5 % to 95 %
    confidence intervals of the field observations.
  •  Se release substantially exceeds MCL for both baseline
    and test cases and is generally worse for test cases.
  •  The fly ash from the test case resulted in greater Se
    concentration at the natural pH of the  material than
    the baseline case (by about 3 times).

Figures D-11. andD-12:
  •  The fly ash from the test case would result in As re-
    lease  less  than expected from the baseline case, with a
    95% probability to be less than 338 and 2095 |jg/kg,
    respectively.
  •  No significant difference in Se release is expected from
    both baseline and test cases.

Figure D-13:
  •  For the 95% probability scenario, arsenic release from
    the baseline case would be greater than the amount
    that would be released if the release concentration was
    at the MCL.
  •  For the 95% probability scenario, a lower arsenic re-
    lease would be expected from the test case. However,
    the fly ash from the test case had lower total As con-
    tent than the fly ash from the baseline case (by about a
    factor of 3).
  •  For the default scenario corresponding to disposal in a
    monofill (leachate pH controlled by the material being
    disposed), no significant difference in arsenic release
    between the baseline and the test cases would be ex-
    pected. Additionally, arsenic release would be less than
    the amount that would be released if the release con-
    centration was at the MCL.
  •  For the default scenario corresponding to the "extreme"
    pH of 3, arsenic release would be greater than the
    amount that would be released if the  release concen-
    tration was at the MCL, for both the baseline and the
    test cases.
  •  For the default scenario corresponding to the "extreme"
    pH of 12.5, arsenic release is expected to  be greater
    for the test case than the baseline case.

Figure D-14:
  •  For the 95% probability case,  selenium release would
    be greater than the amount that would be released if
    the release concentration was at the MCL for both the
    baseline and the test cases.
  •  For the default scenario corresponding to disposal in a
    monofill (leachate pH controlled by the material being
    disposed), a greater selenium release would  be ex-
    pected from the test case.
  •  For the default scenario corresponding to the "extreme"
    pH of 12.5, selenium release is expected to be greater
    for the test case than the baseline case. Selenium re-
    lease from the test case would be greater than the
    amount that would be released if the  release concen-
    tration was at the MCL.
130
-------
                   Characterization of Coal Combustion Residues
        Appendix E
Pleasant Prairie Fly Ashes
                                              131
-------
Characterization of Coal Combustion Residues
List of Figures
Figure                                                                                                 Page

E-l  pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	133
E-2  pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	133
E-3  Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	134
E4  Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
     the Fly Ash with Enhanced Hg Control	135
E-5  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	136
E-6  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio forthe Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	137
E-7  Selenium Release (top) and Spike Recoveries (bottom)  as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	138
E-8  Selenium Release (top) and Spike Recoveries (bottom)  as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	139
E-9  Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH	140
E-10 Regression Curves of Experimental Data of Selenium Solubility as a Function of pH	141
E-ll 100-Year Arsenic Release Estimates as a Function of the Cumulative Probability forthe Scenario of Disposal
     in a Combustion Waste Landfill	142
E-12 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
     in a Combustion Waste Landfill	142
E-13 100-Year Arsenic Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI	143
E-14 100-year Selenium Release Estimates from A) Baseline Fly AshandB) Fly Ash withACI	144
132
-------
                                                       Characterization of Coal Combustion Residues
pH Titration Curves
Baseline Fly Ash
14
1? -
11.2
m
-i- 8
i °
Q.
6 -
4
2 -

:- 4


-


,




$
§ a
o







m.




-r-H-iDj




^
QWS" P

2 0 2 4 6 8 10
meq Acid/g dry
DSR2-PPB-0001 -A
o SR2-PPB-0001 -B
A SR2-PPB-0001 -C
Fly Ash with ACI
14
19
11.9
m
53
I 0 -
a.
-

-\
-t
a





A






)
o
^












D




J^A

DQ
1 -2 0 2 4 6 8
meq Acid/g dry
DSR2-PPT-0001 -A
OSR2-PPT-0001 -B
ASR2-PPT-0001 -C
Figure E-1. pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
pH as a  Function of  LS Ratio
                Baseline Fly Ash
  Q.
14

12

10

 8

 6
                                     I
                      4     6     8    10    12
                      LS ratio [mL/g]

                    nSRS-PPB-0001 -A
                    oSRS-PPB-0001 -B
                    ASR3-PPB-0001 -C
                                                    I
                                                    Q.
                                                           Fly Ash with ACI
19
m


A

(

i
:




!





0











i





\




D 24 6 8 10 12
LS ratio [mL/g]
n SR3-PPT-0001 - A
oSRS-PPT-0001 -B
A SR3-PPT-0001 - C
Figure E-2. pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
                                                                                              133
-------
Characterization of Coal Combustion Residues
Mercury  Release  as  a Function  of pH

                 Baseline Fly Ash

            Hg total content*: 157.7+0.2 ng/g
               Fly Ash with ACI

         Hg total content*: 1180.1+1.2 ng/g
  B)
  D)
MCL
1
0.1
0.011
0.01
.001 -
:
=
; ^
o
p. ,. I-DI-. ••.$
. — ^ — _
B> A



A ,
'"'n , "i
— . 	 . _Q -i —

5%' ' 1 1".2



A
i
1 	
- —

95%
O)


O)
                                             95%     3:
                                             5%
MCL
1
0.1 -
0.011
OfH
.U1 -
n nm

=
E
u
- O Z&
LOOP V



00 n ° n<
o ,
— - — - — - — j
LJ LJ A ^ £
11.9"


i 9
- —
 110
e 100 -
3 90
Q.
w 80 -
D)
x 70
60 -
n SR2-PPB-0001 - A
0 SR2-PPB-0001 - B
A SR2-PPB-0001 - C
2 4 6 8 10 12 14
PH
	 ML
— -MDL
n SR2-PPT-0001 -A
OSR2-PPT-0001 -B
ASR2-PPT-0001 -C


as determined by digestion using method 3052.
-icn

;
-
i




i





p
i














n








=h r








:






r— i -i -an
"C 190
(D
> 110
0
 on
'o.
w an
D)
I 70 -
fin

;
;
; a

;






n n








nn









a D








D








a D




2 4 6 8 10 12 14 2 4 6 8 10 12
pH pH
n SR2-PPB-0001 - A
14
n SR2-PPT-0001 - A
Figure E-3. Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of mercury concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
134
-------
                                                      Characterization of Coal Combustion Residues
Mercury Release as a  Function of  LS  Ratio
                Baseline Fly Ash
                                                                 Fly Ash with ACI
0 01 5
n1
?£ n m
O)
Onn^i
n
c
	 ML
— - MDL
-
-
i


i
r~\"



- o-



- —
,
<

- ^— i
^
>

> —

) 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-PPB-0001 - A
o SR3-PPB-0001 - B
A SR3-PPB-0001 - C
n 01^
IT
§. n m
D)
0 00*1
n
C
	 ML
— - MDL


A [
D i


]
\



D



- —
,
I

, '
>
]

\
) 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-PPT-0001 - A
o SR3-PPT-0001 - B
A SR3-PPT-0001 - C
140 -

^p1 nn
°l
e^ 1?f) -
o>
^
p mn
O)
1/5 80
O) ou
?n
Rn
;

I
:
;
~
;
~
;
:
















u

















[









]





I-JU
14H

To1 1 *^n
ll
>< -ion
o>
E-inn
O)
1/5 80 -
O) ou
?n
Rn .


i
;

;

;
;

















D


















C









]




                     4    6    8    10
                     LS ratio [mL/g]
12
                                                                 2    4    6    8    10   12
                                                                      LS ratio [mL/g]
                   DSR3-PPB-0001 -A
                                                                    DSR3-PPT-0001 -A
Figure E-4. Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                           135
-------
Characterization of Coal Combustion Residues
Arsenic Release as a  Function of  pH

                 Baseline Fly Ash

           As total content*: 21.3+0.3 |jg/g
      Fly Ash with ACI

As total content*: 24.0+0.8 |jg/g
1000 -
100 -
MCLHn
g_ 4
1 ' 1
< 0.1
0.01
n nm
t

1
i - ^itt&Ki
^^m m ^^H • i


' 5'
> 4


. . . -oa£q£ o 5
_ _ 	 _ 	 _ . 	


A 112
6 8 10 1


r
— ^^^


95%
2 1
95%


b%


4
1000
100 -
CL10
4 2
1
0.1 -
0.01 -
n nm
f
i


1 - 3«&-
_ _ _ _ ,


5°
2 4


-oao&r -C&-0- -d
_._._._


/ "1 "1 Q
6 8 10 1

C
*>A
- —


95%
2 1
95%


5%


4
Ml
— - MDL
Total content
1^0 -,
140

"T"1 -ion
^_
^ 110
> I IU
8 100
S qo
'o.
w 80
t/>
< 7Q -
fin
/
2 4 6 8 10 12 14
PH
DSR2-PPB-0001 -A
OSR2-PPB-0001 -B
ASR2-PPB-0001 -C

1
	 ML

^— - IVIUL
2 4 6 8 10 12 14
PH
n SR2-PPT-0001 - A
o SR2-PPT-0001 - B
A SR2-PPT-0001 - C


as determined by digestion using method 3052.
-icn

~
:
:




: ^
[T
1
I
-
]

















D







n
B [








D





I 4 6 8 10 12
pH
i4n

" 'ion

o
o
m inn
S qn
w «n
< 7Q
fin

:


; B
i







D D








n n









n








n







_,
D



14 2 4 6 8 10 12
pH
n SR2-PPB-0001 - A

14
n SR2-PPT-0001 - A
Figure E-5. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of arsenic concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
136
-------
                                                      Characterization of Coal Combustion Residues
Arsenic Release  as a Function of  LS Ratio
                Baseline Fly Ash
       20
        15
% MC1L0
             o
             A
      ML
      MDL
                     468
                     LS ratio [mL/g]
                                   10   12
                 nSRS-PPB-0001 -A
                 oSRS-PPB-0001 - B
                 ASR3-PPB-0001 -C
                                                      20
                                                      15
^ MCL
.3    10
   0

• ML
 MDL
                                                               Fly Ash with ACI
                                                                        D
                                                                        fi
                   468
                   LS ratio [mL/g]
10   12
                  n SR3-PPT-0001 - A
                  oSRS-PPT-0001 -B
                  ASR3-PPT-0001 -C
  £
  o
  o
  £
  o>
  '5.
  (/5
i4n

1?0
nn
100
on
80
70
fin
c



:
|
; D C

;
;





]







D

















[








]




D 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-PPB-0001 - A
                                                o>
                                                o
                                                o
                                                £
                                                o>
                                                'o.
     150
     140
     130
     120
     110
     100
      90
      80
      70
      60

~_

\
- U J

;

~_




]








D


















[








]



D 24 6 8 10 12
LS ratio [mL/g]
n SR3-PPT-0001 - A
Figure E-6. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                           137
-------
Characterization of Coal Combustion Residues
Seleniurr
10000 -
1000
110.9
100
MCL
ZT 10
1 1-
o>
w 0.1
0.01 -
0.001
Ml

— - MDL "
\ Release as a Function of pH
Baseline Fly Ash Fly Ash with ACI
Se total content*: <4.0 ug/g Se total content*: <4.0 ug/g
: D_ <*L
I - Z3QJ -<&
|

i
- - - -o-B A Mi O


1000 -
to QF,%
' ncL100
— 25. 3.
ZT 10
^•y O?
- — CD
w 0.1 -
0.01
I, n nn-i

= RftAl ^7S
-

i
B
on rvKl CA o



'"" ' ' '5% ' '112 95% U'UU1 5%' ' 11.995%
2 4 6 8 10 12 14 2 4 6 8 10 12 14
PH 	 MI PH


n SR2-PPB-0001 - A ~~ ' IVILJL
o SR2-PPB-0001 - B
A SR2-PPB-0001 - C B)

DSR2-PPT-0001 -A
OSR2-PPT-0001 - B
ASR2-PPT-0001 -C
*Total content as determined by digestion using method 3052.
-i en -i en
140

£ 13°
£, 120 -
> 110
8
u 100 -
9 qn
w 80
 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH
n SR2-PPB-0001 - A
a SR2-PPT-0001 - A
Figure E-7. Selenium Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
Fly Ash with Enhanced Hg Control. 5th and 95th percentiles of selenium concentrations observed in typical combustion
waste landfill leachate are shown for comparison.
138
-------
                                                          Characterization of Coal Combustion Residues
Selenium Release as a Function of LS Ratio
Baseline Fly Ash Fly Ash with ACI
1fiO 'lcn
140
120
U 100
9. 80 -
CD
(/) 60
MCL 1
40
20
Oj
C
	 ML
— - MDL
•icn
140
•^ 1 30
^
>, -|00
o>
> -i-in
8
P 100
O>
-^ QO
Q.
w 80 -
0) OU
W 70
RO




: 6

A










1







R
u













<








>





) 2 4 6 8 10 1
LS ratio [mL/g]
nSRS-PPB-0001 -A
oSRS-PPB-0001 -B
ASR3-PPB-0001 -C
140
-i on
IZU
i — i mn
9. sn
.^r OU
O)
60
MCL
40
20









A








i








B
















E








?

T
2 0 2 4 6 8 10 12
_ _ ML LS ratio [mL/g]
— - MDL
n SR3-PPT-0001 - A
o SR3-PPT-0001 - B
A SR3-PPT-0001 - C
i en



|

: [
;







3








n
















r








i




140
^F 1^0
>, -|00
o>
> -I -in
8
P 100
-^ an
._ 90
w 80 -
O) OU
70
en
_
;
;
I
I
; D [
;
;
I





j]








D

















C








]



D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
n SR3-PPB-0001 - A

n SR3-PPT-0001 - A
Figure E-8. Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                                   139
-------
Characterization of Coal Combustion Residues
Arsenic Solubility
                 Baseline Fly Ash
                                                                     Fly Ash with ACI
\J\J\J\J\J -
10000 -
1000 -
100 -
10 -
1 -
0.1 -
0.01 -
n 001

! ^J^ass






f

10000 -
1000 -
950/0 u 100 -
]t 10 -
5% £ 1
0.1
0.01
n nn-i
|
-%$-<*


ux>o-D95i — ca-a — oS

	 9
PA f

U'UU1 ' 5% ' ' '95% 	 ' 5% ' ' '95%
2 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH


n SR2-PPB-0001-A
o SR2-PPB-0001-B
A SR2-PPB-0001-C
Fit rurvp


n SR2-PPT-0001-A
o SR2-PPT-0001-B
A SR2-PPT-0001-C
Fit rurvp


Material log As (|ig/L) pH range of R2 Number of
validity points
PPB 0.0003 pH5 -0.0117 pH4 0.1674 pH3 3-12.5 0.47 33
-1.1600pH2 3.8716 PH -4.2743
PPT 0.0006 pH5 -0.0238 pH4 0.3460 pH3 3-13 0.84 33
-2.3840 pH2 7.7036 PH -8.6786


Figure E-9. Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH.
140
-------
                                                          Characterization of Coal Combustion Residues
Selenium Solubility
Baseline Fly Ash Fly Ash with ACI
1 nnnnn •" ^nnnn
10000 -
1000-
:r 100-
.!• 10-
w 1 -
0.1 -
0.01 -
n nni -



3A
& 	

2 4 &/°6 8 10 ir/01
PH


n SR2-PPB-0001-A
o SR2-PPB-0001-B
A SR2-PPB-0001-C
Fit n \r\fp

10000 -
1000-
95% ^ 100 .
§ 10-
5% » ,
\J j | —
0.1 -
0.01 -
n nm -
4


| £&-<&
i
1

^O-DQS — CA^ r
^*>qj

D
/


2 4 5%6 8 10 il5%1
PH
D SR2-PPT-0001-A
0 SR2-PPT-0001-B
A SR2-PPT-0001-C
Fit n ir\)p


Material log Se (|ig/L) pH range of R2 Number of
validity points
PPB 0.0006 pH5 -0.0234 pH4 0.3605 pH3 3-12.5 0.54 33
-2.6711 pH2 9.4536 PH -10.6432
PPT 0.0012 pH5 -0.0440 pH4 0.6379 pH3 3-13 0.81 33
-4.3855 pH2 14.2483 PH -15.6626
95%
5%
4


Figure E-10. Regression Curves of Experimental Data of Selenium Solubility as a Function of pH.
                                                                                                  141
-------
Characterization of Coal Combustion Residues
100-Year Arsenic  Release Estimates
 Arsenic
IUUUU •
1000 -
^ 100 -
-1— •
s
1_
ro m
(D IU •
O
O
01-
c
5



-[
)


fififi5
HaBUB
!
5%20 40 5(

^



)%60 8095'

)



^(





DO

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
PF
Hg/kg
0.1
0.2
36
473
107
1188
3B
%
0.0006
0.0010
0.2
2.2
0.5
5.6
PF
Jig/kg
0.1
0.2
27
358
81
1049
3T
%
0.0004
0.0007
0.1
1.5
0.3
4.4
                      Percentile
                nPPB
APPT
Figure E-11.100-Year Arsenic Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal

in a Combustion Waste Landfill.
100-year Selenium  Release Estimates
 Selenium
IUUUUU •
mnnn -
O)
^B) innn -
-i— •
^ 100 -
CD
(D
>s -in -
o
o
^~ -I .
01-
c




S

)


HfiSaS
HB
ITI
A
1
i i i i i i i i
5% 20 40 5C

BHBHHBBI
?HA


1 1 1 1 1 1 1 1
)% ' ' 95°/
1/0 60 80

g




'o
1(






DO

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
PF
jag/kg
3.0
4.4
733
4000
2254
4000
3B
%
0.1
0.1
18.3
100.0
56.4
100.0
PF
jag/kg
0.9
3.3
516
4000
1599
4000
3T
%
0.0
0.1
12.9
100.0
40.0
100.0
                     Percentile
               nPPB
APPT
Figure E-12. 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of

Disposal in a Combustion Waste Landfill.

142
-------
                                                        Characterization of Coal Combustion Residues
100-Year Arsenic Release Estimates
                                   PPB -Arsenic
       100000
    iP   10000
S   1000
    o
    TO
    0)
         100
          10
     A)



-

-
100%





473 ug/kg



2.2%


104 ug/kg
0 5%
45 |id/kg_ u.^/u 5Q jl0;/kq


0.2%




0.2%


Total content Combustion Default-pH 3 Default-pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 11.2
                                                                             -MCI     =1000 ug/kg
                                             ™-r=P===^P=- MCI     =125 ug/kg
       100000
    5>  10000
    
-------
Characterization of Coal Combustion Residues
100-Year Selenium Release Estimates
CD

"35


E
2
'c
0)
0)
     6
     o
        10000
         1000
          100
           10
                               PPB -Selenium


          <4000 ug/kg   4000 ug/kg
                                                     4000
     A)
; 100% 100% 100%
:



_
Tol

	
al cont








565 ug/kg
14 1%










^l'




B6jjg7
34.7%


— —




ent Combustion Default-pH3 Default-pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 11.2



                                                                        — MCLLS95% = 5000 ug/kg





                                                                           MCLLS125 = 625 ug/kg
                                   PPT-Selenium
     0)
     (/)
     TO
     TO
     0)
     CD
     O
        10000
         1000
          100
           10
<4000 u9/kg 4000 ug/kg
; 100% 100%




Tol


al cont




.
ent Co
Was
95%


	
mbust
te Lan
confid
740 ug/kg
448 ug/kg 18.5% 316 ug/kg




J.1.2.%









1.9% 	




on Default -pH 3 Default -pH Default -
dfill- 12.5 Natural pH
ence 11.9


                                                                                MCL     = 5000 ug/kg
     B)


Figure E-14.100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI. Release estimates

for percolation controlled scenario are compared to release estimate based on total content. The amount of the selenium

that would be released if the release concentration was at the MCL is also shown for comparison (LSdefaultscenario = 12.5 L/kg

andl_S95%=100L/kg).

144
-------
                                                            Characterization of Coal Combustion Residues
Comments
Figure E-3:
  •  Fly ash from the test case had greater total Hg content
    than the fly ash from the baseline case (by about 7.5
    times).
  •  Hg release is well below levels of potential concern
    (but poor replication) for both baseline and test cases.

Figure E-5:
  •  The fly ash from the test case had similar total As
    content than that from the baseline case.
  •  As release was below the MCL for both the baseline
    and the test cases for most pH conditions.

Figure E-7:
  •  Total selenium content was below detection limits for
    Method 3052 for both fly ashes while significant sele-
    nium release (ranging from around 30 |Jg/L to around
    1000 |Jg/L) as  a function of pH was observed. This
    result is a consequence  of the dilution effects of the
    digestion method and analytical requirements.
  •  Selenium release from both  the baseline and the test
    cases was close to or exceeded the MCL (50 |Jg/L)
    for most pH conditions.

Figures E-11 andE-12:
  •  The fly ash from the test case would result in As re-
    lease slightly less than expected from the baseline case,
    with a 95% probability to be less than 358 and 473 |Jg/
    kg, respectively.
  •  The fly ash from the test case would result in Se re-
    lease less than expected from the baseline case, with a
    95% probability to be less than 4000 |Jg/kg (total con-
    tent) in both cases and a 5% possibility that the total
    content will be released.

Figure E-13:
  •  For all scenarios examined, no significant difference in
    arsenic release would be observed between the fly ash
    from the baseline case and the fly ash from the test
    case.
  •  For the 95% probability scenario, arsenic release from
    both cases would be less than the amount that would
    be released if the release concentration was at the MCL
    and the LS ratio was the resultant LS ratio for the 95%
    case (i.e., about 100 L/kg). However, arsenic release
    would be greater than the amount that would be re-
    leased if the release concentration was at the MCL
    and the LS ratio was the LS ratio of the default sce-
    nario considered (i.e., 12.5 L/kg).
  •  For the three default scenarios considered, arsenic re-
    lease would most likely be less than the amount that
    would be released if the release concentration was at
    the MCL.

Figure E-14:
  •  For scenarios  at alkaline pH, lower Se release would
    be expected for the test case compared to the baseline
    case.
  •  For the 95 % probability scenario, selenium release from
    both cases would be greater than the amount that would
    be released if the release concentration  was at the
    MCL.
  •  In conclusion, Se release will most likely be greater
    than the MCL based on solubility and cumulative re-
    lease. Without controls appears worse than with con-
    trols.
                                                                                                       145
-------
Characterization of Coal Combustion Residues
                           Appendix F
                    Salem Harbor Fly Ashes
146
-------
                                                                Characterization of Coal Combustion Residues
List of Figures
Figure                                                                                                 Page

F-l  pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	148
F-2  pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	148
F-3  Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	149
F4  Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
     the Fly Ash with Enhanced Hg Control	150
F-5  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	151
F-6  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	152
F-7  Selenium Release (top) and Spike Recoveries (bottom)  as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	153
F-8  Selenium Release (top) and Spike Recoveries (bottom)  as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	154
F-9  Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH	155
F-10 Regression Curves of Experimental Data of Selenium Solubility as a Function of pH	156
F-ll 100-Year Arsenic Release Estimates as aFunction of the Cumulative Probability forthe Scenario of Disposal
     in a Combustion Waste Landfill	157
F-12 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
     in a Combustion Waste Landfill	157
F-13 100-Year Arsenic Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI	158
F-14 100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI	159
                                                                                                            147
-------
Characterization of Coal Combustion Residues
pH Titration Curves
Baseline Fly Ash Fly Ash with ACI
14 -1/1
12
11.7
•in
IU
I 8-
Q.



4 -

a
• • •












-1 -0

- S(













j


Q





O

D

.500











A
H
, , a
5














1














12
10.3
•in
IU
I 8-
Q.
6


4
9
; 0








-


-















H












i i i i
52 -2 -1.5 -1 -0

B





" "?









^
-------
                                                         Characterization of Coal Combustion Residues
Mercury  Release  as  a Function  of pH

                 Baseline Fly Ash

            Hg total content*: 650.6+6.8 ng/g
      Fly Ash with ACI

Hg total content*: 1529.6+1.1 ng/g
MCL
1
D) 0.1
0) °-032
I
0.01
0.001
Ml
— -MDL
Total content
150
140
P\
 110
u 100
% 90
Q.
w 80
D)
1 70
60



x^_.
:O" 	
: O O



jTT_A A T!
0 °
______ ^
1


D
- —

MCL
1 -
IT
95% S °-1
5%
n nn-i _
!
E
A A
A
_O - _ -
V 001 <,


A A


95C
5%
5% ' 11.795% W'WW1 5o/n ' ' JOS 95%
2 4S/06 8 10 12 14 2 4 s/0 6 8 10 12 14
pH pH
Ml

D SR2-SHB-0001 - A
0 SR2-SHB-0001 - B
A SR2-SHB-0001 - C

— -MDL

n SR2-SHT-0001 - A
0 SR2-SHT-0001 - B
A SR2-SHT-0001 - C
as determined by digestion using method 3052.
-icn









f




° D J


:
:













D








D








rf




140
— i?n
^T 190,
5 -i -in
i 6 MU
D « 100
Q on
-^ yu
'o.
w sn
ou
"T 7n
/U
— i — i — i— en

|

:n rf.



\













n n n n c









• nn






\j\j 	
> 4 6 8 10 12 14 2 4 6 8 10 12 14
PH pH
n SR2-SHB-0001 - A
n SR2-SHT-0001 - A
Figure F-3. Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of mercury concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
                                                                                                 149
-------
Characterization of Coal Combustion Residues
Mercury Release as a Function of LS Ratio
                Baseline Fly Ash
                                                                 Fly Ash with ACI
008
U 0 OR
O) 0 04
0 02

	 ML '
— - MDL
-
-
; A I
n
O *
: i
-
_ _ .


]
>

_ _ _


o
D

_ _ _





. _ _


[
L




]
^



D 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-SHB-0001 - A
o SR3-SHB-0001 - B
A SR3-SHB-0001 - C
n DR.
u n OR
D) n 04
I U.U4
n n?
n -
Ml ,
— - MDL





— - 1





i- - -





- 71 —





. _ _



/

• — r!



i

r — '
D 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-SHT-0001 - A
o SR3-SHT-0001 - B
A SR3-SHT-0001 - C
140,
^T 1^0-
o^ I OU
o>
> -i -in
o
o>
-^ Qn
w 80
O) OU
7n
en
(












]










n
















[








]




14D
To1 -| *^n
ll
>< -i on
O)
> -i -in
o nu
£ -inn _
o>
1/5 80 -
O) ou
yn
Rn
;
:
;
;
~
;
;
;
:














D















[








]






D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
n SR3-SHB-0001 - A
n SR3-SHT-0001 - A
Figure F-4. Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
150
-------
                                                           Characterization of Coal Combustion Residues
Arsenic Release as a Function of pH
Baseline Fly Ash Fly Ash with ACI
1 nnnn As total content*; 25.9+0.04 |jg/g ^ nnnn As total content*: 26.0+0.03 |jg/g
1000
100
19.30
_i MCL10 -
— 1 -
in
< 0.1 -
0.01 -
0 001
-w^
•CT i^Xj^Wk
\ L^"<>
;
|

	 ?- - eg


1000 -
1?nn
100 -
i — i MO 1 -i r\
-----^-«, O3
	 5% ^3 ^
< 0.1 -
0.01 -
iii n nn-i

i A>

r~"





ft* 95



°-001 5% ' 11795% 	 ' 5% ' "103 95%
2 4 6 8 10 12 14 2 4 6 8 10 12 14
ML PH M, PH

— - MDL


n SR2-SHB-0001 - A ~~ " IVIUL
OSR2-SHB-0001 - B
A SR2-SHB-0001 - C


n SR2-SHT-0001 - A
o SR2-SHT-0001 - B
A SR2-SHT-0001 - C
*Total content as determined by digestion using method 3052.
-i en •|cn
•I/in

sE,
§2 nn
> nu
o
m *i nn
-------
Characterization of Coal Combustion Residues
Arsenic Release as a Function of LS
Baseline Fly Ash
co

1 r on
^ oU -
a
MCL 10 -
0 J
C
Ml
— - MDL
1^0
140

^ 1 20
flj
^ *i *i r\
O 1 lu
o
P 100
O>
-*: QO
Q.
w 80
70
RO
(


. A
: D <




J























^^H • ^^^ • ^^^ • ^^^ • ^^^ • ^^^ • ^^^ 1
1 1 1 1 1 1
) 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-SHB-0001 - A
0 SR3-SHB-0001 - B
A SR3-SHB-0001 - C

|
i
|
;
1 c
; D
1
;
|




3









D
















[








j]





D 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-SHB-0001 - A
> Ratio
250
200
=J 150
a
^ 100
50
MCL
0
	 ML
— - MDL
150 -r-
140 -

^ 1 20
> no
o
E100
o>
-^ QO
Q.
w 80
70

0
Fly Ash with ACI



_




T



A








I




k




0 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-SHT-0001 - A
0 SR3-SHT-0001 - B
A SR3-SHT-0001 - C






n
















D

















r








i





2 46 8 10 12
LS ratio [mL/g]
n SR3-SHT-0001 - A

Figure F-6. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
152
-------
                                                           Characterization of Coal Combustion Residues
Seleniun
10000
1217
1000
100
MCL
5- 10
1 	 1 1
o>
w 0.1
0.01
0.001
	 ML

— - MDL
i Release as a Function of pH
Baseline Fly Ash Fly Ash with ACI
Se total content*: 41 .9+0.06 ug/g -mnnn Se total content*; 44.0+0.04 |jg/g

w> DA


"I
-i

£U
fln a- - *i
"
!

B



141?D700 -
95% 100 -
MCL
5- 10 -
5% §. 1
 110
8
m ^ nn
9J QO
'o_
w 80
(D
W yn
cn




;n J
















n n :
n D








D
n




2 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH

n SR2-SHB-0001 - A
n SR2-SHT-0001 - A

Figure F-7. Selenium Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
Fly Ash with Enhanced Hg Control. 5th and 95th percentiles of selenium concentrations observed in typical combustion
waste landfill leachate are shown for comparison.
                                                                                                    153
-------
Characterization of Coal Combustion Residues
Selenium
5000
4000
5* 3000 -
$ 2000 -
1000 -
MCL Q
	 ML
— - MDL
1 ^n
140
vP 1 ^0
0s" v\J
O>
> -i -in
o
o> mn -
O>
^ 90
Q.
w 80
0> OU
W 70 -
fiO
0
Release as a Function of
Baseline Fly Ash



:
- ** i




n




V








[




1

LS Ratio
Fly Ash with ACI
cnnn

0 24 6 8 10 12
LS ratio [mL/g]
n SR3-SHB-0001 - A
o SR3-SHB-0001 - B
A SR3-SHB-0001 - C





n
	 [







]







D

















[








3





4000
"~T ^000
a
CD onnn
1000
MCL
U H
(
	 ML
— - MDL
1 'in
140
vP 1 ^0
0s" v\J
>* -ion
O)
> -i -in
o
Q> mn -
o>
-^ Qn
._ yu
w 80
0) OU
W 70 -
«n
\
-
t
-
-


J





^









|






D 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-SHT-0001 - A
0 SR3-SHT-0001 - B
A SR3-SHT-0001 - C

;
;
;
: C
;
;
;
;
;



]










D
























n





2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
n SR3-SHB-0001 - A

nSRS-SHT-0001 -A
Figure F-8. Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
154
-------
                                                         Characterization of Coal Combustion Residues
Arsenic Solubility
                 Baseline Fly Ash
Fly Ash with ACI
IUUUUU -
10000 -
1000 -
100 -
10 -
1 -
0.1 -
0.01 -
0.001 -



1
1



_____
&

\\j\j\j\j\j -
10000 -
1000 -
95% _j 100.
^ 10 -
5% < -i .
0.1 -
0.01 -
n nn-i
«*^

jD^-firi^^^Q^Bi

&...

95%
5%
1 ' 5% ' ' '95% 	 ' 5% ' ' '95%
2 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH
n SR2-SHB-0001-A
0 SR2-SHB-0001-B
A SR2-SHB-0001-C
Fit r\ ir\/p


n SR2-SHT-0001-A
0 SR2-SHT-0001-B
A SR2-SHT-0001-C
Fit rurvp


Material log As (|ig/L) pH range of R2 Number of
validity points
SHE 0.0000 pH5 0.0011 pH4 -0.0428 pH3 3-12.5 0.96 33
0.4859 pH2 -2.2393 PH 5.6031
SHT 0.0003 pH5 -0.0106 pH4 0.1194 pH3 3-13 0.92 33
-0.5142 pH2 0.4220 PH 3.3432


Figure F-9. Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH.
                                                                                                155
-------
Characterization of Coal Combustion Residues
Selenium  Solubility
                Baseline Fly Ash
Fly Ash with ACI
1 \J\J\J\J\J -
10000 -
1000-
U 100-
S 10-
% 1-
0.1 -
0.01 -
0 001 -
S. ..P.A^i


jfcn-ja 	 rf?
<%^._ 	


*-


2 4 5/°6 8 10 l!5%1
PH


n SR2-SHB-0001-A
o SR2-SHB-0001-B
A SR2-SHB-0001-C
Fit n ir\
-------
                                                     Characterization of Coal Combustion Residues
100-Year Arsenic Release Estimates
 Arsenic
1 UUUUU -
mono .
D)
7r> innn .
^ mn -
i_
co
CD
>s -in -
0
o
""" 1 .
OH
.1 "I
c




*

)


A Q '-'
A n
A U
A°
iD

5% 2Q 4Q50<
.t
AAAAflc
k n '-'



* 60 80 95°'

!




^c






30

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
Sh
0.8
3.6
546
7922
1749
18969
HB
0.003
0.014
2.1
30.6
6.8
73.2
Sh
2.7
6.1
968
13374
3013
26000
HT
0.01
0.02
3.7
51.4
11.6
100.0
                     Percentile
               nSHB
ASHT
Figure F-11.100-Year Arsenic Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal

in a Combustion Waste Landfill.
100-Year Selenium Release Estimates
 Selenium
IUUUUU -
10000 -
"oS
"TTI 1000 -
i_2T
^ 100 -
CO
CD
^ 10 -
6 IU
o
^~ 1 .
0-1
• T ~\
C



S


)

noHH5
n B
s
JH


5%20 40 5C
naaaHH'





% ' ' 95°/
60 80
l





'o
K






DO

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
Sh
6.6
34.5
5011
41900
17623
41900
HB
0.02
0.08
12.0
100.0
42.1
100.0
Sh
4.7
24.3
3650
44000
14240
44000
HT
0.01
0.06
8.3
100.0
32.4
100.0
                     Percentile
               nSHB
ASHT
Figure F-12. 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of

Disposal in a Combustion Waste Landfill.

                                                                                         157
-------
Characterization of Coal Combustion Residues
100-Year Arsenic Release Estimates
                                   SHE - Arsenic
       100000
        10000
               25900
    0)
    en
    CO
o
'c
0)


CO

CO
    CD
    O
         1000
          100
      10
     A)





100%



7922 |jg/kg
30.6%








19



67|jg/
7.6%



kg
360_ug/kg 241_yg/kg



1.4%





0.9%





Total content Combustion Default-pH3 Default-pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 1 1 7



                                                                            MCLLS95%=1000|jg/kg




                                                                            MCLLS125 = 125ug/kg
                                   SHT - Arsenic
      100000
       10000
0)
CO
CO


"35

o


0)


CO

CO
    CD
    O
        1000
         100
      10
               26000






100%



13;





374 ^g
51.4%



'kg
3326 ng/kg 2590 ug/kg





12.8%








10%



19




50 j^g/
7.5%



kg




Total content Combustion Default -pH 3 Default -pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 10.3
    B)


Figure F-13.100-Year Arsenic Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI. Release estimates

for percolation controlled scenario are compared to release estimate based on total content. The amount of the arsenic that

would be released if the release concentration was at the MCL is also shown for comparison (LSdefaultscenario = 12.5 L/kg and

LS95%=100L/kg).

158
-------
                                                        Characterization of Coal Combustion Residues
100-Year Selenium  Release Estimates
                                   SHE - Selenium
               41900|ig/kg   41900

51 -innnn
O)
QJ
CO
cu 1000
cu
E
'c
cu inn
CO
>s
ci in
0 IU
1

E 100% 100% 20296jig/kg _^,
















1572|ig/kg












48.4%








^8^*9
36.3%


*g-





Total content Combustion Default - pH 3 Default -pH Default -
Waste Landfill - 12.5 Natural pH
95% confidence 11.7



                                                                            -MCLLS95% = 5000ug/kg


                                                                               MCLLS125 = 625ug/kg
    A)
               44000
        SHT- Selenium

44000 na/ka                24978

S5 -innnn
=L
CO
CO
cu innn
cu
E
^
'c
cu -inn
CO
cu
>s
o in
o IU "
1
D\
E 100% 100% 56.8% 18708|ig/kg
















1337|ig/kg
•}<>/„



















42.5%








Total content Combustion Default - pH 3 Default -pH Default -
Waste Landfill - 12.5 Natural pH
95% confidence 10.3
- MOI -5000iin/kn
ivioi_Lgg5% — CNJUU |jy/i\y
- MPI RPR i in/kn
ivioLLS125 - D^O ug/Kg
Figure F-14.100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI. Release estimates
for percolation controlled scenario are compared to release estimate based on total content. The amount of the selenium
that would be released if the release concentration was at the MCL is also shown for comparison (LSdefaultscenario = 12.5 L/kg
andLS95%=100l_/kg).

                                                                                               159
-------
Characterization of Coal Combustion Residues
Comments

Figure F-3:
  •  The fly ash from the test case had similar total Hg
    content to that from the baseline case.
  •  Hg release is greater in baseline than test case, but
    both were below MCL.

Figure F-5:
  •  The fly ash from the test case had similar total As
    content than that from the baseline case.
  •  The laboratory measurements fit within the 5-95% con-
    fidence intervals of the field observations.
  •  Arsenic release  is less in baseline than test case, but
    both are about  10 times greater than the  MCL. Ar-
    senic release at pH higher than 9 is much greater for
    the test case than the baseline case.

Figure F-6:
  •  Initial landfill leachate As concentrations will likely be
    about 20-30 |Jg/L for the baseline case but at least 100
    |Jg/L for the test case.

Figure F-7:
  •  The fly ash from the test case had similar total Se con-
    tent to that from the baseline case.
  •  Se release is similar in baseline and test cases, but sig-
    nificantly above MCL for both cases. The observed
    concentrations are greater than reported in the EPA
    database but consistent with the EPRI database.

Figure F-8:
  •  Initial landfill leachate Se concentrations are expected
    to be around 200 |Jg/L for the baseline case and in-
    creasing with increasing LS ratio, but the initial con-
    centrations for the test case are expected to be around
    3000 |Jg/L and decreasing with increasing LS ratio.

Figures F-11 andF-12:
  •  The fly ash from the test case would result in As re-
    lease greater than expected from the baseline case,
    with a 95% probability to be less than 13,375 and 7,925
    |Jg/kg, respectively.
  •  The fly ash from the test case would result in Se re-
    lease less than expected from the baseline case. At
    the 95th percentile the total content of Se will be re-
    leased (41,900 and 44,000 |Jg/kg, respectively, for the
    baseline case and the test case).
  •  10-100% of the Se can be anticipated to be leached
    from the fly ash for both cases under the projected
    landfill conditions.

Figure-13:
  •  Greater As release would be expected for the test case
    compared to the baseline case, for all scenarios exam-
    ined.
  •  For all scenarios examined, Arsenic release from the
    test case fly ash would be greater than the amount that
    would be released if the release concentration was at
    the MCL.

Figure F-14:
  •  At the 95th percentile, Se release estimate exceeds total
    content for both the baseline and the test cases. This is
    not physically possible. However, this result indicates
    that there is 5% possibility that 100% of the total Se
    content would be released.
  •  For all scenarios examined, Se release would most likely
    be greater than the amount that would be  released if
    the release concentration was at the MCL.
160
-------
                Characterization of Coal Combustion Residues
     Appendix G
Facility C Fly Ashes
                                            161
-------
Characterization of Coal Combustion Residues
List of  Figures
Figure                                                                                                Page

G-l  pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	163
G-2  pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	163
G-3  Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	164
G4  Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
     the Fly Ash with Enhanced Hg Control	165
G-5  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	166
G6  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	167
G-7  Selenium Release (top) and Spike Recoveries (bottom)  as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	168
GS  Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	169
G-9  Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH	170
G-10 Regression Curves of Experimental Data of Selenium Solubility as a Function of pH	171
G-ll 100-Year Arsenic Release Estimates as aFunction of the Cumulative Probability forthe Scenario of Disposal
     in a Combustion Waste Landfill	172
G-12 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
     in a Combustion Waste Landfill	172
G-13 100-Year Arsenic Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI	173
G-14 100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI	174
162
-------
                                                       Characterization of Coal Combustion Residues
pH Titration Curves
                Baseline Fly Ash
12
11.3
1 n
a _

4 -
9 -

"^
_

~


>
\
1
D
6












,, 1





s
         -0.5     0    0.5     1     1.5
                     meq Acid/g dry
             n S R2-GAB - A    o SR2-GAB - B
             A S R2-GAB - C
                                                   I
                                                   Q.
                                                         14
       Fly Ash with ACI
2
^- d
n
3.1
Q

A
9

1
^
cP
I
I
I
I
I
I


t
3
fl
\






.,,.





y






-0.5     0    0.5     1     1.5
            meq Acid/g dry
    DSR2-GAT-A    OSR2-GAT-B
Figure G-1. pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
pH as a Function of LS  Ratio
                Baseline Fly Ash
       Fly Ash with ACI
12
m
8

t\
O
(
~
D '
-
-
-
-

1





B











i










12 -
m
5 8
Q. u
4
9
-
:
i
-
-
-


a





3











<
L




•
1


D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [ml_/g]
pSRS-GAB-A <>SR3-GAB-B
^ SR3-GAB - C
n S R3-GAT - A o SR3-GAT - B
Figure G-2. pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
                                                                                             163
-------
Characterization of Coal Combustion Residues
Mercury
10 -
MCL
1
U
ra 0.1 -
0)
1 0.014
0.01 -
0.001
	 ML

Release as a Function of pH
Baseline Fly Ash Fly Ash with ACI
Hg total content*: 15.8+0.9 ng/g Hn Hg total content*: 1 150.7+14.4 ng/g

m
:* c



i i i | i i i


D pjn ^ ^
A |0
1
A OA Wk
MCL
1
95% 1 °'1 -
D)
I 0.016
n m

CO/
iii n nni


. _ — D - -
A ri
^H • ^^ •
U U


u
- - -0° D
} 	 ?S«... 	

U ' U IS

95%

" ~5%

2 4 65% 8 1011-31295%14 ""' 2 4 65% 88'1 10 1295%14
PH -ML PH
Mni
DSR2-GAB -A 0SR2-GAB-B
ASR2-GAB -C
QS R2-G AT - A » S R2-G AT - B
*Total content as determined by digestion using method 3052.
150 'lcn
140
" 130
"I"! -ion
CD
> 110
0 MU
O
CD 100
CD
jZ qn
'a.
« 80 -
O)
1 70
RO
^






n Pn n









u UCI nri cP

1/1 n
I4U
ion
CD
b ' ' u
o
m mn
S qn
'o.
O)
X 70






D
n
q
D







h DnD
Qj
LJ








4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH
p SR2-GAB - A
DSR2-GAT- A
Figure G-3. Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of mercury concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
164
-------
                                                          Characterization of Coal Combustion Residues
Mercury Release as a Function of LS Ratio
Baseline Fly Ash Fly Ash with ACI
n 1 n •"
n 08
^ 0 06
O)
,3.
o) n r\A
o n?
n
C
ML
— - MDL
150
140
^o1 -i^n -
>s -ion
>
z 11 n
o
P mn -
O>
-^ Qn
Q.
w 80
0) °U
1 70
60
(



. i



g
& i
^.]_..



Q
0
_ _ —





. _ —







i

-—
) 246 8 10 1
LS ratio [mL/g]
nSR3-GAB-A <>SR3-GAB-B
ASR3-GAB -C
0 08
^ 0 06
O)
^
a) n r\A
0 0?
n
2 (
ML
— - MDL






^~ " L





n — -




D
- » —





. _ —








fe
-;-t"-

D 246 8 10 12
LS ratio [mL/g]
n SR3-GAT - A o SR3-GAT - B
•i^n

;
;
I
:
~
;
I
i ii








i i i






n


















C








]

140
^F -i^n
>s -ion
>
^ 11 n
o
P mn
O)
-^ Qn
Q.
w 80
0) °U ~
1 70
Rn
_
;
;
I
:
~
:
:
~















D

















C

I I I






]

I I I


D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
OSR3-GAB -A

D SR3-GAT - A

Figure G-4. Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                                  165
-------
Characterization of Coal Combustion Residues
Arsenic Release as a Function of pH
Baseline Fly Ash Fly Ash with ACI
10000 As total content*; 93.6+5.5 |jg/g -innnn As total content*; 506.3+28.7 |jg/g
1000 -
237.37
100 -
-T MCL-io ,
O)
— 1 -
t/5
< 0.1
0.01 -
n nm
I _ ntaaAVn 100°
jaJ^Jrrn^-HH Qfi% 119.67
" IOO
-T Mri m

, 	 5% 2; 1 -
; tO
i < 0.1
i 0.01
" 	 n nn-i
5% 1 1 3 95%
2 4 6 8 10 12 14
MI nu 	 M|_

— - MDL
*Total content
150 -
140
gr 130-
^ 120
O>
> 110
£ 100
2 90
Q.
w 80 -
t/5
< 70
60
	 - IVIUL
D SR2-GAB - A 0 SR2-GAB - B
A SR2-GAB - C
!» J
^dra ^i^ n._/v IS
! ^y. . .0*P - 'j-f&Q LQP
'- D 1

E
1





95%
5%
'5% "81 ' '95%
2 4 6 8 10 12 14
PH
DSR2-GAT-A <>SR2-GAT-B
as determined by digestion using method 3052.
150
-MO
" -ion
:n n °1
~ •-! - 19D
Cl n £?
CD
D "=• nn
n n o
_ LJ Q

E ° ° ^00
Cl ^ 30
(rt sn
C/)
-------
                                                      Characterization of Coal Combustion Residues
Arsenic Release as a Function of LS Ratio


                Baseline Fly Ash
                                                                 Fly Ash with ACI
tuu

-j 250
"3)
J2 -i en
100 -
50
MCL Q ,
(
	 ML
— -MDL

1 A


n


E (







J






0














5







!






^-TH OCO
i 200
J2 150
mn
en
MCL „





r
<






]
>






R
























D

D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
	 M_
n SR3-GAB - A o SR3-GAB - B — ' IVDL
A SR3-GAB - C
DSR3-GAT-A <>SR3-GAT-B
  o
  o
  0)
150
140
130

11 n
100


80 -
70 -
RD

_
;
=
=
I
= C

- n
=
i






i









D





















[










1





1*50


>* -ion
 -i -i n
o
® inn
-^ 90

Q.
< 70 .
Rn _










i i i










i i i






D



i i i



















[










]


                     46     8    10   12


                     LS ratio [mL/g]
                                                                 2    4    6     8    10    12

                                                                      LS ratio [mL/g]
                      QSR3-GAB-A
                                                                       nSR3-GAT- A
Figure G-6. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and

the Fly Ash with Enhanced Hg Control.
                                                                                           167
-------
Characterization of Coal Combustion Residues
Seleniurr
10000
2851.88
1000
^L1°°
IT 10
~D)
a 1

140
" 1 30
0s
>, 120
~2 1 1 n
o
o
ai 1 nn
2 90
'5.
y) on -
0)
co 7n
60







:PJ P








P D

I I
> 4






«n _rl _«
n LJ n n ^n
III III III II
140
" 1 30
0s
>> 120
i_
¥ 1 m
o
o
m 1 nn
^ 90
'5.
y) on
0)
co 7n
fin






a D c

i i






1 n

i i
6 8 10 12 14 24
PH
DSR2-GAB -A







3 ncti 13 D


6 8 10 12 14
PH
DSR2-GAT- A
Figure G-7. Selenium Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
Fly Ash with Enhanced Hg Control. 5th and 95th percentiles of selenium concentrations observed in typical combustion
waste landfill leachate are shown for comparison.
168
-------
                                                      Characterization of Coal Combustion Residues
Selenium Release as a Function of LS Ratio
                Baseline Fly Ash
                                                                 Fly Ash with ACI
7000
6000
— 5000

<$ 3000
2000
-i nnn
MTI n
C
	 ML
— - MDL




£

t
i






t
I






B













E







I
i


7000 -
6000 -
5000

m 3000
2000
mnn
MP.I n
=
=
=
=
i
1
I
I


]

>







rj















i







y
3 	


) 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] 	 ML LS ratio [mL/g]
— - MDL
QSR3-GAB -A *SR3-GAB-B
ASR3-GAB -C
QSR3-GAT-A oSRS-GAT-B
  o
  o
  
       140
       130
       120
       110
       100
       90
        70
        60
                     468
                     LS ratio [mL/g]
                                     10
12
QSR3-GAB
-A
        o
        o
        P
                                                   »
150
140
130
120
110
100
 90
 80
 70
 60
              468
              LS ratio [mL/g]
10   12
                                                                      QSR3-GAT-A
Figure G-8. Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                            169
-------
Characterization of Coal Combustion Residues
Arsenic Solubility
                 Baseline Fly Ash
Fly Ash with ACI
10000
1000
100
10
1
0.1
0.01
0.001
;
Ea-SL— a
19

K*****W


/


10000 -
1000 -
95% „ 10Q.
S 10-
5% w ,
< ' •
0.1 -
0.01 -
n nn-i .
^

P — nd^-*^^

X

1 5%' ' ' '95% 	 ' 5% ' ' '95%
2 4 6 8 10 12 14 2 4 6 8 10 12 V
pH pH


a SR2-GAB - A
o SR2-GAB - B
A SR2-GAB - C


B;
n SR2-GAT-A
o SR2-GAT-B
Fit curv6


Material log As (|jg/L) pH range of R2 Number of
validity points
GAB 0.0012 PH5 -0.0416 PH4 0.5236 PH3 3-12.5 0.78 33
-3.0400 pH2 8.0912 PH -5.3943
GAT -0.0008 pH5 0.0304 pH4 -0.4500 pH3 3-12.5 0.75 22
3.1407 pH2 -10.1848 PH 14.1412


Figure G-9. Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH.
170
-------
                                                         Characterization of Coal Combustion Residues
Selenium Solubility
Baseline Fly Ash
100000
10000-
1000-
u 100.
1 10-
CO 1 •
0.1 -
0.01 -
n nm


r3p&&*&flr^
^^^^.. 	


•s.


2 4 5%6 8 10 1295%1
PH
100000
10000
1000
95% :j -i oo
S 10
5% 
-------
Characterization of Coal Combustion Residues
100-Year Arsenic Release Estimates


   100000
 ^10000
 O)
 _
  CO
  CD
  ><

  6
  o
1000


 100


  10


   1


 0.1

°"^
0^
•""'
A
[
D S% 20 40 E
0"°°^'
;°" 	




0%' ' 95 °/
60 80
:
l





'o
1C

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
Gf
|jg/kg
5.8
27.5
4086
59706
13789
93600
\B
%
0.0062
0.0294
4.4
63.8
14.7
100.0
G>
|jg/kg
4.1
7.0
1142
15411
3485
48499
M
%
0.0008
0.0014
0.2
3.0
0.7
9.6
                     P ere entile
               nGAB
                          A GAT
Figure G-11.100-Year Arsenic Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal

in a Combustion Waste Landfill.
100-Year Selenium  Release Estimates
 Selenium
100000
"oS
==£ -innnn
O)
1 — ' mnn
^ IUUU
^
m 100
CD

Mg/kg
24.8
69.6
4000
4000
4000
4000
\B
%
0.6
1.7
100.0
100.0
100.0
100.0
G>
Mg/kg
31.1
142.5
21033
206300
78816
206300
VT
%
0.02
0.1
10.2
100.0
38.2
100.0
                     Percent! I e
               DGAB
                           kGAT
Figure G-12. 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of

Disposal in a Combustion Waste Landfill.

172
-------
                                                       Characterization of Coal Combustion Residues
100-Year Arsenic  Release  Estimates
                                   GAB -Arsenic
      100000
       10000
        1000
     CO

     CO
    o
    o
         100
          10
93600 ug/kg 93600 ug/kg
100% 59706 M9/kg 1000/0
:
:
!
-
-
:
1
1








Dd S%



4855 |jg/kg




5.2%










2967 ug/kg




3.2%










MCL= 1000 ug/kg
MCLLS1  = 125 ug/kg
Total content Combustion Default- pH 3 Default -pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence H 3
A)
GAT- Arsenic
506300 ug/kg
1000000
1 — ' 1 00000
O)
CD 10000
CO
"CD
o IUUU -
CO
ro -inn
i- 1 00
CO
o
0 10
•1

E 1 00%
:
-
:
=
=
=




15411 ug
3.0%








/kg
2413 ug/kg
1178 ug/kg
0.2%








0.5%



14



96 ug/kg
0.3%





Total content Combustion Default- pH 3 Default-pH Default-
Waste Landfill - 12.5 Natural pH



                                                                              MCLLS95% = 1000
                                                                              MCLLS12  = 125 ug/kg
     B)
                          95% confide nee
Figure G-13.100-Year Arsenic Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI. Release estimates
for percolation controlled scenario are compared to release estimate based on total content. The amount of the arsenic that
would be released if the release concentration was at the MCL is also shown for comparison (LSd f lt     =12.5L/kgand
LS95%=100L/kg).

                                                                                              173
-------
Characterization of Coal Combustion Residues
100-Year Selenium  Release Estimates
                                    GAB - Selenium
        10000
     O)

     ^5)
     CO
     CO
     cu
     CO

     CD
     cu
     >s
     I
     O
     O
         1000
          100
10
<4000 ug/kg 4000 ug/kg 4000 ug/kg 4000 ug/kg
; 100% 100% 34Z























i/.2ug
80-2%


/k9 100% 100%



























                                                                      MCLLS95% = 5000 ug/kg



                                                                      MCLLS125 = 625 ug/kg
     A)
               Total content   Com busHon   Default - pH 3
                           Waste Landfill -
                           95% confidence
                                            Default - pH  Default - Natural
                                               12.5          pH

                                                            11.3
                                     GAT - Selenium
     1000000
    0)100000
    CO
    JD
     D
    CO
    CD
    O
    O
       10000
        1000
         100
          10
E 206300 ug/kg 206300 ug/kg 206300 u
: 100% 100% 100%

:
-


-
=
1
:
:

























7191. 3 ug/kg
3.5%























3/kg
41101.1 ug/kg







19.9%













                                                                       MCLLS95% = 5000 ug/kg

                                                                       MCLLS125 = 625 ug/kg
               Total content
    B)
                  Combustion
                 Waste Landfill -
                 95% confidence
Default - pH 3
Default-pH
   12.5
 Default-
Natural pH
   8.1
Figure G-14.100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with ACI. Release estimates
for percolation controlled scenario are compared to release estimate based on total content. The amount of the selenium
that would be released if the release concentration was at the MCL is also shown for comparison (LSdefaultscenario = 12.5 L/kg
andl_S95%=100L/kg).

174
-------
                                                            Characterization of Coal Combustion Residues
Comments

Figure G-1:
  •  All extract Hg concentrations are well below levels of
    potential concern.

Figure G-5:
  •  Arsenics extract concentrations for the baseline case
    peak between pH 7 and 9, with maximum concentra-
    tions significantly greater than the range reported for
    field landfill leachates in the EPA database but consis-
    tent with the range of concentrations for field landfill
    leachates reported in the EPRI database.
  •  Arsenic extract concentrations for the test case indi-
    cate somewhat  lower  concentrations than for the
    baseline  case over the  range of pH examined, even
    though the test case as  around 5 times as much total
    As as  the baseline case. These results also suggest
    different chemistry controlling the aqueous-solid equi-
    librium for the two cases.

Figure G-7:
  •  Se extract concentrations as a function of pH exhibit
    similar behavior for the baseline and test cases, even
   though the test case has greater than 50 times the
   amount of total As than the baseline case.

Figure G-8:
  • Initial leachate concentrations for Se are likely to be
   ca. 1000-4000 |Jg/L (at LS=2), which is much greater
   than reported in the EPA database but consistent with
   values  reported in  the EPRI database for landfill
   leachates.

Figures G-11and G-12:
  • A much greater percentage and quantity of As can be
   anticipated to be released from the baseline case than
   for the test case under the scenarios examined.

Figure G13:
  • Arsenic release from the base  case warrants further
   examination.

Figure G-14:
  • Se release from the test case warrants further exami-
   nation.
                                                                                                     175
-------
Characterization of Coal Combustion Residues
                             Appendix H
                         St. Clair Fly Ashes
176
-------
                                                                Characterization of Coal Combustion Residues
List of Figures
Figure                                                                                                Page

H-l  pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	178
H-2  pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	178
H-3  Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	179
H-4  Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
     the Fly Ash with Enhanced Hg Control	180
H-5  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	181
H-6  Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	182
H-7  Selenium Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	183
H-8  Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	184
H-9  Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH	185
H-10 Regression Curves of Experimental Data of Selenium Solubility as a Function of pH	186
H-ll 100-Year Arsenic Release Estimates as a Function of the Cumulative Probability forthe Scenario of Disposal
     in a Combustion Waste Landfill	187
H-12 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
     in a Combustion Waste Landfill	187
H-13 100-Year Arsenic Release Estimates from A) Baseline Fly AshandB) Fly Ash withACI	188
H-14 100-year Selenium Release Estimates from A) Baseline Fly AshandB) Fly Ash withACI	189
                                                                                                            177
-------
Characterization of Coal Combustion Residues
pH  Titration Curves
                Baseline Fly Ash
12.1
1 9
1 n -
8

A
9
• -PI




I I I I
s
hi
6


I I I I


£


I I I I


6


i i i i





i i i i



:
*n
D
i i i i










i i i i i i i i
          -101234567
                     meq Acid/g dry
              n SR2-JAB - A    o SR2-JAB - B
     Fly Ash with B-PAC
                                                         14
                                                        12.2
                                                         12
                                                         10

                                                          8

                                                          6
-101234567
           meq Acid/g dry
    D SR2-JAT- A     o SR2-JAT - B
Figure H-1. pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
pH as a Function of LS Ratio
                Baseline Fly Ash
     Fly Ash with B-PAC
12
m
8
fi -
/\
9
1 5 i
:
-
:
-
-

















E





]





it -
19
m
I 8
Q. °
4
9
1 8 i
1
-
-
-
-
a

















E





]





          0     24     6    8    10    12
                     LS ratio [mL/g]
 0     24     6    8    10    12
           LS ratio [mL/g]
             pSRS-JAB-A     0SR3-JAB-B
   OSR3-JAT-A     <>SR3-JAT-B
Figure H-2. pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
178
-------
                                                           Characterization of Coal Combustion Residues
Mercury
10 -
MCL
1 -
ra 0.1
o, 0.030
I
0.01 -
0.001 -
	 ML
_ Mm

Release as a Function of pH
Baseline Fly Ash Fly Ash with B-PAC
Hg total content*: 1 10.9+5.8 ng/g ^ n Hg total content*: 1 1 63.0+8.9 ng/g



U. LJ
,9
_ . __ .




06 * 
^5o/

' 5% 95% '199
2 4 6 8 10 1212'214
PH
D SR2-JAT - A « SR2-JAT - B
*Total content as determined by digestion using method 3052.
1*50 'l^n

" 130

O>
o
m -inn
£ IUU
* 90
Q.
O)
x 7n
en






[







]
n








a a
an n n



r^' 130 -
t^ -ion
O)
5. -i-i n
0 MU
O
Emn
_ m
a ri on
D in «?n
O)
I 7D
, , , en






:








1° C







n

c








J D 1








ID








cd

4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH
QSR2-JAB-A
QSR2-JAT-A
Figure H-3. Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of mercury concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
                                                                                                    179
-------
Characterization of Coal Combustion Residues
Mercury Release as a Function of LS Ratio
Baseline Fly Ash Fly Ash with B-PAC
01 n 1
0 08
— ' 0 06
O)
a) n 04
I U.U4
n no

(
	 ML
	 MDL
150
140

2^ ion
o>
> 11 n
o
E-inn
O>
Q.
w sn
0) °U ~
1 70
60
(
;
- D

0 r



3

_ _





-^-





. _ ^_


<

[
_ ^_


>

3
_ ^^





D 246 8 10 12
LS ratio [mL/g]
nSR3-JAB-A oSR3-JAB-B

i
;
;
I
I

~
: O [

:
~






:








D
















r








]


I I I

D 2 4 6 8 10 12
LS ratio [mL/g]
DSR3-JAB-A
0 08
— "" n OR -
O)
c5 n n4
I U.U4
n OP
n '



D




771.....





^
_ _ —





. _ —



i




]

__^__

0 246 8 10 12
ML LS ratio [mL/g]
MDL

150
140
^ 130 -
^ 120 -
o>
o 110"
E 100 -
o>
^ 90 -
E 80
1 70
60 -
nSR3-JAT-A oSR3-JAT-B

|
;
;
I
i
~
: 0 I
1
1 II






D








n


















i








:


0 2 4 6 8 10 12
LS ratio [mL/g]
DSR3-JAT- A
Figure H-4. Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
180
-------
                                                          Characterization of Coal Combustion Residues
Arsenic Release as  a Function  of pH

                 Baseline Fly Ash

     „-.„-.  As total content*: 43.4+2.6 |jg/g
  w
  <
1000

100 -
/ICL-in _

1 -
0.92
0.1 -
0.01 -
o 001 -

*
=
i
i
1
=
= - -x&^c
z
=
=
-
5C
2 4






n n



/o 95%
6 8 10 1




9





12.1
2 1

95%



5%





4
                                                          10000
                                                       ^ MCL-H
                                                       ^)
                                                                     Fly Ash with B-PAC

                                                                 As total content*; 40.8+1 .1 |jg/g
000
100 -
I 10

0.54
0.1 -
3.01 -

!
1
=
r—ds^
=
i
5C
2 4



•— •— • —r *- *&m f *—
j Q i?Eiin QD


/o 95%
6 8 10 1


ti
>- -


21Z21
95%


o/u


4
	 ML

— - MUL
2 4 6 8 10 12 1'
nH
Pn
d SR2-JA B - A o SR2-JAB - B
1
	 ML

— - MUL
2 4 6 8 10 12 1
nH
Pn
QSR2-JAT-A <>SR2-JAT-B
*Total content as determined by digestion using method 3052.


"^ i9n
o>
> 11 n
0 MU
m mn
5 qn
'5.
to ftn

60














n
) n



i i i



D







n
" a n








Dl








:P





? 46 8 10 12 14
PH
D SR2-JAB - A
i4n

^! 190
a>
> 11 n
0 MU
" 100

'5.
CO on

60 -
f



-_ C




I II



U L




I I I



n




i i i



=m c




i i i



D




i i



_n
cP



I I I
> 46 8 10 12 14
PH
OSR2-JAT- A
Figure H-5. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of arsenic concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
                                                                                                   181
-------
Characterization of Coal Combustion Residues
Arsenic F
350 -
300
250
| 200-
7 150
<
1 nn
50
MCL n
Release as a Function of LS Ratio
Baseline Fly Ash Fly Ash with B-PAC
•3 en
: fi
;
i
;

;
1 i






a





























0 2 4 6 8 10 1
LS ratio[ml_/g]
	 MDL
150
140
^ 130
>> -ion
0>
> 11 n
0
£ 100
0>
-* QO
•Q. yU
w 80 -
%
70 -
fin
c





300
250
51 200 -
0)
^.
,n 150
<
100
50 -
MCL
2
	 ML
— -MDL
p SR3-JAB - A o SR3-JAB - B
:
;
- D
|
;
:






i,,y,,,




























0 2 4 6 8 10 12
LS ratio [mL/g]
0 SR3-JAT- A o SR3-JAT- B
i^n






: n I








i








n















r








]




140
^ 130
>> -ion
0)
> 11 n
0
£ 100
0)
-* QO
•Q. yU
w 80 -
w ou
"^ 70-
Rn






D [








1








D















:




124681012 024681
LS ratio [mL/g] LS ratio [mL/g]
QSR3-JAB- A





]




0 12
QSR3^JAT-A
Figure H-6. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
182
-------
                                                           Characterization of Coal Combustion Residues
Seleniui
10000
1000
MCL 50.80
IT 10
— 1
O>
W 0.1
0.01
0.001
MI
	 MDL
*Total conten
150
140
^ 130
^ 120
0)
> 110
g> 100
5 90
! 8°
w 70
60
n Release as a Function
Baseline Fly Ash
Se total content*: 10.7+0.1 |jg/g


!
T-NfcC
I
1
5%
2 4

DOD {] D)ISI o



a
i
i
I
i
i
i
i
i
Of
95%
5%

95% 12.1
6 8 10 12 14
PH
Q SR2-JAB - A o SR2^JAB - B
t as determined by digestion using













: C

: 1








1 t
3



> 4





D
D








nDD





PH
10000 -
1000
M01 58.85
ZT 10
" 0.1
0.01 -
0.001
f\/|

— -MDL
Fly Ash with B-PAC
Se total content*: 12.6+0.9 ug/g

-:  S R2- JAT - B

method 3052.
-icn





D[



6 8 10
PH





f




12 14
08^B-A

™

^_
> 1 m
o ' IU
o
5 90
'5_

«n





; c
: 	 Cl 	







^








:,-p c








D





















4 6 8 10 12 14
PH
0 SR2^JAT - A
Figure H-7. Selenium Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
Fly Ash with Enhanced Hg Control. 5th and 95th percentiles of selenium concentrations observed in typical combustion
waste landfill leachate are shown for comparison.
                                                                                                   183
-------
Characterization of Coal Combustion Residues
Selenium Release as a Function of LS Ratio
Baseline Fly Ash Fly Ash with B-PAC
Rnn °nn
?nn
finn
1 — • 'inn
5 4nn
co 300
onn
mn
MCL
n
c
	 ML
— -MDL
1 ^n
140
^F -i-sn
?* i?n
O>
> nn
o
£ mn
O>
-* 90
Q.
w 80
0> OU
W 70 -
60

D



C








\





























r








. ^^_,^, .
?nn
ROO
•~r 'inn
_l OUU
5 400 -
to 300
onn
mn
MCL
n


' &


I








[










n

















6








, J_,^ .

2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] 	 ML LS ratio [mL/g]
— -MDL
n SR3-JAB - A o SR3-JAB - B
n S R3- J AT - A o S R3- J AT - B
i en

1


1
: D
[
1







D








D
















i








\



140
^p1 -ion
>< -ion
O)
> 1 m
o
P mn
O)
-^ Qn
Q.
w an
0. BU "
w 7n
«n
~
;
~
;
~
;
n [

I






]









D





























D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
n SR3-JAB - A
n SR3-JAT - A
Figure H-8. Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
184
-------
                                                            Characterization of Coal Combustion Residues
Arsenic Solubility
Baseline Fly Ash Fly Ash with B-PAC
mnnnn -innnnn
10000-
1000-
U 100-
1 10-
0.1 -
0.01 -
0 001

n

10000 -
1000-
95% :j 100-
1 10-
5% CO ,
^L I "
0.1 -
0.01 -
n nn-i





0 ,
[___^uim>j IMH^_/


k
95%
5%
0.001 -> , go/- i • i950/0 u.uu, , 50/, , , ,g5%
2 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH


n SR2-JAB - A
0 SR2-JAB-B
Fit n \r\&*


n SR2-JAT - A
0 SR2-JAT-B
Fit n \r\fp


Material log As (ug/L) pH range of R2 Number of
validity points
JAB 0.0010 pH5 -0.0324 PH4 0.4137 PH3 3-12.5 0.62 22
-2.4547 pH2 6.71 05 PH -6.6563
JAT 0.0025 pH5 -0.0973 pH4 1.4698 pH3 4-12.5 0.81 22
-1 0.6792 pH2 37.241 9 PH -50.3673



Figure H-9. Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH.
                                                                                                      185
-------
Characterization of Coal Combustion Residues
Selenium  Solubility
                Baseline Fly Ash
Fly Ash with B-PAC
IVJVJVJVJVJ -
10000 -
1000-
•-j 100 -
1 10-
o> ,,
w 1 -
0.1 -
0.01 -
n nn-i .
I
!

"^^CE^. n







liiy^*iJ3(>oc^--:








V






95%

5%


i \j\j\j\j\j -
10000 -
1000-
U 100-
1 10-
w 1 -
0.1 -
0.01 -
n nm -
•
;
^0*
w— — ^





>^&&tor
-------
                                                     Characterization of Coal Combustion Residues
100-Year Arsenic  Release Estimates
 Arsenic
    10000 -B-
  o5  1000
 ^   100
  CO
  (U
10
      0.1



1
n°D°i
a A
n AA
D A
• n . f . . . , .
D 5% 20 40 £

[
n -
n D A A
J4A*AA~
. A

)0%' ' 95°.
60 80

n
L


/O
1(





DO

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
JAB
|jg/kg
0.01
0.05
8
110
25
836
%
0.00003
0.0001
0.02
0.3
0.1
1.9
JAT
|jg/kg
0.01
0.01
2
29
1
346
%
0.00001
0.00003
0.01
0.1
0.02
0.8
                     Percentile
                D JAB
                         A JAT
Figure H-11.100-Year Arsenic Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
in a Combustion Waste Landfill.
100-Year Selenium  Release Estimates
 Selenium
IUUUUU -
mono -
D)
_*:
"a) 1000 -
5 100 -
i_
CO
£ 10-
o
o
^~ 1 -





I



ft'8
a1
i

r
B*11
1H"



g
1





Wn ' ' c;O% ' ' 95% '
0 0/020 40bu/060 80 100

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
J/
^g/kg
1.8
4.5
748
10028
2270
10700
\B
%
0.02
0.04
7.0
93.7
21.2
100.0
J/
|jg/kg
1.9
4.3
710
9602
2172
12600
\T
%
0.01
0.03
5.6
76.2
17.2
100.0
                     Percentile
               D JAB
                        A JAT
Figure H-12. 100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of
Disposal in a Combustion Waste Landfill.
                                                                                          187
-------
 Characterization of Coal Combustion Residues
100-Year Arsenic  Release Estimates
                                    JAB - Arsenic
100000 43400 ug/kg
o) "innnn
O)
=L
CO
co -innn
"35
o
co 1UU
CO
CO
0)
CD -in
0 1U "
1 -
A \




Tol
100%

1
al cont






110|jg/kg
0.3%




noug/kg
0.3%
18 ug/kg
0.04%








12 |jg/kg
0.03%




—
ent Combustion Default -pH 3 Default -pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 12. 1
                                                                                MCLLS95% = 1000ug/kg


                                                                              -MCLLS125 = 125ug/kg
                                    JAT - Arsenic
100000 4080° Mg/kg
-i nnnn
D)
-3 -i nnn
-------
                                                         Characterization of Coal Combustion Residues
100-Year Selenium Release  Estimates
                                    JAB - Selenium
'D)
v> -i nnnn
ij 1 UUUU
0)
^=L
0)
(/)
S -i nnn
£
E
^
S 1 nn
0)
en
s_
TO
0)
=r 10
O
-1
A)

: 10700 ug/kg 10028 |jg/kg
100% 93.7% 607° u9/kg














56.7%





727 ug/kg 635 ug/kg
6.8% 5.9%













Total content Combustion Default-pH3 Default-pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 12.1



                                                                                MCL.   = 5000 ug/kg
                                                                                MCLLS1  = 625 ug/kg
                                    JAT- Selenium
       100000
    O)
        10000
    8>    1000
    E
    2
    TO
    0)
    B)
          100
    S     10
    o
I
: 12600 ug/kg 9602 ug/kg
100% 76.2% 3243 ug/kg



-
:














5§T^%


778 |jg/kg 736 |jg/kg
6 2% 5. 8%













               Total content   Combustion    Default-pH 3   Default - pH
                           Waste Landfill -                  12.5
                           95% confidence
 Default-
Natural pH
   12.2
                                                                                MCL     = 5000 ug/kg
                                                                                MCLLS125 = 625
Figure H-14. 100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with B-PAC. Release
estimates for percolation controlled scenario are compared to release estimate based on total content. The amount of the
selenium that would be released if the release concentration was at the MCL is also shown for comparison (LSdefaultscenario =
12.5 L/kg  and LS95% = 100 L/kg).
                                                                                                189
-------
Characterization of Coal Combustion Residues
Comments

Figure H-3:
  •  All extract concentrations for Hg are well below lev-
    els of potential concern.
  •  Scatter in the extract concentrations for the cased with
    enhanced Hg control most likely results from the ma-
    terial heterogeneity associated with addition of particu-
    late activated carbon.

Figure H-5:
  •  All extract concentrations for As are well below levels
    of potential concern.

Figure H-6:
  •  Initial As leachate concentrations from landfills are ex-
    pected to be substantially greater (i.e., equal to 50 |Jg/
    L at LS=2) than indicated by SR002 (LS=10) because
    of other ionic species at higher concentrations present
    at low LS ratio typical of landfill scenarios. These an-
    ticipated concentrations are consistent with landfill
    leachate  concentrations reported in the EPRI database.

Figure H-7:
  •  Extract concentrations of selenium are greater than
    the MCL but within the range reported in the EPA and
    EPRI databases.

Figure H-8:
  •  Initial Se leachate concentrations from landfills are ex-
    pected to be substantially greater (i.e., more than 200-
    300 :g/L at LS=2) than indicated by SR002 (LS=10)
    because  of other ionic species  at higher concentra-
    tions  present at low LS ratio typical of landfill sce-
    narios. These anticipated concentrations are consis-
    tent with landfill leachate concentrations reported in
    the EPRI database.

Figures H-11 andH-12:
  •  A much greater percentage and quantity of As can be
    anticipated to be released from the baseline case than
    for the test case under the scenarios examined.

Figure H-13:
  •  For the three default scenarios considered and the 95 %
    probability scenario, arsenic release would most likely
    be less than the amount that would be released  if the
    release concentration was at the MCL.

Figure H-14:
  •  For the 95 % probability scenario, selenium release from
    baseline and test cases would be greater  than  the
    amount that would be released if the release concen-
    tration was at the MCL.
  •  For the default scenario corresponding to disposal in a
    monofill (leachate pH controlled by the material being
    disposed) and the default scenario corresponding to
    the "extreme" pH of 12.5, no significant difference in
    selenium release between the baseline and test  cases
    would be expected.
  •  For the default scenario corresponding to the "extreme"
    pH of 3, selenium release is expected to be greater for
    the baseline  case than the test case. In both cases,
    selenium release would be at or greater than the amount
    that would be released if the release concentration was
    at the MCL.
  •  Se release from the baseline and test cases  warrants
    further examination.
190
-------
                Characterization of Coal Combustion Residues
     Appendix I
Facility L Fly Ashes
                                            191
-------
Characterization of Coal Combustion Residues
List of Figures
Figure                                                                                                 Page

I-l   pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	193
1-2   pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control	193
1-3   Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	194
14   Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
     the Fly Ash with Enhanced Hg Control	195
1-5   Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	1%
1-6   Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio forthe Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	197
1-7   Selenium Release (top) and Spike Recoveries (bottom)  as a Function of pH for the Baseline Fly Ash and the
     Fly Ash with Enhanced Hg Control	198
1-8   Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash
     and the Fly Ash with Enhanced Hg Control	199
1-9   Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH	200
1-10  Regression Curves of Experimental Data of Selenium Solubility as a Function of pH	201
1-11  100-Year Arsenic Release Estimates as a Function of the Cumulative Probability forthe Scenario of Disposal
     in a Combustion Waste Landfill	202
1-12  100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
     in a Combustion Waste Landfill	202
1-13  100-Year Arsenic Release Estimates from A) Baseline Fly AshandB) Fly Ash withACI	203
1-14  100-year Selenium Release Estimates from A) Baseline Fly AshandB) Fly Ash withACI	204
192
-------
                                                       Characterization of Coal Combustion Residues
pH Titration Curves
Baseline Fly Ash
14
19
m
-i- 8
i °
Q.
6
5.8
4
o
-0

-


_


^
o
D
*
	 L




;
\
f





1 1 i i
.5 -0.25 0 0.25 0
meq Acid/g dry

Fly Ash with Enhanced Hg Control
14
19
m

Q.
6
5.0-
4 _
o
5 -0
n SR2-LAB - A o SR2-LAB - B

-
-
:
-
-
-
_

n
*
fi
c
<

?





£"
|Q0






.5 -0.25 0 0.25 0
meq Acid/g dry
5
nSR2-LAT-A oSR2-LAT-B
Figure 1-1. pH Titration Curves for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
pH as a  Function of LS Ratio
                Baseline Fly Ash
19
m -
in o
Q. °
6
4
o



- D (





J





9











<
i




>
i

           0     2     4    6     8    10   12
                      LS ratio [mL/g]
             n SR3-LAB - A     o SR3-LAB - B
Fly Ash with Enhanced Hg Control
                                                          14
                                                          12
                                                          10
 0    2     4     6     8    10    12
            LS ratio [mL/g]
    n SR3-LAT - A     o SR3-LAT - B
Figure I-2. pH as a Function of LS Ratio for the Baseline Fly Ash and the Fly Ash with Enhanced Hg Control.
                                                                                              193
-------
Characterization of Coal Combustion Residues
Mercury
10
MCL
1
IT
0> 0.1 -
D)
I 0.010
0.01
0.001
	 ML
— -MDL

Release as a Function of pH
Baseline Fly Ash Fly Ash with Enhanced Hg Control
Hg total content*: 1 3+0.2 ng/g dn Hg total content*: 37.7+1 .3 ng/g
|
-
D
- D C
n
D D '
^^ • ^^m m 1
,? V V V '

D
O
D n

<«> V w
MCL
1
D) 0.1
c>F>% .^
D)
n n-i
> 0.009
i i . n nn-i
5-°5% 35%
2 4 68 10 12 14
PH
	 ML
— -MDL
n SR2-LAB - A o SR2-LAB - B


; D
13 0 P
• - - - -i
^^H • ^^^ '•
' i? V ^JW

D
D
1 Qn


95%
i
~~ RO/_
C> O O ^> < ^ *" ' "
	 r, , ,
2 4 ' 6 ° 8 10 12 °14
PH
n SR2-LAT - A o SR2-LAT - B
*Total content as determined by digestion using method 3052.
-i en -I en
-i/in

"ZT ion
^ \AJ
 110
0 MU
o
m mo
sj iuu
$ qn
'o_
(/) Qf)
O)
X yn
fin
^





n n
: n







DD
n








D _ °
n
n



„
"
^ 120 -
5 -i -in
0 ' IU
V 100
r - 90
'Q-
M RO
D)
I 70
i en





; nn r
3D







n D








n
D







D








D



> 4 6 8 10 12 14 2 4 6 8 10 12 14
pH pH
n SR2-LAB - A
n SR2-LAT - A
Figure I-3. Mercury Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of mercury concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
194
-------
                                                      Characterization of Coal Combustion Residues
Mercury Release  as a Function of LS Ratio
                Baseline Fly Ash
n ns
u n OR
D)
I 0.04 -
n n?
n

- n
o

<




'



D
A









[
<




D
> 	
0 2 4 6 8 10 12
Ml
— -MDL
-icn
140

0s* ' *-'**
>< -ion
O>
> 110-
o
£ mn
O)
-^ Qn
._ 90
w 80 -
£ 70
en
(
LS ratio [mL/g]
n SR3-LAB - A o SR3-LAB - B

~
:
:
~
:
I
: [
n
;
;






i








D
















[












D 2 4 6 8 10 12
LS ratio [mL/g]
n SR3-LAB - A
        Fly Ash with Enhanced Hg Control
                                                   D)
      0.1

     0.08

     0.06

£   0.04

     0.02

       0

	ML
— -MDL
                                                             ~°  I

                                                           0     24    6    8    10    12
                                                                      LS ratio [mL/g]

                                                              n SR3-LAT - A    o SR3-LAT - B
                                                        150
                                                        140
                                                   o>
                                                   o
                                                   o
                                                   e
                                                   O>
      120
      110
      100
      90
      80

      60
                    468
                    LS ratio [mL/g]

                     nSRS-LAT-A
                                                                                      10   12
Figure 1-4. Mercury Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                            195
-------
Characterization of Coal Combustion Residues
Arsenic Release as a Function of pH
Baseline Fly Ash Fly Ash with Enhanced Hg Control
.mr,™ As total content*: 20+1 . 1 |jg/g  110
0
» 100 -
$ gn
'5.
CO
< 70
fin
/


|

n n D n
P LJ 1-1 |JLJ Q








a n
D


140

" 190
(D
~> 110
6 ' IU
m 1QO
*• 90
'5.
w
<£ 70
, , , en
~
|
|
~
:
3 n
I
I
|






a c



I 4 6 8 10 12 14 24
PH
n SR2-LAB - A





n°








D
D







D








a



6 8 10 12 14
PH
n SR2-LAT - A
Figure I-5. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of pH for the Baseline Fly Ash and the Fly
Ash with Enhanced Hg Control. 5th and 95th percentiles of arsenic concentrations observed in typical combustion waste
landfill leachate are shown for comparison.
196
-------
                                                      Characterization of Coal Combustion Residues
Arsenic Release as  a Function  of LS Ratio
                Baseline Fly Ash
       50
                                                           Fly Ash with Enhanced Hg Control
       40
  O)
   MCL 10
                                     1
                                                        50
                                                        40
                                                   O)
                                                    MCL 10
                                                                fi
          0    2     4    6     8    10   12
                     LS ratio [mL/g]
      MDL
                                                           0     2    4     6     8    10   12
                                                     • - - ML             LS ratio [mL/g]
                                                     -MDL
1cn
140 -

>< -ion
o>
o
£ 100
o>
£
70
fin
n SR3-LAB - A o SR3-LAB - B

nSRS-LAT-A oSR3-LAT-B
•icn

;

:
I
: n C
:

;





1








D






























140 -
•xO -I ^O
>< -ion
o>
> no
o
£ 100
o>
w 80
w ou
70
fin
:
;
I
;

- U C
:
;
;





1







n


















[








1



                2    4     6    8    10   12
                     LS ratio [mL/g]
                                                                       4     6    8    10    12
                                                                       LS ratio [mL/g]
                      n SR3-LAB - A
                                                                       nSR3-LAT- A
Figure 1-6. Arsenic Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                            197
-------
Characterization of Coal Combustion Residues
Selenium Release as a Function  of pH

                 Baseline Fly Ash

            Se total content*: 4.1+0.1 |jg/g
Fly Ash with Enhanced Hg Control

 Se total content*: 4.3+0.2 |jg/g
   110 -
0
o
 nn
0 MU
o
-------
                                                      Characterization of Coal Combustion Residues
Selenium Release as a Function of LS Ratio
                Baseline Fly Ash
                                                          Fly Ash with Enhanced Hg Control


O)
^.
o> 20
m

:
1
- ft
- (
-



J




















J


O)
^.
o> 20
m




r
?



]
>




9









(




>
          0    2     4    6     8    10    12
 	ML             LS ratio [mL/g]
 — -MDL
                                                           0    2     4     6    8    10    12
                                                       ML             LS ratio [mL/g]
                                                       MDL

n SR3-LAB - A o SR3-LAB - B

n SR3-LAT - A o SR3-LAT - B
1 ^n •< cr>
140
^F 1^0
>< -ion
O)
> 1 in
o
E 100
o>
-^ QO
Q.
W 70 -
60
(
I

I
I
;
:
; n [
~
~
~





i








D

















[








]



1 AD
^p1 -ion
>< -ion
o>
5 1 1 n
o
Emn
o>
.*: on
w 80
a. °u "
w 7n
cn

-

-

: n E

|






i








u

















[








]



D 2 4 6 8 10 12 0 2 4 6 8 10 12
LS ratio [mL/g] LS ratio [mL/g]
n SR3-LAB - A

nSR3-LAT-A
Figure I-8. Selenium Release (top) and Spike Recoveries (bottom) as a Function of LS Ratio for the Baseline Fly Ash and
the Fly Ash with Enhanced Hg Control.
                                                                                           199
-------
Characterization of Coal Combustion Residues
Arsenic Solubility
Baseline Fly Ash Fly Ash with Enhanced Hg Control
100000 -innnnn
10000-
1000-
U 100-
1 10-
< 1 -
0.1 -
0.01 -
Onni
I
^~QO

cKrO-0 — ^
..°S^_ 	
o

X


10000 -
1000-
95% ^ 100 -
1 10-
5% w ,
< I -
0.1 -
0.01 -
n nn-i
|
i
|T3 <> ^
\ Vr^
!
rJ*^**^
y
D

S


95%
5%
2 4 6 8 10 12 14 2 4 6 8 10 12 °14
pH pH


n SR2-LAB - A
o SR2-LAB - B
Fit curve



n SR2-LAT - A
0 SR2-LAT - B
Fit n ir\jp


Material log As (ug/L) pH range of R2 Number of
validity points
LAB 0.0004 pH5 -0.0120 PH4 0.0966 PH3 3-12.5 0.92 22
0.0151 pH2 -2.4406 PH 7.2162
LAT 0.0000 pH5 0.0040 pH4 -0.1157 pH3 3-12.5 0.92 22
1.2681 pH2 -5.5708 PH 9.6784



Figure I-9. Regression Curves of Experimental Data of Arsenic Solubility as a Function of pH.
200
-------
                                                          Characterization of Coal Combustion Residues
Selenium Solubility
Baseline Fly Ash Fly Ash with Enhanced Hg Control
lOnnnn -mnnnn
10000 -
1000-
U 100-
1 10-
<° -.
w 1 -
0.1 -
0.01 -
Onm

!
|
>°^^w~-*

10000 -
1000-
95% ^ 100 -
D)
a= 10-
5% 5 1 -
0.1 -
0.01 -
n nm .

"^H


^^-^o—^

/
/

5%
2 4 °6 8 10 12 °14 2 4 °6 8 10 12 14
pH pH
n SR2-LAB - A
0 SR2-LAB-B
Fit curvs


n SR2-LAT - A
0 SR2-LAT - B
Fit r*i ir\*p


Material log Se (ug/L) pH range of R2 Number of
validity points
LAB 0.0007 pH5 -0.0239 pH4 0.3133 pH3 3-12.5 0.94 22
-1.8284pH2 4.4667 PH -1.9621
LAT 0.0007 pH5 -0.0230 PH4 0.2978 PH3 3-12.5 0.81 22
-1.7038pH2 4.0275 PH -1.4335



Figure 1-10. Regression Curves of Experimental Data of Selenium Solubility as a Function of pH.
                                                                                                  201
-------
Characterization of Coal Combustion Residues
100-Year Arsenic  Release Estimates
IUUUUU -
10000 -
0)
"5) moo -
IE 100 -
CD
* 10-
6 1°
o
^~ 1 .




i



***
B***
\


BBBH
H"*"*




1






O/i | • • • | • • • | • • • i • • • i • • • i
5% 50% 95%
0 20 40 60 80 100
Percentile

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
LAB
|jg/kg
0.83
13.36
2049
20000
10541
20000
%
0.00413
0.0668
10.25
100.0
52.7
100.0
LAT
|jg/kg
1.50
17.38
2471
18700
10090
18700
%
0.00800
0.09296
13.22
100.0
53.96
100.0
                nLAB
 A LAT
Figure 1-11.100-Year Arsenic Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
in a Combustion Waste Landfill.
100-Year Selenium  Release Estimates
 Selenium
IUUUUU -
10000 -
D)
"3) moo -
S mn -
&_
CD
* 10-
o
o
^~ -1 .
OH
.1 H
(

|
!
:
|

D



D
n ^
i*
11
i

5%20 405C

B1
I-""""1



)% ' ' 95°/
1/0 60 80

\




\(






DO

Mt min
Mt - 5%
Mt - 50%
Mt - 95%
Mean Mt
Mt max
\J
|jg/kg
0.3
1.0
143
2113
484
4100
\B
%
0.01
0.02
3.5
51.5
11.8
100.0
\J
|jg/kg
0.3
1.0
141
2041
466
4300
M
%
0.007
0.02
3.3
47.5
10.8
100.0
                     Percentile
               nLAB
A LAT
Figure 1-12.100-Year Selenium Release Estimates as a Function of the Cumulative Probability for the Scenario of Disposal
in a Combustion Waste Landfill.

202
-------
                                                         Characterization of Coal Combustion Residues
100-Year Arsenic  Release Estimates
                                    LAB - Arsenic
       100000
    O)
    ^
    "DJ
    O
    'c
    0)

    £
    TO
    L_
    CO
    0)
    >,

    CD
    O
10000
         1000
  100
   10
! 20000 |jg/kg 20000 |jg/kg 20000 |jg/kg
: 100% 100% 100%




\












729.3 |jg/kg
3.7%













324.3_yg/kg_



1.6%








-MCLLS95% = 1000ug/kg



   MCLLS125 = 125ug/kg
     A)
        Total content    Combustion   Default-pH 3    Default-pH
                    Waste Landfill -                   12.5
                    95% confidence
                                                                     Default-
                                                                    Natural pH

                                                                      5.8
                                    LAT - Arsenic

O)
v -I nnnn
D)
^_
0)
CO
m mnn

0>
O

Si: 1UU -
CO
CO
0)
S -in
1
D \
M87




:
oo |jg/kg 18700 |jg/kg 18700 |jg/kg
100% 100% 100%












468 ug/kg



2.5%










1.7%








Total content Combustion Default -pH 3 Default -pH Default-
Waste Landfill - 12.5 Natural pH
95% confidence 5 Q
Figure 1-13.100-Year Arsenic Release Estimates from A) Baseline Fly Ash and B) Fly Ash with Enhanced Hg Control.
Release estimates for percolation controlled scenario are compared to release estimate based on total content. The
amount of the arsenic that would be released if the release concentration was at the MCL is also shown for comparison
(LSdefaultscenario= 12.5 L/kg and LS95%= 100 L/kg).

                                                                                                 203
-------
Characterization of Coal Combustion Residues
100-Year  Selenium  Release  Estimates
                                      LAB - Selenium
       10000
        1000
                4100 ug/kg
      CO
      Q)
      0)
      I  100
      en
      CO
      
      >,
      6
      o
      10


-





100%







0-1







1° R ii'

51. 5°/





i/kn

3 58







7.9 ug
14.3%






'kg 98







9.7 ug/
24.1%






kg


10







3.6 ug/
2.5%







kg


                                                                                  MCLLS125 = 625 ug/kg
      A)
           Total content   Combustion    Default -pH 3    Default -pH  Default- Natural
                       Waste Landfill -                   12.5           pH
                       95% confidence                                5.8
CO

6
o
                                     LAT- Selenium
        10000
D)

j±  1000
0)
CO
     E    100
                4300 ug/kg
                   100%
     B)
           10

                                                          22%
                                           559.4 ug/kg
                              _
                                   _
                                                                	MCLLS95% = 5000 ug/kg
                                                                                  MCLLS1  = 625 ug/kg
                                                                     73.1 |jg/kg
                                                                       1.7%
                Total content   Combustion    Default-pH 3
                            Waste Landfill -
                            95% confidence
                                                   Default-pH   Default-Natural
                                                      12.5           pH

                                                                   5.0
Figure 1-14.100-year Selenium Release Estimates from A) Baseline Fly Ash and B) Fly Ash with Enhanced Hg Control.
Release estimates for percolation controlled scenario are compared to release estimate based on total content. The
amount of the selenium that would be released if the release concentration was at the MCL is also shown for comparison
(LSdefaultscenario= 12.5 L/kg and LS95%= 100 L/kg).
204
-------
                                                            Characterization of Coal Combustion Residues
Comments

Figure 1-3:
  •  Fly ash from the test case had greater total Hg content
    than the fly ash from the baseline case  (by about 3
    times).
  •  Hg release is low (but poor replication) for both baseline
    and test cases.

Figure 1-5:
  •  The fly ash from the test case had lower total As con-
    tent than  that from the baseline case  (by about 4.5
    times).
  •  Arsenic release was close to or exceeded the MCL
    (10 |Jg/L) for both the baseline and the test cases for
    all pH conditions.

Figure 1-7:
  •  Fly ash from the test case had greater total Se content
    than the fly ash from the baseline case (by about 4.5
    times).
  •  Selenium release from both the baseline and the test
    cases was close to the MCL (50 |Jg/L) for most pH
    conditions.

Figures 1-11 and 1-12:
  •  The fly ash from the test case would result in similar
    As release than expected from the baseline case, with
    a 95% probability to be less than 18700 and 20000 ug/
    kg, respectively.
  •  The fly ash from the test case would result in similar
    Se release than expected from the baseline case, with
    a 95% probability to be less than 2115 and 2045 |Jg/kg,
    respectively.

Figure 1-13:
  •  For the 95% probability scenario, arsenic release from
    both cases would exceed the amount that would be
    released if the release concentration was at the MCL.
  •  For two of the three default scenarios considered (i.e.,
    pH 3 and natural pH), arsenic release would most likely
    be less than the amount that would be released if the
    release concentration was at the MCL. However, for
    the default scenario at pH 12.5, arsenic release would
    most likely exceed the amount that would be released
    if the release concentration was at the MCL.

Figure 1-14:
  •  Similar Se release would be expected for the test case
    compared to the baseline case for all scenarios exam-
    ined.
  •  For the 95 % probability scenario, selenium release from
    both cases would be less than the amount that would
    be released if the release concentration was at the MCL
    and the LS ratio was the resultant LS ratio for the 95%
    case (i.e., around 100 L/kg). However,  selenium re-
    lease would be greater than the amount that would be
    released if the release concentration was at the MCL
    and the LS ratio was the LS ratio of the default  sce-
    nario considered (i.e., 12.5 L/kg).
                                                                                                      205
-------