EPA-600/4-81-019
March 1981
MONITORING STRATEGIES
FOR FLUIDIZED BED COMBUSTION
COAL PLANTS
by
A.B. Garlauskas, C.E. Hina, T.T. Blair/
M.J. Kangas, and C.L. Cornett
Dalton-Dalton'Newport, Inc
3605 Warrensville Center Road
Shaker Heights, Ohio 44122
Contract No. 68-03-2755
EPA Project Officer
Richard Bateman
Advanced Monitoring Systems Division
Environmental Monitoring Systems Laboratory
P.O. Box 15027
Las Vegas, Nevada 89114
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89114
-------
EPA-600/4-81-019
March 1981
MONITORING STRATEGIES
FOR FLUIDIZED BED COMBUSTION
COAL PLANTS
by
A.B. Garlauskas, C.E. Hina, T.T. Blair,
M.J. Kangas/ and C.L. Cornett
Dalton«Dalton*Newport, Inc
3605 Warrensville Center Road
Shaker Heights, Ohio 44122
Contract No. 68-03-2755
r-
u> EPA Project Officer
i. Richard Bateman
Advanced Monitoring Systems Division
^. Environmental Monitoring Systems Laboratory
ri P.O. Box 15027
Las Vegas, Nevada 89114
o
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89114
-------
DISCLAIMER
This report has been reviewed by the Environmental
Monitoring Systems Laboratory, U.S. Environmental Protection
??e?C^ and aPProved for Publication. Approval does not signify
that the contents necessarily reflect the views and policies of
the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or
recommendation for use.
li
-------
ABSTRACT
This report presents air and water monitoring strategies
for commercial-size Fluidized Bed Combustion (FBC) coal
plants. It is one of five reports developing air and water
monitoring strategies for advanced coal combustion (FBC), coal
conversion (coal gasification and liquefaction), and oil shale
conversion commercial-scale plants. The overall objective of
the five-part project is to assure that, when any of the coal
or oil shale technologies become commercialized, appropriate
and cost-effective monitoring requirements are in effect to
protect the public health and the environment without delaying
commercialization.
A detailed analysis of FBC processes was conducted to
determine the quantities and nature of air and water ef-
fluents. Process data were obtained from former and existing
small experimental and larger pilot plant FBC units. Data on
the types and concentrations of pollutants were compiled and
the effluent process streams identified. A pollutant selec-
tion and priority-ranking procedure was designed and used to
determine which pollutants of the FBC processes needed moni-
toring.
Air and water quality monitoring approaches in the envi-
ronment around an FBC plant were evaluated. Then air quality
monitoring strategies were developed for compliance audits,
for monitoring regulated pollutants from coal-fired plants,
and for monitoring fugitive emissions. Water quality moni-
toring strategies were developed for leachates from fly ash
and other solid wastes and from coal and sorbent material
storage piles, and for pollutants generated by cooling towers,
ponds, or canals.
Appropriate sampling procedures, analytical instrumen-
tation, and general analysis methods were identified and
selected. Monitoring data-interpretation approaches and re-
porting procedures were researched. A basic quality assurance
program applicable to FBC and other fossil energy technologies
was designed.
Conclusions and recommendations demonstrate that FBC
plants present less undesirable environmental consequences and
111
-------
are more efficient than conventional coal-fired plants. How-
ever, significant information gaps exist, preventing accurate
extrapolation of experimental data to full commercial-scale
FBC plants, and the current FBC environmental assessment pro-
gram needs acceleration and better data dissemination. A
number of specific recommendations were developed.
This report covers the period October 1, 1978, to
September 30, 1979.
1v
-------
CONTENTS
Abstract 111
Figures x
Tables x-f-(
SECTION 1 - INTRODUCTION - 1
Purpose and Scope 1
Monitoring Systems 2
Monitoring System Design Considerations 6
Report Organization 10
SECTION 2 - FLUIDIZED BED COMBUSTION OF COAL:
PROCESS, ENVIRONMENTAL POLLUTANTS, AND
POLLUTION CONTROL 11
Background 11
Process Description 14
Basic Principles 14
Atmospheric Fluidized Bed Combustion of Coal . 19
Pressurized Fluidized Bed Combustion of Coal . 22
Environmental Pollutants from FBC of Coal 24
Methods of Determining Pollutants 24
Sources and Major Types of Pollutants 27
Environmental Pollutants 28
Flue Gases and Particulates 34
Solid Wastes and Leachates 47
Sorbent Leachate 48
Boiler Slowdown 48
Cooling System Effluents 48
Sorbent Regeneration 51
Biological Examination 55
Pollution Control 55
Background 55
Pollution Control of Flue Gases 60
General Pollution Control Applications . 60
Carbon Utilization .62
Nitrogen Oxide Reduction 63
Sulfur Dioxide Reduction 68
Particulate Matter 73
Outlook 73
-------
SECTION 3 - SELECTION OF POLLUTANTS FOR MONITORING .... 81
Introduction . 81
Assumptions 85
Categorization 87
OverView 87
Spedies Identification 87
Media Determination 88
Identification of Standards and Criteria
for Specific Chemical Components of the
Effluent Streams 88
Identification of Extremely Hazardous
or Toxic Substances 89
Summary of the Categorization Process 91
Priority Ranking 91
Overview 91
Identification of the Potential for
Receptor Damage from Unregulated
Pollutants 92
Identification of Emission Rates
and Ambient Level Concentrations 101
Methods of Combining Effects and Concen-
tration Values to Estimate Hazard
Potential . .104
Development of a Decision Rule to Select
Pollutants For Monitoring Based on Thei-r
Potential for Environmental Hazard 108
Preparation of the Preliminary List of Pollutants
to be Monitored 109
Final Selection of Pollutants Based on Technical
and Cost Considerations 109
Technical Quality of Available Analytical
Methods 110
Cost Considerations 113
Conclusions 118
SECTION 4 - AIR QUALITY MONITORING STRATEGIES 119
Introduction 119
Air Quality Impact Generators 120
Conditions Where Ambient Monitoring Is Indicated . .120
Regulations 124
New Source Performance Standards 124
Prevention of Significant Deterioration. . . .126
Monitoring Requirements 129
Selection of Pollutants to be Monitored 131
Introduction 131
Categorization and Prioritization 132
Category I Pollutants 132
Category II Pollutants 133
vi
-------
Estimation of Expected Ambient Concen-
tration of Problem Pollutants 133
Preliminary List of Pollutants for Monitoring.133
Monitoring Network Design and Dispersion Models. . .133
Background 133
Air Monitoring Network Design 134
Monitoring Strategies for FBC Installations 137
Background .137
Compliance Audit Monitoring 137
Methods of Measurement 139
Ratio of Components Techniques 139
Upwind/Downwind Sampling 139
Differential Plume Impact
Monitoring 139
Remote Sensing 142
Dispersive Correlation/Spectroscopy . . 145
Non-Dispersive Gas Filter
Correlation Spectrometer (GFC) .148
Ultraviolet Television Sensor. . .148
Emission Rates for Other
Materials '. . .150
Level 1 and Level 2 Assessment Monitoring. . .151
Fugitive Emission Monitoring . .161
Heat of Hydration and Alkalinity . . . .162
Portable Wind Tunnel Monitoring 164
Sulfates, Sulfites, and Organic Adduct
Analysis 165
Long-Range Plume Transformation
Studies 168
Short-Range Plume Transformation
Studies . . .169
Balloon-Borne Sampling System 169
Opacity Measurements 171
Opacity Mass Concentration Rela-
tionship 172
Special Considerations for Fluidized
Bed Combustion Units 176
Applications of Opacity/Mass
Concentration Relationships 176
In-Stack Opacity Measurement 178
Remote Measurement of Plume Opacity. . .179
Trained Observers 179
The Method of Contrasting
Targets 180
Laser Radar 181
Sun Photometer 181
Data Reporting 181
vii
-------
SECTION 5 - WATER QUALITY MONITORING STRATEGIES 183
Introduction 183
Impact Generators 184
Conditions Where Water Quality Monitoring Is
Indicated . .185
Selectioni of Pollutants To Be Monitored 190
Groundwater Monitoring 197
Surface Water Monitoring 197
Physical Sinks 198
Biological Sinks 199
Considerations Specific to Surface Water Monitoring.200
Definition of the Watershed and Hydrological
System 200
Transport of Pollutants in Surface Aquatic
Systems 201
Considerations Specific to Groundwater Monitoring. .202
Evaluation of Ability of Pollutants to
Percolate From Land Surface to the Zone of
Saturation 204
Evaluation of Attenuation of Pollutants
in the Vadose Zone 205
Evaluation of Attenuation of Pollutants
in the Zone of Saturation 206
Sampling 207
General Considerations 207
Number and Location of Sampling Points .207
Frequency of Sampling 208
Sample Volumes 209
Sample Preservation. 209
Sampling Equipment 210
Manual Samplers 210
Analytical Methods 214
Approved and Standard Methods 214
Trace Organics 221
Biological Methods 221
Level I/Level 2 Analysis 223
Data Recording 225
Data Handling 230
Data Analysis 230
Regression Analysis 230
Time Trend Analysis 231
Factor and Principal Component Analysis. . . .231
Water Quality Indices 232
Water Quality Models ." . . . .233
SECTION 6 CONCLUSIONS AND RECOMMENDATIONS 235
Conclusions 235
Recommendations 235
vili
-------
REFERENCES 239
APPENDICES
A. Guidelines for Developing a Quality Assurance Pro-
gram for Ambient Monitoring Associated With Energy
Technologies 259
B. List of Laws, Regulations, and Executive Orders
Pertaining to Environment and Safety and Health
Requirements 297
ix
-------
FIGURES
Number Page
1 Multimedia Sampling Approach Overview 7
2 Selected FBC System Options 18
3 Schematic Diagram of Atmospheric Pressure FBC
Boiler 20
4 Schematic Diagram of Pressurized FBC Boiler
with Sorbent Regeneration 23
5 Generic Schematic of FBC Unit Operations
With Numbered Process Flows Producing
Waste Streams 29
6 Schematic Diagram of Emission Sources Within
' Atmospheric-Pressure FBC Plant 30
7 Scheme of FBC (Materials Flow) 49
8 Air Pollutant Particle Sizes 58
9 Emission Control Equipment Chart 59
10 Effect of Oxygen Concentration on
Hydrocarbon Emission 64
11 Nitrogen Oxides Emissions (expressed as
N02) From Atmospheric Fluidized
Bed Combustion Units 65
12 Nitric Oxide Reduction With -325 Mesh (<44ym)
Tymochtee Dolomite • . 66
13 Effect of Gas Velocity on Nitric Oxide
Reduction in Flue Gas. . 67
14 Composite Plot of Data for NOX Emissions
from Fluidized Combustion of Coal 67
x.
-------
Number Page
15 Sulfur Removal Performance for Typical
Sorbents (Projected using Westinghouse
Kinetic Model) 69
16 FBC Power Plant Energy Conversion
Efficiency . 70
17 Schematic Flow Diagram of the
Precalcination Process 72
18 Total Solids from Combustor (AFBC) 74
19 CFCC Particulate Emissions 75
20 Design of an Ambient Monitoring Program 83
21 Procedure for Selecting Pollutants for a
Source-Oriented Ambient Monitoring Plan. .... 86
22 Methodology for Deriving MATEs from
Empirical Data 95
23 Methods Selection-Technical Aspects Ill
24 Methods Selection - Cost Aspects 114
25 Differential Plume Impact Data 141
26 Bas-ic Level 1 Sampling and Analytical
Scheme for Solids, Slurries, and Liquids . . . .153
27 SASS Analysis Requirements 154
28 Flue Gas Sampling Flow Diagram 155
29 Source Assessment Sampling Train Schematic . . .156
30 Fugitive Air Sampling Train Components 157
31 Integrated Gas-Sampling Train 159
32 Schematic of Second Generation Sampling
Package 170
33 K as a Function of Log Normal Size
Distribution Parameters 173
34 Data Flow in a Source-Oriented Ambient
Monitoring Program 227
xi
-------
TABLES
Number Paqe
1 Arithmetic Mean of Proximate and Ultimate
Analysis and Elemental Composition for
Coals from Various Basins 12
2 Types of Potentially Harmful Pollutants
Associated With FBC Technology 13
3 Alternative Sorbent Selection. . 25
4 Stream Designations for Generic Units 28
5 Radioactivity Measurements 33
6 Reported Ranges of Some Minor Stack Flue
Gas Constituents 35
7 Volatile Element Analysis in Flue Gas 37
8 Projected Flue Gas Emission of Trace
Elements as a Percentage of the Element.
Entering the System 37
9 Potential Enrichment of Elements in the Fine
Particulate Catch 38
10 Inorganic Elements and Compounds Identi-
fied in Flue Gas Particulates 42
11 Polycyclic Organic Material Analysis 43
12 Inorganic Analysis of Particulate Emissions in
Flue Gas .'45
13 Trace Metals in Bed Reject Material 46
14 Pollutants From Leaching of Coal Piles 50
15 Chemical Agents Associated With Boiler
Treatment 50
xii
-------
Number Page
16 Chemicals Used in Recirculative Cooling
Water Systems 52
17 Cooling Tower Corrosion and Scale Inhibitor
Systems .52
18 Representative Concentrations of Elements in
Leachate of FBC Spent Bed Material 54
19 Leachates from Flue Gas Particulates Collected
from FBC Processes 54
20 POM Compounds in Leachates from AFBC and PFBC
Combustion of Coal Solid Wastes 56
21 Anion Analysis of Leachate of FBC Solid
Wastes 56
22 Major Contaminants in the Flue Gas of
Fluidized Bed Combustion and Applicable
Pollution Control Equipment 61
23 Distribution of Combustible Losses 62
24 Predicted Ca/S Requirements for the
Atmospheric Fluidized Bed Combustor. ..... 71
25 Properties of the Particles Suspended in
the Flue Gas 74
26 Factors Influencing the Commercialization of
Fluidized Bed Combustion of Coal 79
27 Criteria Used to Identify Extremely Toxic
Substances 90
28 Emission Level Goals Foundations . 94
29 Basic Data and Derivations for Ambient
Level Goals 96
30 Approaches to Estimating Ambient Level Goals
(ALG) Which Function as Minimum Potential
Hazard Levels in the Ambient Environment . . . 102
31 Approaches to Estimating Ambient Environmental
Concentrations Expected to Impact Human and
Environmental Receptors. ........... 103
32 Sources of Atmospheric Emissions 121
xiii
-------
Number Page
33 Federal Standards of Performance for New
Stationary Sources of Air Pollution 125
34 Allowable Increases in Ambient Pollution
Levels Under PSD Regulations 127
35 Ambient Air Quality Standards 128
36 Assumed Natural Background Concentrations. . . 130
37 Comparison of In-Stack and COSPEC II SC>2
Concentration Measurements 149
38 Summary of Average Values of Remote GFC and
Extractive Data 149
39 Comparison of SC-2 Concentration and Effluent
Velocity Measurements By Video and In-Stack
Techniques 150
40 Basic Level 1 Analyses 152
41 Applicable Ambient Water Quality Standards . . 189
42 Summary of Pollutants Segregated Among
Categories 194
43 Selected EPA Effluent Standards for Steam
Electric Power Generating Plants 196
44 Volume of Sample Required for Determination
of the Various Constituents of Industrial
Water 211
45 Summary of Preservation Methods for Water
Samples 212
46 State-of-the-Art Water Quality Methods:
Non-Metals 216
47 State-of-the-Art Water Quality Methods:
Metals 218
48 Atomic Absorption Concentration Ranges .... 219
49 Criteria for Selection of Analytical
Methods 220
xiv
-------
Number Page
50 Organics for Which Discrete Analytical Methods
Have Been Developed 222
51 NAWDEX Participating Data Systems 228
XV
-------
SECTION 1
INTRODUCTION
PURPOSE AND SCOPE
This report presents strategies for source-oriented am-
bient air and water pollutant monitoring of commercial-scale
Fluidized Bed Combustion (FBC) coal plants. This is the first
of five monitoring strategy reports to be completed under the
overall project, Air and Water Monitoring Guidelines for Ad-
vanced Coal Conversion and Combustion Plants. Each report de-
velops monitoring strategies for one of the following:
Fluidized Bed Combustion of Coal
Low/Medium BTU Gasification of Coal (surface and
in situ)
Oil Shale Conversion
Coal Liquefaction
High BTU Gasification of Coal
The strategy report for each technology focuses analyti-
cally on environmental requirements as these technologies move
to commercialization. Each technology is dealt with generi-
cally wherever possible and appropriate. The reports are not
technical monitoring manuals; they are monitoring policy guid-
ance reports, to help the U.S. Environmental Protection Agency
(EPA) evaluate monitoring requirements for emerging energy
technologies.
The basis for the project is to ensure that when one of
the new fossil technologies (four coal, one shale) is ready to
move to commercial status, it will not be delayed by inade-
quate, inappropriate or unavailable environmental monitoring
requirements. The guidance reports will help to determine the
most likely candidate technologies and to develop the
strategies for environmental monitoring before these technol-
ogies are ready for commercialization, to assure that they
will be moved rapidly ahead on a sound environmental basis.
The monitoring strategies are based on "ambient" condi-
tions, rather than only on the products at the end of the
-------
"stack." Thus, they identify pollutants, their paths upon
leaving the stack or pipe, and their transformation before
they reach the ground, people, plants, or animals.
These monitoring strategies/approaches will provide a
critical link between traditional "stack pipe" monitoring and
the study of their effects on plants, people, and animals, and
will make the final required monitoring programs both more
cost-effective and more closely associated with actual effects,
i
If the development of environmental monitoring strate-
gies is delayed, the implementation of the associated tech-
nologies may also be delayed. It is important to maintain and
expand this effort to:
Ensure that EPA regulations are adequate, appro-
priate, realistic, and cost-effective.
Ensure cost-effective monitoring programs.
Ensure that the development of energy technologies
is not delayed because of lack of regulatory re-
quirements.
Minimize the cost of development, both for tech-
nologies and for environmental control.
MONITORING SYSTEMS
Environmental monitoring, a systematic and scientific
approach to assuring a congruous interface between natural and
man-made systems, is an essential component of environmental
management. It serves to minimize the impact of man's activ-
ities through the resolution of the conflicts that arise from
human activities on the natural environment.
The EPA has the major responsibility to assure that the
country's environment is properly managed. With proper man-
agement tools, the Agency can gather data to develop and main-
tain an adequate environmental monitoring capability. A sys-
tems approach to data-gathering links private and governmental
monitoring systems, to provide data that complies with envi-
ronmental monitoring regulations and degree-of-accuracy re-
quirements, EPA monitoring systems, augmented by other
governmental and private systems, can provide a sound basis
for decision-making. Critical aspects of a problem can be
adequately analyzed and findings incorporated early in the de-
sign and control strategy phases. This is especially impor-
tant with emerging energy technologies that have unknown or
partially known effects on the environment.
-------
The three basic categories of environmental monitoring
are source monitoringr ambient monitoring, and effects moni-
toring. Each has its strengths and weaknesses. When properly
linked in a systems approach, each category fills data gaps,
assures the correlation and continuity of data from source to
effect, and reinforces the data base on any scale of analy-
sis—national, regional, global, or local. The measurement
systems of each monitoring category can be remarkably dif-
ferent but, with clearly defined linkages and data correlation
techniques, they can produce a comprehensive data base that
will demonstrate environmental acceptability or non-accepta-
bility of a specific energy conversion plant or the entire
energy technology. Definitions of each category of monitoring
are given in the following paragraphs.
Source monitoring refers to measurements taken at the
point or points at which pollutants escape or are discharged
into the environment. Source monitoring seeks to determine
what residuals enter or will enter the environment, from what
sources, and in what amounts. It serves in enforcing com-
pliance with emission standards; forecasting those activities
of production, storage, transportation, use and disposal that
may cause pollution; and taking inventory of emissions from
production processes. Sources include point sources (indus-
trial stacks, discharge pipes, and other stationary sources) ,
area or non-point sources (groupings of small sources spread
over regions [cities, farmlands, forests]), and mobile sources
(automobiles, aircraft, trucks, buses, railroads, pipelines,
ships). The latter include spills of toxic and other hazard-
ous materials during transport (National Research Council,
1977).
Ambient ("surrounding on all sides") monitoring refers
to the measurement of concentrations of pollutants in the air,
water, soil and biota; it is designed to determine what con-
centrations of residuals are present in air, water, soil, food
and animal tissues (National Research Council, 1977). A more
restricted definition of ambient monitoring can be developed
from the EPA definition of ambient air as "that portion of the
atmosphere, external to buildings, to which the general public
has access" (U.S. EPA, 1979c); in which case, ambient moni-
toring can be defined as monitoring air and water to which the
general public has access.
Effects monitoring is designed to determine the conse-
quences of current and potential pollutants on humans, ani-
mals, plants and materials, and to evaluate their economic,
social, and aesthetic costs (National Research Council,
1977) . Some examples of effects monitoring are: epidemio-
logical studies of disease occurrence in human populations
-------
where the causal factors are related to environmental pollu-
tion; disease detection programs based on continuing surveil-
lance of particular groups of people who are or have been ex-
posed to chemicals known to be animal carcinogens but not yet
identified as human carcinogens; monitoring of occupational
health statistics; and monitoring of animals and plants to
determine how chronic exposure to low-level pollutants affects
them. Monitoring of ecological systems includes productivity,
diversity, fecundity, bioaccumulation, and other important as-
pects of biological species on plots in natural ecosystems
(such as a watershed for terrestrial life or river for aquatic
life); and productivity of major agricultural crops, forage,
and forests on selected field plots (National Research Coun-
cil, 1977) .
The application of source, ambient, and effects moni-
toring to special purposes and in combination with each other
has created variants of monitoring categories with specialized
meaning. One such sub-category is source-oriented ambient
monitoring, as used in this report. This specialized form of
ambient monitoring is designed to monitor pollutants, hazard-
ous wastes, and toxic substances (pollutants) in an area where
they can be in some way attributed to a specific source.
Source-oriented ambient monitoring also focuses on the deriva-
tives of pollutants, hazardous materials, and toxic substances
which are transformed upon release into the ambient environ-
ment including air, water, land, and living tissue.
The goals and objectives of source-oriented ambient
monitoring are to:
• Separate out the materials in the environment
directly attributable to a source from the back-
ground concentrations.
• Monitor those substances and materials that cannot
be monitored at the point of discharge (e.g.,
fugitive emissions).
• Monitor the substances that are transformed upon
release from the source into, the ambient environ-
ment (including in living tissues).
Serve as a link between source monitoring and ef-
fects monitoring by providing the necessary infor-
mation of what path a substance takes in the am-
bient environment upon release from a source to
its final site of action. In this way, source-
oriented ambient monitoring helps establish
cause/effect relationships.
-------
Source-oriented ambient monitoring, then, traces energy
source effluents from the source to the ambient environment.
This approach, which is of critical importance in establishing
clear pollutant pathways and impacts, relies heavily on the
availability of sound ambient environmental quality baseline
data. Although at least eight different federal agencies
monitor environmental quality systematically throughout the
nation in different regions (Morgan, 1978) , the type of in-
formation generated by these agencies is primarily geared
toward conventional pollutants. The certainty of development
of additional coal and oil shale resources and their pro-
cessing for energy generation has created the need for moni-
toring non-conventional pollutants that will be generated from
these new energy sources. Although FBC of coal does not ap-
pear to be a hazardous or toxic-waste-producing technology,
based on current and anticipated regulations, other fossil
fuel combustion and conversion technologies such as coal gasi-
fication or liquefaction, or oil shale extraction may generate
significant toxic or hazardous pollutants.
Based on the above considerations, a monitoring system
must 1) establish a sound baseline of toxic pollutants in the
ambient environment before energy developmend, and 2) detect
changes in any of the pollutants or introduction of new toxics
following energy source development. The type of monitoring,
the pollutants to be monitored, and the methods to be used
should be determined before new technologies are commercial-
ized, to allow development to proceed within a sound environ-
mental framework.
In this context, monitoring should be "anticipatory,"
focusing on on-going surveillance of air, surface water, and
groundwater for potential changes in concentrations of speci-
fied toxic substances which may be induced by the development
of energy resources and/or technologies in the region of
interest. Its scope is defined by:
Types of energy resources likely to be developed
in the region
• Toxic pollutants expected to be associated with
them
Nature of the technology and its associated pol-
lution controls and waste disposal methods
Extent of the air regions affected
• Types of water uses likely to be impacted by these
pollutants in the region in question
-------
Most probable loci of these impacts.
The goals of anticipatory monitoring are to:
Establish baseline concentrations of toxic sub-
s'tances in the air and water regions of interest
so that future changes can be detected before they
adversely affect water uses
• Focus monitoring efforts on the air and water
regions most likely to be impacted by particular
types of energy resource development
• Focus monitoring efforts on the pollutants most
likely to adversely affect air quality and water
uses
• Determine whether there are reliable correlations
among various pollutants associated with a given
energy resource/technology and/or geographic area,
to identify the significant "indicator species"
that should be routinely monitored.
MONITORING SYSTEM DESIGN CONSIDERATIONS
Development of a source-oriented ambient monitoring pro-
gram requires, as a first step, delineation of specific objec-
tives. Also, the program should be visualized in the proper
perspective; as Figure 1 illustrates, it is multimedia.
Objectives of a source-oriented monitoring program should in-
clude:
Accurate and complete characterization of ambient
air and surface and groundwater conditions in the
vicinity of the sources and for areas expected to
be impacted by discharges.
• Monitoring air, surface water, groundwater, and
associated media for contamination resulting from
operation of FBC units.
• Documentation of existing and continuing ecologi-
cal conditions.
Protection of the public against unexpected plant
discharges.
Maintenance of records for the protection of the
plant operator against any liability claims.
-------
Figure 1. Multimedia sampling approach overview.
-------
Accumulation of data to assist in the improvement
of pollutant dispersion models and estimation of
impacts for future coal-fired FBC units.
Provision of data for use in describing potential
hew source performance standards for FBC units.
The activities involved in monitoring air and water
quality will be similar, whether the program is designed to
determine compliance to ambient environmental quality stan-
dards, to identify sources of pollutants, or to determine
impacts of a specific source. The general operational activ-
ities of a source-oriented ambient monitoring program are:
Network Design
Sample Collection
• Laboratory Analysis
Data Handling
• Data Analysis
Information Utilization
Each major activity contains a subset of specific ele-
ments which must be considered and described in detail prior
to initiation of actual monitoring operations. The following
monitoring system activity list has been adapted from Ward
(1979).
Network Design
Statement of objectives
- Parameter selection
Sample station location (macro-location)
Sampling frequency
Sample Collection
Sampling point (micro-location)
Field measurements
Sampling technique
Sample preservation
- Sample transportation
Quality assurance
Laboratory Analysis
Selection of analytical methods
Operational procedures
Quality assurance
Data recording
8
-------
Data Handling
Data reception from laboratory or outside
sources
Screening and verification
Storage and retrieval
Data reporting
- Dissemination
Data Analysis
Basic summary statistics
- Regression analysis
Air and water quality indices
Interpretation of quality assurance data
Time series analyses
Air and water quality models
Information Utilization
Reporting formats
- Information needs
Evaluation of overall program
Data utilization evaluation
Applications of source-oriented ambient monitoring are
for:
• Detecting pollutant transformations occurring
during transport to their site of action
• Establishing the links between source of pollution
and effects (cause/effect)
Calibrating emission models
Auditing accuracy of emission inventory data ob-
tained by conventional techniques (sources tend to
be more careful of their operations during conven-
tional emission tests than they might otherwise be)
Obtaining entry warrants for conventional moni-
toring, by showing probable cause to suspect vio-
lation of standards
• Determining underground water flow patterns and
dilution. Also can be useful in complex terrains,
where modeling often breaks down
Investigating complaints
Monitoring groundwater around hazardous waste dis-
posal areas, under proposed Resource Conservation
and Recovery Act (RCRA) regulations (12-18-78),
would be required and air monitoring might be
-------
required as a permit condition on a case-by-case
basis. This contract could help to define what to
monitor:
• .'Measuring emission rates from flares and fugitive
'sources that cannot be measured by more conven-
tional techniques
• Researching the health and ecological effects of
emerging technologies
Providing for an alarm system for hazardous fugi-
tive emissions (e.g., polynuclear aromatics [PNAs]
from coal gasification)
Supporting enforcement actions by providing docu-
mentation of the significance of the impact of a
source, for legal action if necessary; helping to
document the seriousness of impact in a way that
is less theoretical than modeling—may help con-
firm the results of modeling
• Increasing the store of knowledge regarding pollu-
tant transport, diffusion, and accumulation on a
small geographic scale.
REPORT ORGANIZATION
Section 2 of this report reviews the FBC process, de-
fines the effluents, identifies the sources of pollutants and
their control, and reviews the commercialization outlook for
this technology. Section 3 presents a procedure for selecting
pollutants to be included in a monitoring program of FBC com-
mercial-scale plants. Section 4 discusses air quality moni-
toring methods and approaches for FBC facilities, and Section
5 discusses water quality monitoring systems and methods
applicable to FBC plants. Sections 4 and 5 also discuss qual-
ity control and quality assurance and reporting procedures.
Quality assurance is treated in detail in Appendix A. Section
6 presents major conclusions and recommendations.
10
-------
SECTION 2
FLUIDIZED BED COMBUSTION OF COAL:
PROCESS, ENVIRONMENTAL POLLUTANTS, AND
POLLUTION CONTROL
BACKGROUND
Coal, which represents about 85 percent of the United
States fossil fuel resources, will almost certainly be used to
fulfill a major portion of the country's energy needs in the
very near future. Coal can be readily extracted and is avail-
able in accessible geographic locations, and its utili.zation
can fill the gap created by diminishing oil supplies. In-
creased use of coal will create different stresses on the
economy, environment, and technology. Coal use currently con-
stitutes about 18 percent of the U.S. energy consumption. How
soon coal will fill the energy gaps and increased requirements
will be a function of conservation, environmental regulations,
tax laws, economics and developing technologies.
Coal is a complex, heterogeneous material containing
organic and mineral matter. The organic matter is composed
primarily of carbon and small amounts of hydrogen, nitrogen,
oxygen, and sulfur. The mineral matter consists chiefly of
clay minerals, mineral forms of sulfur (mostly pyrite), :cal-
cite, and smaller amounts of other minerals. Its composition
is a fossil record of the geochemistry occurring at the !time
the original material from which the coal formed was -depos-
ited. Geochemical and coalification processes effected
changes in heat, water, mineralogy, and, importantly, in sul-
fur content, with changes in the concentration and abundance
of minor and major trace elements. Table 1 summarizes the
elements and concentration ranges for coals from four basins.
The inorganic constituents in coal are minerals derived from
the surrounding rocks and incorporated into the coal seam.
The burning of coal releases most of the constituents
into the environment in some form, creating public health and
environmental problems. Many of the elements and compounds
are injurious to any form of life even in small quantities.
Some elements such as rubidium are toxic only in large quanti-
ties and concentrations. Elements that are extremely toxic
even in small quantities are arsenic, beryllium, cadmium,
lead, and mercury. To understand and deal with toxic elements
11
-------
TABLE 1. ARITHMETIC MEAN OP PROXIMATE AND ULTIMATE ANALYSIS
AND ELEMENTAL COMPOSITION FOR COALS FROM VARIOUS BASINS
Constituent
i
Moisture %
Volatile matter %
Fixed carbon %
Ash %
Hydrogen %
Carbon %
Nitrogen %
Oxygen %
Sulfur %
Heating value MJ/kg
Elements g/kg
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Cadmium
Calcium
Chlorine
Chromium
Cobalt
Copper
Fluorine
Gallium
Iron
Lead
Lithium
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Niobium
Phosphorus
Potassium
Scandium
Selenium
Silicon
Silver .
Sodium
Strontium
Telluvium
Thallium
Thorium
Tin
Titanium
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
Appalachian Pennsylvania
coal3 anthracite3
2.8
31.6
54.6
11.0
4.9
72.6
1.3
7.8
2.3
30
16.
0.0012
0.027 .
0.11
• 0.002
0.0001
0.030
0.011
0.0007
1.2
0.72
0.020
0.007
0.024
0.080
0.007
19.
0.0153
0.0276
0.68
0.62
0.00024
0.003
0.015
0.005
0.09
2.3
0.005
0.0047
27.
0.00003
0.32
0.1
0.00034
0.0001
0.0048
0.0024
0.9
0.0014
0.02
0.001
0.01
0.02
0.05
1.4
6.5
79.5
12.6
2.4
80.1
0.8
3.2
0.8
29.7
20.
0.0009
0.006
0.1
0.002
<0.0001
0.01
0.001
0.0003
0.7
1.5
0.02
0.007
0.0027
0.061
0.007
4.4
0.01
0.033
0.6
0.02
0.0002
0.002
0.02
0.003
0.075
2.4
0.005
0.004
27.
0.003
0.5
0.1
<0.0001
0.05
0.003
0.001
1'.5
0.002
0.02
0.001
0.01
0.016
0.05
Illinois Basin
coalb
10.02
39.80
48.98
11.28
4.98
70.69
1.35
8.19
3.51
29.6
12.2
0.0135
0.1491
0.0172
1.138
0.1527
0.0289
7.4
1.5
0.1410
0.0915
0.1409
0.5930
0.0304
20.6
0.3983
0.5
0.5316
0.0021
0.0796
0.2235
0.6277
1.6
0.0199
23.9
0.5
0.0456
0.6
0.3313
3.1304
0.721
North Dakota
lignite3
29.7
29.6
32.1
8.6
6.1
43.1
0.7
40.0
1.5
16.7
3.7
0.0003
0.0067
0.4
0.0002
0.12
0.0001
12.
0.17
0.002
0.001
0.0052
0.026
0.0014
13.
0.0034
0.0021
3.43
0.048
0.002
0.0022
0.32
0.2
0.0012
0.0008
8.5
0.002
3.57
0.34
0.0035
0.0004
0.33
0.0007
0.004
0.0002
0.003
0.0033
0.01
DeAngelis and Reznik, 1979.
Ruch et al., 1974.
12
-------
is essential in developing viable coal utilization tech-
nologies. Compounding the problem of toxic elements are the
complex organic and less complex inorganic compounds that are
formed during combustion of coal. Many of them are toxic or
otherwise injurious to living forms. Table 2 shows some of
the major toxic elements and compounds released by combustion
of coal.
TABLE 2. TYPES OF POTENTIALLY HARMFUL POLLUTANTS ASSOCIATED WITH FBC
TECHNOLOGY
(Cornaby et al., 1977)
General category
Examples
Emissions of Single Chemicals
Emissions of Aggregated Chemicals
Emissions of Complex Effluents
Emissions of Energy
Nonpollutant Factors
Cadmium, Chromium (Heavy Metals)
Carbon Monoxide (Toxic Gases)
Radionuclides
Other
Total Hydrocarbons
Oxidants
Sulfates
Radionuclides
Particulates/Dust (organic/
inorganic totals)
Complex Mixtures of Effluents
Microorganisms
Noise,
Heat
Electromagnetic Radiation
j
Land Usage
Water Usage
Soil Stability
Aesthetics
Increasing recognition of public health and environmental
problems associated with combustion of coal and the attendant
government regulations, coupled with economic problems like
inflation and severe price increases on imported oil, has led
to reexamination of coal technologies utilized in the pro-
duction of energy. Conventional coal combustion systems are
being upgraded and retrofitted with pollution control equip-
ment, i.e., scrubbers, to comply with stationary source emis-
sion standards. The newly emerging combustion technologies
13
-------
have some inherent advantages over conventional coal combus-
tion processes and will render much of the older equipment
obsolete. One of the most promising coal energy technologies
is fluidized bed combustion (FBC) of coal. This particular
technology is a combination of the invention and development
of 1) pulverized coal for firing in utility boilers and 2)
Fritz Winkl'er's Fluidized Bed Combustor. Although both de-
velopments occurred in 1921, they were not immediately com-
bined. Pulverized coal rapidly became widely used in the
utility industry, whereas FBC of coal did not receive exten-
sive research and development until the 1950's.
The combination of the two processes did not occur at the
time of their invention because the preexisting solid bed coal
combustors were readily modified to handle pulverized coal.
Also, Winkler's FBC invention was developed in a different
context, which was to produce fuel gas by partial combustion
of coal. Recently, spurred by restrictive regulations on sul-
fur dioxide (S02)/ nitrogen oxides (NOX), and particulate
emissions, FBC of coal research and development has gained
momentum. The FBC process is well suited to burn high sulfur
coal without producing high SC>2 and particulate emissions in
the flue gas.
The FBC process combusts coal (or other fuel) at a tem-
perature of 800° to 900°C (1,500° to 1,700°F), in a
bed of granular non-combustible material fluidized by air.
The bed is made up of granulated dolomite and/or limestone,
which absorb the sulfur to a level where the boiler flue gas
sulfur dioxide emissions are reduced below the minimum re-
quired by regulations. Particulate matter is controlled by
electrostatic precipitators or baghouses. Heat from the
fluidized bed is transferred to coils immersed in the bed, and
water is converted to steam as it is pumped through the
coils. The pressurized steam is used either to drive tur-
bines to generate electricity or provide heat for various pro-
cesses, or is used for other purposes.
This section discusses FBC's major processes, possible
environmental contaminants, and applicable pollution control
technologies.
PROCESS DESCRIPTION
Basic Principles
For effective combustion of solid fuels, a combustor de-
signer can manipulate three major interrelated variables by
which adequate contact is achieved between solid fuel and
oxygen, by providing (Farmer et al., 1977):
Large solid fuel surface area
14
-------
Long contact time between gas and solid particles
High relative speed between gas and solid parti-
cles, so:that fuel is not shielded from oxygen by a
thick layer of stagnant burned gases.
Combustion in a conventional spreader stoker emphasizes
the second and third of these variables, while pulverized coal
firing depends chiefly on the first and third. Combustion in
a fluidized bed can take advantage of all three variables,
thus achieving an unusually high intensity of heat release in
a small combustion volume.
Basically, a "fluidized bed" combustor consists of a
chamber in which granular, non-combustible particles are sus-
pended by air forced through the "bed" from the bottom of the
chamber containing holes in the floor. The mass is called
"fluidized," because, under proper air pressure and flow
rates, the bed exhibits characteristics of a boiling liquid
(Yaverbaum, 1977). Depending on the purpose of the combustor,
the chemical composition of the fuel, and environmental con-
straints, the bed material can be inert (i.e., sand), chemi-
cally reactive (limestone), or a mixture of materials.
Pulverized coal particles are injected into the fluidized mass
at a constant rate to maintain a two to five percent fraction
of carbon in the bed mass.
When the fuel is fired, the heat is transferred to the
non-combustible particles which transfer the heat by con-
duction to the boiler tubes immersed in the "fluidized bed."
The process of conduction rather than convection permits
higher rates of heat transfer than a conventional pulverized
coal combustor. FBC pilot plants have achieved heat transfer
rates ten times higher than conventional combustors. Heat
transfer rates o£ over 100,000 Btu/hr/ft3 of expanded bed
volume or about 50,000 to 60,000 Btu/hr/ft3 of firebox have
been recorded. A typical pulverized coal combustor generates
about 20,000 Btu/hr/ft3 of firebox. Consequently, FBC
boilers up to about 250,000 Ib. steam/hr can be designed small
enough for package shipment by rail. By comparison, a conven-
tional boiler that can be package-shipped by rail can generate
only about 50,000 Ib. steam/hr. This package feature should
make the future erection cost of large-sized FBC boilers sig-
nificantly lower than for corresponding field-erected
stoker-fired or pulverized coal-fired units.
Efficient heat generation and transfer means that FBC
units can operate at lower temperatures—about 650° to
950°C (1,200° to 1,800°F)—than conventional pulverized
coal combustion units, which operate at 1,400°C (2,500°F)
and higher. At these lower temperatures significant amounts
15
-------
of NOX are not generated. With additional attention to de-
sign, NOX production can be further minimized.
Addition of a chemically active bed material can serve to
lower gaseous emissions of other pollutants by trapping them
in the bed. For example, high-sulfur, coal-burning, conven-
tional pulverized coal combustion units require extensive,
expensive, and often unreliable flue gas desulfurization sys-
tems to remove S02 from the waste gas stream to meet current
emission standards. FBC units, by chemically trapping the
sulfur in the bed during combustion, produce an easily handled
stable solid. In a simple form, the sulfur trapping reaction
can be described as:
MeO + 1/2 02 + S02 « * MeS04 (1)
MeO (the sorbent) represents a metal oxide chosen for
desirable physical and chemical properties and for economic
reasons. Bed conditions are maintained to favor sulfation of
the metal oxide to form the metal sulfate. The metal oxide-
sorbent-feed rate is determined in accordance with the sulfur
content of the fuel, the reactivity of the metal oxide, and
the emission standards. The sulfur removal is sufficient so
that further treatment is not needed.
Another valuable characteristic of FBC technology is its
wide tolerance to type and quantity of solid fuels. Caking
and non-caking coals, cokes and chars, and solid wastes can
all be burned efficiently in a fluidized bed. It is antici-
pated that FBC boilers will be effective burners for direct
combustion of anthracite culm and low-Btu oil shale, j
j
One of the main operating parameters is uniform introduc-
tion of fuel and SC>2 acceptor at multiple points within the
bed, with generally one fuel/acceptor feed point specified for
about ten square feet of bed area. Not only is the average
bed temperature relatively low, but temperatures throughout
are uniform if good fluidization is maintained. The absence
of hot spots means that fuels having low-softening point ash,
can be effectively burned without serious ash fusion or
clinker formation.
While the FBC concept is relatively simple, the design of
a practical, operating unit can be complex. One of the major
problems is scaling up design parameters from laboratory and
pilot plant sizes to full-scale combustors. At least nine
physical parameters/variables influence the design and thereby
the efficiency and pollutant output: bed depth, bed and
boiler tube geometry, fluidizing grid design, sorbent and coal
16
-------
particle size, fluidization velocity, excess air feed rate,
fuel feed mechanism, pressure and temperature. Other vari-
ables introduced through government regulations prescribe cer-
tain operational modes and pollutant controls, sorbent prep-
aration economics, solid waste disposal, and tax incentives
and disincentives.
Consideration of these variables has promoted the devel-
opment of FBC process design variations. The major design
variants are:
• Combustor operating at or near atmospheric pressure
in the combustion chamber—basic application is in
smaller sizes, primarily for generation of indus-
trial process or space heat or steam.
Pressurized combustor operating at several atmos-
pheres of pressure—most promising for electrical
utilities; power is generated through the expansion
of high pressure off-gases flowing through a tur-
bine; the efficiencies are five to ten percent
higher than those of the atmospheric FBC units.
Absence of cooling units in the bed—Columbus
Battelle Laboratory's researchers envision an FBC
design where the heat transfer tubes are located
outside rather than inside the fluidized bed; bene-
fits include decoupling of combustion and heat ex-
change processes, reduction in combustor size,
reduction of corrosion in the heat exchanger tubes,
and improvement of load response.
High fluidizing air volumes—in this mode, excess
air under pressure rather than boiler tubes is used
to extract the heat of combustion; electricity is
generated by flue gas turbines followed by waste
heat recovery units.
Regeneration of the metal oxide sorbent—regenera-
tion of sorbent material minimizes the solid waste
problem and produces a potentially marketable prod-
uct.
Figure 2 shows several selected FBC system options.
Based on current development efforts, the two major cate-
gories of FBC processes are Atmospheric Fluidized Bed Combus-
tion (AFBC) and Pressurized Fluidized Bed Combustion (PFBC),
both chemically active. The literature and researchers have
treated a number of the AFBC and PFBC variants as separate
major categories, e.g., Pressurized, Combined Cycle FBC of
coal (adiabatic combustor) (Abelson and Lowenbach, 1977) .
17
-------
CO
I. USER
2. PURPOSE
3. PROBABLE TYPICAL SIZE.
I06 Btu/hr FUEL INPUT
1|. OPERATING PRESSURE
5. EXCESS AIR, PERCENT
6. CYCLE TYPE
7. FUEL-TO-ELECTRICITY
EFFICIENCY
8. REGENERATION
9. REGENERATION PROCESS
10. FINAL S REMOVAL
INDUS
STEAM CE
5
ATMOS
15
PKOCES
..X N.
NO
FLUIDIZED BED
COMBUSTION SYSTEMS
1
TRIAL UTILITY
JE RAT ION ELECTRIC GENERATION
10 5000
1
1 1
•HERIC ATMOSPHERIC PRESSURIZED
(10 ATM)
|
25 15-25 15-25
STEAM TURBINE (801)
> STEAM STEAM TURBINE AND GAS TURBINE (20)
A. 36-38 36-38
1 1
1 1 1 1
YES NO YES NO YES
1
1-STEP I-STEP 1-STEP
1
300 (AD1ABATIC)
GAS TURBINE (80t)
AND WASTE HEAT
) BOILER- STEAM
TURBINE (20%)
31-38
YES
1
2-STEP PRESSURIZED
1
S HjSO^ S02 S H^O^
RECOVERY RECOVERY SCRUBBING RECOVERY RECOVERY
PROBABLE Co:S MOLE RATIO IN FEED TO COMBUSTION IS 2:1. COAL SULFUR CONTENT MAY BE A PERCENT IN
INITIAL SYSTEMS. PROBABLE OPERATING TEMPERATURE OF THE MAIN COHBUSTOR IS 850-950*C.
Figure 2. Selected FBC system options.
(Fennelly el al.. 1977)
-------
This section will deal primarily with the two major categories
of FBC of coal—the Pressurized and Atmospheric.
Atmospheric Fluidized Bed Combustion of Coal
Experts agree that AFBC units will establish themselves
in the market first to meet industrial needs for power and
steam. The major breakthroughs will be in retrofitted and
converted units (Patterson and Griffin, 1978). Babcock Con-
tractors Inc. (Pittsburgh, Pa.) and Riley Stoker Corp.
(Worcester^, Mass.) entered in a joint agreement in 1978 to
manufacture and market AFBC boiler units in the United
States. The first commercial units are expected to be
operating in late 1979 in Ohio.
Figure 3 is a schematic diagram of an atmospheric pres-
sure FBC boiler (Farmer et al., 1977). The bed consists of a
mixture of crushed limestone, dolomite or other inert mate-
rial, and large ash particles. This mixture is "fluidized" by
a stream of air or combustion-gases rising through the sup-
porting grid beneath the bed. Original particle size of the
bed material is about 1/8" diameter. The gas velocity is
maintained so that the bed particles are suspended and move
about in random motion, but do not blow away. Under these
conditions, a gas/solids mixture behaves much like a boiling
liquid (e.g., seeks its own level, and can be readily moved
through channels). The boiler tubes submerged in the bed re-
move heat at a high rate so that typical bed temperatures are
in the range of 750° to 870°C (1,400° to 1,600°F).
Crushed coal (1/4" to 1/2" particles) and the required
bed makeup material are continuously added at fuel injection
points. Within the bed, the coal burns very quickly, and the
bed generally contains less than two-to-four percent carbon.
Most of the ash resulting from combustion of the'coal is in
relatively small, light particles which are swept out of the
bed by the flue gas. If high sulfur coal is being burned,
sulfated bed material is continuously withdrawn to maintain
bed volume and activity for sulfur capture. If low sulfur
fuel is being burned, it is assumed that the bed can be com-
posed of inert material, such as alumina or sand, and that bed
makeup and withdrawal rates will be very low or negligible.
FBC can effectively control emissions from high-sulfur
fuels. For this purpose, the sorbent bed is made up of lime-
stone, dolomite, or lime. Assuming a limestone feed, the
first reaction at bed temperature is calcination:
CaC03 «, CaO + C02 (2)
19
-------
PRIMARY
CYCLONE
SECONDARY
PARTICIPATE REMOVAL
CONVECTION
SECTION
WATER
WALLS
BAFFLE
TUBES
ASH; PARTICULATES
PREHEATER, SUPERHEATER
OR REHEATER SECTION
AIR
LIMESTONE
Combustion Pressure - close co atmospheric (usually balanced draft)
Bed Temperature - I'tOO to 1600 F
Gas Velocity - 2 to 12 ft/tec
Bed Material - limestone or dolomite, or inert material for low sulfur fuels
not requiring sulfur capture
Fuel - coal, fuel oil, bark and wood wastes, coke, char, etc.
Figure 3. Schematic diagram of atmospheric pressure FBC boiler.
20
-------
The sulfur in the fuel burns to sulfur dioxide, and bed condi-
tions are maintained to favor sulfation of the lime to
gypsum:
CaO + SO.
CaSO,
(3)
The limestone sorbent feed rate is set in accordance with the
sulfur content of fuel, and sufficient cleanup of S02 is
achieved so that no further treatment is required for com-
pliance with sulfur emission regulations. By contrast, a
stoker or pulverized-coal-fired boiler emits most of the sul-
fur in the fuel as S02, and high sulfur fuels cannot be
burned in these boilers without desulfurization of the flue
gas using some sort of scrubber.
Considerable work is underway to develop sorbent regener-
ation. The incentive for this is to reduce the requirements
for fresh limestone or dolomite, and to reduce spent stone
disposal costs. In the sorbent regeneration process, the sul-
fur absorbed by the stone will be released as a more concen-
trated stream of S02, and subsequent disposition of this
stream will be by conversion to sulfur or to sulfuric acid.
The basic reactions in the regeneration process are
(Newby and Keairns, 1977):
to acid plant
at atmospheric pressure
CO
CaS0
CaO + SO,
H2°
to
air
pre-
heater
(4)
1200° - 1900°F
(650° - 1040°C)
The reducing gases are produced by the incomplete combustion
of coal.
The regenerator off-gas enriched with S02 is piped to
the acid plant for removal of particulate matter. This regen-
eration process is called reductive decomposition and, based
on performance and cost criteria, is the most favorable for
atmospheric FBC processes (Newby and Keairns, 1977).
21
-------
The above reactions of the sorbent regeneration process
represent an idealized chemical process without the considera-
tion of the presence of other active constituents in the sys-
tem such as ash. Sorbent regeneration, as an integral
component of the FBC technology, provides two essential advan-
tages: 1) minimizes Environmental impact by recycling the
sorbent and removing S02r 2) enhances the technology's
economics by reducing the need for raw materials and producing
sulfur, a saleable product. However, it is questionable
whether the first commercial-scale AFBC units will incorporate
sorbent regeneration. The 'state-of-the-art of FBC processes
is rapidly advancing but sorbent regeneration research and
development lags behind. Although new uses for sulfur are
being continuously developed (i.e., new batteries and as an
asphalt additive) overproduction of sulfur is a possibility
and may change the sulfur recovery economics.
Pressurized Fluidized Bed Combustion of Coal
Current projections show the PFBC of coal processes to be
costly, with costs increasing as units decrease in size. PFBC
units are expected to be commercialized for power utility
applications, probably in co-generation applications and as
large utility boilers. The PFBC processes have a much greater
range of operating conditions, higher gas turbine pressures,
alternative cbmbustor temperature ranges, and higher excess
air levels. This greater process flexibility makes PFBC more
applicable for large-scale electrical power generation instal-
lations. Although full-scale commercialization of PFBC is not
expected until 1990, significant R&D efforts are currently
underway to develop these processes.
A schematic diagram of a pressurized FBC boiler (with
j sorbent regeneration) is given in Figure 4 (Farmer et al.,
* 1977). .The higher operating pressures of PFBC achieve higher
thermodynamic efficiencies for electric power generation than
Is possible with an AFBC pressure unit. This is accomplished
by use of a combined cycle, generating power from both a
steam-driven turbo-generator and a gas turbine. The combustor
typically operates at a pressure on the order of ten atmos-
pheres and at temperatures ranging from 900° to 1,040°C
(1,650° to 1,900°F). Air is compressed into the com-
bustor, and coal and limestone are injected through lock
hoppers. Because of the higher pressure, combustion inten-
sities are even higher than with AFBC, and deeper fluidized
beds can be used. Hot effluent flue gas from the boiler is
cleaned to reduce the particulate loading to a very low level,
and then expanded through the gas turbine to generate supple-
mental electricity. High temperature particulate removal is
currently a major research area.
22
-------
to
u>
GAS TURBINE
TO WASTE HEAT
RECOVERY AND STACK
TO SULFUR
"RECOVERY
SEPARATOR
DISCARD
COAL AND
MAKEUP SORBENT
A I
FUEL
REGENERATOR
Figure 4. Schematic diagram of pressurized FBC boiler with sorbent regeneration.
-------
PFBC gives significantly lower SOX emissions than the
atmospheric version when burning high-sulfur coal.
The applicable sorbent regeneration process is a two-step
reaction of sulfate reduction to sulfide followed by reaction
with steam to generate H2S (Newby and Keairns, 1977):
(1) CaS04 + H2 + 3CO ^ * CaS + ^2° + 3C°2
(5)
8 atm
1500°F
(876°C)
to acid plant
1
(2) CaS + C02 + H20 .(steam) * CaCOg +2
12 atm reuse
1100°F
(593°C)
The H2S rich gas is fed to an acid plant for extraction
of sulfur.
There are a number of different sorbents; however, based
on thermodynamic projections of desulfurization and regenera-
tion performance, dolomite and limestone (CaO/MgCa [003]2
CaCOs) appear to be the best and most cost-effective choices
for AFBC (Newby and Keairns, 1977). Table 3 shows the alter-
native sorbents and their projected potential.
ENVIRONMENTAL POLLUTANTS FROM FBC OF COAL
Methods of Determining Pollutants
Theoretical and empirical methods can be used to deter-
mine what pollutants will be generated from fluidized bed
combustion of coal. At this stage of FBC development, how-
ever, both methods provide only approximations for future com-
mercial-scale FBC facilities.
The theoretical method involves the calculation of a
materials balance for all the process streams of the FBC of
coal. The materials balance concept is based on the law of
conservation of matter which states that atoms are neither
created nor destroyed, except in nuclear reactions. The ap-
propriate equations for any system are:
24
-------
TABLE 3. ALTERNATIVE SORBENT SELECTION
(Newby and Keairns, 1977)
to
Ul
Desulfurizer
potential
Sorbent
Na2C03
CaO
SrCO
BaCO
LiA102
LiFeO
Li2Ti03
NaAlO
NaC03-Fe203
CaAl204
SrAl204
SrTiO
BaAl.,0,
2 4
BaTiO
Atmos-
pheric
Limited
temp.
range
Good
Good
Good
Good
Good
Good
Limited
temp.
range
Limited
temp.
range
Good
Good
Good
Good
Good
Pressur-
ized
No
Good
Good
Good
Good
Good
Good
No
No
Good
Good
Good
Good
Good
Reduction Decom-
position
potential
Atmos-
pheric
Limited
so2
Good
No
No
Good
Good
Limited
so2
Good
Limited
Good
Good
Good
Good
Good
Pressur-
ized
No
Limited
so2
No
No
Limited
so2
Limited
so2
No
No
No
Good
Good
Good
Good
Good
2-Step regenera-
tion potential Avail-
Atmos- Pressur- ability
pheric ized potential
Good No Good
Limited Good Good
Good Good Poor
Good Good —
Good Good
Limited Limited
Good Good
Good No Good
Good No Good
Good Good Good
Good Good Poor
Good Good Poor
Good Good
Good Good
Cost
potential Overall
Limited appli-
cability; highly
corrosive nature
Good Good potential
— Poor availability
— Reasonable poten-
tial
Poor Poor cost poten-
tial
Poor Poor cost poten-
tial due to
limited perform-
ance
Poor Poor cost poten-
tial
— Limited applica-
bility; highly cor-
rosive nature
— Limited applica-
bility; highly cor-
rosive nature
Good Reasonable poten-
tial
— Poor availability
— Poor availability
— Reasonable poten-
tial
— Reasonable poten-
tial
-------
Accumulation of total atomic total atomic
atomic species J = species J entering - species J (7)
: within the system system leaving system
By summation of ail atomic species, the materials balance
is calculated
Ma = Me - MA (8)
where:
Ma = total mass accumulating in system
Me = total mass entering the system
mJl = total mass leaving the system
The materials balance takes into consideration the nature
and composition of coal and the sorbent materials and the
chemical reactions and thermodynamics of the process. This
approach cannot account for all the changes in operating^
conditions in a commercial-size process and subsequent dif-
ferences in chemical reactions and thermodynamics producing
different system by-products.
For most elements, the information available will not
permit such calculations. First, the necessary data are
either not recorded or go unreported. In addition, spark
source mass spectrometry (SSMS) as commonly used is notori-
ously unreliable. In one case, a 40 percent relative standard
deviation was reported (Carpenter et al., 1978) and in
another, values were reported to be within an order of magni-
tude, leading to mass closures varying from zero to 1,000
percent (Pillai and Roberts, 1977). With appropriate care and
the proper analytical technique, usually atomic absorption,
closure values nearing 100 percent can be obtained even for
the volatile trace elements (Pillai and Roberts, 1977). For
the initial Battelle work, "The emphasis . . . was mainly on
establishing a general methodology for comprehensive samp-
ling—analysis from FBC units and not to define the precision
and/or accuracy of the techniques employed" (Merryman et al.,
1977). This emphasis is apparent through all Level 1 analy-
ses.
The second approach is empirical. Bench scale, pilot
plant, and demonstration plant test data on environmental
contaminants can be compiled and extrapolated to fit com-
mercial-scale facilities. However, the environmental contami-
nant data from existing experimental FBC projects are highly
variable and not necessarily representative of a com-
mercial-scale operation. The amount and nature of test data
26
-------
obtained under closely controlled conditions from special pur-
pose projects can completely misrepresent the FBC process when
extrapolated to a commercial size facility.
Combining the theoretical and empirical methods will
overcome some of their shortcomings and can provide a rea-
sonably sound basis for predicting what environmental contami-
nants can be expected from commercial-scale FBC processes.
Sources and Major Types of Pollutants
The major sources of pollutants are the principal unit
operations of the FBC process. They are listed below, with
their associated major pollutants:
Sorbent (limestone) preparation and feed unit
operations
air emissions (fine particulates)
leachates
Fuel preparation - coal storage, drying, crusting,
and feeding unit operations
air emissions (particulates, fugitives)
- leachates
solid wastes (pyrite)
Coal combustion in fluidized bed unit operations
- air emissions (S02, NOx' particulates,
HCX, trace elements)
solid wastes (ash leachates, fugitive emis-
sions)
Sorbent regeneration unit operations
- spent sorbent discard
by-products (sulfur recovery - acid plant)
Steam turbine and water treatment unit operations
thermal discharges (air and water)
liquid discharges
sludge
• Air pollution control unit operations
- steam
- flue gas emission (particulates, trace ele-
ments, HCX)
solids (sludge, ash)
Figure 5 is a schematic of the major unit operations of a
generic FBC process; major process waste streams are numbered
and identified in Table 4. Figure 6 is a generic process flow
of AFBC of coal and the associated environmental waste
streams. While PFBC of coal process flows are generically
27
-------
similar to AFBC, the thermodynamics of 'the two processes
differ markedly due to the pressurization of the fluidized bed
combustion chamber and the higher system flow rates.
TABLE 4. STREAM DESIGNATIONS FOR GENERIC UNITS
(Abelaon and Lowenbach, 1977)
Stream No. Designation
1 Stack Gas
2 Particulate Removal Discard
3 Bed Solids Discard
4 Particulate Removal Discard—Regeneration Operations
5 Other Effluents from Regeneration and Sulfur Recovery Operations
6 Slowdown from Steam Turbine Cycle
7 Slowdown from Water Treatment Operations
8 Product from Sulfur Recovery (Sulfur or Sulfuric Acid)
9 Raw Fuel to Preparation
10 Raw Sorbent to Preparation
11 Intake Water to Treatment
12 Air to Combustor
13 Air/Steam to Regenerator
14 Prepared Fuel Feed to Combustor
15 Prepared Fuel Feed to Regenerator
16 Start-up Fuel Feed
17 Prepared Sorbent Feed to Combustor
18 Bed Solids to Regenerator
19 Flue Gas to Particulate Removal
20 Makeup Water to Steam Turbine Cycle
21 Recycle from Particulate Removal
22 Recycle from Regeneration and Sulfur Recovery
• 23 Gas to Gas Turbine Inlet (Pressurized FBC Only)
Environmental Pollutants
Based on the analysis of the raw coal, numerous potential
pollutants can be predicted and their concentration ranges in
waste streams estimated. Fennelly et al. (1977) , in calcu-
lating the potential pollutants from FBC processes, cate-
gorized the potential pollutants into the following major
categories:
Acids and Acid Anhydrides
Organic Acids - Compounds such as carboxylic
acids, dicarboxylic acids, and sulfonic acids
could conceivably form from incomplete com-
bustion of hydrocarbons; however at the
28
-------
ro
IX)
©
PFBC
ONLY
REMOVAL
OPERATIONS
t i
water 1 I
from various i
plant sources *
water
to various
plant locations
FLUIDIZED
BED
COMBUSTOR 22
SORBENT REGENERATION
AND
SULFUR RECOVERY
OPERATIONS
©
©
Figure 5. Generic schematic of FBC unit operations with numbered
process flows producing waste streams.
-------
/ AIR .
( EMISSIONS X
\FUGITIVES '
U)
O
HOT
ELECTROSTATIC
PRECIPITATOR
EFF-0.96+
FROM
BED MATERIAL
STORAGE
VENT
( EMISSIONS »
\FUGITIVES I
/SOLIDN
/ WASTE \
|(LEACHATES.I
BOILER
1350 F (ATM
VENT TO
ELECTROSTATIC
PRECIPITATOR
' WATER N
^EFFLUENTS )
\
SOLID WASTE X
(LEACHATES. \
\FUGITIVE EMISSIONS ,
TO FLUIDIZED
BED BOILER
/• WATER^,
^EFFLUENTS/
FEEDWATER
Figure 6. Schematic diagram of emission sources within
atmospheric-pressure fluidized bed combustion plant.
(Henschel. 1977)
/ WATER N
^EFFLUENTS;
-------
temperatures involved (~850°C), compounds
of this type should quickly decompose to form
small hydrocarbons and
Inorganic Acids - The predominant inorganic
acids should be HC1 and HP. Sulfuric,
sulfurous, nitric and nitrous acid should not
form until the flue gas has cooled to tem-
peratures less than 240°C.
Carbon Compounds
The major carbon compounds, as expected, will be
C02 and CO. Soot (solid carbon) could be a prob-
lem since it would be emitted as small particles
(<0.1ym) which could pass through most parti-
culate control devices. Calcium and magnesium car-
bides could also form in small quantities in the
ash formed in the combustor, the carbon burn-up
cell or the regenerator. After ash disposal, these
compounds could release acetylene upon contact with
water. The quantities generated, however, should
pose no special problems.
Halogen Compounds
Halogens should be emitted primarily as HX (where X
F, Cl, Br) . Experience in coal combustion chemis-
try indicates that the presence of chlorine in coal
enhances condensation reactions via elimination of
HC1; hence, species such as chlorinated hydro-
carbons are not favored. Furthermore, most
chlorinated aliphatics are unstable at the tempera-
tures prevailing in the combustor.
Hydrocarbons
Long chain aliphatics and cyclic hydrocarbons
should decompose within the bed to form H2 and
smaller hydrocarbons. Some of these smaller hydro-
carbons could condense to form polycyclic species
such as pyrene, anthracene, etc. Hydrocarbon con-
centrations are strongly dependent on the amount of
excess air. With excess air levels of about 20
percent, total hydrocarbon levels (measured as
methane) of less than 100 ppm are attainable.
Based on a comparison with data from conventional
combustion systems, 1 ppb of benzo (a) pyrene (or
similar compounds) could be present in fluidized
bed combustion. With excess air levels of 10
percent, benzo (a) pyrene could reach 10 ppb.
31
-------
Nitrogen Compounds
The predominant nitrogen compound should be NO.
Trace amounts (< ppm) of HCNr (CN)2 and
azoarenes may be present. Species such as amines,
pyridine, pyrroles, and nitrosamines, which could
form as combustion intermediates, should quickly
decompose within the bed to form hydrocarbons, NO
and HCN.
Oxygen Compounds
Oxygenated hydrocarbons such as furans, ethers,
esters, aldehydes, etc. could form as combustion
intermediates. These species, however, are un-
stable at temperatures on the order of 850°C, and
they should decompose before escaping from the
bed. Ozone, if formed, would also decompose at
these temperatures.
Particulates
. . . Although there are only very limited data
available on particulate concentrations and parti-
cle size distributions, it seems that no special
problems should result from particulate loadings,
provided the conventional process cyclones and a
control device such as an electrostatic precipi-
tator or a fabric filter are used. The actual per-
formance of these devices on full-scale fluidized
bed boilers, however, has not yet been tested.
Radioactive Isotopes
Based on a "worst case" analysis with lignite coal,
uranium could be emitted at significant levels, but
the radioactive isotopes of uranium as well as
those of other radioactive elements should not be
present in high enough quantities to cause concern.
[See Table 5 for actual measurements.]
Sulfur Compounds
S02 and 803 will be the major sulfur compounds
formed. The presence of limestone in the bed, how-
ever, should keep these emissions below present
emission standards. COS could form in trace quan-
tities (~1 ppm). Compounds such as thiophenes or
mercaptans, if formed as combustion intermediates,
will decompose to hydrocarbons and H2S; the H2S
will then be oxidized to form S02-
32
-------
TABLE 5. RADIOACTIVITY MEASUREMENTS'
Sample description
Radioactivity pico Curies/gin
Test 6/la
Bed material
Coal
Dolomite
Primary cyclone dust
Secondary cyclone dust
Test 6/2a
Bed material
Coal
Dolomite
Primary cyclone dust - coarse
- fine
- total
Secondary cyclone dust - coarse
- fine
- total
Flue gas dust - coarse
- fine
Total in flue gas dust
Total in flue gas
pci/kg gas
18.8
5.0
5.8
8.2
14.8
8.2
10.6
2.4
19.6
9.0
19.4
14.0
5.0
11.3
22.2
7.0
20.7
4.9
+ 4.0
+ 2.4
+ 2.4
+ 3.2
+ 4.0
+ 3.2
+ 3.2
+ 2.4
+ 4.0
+ 3.2
T 4.0
+ 4.0
+ 2.4
T 3.5
+ 5.0
+ 7.0
+ 5.2
46.8
9.0
8.2
44.4
69.4
24.6
16.0
27.4
55.0
61.2
55.1
84.6
94.8
87.7
111.8
78.9
108.5
25.1
+ 6.0
+ 5.2
+ 5.2
+ 6.0
+ 6.6
+ 5.4
T 5.4
+ 5.4
+ 6.4
+ 6.4
+ 6.4
T 7.0
+ 7.2
+ 7.1
+ 7.2
+ 23.7
+ 8.9
Exxon
Flue gas
6 + 3
30 + 10
Pillai and Roberts, 1977.
Murthy and Henschel, 1978.
33
-------
Trace Elements
Data should be acquired with respect to the fate of
trace elements such as Be, As, Pb, V, Ni and Cl.
Also needed :are data with respect to chemical com-
position as 'a function of particle size. For the
most part, the other trace elements commonly en-
countered in coal combustion are captured in the
coarse solids and remain in the bed. The high pH
of the bed material is advantageous in that it
tends to retard the leaching of the heavier metals.
Fennelly and others have also estimated the expected con-
centrations in the flue gas and the solid waste. Their esti-
mates follow:
Flue Gas
One hundred parts per million: CH4, CO, S02, NO
(100 ppm)
Ten parts per million:
(10 ppm)
One' part per million:
(1 ppm)
S03,
C2H6, HC1
HF, HCN, NH3, (CN)2,
One part per billion:
(1 ppb)
Oner-tenth part per billion:
(0.1 ppb)
HN03, Elemental
Vapors, As, Pb, Hg ,
Br, Cr, Ni, Se, Cd, U Be,
Na
Diolefins, Aromatic Hydro-
carbons, Phenols, Azoarenes
Carboxylic Acids, Sulfonic
Acids, Alkynes, Cyclic
Hydrocarbons, Amines,
Pyridines, Pyrroles,
Furans, Ethers, Esters,
Epoxides, Alcohols, Ozone,
Aldehydes, Ketones,
Thiophenes, Mercaptans,
Chlorinated Hydrocarbons
Flue Gases and Particulates
Table 6 shows the reported ranges for some of the minor
flue gas constituents found in FBC research installations.
From the results, it appears that PFBC produces less pollu-
tants than AFBC. The combustion efficiency also appears to be
greater in pressurized units than in atmospheric, based on the
34
-------
TABLE 6. REPORTED RANGES OF SOME MINOR STACK FLUE GAS CONSTITUENTS
(Values in ppm)
Compound Exxon
or Element
Leatherhead
fed
MERC Battelle Estimated
worst case
SO.
0-4
0.14-18.6 200
Cl
33.5-47.1
18
37.3-43.5 10
6.7-14.5
2.5
0.082-0.70
HCN
<0.0004-0.001 2-3.5
<0.05-<0.18 0.07-0.18
NH.
0.06-0.64
0.13
1.8-7.05
CO
53-130
100-400 790-4790g 100
HC
3.3
1-13
85-900
CH-100
, COS
<4.3
Allen et al., 1978a.
Pillai and Roberts, 1977.
Merryman et al., 1977.
Fennelly et al., 1977.
Murthy and Henschel, 1978.
Allen et al., 1978b.
Merryman et al., 1978.
35
-------
outputs of CO and HC. Chloride output is similar in both
types of operation and above the estimated maximum emission.
Fluorine emissions will probably be highly variable
depending upon the presence of excess CaO and the stack tem-
perature at particulate clean-up. At operating bed tempera-
tures, fluorine will be released as HF which will then combine
with any excess CaO below 550°C and MgO below 450°C in the
following reaction:
MeO + 2HF ^ » MeF2 + H20 (9)
Argonne National Laboratories noted that fluorine
retention was 56 to 62 percent with a calcium sorbent compared
to 5 to 23 percent with a sorbent free bed (Alvin et al.,
1977).
Table 7 shows collated information on the emission of
eight toxic volatile elements found in the flue gas after some
small particulate removal. The output of lead is dependent
upon several variables, including the tendency of lead com-
pounds to preferentially deposit upon small particulates, the
total concentration of chlorine in the system, and pressure.
At high temperatures and pressures, gaseous PbCl4 is the
thermodynamically favored compound but, at high temperatures
and low pressures, gaseous PbO is favored. Upon cooling
and/or lowering the pressure, PbO will precipitate out quite
rapidly. In addition, conversion of lead to PbCl4 is re-
lated to the concentration of chlorine in the gas. If the
chlorine concentration falls to 50 percent of its assumed
value, then PbCl4 falls to 6 percent of its previous value
(Alvin et al., 1977; 1978). In conventional combustion, lead
preferentially precipitates upon smaller particulates and
there is some evidence that it behaves the same in FBC (see
Table 8). Thus, lead emission in the exhaust gas is favored
by a low chlorine coal, or, alternatively, a high alkali re-
lease which will scavenge the chlorine, and efficient removal
of the fine particulates.
Mass balance information permits emission calculations as
shown in Table 9. Selenium and mercury behave in fluidized
bed operations as they do in conventional combustion, probably
because the elemental form in both cases is thermodynamically
preferred as in the following reactions:
36
-------
TABLE 7. VOLATILE ELEMENT ANALYSIS IN FLUE GAS
Element
Antimony
Arsenic
Beryllium
Cadmium
Lead
Mercury
Selenium
Tellurium
w
Battelle"
0.039
<0.026
<0.052
<1.3
0.009
<0.039
<0.26
e d
MERC:: Exxon Leatherhead"
' u'g/m yg/m ppb
<0. 27-0. 48 1.2-l.sJ
0.12-0.72 0.8-2.11
<0.4
0.1
<1.2
0.13-0.51 0.50-1,27
<1.4
<1.7
0.4
3.0
0.6
1.1
5.4
3.9
20
Trace
* All analysis done by atomic absorption or optical emission spectroscopy.
b Merryman ot al., 1977.
0 Allen et al., 1978b.
Murthy and Henschel, 1978.
* Pillai and Roberts, 1977.
f Allen et al., 1978a.
TABLE 8. PROJECTED FLUE GAS EMISSION OF TRACE ELEMENTS AS A PERCENTAGE
OF THE ELEMENT ENTERING THE SYSTEM
(Pillai and Roberta, 1977)
Element
Mercury
Selenium
Arsenic
Lead
Cadmium
Tellurium
Beryllium
Antimony
Emissions (% of input)
Leatherhead Exxon
81 80
51
5 15
10 0-20
14
4
8 20-40
5
Conventional
combustion
90 - 100
30" - 70b
SO - 60
0-60
75C
50°
10C
75°
Andrcn et al., 1975.
Kaakinen et al., 1975.
DeAngelis and Reznik, 1979.
37
-------
TABLE 9. POTENTIAL ENRICHMENT OF ELEMENTS IN THE FINE PARTICULATE CATCH
Element
Pb
Tl
Yb
Ho
Gd
Eu
Nd
Pr
Cs
I
Te
Sb
Sn
Cd
Nb
Y
Sr
Br
Se
As
Ge
Ga
Zn
Cu
Ni
CO
Mn
Cr
V
Sc
Cl
P
Na
F
B
Be
Li
Class
II
I
IV
II
II
I
III
II
II
II
II
II
IV
I
I
IV
IV
I
III
IV
a
Morgantown
I
PD
I
PE
I
I
PE
PE
PD
I
I
I
PE
PE
I
PE
PE
PE
I
I
PE
I
PE
PE
PE
PE
PE
PE
ND
b c
Morgantown Battelle
PE
I PE
I
I
I
I
I
I I
ND ND
I
I PE
I
I
PE
ND ND
PE
PE
I
PE
PE
PE
I
I
I
I
I
I
PE
PE
PE PE
Exxon
PE
PE
PE
PE
PE
PE
PE
PE
ND
ND
PE
ND
PE
PE
PE
PD
PE
PE
PE
I
PE
I
PE
PE
PD
I
I
PE
PE
PE
PE
Conven-
G -f-
Leatherhead tional
E E
N
E E
E
N
G
E
E E
E
E
E
S
N
N
S
S
N
G
S
E
Carpenter et •!., 1978.
' Allen «t »!., 1978b.
Merryman et al., 1978*.
Allen tt •!., 1978a.
Filial and Roberta, 1977.
Klein et al., 1975.
KKi CLASS I - llthophllaa - aluminosilieatas -
little volatility
II - chalcophile - sulfides - highly
volatile
III - gases
IV - unclassified - appear to have
properties between Classes I t II.
E - enrichedt measured by atonic
absorption
G - gas
I - Indeterminate
N - not volatile
NO - not determined
PD - probably depleted SSMS
PE - probably enriched SSMS
38
-------
Se02 + 2S02 ^ » Se° + 2S03 (10)
H2SeO + 2S03 + H20 -^—, „ Se° + 2H2S04 (11)
HgCl2 -* Hg + C12 (12)
From thermodynamic data (Alvin et al., 1977), beryllium
would be expected to volatilize without deposition upon cool-
ing, thus showing a high emission. This, however, is not
demonstrated by the available data. Also, from the same theo-
retical evaluation, boron and chlorides would be emitted in a
gaseous form. While chloride emissions are higher than ex-
pected in some cases, no adequate information is available,
nor is any information available for boron.
Arsenic is another element not behaving as predicted from
conventional coal combustion. There is some evidence which
indicates that it is being preferentially deposited upon small
particulates (see Tables 8 and 9) . Cadmium, tellurium, and
antimony also do not behave as predicted.
Without additional mass balance information, the fate of
any element cannot accurately be determined. However, a ten-
tative prediction can be made by establishing a concentration
ratio where:
(Concentration) after combustion
(concentration).^ =• (concentration) * before combustion
After statistically accounting for carbon loss during
ignition of the coal, the elemental values of the small parti-
culate catch are compared to their values in the coal, while
39
-------
the spent bed material is compared to the limestone. Simpli-
fying assumptions ignore the contributions of the elements in
the limestone and consider that none of the bed material will
be carried into the fine particulate catch, while no fly ash
remains behind in the bed. (Leachate information supports
this with the bed reject leachate having a pH of 12 and the
second cyclone leachat'e having a pH of 9.) The assumptions
are, of course, debatable considering that the alumino-
silicate-associated elements are not volatile. However, since
at FBC operating temperatures these elements do not melt to
form a fused slag, they could be carried from the bed as fly
ash. Table 9 shows a qualitative description of the available
enrichment information. Considering the ambiguities of SSMS,
only the most tentative conclusions can be drawn.
Generally, in both pressurized and atmospheric combus-
tion, Class II, sulfide-associated elements behave as
expected, based on pulverized coal combustion (PCC) tech-
nology. That is, they are initially volatilized and then
recondensed on the small fly ash particles. One exception is
cadmium which appears to be found in all waste streams. From
the Leatherhead data, it would appear that antimony is not re-
leased as a gas, contrary to PCC expectations. But it is the
only combustor for which data indicate a concentrating effect
on the small particulate catch. Thallium, another toxic ele-
ment appears enriched in one case and depleted in another.
It would also appear that the Class I elements are being
removed from the bed, preferentially concentrating on the
small particle catch, contrary to their behavior in PCC
units. Considering that SSMS is the source of the data, this
conclusion may be inaccurate. Alternatively, because of the
temperature, the physical behavior of the elements might
differ, in FBC units, from PCC units.
Too little data are available to accurately predict the
fate of the Class III and IV elements. Superficially, it
would appear that there is a difference between Pressurized
and Atmospheric FBC units, but more must be known.
To extract additional energy in the PFBC system, the hot
exhaust gases will be expanded through a turbine. However, if
the gases contain alkali metals, hot corrosion of the turbine
blades will likely occur. Sodium-chloride-doped coal and
charcoal suggest an inverse emission relationship with ash
content (Swift et al., 1977). The scavenging is probably a
function of the clay minerals in the coal for which the fol-
lowing reaction is typical:
Al,0, 'SiO- + 5Si00 + 2KC1 + H00 r 2KAlSi_00 + 2HCL (14)
O J £. £. £ ^ J O
40
-------
The calculated effect of the aluminosilicate is to suppress
about 98 percent of the potassium emissions and about 86 per-
cent of the sodium emissions at 925°C and ten atmospheres
(Alvin et al.,1977). In experiments, thermally activated
bauxite captured 98 percent of vaporized sodium chloride
although alundum ( a- A1203) did not trap any salt (Swift
et al., 1977). Addition of an aluminosilicate mineral to the
bed will probably be necessary in PFBC combustion.
A Level 2 analysis of the second run from the Battelle
miniplant has begun characterizing the inorganic compounds
found in the flue gas particulates (Table 10). The presence
of S"2 in the particle interior but not in the surface is
noteworthy; it might imply that the center of the particles
never reach a complete oxidizing condition. The inability to
detect barium and nickel ions in the particulate leachate or
by instrumental analysis of the particle is interesting.
SSMS data (Merryman et al., 1977) report their presence as "00
and 200 ppm, respectively, in the second cyclone removal sys-
tem. Perhaps the technology is insufficiently sensitive, SSMS
is inappropriate, or the elements are present as insoluble
compounds (BaC03, BaS04, NiC03 or NiO).
With the scarcity of information, the inadequacy of the
analytical technique and the lack of runs where only a single
variable was changed, little can be said about elemental frac-
tionation during different operating conditions. It would be
desirable, for example, to have reported information from a
series of runs where the only variable was the amount of lime-
stone added for sulfur sorption. It would be expensive but
desirable. Considering the economic cost of limestone and the
cost of disposal, it is only reasonable to predict that the
calcium-to-sulfur ratio of limestone to coal will approach 1.1
to 1. At least two changes from the current conditions can be
predicted: less calcium oxide will be present in the waste
stream to cause thermal problems upon disposal and less
fluoride will be trapped in the small particulate catch as it
is now by the excess CaO.
A number of known polycyclic carcinogens have been iden-
tified in the flue gas of fluidized bed combustors (Table
11). While solid carbon has not specifically been determined
in the coarse dust, carbon concentrations are usually very low
(see Table 23) and this is reflected in the low polynucleated
aromatic hydrocarbon (PAH) concentration of the removal sys-
tem. The small particulates of the Battelle AFBC unit have
higher concentrations of PAH material. This could reflect the
tendency for PAH material to deposit on the fines, a dif-
ference between AFBC and PFBC operation, or a difference in
stack temperatures at collection. The higher concentrations
of PAH material in the flue gas may depend on the same factors
as the fine particulates.
41
-------
TABLE 10. INORGANIC ELEMENTS AND COMPOUNDS IDENTIFIED IN FLUE GAS PARTICULATES
(Ryan et al., 1979)
NJ
IR
>27p <27p
>27V
ESCA
<27y
Oxidation Atom Oxidation Atom Percent
state percent state
CaSO4 SiO2
CaSO4-l/2H2O CaS04
CaSO4 • 2H2O CaSO4 •
SiO2
Fe203
Fe304
XRD
>27p <27U
SiO2 Si02
Fe2O3 CaS04
CaSO4 CaS
CaS Fe2°3
CaO
Fe*3
O~2
2H2O Ca*2
C ,
S*6
Si*2
Al*3
Element
Ca
S
O
Si
Fe
K
Al
Mg
Cl
Ti
2 Fe*3
62 0~2
4 Ca*2
C
2 Na*1
22 S*6
8 Si*2
Al*3
EMPA
Relative abundance %
>27p <27p
5-10 >10
5-10 >10
5-10 >10
>10 >10
1-5 5-10
5-10 5-10
1-5
1-5
1-5
1-5
Surface Interior 100 $J
1 1
61 53
5 10
-
2 1
7 5 S04-2.S~2 6:4
13 16
12 16
Elements and compounds not detected
Be SrF, PO"3
MgF Ba** TiO3
MgCO3 Tl FeO2
CaF2 P Ru
Interior 200 §
2
53
9
-
1
3 S04~2,s-2 5.5:4.5
17
14
in flue gas particulates
Co*2 Ni*3 ca
Co*3 Th
Rh U
Ni*2 Rb
KEY: IR = Infrared
ESCA = Electron Spectroscopy for Chemical Analysis
XRD = X-ray Diffraction
EMPA = Electron Beam Microprobe Analysis
-------
TABLE 11. POLYCYCLIC ORGANIC MATERIAL ANALYSIS
(Values in ng/g or ng/m where appropriate)
OJ
Particulates
Cyclone
Battelle3
Anthracene/phenanthrene <0 . 24
Benzo (b) f luoranthene
Benzo (a) pyrene
Ben zo ( e ) py rene
Benzo (g , h , i ) pery lene
Benzo (c) Phenanthrene
Chrysene/benz (a) anthrancene
Coronene
Dibenzo (a,h) anthracene
Dibenzo (a , h) carbazole
Dibenzo ( c , g) carbazole
Dibenzo (ai & ah)pyrenes
7 , 12-Dimethylbenz (a) anthracene
Fluoranthene
I ndeno ( 1 , 2 , 3 , cd) py r ene
Methyl anthracenes
Methyl benzopyrene
3-Methyl cholanthrene
Methyl chrysenes
Methyl pyrene/f luoranthene
Perylene
Pyrene
*value less than
2 or coarse dust
Leatherheaa Exxon
0.13 <0.3-4
*
*
*
*
*
*
*
*
*
*
*
* <0.3-0.4
*
* <0.2
*
*
*
*
*
* <0.2
0.1
Flue gas
Small particles
a
Battelle Exxon
<27y <3p
429.3 <0.3-14
31.2
19.5
3.9
9.8
39 .
*
' *
*
*
175.4 <0.3-2
3.9
39.0 <0.2
3.9
*
5.8
7.8
*
0.7
Leatherhead
exhaust dust
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
0.2
Battellea
2667-57300
8.8
3.5
3.5
14-170
35.1-370
*
*
*
*
*
1404-3200
3.5
561-9400
1.8
*
3.5-170
28.1-730
*
211-950
0.4
Exxon Leatherhead
53 93
1.0 *
0.5 *
*
*
0.2 3
3.8 3
*
*
*
*
*
26 21
*
5 10
*
*
<0.7
1.0 7
*
9 9
0.1
Merryman et al.f 1977.
i
Pillai and Roberts, 1977.
Murthy and Henschel, 1978.
-------
Two other carcinogens, dimethyl and monomethyl sulfate,
have been found in the ambient air around a power plant burn-
ing low-sulfur, high-ash coal (Eatough, 1979). The methylated
sulfate concentrations may be as high as 1,000 ppm on
respirable particles less than one micron, with an atmospheric
residence time of hours or days. The decomposition point of
dimethyl sulfate is 188°C which means the material must be
formed downstream of the combustion chamber. The material
could be formed in the following reaction:
SCO + S02 + 3H20 . (CH30)2S02 + 3C02 (15)
This finding must be further examined.
Dow Chemical Company has also reported the presence of
dioxins in combustion products from fireplaces, refuse in-
cinerators, automobiles, trucks, charcoal grilles, cigarettes
and fossil-fueled power plants. Based on this, the EPA is
initiating a testing program covering all emission sources,
stack, fly ash and wastewater streams in incinerators and
power plants to verify the findings and to determine if regu-
latory activity is warranted (Environment Reporter, 1979).
This finding must also be checked for FBC units.
Actual test data on FBC effluents obtained from bench
scale, pilot plant, and demonstration scale processes shows
that pollutant concentrations vary. As shown in Tables 12 and
13, investigation has focused on flue gas and spent bed and
ash reject material. Caution and restraint must be used in
extrapolating this experimental data to commercial-size FBC
units.
The types and quantities of pollutants formed and dis-
charged from FBC processes depend on many FBC system
variables:
Combustion temperature
Combustion pressure
Fluidized bed height and freeboard in combustion
chamber
Type of coal(impurities)
Fuel feed rate
• Air flow rate
Sorbent feed rate
Superficial velocity
Ca/S mole ratio
Sorbent and fuel particle size
• Pollution control equipment
Mode of operation
Fugitive releases during process breakdowns
44
-------
TABLE 12. INORGANIC ANALYSIS OF PARTICULATE EMISSIONS IN FLUE GAS
(Murthy et al., 1977)
Size range
Substance 1-3 (microns) 3-10 (microns)
Major elements, yg/g
Al 200,000 200,000
Fe 60,000 20,000
Si 200,000 200,000
K 3,000 1,500
Ca 30,000 30,000
C (Total 12,000 11,000
carbon)
Anions, weight percent
Cl~
F~
Co =
i
S°3=
S»
NO *
N02~
0.011
0.031
<0.2
0.001
<0.03
<0.001
<0.001
0.007
0.032
<0.2
0.004
0.03"
<0.001
<0.001
A major group of pollutants, fugitive emissions, are
site-specific and do not lend themselves to generic treat-
ment. These pollutants originate from vents, loose system
components (seals, connections, valves, etc.), careless mate-
rials handling, coal and sorbent storage piles, and other
sources. Airborne fugitive emissions associated with
coal-fired operations commonly originate from piles of stored
coal, combustion residues from furnace cleaning, and collected
particulate matter from air pollution control equipment. The
piles of material are stored outside and are subject to wind
action. Water-borne fugitive emissions usually originate by
leaching of feed and residue storage piles, subsurface dis-
posal, surface runoff, and atmospheric washout of combustion
products not controlled at the site.
The significance of fugitive emissions, in the context of
all other emissions from FBC operations, is relatively un-
defined. They are usually caused by careless and poorly main-
tained operations. Proper materials handling, enclosed
short-term storage of feed stock and residues, and good house-
keeping can probably eliminate any significant fugitive
emissions.
45
-------
TABLE 13. TRACE METALS IN BED REJECT MATERIAL
(Murthy et al.f 1977)
Element
Li
Be
B
F
Na
Mg
Al
Si
P
S
Cl
K
Ca
Sc
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Ga
Ge
As
Se
Br
Rb
Sr
Y
Zr
Nb
Mo
Ru
Rh
Pd
Observation
(ug/g and Percent)
200
0.3
30
0.3
300
20%
1.5%
1.5%
50
1.5%
40
1%
20%
3
200
20
30
100
1%
1.5%
150
15
400
7
40
20
3
3
60
300
1.5
0.5
2
2
<1
<0.5
Element
Ag
Cd
In
Sn
Sb
Te
I
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
W
Re
OS
Ir
Pt
Au
Hg
Tl
Pb
Bi
Tn
U
Observation
(yg/g and Percent)
<0.1
<0.5
<0.6
2
1
<0.4
<0.07
4
30
7
7
1
1.5
0.5
3
1
0.3
<0.1
<0.3
<0.1
<0.3
<0.3
<0.3
<0.4
<0.4
<0.4
<0.4
<0.2
<0.4
<0.1
<0.4
<0.2
2
<0.15
0.2
0.4
46
-------
The fugitive emissions generated during process breakdown
or during malfunction of pollution control equipment can be
readily predicted by estimating what untreated waste streams
produce. Generally only treated waste discharges are moni-
tored, but sufficient data are available on the efficiency of
contaminant-removal of pollution control equipment so that raw
emissions can be back-calculated. Fugitive emission rates
then can be estimated with relatively close accuracy.
Solid Wastes and Leachates
As currently envisioned, the use of FBC for industrial
and utility applications will utilize a once-through-and-dis-
card cycle with respect to the material used as a sulfur ac-
ceptor (probably limestone or dolomite). The basic PBC unit
produces no aqueous discharges; however, disposal of the spent
bed material is likely to result in leachate which could con-
taminate surface or groundwater systems if improperly dis-
posed. For FBC to be functional, additional processes such as
heat transfer systems or electric power generators are re-
quired.
Heat transfer systems and electric generating systems,
however, are potential sources of water pollutants. The pol-
lutants are derived from the system blowdown and water treat-
ment chemicals used. Another source of potential pollutants
common to all facilities using coal is leachate formed by dis-
solution of soluble species in coal by rainwater and melting
snow. In addition to coal pile leachate, FBC operations could
generate leachate from stored limestone or other sulfur ac-
ceptor material.
Figure 7 is a simplified materials flow illustration,
indicating potential sources of aqueous pollutants. Depending
on the materials handling and storage procedures at various
plants, the amounts and type of pollutants discharged may vary
considerably. All plants are expected to utilize coal as the
primary fuel and either limestone or dolomite as the sulfur
absorbent. The coal and sorbent may be stored in the open or
in enclosed silos, as is done at the 30 MW units at Georgetown
and Rivesville. Large, utility-size units would be expected
to store both coal and limestone in the open primarily because
of the quantities that must be kept on hand. Typically, a
coal-fired utility plant would maintain a 75 to 90-day coal
supply (Federal Power Commission, 1974a) which amounts to 810
tons per megawatt.
Whenever coal is stored exposed to the weather, drainage
will occur. Coal pile drainage is typically acidic, high in
suspended .and dissolved solids, high in iron, sulfur,
manganese and hardness. Ranges of concentration of pollutants
47
-------
from coal piles from two studies are summarized in Table 14.
The amount and quality of the leachate depend on several
factors, such as amount of precipitation/ duration, intensity,
etc.; turnaround time of piles; and quality of coal.
i
Sorbent Leachate
Leachate from exposed limestone or dolomite piles will
have a pH generally in excess of nine, and be high in sus-
pended solids. Sodium may also be present in significant
amounts (Stone and Kahle, 1977) . Like the coal pile leach-
ates, the quality of sorbent leachates will vary with the
amount of water and with time of contact of water and sorbent.
Boiler Slowdown
Chemical additives are used to prevent corrosion, control
pH and scale formation, and prevent solids deposition in steam
generators, heat transfer lines, and boilers. The chemical
additives used are summarized in Table 15. Factors which
determine the quality of boiler water blowdown include:
Boiler design pressure
• Chemicals used in water treatment
Quality of water used in boiler
In addition to the chemicals used in water treatment,
boiler blowdown commonly contains calcium and magnesium scale,
iron and copper oxides, and sometimes elemental copper.
Cooling System Effluents
Two types of cooling systems are in common use, once
through cooling systems and recirculating cooling systems. For
the former, the primary pollutant is waste heat. The chemical
composition of the effluent is essentially the same as the
influent water; the solids may be slightly higher due to
evaporation, introduction of corrosion products, and use
of biocides. Commonly used biocides include chlorine and
sodium or calcium hypochlorite. Cooling water is generally
chlorinated by shock methods wherein large doses are added
intermittently.
Recirculating cooling systems may utilize cooling towers,
ponds, or canals to collect cooling water, dissipate heat and
allow re-use of the water. Since evaporation is the major
cooling mechanism, water vapor is a major atmospheric dis-
charge. Along with the water vapor, small amounts of organic
and inorganic water contaminants and chemicals used in water
treatment may become airborne in the form of aerosols.
48
-------
SLOWDOWN
WATER
SLOWDOWN
STEAM
TURBINE
CVCU
JN
NT
3NS
JN
t
-
1 '
COOLING
WATER
DISCHARGE
Figure 7. Scheme of FBC (materials flow).
49
-------
TABLE 14. POLLUTANTS FROM LEACHING OF COAL PILES9
Pollutant
Concentration
Concentration
Phosphorus
Ammonia
Nitrate
Alkalinity
BOD
COD
P«
Acidity
Conductivity
Chloride
Sulfate
Dissolved Solids
Suspended Solids
Iron
Manganese
Silica
Copper
Zinc
Chromium
Aluminum
Nickel
Calcium
Magnesium
Lead
Mercury
Barium
Arsenic
Cadmium
Selenium
Titanium
Beryllium
Antimony
1
2.3-3.1
300-7100
2400-6400
0-660
1800-96,000
270-16,000
8-2300
23-1800
1.8-45
1-390
0.07-1.4
1.1-16.6
<0. 005-0. Oil
20-440
0.15-4.5
31-720
17-480
<0.01
<0. 0002-0. 0072
<0.1-0.5
0.006-0.36
<0. 001-0. 003
<0. 001-0. 030
<1.0
<0. 01-0. 07
<0.1-1.8
0.2-1.2
0-1.8
0.3-2.3
15-80
3-10
85-1100
2.1-7.8
10-22,000
4-480
130-19000
250-29,000
20-3300
0.17-4700
1.6-1.8
0.001-12.5
0-15.7
a All aeasurenents in tag/I except conductivity (nmnos/cm) and pH.
b tox et al., 1979.
c Nichols, 1974.
TABLE 15.
CHEMICAL AGENTS ASSOCIATED WITH BOILER TREATMENT
(Sugarek and Sipes, 1978)
Control
objective
Candidate chemical additives
Residual concentration
in boiler water
Scale
Corrosion
PH
Solids
Deposition
Di- and tri-sodium phosphates
Ethylene diaminetetracetic
acid (EDTA)
Nitrilotriacetic acid (NTA)
Alginates
Polyacrylates
Polymethacrylates
Sodium sulflte and catalyzed
sodium sulfite
Hydrazine
Morpholine
Sodium hydroxide
Sodium carbonate
Ammonia
Morpholine
Hydrazine
Starch
Alginates
Polyacrylamides
Polyacrylates
Polymethacrylates
Tannins
Lignin derivatives
3-60 mg/t as PO4
20-100 mg/t
10-60 »g/t
up to 50-100 mg/l
up to 50-100 mg/l
up to 50-100 mg/t
less than 200 mg/t
5-45 mg/l
5-45 mg/l
Added to adjust
boiler water ph
to the desired
level, typically
8.0 - 11.0.
20-50 mg/l
20-50 rag/I
20-50 mg/l
20-50 mg/l
20-50 mg/t
«00 mg/l
£200 mg/t
50
-------
Chemicals are added to recirculative cooling systems to
prevent or control corrosion, pH, and biological growth. Be-
cause of evaporation, the treatment chemicals and contaminants
native to the water used are concentrated. These pollutants
are periodically removed in the blowdown, at a rate ranging
from 0.5 to 3 percent of the recirculating water flow
(Donahue, 1970). Some of the major chemicals used for water
treatment in recirculating cooling systems are summarized in
Tables 16 and 17.
Additional potential pollutants created in the system may
result when chlorine, added to prevent biological growth,
reacts with hydrocarbon contaminants in the water and forms
undesirable chlorinated hydrocarbons. Another source of water
contaminants may be contact of the water with the atmosphere;
in some cases, 80 percent of the suspended solids in the
cooling water has been attributed to atmospheric contami-
nants. Soluble gases may also contaminate the cooling water.
Sorbent Regeneration
When the process of sorbent regeneration is commercial-
ized, it may produce some water pollutants. The process is
essentially a dry process; however, the off gas containing
concentrated levels of sulfur dioxide may require an acid
rinse prior to reduction to sulfur. Sulfuric acid would be
the expected rinsing acid. Ultimately, the acid would become
contaminated and require renewal. The acid discharge would be
expected to be highly contaminated with both inorganic and
organic species and with suspended solids.
In addition-to the specific process effluents, other
sources of water pollutants which may be derived from opera-
tion of an FBC power generating or energy conversion plant in-
clude:
Equipment cleaning
General plant discharge
Process spills and leaks
Miscellaneous sources such as laboratory or machine
shop wastes
All of these sources, individually or in combination, may
create potentially toxic or hazardous aqueous pollutants. The
wastes from the various processes may be treated on site and
the effluent discharged to a sewer or surface water, or the
wastes may be directly discharged to a sewer system. If
treated on-site, these aqueous waste streams would most likely
be combined. Since some of the wastewater streams are acidic,
and others basic, the combined wastewater will be at least
partially neutralized. Removal of organic and inorganic
51
-------
TABLE 16. CHEMICALS USED IN RECIRCUIATIVE COOLING WATER SYSTEMS
(Federal Power Comission, 1974b)
Chemical
Corrosion inhibition or scale
prevention in cooling towers
Biocides in cooling towers
pH control in cooling towers
Dispersing agents in cooling
towers
Biocides in condenser cooling
water systems
Organic phosphates
Sodium phosphate
Chromates
Zinc salts
Synthetic organlcs
Chlorine
Hydrochlorous acid
Sodium hypochlorite
Calcium hypochlorite
Organic chromates
Organic zinc compounds
Chlorophenatas
Thiocyanates
Organic sulfur compounds
Sulfuric acid
Hydrochloric acid
Lignins
Tannins
Polyaerylonitrile
Polyacrylamide
Polyacrylic acids
Polyacrylic acid salts
Chlorine
Rypochlorltes
Sodium pentachlorophenate
TABLE 17. COOLING TOWER CORROSION AND SCALE INHIBITOR SYSTEMS
(Suprenant et al., 1976)
Inhibitor system
Concentration
of chemical additives
in recirculating water
Chroma te
Chromate + zinc
Chronate + zinc
+ Phosphate (inorganic)
Zinc + Phosphate
(inorganic)
Phosphite (inorganic)
Phosphate (organic)
Organic Biocide
200 - 500 mg/i CrO4
17-65 mg/1 CrO4
8-35 mg/i Zn**
10 - 15 mg/1 Cr04
8-35 mg/i Zn'1"1'
30 - 45 mg/i P04
8-35 mg/i Zn**
15 - 60 mg/l PO4
15 - 60 tng/i PO4
15 - 60 mg/i P04
3-10 mg/l organics
30 mg/i chlorophenol
5 mg/i sulfone
1 mg/i thiocyanate
52
-------
species from the wastewater streams may require some form of
flocculation, precipitation, and sedimentation. Possible
treatment chemicals may include soda-lime combination or spent
bed material. The effluents from the wastewater treatment
process and the resultant sludge may include as contaminants
any of the chemicals or elements listed in Tables 16 and 17.
The accumulated sludge and other solid wastes may be
transferred to landfills for ultimate disposal. The solid
wastes which may require disposal include:
Spent bed materials composed primarily of calcium
and/or magnesium carbonates, sulfates and oxides
Particulate matter from cyclones or baghouses
Particulates, spent bed material or bottom ash from
carbon burnup
If the plant is sufficiently large that regeneration of
sorbent is economically practicable/ the only major sources of
solid wastes requiring disposal would be the particulates cap-
tured by cyclones, baghouses, or other techniques. A similar
situation would occur if the spent bed material were
transported to another facility which specialized in sorbent
regeneration and sulfur recovery.
Whether the spent bed material and the captured flue
particulates, as well as wastewater treatment sludge, are
individually disposed or combined for ultimate disposal, the
major environmental contaminants will result from leachate
formation. The leachate formed within a landfill may contami-
nate surface water and/or groundwater. Groundwater may feed
surface waters and vice versa, resulting in cross-contamina-
tion. Representative concentrations of elements in spent bed
material leachate are summarized in Table 18. The leachate
data for the Exxon Miniplant is taken as the highest concen-
trations measured on several samples by two methods of pro-
ducing leachate. The values presented for the Level 2 analy-
sis are the sum of acidic and basic leachate concentrations.
Table 19 summarizes leachate data obtained from flue gas
particulates. The data from the Exxon plant represent data
obtained from leaching of particulates captured by the second
of two cyclones installed in the exhaust gas line. The data
from the Level 2 analysis represent leachate from captured
flue gas particulates greater than 27 microns.
Table 20 presents data on leachate from Pressurized and
Atmospheric FBC unit solid wastes. These analyses, performed
by the Westinghouse Research and Development Center (Sun and
53
-------
TABLE IB.
REPRESENTATIVE CONCENTRATIONS OP ELEMENTS IN LEACHATE OP PBC
SPENT BED MATERIAL*
Element
Lithium
Beryllium
Boron
Fluorine
Sodium
Magnesium
Aluminum
Silicon
Phosphorus
Sulfur
Chlorine
Potassium
Calcium
Scandium
Titanium
Vanadium
Chromium
Manganese
Iron
Cobalt
Nickel
Copper
Zinc
Gallium
Exxon*5
1,000
1.5
300
6
15,000
20,000
3,000
25,000
3,000
1,500,000
2,000
15,000
1,000,000 1
< 2
1,000
100
200
30
3,000
5
500
150
<40
5
Level"
560
, N.D.
31,000
1 N.D.
17,000
1,400
7,400
260,000
N.D.
N.D.
N.D.
7,200
,300,000
N.D.
N.D.
< 20
N.D.
N.D.
800
N.D.
170
14
220
N.D.
Element
Arsenic
Selenium
Bromine
Rubidium
Strontium
Ytterium
Zirconium
Molybdenum
Rhodium
Silver
Cadmium
Mercury
Thallium
Lead
Tin
Antimony
Tellurium
Iodine
Cesium
Barium
Lanthanum
Cerium
Neodymium
Tungsten
Exxon
150
100
300
300
700
7
30
400
< 1
10
< 2
< 2
< 2
20
100
1
6
7
20
700
70
1
4
20
Level0
N.D.
N.D.
N.D.
N.D.
390
N.D.
N.D.
60
N.D.
4
N.D.
-------
McAdams, 1979) , compare FBC solid waste leachate data to regu-
lations being promulgated under the Resource Conservation and
Recovery Act (RCRA). The leachate tests were performed in
accordance with the RCRA procedure. The material tested in-
cluded spent bed material, carryover (solids captured by the
second cyclone), and fines (solids captured by a third
cyclone, baghouse, or other filter). Pertinent anion data
from some leachate studies on FBC solid wastes are summarized
in Table 21.
Biological Examination
Comprehensive biological examination of the waste streams
from FBC plants is just beginning. The fine particulate
stream from the Exxon miniplant had a positive Ames Test and
showed cytotoxicity with rabbit alveolar macrophages.
Leachate from the second cyclone catch exhibited toxicity to
saltwater organisms, and, along with bed reject material
leachate, exhibited toxicity to freshwater organisms (Murthy
and Henschel, 1978). In a preliminary examination, the fine
particulate matter from the Argonne combuster also showed a
positive Ames Test but only under startup conditions. At
steady state conditions, the fine particulate catch appeared
to be non-mutagenic (Norris, B., Argonne National Laboratory,
to Blair, T., personal communication). Argonne researchers
are also exposing mice to a diluted exhaust gas stream. The
preliminary results indicate low immediate toxicity which
might correlate with the negative results on the plant
ethylene stress test for the Exxon unit. Further examination
is in order and caution is warranted in waste disposal.
POLLUTION CONTROL
Background
One of the major advantages of FBC over conventional pul-
verized coal combustion processes is that it burns high-sulfur
coal and control emissions of S02 and NOX in the combus-
tion chamber. As stated previously, the limestone is cal-
cinated to produce lime (CaO) in the fluidized bed at tempera-
tures of 750 to 950°C.(1,400° to 1,800°F). The combus-
tion of coal releases the sulfur in the bed in the form of
sulfur dioxide (S02) which reacts with lime (CaO) to produce
calcium sulfate (CaS04). The calcium/sulfur molar feed
ratio is about 3:1 for limestone (CaCOs) and reduces S02
emissions by over 90 percent. For dolomite (MgCa[C03l2)
the ratio is 2:1. The sulfur extraction is sufficiently high
to reduce the S02 emissions to meet the air quality stan-
dards for coal-fired boilers. Because of lower combustion
temperatures in the fluidized bed, NOX gas formation is sup-
pressed below the level requiring additional pollution control
systems.
55
-------
TABLE 20. POM COMPOUNDS IN LEACHATES FROM AFBC AND PFBC COMBUSTION OF COAL
SOLID WASTES
1
Component
Anthracene/Phenanthrene
Methylanthracenes
Fluoranthene
Pyrene
Methylpyrene/Fluoranthene
Benzo (c) phenanthrene
Chrysene/Benz(a) anthracene
Methylchrysenes
7 , 12-dimethylbenz (a) Anthracene
Benzo fluoranthenes
Benz (a) pyrene
Benz(e) pyrene
Perylene
Methylbenzopyrenes
3-Methylcholanthrene
Indeno (1 , 2 , 3-cd) pyrene
Benzo (ghi) perylene
Dibenzo(a,h) anthracene
Dibenzo(a,h) carbazole
Dibenzo(ai & ah)pyrenes
Coronene
Leatherhead
yg/kg in solid waste
Bed reject Primary Secondary Exhaust
material cyclone cyclone dust
<0.1 0.1 0.13 <0.2
<0.1 3 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
2.4 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
<0.1 <0.1 <0.1 <0.2
Pillai and Roberts, 1977.
TABLE 21. ANION ANALYSIS OF LEACHATE OF FBC SOLID WASTES
(Allen et al., 1977)
Wastes
Cyclone
#2 Dust
anion - yg/1
Cl F SO,
S0
Bed Reject 176,000 4,000 3,000,000 3,100 8,300 1,900
Material
65,000 2,400 2,500,000 2,300 6,700 71,000
-------
Basic problems—extent of particulate control, toxic
trace elements, and gaseous organic pollutants—are still
relatively undefined. However, experimental evidence (Murthy
et al., 1977), shows that particulates, certain trace ele-
ments, and certain gaseous organic compounds are emitted in
sufficiently high concentrations in the flue gas to warrant
installation of pollution control equipment. Experimental
data and theoretical considerations indicate that conventional
pollution control equipment used on coal-fired systems will be
applicable to FBC operations, with a similar level of con-
trol. The basic classes of applicable air pollution control
equipment are:
Cyclones (single and multiple stage)
Fabric and granular bed filters
Catalytic and thermal oxidizers
Electric precipitators (single and multiple stage—
high volt and low volt)
Scrubbers
Combinations of these categories of equipment can be de-
signed into pollution control systems that would allow FBC of
coal processes to meet the New Source Performance Standards
(NSPS).
The solid wastes and possible secondary water contami-
nants present a set of different problems and need to be
considered in detail within the context of a total waste man-
agement system for coal combustion and conversion processes.
Discharges from atmospheric and pressurized FBC processes
are emitted under different conditions (i.e., temperature,
pressure) and therefore may require different pollution con-
trol system design. However, the general categories of pollu-
tion control equipment will be applicable to both Atmospheric
FBC and Pressurized FBC processes.
Applicability of different pollution control equipment
depends on the physical size and reactivity of the individual
contaminant particles. Figures 8 and 9 show the relationship
of contaminant particle size and the types of pollution con-
trol equipment applicable.
One form of pollution control that must not be overlooked
is coal pretreatment and fractionation. For example, density
fractionation of coal from the Upper Freeport Coalbed, Garett
County, Maryland, indicates that removal of the sink 1.60
fraction would reduce the mercury concentration by 47 percent,
sulfur 42 percent, chromium 29 percent, copper 43 percent,
nickel 28 percent and manganese 50 percent. Other coals, of
course, show different removal patterns. Cadmium in the coal
57
-------
01
00
INGSTROM MICRON KETER
1 108 I008 • 10008 1 IOM 100M 1HM ICH 10CM \f
—
1 i
CAS MOLECULES — Oj
SMALL
1
S°2 N°x H2S OTHERS
ORGANIC MOLEC
FINE
ULES. ODORS
LARGI
SMOKE
SOR
ORGANIC MOLE
SULFUR 1C A
IENT BED PARTI
CULES
CID MIST
PAINT PIG
FOG.
CLES
1ENT
CLOUDS, STEAM
FLY AS
H
GRINDING DUS1
r
CUPOLA EMM I SS IONS
RAII
4
Figure 8. Air pollutant particle sizes.
-------
Ln
vo
ANGSTROM
I IDS
toooS
MICRON
1
IOM IOOM
IrtM
I CM
METER
10CM IM
GR
CHEHI
•m
AtflTY CHAMBER'
CAL REMOVAL-T
— '
1ERHAL OX I DAT I
|
ON. CATALYTIC
LIQUI
COMMON
OXIDATION
D SCRUBBERS
ELECTRIC PR£
AIR FILTERS
CIPITATORS
CYCLONES
Figure 9. Emission control equipment chart.
-------
from Hazard No. 4 Coalbed, Bell County, Kentucky, is evenly
distributed throughout the various density fractions (Schultz
et al., 1975). The cadmium is probably attached to the
carbonaceous material rather than associated with the pyritic
material as are the other elements. The economics of lime-
stone mining, calcination, and spent bed material disposal
versus the amount of removable sulfur will be the major deter-
mining factor for coal pretreatment.
Pollution Control of Flue Gases
General Pollution Control Applications—
Theoretical calculations (Fennelly et al., 1977) and ex-
perimental data (Murthy et al., 1977) show that the major
gaseous contaminants from FBC process flue gas occur in rela-
tively manageable concentrations and emission rates. Conven-
tional commercially available air pollution control devices
and systems can reduce the contaminant emissions from FBC
processes to meet air pollution standards.
Particulate matter emitted by FBC processes can also be
readily removed by commercially available cyclones and elec-
tric precipitators with sufficiently high efficiencies to
comply with current air quality standards. Although a frac-
tion of particulates from FBC processes will be generated from
the fluidized bed media, the sorbent (limestone and/or dolo-
mite) and the general nature of the particulate handling
systems will be nearly the same as for the conventional coal-
fired boilers, except that the highly reactive CaO particles
may have to be stabilized before or during collection. This
similarity will probably place the FBC processes under the
same air quality compliance requirements. The contaminant re-
moval efficiency of each category of equipment varies greatly
with the conditions of FBC operation and the commercial design
of the pollution control equipment. The range of contaminant
removal efficiency of the general types of equipment is given
in Table 22.
A major concern in the pollution control process is to
remove particulates that are less than 3 microns. These
particulates are of much concern because they are less effi-
ciently captured by the particulate control devices, have a
long atmospheric residence time, are deposited efficiently in
the deep lung and only slowly removed, and because the
smallest particles concentrate the most toxic elements. In
the lung, serum can solubilize up to 80 percent of the trace
elements on the particle versus at most 15 percent in the
stomach (Piperno, 1975). Polycyclic organic matter can be
deposited on the particles depending upon the temperature at
which the particulate collection occurred. In one study, an
electrostatic precipitator operating above 100°C collected
60
-------
TABLE 22. MAJOR CONTAMINANTS IN THE FLUE GAS OF FLUIDIZED BED COMBUSTION
OF COAL AND APPLICABLE POLLUTION CONTROL EQUIPMENT
Contaminants
CO
HC
H-SO . (mist) + SO,
srf
NH,
CNJ
Cl
NO
PartiSulates 100 to 1000 microns
Particulates 10 to 100 microns
Particulates 1 to 10 microns
Particulates 1 micron _
Trace elements and compounds
Filters3
X
X
X
X
X
X
X
X
Effective and
Scrubbers
X
X
X
X
X
X
X
X
X
X
applicable pollution control systems
Thermal or1- ^
catalytic Electric
oxidation precipitators Cyclones
X
X
X
X
X
X
X
XX X
Filter efficiency and applicability greatly depends on the filter media - fabric type, granular, etc. Gravel
bed filters can be used for sulfur removal.
Scrubber efficiency and applicability is dependent primarily on the scrubbing liquid used and the pressure
drop.
The need for thermal and catalytic oxidation can be eliminated for CO and HC by increasing combustion ef-
ficiency in the burner.
Electric precipitators can be low volt, high volt, single and multiple stage, depending on the efficiency
desired.
Cyclones remove most particulate matter 100 to 1000 microns.
Trace elements will be removed in fractions by any of these systems in conjunction with other contaminants.
-------
large and small particles, with no mutagenic activity; yet, at
95°C, stack samplers after the precipitator collected muta-
genic particulates (Fisher et al.f 1979). The presence of
particulates along with the PAH material exacerbates the car-
cinogen problems in that the fine solids can cause damage by
themselves or they can act as a cocarcinogen in the two-stage
carcinogenesis model (National Academy of Sciences, 1972).
Also, the fine particulates can combine with S02 to form a
toxic, hazardous material in the atmosphere.
Carbon Utilization—
Efficient carbon utilization is important in reducing
potentially toxic carbon compounds, maintaining thermal effi-
ciency, and in reducing capital and operating costs. The car-
bon efficiency of a coal combustor/boiler can be defined as:
Carbon in Coal - Carbon Lost x 1QO
cu Carbon in Coal
For FBC units, efficiencies near 100 percent can be ob-
tained only if the large particle fly ash trapped in the
particulate removal systems is recombusted. This requires
either a carbon burn-up cell operating at high temperatures
(1,1000C [2,000°F]), or that the fly ash be reinjected
into the fluid bed. Typical combustor carbon losses are shown
in Table 23 for a PFBC. Particulate reinjection into the bed
is relatively simple for an AFBC unit, but a combustor such as
the multi-solid FBC variant requires more sophisticated re-
cycling to achieve high carbon utilization efficiencies (Liu
et al., 1979).
TABLE 23. DISTRIBUTION OF COMBUSTIBLE LOSSES
(Hoy and Roberts, 1977)
Source Typical range
%
Carbon monoxide -- 0.04 - 0.10
Carbon content of
(a) bed material 0.05 - 0.20
(b) 1° cyclone dust 1.00 - 5.00
(c) 2° cyclone dust <0.01 - 4.00
62
-------
Hydrocarbons (HCX) and carbon monoxide (CO) emissions
can be significantly reduced by maintaining high oxygen con-
centrations, as shown in Figure 10. Also, low superficial
velocities help reduce emissions. The conventional pulverized
coal combustion units produce very low HC concentrations in
the stack gases (Merryman et al., 1977). The Babcock and
Wilcox experimental AFBC unit is monitored for CO with auto-
matic coal feed shutoffs when CO concentrations in the flue
gas exceed 0.1 percent (GCA/Technology Division, 1978).
Carbon utilization efficiency can be measured by con-
tinuous monitoring of C02 in the stack. However, if a metal
carbonate (limestone or dolomite) is used for sulfur sorption,
the results can be distorted, because C02 released during
calcination disturbs the mass balance calculations. For
example, Lowellville limestone is 90.4 percent CaC03; in a
molar balance during combustion and sorption in an FBC unit,
carbon from the limestone accounts for 4.45 percent of the
total combustible carbon input to the bed (Babcock and Wilcox
Company, 1979).
Nitrogen Oxide Reduction—
Figure 10 shows that most NOX production from operating
experimental FBC units is below the current EPA New Source
Performance Standards. Data from PFBC units suggest that they
will produce even less NOX than AFBC units, but neither
should have trouble meeting EPA's proposed reductions in the
standards. The high NOX emissions shown in Figure 11 were
apparently generated by a carbon burn-up cell (Henschel,
1977).
The factors affecting nitrogen emissions from FBC pro-
cesses are many and complex. Simply reducing temperatures in
the combustion chamber will minimize the thermal formation of
NOX, although careful analysis has shown that the majority
of nitrogen emissions result from bound nitrogen in the coal
(Mitchell, 1976; Hozio et al., 1977). Therefore, careful
manipulation of various process design parameters will be re-
quired to keep NOX emissions below any required limits.
In addition to combustion temperature, design parameters
affecting NOX production are: pressure, gas residence time,
excess air, calcium/sulfur ratio, water vapor, carbon monoxide
concentration, and char presence. Figure 12 shows a reduction
in NOX emissions as the calcium/sulfur ratio increases in an
AFBC unit (Jarry et al., 1970), but this correlation is not
apparent in PFBC units. Figure 13 shows that longer gas resi-
dence time lowers NOX emissions (Jarry et al., 1970).
Figure 14 (Keairns et al., 1975) shows that nitrogen oxide
emissions increase with increasing excess air, which increases
the oxygen concentration.
63
-------
1500
E
a
a
CO
c
o
JQ
L_
(0
u
o
•a
>»
I
100
500
O
1.0
-O
2.0 3.0
Oxygen in Flue Gas %
4.0
5.0
Figure 10. Effect of oxygen concentration on hydrocarbon emission.
(Glenn and Robison, 1970)
-------
m
250xlOe BTU/Hr (73MWt)
New Source Performance Standard for
Coal-Fired Electric Utility Steam
Generating Units
>250x10* BTU/Hr (73MWt)
_E_P.E_ L-.
p_p,fJZ.
Equi1ibrium NOx
for the reaction:
2NO
50
2200
(1205)
Bed Temperature, °F (°C)
en
c
0)
•o
X
o
01
o
Figure 11. Nitrogen oxides emissions (expressed as NO,)
AFBC units.
(Weber, 1970)
65
-------
50
40
.9
HI
.2
1 30
•o
0)
DC
20 —
10
1 2
Ca/S Mole Ratio
Figure 12. Nitric oxide reduction with -325 mesh K44um) tymochtee dolomite.
(Jarry at al., 1970)
GAS
VELOCITY
ft/sec
0 2.7
• 2.7
A 8.6
A 8.6
0 2.7
V 8.6
LIMESTONE
NO. I359
urn
25
25
25
25
600
UOO
RECYCLE
NO
YES
NO
YES
NO
NO
TEMPERATURE: 1600 F
COAL FEED: l«.5 «t * S
ADDITIVE: LIMESTONE NO. 1359
FLUIOIZED BED: ALUMINA
#*
<
O
uj
.J
u.
z
z
EDUCTIO
flC
X
o
u
t—
z
70
60
50
1(0
30
20
10
0
I 1 1 1 1 1
—
~ LOW-VELOCITY GAS ~
fo j
loo j
i a 1
~ _J._2 ?_D_OJ —
1 j
L* !
— HIGH-VELOCITY GAS —
1 1 1 1 1 I
3 4 5
Ca/S Mole Ratio
Figure 13. Effect of gas velocity on nitric oxide reduction In flue gas
(Jarry at al., 1970)
66
-------
Approximate Percent Excess Air
z
JO
0.8
0.6
0)
I °-4
'(/>
.w
•I o^
O
CURRENT EPA EMISSION STANDARD
PRESSURE: 303.9 to 1013 k Pa (3 to 10 Atm)
Percent O_ in Flue Gas
8
Figure 14. Composite plot of data for.NOx emissions from FBC of coal.
(Keairns et al., 1975)
-------
In one experiment attempting to reduce the NOX output,
decomposition catalysts (aluminum and zirconium oxides) were
added to the fluidized bed but had no effect. Another cata-
lyst, cobaltous-cobaltic oxide, increased NOX emissions
(Jarry, et al., 1970) .
Studies on NOX formation have shown that the oxides are
formed initially in high concentrations near the distributor
plate at the bottom of the fluidized bed and decrease in the
freeboard area (Pereira et al., 1976). This lower concentra-
tion is probably a result of char NH3 catalyzation of NO to
N2« Utilization of this finding requires some design
changes in FBC units to introduce the first cyclone catch onto
the top of the fluidized bed. Alternatively, a second shallow
uncooled bed could be established where some coal and the re-
cycled char is introduced. The design also requires the
introduction of some secondary air into the freeboard (Beer et
al., 1977).
Sulfur Dioxide Reduction—
Desulfurization and other sulfur reduction methods can
significantly lower sulfur dioxide (S02) formation and
emission. Desulfurization involves a competition between
trapping S02 in the bed via chemical reaction and removal
from the bed by the fluidizing air following sulfur oxida-
tion. The balance between these two processes is dependent on
a multitude of design and operating conditions, such as tem-
perature, bed depth, and gas velocity. When limestone
(CaCOs) is used as the sorbent material, other variables
must be considered. Different limestones exhibit different
absorption rates and capacities, and all require calcination
either in the fluid bed or prior to feeding into the com-
bustor. Figure 15 shows several projected sulfur removal
curves based on thermogravimetric analyses (Newby et al.,
1978), and a kinetic model developed by Westinghouse. To meet
the projected EPA standard of sulfur removal, operating condi-
tions must be optimized to maintain low calcium/ sulfur ratios
using limestone as a sorbent. High calcium/sulfur ratios re-
quire larger quantities of limestone, thereby increasing ex-
penditures for raw materials, for solids disposal, and for
capital costs, and also reducing thermal efficiencies. Figure
16 shows several calculated thermal efficiency curves for
various calcium/sulfur ratios in a fluidized bed fired with
coals containing different sulfur content. These thermal
efficiencies can be compared to a thermal efficiency of about
35 percent attained by a conventional pulverized coal combus-
tion plant with flue gas scrubbing.
Calcination conditions have varying effects on sulfur
sorption. Pore size appears to be a critical, if not the
limiting, factor when using limestone or dolomite as
68
-------
90
80
70
I I
Sorbent Type
Average*Diameter, um
Pressure, kPa
Bed Temperature, C
Excess Air, %
Velocity, m/s
Bed Depth, m
OPERATING CONDITIONS
AFBC
Limestone
a b
500
101
8<40
20
1.83
1000
101
81.0
20
3.05
500
1013
950
20
1.52
2000
1013
950
20
.1.52
1.22 1.22 3.05 3.05
It 5 6 7 8 9 10
Calclum-to-Sulfur Ratio (Molar)
Figure 15. Sulfur removal performance for typical sorbents
(projected using Westlnghouse'kinetlc model).
(Newby et al., 1978)
69
-------
)23
-------
sorbents. Salt added to the bed mixture reduces the cal-
cium/sulfur ratio needed to meet standards. Salt appears to
cause an increase in the amount of CaCOs converted to CaO
through changes in the pore structure (Gasner et al., 1977).
Salt, however, creates other problems, such as corrosion, and
is not recommended for addition. Precalcination in a con-
trolled atmosphere changes the limestone sufficiently such
that thermogravimetric analysis predicts decreases in the
amount of limestone needed for sorption as shown in Table 24.
This may not have influence in the design of a small FBC com-
bustor, but in a large plant precalcination may be important
and add another step in the process design (Babcock and Wilcox
Company, 1979) (Figure 17).
TABLE 24. PREDICTED Ca/S REQUIREMENTS FOR THE ATMOSPHERIC
FLUIDIZED BED COMBUSTOR
(Newby et al., 1979)
Coals
Western coala
at 83.3% S retention,
Eastern coal corresponding to
Sorbents at 85% S retention 0.2 Ib S02/M.Btu
Raw Greer Limestone 2.5 0.83
Precalcined Greer 1.7 0.57
Raw Grove Limestone 5.7 2.0
Precalcined Grove 2.0 0.58
The ash CaO utilization is assumed to be 30%.
The chemical reaction rate limiting step in sulfur sorp-
tion appears to be the 'oxidation of S02 to 803 prior to
calcium sulfate formation. This step may be catalytically
enhanced by compounds found in coal, by limestone, or by added
catalysts. Analysis of spent bed material showed large
amounts of iron in the sulfated layer indicating that iron may
be a catalyst for sulfur uptake (Yang et al., 1978).
71
-------
LIMESTONE FEED,
CRUSHED AND SIZED
(76-77'F)
HEATED LIMESTONE
(600-800*F)
HEATED LIMESTONE
(1150-1250*F)
COAL
(77'F)
HEATED AIR
(800-900*F)
AIR
*• BLOWER
3 £
FIRST HEATING STAGE
(600-800*F)
SECOND HEATING STAGE
(II50-1250*F)
CALCINING STAGE
(1562-F)
COOLING STAGE
(800-900°F)
AIR
(302*F)
COOLED EFFLUENT GAS
(600-800° F)
EFFLUENT GASES
(II50-I250*F)
HOT EFFLUENT GASES
(I562*F)
HOT LIME PRODUCT
(I562°F)
COOLED LIME PRODUCT
COAL FEED
PRIMARY
CYCLONE
FLUE GAS
FLY ASH
MAIN ATMOSPHERIC
FLUIDIZED 8ED
COMBUSTOR
-------
Particulate Matter—
In 1977, fly ash produced by coal-fired power plants was
considered to be the sixth most abundant mineral in the United
States (Chemical and Engineering News, 1978). FBC processes
will produce even more non-sintered solid material than the
conventional coal-fired power plants. Particulate matter is
derived from char, ash, and from sulfur sorbent material in
the fluidized bed, and each group has different physical and
chemical characteristics. The mechanisms controlling the
formation of the particulate matter during process are related
to a number of variables. The major variables are fluidizing
velocity, feed material particle size, feed material distri-
bution and feed rates, degradation rates of the bed material,
interparticle attrition rates and buoyancy, and bed geometry
and height, including the height of the freeboard. The major-
ity of particulates in FBC processes are derived from the sor-
bent material, with the total quantity being dependent on the
calcium/sulfur ratios required to meet sulfur removal perfor-
mance standards (Newby et al., 1979). Figure 18 shows calcu-
lated particle loadings for a calcium/sulfur ratio of 2, com-
pared to actual values obtained in a combustor: a total load-
ing of 0.535 gm solids produced per gram of coal. Table 25
shows some of the physical parameters of flue gas particulates
from a PFBC unit.
The release of particulates through the stack to the
atmosphere is a function of the final particulate removal sys-
tem performance as shown in Figure '19. The very small parti-
culates (less than 3 microns) are significant because they
easily adsorb toxic elements and are respirable into the human
lungs.
OUTLOOK
It is relatively certain that FBC will be commercialized;
the timetables for market entry, however, will depend on
various technical, economic, environmental and political
factors. Technical factors weigh heavily in making a choice
and commitment to either AFBC or PFBC or both. Technically,
they are both-good choices. Each offers certain advantages,
over its range of applicability.
Some companies such as Pope, Evans & Robbins, Inc.,
Combustion Engineering, and Babcock and Wilcox, Alliance,
Ohio, are concentrating entirely on atmospheric systems;
others such as Curtiss-Wright Corp. and Exxon are concen-
trating on pressurized systems. Still others, such as Babcock
& Wilcox Ltd., London, are actively involved in both lines of
development (McKenzie, 1978). Companies pursuing both lines
feel that AFBC will be preferred for industrial-sized boilers,
and that utilities will prefer pressurized systems, which can
73
-------
TABLE 25. PROPERTIES OF THE PARTICLES SUSPENDED IN THE FLUE GAS
(Murthy et al., 1977)
<1 micron
1 to 3 microns i
3 to 10 microns
>10 microns
Predominant shape
Evident cleavage:
Structure:
Color:
7% wt.
39% wt.
21% wt.
33% wt.
Irregular
None
3 phase
White, black, red
0)
'w
(5
O
•*•<
>.
a>
u
t_
CD
a.
0,01
0.1
1
5
20
50
80
95
99
99.9
99.99
0.1
Pope, Evans; Robbins Data
Projected
1.0 10.0
Diameter (urn)
100.0
1000.0
Figure 18. Total solids from combustor (AFBC).
(Newby et al., 1978)
74
-------
6.2-
Partlculate
Emissions
lb/10 Btu
0.025-
0.011
Cyclone I
78,000 Ib/hr
0.2.
O.I - Current NSPS
0.05 Projected NSPS
— CFCC Emissions
ItOO 300 200 100
Particle Size (Microns)
Figure 19. CFCC particulate emissions.
(Luck* and Murphy, 1977)
75
-------
be adapted to combined-cycle use. For a large utility plant
generating about 200 megawatts, important capital-cost savings
will be realized by the PFBC's smaller size and improved effi-
ciency is achieved through, combined cycles.
i
The controversy on timing stems from uncertainty as to
scale-up and the usefulness of pilot-plant data. Some United
States research laboratories conduct exhaustive small-scale
investigations on emissions and their control, sorbent be-
havior, and other fluidized-bed phenomena. Others, especially
European firms, insist that small-scale results will not apply
to scaledup units. Researchers from such companies, including
Stal-Laval Turbin (Finspong, Sweden) and Babcock & Wilcox Ltd.
(London), feel that enough data already exist to permit
construction of major prototype facilities.
A different opinion is held by those involved in a Pres-
surized FBC demonstration plant to be built at Grimethorpe,
England, by the British Government's National Coal Board. The
80 megawatt (thermal) unit will be of prototype size but will
be used purely for research. It is expected to be more flexi-
ble than a commercial unit and should better lend itself to
full-scale experimentation.
Researchers at Grimethorpe doubt there are enough rele-
vant data available to successfully design and operate
full-scale, pressurized fluidized bed combustor/gas-turbine
systems. However, Stal-Laval Turbin and Babcock & Wilcox
researchers, among others, disagree. The Grimethorpe team
consider it premature to tackle both the problem of a pres-
surized fluidized bed combustor and that of coupling such a
combustor to a gas turbine. Some US experts feel that the
Stal-Laval/Babcock & Wilcox team is overlooking the problems
of gas-turbine-blade erosion/corrosion due to sodium, potas-
sium and vanadium in the coal.
One project, already considered obsolete by some, is the
30 megawatt (electrical) prototype atmospheric unit, built in
Rivesville, West Virginia, by Pope, Evans & Robbins for the
former U.S. Office of Coal Research, now part of U.S. Depart-
ment of Energy (DOE). The project was an attempt to develop a
relatively inexpensive coal-fired boiler for power genera-
tion. Contracted for in 1972, before the coal price-rise, the
system was designed to compete with oil-fired boilers. The
plant is a multi-cell FBC unit, consisting of four separate
beds. The first bed boils the water, the second heats the
resulting steam, the third superheats the steam, and the
fourth burns up unburned carbon. Problems have been en-
countered, particularly with coal-handling and coal-feed sys-
tems. One persistent problem is clogging of coal in feed
lines, blocking the fuel supply.
76
-------
Fluidized bed combustors must compete economically, not
with oil-fired boilers, but with conventional coal-fired
boilers using scrubbers. Pope, Evans & Robbins researchers
concede that times and priorities have changed. However, the
company questions a wholesale switch to PFBC, believing that
money saved by smaller size units for the same output (pres-
surized unit) will be offset by the additional cost of coping
with higher pressures, especially for feeding coal and re-
moving solid waste. Pope, Evans & Robbins researchers feel
that atmospheric systems are preferable for all size ranges.
Stal-Laval researchers believe that pressurized systems
may be preferable to medium or large-scale FBC applications.
The firm cites the advantage of being able to double a pres-
surized system's useful energy output by using combined cycles
or cogeneration. Its researchers believe this will be the
most effective lever in persuading industries to convert their
boilers to ones that use pressurized fluidized bed coal
firing.
Dr. Steve Freedman, Director of the U.S. DOE's fluidized
bed program, feels that industrial-sized AFBC units are
closest to becoming commercial. Their greatest advantage, he
believes, is their fuel-use flexibility. Freedman sees util-
ity systems as a longer term proposition. DOE hopes to have
both large AFBC and large PFBC plants on-line by 1984, so
utilities can choose between the two systems. There are ad-
vantages in both approaches: atmospheric is simpler and more
reliable, while pressurized has a slightly higher efficiency
and produces slightly fewer sulfur-dioxide emissions.
Other differences of opinion exist. Columbus Battelle
and various U. S. researchers are concerned about possible
corrosion of boiler tubes. Others, especially those in
Europe, see nothing at all to suggest that such problems will
arise (McKenzie, 1978).
A number of factors inhibit the introduction of FBC tech-
nology. Initially, the air pollution rules favor the use of
clean fuels such as natural gas and low sulfur fuel oil. As
the price of these fuels rise, there will be a shift to coal
where fluidized bed combustors must compete with established
technology and add-on flue gas desulfurization.
Another problem arises from the fact that FBC is an un-
tried technology. While it is agreed that FBC solves the
SOX and NOX problems, it is unknown whether new pollutants
in low concentration will be generated. Coal contains many
impurities, and these compounds may be released into the envi-
ronment. The little available evidence from research units
does not suggest any hazard, but these units have not been
adequately tested and the problems of scale-up to commercial
77
-------
units remain to be examined. Additionally, governmental regu-
lations are constantly changing with unknown consequences for
FBC technology. As one person at Babcock & Wilcox said,
"There are hundreds of orders for the second commercial com-
bustor built."
The remainder of the problems are economic. There is
little governmental economic incentive to develop and com-
mercialize FBC. The federal government has approved the
release of low interest bonds to finance pollution control
equipment; however, these bonds can be used to finance the
installation of scrubbers on existing conventional coal burn-
ing facilities, they cannot be used to finance new FBC instal-
lations which are not designed as a basic principle to emit no
pollution. Also, these bonds cannot be used to convert
existing coal-burning facilities with pollution problems to
FBC processes. These negative governmental measures could
substantially retard FBC development.
From a large utility standpoint, there appears to be
little incentive for FBC technology. While it appears that
fluidized bed combustors are more energy-efficient than con-
ventional combustion with add-ons, there is little incentive
to switch as many state utility regulators permit fuel cost
pass on or recovery. Also, allowance is made for capital
recovery for equipment, and thus, there is little incentive to
build cheaper plants. Where utilities cannot recover their
costs (e.g., employee salaries), they are already pared to the
bone. Table 26 shows the wide variety of factors influencing
the commercialization of any new or emerging technology, such
as FBC.
Despite these problems, it seems fairly certain that FBC
of coal will be commercialized. It is probable that the first
units installed will be small industrial boilers. The time-
tables are uncertain since they depend on technical, economic,
environmental and political factors. Because of their tech-
nological complexity, pressurized systems will undoubtedly be
commercialized after atmospheric units.
78
-------
TABLE 26. FACTORS INFLUENCING THE COMMERCIALIZATION OF FLUIDIZED BED
COMBUSTION OF COAL
Degree of current Degree of influence
Influence factor quantification on commercialization
A. ECONOMIC FACTORS
1. Benefit/costs
2. Availability of Commercial
Scale Components
3. Number of Existing Units
4. Existing Unit Size vs
Commercial Unit Size
5. Commercialization Schedule
(DOE Industrial Foreign)
6. Potential Market Size
7. US Government R&D Support
B. ENVIRONMENTAL FACTORS
1. Relative Known Hazard
2. Number of Waste Streams
3. Pollutant Mass Flow Rate
4. Extent and Types of Fugitives ++
5. Relative Effectiveness of
Existing Controls ++
6. Variability of Emissions
with Site +0
7. Availability of Environmental
Baseline Information +0
8. Availability of Commercial
Monitoring Equipment ++
C. POLITICAL/INSTITUTIONAL FACTORS
1. Perceived Relationships to
National Energy Needs +0
2. Current and Anticipated Tax
Advantages ++
3. Regional Differences in
Advocacy and Commitment +++
4. Historical Industrial
Preference +++0
5. Perceived Short Term and
Long Term Advantages ++
6. Regulatory Constraints
7. Regulatory Incentives
D. TECHNICAL FACTORS
1. Magnitude of Unresolved Techni-
cal Problems (i.e., coal feed)
2. Process Differences + +
3. Raw Material Acquisition ++ +-H-
4. Simplicity vs Complexity of
Process/Technology +0 ?
5. Professional Opinion and
Preference +0
High Degree +++
Moderate Degree ++
Slight Degree +
Unknown ?
Extremely Difficult to
Estimate 0
79
-------
80
-------
SECTION 3
SELECTION OF POLLUTANTS
FOR MONITORING
INTRODUCTION
The EPA is interested in monitoring pollutants from coal
facilities utilizing FBC primarily to support pollution con-
trol and abatement efforts, promote compliance with standards,
and protect public health and welfare. The advances made in
the understanding of the complex manner in which pollutants
roay impact human and environmental receptors have led the EPA
to develop holistic approaches to assessment and monitoring
(Cleland and Kingsbury, 1977; Schalit and Wolfe, 1978; Behar
et al., 1979).
Since it is reasonable to assume that a commercial-scale
fluidized bed combustor may emit some unregulated pollutants,
selecting pollutants for monitoring is difficult. However,
Monitoring both regulated and unregulated substances may sup-
port research designed to link a source to known impacts or
health and ecological effects. A serious deficiency in cur-
rent approaches to monitoring is the absence of an organized
Procedure for selecting pollutants to be included in a moni-
toring program. In this section, a conceptual framework is
developed for such a procedure, as part of an approach to the
design of a source-oriented ambient monitoring program.
Objectives of a source-oriented ambient monitoring pro-
gram may include monitoring to:
Assure compliance with existing standards, criteria
and regulations-addressing source emissions, re-
ceptor media, and potential environmental effects
Support effects monitoring and research
Accumulate baseline data in anticipation of proposed
standards, criteria and regulations
Verify dispersion/transport models
Add to existing data banks holding environmental
data
81
-------
Monitoring objectives provide the rationale for identi-
fying:
Which type of pollutants need to be monitored
• Where monitoring should be done in relation to the
source
Where sampling sites should be located in the envi-
ronment
How frequently samples should be collected
How monitoring data should be analyzed
The format in which results of analysis and raw data
are reported
The quality assurance requirements
Analytic methods selected for ambient measurements
The objectives determine the overall perspective and de-
tails of every element of the monitoring program. Therefore,
they must be made explicit at the beginning of the design pro-"
cess to assure that the objectives will be met after the ef-
fort is completed.
As Figure 20 illustrates, before the prioritization
procedure can be used, candidate pollutants and monitoring
objectives are identified. Data analysis and quality control
needs to meet program objectives are formulated concurrently
with the selection of pollutants that are to be monitored.
Once the pollutants have been selected for the program,
sampling and analytic methods can be considered. Ideally, the
entire monitoring program should be designed before any actual
sampling begins.
The specific monitoring objectives used to illustrate how
the proposed pollutant selection scheme works include moni-
toring to:
Assure compliance with existing federal standards,
criteria, and regulations exclusive of Occupational
Safety and Health Act (OSHA)
Accumulate baseline data in anticipation of proposed
standards criteria and source regulations
Support effects monitoring and research
Monitoring program objectives influence the pollutant
selection process by providing the basis for identifying the
82
-------
r
QUALITY
ASSURANCE •
NEEDS
DETERMINED
t
I
QUALITY
CONTROL-
PROCESS
SELECTED
oo
to
REGULATIONS
REVIEWED
POTENTIAL
POLLUTANTS
IDENTIFIED
OTHER
*• NEEDS -
ADDRESSED
(DATA EVAL-
UATION)
METHODS
SELECTED
MONITORING
OBJECTIVES
FORMULATED
Procedure developed in this
section fits within this step.
I
^POLLUTANTS
PRIORITIZED —
AND
SELECTED
I
I
I
I
DATA
^ ANALYSIS —
PROCEDURES
SELECTED
I
I
I
I
DATA
INTERPRETATION
*- PROCEDURES -
FORMULATED
I
SAMPLING
AND —
ANALYTIC
METHODS
SELECTED
DATA
~REPORT ING -
FORMAT
ESTABLISHED
GUIDANCE
- REPORT
COMPLETE
Figure 20. Design of an ambient monitoring program.
-------
series of questions that must be asked during the selection _
process. Once these questions are identified, they are
organized so an efficient and systematic selection procedure
results. This approach also lends itself to careful documen-
tation of the entire program design.
Initially, the questions used to determine whether a
substance, required monitoring were identified. Relationships
among the questions were reviewed and dependencies estab-
lished. The dependencies were utilized to organize the ques-
tions into a hierarchy that represents a logical flow from
identification of the chemical nature of a substance to a
determination that its concentration should be monitored in
the ambient environment. The procedure attempts to achieve
objectives:
Categorization of all potential pollutants
Priority ranking of FBC pollutants
Selection of regulated and problem pollutants for
monitoring in the ambient.environment
The first objective of the procedure is to categorize the
candidate pollutants into three types based on the monitoring
objectives of the program:
Those that must be monitored in the ambient environ-
ment or facility effluent according to existing
legal requirements (Category I pollutants)
Those that could be monitored based on proposed
regulations or criteria (Category II pollutants)
• Those for which there are no existing or proposed
regulations or criteria (Category III)
Unregulated pollutants that are likely to present a
serious hazard to human health and welfare are included in
Category I during implementation of the procedure. Category I
pollutants are the minimum set of pollutants to be included in
the monitoring program, unless cost or other considerations
make this unreasonable. This approach assures that legal re-
quirements for monitoring are met and potentially significant
health and ecological problems are considered. Monitoring of
Category II pollutants is desirable but not required. Design
of a monitoring program should include monitoring for these
substances in anticipation of future needs to meet standards
currently proposed or pending.
84
-------
The second objective is to priority rank the Category III
pollutants to decide which unregulated substances need moni-
toring. Some existing threshold criteria for Category III
Pollutants are identified in available literature; new crite-
ria may be derived from a review of known health and ecologi-
cal effects. Ideally, the threshold criteria represent a ten-
tative level of concentration in the environment below which
significant adverse effects are unlikely. This concentration
is used as a basis for ranking the hazard potential of the
Pollutants prior to selecting those designated for monitoring.
The third objective is to select the pollutants to be
monitored. This set of substances will include, minimally,
Category I pollutants and their support parameters. Support
parameters are the environmental characteristics which are not
associated with selecting sampling sites and frequency, but
are needed in order to interpret whether a standard or crite-
rion is met or exceeded. Typical support parameters are
temperature, wind velocity and direction, and stream or
aquifer flow rate. In addition, Category II substances and
the most hazardous substances in Category III should be in-
cluded in the program. The outcome is a list of pollutants
that will not be monitored and a list of those that will be
monitored to meet specific monitoring program needs. A sum-
mary list of the pollutants in the former group is retained to
document the decision process used in selecting the pollutants
for monitoring.
The hierarchical organization of the procedure is il-
lustrated in Figure 21. The discussion in this section
Presents this systematic approach to the selection of the pol-
lutants for monitoring in the ambient environment.
ASSUMPTIONS
The procedure is designed for application to a specific
PBC process selected as the focal point of a source-oriented
ambient monitoring program. The FBC process is assumed to in-
clude all activities associated with combustion of coal at a
Particular facility, from moving coal and sorbent from a stor-
age pile on-site, to disposal of all wastes at an approved
landfill. Basic assumptions are:
A review of the literature on the FBC technology has
been completed.
Sufficient information on starting materials and
effluents from pilot demonstration, and all com-
mercial plants is available.
A specific set of monitoring objectives guide devel-
opment of the monitoring program.
85
-------
Process Selected
Ye I
Miss Balance
Thermodynamlc Analysis
Starting Materials Analysis
Level I/Level 2 Analysis
Other Effluent Testing
Chemical Species Identified
I
Data Base Completed
— ——Insufficient Data Base?
1 No
Expected Receiving Media Identified
I
Do Program Objectives Consider Stack/Pipe Emissions?
No
Yes
Applicable Existing Ambient Standards
and Regulations Identified for Process,
Media, and Effluent Components Only
Applicable Existing Effluent and Ambient
Standards and Regulations Identified for
Process, Media, and Each Effluent Suedes
I
Does A Standard or Regulation
Currently Exist for the Substance?
No
Yes
Does A Proposed Standard or
Regulation Exist for the Substance?
Yes
No
Do Relevant Laboratory or Field Studies
Indicate Potential for Extreme Toxlclty?
"'
Do Monitoring Objectives
Require Monitoring for
Proposed Criteria and -
Standards ?
Yes
4
No
Collate All Category III Pollutants
Identify or Derive Environmental Effect
Threshold Level for Each Pollutant
Estimate or Derive Expected Emission
Rate or Ambient Concentration Generated
by Process In Each Medium
Establish Hazard Potential
for Each Unregulated Pollutant
I
Does the Hazard Potential of the
Unregulated Pollutant Suggest It Should
Be Monitored in the Ambient Environment?
Yes
No
Category I
Pollutant
Does Toxiclty Data or Other Evidence
Suggest Standard or Criterion Should
Be Changed?
Yes
~i No
Support Parameters Required for All
Problem Pollutants Identified
Preliminary List of Pollutants Identified
for Design of Monitoring Program
Preliminary List of Unregulated
Pollutants Not Selected for Monitoring
Figure 21. Procedure for selecting pollutants for a source-oriented ambient monitoring plan.
86
-------
It is further assumed that:
Monitoring networks are designed to detect specific
substances in the ambient environment around a
facility utilizing the FBC process and potential
waste disposal sites. Therefore, pollutants
generated by a source are priority ranked, in con-
trast to the priority ranking of sources or facil-
ities (Eimutis, 1976; Reznik et al., 1978; Slawson,
1979; Everett, 1979).
• Adequate information is not available in most cases
to establish error associated with estimates of ef-
fects and production rate factors of pollutants from
the processes considered. Therefore, the approach
presented here for identifying the pollutants that
should be monitored in the ambient environment can-
not produce probability statements indicating the
level of confidence associated with the decision
that monitoring is required.
CATEGORIZATION
Overview
Categorization of effluent components involves four pro-
cesses:
1) Identify chemical species in each effluent stream
2) Determine receptor media for each chemical species
3) Identify relevant standards, criteria, and regula-
tions for the facility, effluent streams, media, and
individual pollutants
4) Identify extremely hazardous or toxic substances
These four steps divide the initial list of effluent compo-
nents into three groups, each with different monitoring
requirements. It is possible for design of a monitoring pro-
gram to focus on only one or all of these groups of pollu-
tants.
Species Identification
If there has been extensive analysis of effluents and
emissions from pilot and demonstration plants, the list of
conceivable pollutants may include single chemical species and
families of substances with closely related chemical struc-
tures. When a group of closely related chemical species is
identified in process effluents, refinement of information on
87
-------
chemical structure and properties of the component is needed
so appropriate regulations and health and ecological effects
can be utilized in evaluating whether or not monitoring the
group's ambient concentration is necessary.
i
An effluent component which includes a group of closely
related chemical substances may be identified, but the infor-
mation necessary to determine exactly which species is in the
effluents can be unavailable. The monitoring requirements
with respect to this component would be determined in one of
two ways: 1) the component may contain so few species that
considering each member of the group for monitoring is war-
ranted; or 2) the component may include a variety of species;
however, monitoring the component as a group could be more
appropriate. As a third alternative, the group could be broken
into subsets, and each subset of the component considered for
monitoring. Consideration of available analytical methods is
a significant factor in deciding how to proceed.
In the event that the chemical makeup of an effluent
component is unknown, completion of a Level I/Level 2 assess-
.ment or other effluent characterization procedures would be
recommended before developing a source-oriented ambient moni-
toring program for the component. Level 1 sampling and
analysis is designed to give rough information on the amount
of all pollutants for which Multimedia Environmental Goals
(MEGs) have been established. Level 2 sampling is a focused,
accurate analysis program for a few selected chemical pollu-
tants.
At this point, the best available information on the
kinds of pollutants present in process effluents is assembled
for further consideration.
Media Determination
Further characterization of an effluent component ac-
cording to the environmental medium most likely to receive the
substance(s) is needed prior to identifying relevant standards
and regulations, and potential health and ecological effects.
The description of the process selected for monitoring pro-
vides the information for determining where the substances are
most likely to enter the ambient environment. Otherwise, the
receiving media must be assumed, based on whatever information
is available on operation of the fluidized bed combustor and
disposal methods.
Identification of Standards and Criteria for Specific Chemical
Components of the Effluent Streams
Legal requirements for ambient monitoring based on the
facility or process type and specific location of the facility
88
-------
are identified first. Next, existing and interim federal
environmental laws and regulations relevant to the PBC process
are reviewed to determine appropriate standards and criteria
for each receptor medium and waste substance. A sample list
of these laws, including air, water, and solid waste statutes,
regulations, and other environment-related governmental
orders, is provided in Appendix B.
Category II pollutants are identified through review of
the status of laws which are being translated into regu-
lations. Normally, regulations are proposed, the proposals
reviewed by the public, and the proposed rules revised
accordingly, before final promulgation. Any applicable pro-
posed regulations are reviewed to identify Category II pollu-
tants. Proposed interim regulations would be considered for
identifying Category II pollutants, since they are not cur-
rently enforceable (Section 5).
If emission standards are considered in scope, chemical
species for which emission rates or discharge are regulated
should be placed in Category I even though monitoring in the
ambient environment may not be explicitly required. For these
substances, monitoring can be used to verify compliance with
emission standards in a potentially cost-effective manner.
Any changes, revisions, or amendments to regulations and
standards can be identified through review of the Federal
Register and secondary sources. In the event that evidence
becomes available that a standard or criterion is excessive,
deficient, or recently revised, the need to monitor the pollu-
tant can be reconsidered based on the new evidence.
Identification of Extremely Hazardous or Toxic Substances
At this point, both regulated and unregulated pollutants
have been identified and characterized according to chemical
species and receptor medium. If available evidence indicates
that some of the unregulated substances are extremely toxic or
hazardous to human or environmental receptors, they can be
placed in Category I and automatically considered for moni-
toring. The implication is, in this case, that discharge of
even very small quantities carries some risk to environmental
receptors. If data indicate only very small emissions of
these substances are expected from the source, the data can be
used to support a position that they are unlikely to cause
impacts and should not be monitored.
Evidence of extreme hazard can be obtained from two
primary sources. Lists can be consulted which reflect at-
tempts by government agencies and other organizations to iden-
tify substances posing a threat to health and the environment
(International Joint Commission, 1978; Keith and -Telliard,
89
-------
1979). Alternately, the EPA-NIH Chemline can be reviewed to
locate toxicological data on substances not on these lists.
The open literature can also be used to identify potential
hazards of the unregulated pollutants, including acute
toxicity, sub-acute or chromic toxicity, persistence, or bio-
accumulation.
Using criteria proposed in Table 27, extremely toxic sub-
stances in process effluents can be identified.
TABLE 27. CRITERIA USED TO IDENTIFY EXTREMELY
TOXIC SUBSTANCES
Type of data
available
Threshold level
indicating extreme
toxicity
Reference
Acute Oral Toxicity,
Mammalian LD,-0
Acute Ingestion
Toxicity, Fish
24, 48 or 96 hr LC5Q
96 hr LCsg,
Fish or Invertebrate
Longevity Factor
(Soil Life
Persistence)
Bioaccumulation
(highly accumulative)
<20 mg/g
<0.01 ppm
mg/1
>5 wks
>8°°0
Weber, 1977
Weber, 1977
NIOSH, 1976
Weber, 1977
. »"
If toxicological data are available in one of these cate-
gories, it can be compared to the threshold levels for extreme
toxicity. Although unregulated, a substance revealed by this
method to be extremely toxic or hazardous deserves serious
consideration for monitoring to link source emissions to
potential health or ecological effects. For a less toxic
substance, a more refined approach is needed to be able to
select, with confidence, whether it should be monitored in
the ambient environment.
90
-------
This simplified screening process is most useful when the
contribution of the FBC process to ambient levels of a pollu-
tant is unknown/ and should be replaced by a more sophisti-
cated priority ranking procedure (discussed in the following
sections) when site/facility data become available.
Summary of the Categorization Process
The result of completing the first phase of the selection
procedure is a grouping of effluent components by categories.
Category I pollutants are to be monitored to meet the stan-
dards and regulations addressing operation of FBC processes
and gather information on unregulated pollutants strongly
suspected of having the potential to impact public health and
ecological systems. Category II pollutants are monitored in
anticipation of federal requirements for compliance. Category
III pollutants are unregulated substances for which there is
either only limited effects information or evidence of only
slight-to-moderate hazard to potential receptors. The need to
monitor substances in Categories II and III is based primarily
on the monitoring objectives of the program. It is assumed
that these objectives recognize the need to support health and
ecological research programs. The primary criterion for
selecting a substance in Categories II and III for the moni-
toring program is the likelihood that the substance would pose
a problem if it were to enter the ambient environment. This
is determined in the second phase of the selection procedure,
which identifies an effects threshold level and an estimate of
potential hazard for substances not previously assigned a
legal criterion. Also in this phase, the substances in the
three categories may be readjusted, e.g., if health or ecolo-
gical studies suggest that a legal standard or criterion for a
pollutant should be changed, a new effects threshold level can
be derived for the substance and it can be processed further
as a Category III substance.
PRIORITY RANKING
Overview
Priority ranking is a procedure developed to select from
the unregulated pollutants produced by a source those sub-
stances which should be monitored in the ambient environment.
Since priority ranking estimates the potential damage to
health or the environment from expected emissions generated by
the source, it addresses monitoring objectives to link source
to effects. The procedure consists of two major activities.
The first, estimating the potential for receptor damage
for each unregulated pollutant generally requires at least two
data elements—a factor representing known damage effects of
91
-------
the substance and the amount or rate of production of the sub-
stance by the source. In the absence of actual data, values
for either of these elements should be assumed or estimated
for a specific FBC source. :The second activity combines the
two factors derived in the first step as a simple ratio or
more complex algorithm to formulate a hazard index. A deci-
sion rule is then developed that serves as a basis for com-
pleting a preliminary list of Category III pollutants that
need to be monitored according to information contained in the
hazard indices.
Identification of the Potential for Receptor Damage from Un-
regulated Pollutants
The primary criterion for evaluating the damage potential
of an unregulated pollutant is either a suspected health or
ecological effect. Other effects, on non-living materials for
instance, can also be considered. In any case, information
sources on effects of each of the unregulated pollutants are
reviewed and the available data are translated into a thres-
hold concentration, derived from the concentrations known to
produce effects on receptors. Ideally, a minimum effects
threshold concentration can be estimated, the receptor
exposure concentration below which effects should not be ob-
served. If appropriate data are available, this should be
estimated for critical or sensitive receptors in the environ-
ment (children, hypersensitive individuals, endangered
species, etc.)
The minimum effects threshold level concentration serves
as a scaling factor or index of the minimum potential hazard,
reflecting the maximum acceptable environmental concentration
of the substance that is unlikely to cause serious acute or
chronic health or ecological effects. This minimum threshold
concentration is analogous to a standard or criterion. Since
this concentration is estimated by one of many available
procedures, the error associated with the estimate should also
be calculated to determine to what degree the threshold is
likely to be exceeded.
Assigning minimum hazard indices requires an extensive
understanding of the methods (and pitfalls) associated with
utilizing toxicological and ecological effects data on a sus-
pected pollutant (Cairns and Maki, 1979; Draggan and Giddings,
1978).
Most variations in the derivations can be traced to quan-
tification of value judgments concerning relative environ-
mental impact or potential for causing damage to receptors in
the absence of primary data (Jones, 1978).
92
-------
Representative systems were reviewed and evaluated for
determining potential minimum hazard levels for pollutants.
Implementation of the selection procedure should begin with a
review of all available systems before selecting the most
appropriate procedure consistent with the objectives of the
monitoring plan.
The EPA has recently funded a program to identify minimal
effects concentrations for pollutants generated by fossil fuel
combustion and conversion processes (Cleland and Kingsbury,
1977; Kingsbury and White, 1979). This system, the Multimedia
Environmental Goals (MEGs), was developed by several support
contractors for the EPA. Because it is undergoing continued
refinement within the EPA, it was given the greatest atten-
tion in this review.
MEGs are concentration-level goals for hundreds of trace
contaminants found in the effluents of coal and oil conversion
processes (Cleland and Kingsbury, 1977). MEGs have been de-
rived for pollutants based on both health and ecological ef-
fects and are tabulated for the air, water, and land media.
Currently, MEG values are available for many of 650 substances
in the original publication (Cleland and Kingsbury, 1977;
Kingsbury and Sims, 1978); MEGs are updated periodically for
various organic and inorganic compounds.
The initial intent of establishing MEG values for chemi-
cal substances was to provide scaling factors to 1) provide
decision criteria in assessing the impact of a source on the
environment, 2) use in systems to establish priorities among
pollutants considered for possible regulation, and 3) in-
fluence developments in control technology. These reports
collate diverse data sets in a coherent, efficient fashion and
explore related subjects.
One major category of MEGs is Minimum Acute Toxicity Ef-
fluents (MATEs). MATE values are approximate concentrations
of contaminants in undiluted source emissions to air, water,
or land which probably will not immediately evoke harmful or
irreversible responses in exposed humans or other organisms,
when those exposures are limited to short periods of time
(Table 28). Figure 22 shows how MATEs are derived for each
medium.
The second major category of MEGs is Estimated Permis-
sible Concentrations (EPCs). These are concentration levels
in the ambient environment that are considered to be safe for
continuous exposure. Twenty-two different kinds of EPCs are
defined, with some EPCs secondarily derived from other EPCs.
In the event that multiple data exist for a substance, only
the lowest EPC for a given medium is reported.
93
-------
TABLE 28. EMISSION LEVEL GOALS FOUNDATIONS
(Cleland and Kingsbury, 1977)
EMISSION LEVEL GOALS
Air
Water
Land
I. BASED ON BEST TECHNOLOGY II. BASED ON AMBIENT FACTORS
\. Existing Standards
i
NSPS, BPT, BAT
Methodology not yet
developed .
B. Developing
Technology
Engineering Estimates
(R&D Goals)
Methodology not yet
developed
A. Minimum Acute
Toxicity Effluent
Health
Effects
Foundation:
OMATEAHS
OMATEAHl(a)
OMATEAH2
OMATEAH3
OMATEACI
"•««**
Foundation:
OMATEWHS
OMATEWHI
OMATEWC
Foundation :
°^«U.l
Ecological
Effects
Foundation:
OMATEAE
Foundation:
OMATEWES
OMATEWEI
Foundation :
OMATELEI
B. Ambient Level
Goal
Health
Effects
Foundation:
°EPCAHS
°EPCAH1
°EPCAH3
°EPCAC1
EPCAC2
°EPCAT
Foundation:
°EPCWHS
°EPCwc
°EPCwc
°EPC
WT
Foundation:
°EPCLH
°EPCLC
°EPCLT
Ecological
Effects
Foundation:
°EPCAE
Foundation:
°EPCWE1
°EPCWE2
Foundation:
°EPCLE
C. Elimination of
Discharge
Natural Background
Foundation:
Rural background con-
centrations . ..
- .-
Foundation:
Natural concentrations
in surface waters
o
Concentrations
measured in drinking
water
Natural seawater con-
centrations
Foundation:
Typical soil con-
centrations •
vo
-------
MATE-10 xTLV OR
NIOSH VALUE
IN mg/m1
MULTIMEDIA MINIMUM ACUTE TOXICITY EFFLUENTS (HATES)
MOST STRINGENT
MATE FOR WATER
MATE=2xlo>LOWER
'""WATER IN «"'
MOST STRINGENT CRITERION FOR
PROTECTION OF AQUATIC LIFE
MATE=5xCRITERION
USED IF TLV
THAT CONSIDERS
USED IF NO
APPLICABLE
STANDARDS OR
CRITERIA
TLV OR NIOSH
RECOMMENDATION
THAT CONSIDERS
CARCINOGENICITY
MATE-lo'xTLV OR
NIOSH VALUE
IN mg/m1
— — 1 CARC 1 N
\IS NOT
CARCINOGENIC POTENTIAL
ORDERING NUMBER
MATE-7xlO*xTDLO IN mg/kg
EPA/NIOSH ORDERING NO.
LOWEST DRINKING WATER
STANDARD OR CRITERIA
MATE=5xLOWEST STANDARD
OR CRITERIA IN ug/l
PREDICTED FROM AIR
MATE BASED ON HEALTH
EFFECTS
MATE-15xMATEA|R
IN pg/ra1
1
L°50
nATE="l5
-------
TABLE 29. BASIC DATA AND DERIVATIONS FOR AMBIENT LEVEL GOALS
(Cleland and Klngsbury, 1977)
AMBIENT LEVEL GOALS
VD
Air
Hater
Land
I. Current or Proposed Ambient
Standard or Criteria
A. Health Effects
Basic Data i
°Prlnary Ambient Air
Quality Standard
(existing or proposed
"Criteria for Hazardou
Air Pollutant Emissloi
Standards
Derivation t
°Basio Data (EPC^)
Basic Datat
°Drinking Water Re-
gulation or Criteria
Derivation i
°Lowest Value from
Basic Data (EPC^)
Basic Data:
DNone Available (in
appropriate units)
Derivation!
None
B. Ecological Effect.
Basic Data:
°Secondary Ambient Air
Quality Standard
(existing or proposed
Derlvationi
°Basic Data
Basic Datat
°Water Quality Criteria
Established for
Protection of Aquatic
Life
Derivation:
°Lowest value from
Basic Data 'EPCyggl
iasic Data:
'None available
)erivation: t
i o
None
II. Toxicity Based Estimated
Penaissible Concentrations
: A. Health Effects
Basic Data:
OTLV
ONIOSH Recommendatio
°LDSO (or substitute
Derivation i
°EPCAH1(.)
°*V»
Basic Data
°TLV
o._, (or substitute)
"'so
Derivation:
°EfCm2
Basic Data:
OEPCWH(S2)
erivation:
EPCLH
B. Ecological Effects
Basic Data:
"Lowest concentration
affecting sensitive
vegetation (24 hrs)
Derivation'
°-AK
Basic Data:
Lowest Aquatic LC
and Application 5
Factor
"Tainting Leliel
°Application Factor
or Hazard Level
from Hater Quality
Criteria
"Accumulation Factor
and Allowable Flesh
Concentration
Derivation:
^Srci
EPCHE2
OEPCHE3
x«
Basic Data:
o
EPSlB(S,l,2,3.4)
Derivation:
o
EPCLE
III. Zero Threshold
Pollutants Estimated Per-
missible concentrations
Health Effects
Basic Data:
°TLV (oncogenlclty)
°NIOSH Recommendation
(oncogenlcity)
"Adjusted Ordering
Number (species,
route of adminis-
tration, lowest
effective dosage
as carcinogen or
as teratogen)
Derivation:
°E-AC1
OEPCAca
X.
Basic Data:
°EPC
ACU.2)
°*~AT
Derivation!
o
EPCwc
o___
EKm
Basic Data:
°-v
X*
Derivation:
"Sc
°EPCLT
-------
EPCs and MATEs are based on data of diverse origins
(Table 29). Currently available EPCs and MATEs are based on
threshold limit values (TLVs), current and proposed ambient
standards, criteria, regulations, and recommendations;
toxicityr carcinogenicity, mutagenicity, and teratogenicity
data; and nuisance factors. Additional MATEs are expected to
be developed from data-on technological factors in the future
(Kingsbury, G.L., EPA, to Kangas, M.J.r personal communica-
tion) .
To keep the MEG values conservative, simplified "worst
case" assumptions were made in deriving algorithms for speci-
fic EPCs and MATEs. In general, MATE values for a substance
are higher than the corresponding EPC value, and both values
are higher than the reported natural background concentra-
tions.
A standard nomenclature is used to identify MATEs and
EPCs that indicate the media and type of effect addressed,
data used to derive MEG value, and model number. The sub-
scripts and their meaning are:
A Air
W Water •
L Land
H Health effect
E Ecological effect
C Carcinogenicity
T Teratogenicity
S Standard or criteria
1-4 Reference to specific models or algorithms
a MEG expressed in ppm instead of micrograms/
per liter, gram, or cubic meter
In this discussion, it is assumed that the MEGs are used
primarily in the EPA project as scaling factors for ranking
pollutants or emission streams for a source-oriented ambient
monitoring program. Furthermore, this preliminary evaluation
does not include a comparison of MEGs to other similar scaling
factors or priority ranking systems. Disadvantages associated
with the MEGs include the following:
The MEG methodology does not consider synergisms,
antagonisms, or other secondary pollutant associa-
tions.
Significant data gaps exist in MEGs for land, which
are based on simplified assumptions and air and
97
-------
water data. Modifications to the models for de-
riving land MEGs were adopted in February 1978, to
overcome the overly conservative aspects of the end
values published in the original volume (Kingsbury,
G.L., EPA, to Kangas, M.J., personal communication).
Human health effects are extrapolated from animal
toxicity data. Models and assumptions frequently
include arbitrary factors or constants, e.g., safety
factors used to translate animal LDsg data to
human health effects.
Inappropriate use of MEGs as a substitute for stan-
dards, criteria, or recommendations is possible, if
they are not recognized as being only preliminary
scaling factors.
No explicit provision is made for error estimates or
confidence intervals associated with MEG values.
MEGs are meaningful for primary pollution effects
only. If any changes in the chemical nature of a
substance occur in the effluent stream or ambient
environment, a different MEG value would have to be
referenced.
The overall level of confidence associated with any
MEG is no greater than the health effects or
toxicity data used in the particular model. All
standard criticisms applied to these data (inade-
quate or unspecified sample size, conditions under
which the data were obtained, etc.) apply also to
the MEGs.
Safety factor for MATEs is less than that for EPCs,
on a proportional basis, e.g., if an EPC is exceeded
by some fraction (%), the actual impact may be much
less than if a MATE were exceeded by the same frac-
tion (%).
Non-exceedance of a MATE may be mistaken to mean
that effluents do not have the potential of creating
problems in the ambient environment. However, MATEs
= TLVs for many substances, and effects can occur
below TLVs on some sensitive receivers. TLVs are
for healthy workers, not for the more sensitive mem-
bers of the general public.
MATEs for presumptive or suspected carcinogens do
not accommodate factors involving synergisms, co
carcinogenicity, promotion, and metabolic altera-
tions into more or less active metabolites.
Certain application factors used in MATE models have
been selected subjectively.
Values for solid waste (land) MATEs are simply
scaled from water MATEs, and some water MATEs are
scaled from air MATEs.
Some EPCs for land and water are secondarily based
98
-------
on health effects and animal toxicity data. Intro-
duction of additional arbitrary factors and con-
stants has the potential of compounding or magni-
fying error in derived values, e.g., EPO^m based
on EPC^H; EPCLH based on EPCWfl.
"Land" MEGs are scaled from water MEGs, which are
more likely to reflect biological effects since im-
pacts of pollutants in solution are probably more
serious than those generated by "dry" material.
Advantages associated with the MEGs include the following:
MEGs are readily available for many chemical sub-
stances likely to be components of effluents from
facilities utilizing advanced fossil fuel combustion
and conversion processes.
• The MEG methodology is currently being extended to
non -chemical degradants.
• Algorithms for deriving MEG values are designed to
address health and ecological effects in all media,
based on a wide variety of data. Data used to de-
rive published MEG values have been priority ranked
to use best available information (standards and
criteria, epidemiological data, TLVs and NIOSH
recommendations before animal toxicity data;
LD5Q[oral, rat] before other toxicity data, etc.)
MEGs for Zero Threshold Pollutants (carcinogens,
teratogens, and mutagens) are published for all
media and consider such factors as animal test
species, recorded human effects, etc.
Phytotoxicity, reported tainting levels, bio-
accumulation and potential biomagnification data are
incorporated in deriving MEGs when available.
When several data sets are available for a sub-
stance, each is used to calculate a MEG value, but
only the most stringent value is reported.
• Further development and extensions of the MEG
methodology is in progress. The existing system is
designed to use new information as it becomes avail-
able, without modifying models, although new models
may eventually be developed.
MEGs are designed for direct use in Source Assess-
ment Models (SAMs).
Selection of substances for MEGs emphasizes many of
the contaminants and degradants expected in PBC ef-
fluents.
In the event that the MEG methodology is not
utilized, the background information summarized in
the MEG charts is useful in priority ranking, selec-
tion of sampling and analysis methods, and Level 1,
2, or 3 source analysis.
99
-------
The MEG background information summaries are a signifi-
cant collection of data on chemical substances that could be
expected in the emissions from FBC processes. In the event
that a system of priority ranking of pollutants other than the
MEG/SAM model system is selected, the data in these summaries
may be directly applicable in the alternative system selected.
An alternative approach that has some similarity to the
MEGs developed' as a response to assessing applications for
permits for new or modified air pollution sources submitted to
the Texas Air Control Board (Price et al., 1979). The bases
for decisions associated with their permit review process were
ambient levels called Acceptable Public Exposure Levels (APEL)
derived from existing toxicological data to indicate accept-
able levels for short duration exposure. These levels were
derived by applying a reduction factor for healthy adult males
to available occupational health or toxicity data on each sub-
stance. A reduction of 0.03 is used when only slight-to
moderate effects are indicated and available toxicity data
indicate none of the following acute effects: irritation from
short-term exposure, sensitization, "relatively severe ef-
fects," or an occupational ceiling limit and exposure.. Other-
wise, a factor of 0.01 is used. The data bases reviewed'for
each substance include TLVs (ACGH), OSHA standards, NIOSH
recommended levels, and available oncogenic and genotoxic
information. Currently, the authors are compiling a list of
APELs, which could be used for priority ranking pollutants
produced in the effluents of FBC processes.
The main advantage of this approach is, however, that it
uses health effects data and compiles the original data and
comments into a clearly formatted list; its major drawback is
the subjective assignment of a reduction factor to be applied
to original health effects data to determine the APEL.
Walsh et al. (1978) derive a threshold effects concentra-
tion which is used to define a hazard index for assessing
limited exposures to environmental pollutants. The threshold
concentration (QL) is a value selected to establish a limit
for humans which should not be exceeded because of health
risks. Stated in terms of total-body, organ, or tissue
burden, QL involves identifying a human health endpoint such
as a statutory limit, standard organ level, total-body burden,
or specific disease state.
A second study applied this approach to cadmium release
from a smelter operation in Helena, Montana (Rupp et al.,
1978). The critical human health endpoint selected for the
test of the procedure uses the kidney cortex, based on a
review of cadmium toxicity. Therefore, 200 mg Cd/g was the
value selected for QL. Utilizing human organ effects to
establish an ambient level goal may not always be possible,
100
-------
due to the lack of effects data for many organic pollutants
and less toxic organics. In calculating QL, however, a
thorough consideration is given to the physiological prop-
erties of the receptor, chronic exposure, and tissue bio-
accumulation. While too detailed as a preliminary screening
process, this approach could be valuable once a specific
facility utilizing the FBC process, site factors, and receptor
populations have been identified.
A number of systems utilize a rank order number to indi-
cate potential receptor damage (Ingraham and Zechel, 1979;
Jones, 1978; TSCA Interagency Committee, 1977).
Ingraham and Zechel (1979) develop a hazard index for a
substance based on expected impacts on an individual human
receptor, incorporating this value into a simple model to
indicate the potential hazard a particular source presents to
the population in nearby areas. This approach could poten-
tially be utilized for sensitive ecological receptors.
The hazard is a percentage rank estimated subjectively
from a review of medical reports of damage. Four percentage
categories were used by the authors in comparing the relative
receptor damage from lead and cadmium: 25, 50, 75 and 100.
In this case, lead was assigned value of 50 and cadmium a
value of 100, since medical reports indicated cadmium damage
to humans is considerably more severe than that of lead.
This approach has several disadvantages. If a large
number of substances have to be considered, assigning relative
percentages would become almost impossible. The manner in
which relative degree of hazard to receptors is estimated from
a review of the relevant literature is not given by the
authors. Finally, an index based on actual data on the damage
to potential receptors would be more desirable; it would
represent a level which could be compared to an expected
emission rate or an environmental concentration.
Some of the systems used for deriving threshold concen-
trations for single pollutants are summarized in Table 30.
The approaches estimating values for effects threshold concen-
trations utilize available information more effectively than
systems deriving a single rank value to indicate potential
hazard to receptors.
Identification of Emission Rates and Ambient Level Concentra-
tions
The second data elements needed for priority ranking are
emission rates or ambient level concentrations for each un-
regulated pollutant. Threshold levels for potential effects
101
-------
TABLE 30.
APPROACHES TO ESTIMATING AMBIENT LEVEL GOALS (ALG)
WHICH FUNCTION AS MINIMUM POTENTIAL HAZARD LEVELS
IN THE AMBIENT ENVIRONMENT
Reference
Method of deriving an
ambient level goal
Cleland and Kingsbury, 1977
Kingsbury and White, 1979
Price et al., 1979
Walsh et al., 1978
Rupp et al., 1978
Ingraham and Zechel, 1979
Simple models are applied to
all relevant health and eco-
logical hazard data
Simple reduction constants are
applied to all relevant health
hazard data
Total exposure burden/dose
limit is calculated for human
receptor organ most sensitive
to pollutant.
Simple relative rank order'
value (a percentage) is subjec-
tively assigned to all pollu-
tants considered based on a
literature review of known
health effects.
can be adjusted by a factor representing production expected
from the source. It is assumed that this parameter is esti-
mated during the analysis of the FBC process that occurs at the
beginning of the monitoring design procedure. Otherwise,
emission rates or environmental concentrations must be assumed
from a hypothetical scenario. In principle, as data on ef-
fluents from many demonstration level FBC facilities become
available/ scaling factors can be derived to project emission
rates of specific pollutants from commercial-scale plants.
In addition to estimating potential ambient concentra-
tions for each pollutant addressed during priority ranking,
careful consideration of the emission rate expected from the
FBC process with respect to background ambient levels is re-
quired. If expected emission rates are significantly below
ambient background levels, serious problems of measurement may
result. This effect must be taken into account in deciding
whether monitoring this pollutant will be technically feasi-
ble.
102
-------
Three basic approaches were identified in a review of the
literature on priority ranking substances based on their po-
tential for causing environmental effects. (They are sum-
marized, with appropriate references, in Table 31.) Each
approach is reviewed briefly in the following paragraphs.
TABLE 31. APPROACHES TO ESTIMATING AMBIENT ENVIRONMENTAL
CONCENTRATIONS EXPECTED TO IMPACT HUMAN
AND ENVIRONMENTAL RECEPTORS
Reference
Method of deriving
expected ambient concentrations
Walsh et al. , 1978
Behar et al, 1979
Rupp et al. , 1978
Anderson et al., 1978
Eimutis, 1976
Reznik et al. , 1978
Berkowitz et al. (In
Jones, 1978)
TSCA Interagency Testing
Committee, 1977
Authors use site-specific atmo-
spheric, terrestrial and aquatic
transport models to estimate
human exposure and physiological
transport models to estimate
organ exposure concentrations
Ambient level concentrations
are estimated by simple disper-
sion models for gaseous, liquid
and solid discharges from fossil
fuel combustion and conversion
facilities.
"Two reservoir" model is used
to evaluate the rate of entry
of water pollutants to the
aqueous environment.
Behar et al. (1979) present the conceptual framework for
modeling the flow of a quantity of a pollutant through all en-
vironmental media once it leaves a source, but give no
specific mathematical models. Walsh et al. (1978) provide
specific mathematical formulations utilizing a similar ap-
proach.
The models proposed by Walsh et al. are implemented for
cadmium transport by Rupp et al., 1978. Walsh et al. offer
the most detailed approach available, one which has the poten-
tial for most closely estimating the actual concentrations of
a substance which may impact a specific receptor. Two serious
problems, however, are that models cannot be simplified to ac-
commodate missing data; and that the models, in their present
103
-------
form, do not allow probability statements to accompany esti-
mates of exposure concentration.
Anderson et al. (1978), Eimutis (1976), and Reznik et al.
(1978) also use a multimedia approach but employ simpler
modeling methods to estimate quantities of a pollutant
generated by a source. These simpler approaches are valuable
when information availability is low.
An example of the many other studies relevant to tracing
environmental concentrations is the approach developed by
Berkowitz et al. (Jones, 1978), which should be reviewed for
modeling ambient environmental concentrations of a pollutant
in aquatic environments. Dispersion models for predicting
movement of airborne pollutants could also be used to estimate
potential exposure concentrations when appropriate emission
data are available.
Methods of Combining Effects and Concentration Values to Estimate
Hazard Potential
A method of combining appropriate data on potential ef-
fects and expected ambient concentrations is desirable for
estimating the hazard potential of each pollutant. Where rank
values indicate potential effects and ambient concentrations,
various weighting algorithms are used to estimate potential
hazard (Jones, 1978; TSCA Interagency Testing Committee,
1977). When expected concentrations are used directly, some
form of ratio of the expected environmental concentration or
emission rate to potential effects concentration is used to
indicate potential damage to receptors (Petrie and Kingsbury,
1978; Reznik et al., 1978; Walsh et al., 1978; Rupp et al.,
1978).
A method utilizing assigned rank values was developed by
the TSCA Interagency Testing Committee to select substances
which should be added to the inventory list of substances
required under Section 4(e) of the Toxic Substances Control Act
(TSCA), Public Law 94-469. The approach rank orders chemicals
for priority in testing to determine their effects on human
health and the environment. The initial priority of the jth
chemical is given by the equation:
Wi i (17)
104
-------
Where:
wi = weight assigned to the itn factor
fij = the itn factor score for the j tn chemical
Si = a scaling factor chosen to normalize the
factor scores.
The following factors were considered:
Annual national production of the chemical
Environmental burden of the chemical in terms of:
The quantity released into the environment
nationwide
The persistence of the chemical
• Occupational exposure
Extent to which the general population is exposed,
which involves four subfactors:
Number of people exposed
Frequency of exposure
Intensity of exposure
Penetrability (i.e., ability of the chemical to
enter the body)
The chemicals that scored high were further evaluated by
the method described below (unless they were eliminated for
another reason, e.g., already a regulated substance). Eight
factors were considered with each factor having a possible
numerical value of from 0 to 3. The values for the eight
factors are then summed to obtain the priority value for the
chemical in question. The factors are:
Carcinogenicity
Mutagenicity
Teratogenicity
Acute toxicity
Other toxic effects
Bioaccumulation
105
-------
Ecological effects (includes toxic effects on
non-human organisms and effects on ecosystems,
atmosphere, and climate)
Hazardous contaminants and environmental degradation
or conversion products
The TSCA scheme is not entirely appropriate. First, it
includes factors to evaluate a chemical's annual national pro-
duction by chemical manufacturers and its release by all users
of the chemical. While this is essential to TSCA, it is inap-
propriate for selecting pollutants for ambient monitoring of
an FBC process. Second, the ecological factors considered do
not include site-specific considerations, such as the evalua-
tion of probable effects on socially valuable species in the
immediate vicinity of the plant.
An example of the second approach to estimating hazard
potential are the Source Assessment Models (SAMs) developed
for the EPA. At least two SAMs, developed by Acurex Corpora-
tion, are currently under consideration by the EPA: SAM 1 and
SAM 1A. Although still in draft form, SAM 1 can be used to
illustrate the approach. This model considers peak ambient
concentrations and derived emission level goals based on goals
for peak allowable ambient concentrations (EPCs) (Anderson et
al.r 1978). The model proposed for SAM 1 defines the degree
of hazard for a pollutant species as:
Pollutant
u - cna^-iAd - _£ - Discharge Pollutant Concentration
Hf - Species — (18)
Degree of K ALG Discharge Level Goal
Hazard
Discharge level goal = Emission level goal = K ALG m
Where
K = dispersion or _ Source Stream
dilution factor Dilution Factor
C = Discharge level concentration
X = Ambient concentration
ALG = Ambient level goal (e.g., an EPC)
ALGm ~ Smallest of EPC values
The SAM 1A model priority ranks pollutants according to the
ratio of the total emission rate for an individual chemical from
106
-------
all air or water pollution sources at a facility, divided by the
MATE value for the chemical (Schalit and Wolfe, 1978). Priority
ranking based on the SAM 1A system is independent of the re-
sulting ambient concentration. It assigns the same relative
priority to an emission of 5 grams per second of beryllium if it
were emitted 200 meters above the ground or at ground level, re-
gardless of the fact that the ground level emission would cause
higher ambient concentration. This system tends to consider the
total burden of a pollutant in the environment and how wide-
spread the impact.
Under SAM 1A, pollutants would be assigned higher prior-
ities according to increasing values of the toxic unit discharge
rate for the pollutant (TPj, where i identifies the pol-
lutant) :
n n
P
TP. = V* H., Q. = > .„ ifc,, . Q (19)
i £_j ik k £_j MATEifc k
Where
i = pollutant
k = effluent stream
n = number of effluent streams
cik ~ concentration i'th pollutant in k'th effluent
stream
= MATE value for i'th pollutant in the medium
corresponding to k (8 hour MEG goal)
Qk = volume rate of flow in effluent stream k
u s Trs-s^r =- Hazard factor i'th pollutant,
lk m*E k'th stream
This procedure would be performed separately for those
MATEs based on ecological effects and those based on health
effects, in order to priority rank each separately, according
to effects. If a priority ranking based on both types of ef-
fects is desired, the smaller of the MATEs (based on health
versus ecological effects) could be used.
107
-------
In principle, pollutants in air and water effluents could
be priority ranked together by adding the TPj's in air and
in water. This might be useful for multimedia health-related
monitoring programs—similar to the type of program in the EPA
Las Vegas Integrated Exposure Assessment Monitoring Program
(Behar et al.f 1979). It is also possible to priority rank in
terms of a MATE based on carcinogenicity only, toxicity only,
or any other basis for which separate MATE values have been
calculated. •
i
A distinct advantage of SAM 1A is that it is currently
being applied to effluent data from fluidized bed boilers and
from coal gasification plants by EPA contractors. Priority
ranked lists of pollutants using this system are available for
some pilot-scale facilities (Anderson et al., 1978; Murthy et
al.f 1977; Murthy and Henschel, 1978).
The SAM models have an important limitation. The degree
of hazard estimated by a SAM model is influenced by a MEG
value. If the MEG value is known to be invalid, the estimated
hazard potential is also questionable. Likewise, the hazard
index of a substance estimated by any other procedure could be
questioned if evidence indicates that available data under-es-
timate potential effects or ambient concentrations. SAMs and
MEGs are meant only as a "first iteration" priority ranking
system, a mechanism to make an intelligent guess on what
should be monitored, what is hazardous and what is not, and
the relative degrees of hazard.
DEVELOPMENT OF A DECISION RULE TO SELECT POLLUTANTS FOR MONI-
TORING BASED ON THEIR POTENTIAL FOR ENVIRONMENTAL HAZARD
Two approaches were considered to decide whether a pollu-
tant should be considered further for monitoring, the princi-
pal assumption being the existence of a hazard index incorpo-
rating effects and expected concentrations or emission rates.
The first approach considers each pollutant independent of all
other pollutants generated by a source. If the expected emis-
sion rate or environmental concentration exceeds the estimated
minimum effects threshold concentration or emission rate, the
substance is considered further for monitoring.
An alternative approach involves ranking all pollutants
by hazard index value and considering the pollutants having'
the largest hazard values for monitoring. The group of Cate-
gory III pollutants can be handled as a population or sample,
and appropriate statistical parameters of the array selected
as possible cutoff points for the decision to monitor. The
median, for example, would segregate the array symmetrically,
with half the pollutants ranking above the median and half be-
low. Instead of using a statistical parameter as a cutoff
108
-------
point in the array as a decision criterion, "orders of magni-
tude" values could be selected (e.g., greater than 1, between
1 and 0.1, between 0.1 and 0.01).
These alternatives offer objective ways of selecting a
pollutant for monitoring when no clear criteria are avail-
able. Other alternatives are possible, and may in fact
address monitoring objectives more effectively. Whatever sys-
tem is selected, professional judgment must be balanced with
explicit rules or guiding principles in selecting unregulated
pollutants for monitoring.
PREPARATION OF THE PRELIMINARY LIST OF POLLUTANTS TO BE MONI-
TORED
After categorizing and priority ranking of all pollutants
are completed, all support parameters required to interpret
Category I, Category II, or potentially hazardous Category III
pollutants are identified. A combined listing of all of these
substances and parameters is compiled as a basis for further
refinement in the selection process. At this next level of
analysis, availability, cost-effectiveness, accuracy, and
quality assurance requirements of sampling and analytic
methods for. each substance and its support parameters are con-
sidered. This final phase of the selection process is
presented in the next section.
FINAL SELECTION OF POLLUTANTS BASED ON TECHNICAL AND COST CON-
SIDERATIONS
The first phase of this priority ranking scheme has de-
veloped an outline for 1) chemically characterizing source
effluents, 2) identifying potential health and ecological
problems associated with potential pollutants, and 3) estab-
lishing the potential for exceedance of existing or proposed
standards or other threshold criteria. These determinations
were used to select a preliminary list of pollutants for moni-
toring. This section extends the priority ranking procedure
to evaluate the need for source-oriented ambient monitoring
with respect to cost and technical problems and alternative
approaches to monitoring sources.
The first part of the priority ranking scheme allowed one
of the following conclusions:
Insufficient data are available to determine with
any degree of confidence that one or more substances
need to be monitored,
Sufficient data are available to show that one or
more substances do not require monitoring, or
109
-------
Sufficient data are available to support a decision
that a substance needs to be monitored.
An action plan to support the first conclusion would
begin by exploring appropriate procedures for effluent charac-
terization. Any monitoring other than that required for con-
struction permits would be delayed until an evaluation of the
facility's emissions is completed.
The second conclusion, with regard to specific substances
in source effluents, requires no continuous monitoring of the
FBC process. Periodic source checks are valuable, however, to
detect an increase in these substances to problem levels. A
standard plan to monitor during unanticipated process upsets
could also be developed under this conclusion.
If the analysis used in the priority ranking procedure
suggests that a substance should be monitored, an appropriate
action plan could be developed only after additional analysis
is completed. This additional analysis is the subject of the
following discussion.
Technical Quality of Available Analytical Methods
It is assumed that a preliminary sampling plan has been
developed which specifies the frequency, location, and
intensity of sampling for each problem pollutant produced by
the facility. This information is required to specify the
analytical methods appropriate for the monitoring program.
Technical quality of available analytic methods must be
considered as part of the selection process. For many pollu-
tants, if an EPA-approved method is applicable in the situa-
tion being considered, this will eliminate the need for
further consideration of other technical methods. However,
care must be taken to ensure that the approved method is
indeed applicable (e.g., no interferences, equipment available
for required sampling locations and frequencies), before the
resultant cost can be estimated for the overall monitoring
program. The following text, with reference to Figure 23,
presents the technical aspects in detail.
As shown in Figure 23, the pollutants and parameters con-
sidered for monitoring are divided into two groups:
chemical pollutants and physical and support parameters. The
major portion of the following discussion focuses on method
selection procedures for chemical pollutants.
Chemical pollutants for which EPA-approved methods are
indicated by regulations, permit conditions, or program objec-
tives (Step 2) must be considered, together with anticipated
concentrations of interfering co-pollutants (Step 3). If a
110
-------
PARAMETER IS A PMTSICA'. POUU1AHI
Oft A SUPPORT PARMEH*
ABE AN< OF TU£ APPROVED
HETMOOS IKSEHSlllYE TO
ANTICIPATED COHCEH7RAT10N5
Of IKTERfERIPIC CO-POUUTAKTS
AND FIELD COCOl T IONS'
IS THE KEFERCNCE METHOD
(IF THERE IS ONE AKD IT IS
IKTERrCKEHCE-FRCE)
iinouiTH nail EXPEHSKE
THAN «N> OF THE CERT If ICO
EquIVAltNT HETHOOS?
U5.E Of THE HOST COS7-
£rr£CT|VE EQUiyALCNr METHOD
(FOR A(R) OR OTHER APPROVED
«IHOO IS IHtfiCftUD
IS THERE AN CPA APPROVED tCIHOQ
8V R£GUIAtlWt OR KftHIT COHDU10M,
OR nwtwiro BY PKQJECI OBJECTIVES?
ELIMINATE N.L THOS£ CAHDIOAIt
HElHOOS EXCESSIVELY SENSMIKE TO
AHTIC1PATID CONCENTRAriOHS Of
1NTERFERIW CO-POLLUTANTS OR
PRQCtSS CONDITIONS
ust of THE HOST cosi-trftcTivt
eQyUAlEHT ntTHOO, OS OTHER
APPROVED METHOD. IS INDICATED
ELIHltlATE ALL REMAIN If*. CAlJDIUAtE
HETrfOOS WHICH ARE NOT CfJMPAtieiC
MUM PREV10USLT UUCKO TlHf ASPECT
Of LEKEt Of IWEHSITt
Figure 23. Methods selection-technical aspects.
-------
permit or regulatory requirement is involved, the appropriate
enforcement office personnel must be informed. If the ap-
proved methods are inaccurate due to interfering co-pollutants
or field conditions, the EPA should be contacted for permis-
sion to use an alternative method (Step 4).
If the EPA-approved method is appropriate, its relative
cost should be compared to certified equivalent methods (Step
5). If the approved method is not significantly more expen-
sive, it should be used (Step 6). Otherwise, use of the most
cost-effective approved or certified alternative is indicated
(Step 7).
In the event that EPA-approved methods are not indicated,
the desired accuracy of data for the particular stage of the
project in question must be ascertained (Step 8). Standard
methods as tested and approved by various professional organi-
zations (ASTM, NBS, USGS) should be used if high accuracy is
desired (Steps 10 and 11), with preference given to any stan-
dard methods which EPA may approve in the near future (Step
13). If no standard methods are available, EPA Laboratories,
ASTM committees, universities, equipment manufacturers, or
other research institutions should be consulted (Step 12).
For various trace materials, particularly organics, sophis-
ticated research methods and accurate equipment are avail-
able. Appropriate consultation will reveal not only possible
methods for the program but also the availability of the
equipment and trained personnel.
Various "range-finding" methods may be adequate when
highly accurate data are not necessary, even if these methods
are not sensitive or reliable enough to gain the approval of
EPA or other groups (Step 9). Anticipated concentrations of
co-pollutants which might interfere must also be considered
for any non-EPA approved methods (Step 14).
With reference to physical and support parameters, appro-
priate EPA-approved methods should also be used (Steps 19, 21,
22, and 25). Where no standard or EPA-approved methods exist,
certified equivalent methods should be identified through con-
sultation with NBS, ASTM, or other sources (Step 23).
In the final steps of the methods selection, after ac-
ceptable candidate methods have been identified for each type
of parameter, any method which is not compatible with the
previously chosen sampling interval should be eliminated (Step
15). Methods lacking needed accuracy or reliability under
actual operating conditions can also be eliminated (Step 16).
Use of the most cost-effective remaining methods is indicated
(Step 17).
112
-------
Cost Considerations
An adequate monitoring plan carefully considers selection
of parameters, methods, and sampling intensity in meeting pro-
gram goals. Cost enters the decision-making process through
selection of methods and sampling intensity. Unnecessarily
intense sampling or use of overly sophisticated analytical
methods may generate excessive total program costs.
Cost aspects of methods selection, displayed in Figure
24, involve deciding whether to use in-house or contracted
services. If contracted services are considered (Steps
25-27), care should be taken to specify required method, ac-
ceptable alternatives, and level of accuracy (Step 25). The
sampling route and schedule should be analyzed to identify for
contract proposers the possible economies of scale resulting
from multiple-sample lots of multiple-parameter methods (Step
26). Preliminary cost estimates can then be solicited from
qualified contractors (Step 27).
For estimates of in-house monitoring costs (Steps 2-37),
two general categories of methods are distinguished: instru-
mental and non-instrumental (Step 2). Instrumental methods
include electronic sampling and analyzing equipment and other
automated analytical methods; non-instrumental methods refer
generally to "wet chemical" techniques.
For instrumental monitoring methods, the cost of pur-
chasing or leasing various makes of satisfactory instruments
is obtained first (Steps 3-4). For the use of each instru-
ment, estimates are made of:
Purchase or lease costs of the instrument
Purchase or Tease costs for all necessary ancillary
equipment of requisite accuracy and reliability
(e.g., calibrators, data loggers, and computer link-
ages) (Steps 5-8)
Costs of expendable supplies for the operation of
each set of primary and ancillary instruments (in-
cluding computer time) (Steps 9-10)
Costs of each type of appropriate housing needed to
protect instruments from the elements, vandalism, or
theft; to control temperature, humidity, and light
conditions if required; or to provide mobility
(Steps 11-12)
Costs for all utility needs for each instrument set,
including service lines, on-site generators, bottled
113
-------
IS METHOD IK QUESTION
FUNDAMENTALLY AN
INSTRUMENTAL METHOD?
ARE CONTRACT SERVICES TO BE
CONSIDERED FOR SERVICES FOR
KHICN THERE CUSTOMARILY !S
A PRICE SCHEDULE («.?.,
ANALYTICAL CHEMISTRY)
CONSIDER ALL HAKES OF
INSTRUMENT HAVINO. THE
REQUIRED TECHNICAL
4
OBTAIN PURCHASE OH
LEASE COST FOR EACH
INSTRUMENT
OBTAIN Oil DEVELOP A
PROTOCOL FOR THE USE
OF EACH HAKE OF
INSTRUMENT
DETERMINE ALL ANCILLARY
EQUIPMENT NEEDED TO USE
INSTRUMENT (l.q.,
CALI8RATORS. DATA LOGGERS,
COMPUTES LINKAGES, ETC.)
CONSIDER ALL MAKES OF
ANCILLARY EQUIPMENT HAVING
COMPATIBILITY ylTH
INSTRUMENTS), RELIABILITY,
ACCURACY, ETC.
DETERMINE KINCS AND
AMOUNTS OF EXPENDABLE
SUPPLIES NEEDED TO USE
EACH SET OF INSTRUMENT
AND ANCILLARY EQUIPMENT
(INCLUDING COMPUTER TIME)
DETERMINE INSTRUMENT
HOUSING NEEDS FOR
PROTECTION FROM THE
ELEMENTS, VANDALISM,
THEFT; FOR CONTROLLED
ENVIRONMENT FOR
SENSITIVE INSTRUMENTS;
FOR MOBILITY
DETERMINE UTILITY
NEEDS FOR EACH
INSTRUMENT SET (».g,,
ELECTRIC POKER, TELEPHONE,
UATER, GAS)
• SEED TO INSTALL SERVICE
LINE(S)
• NEED TO GENERATE
ELECTRICITY ON STATION
• NEED TO USE BOTTLED
GAS OR UATER
• NEEO FOR RADIO-
TELEMETRY
DETERMINE EQUIPMENT NEEDS
CONSIDER ALL MAKES HAVING
THE REQUIRED TECHNICAL
CHARACTERISTICS
DETERMINE NEEDS FOR ANY
SEMI-PERMANENT INSTALLATION
OF EQUIPMENT AT SAMPLING
STATION(S), {I.e., MARKERS,
HELLS, ETC.)
ESTIMATE INSTALLATION
FOR EACH CANDIDATE METHOD,
DETERMINE THE FOLLOWING
MANPOWER COSTS
ESTIMATE SAMPLE COLLECTION TIME:
ORGANIZATION TIME
TRANSPORTATION TIKE TO FIELD
WORK TINE ON STATION
TRANSPORTATION TIME BCTUEEN
STATIONS
MARGIN FOR INCLEMENT UEATHER
TIMS FOR DELIVERY OF SAMPLES
TO LAB
TIME FOR PROPER CUAN-UP AND
STORAGE OF FIELD EQUIPMENT
ESTIMATE COSTS OF:
• PURCHASE, LEASE Q« CHARTER
• MILEAGE
• MAINTENANCE
• INSURANCE
DETERMINE TRANSPORTATION NEEDS
ASSOCIATED UITH EACH METHOD
OPTION:
• ORDINARY AUTOMOBILE OR VAN
. SPECIAL VEHICLE
• BOAT
• AIRCRAFT
ARRAY TOTAL HANPOUER COSTS
FOR EACH OPTION METHOD
• PROFESSIONAL
. TECHNICIANS
• INDIRECT SUPPORT PERSONNEL
ESTIMATE PROFESSIONAL SUPERVISORY
AND INTERPRETATION^ TIME
ESTIMATE ANALYTICAL TIMS:
TEST PREPARATION TIME
ACTUAL TEST TIME
CLEAN-UP TIME
QUALITY CONTROL TIME
ECONOMIES OF SCALE
,
ESTIMATE INDIRECT SERVICES
IME:
TYPISTS
KEYPUNCH OPERATORS
JANITORS
OTHERS
ESTIMATE INSTRUMENT CALIBRATION
AND MAINTENANCE TIME:
• ROUTINE SCHEDULE PLANNED
• EXPECTED INSTRUMENT
RELIABILITY
CONSIDER REDUCING NUMBER
OF PARAMETERS, INTENSITY
OF SAMPLING. AND/OR LEVEL
OF ACCURACY (ANY SUCH
CHANCE MAY IMPLY A
CORRESPONDING CHANGE IN
MONITORING OBJECTIVES)
($«« II III
ESTIMATE METER READING TIME:
• TRANSPORTATION TIME TO FIELD
• WORK TIME ON STATION
• TRANSPORTATION TIME 8ETUEEN
STATIONS
Figure 24. Methods selection-cost aspects.
114
-------
gas or water, and radio-telemetry (Steps 13-14).
(Service roads may also be necessary.)
For non-instrumental monitoring methods, after equipment
needs are determined (Step 18), estimates can be made of:
Purchase or lease costs for each type and make of
equipment (Steps 19-20)
Costs of expendable supplies (Steps 21-22)
Costs of obtaining and installing any semi-permanent
facilities at sampling stations (e.g., markers,
wells, service roads, and automatic samplers) (Steps
23-24)
These hardware costs for both instrumental and non-in-
strumental methods can then be arrayed and compared (Step 15).
The effort required to obtain cost estimates for all
potentially applicable instruments (Step 4), equipment options
(Step 20), or contractors (Step 27) is often very time-con-
suming and/or impractical—there may be a dozen or more sup-
pliers of the more common types of equipment and services.
However, as many cost estimates as reasonably possible should
be acquired to ascertain an accurate range of costs.
If the equipment is to be purchased by competitive bid-
ding, all identified potential suppliers will bid on supplying
the necessary instrumentation and services at the lowest
cost. When writing specifications for purchases through com-
petitive bidding, it is important to note that, while the
project is to obtain material which meets specifications at
the lowest possible cost, some suppliers may minimize the
price of their bids by proposing equipment that only mar-
ginally meets the minimum requirements in the specifications.
Therefore, a vital part of writing specifications for equip-
ment involves various technical considerations regarding the
choice of the method of monitoring and the internal cost esti-
mates, and such factors as:
• The availability of spare parts on short notice
The availability of maintenance and repair contracts
from the equipment supplier and the types of support
services available
The amount of protection available through the war-
ranties
Whether the equipment has been successfully used in
similar projects
115
-------
The ability of the equipment to interface with
equipment on hand and on order
Exceptions to the specifications
The maintenance schedule and the apparent ease of
equipment calibration, operation, and maintenance
The delivery date
Bidders for major pieces of equipment and types of ser-
vices should list references to independent organizations
which have used their equipment and services. These users
should be contacted and their viewpoints considered in the
final selection.
For both instrumental and non-instrumental methods, man-
power, transportation, and related costs may contain a
substantial portion of the costs of the monitoring plan.
There- fore, a careful examination of these costs must be made
when developing the plan.
Manpower costs (Steps 16-33) require consideration of the
following types of expenses as appropriate for each candidate
method:
Sample collection time—organizational efforts,
portal-to-portal trip, clean-up, and contingencies
(Step 17)
Meter reading time—portal-to-portal trip (Step 28)
Calibration and maintenance—routine and unexpected
(including downtime during service by manufacturer's
representative) (Step 29)
Indirect services time—clerical, maintenance,
training, and support (Step 30)
Analytical time—preparation, analysis, clean-up,
quality control; consider schedule of analysis for
economies of scale (Step 30)
Professional supervisory and interpretational time
(Step 32)
These manpower costs for each alternative method can then
be arrayed and compared (Step 33).
Transportation costs (Steps 34-36) for publicly owned and
private vehicles include:
116
-------
Purchase, lease, charter
Mileage and depreciation
Maintenance
Insurance
These, too, can be arrayed by alternative and compared.
Hardware, manpower, and transportation costs can then be
combined and compared for each alternative (Step 37).
If monitoring service is contracted, all of the factors
discussed above will be the concern of the contractor. The
methods and level of accuracy needed are stipulated, and cost
comparisons are made among various contractors1 estimates
(Steps 25-27).
Finally, all costs must be considered in the context of
the entire monitoring scheme (Step 38). Savings in cost are
sometimes possible if ambient environmental monitoring person-
nel and instrumentation requirements overlap with OSHA and any
process monitoring resources. For example, the process work
may require an analytical chemist for two-thirds of his time
and the monitoring work may require a chemist on site
one-third of the time; or, OSHA monitoring could require in-
strumentation and personnel which could be useful when mea-
suring fugitive emissions. Where possible, these savings
should be factored into cost estimates.
If anticipated costs for monitoring are too high (Steps
42- 45), reductions of the following should be considered:
Number of pollutants and parameters
Intensity of sampling
Level of accuracy
Monitoring of lower-priority pollutants should be reduced
or eliminated first. The proposed reductions are then con-
sidered in terms of the original program objectives. If the
objectives can be met with reductions, the design and cost
estimates should be revised, and the design considered com-
plete. If the objectives cannot be met, sufficient funds to
eliminate the need for reductions should be sought. If funds
can be obtained, the program, as designed, is complete. If
not, monitoring objectives must be redefined, and design and
cost estimates revised.
117
-------
CONCLUSIONS
This section focuses on the procedure for selecting pol-
lutants generated by a facility using a FBC process for moni-
toring. Two approaches are possible to meet this goal for a
particular problem pollutant:
Stack/pipe monitoring with appropriate dispersion
modeling
• Source-oriented ambient monitoring
The preliminary screening processes described in the
first part of this section allow identification of all legal
requirements for monitoring and problem pollutants. The tech-
nical and cost factors are useful in determining the appro-
priate mix of these alternative approaches to monitoring pol-
lutants. Additional factors which have not been discussed but
are nevertheless relevant include:
Adequacy of sampling procedures
Sensitivity and suitability of analytical methods
• Assumptions and models for relating source to am-
bient concentration levels
Applicability and limitations of stack monitoring
and modeling ambient pollutant concentrations
Finally, while ambient concentrations of some problem
pollutants may be minute, exposure levels of these substances
can remain potentially hazardous because of their bioaccumu-
lation potential, potential for synergistic effects, or
proximity to sensitive receivers. These substances present a
serious problem in the ambient environment because the best
available monitoring equipment may be incapable or detecting
expected concentrations or the ambient levels of harmless
chemical analogs may interfere with detection of the problem
pollutant. Undoubtedly, this problem will not be resolved
until FBC processes reaches commercialscale operation and
effects develop that are more readily measureable.
If these factors cannot be incorporated into the priority
ranking and selection procedure in an explicit and systematic
fashion, their presence should, at least, influence qualita-
tively the monitoring system design process.
118
-------
SECTION 4
AIR QUALITY MONITORING STRATEGIES
INTRODUCTION
PBC is a coal combustion process that, in respect to air
quality/ emits lower quantities of gaseous pollutants for the
same amount of coal burned than does a conventional pulverized
coal combustion process. However, FBC can potentially emit
concentrations and quantities of particulate matter higher
than a conventional coal-burning plant. Most of the ash
present in the coal feed to FBC is normally elutriated from
the bed and must be removed (along with attrited limestone and
other particulates) prior to release of the flue gas to the
atmosphere.
Except for the design of the combustor, FBC is basically
a variant of a conventional coal-fired plant, and therefore
should be subject to the same air quality standards as any
coal-fired plant. Air quality monitoring requirements and
methods applicable to conventional coal-burning plants are
equally applicable to FBC plants. Based on available theo-
retical and experimental information on air emissions from FBC
plants, commercial-size FBC units will probably meet New
Source Performance Standards for air emissions applicable to
coal-fired facilities.
The FBC air quality monitoring strategies proposed in
this section address primarily regulated pollutants, because
experimental data from FBC pilot plants show that these pollu-
tants will require monitoring. Based on available informa-
tion, no other pollutants are projected to cause serious air
quality problems or violate current air quality standards.
This section discusses the following important air qual-
ity parameters:
Air quality impact generators
Conditions where ambient monitoring may be required
Applicable regulations and monitoring requirements
Selection of pollutants to be monitored
119
-------
Air monitoring network design and dispersion models
Monitoring strategies for FBC installations
AIR QUALITY IMPACT GENERATORS
This discussion of atmospheric emissions from FBC of coal
will only cover flue gas emissions, as experimental data from
other sources of emissions are not available. Figure 5 in
Section 2 depicts the generic FBC process scheme, and Table 4
shows the .sources of atmospheric emissions. Not all of these
sources wi,ll be present in all commercial units. For example,
sorbent regeneration (and associated waste streams) will not
be present in the first commercial FBC units currently under
consideration. Furthermore, there could be some process
variations involving currently undefined effluent streams.
The plans for individual FBC units should be analyzed
case-by-case, to develop process specific air pollutant in-
ventories for all waste streams.
The nature and extent of fugitive atmospheric emissions
from FBC processes are undefined; whether there will be sig-
nificant adverse environmental impact generators of this type
cannot be predicted. Table 32 explores the sources of atmos-
pheric emissions for the FBC process. By expanding and modi-
fying the listing in this table for each commercial-scale
facility, a full air emission inventory can be developed.
Some sources listed may require careful and attentive mainte-
nance, not specialized monitoring and pollution control sys-
tems; others, occupational exposure problems contained at the
FBC facility, will fall under the jurisdiction of the Occupa-
tional Health and Safety Administration rather than under the
EPA.
CONDITIONS WHERE AMBIENT MONITORING IS INDICATED
The variations in processes which could be used and types
of pollutants which may be generated as a result of FBC tech-
nology are summarized in Section 2. The complex operations at
a facility employing FBC as an energy conversion process will
determine the need for and requirements of a source-oriented
monitoring program. Two basic types of atmospheric emissions
can be identified: stack emissions and fugitive emissions.
Each process associated with FBC technology should be
monitored for fugitive and stack emissions, and, as well, -for
wastes generated by processes not directly generated by an FBC
facility. In some cases, pollutant discharges may be altered
by contaminants introduced by the FBC process, e.g., in cool-
ing towers or cooling ponds gaseous or aerosol discharges
could contain enhanced or altered pollutant levels due to con-
tamination by wastes associated with the FBC process.
120
-------
TABLE 32. SOURCES OF ATMOSPHERIC EMISSIONS
Activity
Source
Coal Preparation and Feed
Coal Combustion and Associated
Pollution Control
Stac.k (Flue Gas) Particulate Disposal
Special Start-Up
Leaks
Sorbent Preparation and Feed
Spent Sorbent Discard
Shipping Coal to Plant
Coal Unloading
Fugitive Emissions During Storage
Loading of Coal to Preparation Plant
Crushing, Sizing, Washing, Drying, Fine
Particulate Removal or Preparation to Recycle
Materials Handling in Coal Preparation Plant
Coal Feed Hopper Backflow
Normal Stack (Flue) Gas
Stack (Flue) During Soot Conditions
Blowing Stack During Start-Up
Stack Gas Particulate Removal—Hopper Unloading
Stack (Flue) Gas Particulate Storage
Stack (Flue) Gas Particulate Recycling
Coal Feed
Materials Handling
Stack Gas Particulate Loading for Transport
Stack Gas Particulate Transport
Stack Gas Particulate Unloading and Processing at
Disposal Site
Stack Gas Particulate Storage at Disposal Site
Stack Gas Particulate Use in Commercial Application
Start-Up Combustion of Coal, Liquid Fuel, or Gas
Air Preheater (if any) with liquid or gaseous fuel
Fugitive Emissions from Leaks in Boiler and
Directly Associated Systems
Sorbent Shipping
Sorbent Unloading
Sorbent Storage
Sorbent Loading to Preparation Plant
Sorbent Preparation
crushing
sizing
washing
drying
fine particulate removal or preparation
for recycle
Material Transfer in Sorbent Preparation Plant
Sorbent Feed Hopper Backflow
Bed Solids Removal from Active Boiler
Used Bed Solids Storage
Waste Bed Solids Transfer for Shipping
Waste Bed Solids Shipping
Waste Bed Solids Unloading and Processing at
Disposal Site (or Use Site)
Waste Bed Solids Storage at Disposal Site
Waste Bed Solids Use in Commercial Application
(Continued)
121
-------
TABLE 32 (Continued)
Activity
Source
Sorbent Regeneration Unit Operations
Maintenance
Steam Turbine and Water
Treatment Unit Operations
Unloading Bed Solids to Regenerator
Bed Solids Regeneration Flue Gas
Regeneration Flue Gas Fine Particulate
Collector Hopper Unloading
Regeneration Flue Gas Shipment
Regeneration Flue Gas Fine Particulate Disposal
(or Use)
Regeneration and Sulfur Recovery
bed solids regeneration sulfur recovery
system stacks (stacks and vents)
- .sulfur unloading
Sulfur storage
Sulfur shipment
Sulfur unloading at disposal site
sulfur disposal or use
sulfur recovery fugitive emissions
Fuel for Regeneration
fuel shipment
fuel storage
fuel preparation
fuel combustion
ash collection
flue gas particulate collection
flue gas particulate disposal
ash disposal
special gases used for regeneration
Bed Shutdown Solid Waste Removal
Solids Bed Cleaning
Removal of Solids from Metal Cleaning and Scrapin?
Plant Maintenance Fugitive Emissions
cleaning
lubrication
repair
Cooling Tower Plume.(or Cooling Pond)
Gas from Turbine
stack
fugitive
Turbine Gas Pollution Control System Solids
Disposal
Boiler Slowdown Water Off-Gases
Cooling System Cleaning Gases and Suspended
Particulates
steam System Cleaning Gases and Particulate
Matter
Other Processes Using Steam from Boiler
Slowdown from Water Treatment Operations
Gas Turbine Fugitive Emissions
122
-------
Disposal of the solid wastes generated by the FBC tech-
nology would require at the very least a source monitoring
program during disposal operations and at the site after
covering the solids with soil or other material. Monitoring
programs at the disposal site should be directed towards mea-
suring amounts of fugitive emissions resulting from handling
of the solid wastes at the landfill site, and monitoring for
potential gases or fumes generated by bacterial or chemical
action upon the FBC solid wastes or combination wastes.
In the absence of evidence that they will not constitute
a significant problem, effluent streams should be monitored.
The need for testing should be determined case-by-case and the
reasons for not testing (and for testing) documented/ based on
the inventory of effluent streams. An effluent stream may not
require monitoring if:
Data from similar or related facilities indicate
that the process stream is not a problem. (The
applicability and limitations of applying data from
other sources to the specific installation must be
considered before such a decision can be adequately
justified.)
Emission factors for pollutants are not available.
The limitations of the data base and the need for
site-specific information should be considered in
this determination.
• Control systems included in the specific instal-
lation will render the impact of the effluent insig-
nificant. In this case, the need for verification
of the effectiveness of the control system, the im-
pact of control system failure, and of byproducts of
control should be considered.
In some cases, while it may be desirable to monitor an
effluent stream, there may be mitigating reasons why moni-
toring is not carried out:
It may be considered too expensive or not cost-ef-
fective.
The parameters to be studied are unmeasurable-by
current technology, or the accuracy of such measure-
ment is unknown.
123
-------
REGULATIONS
New Source Performance Standards
Federal New Source Performance Standards (NSPS) , pub-
lished by the EPA, for atmospheric emissions from all fos-
sil-fuel-fired plants are summarized in Table 33, which is de-
rived from 40 CFR Part 60.
Some states have regulations and individual emission
standards for boilers smaller than 73 megawatts that are
tighter than the above restrictions. Often, these regulations
are genera'ted on a case-by-case basis to assure compliance
with state-wide ambient air quality standards, Prevention of
Significant Deterioration (PSD) regulations, or emission off-
set regulations. During the following conditions, the NSP
Standards do not apply: during startup; during shutdown; or
during emergency conditions when sulfur dioxide emissions are
minimized by operating all flue gas desulfurization modules
that can be operated (with a spare module for units larger
than 365 megawatts (1,250 million Btu/hr) under some circum-
stances) .
Owners and operators of federally regulated FBC steam
generators must install, calibrate and maintain an S02 emis-
sion monitoring system and record its output. For electric
utility FBC units, an "as fired" fuel monitoring system is
also required for measuring potential S02 emissions before
control (S02 monitoring upstream of control equipment is not
applicable to FBC units for which S02 is controlled in the
combustion chamber) .
The owners and operators also must install, calibrate and
operate a continuous monitoring system for C02 or 02 at
locations where S02 is monitored. This allows the readings
of the S02 emission monitor to be converted to ng/J. For
units equipped with C02 monitoring, the formula is:
where :
E = Pollutant emission rate, ng/J
C = Pollutant concentration, ng/scm on a dry or
wet basis (Subscripts "d" or "w" are used to
note this)
124
-------
TABLE 33.
FEDERAL STANDARDS OF PERFORMANCE FOR NEW STATIONARY SOURCES OF
AIR POLLUTION
Source
category
Affected
facility
Pollutant
Emission level
Monitoring
requirement
Subpart D:
Fossil fuel fired
steam generators >73 MW
heat input (>250 million
Btu/hr), 63 Kcal/hr)
Coal-fired boilers
Particulate
Opacity
SO,
Anthracite,
Bituminous,
or Subbitu-
minous coal
Lignite
More than 25%
coal refuse
0.10 Ib/million Btu
20% (27% for 6 min/hr)
1.20 Ib/million Btu
0.70 Ib/million Btu
0.60 Ib/million Btu
Exempt
No requirement
Continuous
Continuous
Continuous
Subpart Da:
Electric utility steam
generating units >73 MW
heat input (>250 million
Btu/hr)
Coal-fired boilers
(and coal-derived
fuels)
Particulate
Opacity
SO,
100% anthra-
cite
N0_
Anthracite,
Bituminous,
and Lignite
Subbitu-
minous coal
Coal-derived
fuels and
shale oil
More than 25%
coal refuse
0.03 Ib/million Btu
20% (27% for 6 min/hr)
1.20 Ib/million Btu
and 90% reduction,
except 70% reduction
when emissions are
less than 0.60
Ib/million Btu
Percent reduction does
not apply
0.60 Ib/million Btu
0.50 Ib/million Btu
0.50 Ib/million Btu
Exempt
No requirement
Continuous
Continuous
compliance
Continuous
compliance
Subpart Y:
Coal preparation plants
Subpart HH:
Lime manufacturing
plants
Thermal dryer
Pneumatic coal
cleaning equipment
Processing and
conveying equipment,
storage systems,
transfer and loading
systems
Rotary lime kilns
Particulate
Opacity
Particulate
Opacity
Opacity
Particulate
Opacity
0.031 gr/dscf
(0.070 g/dscm)
20%
0.018 gr/dscf
(0.040 g/dscm)
10%
20%
Temperature
scrubber
pressure loss
Water pressure
No requirement
No requirement
No requirement
No requirement
0.30 Ib/ton
10%
Scrubber
pressure
loss; Scrubber
liquid supply
pressure
Continuous
125
-------
%CC-2= Percent carbon dioxide on a dry or wet basis,
depending on the subscript
Fc = A factor representing the ratio of the volume
of carbon dioxide generated to the calorific
value of the fuel combusted. F is defined as
follows:
for anthracite, Fc = 0.532xlO~7 scm C02/J
for subbutuminous and bituminous coal,
: Fc = 0.486xlCr7 scm C02/J
i
for lignite Fc = 0.516 x 10 ~7 scm
C02/j
special formulas are given for mixtures
of different fuels
In chemically active FBC units, some C02 is emitted and
some 02 is removed when limestone is converted to calcium
oxide in the boiler, causing a slight systematic underestima-
tion of the ng/J emissions when the above formulas are used.
These effects are of minor significance under normal operating
conditions, but should be considered, along with the errors in
monitoring, in borderline cases between compliance and
non-compliance with emission regulations.
All federally regulated FBC installations are required to
process the stack monitoring data and report any emissions
above those allowable by the standards. For electric utility
steam generators, the reporting of running 30-day average ng/J
emissions and percent reduction in potential emissions is also
required.
Prevention of Significant Deterioration
The Clean Air Act Amendments of 1977 contain compre-
hensive requirements for the prevention of "significant de-
terioration" of air quality in areas that have pollution
levels better than the national ambient air quality stan-
dards. Such areas may be classified as Class I, Class II or
Class III, according to the amount of deterioration of air
quality allowed. Table 34 shows the maximum a-llowable
increases. For any period, other than an annual period, the
applicable maximum may be exceeded no more than once per year
at any location. Allowable increases in annual average levels
may never be exceeded.
It is possible to obtain a presidential or gubernatorial
variance in Class I areas, providing that certain allowable
increments for S02 are not exceeded more than 18 times per
126
-------
TABLE 34. ALLOWABLE INCREASES IN AMBIENT POLLUTION LEVELS
UNDER PSD REGULATIONS
Area designations
Class-IClass.IIClass~III
Pollutant (yg/nr) (yg/mj) (yg/nr)
Particulate Matter
Annual Geometric Mean 5 19 37
24-Hour Maximum 10 37 75
Sulfur Dioxide
Annual Arithmetic Mean 2 20 40
24-Hour Maximum 5 91 182
3-Hour Maximum 25 512 700
Carbon Monoxide
Nitrogen Oxides
b
Ozone
Visibility0
a
Class I is defined by EPA as environmentally "pristine";
Class II as approaching EPA pollution maximums; Class III as
"all others."
Federal regulations currently under development.
Q
Visibility regulations were proposed by EPA in the Federal
Register (45 FR 34762) on May 22, 1980. Final versions have
not been promulgated but are expected to differ from the pro-
posed' versions. This phase of the regulations addresses
Federal Class I regions only. Future phases are expected to
address regional haze and plume controls.
year. The allowable increase in low terrain areas is a 24-
hour value of 36yg/m3 and a 3-hour value of 130
Ug/m3. In high terrain, it is a 24-hour maximum of 62
Ug/m3 and a three-hour level of 221 yg/m3. In addi-
tion, the air pollution levels may never exceed any applicable
ambient air quality standard (see Table 35).
127
-------
TABLE 35. AMBIENT AIR QUALITY STANDARDS
Pollutant
Duration
Restriction
Federal Standards
Primary
Secondary
Suspended
Particulates
Suspended
Particulates:
Annual
Meanb
24 hour mean
Concentration
Not to be exceeded 75
Not to be exceeded more 260
than once per year
60
150
Sulfur
Dioxide
Sulfur
Dioxide
Sulfur
Dioxide
Lead
Carbon
Monoxide
Carbon
Monoxide
Photochemical
Oxidants
Non-methane
Hydrocarbons
Nitrogen
Dioxide
Annual
Mean
24 hour mean
Concentration
3 hour mean
Concentration
3 month
Mean
8 hour mean
Concentration
1 hour mean
Concentration
1 hour mean
Concentration
3 hour mean
Concentration
Annual
Mean
Not to be exceeded
Not to be exceeded more
than once per year
Not to be exceeded more
than once per year
Not to be exceeded
Not to be exceeded more
than once per year
Not to be exceeded more
than once per year
Not to be exceeded more
than once per year
Not to be exceeded between
6:00 AM and 9:00 AM
Not to be exceeded
80 (.03)
365(.14)
-
1.5
10(9. 0)c
40(35.)°
235( .12)
160 ( .24)
100( .05)
"™
-
1300(.50)
1.5
10(9.0)°
40(35.)°
235( .12)
160 ( .24)
100( .05)
aUnless otherwise noted, values without parentheses are in micrograms per cubic
meter. Values in parentheses are in parts per million by volume.
Geometric mean. All other averages are arithmetic means.
cValues without parentheses are in milligrams per cubic meter.
All areas with air quality better than the ambient air
quality standards are classified as Class II (except for cer-
tain kinds of national and international parks) unless state
governments (or the Indian governing body, in the case of an
Indian reservation) follow specified procedures to have the
area reclassified, to allow for development. The S02 incre-
ments are applicable to the combined impact of all new sources
in an area operating after the date of the first application
for a permit subject to PSD regulations. Compliance with the
increments is generally determined by modeling, and compliance
with the ambient air quality standards is determined by super-
128
-------
imposing the modeled impact of the sources on observed back-
ground ambient air quality concentrations.
All new major emitting facilities must obtain a PSD per-
mit before they are built. Such facilities include new,
isolated FBC facilities with a heat capacity above 250,000,000
Btu/hr (when used as steam electric power plants or boilers)
which have the potential to emit more than 100 tons/year of
any regulated air pollutant when control equipment is opera-
ting in a reasonably anticipated manner. In other applica-
tions, the limit is 250 tons/year. All sources at an instal-
lation, including fugitive sources, may be included in the
total tonnage, at the discretion of the EPA Regional Adminis-
trator. No current EPA regulations explicitly cover atmos-
pheric fugitive emissions from coal combustion solid waste
disposal operations.
If an FBC unit is built at an existing facility, and an
existing source at the facility is shut down or otherwise
modified so that there is no net increase in emissions, a PSD
permit would not be required. Retrofitting an existing boiler
with FBC technology may also result in an exemption if there
is no net, increase in emissions. Appropriate emission factors
for fugitive atmospheric emissions from FBC solid waste hand-
ling activities are not available. Fugitive emission testing
before and after modification would be appropriate to provide
a data base for such determinations in cases where significant
increases in emissions are possible. Requiring full PSD moni-
toring, unless the source can demonstrate no significant
impact, might also be considered on a case-by-case basis.
Monitoring Requirements
When necessary, new sources subject to the PSD regula-
tions are required to monitor the ambient air quality for up
to 1 year to establish that the impact will not cause or con-
tribute to an exceedance of the ambient standards. Monitoring
for less than 1 year may be acceptable if it demonstrates to
the satisfaction of the regulatory agencies involved that the
standards will not be exceeded. If adequate data are already
available to establish background levels, monitoring may not
be required.
If there are no significant man-made sources nearby, cer-
tain background levels may be assumed constant (see Table
36). Although background particulate levels can be far above
40 yg/m during dust storms, these background levels can be
used to allow construction of new facilities, without emission
offsets or other sometimes impractical controls, in areas
where standards are exceeded because of dust storms.
129
-------
TABLE 36. ASSUMED NATURAL BACKGROUND CONCENTRATIONS
(EPA, 1978b)
Pollutant Concentration (pg/m3)
S02 20
CO 1
N02 : .01
Total Suspended Particulates 30-40
(TSP)
°3
The EPA also has the authority to require ambient air
quality monitoring after a source is built to establish the
effects of the source on ambient pollution levels. Seldom
used, this authority might be appropriate in areas where the
theoretical impact of a source combined with the background is
only marginally better than the standards, and the uncertainty
in modeling or fluctuations in annual average levels from
other sources might mean that there actually are exceedances.
It might also be appropriate if the impact of a source can be
separated from background levels and if there is considerable
uncertainty in the appropriateness of the model or the emis-
sion factors used for determining PSD increment consumption.
EPA has the authority to require a rollback of operations or
application of retrofit air pollution control technology if it
is determined that PSD increments are being exceeded.
Maximum ground level concentrations from isolated point
sources in simple, relatively flat terrain can typically be
predicted to within a factor of two or three, using the equa-
tions in the EPA Workbook of Atmospheric Dispersion Estimates
(Turner, 1974) or the UNAMAP computer program package (Briggs,
G., to Cornett, C.L., personal communication). The exact
location of the maximum may differ from the modeling; accuracy
only to an order of magnitude is not unusual when predictions
are made for a specific location. If there are no conditions
that the modeling did not account for (e.g., lake-effect-in-
duced fumigations, particle settling, reactions and washout of
gases, or plume behavior in complex terrain), higher accuracy
may be possible.
130
-------
The amount of error in the ability to separate the impact
of a source of interest from background concentrations needs
to be considered if increment consumption caused by a source
is to be monitored. Fluctuations in background air pollution
levels can occur that cannot be accounted for by simply com-
paring total pollution concentrations before and after a plant
is built. Some form of tracer for the plant may be needed
when the background is significant. Fluctuations in back-
ground air pollution levels are dependent on variations in
emission rates from background sources and on meteorological
conditions. Monitoring over several years' time is usually
required to measure these long-term fluctuations.
Legal requirements for PSD monitoring are contained in 40
CFR Parts 51, 52, 53 and 58 (Federal Register, August 7,
1980), and additional general guidance is available from the
PSD Guidelines (EPA, 1978a). The Guidelines cover all aspects
of PSD monitoring, including network design, instrument selec-
tion and operation, quality assurance and data reporting.
Generally, monitoring is performed to represent levels where
the source has its maximum impact and where the source plus
the maximum background is maximum. For S02 monitoring, the
3-hour and 24-hour maximum impact is recommended. When fugi-
tive emissions could be significant, monitoring at the site of
the proposed facility is also recommended. The exact number
and location of stations must be determined on a case-by-case
basis by the owner or operator and reviewed by the per-
mit-granting authority. A major limitation of the PSD Guide-
lines is that they do not cover monitoring network design in
any detail. Available references should be consulted (Ball
and Anderson, 1977,- Ludwig et al., 1977; Ludwig and Shelar,
1978).
SELECTION OF POLLUTANTS TO BE MONITORED
Introduction
The procedures described in Section 3 were followed,
where applicable, to select pollutants for air quality moni-
toring of the commercial-scale FBC process. The review of
impact generators and regulations described in this section
formed the basis for determining how the pollutant selection
procedure had to be modified for air quality monitoring.
It is assumed that future commercial-scale FBC processes
will be subject to the same air quality standards as those
which apply to conventional coal-fired plants. Furthermore,
commercial units will probably meet the New Source Performance
Standards for air emissions.
131
-------
Categorization and Prioritization
Currently/ data are unavailable for estimating expected
ambient concentrations of air pollutants generated by FBC
units similar to commercial-size conventional coal-fired
boilers. Therefore, no Category III pollutants were identi-
fied for priority ranking. The preliminary list of pollutants
contains only Category I and Category II pollutants and para-
meters. Once data become available to complete a full air
emission inventory of a commercial-scale facility and disposal
sites, Category III pollutants can be identified and added to
the preliminary list to determine the need for additional
monitoring!.
Category I Pollutants—
The Category I chemical pollutants and parameters, and
the reference documentation are listed below (specific values
for each standard and detailed descriptions of present moni-
toring requirements were previously noted.)
Pollutant or air
quality parameter Source
Opacity of flue gas NSPS for fossil-fuel-fired
particulate emission rate steam generators
reduction in S02 emissions 40 CFR 60.4 to 60.44
C02 or 02
Opacity of all emissions Federal standards for new
particulate emissions coal preparation plants
40 CFR 60.250 to 60.254
Particulate matter Clean Air Act Amendments
S02 (1977); Prevention of
Significant Deterioration
of Air Quality
Suspended particulates Ambient Air Quality Stan-
S02 dards
Pb
CO
Ozone
Non-methane hydrocarbons
NOX
Most of these pollutants, with the exception of the ambient
air quality standards (Table 35), will require, primarily,
emission monitoring of the FBC source. The implications of
these standards to development of a source-oriented ambient
air quality monitoring program for an FBC unit is discussed in
the section, "Monitoring Requirements."
132
-------
Category II Pollutants—
Category II pollutants identified for the FBC process
(carbon monoxide, nitrogen oxides, ozone, and visibility) are
given in Table 34 with expected promulgation dates.
Estimation of Expected Ambient Concentration of Problem Pollu-
tants—
Since the primary problem air pollutants are already
regulated, priority ranking based on expected ambient concen-
trations is not required, nor possible, because of the lack of
emission data. However, dispersion modeling procedures appro-
priate for estimating maximum ground level concentrations from
isolated FBC sources are discussed below.
Preliminary List of Pollutants for Monitoring
The air pollutants and air quality parameters identified
through the selection protocol are listed in Tables 34 and
35. Only Category I and Category II pollutants can be identi-
fied, based on currently available information.
AIR MONITORING NETWORK DESIGN AND DISPERSION MODELS
Background
Most evaluations of the impact of point sources on am-
bient air quality employ process data, emission factors, and
meteorological data to predict ambient concentrations through
dispersion modeling. Sometimes, determining which meteorolo-
gical data and dispersion model to employ can be quite contro-
versial, and a wrong choice can lead to errors.
This can be illustrated with reference to the June 12,
1979, Federal Register in which EPA proposed relaxing the
S02 emission regulations for two power plants located along
the southern shore of Lake Erie (EPA, 1979b), because
source-oriented ambient monitoring allegedly showed that the
urban RAM model (used to derive the emission regulations for
the plants) was inappropriate. The plant owners indicated
that the rural RAM model seemed more appropriate, and their
environmental consultant did not believe lake-effect-induced
fumigation was important in contributing to exceedances of the
ambient air quality standards.
EPA is proposing to require the installation of an exten-
sive ambient air pollution monitoring network to help verify
that the reviewed emission limitations will protect the am-
bient air quality standards or to develop an appropriate data
base for establishing new emission limitations, if the revised
limitation does not. The details of this monitoring program
133
-------
have not been announced by the EPA, but the plant owners plan
to install ten stationary air pollution monitoring stations
for each power plant affected.
Ambient air pollution monitoring can easily be misused,
underestimating the true peak impact of an individual instal-
lation. Detailed EPA guidelines and regulations—stating what
constitutes an acceptable program for checking the accuracy of
air pollution models and how to petition for changes in emis-
sion regulations—are needed. In the absence of such guide-
lines and regulations, the following discussion presents some
general guidance.
Air Monitoring Network Design
Whenever possible, meteorological data spanning at least
a 5-year period should be used to design the monitoring net-
work. The monitoring stations, ideally, should be placed at
locations where the maximum impact of the source is predicted,
according to the different models being studied. Peak 3-hour
and 24-hour impacts are usually the most important for point
sources. Since the locations where these impacts occur change
from year to year, these differences should be considered in
monitoring network design and data interpretation. This will
allow some flexibility in locating monitoring stations.
A series of stations should be placed along radial lines
emanating from the source, at distances where the various
models being tested predict measurably different ambient pol-
lution concentrations. This will help to facilitate selecting
the most accurate model. Available information and guidance
should be considered for all pollutants of interest emitted
directly from the source (Ball and Anderson, 1977) and for
further refinement and expansion of the monitoring station
network (evaluating, upgrading, enforcing, and assessing sup-
plementary control systems) (EPA, 1976a; 1976b). The moni-
toring methods and probe placement should conform to the re-
quirements for state and local air monitoring stations.
The ability to separate the impact of the source or
sources of interest from background levels must be given ade-
quate consideration in monitoring program design. Sometimes,
the use of tracers may be necessary; if so, the toxicity of
these materials should be considered.
At a minimum, meteorological towers and equipment, ad-
hering strictly to requirements and recommendations (Nuclear
Regulatory Commission, 1972) should be employed, with one ex-
ception. Bivanes (or approved equivalent) meeting all NRC
specifications in the horizontal plane, extended to the verti-
cal plane, should be used. The data should be processed to
generate the usual wind direction and the velocity data used
134
-------
in modeling and direct estimates of the cry and oz values.
Horizontal and vertical wind direction should be read by a
data processor at least once every ten seconds and processed
to determine the appropriate standard deviations over the
three-minute to one-hour time periods most models are meant to
represent (the exact time will depend on the model). After
computing Oy and 02 (Pasquill, 1974), the values of tfy and 03
can be compared to the values predicted from the models and
the meteorological data sources used to model impacts to
determine if they are appropriate. They can also be used
directly in modeling.
Caution should be exercised in interpreting the applica-
bility of the resulting monitoring data, especially near large
bodies of water and in complex terrain, where the representa-
tiveness of meteorological data can be quite limited. There
can be large localized shears in wind direction and changes in
turbulence that are relatively unpredictable, without an ade-
quate observational data base to study the detailed climatol-
ogy. Under such circumstances, it can be appropriate to use
certain types of advanced instrumentation such as:
• Acoustic radar to measure mixing heights.
Doppler acoustic radar to measure winds aloft.
Tristatic rather than bistatic systems are recom-
mended (when possible) in complex flow regimes to
assure that the vector wind parameters are for the
same parcel of air.
• Laser doppler anemometers.
Tethered balloons (wiresondes) with wind direction,
velocity, air temperature, and wet bulb temperature,
with telemetry. These can be operated on automatic
winches to provide vertical profiles. They can also
be operated at plume centerline height. These de-
vices are also useful for temporary installations
aimed at studying complex wind fields over an area.
Pilot balloons (pibals) tracked by balloon theodo-
lites (with lights for use at night, when needed).
Rawinsondes tracked by radar or radio direction
finders.
• Radiosondes to transmit vertical wet bulb, dry bulb,
and pressure soundings to ground stations.
Rocketsondes.
135
-------
Aircraft with monitoring equipment for air tempera-
ture, relative humidity, turbulence (to pinpoint
thermal inversion boundary layers for fumigation
studies), pollution, infrared ground and water tem-
perature, and recording navigational equipment.
• Mobile air pollution monitoring vehicles, with pol-
lution .monitoring instruments, calibrators, 502
remote(sensing devices (correlation spectrometers),
wet and dry bulb temperature recording, navigational
equipment, event recorders, data loggers and com-
puters for practical real-time displays of the
monitoring results (EMI MAP system), all of which
work while the vehicle is in motion.
• Semi-mobile air monitoring vans that can be dis-
patched to a site to immediately begin measuring
pollution levels for up to 24 hours when parked. It
should have power onboard to keep instruments warmed
up in transit and either a network of power' drops or
other power supply for use when parked. If genera-
tors are used for power when parked, care must be
taken that the system does not measure its own ex-
haust.
Stationary S02 remote sensors for measuring plume
rise.
The aircraft sampling and ground-based S02 sending can
be used to measure (7y and az values and plume rise (although
problems may arise when multiple intersecting plumes are
involved unless they are mixed completely). They can also be
used to dispatch the mobile ground based pollution moni-
toring vehicles to locations of maximum plume impact.
The exact set of equipment and monitoring program must be
determined on a case-by-case basis. In one case, a network of
ten stationary S02 monitoring stations, selected on the
basis of a study of lake effect, was deployed around a power
plant in Waukegan, Illinois, on the shore of Lake Michigan
(Dooley, 1976; Lyons, 1977). A single mobile van under air-
craft guidance measured 11 S02 samples above 0.15 ppm on the
field, while the 10 stationary sites together measured only
1. This suggests that, if the purpose of ambient monitoring
around a source is to measure compliance or non-compliance
with standards, the possible existence of unobserved peak con-
centrations should be considered in the data interpretation,
along with the amount of error in the data. The extensive
meteorological and mobile monitoring conducted also revealed
significant details about the lakeshore meteorology and atmos-
pheric dispersion that could not be obtained by more conven-
tional stationary air pollution monitoring techniques.
136
-------
In addition to ambient pollution monitoring and meteoro-
logical sensing, a model validation study using a real pollu-
tion source should include continuous monitoring of the types
of process data to be used to project future emissions, and
the continuous monitoring of effluent emission rates, volume
rates of flow, and effluent temperature. The latter two can
be used to calculate plume rise, to allow comparison with
observed plume rise. Measurement of moisture and major gas
constituents in the stack can also be valuable in plume rise
calculations. The specific heat and phase changes in ef-
fluents can have significant effects. Remote sensing equip-
ment can be quite useful for the direct measurement of plume
rise; sampling with aircraft can also be appropriate.
The cost of extensive monitoring programs should be
balanced against the value of the data generated, to determine
if it is worth the expense. It is difficult to ascribe a
dollar value to the protection of human life or to an
aesthetically pleasing environment. Professional judgment
must be exercised in such situations, to assure the most
cost-effective program to obtain the needed information.
For guidance on past monitoring verification models, con-
sult Mills and Stern (1975) and Mills and Record (1975).
MONITORING STRATEGIES FOR FBC INSTALLATIONS
Background
The air quality monitoring programs for commercial-scale
installations for FBC of coal fall into five categories: 1)
baseline monitoring prior to construction, 2) Level 1 and
Level 2 assessment monitoring, 3) compliance audit monitoring,
4) source monitoring required by current regulations of all
coal-fired facilities/boilers and 5) fugitive emission moni-
toring. The first category is well regulated and currently
implemented at federal, state, and local levels and need not
be addressed in this report. The second category is a
research approach that has use in initial operational envi-
ronmental assessment of new commercial FBC installations. The
other three require different strategies implementable through
source-oriented ambient monitoring filling informational and
system gaps created by some inherent shortcomings of emission
monitoring systems.
Compliance Audit Monitoring
Compliance audit monitoring is performed periodically to
assure that source monitoring systems are producing valid data
and that the source is complying with applicable emission
standards.
137
-------
A serious deficiency in the emission monitoring require-
ments is that they do not include comprehensive quality
assurance auditing programs to quantify the precision and
accuracy of the data. For extractive gas monitoring systems,
reactive gases are more sensitive to systematic losses due to
contamination of in-stack filters and other forms of contami-
nation of extractive systems than C02» In addition, extrac-
tive systems can leak (especially when in-stack filters are
clogged). This'can cause systematic underestimation of
lbs/106 Btu emission rates. Even without the systematic
errors, random errors can occur in the data.
At an absolute minimum, all quality assurance and quality
control procedures required and recommended for Prevention of
Significant Deterioration (PSD) monitoring should be per-
formed. In addition, auditing the accuracy of the gas moni-
toring systems by introducing calibration gas at the sample
line inlet should be mandatory because of the possibility of
systematic losses of reactive pollutant gases or leaks between
the environment and the detectors. The temperature of the
lines should be kept above the dew point, using heated sample
lines, if necessary. The appropriate state and local agency
should also perform compliance audits.
Following are some stack monitoring problems, in addition
to leaks and contamination of sampling systems (Floyd, 1979):
Design specifications of instruments in stacks often
are not closely checked by agencies or users of
stack monitoring equipment.
• Agencies do not always check instrument and probe
placement on a stack.
Performance specifications are not always met when
the equipment is operating.
Several years may pass between the initial perfor-
mance evaluation for a stack monitoring system and
the next evaluation.
A quality assurance audit of the system design and in-
stallation should be performed, supplemented by a spot check
of system performance. Some value can be derived from com-
pliance audit stack tests using federal reference methods'and
auditing by challenging the instrument, as close to the probe
tip as possible, with known concentrations of the gases being
monitored. It would also be useful to supplement the
"in-stack" auditing programs with non-intrusive compliance
auditing techniques. The techniques for doing this may not be
as accurate as in-stack methods, but they might provide rea-
sonable grounds to obtain warrants for definitive stack tests
138
-------
in the event of a gross problem with the in-stack measure-
ments. The compliance audit values would have to be compared
to the reported in-stack data or to the hourly data that
sources are required to record. Comparing the reported data
with that which could be reasonably anticipated from the type
of equipment in use is recommended.
Several techniques for performing such auditing are
presented below. Further research is necessary, however, to
validate these techniques.
Methods of Measurement—
Ratio of Components Techniques—There is no mechanical
separation between the molecules of C02 and S02 , which
are both part of the emission from coal-fired power plants.
There can be chemical transformations, adsorption and absorp-
tion of the gases at different rates, but these are often
slow. Typical rates of C02 loss are much slower than for
S02. If the concentration of S02 and CC>2 caused by a
source can be separated from background concentrations close
enough to the stack to minimize transformations, the ratio of
these concentrations in the ambient environment might be used
to measure emission rates. The technique also applies to
particulate and oxides of nitrogen, when compared to carbon
dioxide.
Upwind/Downwind Sampling—
Upwind/downwind sampling .works best when the concentra-
tion caused by the source is far above background levels, to
damp out the importance of fluctuations in background concen-
trations. Samplers could be hoisted into smoke plumes, and
upwind of smoke plumes, to measure these concentrations. Air-
craft, kites, tethered balloons or cranes could be used to
hoist the necessary monitoring equipment into plumes.
Naturally, there can be problems if several sources share the
same stack.
Differential Plume Impact Monitoring—
The differential plume impact technique involves ground
level monitoring of S02 and C02. Monitoring stations are
located around the source, according to the site location
criteria for source-oriented ambient monitoring (Ball and
Anderson, 1977), to minimize the background effect on the
readings when the wind is coming from the direction of the
source being monitored. S02 and C02 levels are monitored
and recorded continuously by fast-response instruments.
Ideally, the response times and characteristics from back-
ground levels to plume centerline concentrations should be the
same for both instruments (or adjusted numerically to be the
139
-------
same during data processing) . By measuring the amount of
change in S02 and C02 concentrations as the source plume
sweeps back and forth across the monitoring site, the relative
concentrations of S02 and C02 caused by the source can be
measured. Areas under peaks might also be used. .
The results of a similar measurement program in
Frankfurt-am-Main, Germany/ are illustrated in Figure 25
(Weber, 1970). : Meteorological and pattern recognition tech-
niques are used" to identify the concentration peaks caused by
the source. While there is considerable variation in the
C02 to S02 ratios, enough statistics might be gathered
after a 30-day period to average out the variations. Since
S02 is generally removed from the air faster than C02*
emission rates are underestimated.
In Germany, the measured ambient ratios were compared to
the ratios measured in the stack to help measure removal rates
for S02 from power plant plumes. The data were sorted
according to wind direction, velocity, stability, class and
relative humidity as part of the program. Very rapid removal
rates (over 50 percent per hour of transit time) were measured
in the highly polluted Frankfurt-am-Main area (Weber, 1970).
Typical loss rates for relatively clean environments with low
relative humidities are less than five percent per hour in
other studies using other techniques (Koch et al., 1977).
Similar studies could be performed around conventional power
plants in the United States with reliable in-stack monitoring
programs.
The differential plume impact technique could, in princi-
ple, be adopted to study NOX emission rates. Continuous
monitoring of total NOX and C02 levels would be needed for
this purpose. Monitoring of NO and C02 levels in the am-
bient environment and in stacks could yield information on NO
to N02 conversion rates (especially if supplemented with
continuous N02 or NOX measurements). Potential advantages
of the technique are its low cost and that artificial tracers
need not be injected into plumes to study transformations.
Limitations include:
• A variable systematic error may result in emissions
being underestimated if transformations are not
accounted for.
(If transformations were modeled and emissions "back
calculated", large random errors could be intro-
duced. )
Applying the technique to an individual boiler at a
plant with several boilers may be difficult. (There
can also be problems if it is used to study emis-
sions from the plant as a whole.)
140
-------
ProbemeRstation Frankfurt — Auflenstelle West 22.8.1967
ID —
u00-
loOQp
11°°-
10
gOO-
noo
3
flUJ2
3:^ irr pptn
5-i* '15ppm
'- > luppm
^^* j!?0nftrn
^
^r- : ; 30ppm
r^ '30ppm
/7 12Bppm
\
X) 400 500
CO^ppml
b1"-
.
U°°-
13°°-
I2°°-
-
|0UO-
gOO_
BOO.
% ASOj
r-3^ 1 0.08 ppm
— <•„ ' 100'ippm
<^-« • U.UA ppm
"*• »••**"" * 1 0 09 nnm
w^ O.Uppm
-^^^ I0075ppm
t
0 0.1 0.2 OJ
S02(ppmJ
'AGO? ~
ASOj
270
OQf\
250
500
Q(Y1
210
380
360
Winddirection —
ENE 2jB
Eon
ENE . 3.0
CMC 21
FKIC 77
CMC 70
ENE 25
E7/
2.1
m/sec-"
Windsoeed
Figure 25. Differential plume Impact data.
(Weber. 1970)
-------
• If stationary ambient air monitoring is used, the
surveillance is not continuous. Any significant
short-duration variations in emission rates from FBC
units make it difficult to compare the results to
average hourly reported or recorded emission rates.
The magnitude of these variations needs to be
studied, to help provide some information on the
limitations of grab samples and of such spot tests.
Monitoring from a vehicle using ratio of components
technique of data processing as the vehicle passes
through the plume could provide many measurements
during an hour, but significantly increase the
expense of the technique.
The peaks caused by a power plant may not be easily
separated from other peaks caused by background
sources, especially in the winter for some installa-
tions.
When performing plume transformation studies, trans-
formation rates may differ in various parts of the
plume, e.g., studies indicate that the removal rate
for S02 in power plant plumes is higher at the
edges of the plume than at the center (Koch et al.,
1977).
Remote Sensing—Various optical devices are commercially
available, or under development, for the remote measurement of
S02 at the mouth of smoke stacks and/or in the ambient envi-
ronment. Most such devices are passive, measuring only the
amount of S02 in "line of sight." This is known as the S02
"burden," and is measured in micrograms per square meter. As
of Summer 1979, S02 laser radar units were experimental, ex-
pensive, not available as standard commercial units, and re-
quire more field testing. In addition, some dual absorbance
systems may not have the full-scale range necessary to measure
concentrations at the mouth of smoke stacks.
Equipment for the remote measurement of C02 is rarely
available, even experimentally. Instruments such as the
Barringer Non-Dispersive Gas Filter Spectrometer can be set up
for C02, but the utility of the device has not been demon-
strated in.power plant studies. Background C02 levels
exceed 300 ppm, and it could be difficult to detect a plume
against the background with a device that only measures the
amount of C02 in line of sight. The dependence of the
response of a two-channel infrared to the temperature of the
gas could also cause some difficulties. As an alternative,
Fourrier Transform Interferometer (FTIR) with a telescope
could be aimed above the mouth of smoke stacks and next to the
stack for background measurements. This device could dis-
criminate between stack C02 and background C02 by the
142
-------
changes in absorption band characteristics caused by the
temperature of the effluent. The changes in spectral charac-
teristics can also be used to measure effluent temperature.
Remote sensor mobile PTIR devices are expensive ($150,000 for
a working mobile system) and require highly trained operating
personnel. Nevertheless, testing of the accuracy and limita-
tions of FTIR for this purpose could be worthwhile. The EPA
has used its mobile PTIR to investigate sources for enforce-
ment purposes, and demonstrating its utility and limitations
for quality assurance auditing could be of some value. The
device might also be used to measure S02 concentrations,
although an error of 5°C in temperature can cause a 20-per-
cent error in concentration.
In the absence of established methods for measuring C02
concentrations and burdens in power plant plumes remotely,
some information can be derived from measurement techniques
for S02» Concentrations measured at the mouth of a stack
can be compared to in-stack measurements. If the in-stack
monitor measures pig/m^ at stack conditions, the measure-
ments at the mouth of the stack could be compared directly
(provided it is not diluted before exiting the stack). If it
measures pg/m^ at standard temperature and pressure, or
ppm, correction to stack effluent temperature would be needed
before data can be compared. If the measurements are made per
dry cubic foot, correction for moisture would also be needed
(or the amount of error from not considering it would have to
be considered in data interpretation).
Instruments that measure the burden of SC>2 and which
scan across the mouth of smoke stacks can be used to measure
effluent concentrations. The measurements from such instru-
ments can be converted to emission rates by multiplying the
concentrations by the volume rate of flow out the stack.
Emission rates can also be calculated from the velocity of the
gas exiting the stack and the concentration of materials in
the plume. Laser doppler anemometers and the Viviplume remote
sensor are two instruments capable of measuring effluent
velocity to calculate volume rates of flow, although the
accuracy and limitations of the devices need further study.
S02 emission rates can also be measured by remote
sensors that measure the burden of S02 overhead. Measure-
ments of the overhead S02 burden can be combined with data
on the wind direction and velocity in an area surrounding the
plant, and the matter flux of S02 in and out of the area can
be calculated. The difference is the S02 emission rate
minus the decay rate in the area. These emission rates can
then be compared to the maximum allowable emission rates and
to the emission rates assumed in emission inventories.
143
-------
The types of data required to make these comparisons are:
Design assumptions for the plant concerning the
boiler efficiency
• Test reports during start-up of the boiler con-
cerning the efficiency under various percentages of
capacity loads
i
• Records on the source of the coal/ the Btu/lb
rating/ the percent sulfur (or at least the maximum
percent sulfur, and the amount burned (if available)
Pounds of steam produced, its temperature, flow rate
and pressure.
There is no reason to believe that a remote sensor that
works for conventional boilers without significant inter-
ferences will not work for FBC units. One test of this
assumption is to use a remote sensing spectrometer or inter-
ferometer to measure the spectral "signature" of the plume.
The spectrum could be checked for interfering gases such as
absorbance bands due to other gases. If they are not present,
the applicability of the remote sensor would be established
for similar sources. It is necessary to obtain average spec-
trum because a single slow scan would have problems with
turbulent eddies in the plume and fluctuations in concentra-
tion and plume geometry during scanning FTIR. Many remote
sensors operate in the visible/ultraviolet region where
instruments similar to FTIR are not available within EPA; how-
ever/ some university astronomy departments or other organi-
zations such as NASA may be able to provide appropriate
instrumentation. If 3-minute scans with a relatively high
resolution visible/ultraviolet spectrometer were integrated
and averaged over a long enough period, any significant fluc-
tuations could be dampened out. I '•
An alternative test would be simply to repeat experiments
used to establish the applicability of the remote sensor for
conventional boilers to FBC units. Generally, in-stack mea-
surements for parameters of interest, using federal reference
methods or the equivalent, when possible, are compared to the
results of remote measurements. The best experiments involve
independent teams taking the stack measurements and the remote
sensing measurements and interpreting the results/ and a
referee compiling and comparing the results and coordinating
the timing of the tests. Duplicate simultaneous measurements
of in-stack and remote measurements could be used to establish
the amount of precision in the measurement methods. The cost
of such a program would probably be less than that of devel-
oping visible/special ultraviolet spectral scanning data pro-
cessing techniques; however, the latter techniques could be
144
-------
useful for many other types of sources than FBCs and are
faster.
Dispersive Correlation Spectroscopy—The operating prin-
ciple of dispersive correlation spectrometers is summarized as
follows (Moffat and Millan, 1969):
The principle of correlation spectrometry is based on
measurement of the degree of similarity between the mole-
cular absorption spectrum of a chosen gas and the actual
absorption spectrums of the gases under observation. A
correlation spectrometer system typically consists of a
light source, field-defining fore-optics, an entrance
slit, collimating elements, dispersive element, exit
mask, and photodetector. Radiant power from the light
source is focused on the entrance slit, dispersed by a
prism or grating, and refocused in the plane of the exit
mask. The incoming spectrum is then made to cross-cor-
relate against the mask by moving one relative to the
other in some cyclical fashion and collecting the output
on a photodetector. The exit mask is an enhanced and
modified replica of the absorption spectrum of the gas
being sought and is usually made by depositing a film
of aluminum on quartz or glass blank and then stripping
or photoetching a series of slits in the aluminum which
correspond in some predetermined fashion with the mole-
cular absorption bands of the desired spectrum.
The sensitivities of some remote sensing correlation
spectrometers are as follows (set with an 8-second integration
time) (Lawrence-Berkeley Laboratory, 1976):
S02 N02
Instrument ppm-meter ppm-meter
COSPEC II 6 3
COSPEC III 6 3
COSPEC IV 2 1
The COSPEC dispersive correlation spectrometer, operated
as a passive S02 and/or N02 remote sensing system, may be
useful in experimental studies, for mapping plumes, dis-
patching mobile air sampling equipment, measuring ay and az
values in model verification studies, and measuring emission
rates. The sensor, mounted in a vehicle equipped with recording
navigational equipment, usually measures the burden overhead and
is driven around the source to be monitored. The resulting
readings, in grams per square meter of S02 and/or N02 in line
of sight overhead as a function of vehicle position, are con-
verted to pollutant mass emission rates, using meteorological
data (preferably at plume centerline height and close to the
position of the measurements, especially the position of the
145
-------
plume). Pibals and wiresondes are the most common devices for
measuring wind direction and velocity, although stationary
meteorological towers or laser anemometers may also be used.
There are interferences to the S02 burden readings from
correlation spectrometers. Among the major characteristics of
the instrument that need to be considered for accurate results
are: .
i
The device may measure different wave lengths at once.
As a result, the response characteristics vary with the
spectral distribution of the background light. Adjust-
ments can be made to the instrument to minimize the
baseline drift caused by changes in sky spectral
radiance (Millan and Hoff, 1976). The unit can also
recalibrate very quickly to compensate for changes in
sky spectral radiance during the day. These instru-
ments do not work at night.
The COSPEC II and III respond to changes in the
polarity of the background light relative to the
detector. This can cause problems when circling a
source, especially when the sun is within 30° of
the horizon. Certain internal adjustments can be
made to minimize this problem: the unit can be
rotated during traverses to maintain the same
polarity angle to the sun, or optical devices can be
inserted in the field of view to randomize the
polarity at a sacrifice in the amount of signal into
the instrument (Millan, M.M., Barringer Research, to
Cornett, C.L., personal communication).
• The angle of view of the telescopes on the units,
while adequate for most sky viewing, is finite. If
an object or feature o'jf the sky is smaller than the
angle of view, it actsrllike its own slit and changes
the response of the instrument. This can cause
serious problems when viewing plumes exiting smoke
stacks from too far back. Also, under these cir-
cumstances, varying distances through the plume and
the background will be read simultaneously and the
automatic gain control will not compensate properly
for correlated non-homogeneous light intensities and
S02 burdens in the field of view. When viewing
across smoke stacks, the size of the field of view
must be considered when determining how far back to
place the instrument. When viewing the sky, rec-
ording of the automatic gain control can be used to
edit out extraneous peaks caused by cloud fringes on
partly cloudy days. Recalibration is required with
changing sky conditions.
146
-------
Cloud fringe spikes are also caused because the in-
strument combines two channels of measurement with
different time constants.
There is very little ultraviolet background energy
near the horizon—the atmosphere is heavy in the
ultraviolet spectrum. Measurements should be made
at least 15° above the horizontal.
Under high particulate loadings, light may be scat-
tered into the field of view of the instrument that
has been affected by S02 before this scattering.
This effect is usually less than 10 percent of the
apparent pg/m2 readout.
Different instruments have different full-scale con-
centration ranges, depending on the mask chosen and
degree of tuning. The response also becomes
non-linear for high S02 burdens. It is .necessary
to choose the proper mask, instrument and tuning
combination for the application. The masks are not
easily changed.
The instrument has a finite and adjustable response
time that must be considered when setting up for a
given application. This sometimes is important in
data interpretation.
In an EPA-sponsored study at a coal-fired power plant
(Sperling, 1975), a 95 percent confidence of the results being
within + 20 percent of simultaneous in-stack monitoring of
SC>2 emission rates was obtained using standard EPA methods.
This study involved circling the power plant 30 to 85 times
over 2 to 6 hours, measuring the S02 burden overhead as a
function of vehicle position, and tracking many balloons to
gather meteorological data. The COSPEC II yielded this con-
fidence in two hours; the COSPEC III and IV in about 6 hours.
More recent work yielded the following approximate differences
in emission rates relative to reference methods for a power
plant emitting more than 60 metric tons per day of S02 ,
using COSPEC II and III instruments (Sperling et al., 1979):
± 50 percent in single traverses
i 35 percent for 20-minute averages or 2 to 5 tra-
verses
+ 20 percent for 60-minute averages or 4 to 13 tra-
verses
+ 10 percent for 7 to 12-hour averages or 25 to 75
traverses.
147
-------
Instrumental inaccuracies, fluctuations in real source
emission rates, and especially the non-representativeness of
meteorological data at one location or along a trajectory (for
balloon wind measurements) compared to real conditions at the
locations where the remote sensor is viewing, contribute to
measurement errors. Nevertheless, the applicability and ac-
curacy of correlation spectroscopy has been reasonably estab-
lished.
i
A COSPEC II remote sensor has been used in at least one
study scanning across the mouth of a smoke stack to measure
S02 concentrations for comparison to in-sta-ck measurements
(Table 37). At a range of 20 meters, the results look en-
couraging; at more distant ranges, large discrepancies are
apparent. It is believed that the cause of the problem is the
size of the field of view compared to the size of the stack.
More work of this sort needs to be performed. The sensitivity
of the correlation spectrometer should be more than adequate
for even the smallest commercial-size FBC units, at the mouth
of the stack. It could be necessary to desensitize the in-
strument so that it is not too sensitive in stack mouth appli-
cations.
Non-Dispersive Gas Filter Correlation Spectrometer
(GFC) —
This device uses cells containing various amounts of the
pollutant gas to be measured to modulate light signals between
a telescope and light detectors. An ultraviolet S02 remote
sensor, commercially available in France, costs less than
$10,000, versus around $25,000 for a COSPEC IV. Its accuracy
and performance have not been established.
A Science Applications Incorporated instrument works in
the infrared re'gion, with an 8-milliradian field of view. The
sensitivity and:1, noise of the device varies with the tem-
perature of the gas being monitored. Table 38 presents some
results of the use of an early prototype of the instrument to
measure gas concentrations at the mouth of smoke stacks, com-
pared to concentrations measured within the stack.
Ultraviolet Television Sensor—
An ultraviolet television-based remote sensor for S02
concentrations, emission rates, and effluent velocity has been
developed by the National Aeronautics and Space Administration
and is commercially available for less than $33,000. The in-
strumentation system measures the opacity of the S02 exiting
the stack, converts it to effluent concentration, tracks S02
eddies to measure the average effluent velocity, and calcu-
lates the resulting mass emission rate. The sensitivity is
about 50 ppm meters. It also supplies a television image of
148
-------
TABLE 37. COMPARISON OP IN-STACK AND COSPEC II SO.
CONCENTRATION MEASUREMENTS
(Lugwig and Griggs, 1976)
Range
(meters)
In-Stack
(ppm)
Remote
(ppm)
(IS) - (REM)
x 100
20
20
200
300
300
300
500
585
700
460
450
700
465
520
240
224
205
275
7
10
65
52
54
60
TABLE 38. SUMMARY OF AVERAGE VALUES OF REMOTE GFC AND EXTRACTIVE DATA
(Ludwig and Griggs, 1976)
[S02], ppm
Extractive3
Range Date
(m)
160
390
390
390
390
390
160
8-12-74
8-12-74
8-13-74
8-13-74
8-14-74
8-14-74
8-15-74
Time
10:00-11:45
13:45-16:50
09:00-12:00
18:45-21:00
09:15-12:45
14:30-16:30
09:30-11:45
Remote
T, C Cone.
150
150
150
150
150
150
110
540+200
630+170
530+120
550+120
490+160
440+110
750+ 70
DuPont
Analyzer
• »•»
412+18
422+28
472+32
519+21
EPA
Method 6
H -»
525+25
398+23
401+17
475+7
499+11
%
Deviation
-17
-25
-25
-16
+ 7
-32
All extractive data taken from plume with a temperature of 150 + 5°C.
149
-------
the S02 plume. Plume visualization can be used to assess
plume geometry and check for downwash. Such phenomenon could
cause inappropriate path lengths to be used for calculating
concentrations from opacities, if unobserved. To operate the
unit, the inside and outside stack diameter is fed into the
keyboard, the unit is calibrated, the camera is pointed at the
black target, the distance to the target is fed in, and the
camera is then pointed at the stack. Cursors are set around
the outside stack diameter using the TV image (to scale the
results). The unit can then be commanded to measure concen-
trations, velocities and emission rates once or repeatedly
every minute to every 5 m-inutes and to print the results.
It would be desirable to program the microprocessor to
output hourly average values also, but the lack of this fea-
ture is only a minor inconvenience. Table 39 compares the
results of monitoring using a prototype unit and EPA reference
methods at 400 meters. The results were extremely close to
the EPA reference method results.
TABLE 39.
COMPARISON OF S02 CONCENTRATION AND EFFLUENT VELO-
CITY MEASUREMENTS BY VIDEO AND IN-STACK
TECHNIQUES
(Exton, 1977)
Date
Concentration, ppm
Velocity, m/sec
Video UV (in-stack) Method 6 Video LDV Method 2
i. 8/3/76
2 8/4/76
790
860
650
660
740
840
22.8
21.6 19.1
22.7
22.4
In extremely polluted environments, background plumes,
non-uniform background particulate light scattering, and high
opacity plumes may present some problems.
Emission Rates for Other Materials—
; While certain types of remote sensors can give infor-
mation on emission rates of certain pollutant gases of
interest, they will not work for all gases, nor for trace
materials within particulates. The National Center for Atmos-
pheric Research in Boulder, Colorado, has been applying a
technique for studying emission rates from volcanoes. Sulfur
150
-------
dioxide emission rates were measured with a COSPEC correlation
spectrometer mounted in an airplane equipped with inertial
navigation equipment and meteorological instrumentation. The
ratios of the concentrations of other chemicals and components
to S02 in the volcano plume were determined by sampling the
plume from the airplane. Emission rates for materials of
interest were then calculated by multiplying the S02 emis-
sion rate by the ratios.
This basic technique should be validated by applying it
to measure emissions from air pollution sources, while conven-
tional stack monitoring is taking place, and comparing
results. Once its accuracy and limitations are documented,
the technique may be considered for FBC units.
Similar techniques to those described for SC>2 moni-
toring are also applicable to NOX monitoring. The main
limitation is that most of the NOX emitted from sources is
in the form of NO, and most remote sensors measure N02
rather than NO.
Level 1 and Level 2 Assessment Monitoring
The EPA Industrial Environmental Research Labs has devel-
oped a detailed approach to effluent characterization (Lentzen
et al., 1978). The basic approach is to characterize ef-
fluents in three phases of increasing specificity and
accuracy:
Level 1 - Identify pollutants in specific waste
streams (and fugitive emissions) with an accuracy
factor of j^3; no special procedures are employed to
obtain statistically representative samples or
analyses.
Level 2 - Focus on waste streams indicated as poten-
tially troublesome; characterized by statistically
representative samples, accurate stream flow mea-
surements, and quantitative identification of speci-
fic organic species and/or classes and inorganic
elements and/or species
Level 3 - Identify critical components in selected
waste streams using Level 2 or better methods as a
function of time and process variation for control
device development.
Figure 26 illustrates the Level 1 analysis scheme; Table
40 shows the basic Level 1 analyses. The source assessment
sampling system (SASS) used to test ducted emissions is
presented in Figure 27. Grab samplers are also used. Figures
151
-------
TABLE 40. BASIC LEVEL 1 ANALYSES
(Lentzen et al., 1978)
Category
Parameter
Physical
Chemical
Biological
Cyclone particle size
Optical microscopy
Spark source mass spectrometry (SSMS)
Atomic absorption spectroscopy (AAS)
Wet chemical (selected anions)
Gas chromatography (GC)
Elution chromatography (LC)
Ion chromatography (1C)
Infrared spectrometry (IR)
Low resolution mass spectrometry (LRMS)
Total chromatographable organic (TCO)
Rodent acute toxicity
Microbial mutagenesis
Cytotoxicity
Fish acute toxicity
Algal bioassay
Soil microcosm
Plant stress ethylene
28 and 29 outline the types of analysis performed on the mate-
rial collected in the ducted effluents, and Figure 30 shows a
schematic diagram of the Fugitive Air Sampling Train (FAST).
FAST provides a 500 mg sample in 8 hours for unenclosed emis-
sions that cannot be measured by the SASS train and is used in
upwind/downwind sampling (Kolnsberg, 1976).
FAST separates particles into four size ranges:
Those greater than 100 microns, separated on the .
louvers at the inlet. These are not ordinarily mea-
sured because they are too big to usually be blown
far from the source.
Those greater than 15 microns and less than 100
microns. These are separated on an impactor in the
diaphragm. The impactor is to be replaced by a set-
tling chamber elutriator because the performance of
152
-------
— fr-
SOLID/
LIQUID
' *
ORGAN ICS
SUSPENDED
SOLIDS
BIOASSAY
ELEMENTS AND
SELECTED AN IONS
PHYSICAL SEPARATION
INTO FRACTIONS
». INORGANICS SELECTED ANIONS
QA PHYSICAL SEPARATION
INTO FRACTIONS LC/IR/MS
ELEMENTS AND
SELECTED ANIONS
ELEMENTS AND
SELECTED ANIONS
PHYSICAL SEPARATION
INTO FRACTIONS
LC/IR/MS
PHYSICAL SEPARATION
INTO FRACTIONS
LC/IR/MS
ALIQUOT FOR GAS
CHROMATOGRAPHIC
ANALYSIS
Figure 26. Basic level 1 sampling and analytical scheme for solids,
slurries, and liquids.
153
-------
SAMPLE
SOR8ENT CARTRIDGE —
SECOND AND THIRD
1MPINCER3 COMBINED
IM K 0 — |_ .
— UJ O 1- VI
O X ~ U UJ
i04 o V I 2
x *~* o a. uj 5 o
t~ U W u t- «l ' U
" «=5!qw A5
3 £5 - S ^ > 2 i ^
5 slilsSSsl*
«BnT "\ -i ., , , . • •
5 ORAHS COMBINE
AO.UEOUS PORTION
^\ ORGANIC EXTRACT ^ ^ ^ \
s
w> >
< O3
| 2
— •
TOTALS
525 * 5
* If required, sample should b« set islde for biological analysis at this point.
' This step Is required to define the total mast of particulars catch. If the sample
exceeds lot of the total cyclone and filter sample weight, proceed to analysis. If the
sample Is less than 10* of the catch, hold In reserve.
Figure 27. SASS analysis requirements.
(Lentzan et a I., 1978)
154
-------
r-1
Ul
Ul
NO EPA METHOD 7
COMBINE
CONDENSATE
Onsite gas
SASS SAHPLE
* ,
CULATE 1
|
I I
=l*i 1 to 3w 3
IGHT WEIGHT
1 I
to I0w >10w
WEIGHT WEIGHT
\
GAS
SAMPLES
GASES
l
-
XAD-2
MODULE
INORGANIC
(GRAB)
ORGANIC
bp
-------
01
ISOLATION
BALL VALVE
FILTER
STACK T.C
CONDENSATE
COLLECTOR
DRY GAS METER/ORIFICE METER
CENTRALIZED TEHPERATURE
AND PRESSURE READOUT
CONTROL MODULE
IMP/COOLER
TRACE ELEMENT
COLLECTOR
TWO 10-ft3/min VACUUM PUMPS
IMPINGER
T.C.
Figure 29. Source assessment sampling train schematic.
(Lentzen et al.. 1978)
-------
CYCLONE
Ul
CYCLONE
VACUUM GAUGE
HAIN VACUUM
BLOWER
EXHAUST
VACUUH GAUGE
e>
EXHAUST
TRAP OUTLET T.C.
FILTER
VACUUH
PUMP
Figure 3O. Fugitive air sampling train components.
(Lentzen el al.. 1978)
-------
the impactor was found to be unsatisfactory
(Kolnsberg and Severance, 1979) .
Those between 15 microns and 3 microns (50 percent
collection efficiency at 3 microns) . These are col-
lected on the cyclone. The cyclone should be
adjusted to cut at 2- 1/2 microns to simulate the
cut size for dichotomous samplers, but the dif-
ference is not very important for Level 1 analysis.
Smaller particles collected on a glass fiber filter.
Organic species passing through particulate collection
species that are Cg and higher are collected on the XAD-2
sorbent module. The analysis resembles the SASS train, except
that there are no impingers to collect volatile metal species
(Hg, As, Sb) and enough samples to perform all biological
tests may not be available. It is quite unlikely that enough
hydrocarbons would be collected on the XAD-2 cartridge. AMES
mutagenicity testing can be performed on 150 to 300 mg of
sample, and RAM can be performed on 150 mg of sample. Deter-
mining an LDso would take more sample.
According to the Level 1 manual, low molecular weight
gases are to be collected in an integrated gas sampling train
similar to that used to collect these species in flue gas (see
Figure 31). The condenser would be needed only under ex-
tremely wet conditions, such as in a cooling tower plume.
In the 'absence of defined standard Level 1 procedures for
these gases, the following are recommended:
CO should be measured from the integrated gas samp-
ling bag using the federal reference method
(non-dispersive infrared analysis) or approved
equivalent.
• The federal reference method (chemilluminescent) , or
approved equivalent, should be used to measure NOX
by sampling the air directly, not from the bag.
• Sulfur gas grab samples can be highly unstable under
real ambient conditions. S02 should be measured
by the federal reference method, or approved equiva-
lent. Other sulfur gases should be measured using a
sensitive gas chromatograph (at least 0.05 ppm
sensitivity for hydrogen sulfide) with a flame
photometric detector and ambient air gas sampling
valve. The AID portable gas chromatograph is
marginally sensitive enough for this purpose, but
more stable and sensitive devices would be de-
sirable.
158
-------
FILTER
(GLASS
WOOL)
AIR-COOLED
CONDENSER
QUICK
DISCONNECT
RIGID CONTAINER
VO
Figure 31. Integrated gas-sampling train..
(Lentzen et aL. 1078)
-------
Mercury should be measured by NIOSH P and C AM
Method Number 175 (National Institute for Occupa-
tional Safety and Health, 1974).
Low molecular weight hydrocarbons should be measured
by gas chromatography equipped with a Flame loniza-
tion Detector, but the details should be researched
to achieve as much sensitivity as possible for
screening for substances with MEGs.
Guidance is needed on what constitutes screening proce-
dures equivalent to Level I/Level 2 sampling and analysis.
Alternatives to the precise Industrial Environmental Research
Laboratory Level 1 procedures should be acceptable, if the
alternatives suggested are valid and provide at least as much
useful information as the specified Level 1 procedures. Since
Level 1 is designed to screen for the unexpected, alternatives
should have at least as much sensitivity and accuracy for
pollutant substances possibly relevant to the source that are
screened for by Level 1 analysis.
Often, enough is known about the environment or process
stream being sampled to justify certain types of Level 2
analysis without the need for Level 1 analysis of the para-
meter. Variations on the Level 1 technique may provide the
necessary information at lower cost than separate Level 1 and
Level 2 procedures. This is especialy true of organic analy-
sis, where the procedures specified are expensive, and in-
corporation of fluorescent screening techniques and of Gas
Chromatography/Mass Spectrometry analysis instead of certain
portions of the Level 1 procedures may have some value under
certain circumstances. Any variations should be approved by
the EPA prior to implementation.
All Level 1 monitoring data should be reported into the
Environmental Atmospheric Data Systems (EADS) and be systema-
tically processed to provide upper and lower limits to the
concentrations of substances with multimedia goals. When the
analysis yields no information whatsoever, this should also be
listed, along with the best estimate of an upper and lower
limit to the concentration of the substance based on theo-
retical considerations (if any). The substance and the con-
centration range should be required inputs.
The data gathered should also be used to project peak am-
bient concentrations. The techniques to be used for modeling
should be worked out on a case-by-case basis. Ignoring the
uncertainty of the modeling, the need for further monitoring
can then be priority ranked according to the uncertainty in
the concentration (as a result of the limitations of Level 1
analysis)/Ambient Level Goal ratios. This priority ranking is
based on ambient concentrations, but ignores the geographical
extent of the problem.
160
-------
Pollutants can then be selected for further monitoring
based on the need to reduce the uncertainty in toxic unit dis-
charge rates and on the need to reduce uncertainty in ambient
concentration/Ambient Level Goal ratios. The need for model
validation studies to reduce the uncertainty in modeling
should also be considered.
Fugitive Emission Monitoring
For FBC units equipped with an effective system of hoods
and fugitive air pollution control equipment, stack testing
upstream and downstream of control equipment would be appro-
priate, using the standard stack sampling method, quasi-stack
sampling method or the roof monitor method (Kolnsberg et al.,
1976; Kenson and Bartlett, 1976). For installations not so
equipped, and at major solid waste disposal sites, up-
wind/downwind testing will be necessary. The program should
include testing both before and during operation of the facil-
ity.
If fugitive emissions are significant, PSD monitoring on
the site may be required, including upwind/downwind moni-
toring.
One monitoring station should be located slightly down-
wind of each type of facility. For simple building geo-
metries, and relatively flat terrain, a location two to three
building heights or widths (whichever is less) downwind, along
the prevailing wind direction, should be considered, and an
upwind station should also be installed. The sites chosen
should be undisturbed by construction and meet all site loca-
tion requirements. The downwind site should be reasonably
close to the ash discharge chute for truck loading, or the
active portion of the landfill where FBC waste will be dis-
posed, and within line of sight.
While only the site on the prevailing downwind side of
the source usually needs to be operated on the full PSD moni-
toring schedule, at least four sets of upwind/downwind samp-
ling sets employing duplicate samplers and analysis should be
run, using data from a single meteorological tower with auto-
matic wind speed and direction controlled switching equipment,
or equivalent manual control. In addition to the use of a
FAST train, the upwind/downwind monitoring should have:
High volume samplers, with teflon filter pads,
analyzed for sulfates, sulfites and organic adducts,
along with alkalinity.
Virtual impactor style dichotomous samplers meeting
the specifications of the Office of Research and
Development (Stevens and Dzubay, 1978a).
161
-------
Modified high volume samplers equipped with 4-inch
(101.6-millimeter) teflon nucleopore filters with
0.4-micron pores to collect samples for automated
X-ray fluorescence and electron microscopic finger-
printing of individual particles.
Battelle ultravol 600 CFM high volume sampling for
collection of enough sample for full Level 1 bio-
logical1 testing.
Portable back pack samplers with silver membrane
filters for X-ray defraction analysis.
The PSD site for monitoring at the location of maximum
stack impact should also be included in the upwind/downwind
monitoring program, providing it is in a downwind trajectory
from the point source to be monitored while the fugitive emis-
sion site is also downwind. The data would be processed to
determine any statistically significant differences in up-
wind/downwind concentrations, and the need for further moni-
toring would be determined.
The upwind/downwind monitoring program could be repeated
for Level 2 analysis after the FBC facility and FBC waste
landfill is in full operation and again a few years after the
material has been permanently landfilled and the portion of
the landfill being tested is no longer active. This could be
used to check for off gases from bacterial action and other
long-term effects.
New data gathered from pilot, demonstration or similar
commercial units, and from experience with monitoring should
be reviewed and used to modify this monitoring program design
as necessary. An intense program such as this will not be
necessary at all FBC facilities. After enough data are
gathered at FBC facilities to predict impacts with reasonable
accuracy, the monitoring program can be focused on those pol-
lutants which could cause problems.
Heat of Hydration and Alkalinity—
The most important unregulated pollutant associated with
FBC solid waste is calcium oxide. This is based on a com-
parison of projected ambient concentrations (assuming com-
pliance of total suspended particulate levels with Class II '
PSD limits) with 1/420 of a Threshold Limit Value (TLV).
The only very precise technique for detecting CaO in
atmospheric particulate matter is X-ray diffraction. This
technique measures crystalline CaO in various states of hydra-
tion, but does not measure amorphous CaO. A review of the
NIOSH computerized data base on CaO has revealed that it is
162
-------
significant mostly because of the heat given off when it comes
into contact with water and its alkalinity. These effects may
also synergize with other environmental insults to produce ad-
verse effects. If stinging afterburn sensations are observed
at locations where the general public has access downwind of
an FBC facility, routine ambient air surveillance and plume
transformation studies should be instituted to measure the
rate of change of heat of hydration and alkalinity per micro-
gram of sample as a function of travel time, relative
humidity, and 502 concentrations.
Automated scanning electron microscopic and energy-dis-
persive X-ray fluorescence analysis may be applicable to
separate the amount of particulates from the bottom ash from
total ambient levels in the study. Equilibrating all samples
for electron microscopic X-ray fluoresecent measurement of
relative concentration of FBC waste with a humid atmosphere
could be used to normalize the results. For a given mass of
dry waste, they would show the same cross-sectional area and
this would help in eliminating this problem for mass concen-
tration determinations. On the other hand, separate samples,
stored very dry and introduced directly into the electron
microscope immediately after analysis, could provide some
information on the rate of hydration of the particles by
examining their "puffiness" through morphological identifica-
tion of how expanded the particles are.
The first few heat of hydration tests should be performed
on the fugitive dust emitted when the bottom ash is emptied
directly from the boiler to the hopper, collected while it is
still hot, if possible. The sample should be of minimum size
to eliminate spontaneous hydration during sampling, and it
should be segregated by particle size. Crushable filters are
preferred for heat of hydration measurements. The heat of
hydration of the particles collected should be measured with a
sensitive microcalorimeter with dual cells (one for a blank
filter and water capsule, and one with an exposed filter and
water capsule), with plungers to facilitate breaking the cap-
sules. Sensitivities for heat of hydration of a few micro-
calories can be obtained using available state-of-the-art
equipment. Marche Instruments, Inc., in Lyon, France, and
possibly Tronac (Orem, Utah) manufacture instrumentation and
measurement cells. Blank filters should be run to test the
amount of uncertainty in the zero heat of hydration measure-
ments. If the heat of hydration of the fresh bottom ash fugi-
tive particulates is appreciable, methods should be developed
for ambient air analysis for heat of hydration. Plume trans-
formation and particle size and deposition-related studies
should be performed to determine how far downwind this effect
could have significant impacts, as a function of environmental
conditions and emission rates. The main potential problem is
hydration of particles collected on the filter or in the
163
-------
sampler while air is passed over them during the process of
collection. An ambient sampler that dilutes the sample with
ultra-dry air before collection might help; another approach
is to create a wet filter medium impregnated with pigments
that permanently change color when a certain temperature is
exceeded. Spots on the filter would be a crude index for the
number of particles having an appreciable heat of hydration.
i
The heat of hydration of the particulates from the spent
bed material can combine with the alkalinity to produce sensa-
tions similar to being "bit" by the dust, with an alkaline
afterburn. A program of skin sensitivity testing could be
established, utilizing varying doses of FBC spent bed material
with varying documented heats of hydration, surface tempera-
tures during hydration, alkalinity and size. Similar-ap-
pearing inert material would be needed as a control, and the
program should be performed using double blind experimental
procedures. The chromium content of the dose would also be
measured, and the locations where doses were administered
would be examined for surface skin allergic reaction or any
damage a few days after testing. Effects on vegetation,
scratched and unscratched paint panels and other sensitive
receivers should also be studied. The end point of this
research would be to determine ambient goals for these para-
meters based on health effects, ecological effects (including
property damage), and nuisance value. The research could also
be used in the design of monitoring programs. Pigments in im-
pregnated filters set to change color at temperatures and heat
inputs similar to those that cause adverse impacts could be
designed, based on the results of the program. The necessary
sensitivity and other details for calorimetric experiments
could also be determined.
Portable Wind Tunnel Monitoring—
Midwest Research Institute has developed a new and
definitive method of studying emission rates from aggregate
storage piles and landfills. A portable wind tunnel, with an
open bottom and skirts (to seal the open area around the
edges) is placed over the surface to be tested. High volume
isokinetic samples are taken with a cascade impactor equipped
stack test train as the tunnel is operated at different
velocities-(Coherd and Cuscino, 1979).
The technique brings out certain features of the behavior
of windblown dust emissions in a more direct and cost-effec-
tive manner than upwind/downwind sampling. It easily
generates information on:
• The threshold velocity for the onset of wind erosion
on crusted and uncrusted surfaces for different
particle sizes
164
-------
• The dependence of the emission rate on wind velocity
The rate of decrease of emission rates from surfaces
when exposed to a constant wind velocity
Emission rates, however, are not completely uniform over
material pile surfaces. Rates vary with the wind velocity and
the properties of the surface. Measurements of the wind field
close to the surface of a pile (or a simulated pile in a wind
tunnel or theoretical model)r combined with information on the
amount of crusted, aged, and disturbed surface area, and the
results of field testing with the wind tunnel, would have to
be combined to obtain an accurate estimate of the total mass
emission rate for the pile, using the portable wind tunnel
technique. Research is underway to develop cost-effective
techniques to do this and to compare the results to up-
wind/downwind monitoring. The technique is also not appli-
cable to measuring emission rates from portions of a pile
being actively worked. This technique should be considered
for detailed studies of windblown dust from raw material and
solid waste piles associated with FBC processes. The results
of such testing could also be used to determine at what wind
velocities it might be appropriate to perform upwind/downwind
testing of emission rates and to generate those emission
rates, once suitable techniques for simulating wind fields and
making the necessary calculations are validated in field
tests. When the technique is used, it is suggested that the
runs be made on the following types of surfaces:
fresh dry surface
crusted dry surface, both disturbed and undisturbed
wet surface, both disturbed and undisturbed
• aged weathered crusted surface, both disturbed and
undisturbed.
Replicate runs on different portions of the same type of sur-
face should be performed to measure precision. The use of the
EPA high volume Level 1 SASS train, in combination with the
wind tunnel, should also be considered.
Sulfates, Sulfites and Organic Adduct Analysis—
There are varying methods for measuring sulfates and sul-
fites (Hansen et al., 1976). The EPA routinely uses boiling
water extractions and the methyl thymol blue method (or the
"thorin" method) to measure sulfates on high volume sampler
filters and ion chromatography; and a cold water extraction
procedure to measure sulfate and sulfite on dichotomous
165
-------
samples (Stevens and Dzubay, 1978a) and during Level 1 analy-
sis (Lentzen et al., 1978). The ion chromatograph is cali-
brated against known sulfate and sulfite solutions. X-ray
fluorescence analysis of sulfur was shown to give reasonably
consistent results, when compared to the ion chromatographic
analysis of sulfates, for typical ambient dichotomous samples,
if the sulfur on the filter is assumed to be sulfate. X-ray
fluorescence may also be used to measure sulfates in the am-
bient air. Measurements of ammonium and hydrogen ions suggest
that the sulfate is usually in the form of ammonium sulfate.
EPA-sponsored analysis of ambient samples by ESCA X-ray photo-
electron spectroscopy also has shown the sulfur to be in the
form of sulfates (Stevens and Dzubay, 1978b).
In an alternative technique, sulfur (IV) is measured by
thermometric titration with K2Cr207, followed by the
determination of sulfate by BaCl2 titration (Hansen et al.,
1976). This technique reputedly measures true sulfate, while
the hot water extraction technique converts other sulfur
species to sulfates, resulting in less accurate determina-
tions. Evidence suggests that the ion chromatrographic tech-
nique is subject to interference because of the conversion of
S (IV) to sulfates during the alkaline phase of the analysis
because of the catalytic activity of Fe(III) and Cu(II)
(Eatough, 1979). Hansen et al. (1976) and Eatough (1979),
using Electron Spectroscopy for Chemical Analysis (ESCA), also
demonstrated the presence of sulfate and sulfite in some
samples. For smelters, sulfate plus sulfite by the thermo-
metric method is often less than half the sulfate measured by
approved EPA methods. For power plant plumes, thermometric
sulfate may sometimes be half the sulfate measured by EPA
methods. For typical ambient air the difference may only be
10 percent.
There is evidence suggesting that organic adducts of
sulfur change to inorganic sulfates if the sample ages for a
week. Significant conversion can even occur overnight
(Eatough, 1979). As a result, sulfur species analysis should
be performed immediately after sampling in the field. Evi-
dence exists suggesting that these sulfur (IV) species and
organic adducts form in power plant plumes downwind (Eatough,
1979) and that some of the organic adducts such as monomethyl
sulfate and dimethyl sulfate are mutagenic and carcinogenic
(Druckrey, et al., 1966). The formation of inorganic sulfur
(IV) may be promoted by moisture, soluble iron or copper, and
relatively low acidity. These properties of the FBC flue gas
particulates should be measured and compared to the particles
from conventional boilers.
A method needs to be developed for preparing stable sul-
fate, sulfite sulfur and organic adduct of sulfur compound
standards in matrices similar to fly ash, coal, limestone, and
166'
-------
urban dust, on filters stored cryogenically with different
reactive components micro-encapsulated. National Bureau of
Standards might help to referee tests, establishing such a
method through collaborative testing of NBS audit samples by
the EPA, the Electric Power Research Institute and the U.S.
Department of Energy contractors and in-house labs. Samples
would contain varying (but known) amounts of sulfates,
sulfites, sulfides, elemental sulfur and organic adducts on
matrices simulating environmental samples (with substances
that may cause interference) . Naturally, the stability of the
samples and sample handling procedures would have to be estab-
lished prior to the collaborative testing, and the parties
analyzing the samples (outside NBS) would not know the content
before analysis and data reporting.
In the absence of sulfate, sulfite and organic adduct
standards, it is still possible to test how well the various
techniques measure sulfates in isolation. The recent National
Bureau of Standards Standard Reference Material 2673 pre-
scribes three pairs of strips of fiberglass high volume filter
with known sulfate content and two blank filters. The tech-
niques should be calibrated by the usual methods, the NBS
filters should be analyzed, and NBS concentration compared to
the measured concentrations.
Another type of test to resolve the problem would be to
take ambient high volume samples in a power plant plume, split
the sample into four equivalent strips, and spike two of them
with equivalent known amounts of sulfites (preferably Fe S and
other species found in ambient air). The differences between
the spiked and unspiked samples could then be processed, along
with the amounts of sulfites added, using standard statistical
techniques, to yield information on the recovery of sulfites
and on the error in the measurements. The experiment could be
repeated with sulfate spiking, organic adduct spiking, and
sulfide spiking. Care must be taken that the spiked substance
is not converted to other materials because of interaction
with the collected particulates before analysis begins.
If the thermometric technique gives consistent results
When compared to NBS sulfate standards and EPA quality
assurance sulfate samples, serious consideration should be
given to measuring equivalent sulfates and sulfites by both
the standard EPA techniques and thermometric techniques in FBC
testing programs. Ideally, both techniques should be per-
formed in the field immediately after sampling. A portion of
the solution used for thermal analysis should be aged for a
few days and differences in the Fe S(IV) species concentra-
tions as a result of this aging can be used to measure total
organic adducts (Eatough, 1979) . The thermal technique
requires at least 2 to 4 mg of sample; therefore, it is
suitable for samples from FAST trains and high volume
167
-------
samplers, but unsuitable for most dichotomous samples. For
accurate analysis, samples larger than 2 to 4 mg are
recommended.
Polynuclear aromatic hydrocarbons and polar and non-polar
organic and inorganic mutagens have been observed on fly ash
surfaces from conventional power plants (Fisher et al.,
1978) . Available information on FBC shows very low levels of
polynuclear aeromatic hydrocarbons, but the material is muta-
genic, according to AMES testing (Murthy et al., 1977).
Recent research shows dimethyl sulfate and monomethyl sulfate
at concentrations as high as 830 ppm in fly ash and airborne
particulate matter from conventional coal combustion processes
(Lee et al., 1979). These substances may be, in part, re-
sponsible for the observed mutagenicity of fly ash because
these substances have proven mutagenic and carcinogenic prop-
erties (Druckrey et al., 1966; Druckrey et al., 1970; Couch et
al., 1978). The available data base should be examined to
create zero threshold MEGs for these substances. Testing
should be considered in FBC effluent and FBC plumes, if the
polar organic fractions of the solid waste and fly ash are
found to be mutagenic.
To measure dimethyl sulfate in the ambient environment,
an acid washed high volume sampler filter is used to collect
at least 20 mg of sample (preferably more). It is extracted
with methanol and analyzed by ion chromatography. The mate-
rial is believed to oxidize to sulfate in the ion chromato-
graph, so the method gives a lower limit to concentrations.
Dimethyl sulfate is measured by gas chromatography/mass
spectroscopy (GC/MS) (Lee et al., 1979).
Long-Range Plume Transformation Studies—
One of the important impacts of SC-2 emissions is the
eventual formation of acid rain and sulfates downwind of the
source. Several monitoring programs are being performed to
study this phenomenon for conventional power plants. They in-
clude:
Multistate Atmospheric Power Production Study
(MAPPS), sponsored by the U.S. Department of Energy
• Sulfur Transport and Transformation in the Envi-
ronment (STATE), sponsored by the U.S. Environmental
Protection Agency
• Major studies in Europe and Canada
• Studies by the Electric Power Research Institute
The usual rate of sulfur dioxide oxidation observed in
relatively clean surroundings is less than five percent per
168
-------
hour, although higher rates are occasionally reported. There-
fore plume transformation studies are usually performed at
large, isolated plants, to facilitate separation of a plume
from background levels after it has traveled for many hours
downwind. The techniques in these studies do not yield impor-
tant information on the oxidation rate of S02 from FBC units
unless the unit is:
Isolated from other boilers during the test
At least 200 megawatts in size or has very high
S02 emission
Short-Range Plume Transformation Studies—
Physical and chemical changes often occur in effluents
after they have been emitted to the atmosphere. As hot flue
gas and particulates exit a smoke stack and rapidly cool,
fumes may condense and the chemical composition of the ef-
fluent may change. Portable stack test trains should be.
developed that dilute and cool flue gas with clean cool air
and allow a reasonable amount of residence time for the mate-
rial to establish chemical and physical equilibrium before
analysis. A low volume dilution stack test train could be
useful for collecting particulate samples for automatic elec-
tron microscopic/X-ray fluorescent "fingerprinting". This
would aid in their positive identification and separation from
total suspended particulate levels downwind in ambient moni-
toring programs. Comparison of diluted and undiluted parti-
culate size distribution and morphology would also help in
determining whether fumes could cause significant plume
opacities not measured inside a stack.
Balloon Borne Sampling System—
The Denver Research Institute has developed a lightweight
battery-operated balloon borne sampler that is small enough
not to require an FAA license for operation. As Figure 32
illustrates, it consists of eight remotely controlled, indi-
vidually selected filters, gas sampling tubes, a pump, dry and
wet bulb temperature telemetry, and a lightweight plume de-
tector with telemetry. The sensor detects at least methane,
carbon monoxide and isobutane. The device can also be used to
fill sample bags. Currently being tested at conventional
power plants, its use for FBC should be given serious con-
sideration. Particle samples could be taken within and out-
side the plume, and analyzed by automated scanning electron
microscopy with X-ray fluorescence detection. The source
"fingerprint" obtained in this manner could then be compared
to that measured by more conventional extractive stack
testing, and any differences noted. If the projected
cross-sectional area size distribution of the particles
169
-------
RECEIVER. SERVOS.
VERIFY CIRCUIT.
TRANSMITTER,
VOLTAGE REGULATOR
8 POSITION I GAS
ROTARY VALVE SENSOR
• ELECTRONICS FOR
PARTICULATE AND
GAS SENSORS
Y
GAS SAMPLING MODULE ROTARY VALVE MODULE BASE MODULE
SIDE VIEW
Figure 32. Schematic of second generation sampling package.
(Denver Research Institute. 1979)
-------
changes appreciably between in-stack measurements and out of
stack measurements, the implications on plume opacity should
be derived. Attention should be given to any evidence of fume
condensation in the morphological analysis.
The S02 is measured in gas sampling tubes or on im-
pregnated filters, and the particulate concentration is de-
termined by weighing the filter and automatic electron
microscopy, the ratio of concentrations (after subtracting
background levels) can be compared to those measured within
the stack. If they are within acceptable limits, the method
has some value for roughly estimating particulate emission
rates. The ratio of Total Suspended Particulates (TSP) to
S02 concentrations could be multiplied by the S02 emission
rate to estimate emission rates. The ratio of concentrations
could even be multiplied by the in-stack S02 concentration
to estimate the particulate concentration in the stack, pro-
viding fume condensation and secondary particle formation are
not significant. The values thus derived could be compared to
those indicative of compliance with regulations. If C02
were measured in the gas sampling tube (and upwind), the
results could be converted to emissions.
The balloon-borne system can be modified by attaching a
gas sampling bag on the pump exhaust. Fewer particulate and
gas samplers could be used at the pump inlet to compensate for
the weight of the bag. When used upwind and downwind from the
source, the amounts of particulates, S02 , and C02 caused
by the source could be estimated. These data would then be
processed to determine emissions in the same way that it is
done in the stack. The system is currently being tested for
the quantitative measurement of sulfates, and sulfuric acid
mist; it may prove to be applicable to FBC.
A modification of the balloon borne sampling system is
under consideration for collecting samples for biological
testing. Samples would have to be collected over a number of
days to obtain enough samples for even simple AMES testing.
When tethered balloons are used, attention must be given to
power lines and to the stability and suitability of meteoro-
logical conditions.
Larger samples could be obtained over a short period of .
time by hoisting high-volume samplers, FAST samplers or
Battelle ultravol samplers into plumes with aircraft, cranes,
meteorological towers or by placing them on strategically
located buildings.
Opacity Measurements
The opacity caused by water droplets is generally ex-
cluded from the regulations, except for some local ordinances
171
-------
aimed at eliminating hazardous driving conditions on public
roads caused by fog from man-made moisture plumes.
Opacity Mass Concentration Relationship—
Plume opacity is related to the particular mass con-
centration by the Beer-Lambert Law (Ensor and Pilat, 1970):
I/I. =
(21)
Where:
Opacity = — = the fraction of a light beam inci-
*o dent on a plume that is scattered
or absorbed
I = the intensity of that portion of a
light beam transmitted through a plume
without being scattered or absorbed
Io = the intensity of a light beam before
passing through a plume
W = the particulate mass concentration
(g/m3)
L = the total distance over which the
opacity is measured (in meters)
K = the specific particulate volume/light
extinction coefficient ratio (cm3/M2)
p = the density of the particles in the
plume (g/cm3)
Ensor and Pilat (1970; I971a and I971b) have published
computerized graphs showing the value of K versus the geo-
metric mass mean radius for spherical particles with a uni-
form density and log-normal size distributions (Figure
33). Several curves are drawn on each graph, each corres-
ponding to different standard geometric deviations of the
particle radius. Each graph corresponds to a given wave
length of light and refractive index. Several conclusions
can be made from these curves:
172
-------
Geometric sUndaid deviation
R«frzc!rve index* 1.33
Wave length of lii!ht=550 nni
JO ' 10'1 10° 10'
Gcom**ric man mean radius,
Geometric standard deviation,
Refractive index-1.50
Wave lenclH of liE'>t-=550 nm
Perametor K as'a function ol th«
size distribution parameters -tor
Id water.
ib7! _,
10'' 10^ 10° \Q' 10*
Geometric most mean radius, rc* (microns)
Flgu'r* P?r?rn*l»r K as a fi^nct'on of the
log-ncrmjl size distribution parameters for a
white aerosol.
. Geometric standard deviation.
Refractive index~1.96 U.tbij:
V/av« length of lijhl^SSO nmj.
10-"
'Gf orr.elric nuss .r.e.in radus. r^ (microns)
Figure ' Parameter K as a function ol the
lo^. normal site distribution parameters lor a
bUck a»rosol.
|R*!MCtrv* -nde»-Z,SO .
JY/av* ten^lh ol light =«S50 nm
• Geometric standard deviation. 3t
1C'
10*
10-' 10-' iou jo1
Geometric in is* me*n radius. r_.!
Figure . Parameter K as a functinn of the
hypothetical rcl.'O'.li«« uulex lor iron oxide.
Figure 33. K as a function of log normal size distribution parameters.
' (Ensor and Pilat, 1971a)
173
-------
The magnitude of K is critically dependent on the
particle size distribution.
The particle refractive above a geometric mean
radius equal to twice the wave length of the inci-
dent light index does not significantly influence
the plume opacity. For particle sizes near to and
belowithe wave length of light, the refractive index
is a significant variable.
Variations in the wave length of light do not cause
large variations in K for particles much larger than
the wave length of light.
• K becomes increasingly insensitive to changes in the
size distribution for polydisperse aerosols with
diameters near the wave length of light.
A proportion of particles from FBC is irregular. Pre-
liminary testing with a dilution stack test train and photo-
microscopy shows very few spherical particles for the U.S.
Department of Energy Morgantown FBC (Carpenter et al., 1978) ,
suggesting that the effect of fume condensation on opacity is
likely to be negligible. For these particles, the concept of
a single particle radius breaks down, and the type of radius
that has been found to be most useful in light scattering
calculations is the radius of a circle with the same area as
the projected cross-sectional area of the particle, averaged
over all possible particle orientations. For highly irregular
particles (such as fibers), with dimensions near to or smaller
than the wave length of light, K can be sensitive to the
particle shape.
When the specific particulate volume/light extinction co-
efficient ratio (K) of the particles in a plume and the
density of the particles are stable, the mass concentration of
the particles in the plume may be monitored by measuring the
opacity of the plume across a measured distance. This can be
accomplished by solving the Beer-Lambert Law for the parti-
culate mass concentration (W):
p K (ln(I/IQ)
W = - ~- . (22)
L
When considering the application of this equation to com-
bustion sources, it is important to differentiate between
dust, smoke and fumes (White, 1968):
Dust is formed by the pulverization or mechanical dis-
integration of solid matter into particles of small size
174
-------
by such processes as grinding, crushing, blasting, and
drilling. Particle sizes of dust range from a lower
limit of about 1 micron up to about 100 or 200 microns.
Larger particles, although formed, settle too rapidly to
remain in suspension an appreciable time. Dust particles
are usually irregular in shape, and particle size refers
to some average dimension for any given particle. Common
examples of dusts are fly ash, rock dusts, and ordinary
flour ...
Smoke implies a certain degree of optical density and is
derived from the burning of organic materials such as
wood, coal, and tobacco. Smoke particles are very fine,
ranging from less than .01 micron up to 1 micron. Smoke
particles are spherical in shape if of liquid or tarry
composition such as tobacco smoke, and irregular if of
solid composition such as soot or carbon black . . .
Fumes are formed by processes such as sublimation, con-
densation, or combustion, generally at relatively high
temperatures. Perhaps the commonest fumes are derived
from the oxidation of metallic vapors or compounds, e.g.,
lead oxide fume . . . Fumes range in particle size from
about 0.1 micron to 1 micron . . .
When the size distribution of the smoke, fumes and dust
associated with a combustion operation are constant, there
will be a characteristic P times K value associated with each
component of the plume. The transmittance of the plume
(I/I0) is the transmittance of the smoke component multi-
plied by the transmittance of the fume component multiplied by
the transmittance of the dust component. Small particles
(smoke and fumes) can be expected to usually have smaller K's
than dust particles, when measured at wave lengths in the
visible region (0.4-0.7 microns). For particles .-with smaller
K's, smaller amounts of mass will produce the same opacity.
Thus, the ability of a given mass of smoke or fumes to produce
a given opacity in the visible region is usually greater than
for dust. If a stable relationship between mass concentration
and opacity is to be found for a source emitting smoke, fumes,
and dust, the following factors must be present:
Either two of the three components have optically
negligible effects on opacity
The volume fractions of the components are stable
• The Ks of the components are nearly equal
Data accumulated in Germany indicate that fly ash concen-
trations from some pulverized coal-fired power plants may be
175
-------
monitored by plume opacity measurements with a 95 percent con-
fidence of being within about ^30 percent of the actual
in-stack mass concentration (determined by isokenetic samp-
ling) , except during periods of soot blowing (Buenter, H.P.,
Lear Sieglerr Inc., to Cornett, C.L., personal communication).
Special Considerations for FBC Units—
i
Particles from chemically active FBC units are mostly un-
fused. This makes the size of the flue gas particles more
dependent on the physical coal characteristics than conven-
tional pulverized coal-fired boilers. This would cause a less
stable relationship between particulate mass concentration and
plume opacity than for pulverized coal-fired boilers. Never-
theless, with emission limitations requiring a 99 percent
efficiency of collection of particulates, and EPA's plans to
regulate inhalable particulates, it would be worthwhile to
investigate the relationship between plume opacity and parti-
culate mass concentration for FBC units and associated equip-
ment (coal and limestone preparation plants, solid waste
transfer and disposal operations, etc.).
Many series of simultaneous mass concentration and plume
opacity measurements need to be made for a broad range of
process variations using a variety of coals to study K x P
patterns for the technology as a whole. Studies at single
installations over a long period of time could help determine
the applicability and limitations of such optical monitoring
at individual sources.
Applications of Opacity/Mass Concentration Relationships—
Air pollution control agencies in Germany require empiri-
cal calibrations of the relationship between particulate mass
concentration and plume opacity for large pulverized fuel
fired installations. Based on such calibrations and process
related information (such as the relationship between mass
concentration and pounds of emissions per million Btu or other
relationships that can be related to standards) , individual
plume opacity standards are being set which serve to auto-
matically enforce appropriate mass concentration limits for
such sources (Peeler, 1979).
If the relationship is determined downstream of pollution
control equipment, and the equipment breaks down, the relative
large particulate concentration usually is enhanced. This
often causes the opacity/mass concentration calibration, when
control equipment is working, to underestimate the true parti-
culate concentration. If the opacity implies violation of
standards under these circumstances, the exceedance is greater
than implied by opacity.
176
-------
Ratios of average concentrations of particulates to C02
concentrations or of particulate to C02 emission rates are
routinely used to measure emissions. Upper limits to C02
concentrations for flue gas from a boiler can be used to
establish upper limits to the particulate concentration. If
particular emissions are related to opacity in a known manner,
it should then be possible to establish upper limits to
opacity indicative of compliance with standards. Oxygen moni-
toring of the flue gas may also be used to measure combustion
efficiency. The continuous monitoring of oxygen or C02 con-
tent of flue gas from federally regulated FBC units is
required to help process S02 monitoring data to determine
compliance with regulations. Both types of monitoring suffer
from interference from the limestone. Estimated mass con-
centrations from opacity measurements could be processed in a
manner similar to the processing of S02 concentration mea-
surements to derive information on compliance with emission
standards.
If the relationship between opacity and compliance is
known, along with the errors, but the results are not con-
sidered conclusive, opacity might be used to show probable
cause for suspecting a source to be not in compliance .and to
obtain entry warrants and dispatch a stack test team for ac-
curate measurements.
Continuous records of plume opacity before, during, and
after stack testing can also be used to indicate if conditions
during stack testing represent normal operating conditions.
Transmissometers are also often used as "broken bag de-
tectors" for baghouses. The sensitivity of such devices to
particulate mass loadings from FBC units needs to be deter-
mined, along with the amount of uncertainty in the relation-
ship.
If the relationship between particulate concentration and
opacity is stable, opacity monitoring could be combined with
monitoring the total volume rate of flow out the stack to
yield an estimate of the total particulate emission rate.
This emission rate could be compared with the maximum al-
lowable, when the installation is operating at full capacity.
Exceedances of the maximum allowable emissions (with a rea-
sonable margin determined by the stability of the opacity/mass
concentration relationship) could indicate non-compliance with
emission regulations. Laser anemometers or the Visiplume
S02 television remote sensor (Exton, 1977) could be used to
measure average velocities remotely, when applicable. These
velocities can be combined with the dimensions of the stack
mouth to calculate the volume rate of flow out of the stack.
177
-------
In-Stack Opacity Measurement—
For those sources with in-stack transmissometers already
installed, the existing equipment should be used, providing
they:
Meet specifications and are properly installed,
operated, and maintained.
• Have an acceptable history of low zero and span
drift at the actual installation
A long-term program of stack testing should have a per-
manently installed transmissometer. When this is not practi-
cal or cost-effective, a portable transmissometer with a stack
test probe should be used where applicable. Care should be
taken, however, to clean the stack sampling port before
inserting the probe, especially for units under considerable
negative pressure, because dirt can contaminate the optics
when the probe is inserted or during operation. Random drift
of instrument response due to thermal expansion can be avoided
by allowing the in-stack probe to come to thermal equili-
brium. The device should also have suitable structural design
and integrity to withstand both the air flow conditions and
temperature of the plume without damage or calibration drift.
Units that have proven their accuracy against permanently
mounted transmissometers at boiler installations are pre-
ferred. Compressed air should be used, when ecessary, to keep
the optics clean throughout the test period and to confine the
plume to a measurement slot of known' length.
The opacity should be recorded continuously, and trans-
lated into an average value of In (I/I0) during the time
period of the particulate sampling. Automatic equipment is
available for this purpose. The distance over which the
opacity was monitored, the particulate mass concentration, the
location where opacity and mass concentration was measured,
and appropriate process data to translate the concentration
into terms comparable with regulations, should also be rec-
orded.
Some information about the potential variability of K can
be obtained by calculating K from measured particulate size
distributions, particularly if cut sizes less than 0.5 micron
are included. (See Figure 33 for size distributions that are
approximately log normal.) Aerodynamic size distributions can
be of some value in such studies although projected cross-sec-
tional size distributions are preferred.
If there are no visible emissions, transmissometers in
this effluent may not yield useful information and may not be
appropriate (except as an indicator of non-control if there is
178
-------
an appreciable opacity upstream of control equipment), unless
the equipment is extremely sensitive. Opacity monitoring
would not be appropriate in wet plumes (plumes containing
enough .uncombined water as to cause an appreciable opacity
compared to that of the pollutants in it), unless the purpose
is to measure the water droplet concentration.
Remote Measurement of Plume Opacity—
Trained Observers—Plume opacity can also be measured
outside smokestacks. EPA Method 9 involves the use of trained
observers to measure the plume opacity by its appearance. The
procedure contains detailed procedures for the training and
certification of observers every 6 months and procedures to be
used in field tests. When the plume does not contrast well
against the background, the method can result in gross under-
estimates of opacity. Under contrasting conditions, field
experiments indicate that 99 percent of Method 9 determina-
tions were no more than 5 percent above the results of
in-stack measurements for black plumes and no more than 7.5
percent of in-stack measurements for white plumes. The
accuracy of local smoke inspectors could be different and
should be periodically audited.
Auditing the accuracy of certified smoke inspectors by
researchers at Rutgers University, using standard smoke
generators and EPA Method 9 testing procedures, showed that
only 40 percent of the inspectors could pass all criteria for
certification at least 6 months after their last certifica-
tion. Average errors for white smoke were 6.2 percent; for
black smoke, the average error was 5.5 percent. Some in-
spectors did twice as well as this and some were twice as
bad. Errors above 15 percent occurred about 3.6 percent of
the time for white smoke and 2.4 percent for black smoke.
Some inspectors erred by this much about 10 times as often as
the 'average.
Data on secondary aerosol formation (fume condensation)
in fluidized bed boiler flue gas should be gathered to deter-
mine if such particulates could have an appreciable opacity
which could influence the relationship between in-stack and
out-of-stack measurements. The use of stack sampling trains
that dilute the particles with cool air before collection,
along with conventional samplers, could be useful for checking
for such aerosols. Ambient plume sampling could also be
used. The projected cross-sectional area size distribution of
the particles in the plume after cooling should be contrasted
to that of the particles at stack temperature for this pur-
pose. Fume particulates tend to be spherical.
When changes in particulate size distribution are
negligible, out-of-stack measurements of opacity can be used
179
-------
as a quality assurance auditing procedure for in-stack opacity
measurements (considering the errors in such measurements).
EPA Method 9 could be used during the daytime to check for
gross in-stack measurement errors. The remote measurements
are valuable because regulations cover out-of-stack opacity
not caused by water.
The Method'of Contrasting Targets—The successful appli-
cation of photographic and telephotometric techniques to the
measurement of plume opacities has been documented (Conner and
Hodkinson, 1972). This was accomplished by measuring the
difference in the apparent luminances of contrasting targets
when seen through a plume ( B( - B£ ), and the apparent
difference in luminances of these targets when seen clear of
the plume (BI - 82). The transmittance, T (=I/I0), of
the plume is calculated from the relationship:
Bi - B2
T = g± g±. (23)
. Bl B2
The relative luminances of the targets may be obtained
with direct telephotometer measurements, or by photographing
the targets and obtaining the measurements from the negative
in a laboratory using a densitometer. A high quality camera
should be used to minimize changes in the relative luminance
calibration across the negative or photographic plate.
Neutral density filters are positioned along the film plane to
produce a calibration scale on the negative.
Telephotometer measurements of plume opacity, using the
method of contrasting targets, yielded results which were
within, on the average, 4 percent of the opacity of a white
experimental plume as measured in the stack (Conner and
Hodkinson, 1972) . The agreement between the in-stack and
out-of-stack opacity measurements was within the range of
random experimental error and differences between the wave
length dependent light energy response of the devices. The
photographic method also showed promising results.
Laser Radar—Laser radar may be used to measure plume
opacity during day and night. This technique works by mea-
suring the amount of backscatter of light (normalized for the
response characteristics of the laser radar with distance)
from particles in front of the plume (Io) to the backscatter
from the particles behind the plume (I). Opacity is
1-I/IO. Preliminary indications are that it is accurate to
within + 3 percent to +6 percent of in-stack measurements
(Dybdahl and Cunningham, 1979). Inhomogeneous background con-
centrations of particles near the plume in heavily polluted
environments can sometimes cause greater (up to 12 percent)
errors (Conner, 1979).
180
-------
Laser radar is especially useful at night, when other
less exotic techniques of opacity; measurement (except for
in-stack measurement) are not applicable. The National
Enforcement Investigations Center; (Denver, Colorado) laser
radar unit has proven to be useful in enforcement-related
activities. It can be dispatched for surveillance of sources
suspected of "dumping at night" or falsifying in-stack opacity
measurements. Most laser radar units usually measure plume
opacity at one wave length only, rather than across a range of
wave lengths. This can cause the raw readings to be up to 50
percent lower when opacity is dominated by submicron particles
for some laser radar units, especially for particles less'than
the wavelength of light in size (Conner, 1979). There are
simple visual checks for submicron particles—plumes with
opacities dominated by very small particles tend to appear
blue and cast red shadows. It is possible to correct for the
spectral response of a laser radar using sun photometers to
measure the wavelength dependence of the extinction coeffi-
cient (Conner, 1979) . Multiple wavelength laser radar units
are also being developed to measure opacity and to give some
information on particulate size distribution.
Sun Photometer—Opacity can also be measured by measuring
the intensity of light from the sun when viewed through a
plume and clear of the plume. The angle viewed through the
plume should also be considered, since opacity measurements
should reflect measurements perpendicular to the plume axis.
DATA REPORTING
The results of all FBC monitoring programs should be re-
ported to 1) the permit-granting authority for the facility
tested, 2) the EPA Industrial Environmental Research Labora-
tory, 3) EADS computerized data bank, (when it is ready to
accept data) and 4) to the EPA Regional office, if required.
The data should also be sent to:
U.S. Environmental Protection Agency Advanced Process
Branch
Industrial Environmental Research Laboratory
Research Triangle Park, North Carolina 27711
(919) 541-2825
The results of any programs comparing the results of in-
stack monitoring against those of non-intrusive methods should
be reported to:
U.S. Environmental Protection Agency Quality Assurance
Division
Research Triangle Park, North Carolina 27711
(919) 541-2580
181
-------
and
U.S. Environmental Protection Agency Emission Measurement
and Characterization Division
Mail Drop 46
Research Triangle Park, North Carolina 27711
(919) 541-3034
i
The results of opacity measurement programs (and the
methods of testing used) should be reported to:
U.S. Environmental Protection Agency Emission Measurement
and Characterization Division
Stationary Source Emission Measurement Research Branch
Mail Drop 46
Research Triangle Park, North Carolina
(919) 541-3173
Technical Coordinator
Technical Support Branch
U.S. Environmental Protection Agency Division of Sta-
tionary Source Enforcement
Mail Drop 7
Research Triangle Park, North Carolina 27711
(919) 541-4571
182
-------
SECTION 5
WATER QUALITY MONITORING STRATEGIES
INTRODUCTION
FBC of coal is a fossil fuel combustion technology that,
in respect to the emission of water pollutants, is similar to
those from conventional pulverized coal combustion plants.
Conventional pulverized coal combustion plants which burn
high-sulfur coal are required to use some form of flue gas
desulfurization to remove the sulfur dioxide prior to dis-
charge of the gaseous emissions to the atmosphere. Many of
the flue gas desulfurization techniques produce an aqueous
mixture or sludge as the waste product which must be disposed
of in a landfill or by other methods. FBC technology, on the
other hand, removes the sulfur dioxide at its source,'pro-
ducing only solid wastes, and therefore has no aqueous or
liquid effluents associated directly with the technology.
The use of FBC technology for commercial purposes would,
of course, require other processes such as steam generation
and water cooling. The aqueous discharges associated with the
use of the FBC technology should, therefore, be subject to the
same regulations and standards as any coal-fired facility.
Any water quality monitoring requirements applicable to con-
ventional coalfired facilities are equally applicable to FBC
of coal plants. Based on theoretical and available experi-
mental information, the waste streams from FBC installations
and the appropriate pollution control systems will meet appli-
cable discharge standards, including those directed toward
disposal of solid wastes in landfills or by other means.
The FBC water quality monitoring strategies proposed in
this section primarily address regulated pollutants. However,
other pollutants are included in the proposed monitoring pro-
grams either because of proposed regulations or because such
pollutants are present or could potentially be present in the
FBC waste streams. Some of the other pollutants are included
in the monitoring programs because of their potential as indi-
cators of pollution, or because of their hazard potential.
This section covers the water quality impact generators,
conditions under which ambient (or other type of monitoring)
183
-------
is required, the applicable regulations, selection of pollu-
tants to be monitored, design of the monitoring systems, ap-
plicable water quality measurement and data analysis method-
ologies.
IMPACT GENERATORS
Several potential impact generators were identified in
Section 2. Proper engineering should prevent degradation of
the groundwater at the plant site and at the site of disposal
of solids. Proper design and operation of wastewater treat-
ment facilities at the plant site should also prevent contami-
nation of any surface waters potentially impacted by the
plant.
Three impact generators may have a potentially signif-
icant impact upon groundwater quality during operation of an
FBC facility:
• Leachate from coal and sorbent storage piles
Infiltration from cooling water ponds or canals,
if used
Leachate from disposal of solid wastes
Impact generators which may affect surface water quality in-
clude:
Chemicals used for treatment of water used in
boiler and cooling tower operations
Leachate from coal and sorbent storage piles
• Aqueous wastes from regeneration operations
• Potential aqueous waste resulting if on-site
hydration is used to treat solid wastes prior to
landfill disposal
The aqueous effluents generated by use of FBC technology
and accessory power-generating equipment would probably be
regulated under the National Pollutant Discharge Elimination
System (NPDES). Under-this system, the state in which a •
facility is located would determine: 1) limits of pollutant
concentrations in the effluents, and 2) effluent monitoring
requirements based on effluent standards for steam electric
power-generating facilities; the use, classification, and
water quality of the receiving surface water; and additional
information on specific processes which generate regulated
pollutants and/or pollutants which may be hazardous.
184
-------
CONDITIONS WHERE AMBIENT WATER QUALITY MONITORING IS INDICATED
The variations in processes which could be used and
types of pollutants generated as a result of the processes
have been summarized in Section 2. The complexity of opera-
tions at a facility employing FBC as an energy conversion pro-
cess will determine the need for or requirements of an ambient
water quality monitoring program.
A potential source of water pollutants common to all
coal combustion facilities results from storage of the coal,
and, in the case of FBC, from the storage and handling of the
sorbent and the solid wastes from the combustion process. If
coal, sorbent, and the generated solid wastes are stored in
areas exposed to the weather, leachate formation, and subse-
quent contamination of surface and groundwaters may occur.
The coal leachate may contain contaminants which are poten-
tially mutagenic. Therefore, a facility which stores solid
materials in a manner which allows formation of leachate and
subsequent contamination of water resources would be a candi-
date for source-oriented ambient monitoring. If the solid
materials are not exposed to weather, as at the Georgetown
University FBC facility, the potential for generation .of con-
taminants and for contamination of the water resources would
be minimal and ambient monitoring of the water resources would
not be indicated.
The processes summarized in Section 2 list several
direct aqueous effluents, each of which may contain numerous
pollutants. If a facility employing the FBC technology does
not utilize boilers or other processes which would generate
aqueous effluents, an ambient monitoring program would not be
indicated. If processes or equipment such as boilers, cooling
towers, or ponds are employed, effluents from the facility
would be regulated in the same manner as conventional
coal-fired facilities.
Other processes that could be employed at an FBC facil-
ity, e.g., sorbent regeneration and hydration of solid resi-
dues prior to transportation to a disposal site, would indi-
cate a need for an ambient monitoring program. If either of
these two processes are employed, aqueous effluents specific
to the FBC process will be generated. If the generated wastes
are discharged untreated to a water resource, an ambient moni-
toring program would be indicated. If, however, the waste-
water is either treated on-site and discharged to a sanitary
sewer or discharged directly to a sanitary sewer, an ambient
monitoring program would not be necessary.
Disposal of the solid wastes generated by the FBC tech-
nology would, in general, require an ambient monitoring pro-
gram at the site of disposal or in the water resources which
185
-------
may become contaminated by leachate or direct spillage of the
solid wastes. An exception would occur if the solid wastes
were used to produce "concrete" blocks or were used as roadbed
fill. In this event, their suitability for the intended use
would have to be determined, as would the amounts and types of
pollutants generated and their fate as a result of such use.
An ambient monitoring program, however, would not be required
at that time. '
In summary, the conditions under which an ambient moni-
toring program would be indicated are restricted to the dis-
posal of solid wastes and aqueous effluents or pollutants uni-
que to FBC technology. Monitoring of aqueous effluents is
further restricted to cases where the effluents are discharged
directly to a receiving water or storm sewer rather than to a
sanitary sewer. The aqueous effluents unique to FBC tech-
nology include those resulting from the regeneration process
and possibly from hydration of the FBC solid wastes.
The aqueous effluents and water pollutants resulting
from operation of an FBC unit include two generic sources.
Those resulting from coal pile drainage and use of power
generating equipment are similar to those produced by conven-
tional coal-fired power plants, with the exception of drainage
from piles of sorbent and accessory spent bed material hand-
ling or regeneration operations. The second primary source of
water pollutants is from disposal of the solid wastes, bed re-
ject material and fly ash which has been collected by precipi-
tators, baghouses, or cyclones.
Solid wastes are either buried in a landfill or are re-
used in some form. Three main areas for use of the solid
wastes can be identified:
Agricultural
Commercial/Industrial
Environmental (Land Reclamation)
The primary uses of solid wastes for agricultural pur-
poses are for application of the spent bed material to acid
soils to increase their fertility or for soil conditioners.
For some crops, the materials may be employed as a direct •
source of soil nutrients.
The commercial/industrial uses of the solid wastes may
include the spent bed material and the collected fly ash. The
uses may include;
"Concrete" blocks for building construction
186
-------
Aggregate for road-base construction
Use as gypsum products or raw material for such
construction
Recovery of metals
Use for construction of artificial reefs
In land reclamation, the high pH and alkalinity poten-
tial of the solid wastes may ameliorate or neutralize the ef-
fects of "acid mine leachate,11 which occurs in conjunction
with both strip mining and deep mining of coal. The use of
the material to neutralize the effects of acid rain upon areas
such as the Adirondacks would also be considered as construc-
tive landfilling/land farming/land reclamation.
Monitoring to determine the effects of the FBC solid
wastes and the individual pollutants on the environment from
the utilization of solid wastes for any of the cited uses will
not be included in this report. The EPA should, however, con-
tinue its programs to determine the environmental effects of
emissions or effluents from solid waste utilization processes
-(Henschel, 1979).
The following discussion will concentrate on the pos-
sible effects and consequent ambient monitoring requirements
necessitated by the direct aqueous effluents of the FBC tech-
nology and its accessory power generation or conversion pro-
cesses, and the indirect aqueous effluents resulting from
landfill disposal of FBC solid wastes.
Although nearly all EPA regulations may apply to some
portion of the aqueous effluents generated by operation of an
FBC unit, two sets of regulations are likely to predominate.
Direct aqueous effluents discharged to surface or subsurface
waters would be covered by the NPDES regulations. Disposal of
solid wastes generated by FBC would be covered by the Resource
Conservation and Recovery Act (RCRA) regulations.
In 40 CFR, Part 261 of RCRA, the four categories under
which a waste material may be classified as hazardous are: 1)
toxicity, 2) ignitability, 3) corrosivity, and 4) reactivity.
Toxicity of a waste is based on the concentration of eight
trace metals present in the leachate of the waste when the
leachate is generated by a specified procedure, using a dilute
acetic acid solution as the solvent. If the concentrations of
any of the eight trace elements in the leachate are greater
than one hundred times the allowable concentrations stated in
the National Interim Primary Drinking Water Regulations
(NIPDWR), the waste may be considered toxic and therefore haz-
ardous. In a recent test of several solid residues from both
Atmospheric and Pressurized FBC units, none of the residues
187
-------
were found hazardous according to the leachate toxicity test
(Sun and McAdams, 1979).
The remaining three classes of hazardous materials are
not likely to apply to FBC solid wastes. The solid wastes are
not expected to contain sufficient unburned carbon to be
ignitable, nor would they be expected to cause fires through
friction, spontaneous chemical changes, or retained heat.
This assumes, however, that the solid wastes, particularly the
spent bed material, have been allowed to cool to ambient tem-
perature prior to introduction into a landfill site. If the
wastes are landfilled while still hot (in excess of 250°C) ,
the potential exists for causing fires if combustible wastes
such as cloth, paper, wood, or other organics are also present
in the landfill.
Corrosivity may apply to FBC wastes. If the spent bed
material is hydrated prior to shipment to a landfill for dis-
posal, the resultant slurry can be defined as corrosive, since
the pH of the slurry would be greater than 12.
The residual calcium oxide content in the FBC solid
wastes, particularly the spent bed material, could cause the
release of appreciable quantities of heat upon hydration.
However, the amount and nature of the heat released is not be-
lieved to be sufficient to cause the wastes to be classified .
as reactive (Henschel, 1979).
In addition to the pollutants listed in Table 41, the
EPA has announced water quality criteria for 27 of the 65
criteria pollutants (EPA, 1979d) . Of the 27 pollutants, 8 may
be significant in assessing potential impacts of FBC effluents
on the environment. Seven are metals, and one an organic com-
pound, fluoranthene, which has been detected in ;Level 1 or
Level 2 analysis of FBC effluent streams. Some .of the ele-
ments have significantly different values than those listed in
the National Interim Primary Drinking Water Regulations. For
example, the criterion listed for silver is 10 ug/& for
protection of health, and 0.0090 pg/i for protection of
freshwater aquatic life. Beryllium and thallium are included
in the criteria, and values are provided for protection of
human health. The lead, beryllium and cadmium criteria for
freshwater are based on the hardness of the water. The sug-
gested criteria for fluoranthene are 200 ug/A for protec- .
tion of human health, 250 ug/fc for protection of fresh-
water aquatic life, and 0.30 yg/& for protection of marine
organisms.
These criteria were developed in accordance with Section
304 (a) of the Clean Water Act and are not water quality stan-
dards. The criteria would become standards only if adopted by
188
-------
TABLE 41. APPLICABLE AMBIENT WATER QUALITY STANDARDS'
Pollutant
NIPDWR
NSDWR
Irrigation
Aluminum
Arsenic
Barium
Beryllium
Boron
Cadmium
Chloride
Chromium
Cobalt
Copper
Fluoride
Iron
Lead
Lithium
Manganese
Mercury
Molybdenum
Nickel
Nitrate-N
Selenium
Silver
Sulfate
Total Dissolved Solids
Zinc
pH
0.05
1.0
0.01
0.05
£
2.4f
0.05
0.002
10.0
0.01
0.05
250
1.0
0.3
0.05
250
500
5.0
6.5-8.5
5.0
1.0(0.1)
0.1
0.75
0.01
0.10
0.05
0.20
1.0
5.0
5.0
2.5
0.2
6.01
0.2
0.02
5000
2.0
4.5-9.0
a All concentrations in mg/£ except pH.
b EPAf 1978c.
c EPA, 1977a.
d National Academy of Engineering, 1973.
e EPA, 1976c.
f Limit based on mean air temperature for the region of use.
189
-------
a state or the EPA under Section 303 of the Clean Water Act.
The criteria are useful in assessing the importance of pollu-
tants. :
Sections 265.90-94 (Subpart F) of RCRA also describe
the testing and analysis required to determine the background
levels of the quality of groundwater underlying the facility.
In general, the parameters to be monitored during the various
phases of the program include:
Specific conductivity
Chloride
Total dissolved solids
Dissolved organic carbon
Principal hazardous constituents or indicators
thereof found in the largest quantity in the haz-
ardous waste
National Interim Primary Drinking Water Regulation
parameters
National Interim Secondary Drinking Water Regula-
tion parameters
Beryllium
Nickel
Cyanide
Phenols
Organics (scanning by gas chromatography)
Some pollutants listed are not expected to be contami-
nants in the solid wastes from PBC units. These parameters
include phenols, organics, pesticides, radioactivity, cyanides
and organic carbon. While these pollutants are not expected
to be a problem with FBC solid wastes, their background con-
centrations :should be determined in the groundwater. Determi-
nation of background concentrations and continued monitoring
for these pollutants would be desirable because of the pos-
sible range of types of solid wastes which may be disposed of
in a landfill with the FBC wastes. Mixtures of different
wastes could result in alteration of leachate composition with
respect to that generated by either of the wastes, with poten-
tial significant adverse affects upon the environment.
SELECTION OF POLLUTANTS TO BE MONITORED
Selection of pollutants which should be monitored will
use information generated in Level 1 and Level 2 analysis of
FBC unit effluents, and other applicable tests. Applicable
existing and proposed effluent and ambient standards are used
to provide an initial basis for segregating the potential pol-
lutants among three categories. The applicable regulations
and sources of desirable pollutant limitations include:
190
-------
National Interim Primary and Secondary Drinking
Water Regulations
Resource Conservation and Recovery Act, Section
265, Subpart F
EPA Quality Criteria for Water
• EPA Consent Decree for Criteria Pollutants
Recommended Limits of Pollutants in Water Used for
Continuous Irrigation
Electric Power Plant Effluent Guidelines
Determination of toxicity or hazard potential of the
pollutants was made by use of Minimum Acute Toxicity Effluent
values, and by the carcinogenicity ratings of compounds listed
by the National Academy of Science (1972). The procedure for
selection of pollutants described in Section 3 was employed.
Minimum Acute Toxicity Effluent (MATE) values have been
developed for most of the elements (Cleland and Kingsbury,
1977) . MATEs describe maximum desirable pollutant concen-
tration in effluents, and have been developed for most ele-
ments as well as numerous organic species. Comparison of the
concentration of species in an effluent, in this case the
leachate from FBC solid wastes, with the respective MATE
values provides an estimation of which of the elements may
pose a threat to public health or ecological stability. For
the water media as well as for the air and land media, MATE
values are available for both health and ecology. Where they
differ, the most restrictive value was chosen for comparison
with the leachate data. The significance of a given element
is determined by the ratio of concentration to the MATE
value. Three levels of significance have been identified:
the first level includes those elements whose ratios are
greater than 1; the second, elements with ratios ranging from
0.1 to 1; and the third, elements having ratios less than 0.1.
Elements whose concentrations in the leachate of the FBC
solid wastes exceed the most restrictive MATE values include:
Lithium Sodium Titanium
Boron Iron Vanadium
Magnesium Zinc Nickel
Aluminum Manganese Copper
Silicon Potassium Chromium
Phosphorus Calcium Arsenic
Selenium Lead Silver
191
-------
Elements whose ratio of concentration to MATE values
were between 0.1 and 1 include:
Barium
Beryllium
Elements whose ratios of concentration to MATE values
were less than 0.1 include the following, plus all other ele-
ments not included in Tables 19 or 20:
Fluorine Yttrium Tin
Chlorine Zirconium Antimony
Scandium Molybdenum Thallium3
Cobalt Lanthanum Rhodium3
Gallium Cerium Mercury3
Rubidium Niobium Cadmium3
Strontium Tungsten
Sulfur, although high in concentration, was not compared
to the MATE values, since it was predominantly present in the
form of sulfate, particularly calcium, magnesium, sodium or
potassium sulfates. The sulfates of the metals are included
in determination of respective MATE values.
Some forms of polycyclic organic matter (POM) were mea-
sured on the solid wastes in three studies (Merryman et al.,
1977; Pillai and Roberts, 1977; Ryan et al., 1979). The data
obtained in the former two studies are summarized in Table
11. To assess the impact of solid waste upon the water media,
it is necessary that the contaminants in the solid waste be-
come dissolved in water or at least be in suspension and
available for absorption or ingestion by organisms. If we
assume that all of the contaminants measured will be dissolved
in a single aliquot of water, and the volume of water is
equivalent to the weight of the solid waste, none of the POMs
would exceed the MATE values for water. Only two of the POMs,
benzo(a) pyrene, and dibenzo(a,h) anthracene, would have a
concentration to MATE value ratio greater than 0.1.
Major anionic species such as nitrate, nitrite, sulfate
and sulfites are not included in the list MATE values separate
from a cation, although the anions may have toxic properti.es
of their own separate from any cation. Pertinent anion data
from some leachate studies on FBC solid wastes are summarized
in Table 21.
3These metals have low MATE values or are highly toxic;
however, their concentrations were consistently below the
detectable limit of the analytical method employed.
192
-------
Two distinct process effluents can affect the quality of
water; therefore, each effluent type will be-processed sepa-
rately: 1) solid wastes, including bed reject material and
collected flue gas particulates, and 2) the liquid waste
streams associated with power generation and those associated
with sorbent regeneration and on-site hydration of the solid
wastes. Based on the National Interim Primary Drinking Water
Regulations, the pollutants belonging to Category I include:
Arsenic Chromium IV Nitrate-N
Barium Lead Selenium
Cadmium Mercury Silver
Fluoride
The standards for the pesticides, radioactivity, and
bacteria do not apply to FBC solid wastes, and therefore are
not included in Category I.
Development of the Category II pollutants list uses the
proposed Secondary Drinking Water Regulations, the Resource
Conservation and Recovery Act Section 265, and the 27 pol-
lutants listed in the EPA Consent Decree for Criteria Pollu-
tions, for which ambient standards have been proposed. Once
again, pollutants not applicable to FBC solid wastes are not
included in the list. The pollutants in Category II include:
Chloride Total Dissolved Solids
Copper Zinc
Iron Thallium
Manganese Fluoranthene
Sulfate Beryllium
Nickel Dissolved Organic Carbon
Organics Specific Conductivity
Pollutants included in a list used to develop Category
II pollutants already included in Category I were not dupli-
cated.
The Level I/Level 2 analysis of leachate of FBC solid
wastes indicates that 11 additional elements are present in
concentrations which exceed MATE values but are not regulated
at the federal level as a drinking water or ambient standard.
These include:
Lithium Silicon Calcium
Boron Phosphorus Titanium
Magnesium Sodium Vanadium
Aluminum Potassium
All of these elements are candidates for inclusion in an am-
bient monitoring program. They are present in high concentra-
tions and therefore may indicate pollution of the ground
193
-------
water. Major elements on this list, such as calcium, mag-
nesium, sodium, potassium, or silicon, could be used as indi-
cator or support parameters, while the remainder would not be
considered as pollutants for monitoring. If a hydrologic sur-
vey indicates that an aquifer being used as a source of irri-
gation water has the potential to be contaminated by FBC solid
waste leachate from a landfill or other source, additional
pollutants could be added to Category II. The pollutants
which could be added include boron, aluminum, cobalt and
lithium.
Those pollutants not included as candidates for moni-
toring could be included if the hazard potential for the pol-
lutants so indicate. Determining the actual hazard potential
of these pollutants requires information such as the amount of
leachate formed, the concentration of pollutants in the leach-
ate, the dilution ratio of the leachate by the groundwater
into which the leachate discharges, and the rate of attenua-
tion of the pollutants by soil and other geologic formations.
The pollutants in Categories I and II could be increased
in numbers if state regulations for maximum concentration of
pollutants are included in the decision model. If state regu-
lations are not considered, Category III pollutants will in-
clude all of those not included in Categories I and II. The
pollutants included in the three categories are summarized in
Table 42.
TABLE 42. SUMMARY OP POLLUTANTS SEGREGATED AMONG CATEGORIES
Category I Category II Category III
Arsenic, Barium
Cadmium
Fluoride
Chromium IV
Lead
Mercury
Nitrate-N
Selenium
Silver
Calcium
Magnesium
Chloride, Copper
Iron, Manganese
Sulfate, Nickel
Zinc, Thallium
Beryllium, Fluoranthene
Total Dissolved Solids
Dissolved Organic Carbon
Specific Conductivity
Organics (specific and
general)
Benzo(a)pyrene
Benzo (a , h) anthrance
Lithium
Aluminum
Phosphorus
Sulfide
Titanium
Vanadium
Boron
Silicon
Sodium
Potassium
Nitrite
Sulfite
Rubidium
Gallium
Uranium
Thorium
Cobalt
Palladium
Platinum
Lanthanides
Actinides
194
-------
The pollutants discharged in aqueous effluents at an FBC
site will depend on the processes used; the type and number of
pollutants included in each category also depend on the pro-
cesses used, including those used to treat the wastewater.
Table 43 lists those pollutants regulated in effluents in
power plants using chemicals for water treatment. The pollu-
tants produced by the FBC process may include any of those
listed in Tables 11, 18, and 19. Since the effluent streams
containing these pollutants would be treated along with other
effluent streams generated at a site, the potential number of
pollutants discharged is very high. The number could increase
if chlorinated effluents from cooling tower blowdown are
treated along with coal pile drainage or liquid effluents from
hydration of solid wastes. Many of the organic compounds
would, however, be absorbed into the suspended solids, most of
which would be removed during the wastewater treatment pro-
cess.
While most of the pollutants would in all likelihood be
removed from the wastewater prior to discharge, if a suffi-
cient number and quantity were discharged, the result would be
a low-level, long-term impact upon the aquatic ecosystem. The
solid waste pollutants categorized in Table 42 are also to be
included in monitoring for aqueous effluents which may be dis-
charged directly to a surface water or infiltrate from sedi-
mentation basins into the groundwater. However, because of
the higher potential for organic pollutants to be discharged
at the plant site, more specific organics would be included in
the monitoring requirements.
The organic pollutants which should be included in the
three categories for ambient water quality monitoring programs
are: Category I, Category II (fluoranthene, benzo(a)pyrene,
and dibenzo(a,h)anthracene), and Category III (all other
organics listed in Table 11).
The pollutants listed in Table 43 would be included in
Category I for sampling and analysis of plant effluents, but
would be included in Category II for monitoring programs aimed
at determining water quality of the receiving waters. In case
of duplication, of course, pollutants would remain in the
lowest numbered category consistent with applicable standards.
The four areas of ambient water quality monitoring re-
quired for FBC facilities are listed below and discussed in
the following sections:
Groundwater
Surface Water
Physical Sinks
Biological Sinks
195
-------
TABLE 43. SELECTED EPA EFFLUENT STANDARDS FOR
STEAM ELECTRIC POWER GENERATING PLANTS
(Greenwood et al., 1979)
Pollutant or
characteristic
Maximum
one daya
Average of
daily values for
30 consecutive
days shall not
exceeda
pH
(all discharges)
Polychlor inated
Biphenyl Compounds
TSS
Oil and Grease
Total Copper from Metal
Cleaning or Boiler Slowdown
Total Iron from Metal
Cleaning or Boiler Slowdown
Free Available Chlorine
from Cooling Tower
Slowdown
Materials Added for Corro-
sion Inhibition in Cooling
Tower Slowdown
From Cooling Tower Blow-
down
6.0 - 9.0
No discharge
100.0 30.0
20.0 15.0
1.0 1.0
1.0 1.0
0.5 0.2
No detectable amount
Zinc
Chromium
Phosphorus
Heat from Main
Condensers
1.0
0.2
5.0
None
1.0
0.2
5.0
a Expressed in mg/A except pH.
196
-------
GROUNDWATER MONITORING
The groundwater monitoring which would be required as a
result of operation of FBC facilities includes areas around
landfill sites where FBC wastes are disposed of, and areas or
aquifers around the FBC facility if that facility has the
potential to contaminate the groundwater from coal sorbent
waste solids, pile leachate, canals, or ponds used for cooling
water or for sedimentation.
An additional source of potential contaminants is flue
gas particulates which settle on the ground around the facil-
ity. Little is known about the amount of flue gas particu-
lates which may settle on the ground, the area which may be
affected, or the potential for groundwater contamination.
Most probably the filtering and absorption capacity of soil
would prevent groundwater contamination from this source.
The methods to be employed in selection of groundwater
sampling locations, the methods of sampling and the costs in-
volved, have been adequately summarized (Todd et al., 1976;
Everett et al., 1976). When applying the methods detailed in
summaries, the specific requirements of the states which have
been implemented in groundwater monitoring or quality programs
must be adhered to, if a facility is located in a state with
such a program. The groundwater monitoring program must also
accommodate the monitoring requirements proposed in the
Resource Conservation and Recovery Act, Section 265, Subpart
F. The RCRA monitoring requirements are addressed to monitor
the source for any pollutants so that appropriate actions can
be taken to prevent groundwater contamination if the leachate
quality demonstrates the need for such actions. The required
RCRA monitoring system can be a functional part of the source-
oriented groundwater monitoring system and can be used as an
early warning system for potential problems.
SURFACE WATER MONITORING
The surface water monitoring which would be required as
a result of operation of FBC facilities includes the surface
waters receiving direct discharges from a facility, and those
receiving indirect discharges from airborne particulates or
from contaminated groundwater which discharges to a surface
water. The surface water for which an ambient monitoring pro-
gram must be designed includes rivers and streams, lakes and
ponds, estuaries, and oceans.
Each surface water resource may require a slightly dif-
ferent strategy to effectively monitor water quality and
changes in water quality induced by operation of an FBC facil-
ity.
197
-------
PHYSICAL SINKS
Physical sinks are caused primarily by precipitation,
sedimentation or absorption of pollutants into or onto the
bottom sediments of a body of water. The characteristics
which cause these actions include solubility, density, and
partition coefficients. Physical sinks are not permanent, and
materials contained in the sinks may be recirculated through
the water column or biotic component of the ecosystem if con-
ditions change. An example of recirculation is the mercury
contamination of Lake St. Clair. Mercury metal discharged to
the lake was, in theory, to remain permanently in the sedi-
ments and remain harmless to the environment. However,
long-term action by bacteria produced methylmercury which is
soluble and assimilatable; it consequently bioaccumulated in
high concentrations in fish, rendering them unfit for con-
sumption. Similarly, a physical-chemical process may move
contaminants out of a sink. When a lake such as Lake Erie be-
comes anoxic, a reducing environment is initiated and pollu-
tants such as iron and phosphorus can be solubilized and
"move" from the sediments to the water layer.
A third mechanism of pollutant removal from physical
sinks is direct ingestion of the pollutant by animals. The
animals may be oligochaetes who are subsequently consumed by
higher forms, or fish such as carp, sturgeon or catfish. In
any event, pollutants deposited in physical sinks may not re-
main there permanently.
Physical sinks may occur in all types of surface waters;
however, their importance varies with the type of surface
water, A shallow stream with a moderate to swift current,
while not amenable to formation of physical sinks, often has
pools formed at bends or wide parts where a sink can form. In
general, all surface waters where the current is drastically
reduced, including lakes, ponds and oceans, may have physical
sinks that can be sampled.
The conditions under which a physical sink should be in-
cluded in the source-oriented ambient monitoring are:
All surface waters where an FBC facility is the
primary discharge to the system either from direct
aqueous effluents or indirect effluents from land-
fill leachate.
Surface waters where the type of pollutants dis-
charged by an FBC facility differ significantly in
speciation from other discharges to that surface
water.
198
-------
Times when biological surveys indicate that the
biota of the receiving water is changing in diver-
sity or quality.
After any spills of FBC process wastes reach the
surface water due to either direct discharge or
discharge via leachate.
BIOLOGICAL SINKS
Biologic uptake of a pollutant represents another sink.
Bioaccumulation of a toxic compound can have extremely adverse
consequences for an organism, ranging from ill health, to
failure to reproduce, to death. There are two mechanisms for
uptake, ingestion and absorption through the surface mem-
branes. Internally, a number of processes occur: the material
may be stored; the. material may be excreted; or the material
may be chemically altered, producing a product which again may
be stored, excreted, or further altered. Occasionally, the
secondary reaction products are more toxic than the compound
initially taken in.
Even if a pollutant is present in the ambient environ-
ment in concentrations below the minimal detection limits,
bioaccumulation can magnify its concentration by factors of
10^ or 1()4. For example, the ecological magnification of
chlorobenzene in a model aquatic ecosystem was 650. In
another study, the factor for tetrachlorobiphenyl was 17,840
(Branson, 1978). Polynuclear aromatic hydrocarbons, because
of their high partition coefficients, are expected to bio-
accumulate.
The next problem is assessing the hazard to an organism
as a result of compound accumulation. An additional variable
is introduced by unpredictable synergistic and antagonistic
interactions of pollutants.
Monitoring of the biological components of the aquatic
environment in a source-oriented ambient program is indicated
when:
A process analysis of the FBC facility indicates
the potential for release of compounds known to be
toxic or to bioaccumulate.
Species in the receiving water are known to be
unusually sensitive to an effluent from the facil-
ity.
• The type of pollutants discharged by an FBC facil-
ity differ significantly in speciation from other
discharges to that water.
199
-------
Biological monitoring is the best method to deter-
mine the lowest detectable concentrations allow-
able.
Surveys indicate that the biota of the receiving
water is changing in diversity or quality.
Pollutant containment systems fail, resulting in a
catastrophic discharge to a receiving body of
water followed by reestablishment of the biologi-
cal community.
In lieu of a specific need, synoptic surveillance is all
that would be required for a monitoring program. In such a
program, the presence and abundances of the major populations
in the ecosystem are determined on an overview basis. If
needed, target monitoring must be done: this might include
necropsy, histopathology and chemical analysis.
CONSIDERATIONS SPECIFIC TO SURFACE WATER MONITORING
Because different areas possess different hydrological
conditions, a good monitoring program must take into account
local variations. To do this, the hydrological charac-
teristics of the area or basin to be monitored must be deter-
mined. In this and the following section on groundwater
monitoring, general strategies for hydrological and geological
characterizations as they, relate to developing a good, compre-
hensive water quality monitoring program will be discussed.
Definition of the Watershed and the Hydrological System
The term "watershed" refers to a given geographical area
in which all the water flows to or through one point. It in-
cludes both the land and the bodies of water in the area.
Watersheds vary widely in their size and configuration, and a
number of small watersheds may comprise a larger one if all
water flows to a common point. To define a given watershed,
the common point of flow is identified and then, "working
upstream" on a topographic map, all contiguous areas from
which water will run off or flow downhill to that common point
are determined. The watershed is not equatable with the
hydrologic system. The latter also includes groundwater,
which may have a net inflow or outflow relationship with the
surface waters. Their geographical limits are seldom identi-
cal.
The goal of characterizing the hydrologic system is to
determine where the water is, how much there is, and where and
how it is moving through the system. In order to determine
this, the following information is needed:
200
-------
Precipitation—the amount, duration and intensity
of rain and snowfall. Seasonal distribution is
important.
Surface Waters—number, location, size, source and
destination. This information indicates where the
water is going in order to help identify possible
pollutant sinks.
Evaporation and Transpiration Levels—based on the
climate, area of open water, topography, volume,
and type and distribution of plant cover. Trans-
piration may be the most important component of
total evaporation. Vegetation surveys are re-
quired to determine this parameter.
Run-off—direction, magnitude, and destination of
run-off. This is a major contribution to stream
flow and major cause of erosion. This is espe-
cially important during the construction phase of
a project.
Infiltration—rate at which surface waters enter
the groundwater system. This is a function of
local soils and bedrock geology.
It is also important to ascertain all the administrative
jurisdictions in the watershed to identify which environmental
regulations may apply and who may already be monitoring water
quality within the watershed.
Transport of Pollutants in Surface Aquatic Systems
The ability of pollutants to be transported through the
surface aquatic systems is of great importance, especially
when evaluating the effects of a technology on the biological
components of the surface areas. Factors of importance are:
How fast will the pollutant travel through the
system? (What is the velocity of the hydrological
components receiving the pollutant?)
Where (and how far) will the pollutant be carried?
How concentrated will the pollutant be at some
point downstream. (What is the dilution
factor/capacity of the hydrological systems?)
What is the final receiving system for the pollu-
tant? Is it a particularly sensitive system?
201
-------
How will the pollutant disperse? This is espe-
cially important in standing bodies of water which
may not have an obvious direction of current flow.
How will the pollutant be chemically transformed
in ttye environment?
These factors evaluate the mobility of pollutants in a
hydrologic system and help to evaluate the impact of an FBC
facility on an area. A pollutant may move through the system
in question so rapidly that there will be no impact on that
system. This does not mean that a system downstream will not
be affected. The possible impacts on surrounding systems may
also have to be assessed.
CONSIDERATIONS SPECIFIC TO GROUNDWATER MONITORING
The following sections present a general strategy or ap-
proach for designing a groundwater monitoring program. Since
geologic and hydrogeologic conditions vary from site to site,
the strategy must be flexible to allow for local conditions
while at the same time providing a logical and consistent
design approach which can be applied to any and all potential
sites. Although many different types of monitoring have been
identified by the EPA (i.e., source, case preparation, re-
search, and ambient trend monitoring), the focus here is on
ambient trend monitoring, which characterizes existing condi-
tions and temporal and spatial trends in groundwater quality
relative to federal, state, and local standards (Todd et al.,
1976).
The following is a step-by-step approach to be used
in developing a groundwater quality monitoring program: (1)
define the area or basin to be monitored, (2) define the
physical/geological system of the basin, (3) define the hydro-
geologic system of the basin, (4) evaluate the ability of
wastes to infiltrate at the land surface, (5) evaluate the
ability of pollutants to percolate from the land surface to
the zone of saturation, and (6) evaluate the attenuation of
pollutants in the saturated zone (Todd et al., 1976). These
steps are discussed in more detail below.
The first step is to define the area or basin in term's
of hydrogeological and administrative and political frame-
works. Groundwater basins may, but usually do not, have the
same drainage boundaries as the overlying surface water sys-
tem. Federal, state, and local agencies can assist in de-
fining the basin. The exact lateral and vertical extent of
the basin cannot be identified with certainty until the second
and third steps have been completed. Administrative and
political frameworks should be identified early in the project
to determine (1) who may be monitoring groundwater in the
202
-------
basin, (2) what agency(ies) has (have) jurisdiction over the
basin, and (3) what permits or other legal requirements may be
necessary to implement the program. Early identification of
these items will prevent duplication of efforts and unneces-
sary expenditure of funds.
The second step is to define the physical system of the
basin which consists of the soils, vegetation, and unconsoli-
dated and bedrock deposits found on and underlying the basin.
These materials should be inventoried as to number and spatial
distribution. Particular emphasis should be placed on the
physical properties such as permeability, porosity, and thick-
ness which affect the ability of the materials to transmit
groundwater. Mineralogy and chemical characteristics of the
materials should be identified in order to determine potential
effects on water quality. For example, clay minerals have the
ability to remove ions from solution while other minerals,
e.g., sulfides, may go into solution to be deposited later as
sulfates.
Information on the physical system can be obtained from
federal and state EPA's and geological surveys, soil surveys,
and from universities. In the event that data are not already
available, it becomes necessary to generate data using field
techniques. Exploratory drill holes and soil samples and
seismic and earth resistivity surveys will usually provide the
information necessary to characterize the system. Spacing be-
tween drill holes, soil samples, and seismic and earth resis-
tivity surveys will vary with the geologic complexity of the
site.
The goal of this third step is to determine where the
water is going, how much of it is going, how it is going, and
when it is going. In order to achieve this goal the following
hydrogeologic data are required (Todd et al., 1976):
Aquifer characteristics—locations, depths, ex-
tent, transmissivities/transmissibilities
(source: geologic and pump test data)
Groundwater levels—elevations of water in wells
(source: well observations)
Depths to groundwater—interpretation of ground-
water level and topographic data (sources: well
observations and topographic data)
Locations and magnitudes of groundwater recharge
areas—areas where surface water infiltration con-
tributes to the groundwater system (source:
precipitation, evapotranspiration, soils, land
use, and water level data)
203
-------
Locations and magnitudes of groundwater discharge
areas—areas where groundwater contributes to the
surface water system (source: stream flow data,
water level data, and aerial photographs)
Directions and velocities of groundwater flows—
characterization of flow system (source: water
level and transmissivity data).
Groundwater flow system diagrams will indicate direc-
tions of flow at various depths and the piezometric surface of
the groundwater. Once the flow system(s) has(have) been de-
fined/the geometry of pollution plumes migraling in the zone
of saturation and the potential for pollution of other water
resources can be identified. Mapping of the flow system iden-
tifies groundwater recharge and discharge zones and whether
the flow enters the local or regional flow systems. The rela-
tionship of pollutant sources to recharge and discharge zones
will indicate the potential for pollution of other water re-
sources. For example, the release of pollutants in a dis-
charge zone will pollute surface waters but will not affect
groundwaters.
Evaluation of Ability of Pollutants to Percolate From the Land
Surface to the Zone of Saturation
The ability of pollutants to leach from coal piles,
landfills, and holding ponds is the chief factor creating the
need for a groundwater monitoring program. If the wastes can-
not infiltrate the land surface, they cannot pollute the
groundwater system.
The quantity of leachate from coal piles and landfills
can be computed using the water balance method. The quantity
of leachate (Qp) is equal to the precipitation (P) minus
runoff (R), evapotranspiration (ET), and the field capacity of
the material, excluding the existing moisture content (S):
Qp = P-R-ET-S (24)
The quantity of leachate leaving a coal pile or landfill
is determined by the ability of the underlying soils, uncon-
solidated deposits, and bedrock to transmit water. The quan-
tity of leachate can be determined by comparing the quantity
of leachate produced (Qp) and the maximum quantity of infil-
tration (Qi), determinea as the product of the materials
permeability (K) and the area of the pollutant source (A), if
Qp exceeds Qj. However, if Qj exceeds Qp, then the
quantity of leachate leaving the source TQL) equals Qp.
The ultimate quantity of leachate reaching the zone of satura-
tion is determined by the storage capacity requirements of the
material between the water table and the surface, the
204
-------
attenuation ability of the earth materials, and the leachate
control method employed on-site.
Evaluation of Attenuation of Pollutants in the Vadose Zone
As stated, the ultimate quantity of pollutants reaching
the zone of saturation is partially a function of the storage
capacity requirements of the materials between the water table
and the surface. In arid and semi-arid lands, pollutants may
be taken up into storage and effectively prevented from reach-
ing the groundwater table. However, the vadose zone, the area
between the zone of saturation and the surface, does not al-
ways remove pollutants. At times, the vadose zone is bypassed
by the discharge; in humid environments, infiltration is a
virtually continuous process.
A discussion of the ability of pollutants to percolate
through the vadose zone and reach the groundwater system must
focus on the ability of the materials in the vadose zone to
attenuate pollution. Pollution may be attenuated by the fol-
lowing processes; however, the attenuation must be evaluated
for each process for each pollutant. The attenuation pro-
cesses are (1) dilution, (2) filtration, (3) sorption, (4)
buffering, (5) chemical precipitation, (6) oxidation and re-
duction, (7) volatilization, (8) biological degradation and
assimilation and (9) radioactive decay (Todd et al., 1976).
Dilution occurs when water from other sources infil-
trates the land surface. Dilution in the vadose zone can be
significant in humid regions and non-existent in arid and
semi-arid regions. Dilution is computed in the same manner as
infiltration but all sources of infiltration are considered,
e.g., leakage from streams, lakes, and canals, and artificial
recharge.
Filtration removes particles with effective diameters
larger than the pore spaces of the earth material. Almost all
suspended solids are removed by this process. The filtration
process eventually prevents the passage of all water as the
pore spaces become blocked with solid material. Filtration is
essentially ineffective in removing elements and inorganic
compounds.
Sorption is the dominant attenuation process in
clay-rich earth materials. In sorption, electrically neutral
pollutants, such as most organics and some metals, are removed
from solution and bound to the clay minerals. However, the
sorbed materials may be taken back into solution by migrating
groundwater fluids. The ability of earth materials to sorb
inorganic substances is finite, but the sorptive capacity for
biodegradable substances is renewable.
205
-------
Buffering, the ability of the soil solution to resist pH
change, is directly linked to the sorption of bases; the
higher the amount of sorbed bases, the higher the pH. Buffer-
ing is usually not important if pH values range from six to
nine; however, for other pH values, buffering may affect the
ability of the waste solution to carry pollutants.
i
Chemical precipitation is important relative to calcium,
magnesium, bicarbonate, and sulfate, and these ions may
precipitate readily in the vadose zone. Other trace elements
and compounds which have significant precipitation potential
are arsenic, barium, cadmium, copper, fluoride, cyanide, iron,
lead, mercury, molybdenum, zinc, and radium.
Oxidation and reduction are important attenuation pro-
cesses in the top soil layer. Reducing conditions not only
cause chemical precipitation but also may cause formation of
native elements, including arsenic, copper, mercury, selenium,
silver, and lead.
Reactions in aerobic and anaerobic environments may
cause volatilization of certain elements and compounds, re-
sulting in their loss to the atmosphere. The elements and
compounds which may be affected are mercury, arsenic,
selenium, sulfates, and nitrates.
Biological degradation and assimilation are important
processes in the removal of organic and biologic materials.
Plant uptake effectively prevents pollutants from reaching the
zone of saturation; however, unless the plants are disposed
of, the pollutants eventually will be recycled to the soil and
groundwater systems. The following pollutants are most likely
to be affected by this process: sulfate, nitrate, cyanide,
mercury, selenium, molybdenum, phosphorus, and potassium.
The final attenuating process affecting pollutant
mobility is radioactive decay. Measured as the half-life of
the radionuclide, radioactive decay of many radionuclides is a
slow process. In most flow systems, its effects are not sig-
nificant as a means of pollutant attenuation.
Evaluation of Attenuation of Pollutants in the Zone of
Saturation
Generally, the same processes of attenuation occur in
the zone of saturation as occur in the vadose zone: (1) fil-
tration, (2) sorption, (3) radioactive decay, (4) dilution,
(5) oxidation and reduction, (6) buffering, (7) volatiliza-
tion, and (8) chemical precipitation. Filtration differs only
in that movement is horizontal rather than vertical; hence,
aquifer area is more important than thickness. There are
essentially no differences for sorption, radioactive decay,
206
-------
buffering, and chemical precipitation. Reduction dominates
oxidation in the zone of saturation, while oxidation dominates
in the vadose zone. Dilution is the primary process of pollu-
tant attenuation in the zone of saturation. The amount of
attenuation is determined by (1) the volume of waste reaching
the water table, (2) the transmissivity of the aquifer, (3)
the vertical and areal hydraulic head distributions, (4) the
existing groundwater quality, (5) the quantity and quality of
recharge, and (6) the quantity of groundwater withdrawals.
Evaluation of pollutant attenuation in the zone of saturation
requires considerable judgment by a skilled professional.
SAMPLING
General Considerations
Water quality data can be collected in two basic ways.
The first involves collecting water samples which are returned
to the laboratory for chemical analysis; the second involves
the use of automatic sensor equipment which measures the
presence and concentration of particular parameters. Design
of a monitoring plan should include an evaluation of the kinds
of data collection systems available and their general appli-
cability to the proposed plan. The general physical setting
of the facility being studied, parameters to be tested, inten-
sity of sampling and the type of information sought will all
be factors in determining what sampling system should be
used. In general, a comprehensive monitoring program will
probably include elements from both types of systems.
The major elements of any sampling routine include:
Number of sampling points
Location of sampling points
• Frequency of sampling
Volumes collected
Sample preservation
The importance of each element varies according to the
process involved, physical setting of the facility, sensi-
tivity of the receiving waters, goals of the program and the
analytical methods used.
Number and Location of Sampling Points—
This factor will be affected by the number of real and
potential effluent streams and discharge points associated
with the site. The effluent from a holding (settling) pond or
treatment plant would usually be sampled as part of a moni-
toring program. This will be the major water pollution poten-
tial of a well-engineered surface facility. However, all
207
-------
possible effluent streams should be identified. Area runoff
may carry "fugitive emissions" and such runoff, if not col-
lected into a holding pond, should be monitored.
Care should be taken to identify all aquifers that may
be contaminated during process testing and development, and
they should be mdnitored up-gradient and down-gradient of the
test site. For ambient monitoring, one or more sampling loca-
tions may be appropriate downstream of the facility with at
least one upstream sampling location.- If the hydrologic
characteristics of the system are adequately known, a disper-
sion model may be useful for identifying the best sampling
locations in terms of likelihood of detecting pollutants.
Representative sampling locations should be selected at
sites where the influent, waste, or receiving stream is well
mixed. This is at some point below the "mixing zone." The
mixing zone may be defined as the point where stream water
meets (contacts) water of a different quality, usually ef-
fluent wastes (thermal differences are included). The size of
mixing zones is controlled by the flow rate of the ef-
fluent(s). The EPA guidelines for developing water quality
standards state that (EPA, 1976c):
The total area and/or volume of a receiving stream as-
signed to mixing zones [should] be limited to that which
will: [1] not interfere with biological communities or
populations of important species to a degree which is
damaging to the ecosystem; [2] not diminish other bene-
ficial uses disproportionately.
The EPA makes no recommendations concerning effluent
concentrations within a mixing zone. However, in many states,
sampling stations are required within the mixing zone.
It is important that the site be accessible but pro-
tected from vandalism, weather extremes, or other damaging
factors. The flow of water past the sampling point should be
known, either by actual measurement or calculation.
Frequency of Sampling—
The frequency of sampling depends on the flow rate, •
waste characteristics, and the variability of both. The more
uniform the flow and waste composition, the less frequently is
sampling required. For a variable flow in which the waste
composition is flow-dependent with a known relationship, con-
tinuous flow measurement with infrequent checking of pollutant
levels may be adequate. However, the usual case involves com-
plicated variability in both flow and composition.
208
-------
The number of samples taken per unit of time may also
reflect the need for statistical analysis of the data. Gen-
erally, the more samples taken, the greater confidence ac-
corded to the statistical analyses. This may be limited, how-
ever, by the cost constraints placed on the monitoring pro-
gram. Methods for determining sampling frequency, based on
the needs for statistical representativeness of data, cover a
variety of conditions ranging from single variable-single to
multiple variable-multiple stations (Sanders, 1979). It is,
of course, essential that sampling frequency be consistent
with the type of data required, the methods by which it will
be analyzed and the uses that will be made of the data.
Manual grab samples taken at specified time intervals or
following certain waste-generating occurrences are the
simplest approach if flows and wastes are uniform or pre-
dictable. The approach becomes too manpower-intensive if
samples must be taken often (e.g., hourly) or unreliable if
inadequate consideration is given to flow or waste varia-
bilities. Automatic sampling is appropriate if flow and waste
conditions require frequent sampling. If flow is nearly con-
stant, a non-proportional sampler which obtains the aqueous
sample at continuous flow rate is adequate. If the flow is
variable, a flow-proportioning sampler at irregular time
intervals or a variable volume at regular time intervals is
needed. High variability of either flow or composition sug-
gests that an automatic sampler or continuous monitor be in-
stalled.
Sample Volumes—
Total volume of samples will be determined by the needs
of all the individual waste analyses. The required volume of
sample for several water constituents is indicated in Table
44. In general, 1 to 2 liters per sample is necessary at a
minimum. Individual portions of a composite sample should be
at least 25 to 100 ml.
If more than one laboratory is to analyze the samples
for comparative purposes, larger samples should be taken as
opposed to taking multiple samples at the same time, because
chemical composition may vary between two samples regardless
of location and time of sampling.
Sample Preservation—
Perfect or complete preservation of a water sample so as
to retain its chemical integrity indefinitely is a practical
impossibility. Complete stability of all chemical con-
stituents can never be achieved; for some parameters, their
integrity cannot be preserved. Certain chemical and bio-
209
-------
logical processes change the chemical structure and com-
position of a water sample as soon as it is removed from the
parent water source. Preservation is an attempt to retard
these chemical and biological processes long enough to return
the water sample to the laboratory and perform the appropriate
chemical analyses. Methods of preservation are relatively
limited, serving'to (a) retard biological action, (b) retard
hydrolysis of chemical compounds and complexes, and (c) reduce
volatility of constituents. Preservation methods are gen-
erally limited to pH control, chemical additions, and re-
frigeration (approx. 4°C); freezing (at 4°C or below) is
the best general preservation method. However, it is not
applicable to all types of samples and parameters and may not
be feasible if the sample site is far from the laboratory. A
summary of general preservation methods is given in Table 45.
Sampl ing Equ ipment.
There are two types of sampling equipment, manual and
automatic. The second type can be left in the field for
varying periods of time to collect one or several samples.
However, these samples must be transported to the lab for
analysis.
Manual Samplers—
Most manual water sampling bottles are simply open tubes
with movable plugs at both ends. The sampler is lowered into
the water to the desired depth and a "messenger" (metal
weight) is sent down the cable to activate a release which
closes both ends of the bottle. The sample is then brought to
the surface, treated as necessary, and stored.
There is a wide variety of sampling bottles used to
collect water samples. The choice of one type over another is
largely a matter of personal preference. In the absence of a
sample bottle, a bucket will suffice for collection of surface
water. In general, however, one of the samplers described
below is needed when sampling water at depths greater than 3
feet.
The types of manual samplers in common use include
Kemmerer, Van Dorn, Nansen, and Fjarlie Samplers. Each
sampler operates by slightly different means but will collect
similar samples. The principal difference is their adapta-
bility for taking multiple samples. Manual samplers are
primarily used when it is necessary to collect only a single
sample at a given point in time and space. If a series of
samples is desirable at a given point or if a composite sample
(over time) is required, the use of automatic samplers would
be indicated. A variety of automatic samplers are com-
mercially available which can fulfill the requirements of a
210
-------
TABLE 44. VOLUME OP SAMPLE REQUIRED FOR DETERMINATION OF THE VARIOUS CONSTITUENTS OF INDUSTRIAL
WATER
(EPA, 1973)
Volume of
Sample,8(ml)
Volume of
Sample.•(ml)
PHYSICAL TESTSb
Color and Odor 100 to 500
Corrosivity flowing sample
Electrical conductivity 100
pH, electrotnetrlc 100
Radioactivity 100 to 1000
Specific gravity 100
Temperature flowing sample
Toxicity 1000 to 20,000
Turbidity 100 to 1000
CHEMICAL TESTS
Dissolved Gasesic
Ammonia, NHj 500
Carbon dioxide, free COj 200
Chlorine, free C12 200
Hydrogen, H2 1000
Hydrogen tulfide, H2S 500
Oxygen, Oj 500 to 1000
Sulfur dioxide, free SOj 100
Miscellaneous!
Acidity and alkalinity 100
Bacteria, iron 500
Bacteria, sulfate-reducing 100
Biochemical oxygen demand 100 to 500
Carbon dioxide, total CO2
(including COj", HC03,
and free) 200
Chemical oxygen demand (dlchromate) 50 to 100
Chlorine requirement 2000 to 4000
Chlorine, total residual C12
' (including OC1", HOC1,
NH2C1, NHClj, and free) 200
Chloroform-extractable matter 1000
Detergent* 100 to 200
Miscellaneousi
Hardness 50 to loo
Hydrazine 50 to 100
Microorganisms 100 to 200
Volatile and filming amines 500 to 1000
Oily matter 3000 to 5000
Organic nitrogen 500 to 1000
Phenolic compounds BOO to 4000
pH, colorimetric 10 to 20
Polypnosphatea 100 to 200
Silica SO to 1000
Solids, dissolved 100 to 20,000
Solids, suspended SO to 1000
Tannin and lignin 100 to 200
Cationsi
Aluminum, M+++ 100 to 1000
Ammonium, NH4C 500
Antimony, Sb+++ to SD+++++ 100 to 1000
Arsenic, AS+++ to AS+++++ loo to 1000
Barium, Ba++ 100 to 1000
Cadmium, Cd++ 100 to 1000
Calcium, Ca++ 100 to '1000
Chromium, Cr+++ to Cr++++++ 100 to 1000
Copper, Cu++ 200 to 4000
Iron, FE++ and Fe+++c 100 to 1000
Lead, Pb++ 100 to 4000
Magnesium, Mg++ 100 to 1000
Manganese, Mn++ to Mn+-M-++*+ 100 to 1000
Mercury, Hg+ and Hg++ 100 to 1000
Potassium, K+ 100 to 1000
Nickel, N1++ 100 to 1000
Silver, Ag+ 100 to 1000
Sodium, Na+ 100 to 1000
Strontium, Sr++ 100 to 1000
Tin, Sn++ and Sn++*+ 100 to 1000
Zinc, Zn*+ 100 to 1000
Anlonsi
Bicarbonate, HC03
Bromide, Br~
Carbonate, C0j~~
Chloride, Cl"
Cyanlde, Cn"
Fluoride, PI"
Hydroxide, OH"
Iodide, I-
Nitrate, N03"
Nitrite, N02
Phosphate, ortho, PO^"",
HP04*", H2P04"
Sulfate, S04", HS04"
Sulfide, S~~, HS"
Sulfite, S03"~ , HS03"
100 to 200
100
100 to 200
25 to 100
25 to 100
200
50 to 100
100
10 to 100
50 to 100
50 to 100
100 to 1000
100 to 500
50 to 100
a Volumes specified in this table should be considered as a guide for the approximate quantity of
sample necessary for the particular analysis. The exact quantity used should be consistent with the
volume prescribed In the standard method of analysis, whenever the volume in specified.
b Aliquot may be used for other determinations.
0 Samples for unstable constituents must be obtained in separate containers, preserved as
prescribed, completely filled, and sealed against all exposure.
211
-------
TABLE 45. SUMMARY OF PRESERVATION METHODS FOR WATER SAMPLES9
(EPA, 1979a) ;
Preservative
Action
Applicability
Hgci2
Acid (HN03)
Acid (H2S04)
Alkali (NaOH)
Refrigeration
Bacterial inhibitor
Metals solvent, pre-
vents precipitation
Bacterial inhibitor
Salt formation with
organic bases
Salt formation with
volatile compounds
Bacterial inhibitor,
retards chemical
reaction rates
Nitrogen forms,
phosphorus forms
Metals
Organic samples
(COD, oil & grease
organic carbon),
nitrogen-phosphorus
forms
Ammonia, amines
Cyanides, organic
acids
Acidity-alkalinity,
organic metals,
BOD, color, odor,
organic P, organic
N, carbon, etc.,
biological organism
(coliform, etc.)
a For a more complete list of parameters and associated preser-
vation lists, see EPA, 1979a.
b Because of mercury residuals, HgCl2 poses problems in disposal
of the sample; research is underway to develop a preservative
to replace HgCl2-
212
-------
monitoring program. These samplers may collect samples over
discrete time intervals, or an event such as increased flow
volume either in an effluent or a stream may be used to
trigger or alter a sampling sequence.
A powerful option with automatic samplers is to combine
the sampler with an on-site pollutant monitor. A pollutant
associated with the effluent stream of the FBC facility could
be monitored continuously. If the pollutant concentration
exceeds a predetermined concentration (statistically derived),
the sampler would be triggered to collect a sample or series
of samples. The data derived from samples collected by such
means could be used to monitor pollution abatement equipment
efficiency and relate changes in environmental condition to
plant operating conditions.
Collecting samples for analysis of trace amounts of
organics such as benzo(a)pyrene or fluoranthene often requires
different techniques because of large volumes of sample re-
quired. These methods normally involve pumping the water to
be sampled through an absorbent which is selected to remove
the organic pollutant (s) from the water. Commonly used' sor-
bents are XAD resin and granular-activated carbon. The use of
these techiques and their application to Level I/Level 2
analysis has been summarized by Harris and Levins (1978).
Whether manual or automated methods are used to collect
samples, a number of parameters in the field should be moni-
tored at the time of sample collection. The parameters which
should be measured at time of sampling include: stream flow,
temperature, dissolved oxygen, and in the case of lake or
ocean sampling, the transparency/visibility.
Since many of the pollutants of interest may be absorbed
by particulate matter, sampling of sediment may be required to
determine the fate and rate of accumulation of pollutants.
The samplers employed may be of two types: 1) traps to col-
lect sediments deposited over a known time period, and 2) grab
samplers to collect a profile sample or composite sample.
Profile samples may be collected using a core sampler while
composite samples may be collected by an Ekman, Ponar or
Peterson dredge. The preferred methods of collecting samples
for monitoring pollutants discharged "in FBC effluent streams
are in-place sediment traps and core samplers. Core samplers
would be employed to collect samples for analysis of movement
of pollutants within the sediments and determination of zones
of concentration of pollutants.
Collection of aquatic organisms requires methods similar
to those used for sampling sediments. The methods include
in-place samplers (artificial substrates) and grab samplers.
Numerous types of samplers and methods of collection have been
213
-------
described for various types of organisms (American Public
Health Association, 1976; Swartz, 1978; Jacobs and Grant,
1978). For purposes of both substrate and statistical con-
tinuity over time, the use of artificial substrates is recom-
mended as the method to be employed for collection of aquatic
organisms where such methods are applicable.
Artificial substrates are, of course, applicable to sessile
organisms, but no't to fish or planktonic species.
ANALYTICAL METHODS
Analysis of water samples for collection of water qual-
ity data can be achieved in two basic ways. One way is to
collect water samples at the desired sampling locations,
either manually or with automated sampling apparatus, and
transport the samples to a laboratory for analysis. The
second way involves the use of on-site and analytical appara-
tus. In the second method, the analyses are performed in the
field automatically, without the aid of an analytical labora-
tory, except for necessary reagent preparation and required
quality assurance checks. Data so collected may be either
stored on-site for later retrieval, or transmitted directly to
a central data collection point.
Some parameters may be conveniently and accurately mea-
sured by on-site monitoring equipment. However, because of
the low concentration of many parameters and the presence of
interferences, extensive sample preparation is required. This
tends to rule out the use of on-site analytical equipment ex-
cept for a limited number of parameters. Even if the equip-
ment were available to adequately measure concentrations of
pollutants with the desired level of accuracy and precision,
the cost of the equipment would be prohibitive for more than a
very limited number of sampling points.
The level and type of monitoring system or sampling and
analytical techniques to be used should always be determined
during the developmental stages of the monitoring program,
prior to implementation. This will be very important in de-
termining the types of equipment needed, procedures to be
used, and the cost of the program.
Approved and Standard Methods
Methods for water quality analysis are varied and widely
published. New methods are also continuously being developed
and tested. The validity and acceptability of methods for
quantitative determination of various water constituents are
continuously being reviewed and updated by several organiza-
tions. Current editions of the primary reference works are
available (EPA, 1979a; American Public Health Association,
1976; American Society of Testing and Materials, 1979).
214
-------
The EPA has published a list establishing test proce-
dures for the analysis of pollutants as required by Section
304(g) of the Federal Water Pollution Control Act Amendments
of 1972 (Federal Register, December 1, 1976). These methods,
or approved alternate test procedures, must be used by NPDES
permit holders to ensure that effluents meet permit require-
ments.
Since FBC units and their accessory processes are likely
to produce aqueous discharges, the technology would ultimately
be regulated by the aforementioned regulations, and therefore
EPA-approved methods would be required for at least those
parameters for which approved methods are available.
The EPA has expressed concern that the analytical
methods used to measure water quality be reliable and repro-
ducible. They suggest the following set of criteria in se-
lecting physical and chemical analytical methods:
The method should measure the desired constituent
with precision and accuracy sufficient to meet the
data needs in the presence of the interferences
normally encountered in polluted waters.
The procedure should utilize the equipment and
scales normally available in the average water
pollution control laboratory.
The selected methods should be in use in many
laboratories or have been sufficiently tested to
establish their validity.
The method should be sufficiently rapid to permit
routine use for the examination of large numbers
of samples (EPA, 1979a).
EPA-approved analytical methods and additional instru-
mental techniques for non-metals and metals are given in
Tables 46 and 47. Table 48 summarizes the current preferred
analytical techniques for the metals. The lack of EPA ap-
proval does not reflect a lack of accuracy or precision; in
many cases, it indicates a lack of widespread availability of
the instrument, a need for highly skilled operators, or high
costs associated with operation and/or maintenance.
The selection of analytical techniques to be used in a
monitoring program must be based on the needs of the data
users and the appropriateness of the techniques available.
Table 49 presents a set of criteria for the selection of
analytical techniques that are weighted differently for quali-
tative and quantitative analyses. While the same instrumenta-
tion can often be used for both types of analyses, in the
215
-------
TABLE 46. STATE-OF-THE-ART WATER QUALITY MONITORING METHODS: NON-METALS
General Physical and
Chemical Parameters
EPA-Approved Methods8
Other Instrumental Techniques
Acidity
Alkalinity
Color
Conductivity
Hardness
PH
Odor
Residue, total
Residue, total dis-
solved (filterable)
Residue, total sus-
pended (nonfilter-
able
Residue, total
volatile
Settleable Matter
Salinity
Temperature
Turbidity
Nitrogen, Phosphorus
and Sulfur
Ammonia
Nitrate
Nitrate
Organic Nitrogen
Total Nitrogen
Phosphate
Total Phosphorus
Or thophospha te
Sulfate
Sulfide
Sulfite
Surfactants
Dissolved Gases
Chlorine
Hydrogen Sulfide
Oxygen
(Continued)
Titrimetric
Titrimetric; automated
colorimetric
Colorimetric, Spectro-
photometrie
Conductance bridge
Titrimetric; automated
colorimetric; (Ca + Mg)
No EPA method
Gravimetric 103-105°c
Glass fiber filtration,
180°C
Glass fiber filtration,
103-105°C
Gravimetric, 550°C
Volumetric or Gravimetric
Thermometer
Nephelometric
Manual or automated colori-
metric; titrimetric;
electrode
Manual or automated colori-
metric
Manual or automated colori-
metric
Total nitrogen minus ammonia
Digestion followed by ammonia
method
Manual or automated colori-
metric
Digestion followed by phos-
phate method; gas chroma-
tography
Manual or automatic ascorbic
acid reduction
Gravimetric; turbidimetric;
automated colorimetric
Titrimetric; colorimetric
Titrimetric
Colorimetric (Methylene
Blue)
Titrimetric; colorimetric
(none)
Titrimetric; electrode
Colorimetric
Electrochemical, colorimetric,
automated colorimetric
Ion selective electrode
Electrochemical
Conductance
Thermistor
Thermocouple
Turbidimeter
Activation analysis
UV
Activation analysis;
spectrophotometry
Activation analysis; ion selec-
tive electrode
Activation Analysis
Automated Colorimetric
Colorimetric
Activation Analysis
Automated Colorimetric
Colorimetric
Ion selective electrode
Colorimetric
Elec trochemical
Colorimetric; titrimetric
216
-------
TABLE 46 (Continued)
General Physical and
Chemical Parameters
EPA-Approved Methods'
Other Instrumental Techniques
Halides and Cyanide
Fluoride
Chloride
Bromide
Iodide
Cyanide
Organic Components
Total Organic Carbon
Petrochemicals
Phenols
Oil & Grease
Chlorinated Organic
Compounds
Ion electrode; manual or auto-
mated colorimetric
Titrimetric; automated coiori- Ion selective electrode
metric
Titrimetric
(none)
Titrimetric; colorimetric
Combustion-infrared analysis
(none)
Colorimetric
Gravimetric
Gas Chromatography
Colorimetric
Ion selective electrode;
colorimetric
Ion selective electrode
a Federal Register, December 1, 1976.
b-L'awrence-Berkeley Laboratory, 1976.
Gas fi thin layer chromatography;
UV spectrophotometry
Gas Chromatography
Gas Chrotnatography/Mass Spec-
trome try/Compu ter
Thin-Layer Chromatography
Gas Chromatography
Thin-Layer Chromatography
(Fluorescence)
UV Spectrophotometry
Automic absorption; gas &
thin layer Chromatography;
IR and UV spectrophometry;
sensors; reflectance
217
-------
TABLE 47. STATE-OF-THE-ART WATER QUALITY MONITORING METHODS:
METALS
Metals
Aluminum
Antimony
Arsenic
Barium
Beryllium
Boron
Cadmium
Calcium
Chromium
Cobalt
Copper
Gold
Iridium
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Osmium
Palladium
Platinum
Potassium
Rhodium
Ruthenium
Selenium
Silica
Silver
Sodium
Strontium
Thallium
Tin
Titanium
Vanadium
Zinc
11
o S
il
A a<
° 5
1 8,
3 w
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
E
. E
E
E
E
E
E
B
E
E
B
E
E
E
E
E
l|
11
0 *J
B
B
B
B
B
B
B
B
B
B
B
B
B
E
B
B
B
E
B
B
B
B
8
O -P
HI
ll&
O Q
0 C -H
•H 4J TJ
go O
6 y «H
SS1 8
E
E
E
E
B E
E
B E
B E
E
B
E
B
E
E
E
E
B
E
E
8
11 is
w w '« B
o. 3 e 5
B
G
B G
B
B G
B
B G
B
B
B
B B
G
B
B
B
B
B
escence 1
i-i
CM
1
B
B
B
B
B
B
B
B
B
B
B
B
B
1
1
§
I
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
KEYi E - EPA-approved guidelines
(Federal Register, December, 1976).
B - Additional instrumental techniques
(Lawrence-Berkeley Laboratory, 1976).
G - Additional techniques such as "unproven sensors"
(U.S. Geological Survey, 1976).
N - Trace metals detectable by neutron activation
(Robertson and Carpenter, 1974).
218
-------
TABLE 48. ATOMIC ABSORPTION CONCENTRATION RANGES3
(EPA, 1979a)
Metal
Aluminum
Antimony
Arsenic0
Barium (p)
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Gold
Iridium(p)
Iron
Lead
Magnesium
Manganese
Mercury*3
Molybdenum (p)
Nickel (p)
Osmium
Palladium (p)
Platinum (p)
Potassium
Rhenium (p)
Rhodium (p)
Ruthenium
Selenium0
Silver
Sodium
Thallium
Tin
Titanium (p)
Vanadium (p)
Zinc
Direct
Detection
limit
mg/1
0.1
0.2 .
0.002
0.1
0.005
0.005
0.01
0.05
0.05
0.02
0.1
3
0.03
0.1
0.001
0.01
0.0002
0.1
0.04
0.3
0.1
0.2
0.01
5
0.05
0.2
0.002
0.01
0.002
0.1
0.8'
0.4
0.2
0.005
aspiration
Furnace procedure13
Optimum Optimum
concentration Detection concentration
Sensitivity range limit range
mg/1 mg/1 ug/1 ug/1
1
• 0.5
-
0.4
0.025
0.025
0.08
0.25
0.2
0.1
0.25
8
0.12
0.5
0.007
0.05
-
0.4
0.15
1
0.25
2
0.04
15
0.3
0.5
-
0.06
0.015
0.5
4
2
0.8
0.02
5
1
0.002 -
1 ••
0.05 -
0.05 -
0.2 -
0.5 -
0.5
0.2 -
0.5
20
0.3
1
0.02 -
0.1
0.0002-
1
0.3
2
0.5 -
5
0.1
50
1
1
0.0002-
0.1 -
0.03 -
1
10
5
2
0.05 -
50
40
0.02
20
2
2
7
10
5
5
20
500
5
20
0.5
3
0.01
40
5
100
15
75
2
1000
30
50
0.02
4
1
20
300
100
100
1
3
3
1
2
0.2
0.1
-
1
1
1
1
30
1
1
-
0.2
—
1
1
20
5
20
—
200
5
20
2
0.2
-
1
5
10
4
0.05
20
20
5
10
1
0.5 -
-
5
5
5
5
100
5
5
-
1
—
3 •
5
50
20
100
—
500
20
100
5
1
-
5
20
50
10
0.2 - .
200
300
100
200
30
10
-
100
100
100
100
1500
100
100
-
30
—
60
100
500
400
2000
—
5000
400
2000
100
25
-
100
300
500
200
4
a The concentrations shown are not- contrived values and should be obtainable
with any satisfactory atomic absorption spectrophotometer.
b For furnace sensitivity values, consult instrument operating manual; values
listed are those expected when using a 20 ul injection and normal gas flow
except in the case of arsenic and selenium where gas interrupt is used. The
symbol (p) indicates the use of pyrolytic graphite with the furnace procedure,
c Gaseous hydride method.
d cold vapor technique.
219
-------
TABLE 49.
CRITERIA FOR SELECTION OF ANALYTICAL METHODS
(Fennelly et al., 1976)
Requirement
Definition
Relative importance
in qualitative analysis
Relative importance
in quantitative analysis
ro
to
o
Sensitivity Amount of material that
gives a specified response.
Specificity Ability to measure only the
substance of interest by
eliminating interferences.
Precision Error of the method (often
expressed as the coeffici-
ent of variation) estab-
lished by the analysis of
many samples containing
equivalent amounts of the
species of interest.
Accuracy Ability to determine the
true value.
Range of Concentration range in
analysis which reliable results can
be obtained.
Stability Ability of the materials
to remain intact over a
period of time.
Ease of Skills required to prepare
handling the sample and execute the
analytical methods.
Response Time required to analyze
time one sample completely.
High - Being able to "see" any
substance on first analysis
can save considerable time in
later testing.
Low - Detecting even a "hint"
of a species is important in
screening. Later tests can
provide confirmation.
High
make
a test useless.
Low — An order of magnitude
accuracy would be adequate.
High - Ideally, a range which
spans 10~9 to io~3 g-molc/1
is desirable.
High ~ Consistency is impor-
tant in screening tests;
instability can also affect
precision.
Moderate - Convenience and
speed are important in screen-
ing tests. Methods requiring
advanced skills tend to re-
duce precision.
Moderate - Again convenience
is high priority, but low
response time could be traded
off against ability to measure
many different species.
High - Must be able to "see" at
least down to lowest level of
concern.
High - Interference can cause
erroneous data interpretation.
High - Lack of precision can
make a test useless.
High - Should be good to within
100% at parts per billion; 50%
at parts per million.
Low - Screening tests should
allow one to preselect optimum
range of interest; hence flexi-
bility here is not important.
Moderate - One will often have
option of selecting optimum
time for analysis; hence prob-
lems of instability may be
avoided.
Moderate - The importance of
the final result should dictate
the level of effort.
Moderate - Same as in guali-
tative.
Cost
Availability
of equipment
Actual monetary expenditure
for materials, equipment,
and personnel needed
Ability to purchase the
required equipment and
materials needed for an
analysis.
Examination of numerous
samples requires low cost per
sample.
High - When many tests are to
be performed, one needs easily
available supplies.
Cost can vary with the impor-
tance of the result.
Moderate - Most important
samples~can receive special
Priority
-------
former, sensitivity, speed, and ease of performance are most
important, while the latter requires precision, accuracy and
specificity. The criteria provide guidance concerning the
relative importance of each of these factors for both qualita-
tive and quantitative studies.
Since publication of the list of approved methods in the
Federal Register (December, 1976), a new edition of "Methods
for Chemical Analysis of Water and Wastes" has been pub-
lished. Some changes in approved analytical methods were
made. Atomic absorption is no longer the preferred method for
boron, and analysis of rhenium was added as an atomic absorp-
tion method.
Trace Organics
As indicated in Table 46, gas chromatography is the
preferred method for chlorinated organic compounds and for
other trace organic compounds as well. The analysis of
organic compounds and development of methods of analysis are
under the direction of the Environmental Monitoring and Sup-
port Laboratory in Cincinnati with the cooperation of the En-
vironmental Research Laboratory, Athens, Georgia. Gas
chromatography-mass spectrometry (GS-MS) methods of analysis
have been developed for priority pollutants and for numerous
other organics. A partial listing of the organics for which
GC-MS methods have been developed is indicated in Table 50.
GC-MS methods are preferred when the number of organics to be
monitored is more than four or five, or when the actual types
of compounds are uncertain. Routine gas chromatography proce-
dures could be used; however, identifying the compounds de-
tected would require additional analytical steps. The addi-
tional steps are repetitive analyses of the sample(s) under
varying analytical conditions such as different columns, tem-
perature or carrier gas flow rate.
The other major analytical method for measurement of
small quantities of organics is high performance liquid
chromatography (HPLC). This technique has recently been de-
veloped to measure a wide variety of organic compounds, in-
cluding many of those found in solid wastes and liquid ef-
fluents that may be byproducts of FBC technology.
Biological Methods
The biological methods useful in providing information
for assessment of environmental impacts of FBC effluents may
include any of several methods depending on the nature and
quality of a receiving water, the objectives of the program
and the funds available. The basic types of biological assay
methods are:
221
-------
TABLE 50. ORGANICS FOR WHICH DISCRETE ANALYTICAL METHODS
(GC-MS) HAVE BEEN DEVELOPED
(EPA, 1977b)
1, 3-dichlorobenzene
1, 4-dichlorobenzene
hexachloroethane
1,2-dichlorobenzene
bis (2-chloroisopropyl)
ether
hexachlorobutadiene
1,2,4 trichlorobenzene
naphthalenea
bis (2-chloroethyl)' ether
hexachlorocyclopentadiene
nitrobenzene
bis (2-chloroethoxy) methane
2-chloronaphthalene
ac enaphthy1ene
acenaphthene
isophorone
fluorene
2,6-dinitrotoluene
1,2-diphenylhydrazine
2,4-dinitrotoluene
N-nitrosodiphenylamine
hexachlorobenzene
4-bromophenyl phenyl ether
phenanthrenea
anthracene3
dimethylphthalate
diethylphthalate
fluoranthenea
pyrenea
di-n-butylphthalate
benzidine
butyl benzylphthalate
2-chlorophenol
phenol
2, 4-dichlorophenol
2-nitrophenol
p-chloro-m-cresol
2,4,6-trichlorophenol
2,4-dinitrophenol
2,4-dimethyIphenol
4,6-dinitro-o-cresol
4-nitrophenol
pentachlorophenol
chrysenea
bis(2-ethyIhexyl)phthalate
benzo(a)anthracene3
benzo(b)fluoranthene3
benzo(k)fluoranthene3
benzo(a)pyrenea
indeno(1,2,3-cd)pyrenea
dibenzo(a,h)anthracene3
benzo(g,h,i)perylenea
N-nitrosodimethylamine
N-nitrosodi-n-propylamine
4-chloro-phenyl phenyl ether
endrin aldehyde
3,3'-dichlorobenzidine
2,3,7,8-tetrachlorodibenzo-
p-dioxin
bis(chloromethyl)ether
deuterated anthracene (dlO)
alndicates organics measured for or detected in FBC effluent
streams.
222
-------
Determination of biomass
Identification and enumeration of organisms col-
lected from the aquatic system
Determination of toxicity of effluents to test
organisms
Determination of mutagenicity and cytotoxicity
Biomass determinations are useful in measuring growth
rates of aquatic organisms in relation to responses induced by
pollutants discharged from a particular source. Methods for
measuring biomass have been summarized by Weber (1973) and
Cairns and Dickson (1973). Generally, when biomass determina-
tions are made, identification of at least the dominant types
of organisms present are also done. The conventional method
of measuring biomass is to weigh the collected organisms, and
express the mass as dry weight, or ash-free dry weight per
unit area or volume of the environment from which the sample
was taken. Instrumental methods applicable to measurement of
biomass include determination of chlorophyll and adenosine
triphosphate.
Identification and enumeration of organisms and analysis
of the resultant data by any of several biotic or species
diversity indices should be specified. Methods of collection,
and identification of organisms, and use and interpretation of
indices have been documented (Dalton-Dalton-Newport,
1978). Because of the wide variety of collection methods,
organism identification aids, and methods of interpretation of
the data, the EPA should recommend acceptable complementary
methods. The recommended methods could then be employed for
analysis of impacts of FBC effluents upon surface waters for
those pilot and demonstration-scale plants where such methods
would be applicable.
The methods for testing the toxicity of effluents to
aquatic organisms and determining mutagenicity and cyto-
toxicity have been adequately described (Cairns, et al.,
1976; Duke et al., 1977; Peltier, 1978; Dalton'Oalton-
Newport, 1978) and are not repeated here. The EPA should
vigorously apply these methods to testing of effluents from
FBC units, particularly the long-term tests. It would also be
desirable to compare test results of FBC effluents with
results of tests of conventional coal combustion effluents to
obtain a measure of the relative hazard of each technology.
Level I/Level 2 Analysis
Investigations are currently underway to determine the
exact nature and composition of the effluents of the FBC
223
-------
process, how they might affect the environment or react to
environmental conditions, and the fate of the individual pol-
lutants or classes of pollutants in the environment. The
methods used to identify and quantify the pollutants in the
effluent streams of the FBC technology are based on the phased
approach to sampling and analytical strategy (Dorsey et al.r
1977). The phased approach has three levels of effort for
determining what parameters should be monitored for the FBC
technology (Harris and Levins, 1978):
Level 1 Assessment
provides preliminary environmental assess-
ment data
identifies principal problem areas
provides the data needed for priority rank-
ing of energy and industrial processes, com-
ponents within a waste stream, and classes
of materials for further consideration in
the overall assessment
• Level 2 Assessment
provides additional information that will
confirm and expand the information gathered
in Level 1 assessment
defines control technology needs and in some
cases provides the probable or exact cause
of a given problem.
Level 3 Assessment
uses sampling and analytical methods whose
precision and accuracy are sufficient to
permit quantitative monitoring of specific
pollutants identified in Level 2 assessment
- may be used in the design and development of
suitable pollution control devices
All three assessment levels are appropriate for applica-
tion to commercial-scale operation of an FBC coal boiler.
Whether or not a particular level of assessment is used de-
pends on the amount of information available concerning the
energy technology.
Specific methodologies have been described for Level 1
and Level 2 analysis (Lentzen et al., 1978; Ryan et al., 1979;
Harris et al., 1979). Thus far, the major emphasis in utili-
zation of the methods has been directed toward their full de-
velopment. When the methods have been fully developed, they
can be applied to pilot and demonstration-scale FBC units to
determine scale-up factors for effluents and pollutants from
commercial-scale FBC units. The limitations on use and degree
of uncertainty of the Level I/Level 2 data must, of course,
also be determined.
224
-------
The quality assurance requirements for ambient moni-
toring programs are outlined in Appendix A. The guidelines
are also applicable to other types of monitoring such as ef-
fluent monitoring and the type of monitoring performed using
Level I/Level 2 methodologies. A specific guideline (Smith,
1978) has been developed for Level 1 analysis which specifies
the degree of quality assurance required.
Data Recording
Data recording is basically the transfer of new data
generated during analysis into a form which will allow its
future use. The data obtained from meter dials, vernier,
digital displays or other devices may be recorded manually or
obtained from automatic data loggers or from computer inter-
faces.
In general, two steps are involved in data recording.
The first involves transfer of new data to either laboratory
or field notebooks or to raw data forms. Some analytical
equipment such as automatic field monitors and automated
laboratory equipment may feed data directly to a computer for
both storage and performance of additional calculations such
as statistics for the data and eventual printout of data sum-
maries. In this step all operational parameters, methods,
deviation and results are recorded, and calculations per-
formed.
The second step involves transfer of the data generated
during the laboratory or field analysis into a format which is
useful during data analysis. The data from Level I/Level 2
analysis may be transferred to the forms developed for SAM 1A
(Schalit and Wolfe, 1978) or to a similar format. Ambient
monitoring data may be transferred to forms similar to the
305b Reports, currently an EPA requirement for state moni-
toring networks.
DATA HANDLING
The chief purpose of prescribing data handling proce-
dures is to assure that data reach the appropriate personnel.
The data may be used for interpretation of plant operating
conditions, to determine impacts upon water quality, to gen-
erate information useful for designing wastewater treatment
facilities, or for enforcement activities. However, data can-
not serve any useful function unless they reach the appro-
priate personnel.
Data may be obtained from three basic sources in an am-
bient monitoring program: 1) samples taken to a laboratory
225
-------
for analysis, 2) automatic field analyzers, or 3) sources out-
side the program such as USGS monitoring programs or
state/local ambient monitoring programs. Whatever their
source, the data must be either initially put into a format or
transcribed into a format consistent with thos;e required by
data processing and data analysis procedures. The format will
normally be a preprinted form which contains spaces for all
information pertinent to the data or their interpretation.
Figure 34 describes the flow of data in an ambient monitoring
program. At each step in the data flow process where infor-
mation is produced, collated or changed in form, a quality
assurance check must be performed to assure that no errors are
introduced and that any errors are within acceptable limits.
Whenever data are transmitted via telemetry or other
devices, priorities must be established. The considerations
include the critical need for the data in a proper time frame,
and the ability of the various transmission systems to ac-
complish the task with accurate understanding of the time and
cost requirements. The quality assurance of various trans-
mission systems must also be evaluated, and the ability to in-
corporate feedback mechanisms is of prime importance.
The ambient monitoring programs operated in conjunction
with development and implementation of commercial-scale FBC
facilities are likely to be financed either directly or in-
directly by the EPA. The data produced by ambient monitoring
programs should therefore be available on call in acceptable
format. The STORET system should be employed to receive in-
formation produced in the monitoring- programs. It may also be
desirable to store pertinent data on in-house computer banks
at the various EPA facilities which are engaged in coordina-
ting energy/environment research activities. This would allow
personnel responsible for decision making to have immediate
access to information required for adequate and responsible
decisions.
Other data banks that can be used to store data gen-
erated during the ambient monitoring programs and to obtain
useful information include the National Water Data Exchange
(NAWDEX) which is operated by the U.S. Geological Survey.
Table 51 summarizes the data systems which participate in
NAWDEX.
Data dissemination means that data are reaching the
personnel who are responsible for utilizing the data whether
for research purposes, enforcement activities or preparation
of special reports. Each level of operation in the ambient
monitoring program must be supplied with a list of individuals
and organizations who are to receive information with the type
and format of that information specified. Data handling and
226
-------
COLLECTION ~~|
STATIONS |
I I
Figure 34. Data flow in a source-oriented ambient monitoring program.
(NUS Corporation, 1970)
-------
TABLE 51. NAWDEX PARTICIPATING DATA SYSTEMS
(Edwards, 1978)
Name of System
Organization
Location
Type of System
NJ
to
oo
National Water Data Storage and Retrieval
System (WATSTORE)
Storage and Retrieval System (STORET)
Water Resources Scientific Information
System (WRSIC)
Environmental Data Index (ENDEX)
Oceanic and Atmospheric Scientific
Information System (OASIS)
National Climatic Center (NCC)
National Oceanographic Data Center
(NODC)
Environmental Science Information Center
(ESIC)
Center for Climatic and Environmental
Assessment (CCEA)
Center for Experiment Design and Data
Analysis (CEDDA)
National Geophysical and Solar
Terrestrial Data Center
U.S. Geological Survey
U.S. Environmental Pro-
tection Agency
Office of Water Research
and Technology, U.S. Dept.
of Interior
Environmental Data and
Information Service, NOAA
Environmental Data and
Information Service, NOAA
Environmental Data and
Information Service, NOAA
Environmental Data and
Information Service, NOAA
Environmental Data and
Information Service, NOAA
Environmental Data and
Information Service, NOAA
Environmental Data and
Information Service, NOAA
Environmental Data and .
Information Service, NOAA
Reston, VA
Washington, DC
Washington, DC
Washington, DC
Washington, DC
Asheville, NC
Washington, DC
Washington, DC
Columbia, MO
Washington, DC
Boulder, CO
Water data
Water quality data
Water-related biblio-
graphic references
Index of environmental
data files
Environmental biblio-
graphic references
Meteorological data
Oceanographic data
Environmental data
Climatic impacts on
the environment
Oceanographic and
atmospheric data
Seismology, marine
geology and geophysics,
geomagnetism, solar
activity, interplane-
tary phenomena
(Continued)
-------
TABLE 51 (Continued)
Name of System
Organization
Location
Type of System
Water Resources Document Reference Centre
(WATDOC)
Texas Natural Resources Information
System (TNRIS)
Inland Waters Directorate,
Canadian Department of
Fisheries and the Environ-
ment
Texas Natural Resources
Information System
Iowa Water.Resources Data System (IWARDS) Iowa Geological Survey
to
N)
Nebraska Natural Resources Information
System
Water Resources Data System
REAP Resources Reference System
Colorado Water Data Bank (CWDB)
Snow Telemetry System (SNOTEL)
Water Supply Data Base
Great Lakes Regional Information
Referral Center
Nebraska Natural
Resources Commission
Pennsylvania Department
of Environmental
Resources
North Dakota Regional
Environmental Assessment
Program
Colorado Division of Water
Resources
Soil Conservation Service
Electric Power Research
Institute
Great Lakes Basin
Commission
Ottawa, Canada
Austin, TX
Iowa City, IA
Lincoln, NE
Bismarck, ND
Denver, CO
Portland, OR
Palo Alto, CA
Ann Arbor, MI
Water-related biblio-
graphic references
Water and environmental
data
Water data
Water and environmental
data
Harrisburg, PA Water data
Water and environmental
data
Water data
Hydrometeorological data
Index of water data
systems
Water resources of the
Great Lakes
-------
dissemination systems are the final step in placing data into
a proper framework for use.
DATA ANALYSIS
Data analysis techniques are used to derive information
from data acquired in an ambient monitoring program, e.g.,
Screen basic data to set priorities for pollutant
or parameter selection.
• Provide information concerning the quality of the
data produced.
Determine variables important in determining
changes in water quality.
Relate changes in water quality to the source of
the pollutants and changes in concentration of the
source.
Analysis techniques may range from simple measures of
data quality such as standard direction to complex water qual-
ity models. The methods available include:
• Regression Analysis
• Time-Trend Analysis
Factor Analysis
• Water Quality Indices
Chemical
Biological
Water Quality Model
Regression Analysis
Using regression analysis, a single variable of interest
can be expressed as a function of the concentration of one or
more different (independent) variables (pollutants). For
example, if the concentration of a pollutant such as
benzo(a)pyrene were shown to be related to the concentration
of a more easily measured parameter such as nitrite or sul-
fite, by measuring the concentration of nitrite or sulfite,
the concentration (within certain statistical limits) of
benzo(a)pyrene can be calculated. If predictive functions can
be developed relating dependent variables such as benzo(a)
pyrene to one or more easily independent variables, the need
to measure dependent variables decreases and the costs of
230
-------
monitoring may also decrease. As long as the functional rela-
tionship of variables is valid, and the errors of the esti-
mates of the dependent variables made in this manner are
within acceptable limits, only independent variables need to
be monitored (Ward, 1979).
While the form of the analysis should be kept simple,
prediction capabilities can be improved by adding more inde-
pendent variables or changing independent variables to produce
a usable water quality model. With more variables, the model
may represent a set of conditions unique to the data set used
for the determination.
Time Trend Analysis
Water quality, the pollutants present in the water, and
the pollutants that can be in the water vary with diurnal,
seasonal, or yearly fluctuations. Long-term changes in water
quality must be distinguished from short-term changes. This
requires consistent sampling and methods over time for all
variables. It has been demonstrated that trend analyses are
better performed on harmonic-analysis coefficients rather than
on statistical measures based on the data themselves (Steele
and Dyar, 1974). While this method may not be directly appli-
cable to source-oriented ambient monitoring, it can be useful
in determining the state of water quality during the period of
obtaining background data. Once the state of water quality
and variation therein are determined, changes in water quality
induced by a particular source of pollutants can be more
easily determined.-
Factor and Principal Component Analysis
The general goal of factor analysis is the reduction of
a set of basic data variables to a smaller set of new un-
correlated variables which are defined solely in terms of the
original data, and retain the most important information con-
tained in the original data (Ward, 1979). These analytical
procedures (described by Veldman, 1967) can be used to iden-
tify means of achieving cost-effective monitoring networks by
reduction of some water quality variables.
Principal-component analysis is one of a number of fac-
tor extraction methods. The factor-loading matrix obtained
from a principal-component analysis reflects the charac-
teristics of the extraction procedure, which maximizes the
variance in each succeeding column of the matrix. The first
principal component is easily interpreted and requests the
central focus of a set of original variables (Ward, 1979). In
many water quality studies utilizing factor analysis and
principal component analysis (Daudy and Feth, 1967; Symader
and Thomas, 1978; Hermann et al., 1979), explanations for
231
-------
observed component loadings were developed and sets of
variables were identified that tended to behave together in
terms of spatial or temporal variability (Ward, 1979).
Water Quality Indices
i
Water quality indices are of two basic types. Those
that aggregate a set of different chemical parameters to ar-
rive at a single number which describes the water quality at a
specific point in time and space are chemical water quality
indices. Biological water quality indices aggregate biologi-
cal data (e.g., numbers of benthic macroinvertebrates).
Many chemical water quality indices and their uses have
been described (Ott, 1978) . The number of parameters that
have been incorporated into a chemical water quality index has
ranged as high as 31 (Stoner, 1976) , but the normal range is 8
to 14. The parameters most frequently incorporated in indices
are those which are associated with organic pollution,
nutrients, and effects of high levels of nutrients, although
the same parameters may be equally associated with natural
nutrient loading conditions.
Water quality indices are used primarily to derive quan-
titative measures that can be understood and used by laymen
and water resource planners (Landwehr, 1974). In many
indices, the individual parameters are weighted with regard to
their relative importance in determining water quality, and
each parameter has a unique mathematical relationship to water
quality. The weighting system and the relationship to water
quality may be derived from subjective or objective analysis,
with the latter employing the statistical methods described
earlier.
Increases in many pollutants (individually or in com-
bination from a particular source) will alter the stability of
the ecosystem (including its chemical stability), causing
changes in the biotic system and alterations in the chemical
system. These changes in chemical structures may be monitored
by an "indicator parameter" and mathematically presented in
the form of an index value. When a specific index is used,
both the parameters employed and the sampling and analytical
methods must remain consistent throughout the monitoring pro-
gram, unless it can be demonstrated that other methods produce
the same data.
Aquatic organisms tend to integrate all the stresses
placed on the aquatic system and reflect the combined effects
over extended periods of time (Landwehr, 1974). Using this
principle, "biotic indices" have been derived to produce a
single.number which represents the water quality at a given
location and point in time. The indices in common use are of
232
-------
two general types. One type segregates aquatic organisms into
class according to their sensitivity to "pollution" and de-
rives a ratio based on the number of organisms in each class
to the total number of organisms (Scott, 1969; Chutter, 1972;
Brinkhurst et al., 1968). A second common type of index,
based on information theory (Pielou, 1966; Wilhm, 1967; Wilhm,
1970), includes diversity and ordination indices. Rather than
being based on information denoted by a single class of or-
ganism, these indices are based on information contained in
the entire aquatic community or some segment thereof. Gen-
erally, both indices utilize benthic macroinvertebrates as the
indicator group of organisms, although fish, algae and other
types of organisms have been employed.
Biotic indices may be very useful in determining
long-term environmental impacts of particular pollutants upon
the aquatic ecosystem. However, since the organisms will re-
spond to the sum total of pollutants present, the specific
usefulness of these methods will be reduced in direct correla-
tion to the number of sources of pollutants.
Water Quality Models
Three types of water quality simulation models have been
employed by various investigations. These are:
Steady State - Non-Dispersive
Steady State - Dispersive
Time Variant
Most of the applications of these models involve waste
load allocation of major pollutants, and are not directly ap-
plicable to determining effects of highly toxic micropollu-
tants. They are useful, however, in determining the potential
distribution of pollutants in the aquatic environment, parti-
cularly if the PBC facility discharges to a lake or ocean.
The use of such models may be necessary in large bodies of
water to determine the distribution of pollutants, and the
most desirable or cost-effective locations for sampling.
233
-------
234
-------
SECTION 6
CONCLUSIONS AND RECOMMENDATIONS
CONCLUSIONS
Based on currently available information,. FBC of coal is
a more efficient and less environmentally damaging way of ex-
tractinq coal energy than conventional pulverized coal combus-
tion. As an energy technology, FBC is a subcategory of coal
combustion processes, and therefore, is subject to the same
technical criteria and standards applicable to coal-fired
facilities of similar size and application.
Although a substantial data base has been developed on
FBC processes, most of the environmental data have been gen-
erated from bench scale and pilot plant units. The data have
significant gaps on the chemical Profiles of effluents gen-
erated during different operating conditions. Most of the
database on effluent characterization was generated from re-
seat project" ?Se extrapolation of such data to commercial-
scale FBC units cannot be done with any accuracy. In this
study we have treated the research data as "worst case" for
any Lc unit" without introducing ^Jl^'^^Tts "t
calculating emission rates and quantities of f ;4"^s. At
this time this approach is the most acceptable alternative.
not current B«ort. „
o co-ercial-scale facll-
formation system.
RECOMMENDATIONS
Recommendations for additional research and future moni-
toring of FBC units are:
I/Level 2 analytical methodologies should be
«J to clants larger than the bench scale units
hfve^been used tl demonstrate the phased
235
-------
approach techniques, including units exceeding 200
megawatts in power output.
Sulfur compound speciation testing in Level I/Level
2 analysis should be conducted on all atmospheric
and fugitive emission streams.
Level 1 biological testing procedures should be
utilized on liquid, solid and gaseous effluents
(Lentzen et al.r 1978).
Effluent streams that are generated during start-up
and shutdown cycles of both the Atmospheric and
Pressurized FBC unit should be tested. This effort
should be directed towards units larger than the
bench scale units that have been used to generate
the currently available Level I/Level 2 information.
Effluent streams should be monitored, particularly
bed reject material, flue gas particulates, and
gaseous emissions during combustor unit upset condi-
tions. The latter condition should include:
Substantial changes in flow rate of air to the
combustor
- Changes in feed rate of coal and sorbent
- Both increases and decreases in combustor
operating temperature
The upset conditions should be artificially gen-
erated on combustor units initially operating at
steady state conditions.
Bed reject material and other FBC solid wastes gen-
erated during normal and upset operating conditions
should be tested for their ability to release toxic or
other environmentally injurious substances when dis-
posed of in landfills individually, when mixed, and
in combination with solid wastes from other
sources. Suggested types of solid wastes which
should be tested with FBC wastes include:
- Flue gas desulfurization wastes
Bottom ash from conventional coal-fired corn-
bus tors
Municipal and commercial incinerator wastes •
Sludge and other solid wastes from municipal
and other wastewater treatment plants
Other types of solid wastes listed as "special
wastes" under proposed RCRA regulations
Other solid wastes with which FBC wastes might
not be compatible, but which could be disposed
of in the same landfill
236
-------
Plume transformation studies should be conducted on
large PBC facilities such as the proposed TVA 200
megawatt plant.
The IERL-RTP data storage system should be used to
store data relating to all FBC process, analytical,
and environmental information. This information
should be readily available to EPA planners, deci-
sion makers and contractors.
Sufficient data should be generated on conventional
coal combustion units to make a complete comparison
of environmental impacts of the effluents generated
by the two processes, not simply to compare the
relative amounts of sulfur dioxide and nitrogen
dioxide produced.
Potential impacts from wastes generated by and re-
sulting from the sorbent regeneration .process should
be investigated, to determine:
The fate of pollutants concentrated in bed re-
ject material during the sorbent regeneration
process.
The chemical composition of sorbent material
when the useful life of the material has been
reached.
The leachate composition of the spent sorbent
and comparisons of leachate characteristics to
that of once through bed reject material.
The leachate quality when spent sorbent is
mixed with other FBC solid wastes, and investi-
gation into compatibility for disposal with
other types of solid wastes.
The nature, composition, and quantities of
liquid, atmospheric, and fugitive emissions
from the sorbent regeneration and allied sulfur
recovery processes.
237
-------
238
-------
REFERENCES
Abelson, H.I., and W.A. Lowenbach. 1977. Procedures Manual
for Environmental Assessment of Fluidized-Bed Combustion
Process. EPA-60017-77-009, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
Allen, J.M.,
1978a.
J.E. Howes, Jr., S.E. Miller, and K.M. Duke.
Comprehensive Analysis of Emissions for EXXON
Fluidized-Bed Combustion Miniplant Unit. Ln
Compilation of Level 1 Environmental Assessment
Data, eds. N.H. Gaskins and F.W. Sexton.
EPA-600/2-78-211, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
1978b. Comprehensive Analysis of Emissions from MERC
Fluidized-Bed Combustion Unit. In Compilation
of Level 1 Environmental Assessment Data, eds.
N. H. Gaskins and F.W. Sexton.
EPA-600/2-78-211, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
Alvin, M.A., E.P. O'Neill, and D.L. Keairns. 1977.
Thermodynamic Projections of Trace Element Release in
Fluidized- Bed Combustion Systems. Proceedings of the
Fifth International Conference on FBC, Volume 2. The
Mitre Corporation, Washington, D.C.
Alvin, M.A., E.P. O'Neill, L.N. Yannopoulos, and D.L.
Keairns. 1978. Evaluation of Trace Element Release from
Fluidized- Bed Combustion Systems. EPA-600/7-78-050, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina.
American Public Health Association. 1976. Standard Methods
for the Examination of Water and Wastewater, 14th
Edition. American Public Health Association, Washington,
DC.
American Society for Testing and Materials. 1979. Annual
Book of ASTM Standards, Part 31 - Water. American Society
for Testing and Materials, Philadelphia, Pennsylvania.
239
-------
Anderson, L.B., M.A. Herther, and R.J. Milligan. 1978. SAM
1: An Intermediate Screening Method for Environmental
Assessment of Fossil Energy Process Effluents. Acurex
Report TR- 79-154.
Andren, A.W., D.H* Klein, and Y. Talmi. 1975. Selenium in
Coal- Fired Steam Plant Emissions. Environmental Science
and Technology, 9(9): 856.
Babcock & Wilcox Company. 1979. FBC Development Facility
and Commercial Utility AFBC Design Assessment. Technical
Progress Report No. 8, Electric Power Research Institute,
Palo Alto, California.
Ball, R.J., and G.E. Anderson. 1977. Optimum Site Exposure
Criteria for SC^ Monitoring. The Center for the
Environment and Man, Inc., Hartford, Connecticut.
(EPA-450/3-77-013.)
Beer, J.M., A.F. Sarofim, L.K. Chan, and A.M. Sprouse. 1977.
NO Reduction by Char in Fluidized Combustion. Proceedings
of Fifth International Conference on FBC, Volume 2. The .
Mitre Corporation, Washington, DC.
Behar, J. V., E. A. Schuck, R. E. Stanley, and G. F. Morgan.
1979. Integrated Exposure Assessment Monitoring.
Environmental Science and Technology, 13(l):34-39.
Branson, D.R. 1978. Predicting the Fate of Chemicals in the
Aquatic Environment from Laboratory Data. In Estimating-
the Hazard of Chemical Substances to Aquatic Life, eds.
John Cairns, Jr., K.L. Dickson, and A.W. Maki. ASTM STP
657, American Society for Testing and Materials.
Brinkhurst, R.O., A.L. Hamilton, and H.B. Herrington. 1968.
Components of the Bottom Fauna of the St. Lawrence Great
Lakes. Publication No. 33, Great Lakes Institute,
University of Toronto.
Cairns, J., and K.L. Dickson. 1973. Biological Methods for
the Assessment of Water Quality. ASTM Special
Publication 528, American Society for Testing and
Materials, Philadelphia, Pennsylvania.
Cairns, J., K.L. Dickson, and G.F. Westlake. 1976.
Biological Monitoring of Water and Effluent Quality.
ASTM Special Technical Publication 607, American Society
for Testing and Materials, Philadelphia, Pennsylvania.
240
-------
Cairns, J.f and A.W. Maki. 1979. Hazard Analysis in Toxic
Materials Evaluation. Journal of the Water Pollution
Control Federation, 51(4):666-671.
Carpenter, R.L., S.H. Weissman, G.J. Newton, R.L. Hanson,
E.R. Peele, M.H. Mazza, J.J. Kovach, D.A. Green, and U.
Grimm. 1978. Characterization of Aerosols Produced By
an Experimental Fluidized Bed Coal Combustor Operated
with Sub- Bituminous Coal. LF-57, Lovelace Biomedical
and Environmental Research Institute, Albuquerque, New
Mexico.
Chemical and Engineering News. 1978. Power Plant Ash
Disposal a Growing Problem. Chemical and Engineering
News, 56(45):26.
Chutter, F.M. 1972. An Empirical Biotic Index of the Quality
of Water in South African Streams and. Rivers. Water
Research, 6:19-30.
Cleland, J.G., and G.L. Kingsbury. 1977. Multi-media
Environmental Goals for Environmental Assessment, Volumes
1 and 2. EPA-600-77-136a and 136b, U.S. Environmental
Protection Agency, Research Triangle Park, North
Carolina.
Coherd, C., and T. Cusino. 1979. Development of Emission
Factors for Wind Erosion of Aggregate Storage Piles. 72nd
Annual Meeting of the Air Pollution Control Association,
Cincinnati, Ohio.
Conner, W.D. 1979. Comparative Study of Plume Opacity
Measurement Methods. U.S. Environmental Protection
Agency, Stationary Source Emission Measurement Research
Branch, Research Triangle Park, North Carolina.
Conner, W.D., and J.R. Hodkinson. 1972. Optical Properties
and Visual Effects of Smoke-Stack Plumes. Publication
Number 999-AP-30, U.S. Public Health Service, Washington,
DC.
Cornaby, B.W., D.A. Savitz, L. Pomerantz, and K.S. Murthy.
1977. Development of Environmental Objectives Based on
Health and Ecological Effects. Proceedings of the Fifth
International Conference on Fluidized Bed Combustion,
Volume 2. The Mitre Corporation, Washington DC.
Couch, D.B., N. L. Forbes, and A.W. Hsie. 1978. Mutation
Research 57, 217.
241
-------
Dalton-Dalton-Newport. 1978. Guidelines for Multimedia
Environmental Monitoring of DOE Fossil Energy RD&D
Facilities, Volumes 1 and 2. Dalton-Dalton-Newport,
Cleveland, Ohio. (NTIS No. FE-2495-6 A01.)
Daudy, D.R., and J.H. Feth. 1967. Applications of Factor
Analysis in a Study of Chemistry of Groundwater Quality,
Mojave River Valley, California. Water Resources
Research, 3(2):505-510.
DeAngelis, D.G., and R.B. Reznik. 1979. Source Assessment:
Residential Combustion of Coal. EPA-600/2-79-019a, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina.
Denver Research Institute. 1979. Balloon Plume Sampler.
Technical Progress Report, University of Denver, Denver,
Colorado.
Donahue, J. M. 1970. Chemical Treatment.
Engineering, 7(5):35.
Industrial Water
Dooley, J.C. 1976. Fumigation From Power Plant Plumes in the
Lakeshore Environment. University of
Wisconsin-Milwaukee, Milwaukee, Wisconsin.
Dorsey, J.A., L.D. Johnson, R.A. Statnick, and C.H.
Lockmuller. 1977. Environmental Assessment Sampling and
Analysis: Phased Approach and Techniques for Level 1.
EPA-600/2-77- 115, U.S. Environmental Protection Agency,
Industrial Environmental Research Laboratory, Research
Triangle Park, North Carolina.
Draggan, S., and J. M. Giddings. 1978. Testing Toxic
Substances for Protection of the Environment in the
Science of the Total Environment, Volume 9. Amsterdam:
Elserrier Scientific Publishing Co.
Druckrey, H., H. Kruse, R. Preussmann, S. Ivankovic, and
C. Landshutz.
1966. Z. Krebsforsch 68, 103.
1970.
Z. Krebsforsch 74, 241
Duke, K.M., M.E. Davis, and A.J. Dennis, 1977. IERL-RTP
Procedures Manual: Level 1 Environmental Assessment
Biological Tests for Pilot Studies. EPA 600/7-77-043,
U.S. Environmental Protection Agency, Industrial
Environmental Research Laboratory, Research Triangle
Park, North Carolina.
242
-------
Dybdahl, A.W., and M.J. Cunningham. 1979. Utilization of
the Omega-I Lidar in EPA Enforcement Activities.
Symposium on the Transfer and Utilization of Particulate
Control Technology, National Enforcement Investigations
Center, Denver, Colorado.
Eatough, D. J. 1979. Determination of Sulfur Speciation in
Industrial Aerosols in an 302 Rich Environment -
Summary Progress Report for Period 15 June 1976 - 14 June
1979 (DOE Contract EY-76-S-02 2988). U.S. Department of
Energy, Thermochemical Institute, Brigham Young
University, Provo, Utah.
Edwards, M.D. 1978. NAWDEX - A Program to Improve Access to
Water Data. Water Resource Division, U.S. Geological
Survey. Presented at National Conference on Quality
Assurance, Denver, Colorado.
Eimutis, E.G. 1976. Source Assessment: Prioritization of
Stationary Air Pollution Sources - Model Description.
EPA- 600/2-76-032a, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Ensor, D. S., and J. J. Pilat.
1970. Plume Opacity and Particulate Mass
Concentration. Atmospheric Environment, 4:
163-173.
1971a. Calculation of Smoke Plume Opacity from
Particulate Air Pollutant Properties. Journal
of the Air Pollution Control Association,
21:496-501.
1971b. Comparison Between the Light Extinction Aerosol
Mass Concentration Relationship of Atmospheric
and Air Pollutant Aerosols. Atmospheric
Environment, 5:209-215.
Environment Reporter. 1979. EPA Initiates Testing To
Evaluate Dioxins Created by Natural Combustion.
Environment Reporter, 10(18):1077.
Everett, L.G., ed. 1979. Groundwater Quality Monitoring of
Western Coal Strip Mining: Identification and Priority
Ranking of Potential Pollution Sources.
EPA -60017-79-024, Research Triangle Park, North Carolina.
Everett, L.G., K. D. Schmidt, R. M. Tinlin, and D. K. Todd.
1976. Monitoring Groundwater Quality: Methods and
Costs. EPA-600/4-76-023, U. S. Environmental Protection
Agency, Office of Research and Development.
243
-------
Exton, R.J. 1977. An Ultraviolet Video Technique for
Visualization of Stack Plumes and for Measuring Sulfur
Dioxide Concentration and Effluent Velocity. Technical
Paper 1014, National Aeronautics and Space
Administration, Hampton, Virginia.
Farmer, M.H., E.M. Magee, and P.M. Spooner. 1977.
Application'of Fluidized-Bed Technology to Industrial
Boilers. EPA- 600/7-77-011, U.S. Environmental
Protection Agency, Industrial Environmental Research
Laboratory, Research Triangle Park, North Carolina.
Federal Power Commission.
1974a. Electric Power Statistics. U.S. Department of
Commerce, Washington, D.C.
1974b. Steam Electric Plant Air and Water Quality
Control Data for the Year Ended December 31,
1971. FPC-S- 239, U.S. Department of Commerce,
Washington, D.C.
Federal Register.
12-1-76. EPA-Approved Guidelines Establishing Test
Procedures for the Analysis of Pollutants.
Federal Water Pollution Control Act Amendments
of 1972, Section 304(g), 40 CFR 136(41 FR
52780) .
5-10-79. Ambient Air Quality Monitoring, Data Reporting,
and Surveillance Provisions. 40 CFR 51, 52,
53, 58.
Fennelly, P.F., D.F. Durocher, H. Klemm, and R.R. Hall.
1977. Preliminary Environmental Assessment of Coal-Fired
Fluidized-Bed Combustion Systems. EPA-600/7-77-054, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina.
Fennelly, P.F., D.F. Durocher, A.S. Werner, M.T. Mills, S.M.
Weinstein, A.H. Castaline, and C. Young. 1976.
Environmental Assessment Perspectives. GCA Corporation,
Bedford, Massachusetts. (EPA-600/2-76-069.)
Fisher, G. L., C. E. Chrisp, and W. E. Jennings. 1978.
Physical and Chemical Properties of Mutagens in Coal Fly
Ash. 12th Annual Conference on Trace Substance in
Environmental Health, Columbia, Missouri.
Fisher, G. L., C. E. Chrisp, and O.G. Raabe. 1979. Physical
Factors Affecting the Mutagenicity of Fly Ash from a
Coal-Fired Power Plant. Science, 204:879.
244
-------
Floyd, J.R. 1979. Quality Assurance Audit Techniques for
Validating Data from Continuous Emission Stack Monitoring
(GEM) Systems. 72nd Annual Meeting of the Air Pollution
Control Association, Cincinnati, Ohio.
Gasner, L.L. 1977. Enhancement of Sorbent Utilization By the
Use of Additives. Proceedings of the Fluidized-Bed
Combustion Technology Exchange Workshop, McLean, Virginia,
Conf-770447- P-2, Volume 2. U.S. Energy Research and
Development Administration.
GCA/Technology Division. 1978. Test Plan for Comprehensive
Analysis of Emissions from the Babcock & Wilcox Fluidized-
Bed Combustion Unit. GCA Corporation, Bedford,
Massachusetts.
Glenn, R.D., and E.B. Robison. 1970. Characterization of
Emissions from Fluidized-Bed Combustion of Coal and
Control of Sulfur Emission with Limestone. Proceedings of
the Second International Conference on Fluidized Bed
Combustion, College Corner, Ohio. U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Greenwood, D.R., G.L. Kingsbury, and J.G. Cleland. 1979. A
Handbook of Key Federal Regulations and Criteria' for
Multimedia Environmental Control. Research Triangle
Institute, Research Triangle Park, North Carolina.
Hansen, Lee D., L. Whiting, D.J. Eatough, T.E. Jenson, and
R.M. Izatt. 1976. Determination of Sulfur (IV) and
Sulfate in Aerosols by Thermometric Methods. Analytical
Chemistry, 48(4):634-638.
Harris, J.C., M.J. Hayes, P.L. Levins., and D.B. Lindsay.
1979. Procedures for Level 2 Sampling and Analysis of
Organic Materials. Arthur D. Little Inc., Cambridge,
Massachusetts. (EPA-600/7-79-033.)
Harris, J.C., and P.L. Levins. 1978. Interim Procedures for
Level 2 Sampling and Analysis of Organic Materials. EPA-
600/7-78-016, U.S. Environmental Protection Agency,
Industrial Environmental Research Laboratory, Research
Triangle Park, North Carolina.
Henschel, D.B.
1977. Environmental Emissions from Coal-Fired
Industrial Fluid Bed Boilers. Conference on
Engineering Fluidized-Bed Combustion Systems for
Industrial Use, Columbus, Ohio.
245
-------
1979. The EPA R&D Program to Assess the Solid Residue
From the Fluidized-Bed Combustion Process.
Workshop on Solid Waste Research and Development
Needs for Emerging Coal Technologies, San Diego,
California.
Herrmann, R., U. ,Bolz, W. Symader and H. Rump. 1979.
Interpretation and Prediction of Spatial Variation in
Trace Metals in Small Rivers by Cononical and Discriminant
Analysis. Proceedings of the Third International
Symposium in Hydrology, Colorado State University, Fort
Collins, Colorado.
Horio, M., S. Mori, and I. Muchi. 1977. A Model Study for
the Development of Low NOx Fluidized-Bed Coal Combustors.
Proceedings of the Fifth Conference on Fluidized Bed
Combustion, Volume 2. The Mitre Corporation, Washington,
DC.
Hoy, H.R., and A.G. Roberts. 1977. Further Experiments on
the Pilot Scale Pressurized Combustor at Leatherhead,
England. Proceedings of the Fluidized Bed Combustion
Technology Exchange Workshop, McLean, Virginia,
Conf-770447-P-2, Volume 2. U.S. Energy Research and
Development Administration.
Ingraham, T.R., and E.M. Zechel. 1979. Selection of
Candidate Substances for Control in an Environmental
Regulation Agency. Journal of the Air Pollution Control
Association, 29:38-42.
International Joint Commission (IJC). 1978. Great Lakes
Water Quality Agreement. Great Lakes Regional Office,
Windsor, Ontario.
Jacobs, F., and G.C. Grant. 1978. Guidelines for Zooplankton
Sampling in Quantitative Baseline and Monitoring
Programs. Virginia Institute of Marine Science,
Gloucester Point, Virginia. (EPA-600/3-78-026.)
Jarry, R.L., L.J. Anastasia, E.L. Carls, A.A. Jonke, and
G.J. Vogel. 1970. Comparative Emissions of Pollutants
During Combustion of Natural"Gas and Coal in Fluidized
Beds. Proceedings of the Second International Conference
on Fluidized Bed Combustion, College Corner, Ohio. U.S
Environmental Protection Agency, Research Triangle Park,
North Carolina.
Jones, C.J. 1978. The Ranking of Hazardous Materials by
Means of Hazard Indices. Journal of Hazardous Materials,
2:363- 389.
246
-------
Kaakinen, J.W., R.M. Jordan, M.H. Lawasani, and R.E. West.
1975. Trace Element Behavior in Coal-Fired Power Plant.
Environmental Science and Technology, 9(9): 862.
Keairns, D.L., D.H. Archer, J.R. Hamm, S.A. Jansson,
B.W. Lancaster, E.P. O'Neill, C.H. Peterson, C.C. Sun,
E.F. Sverdrup, E.J. Vidt, and W.C. Jang. 1975.
Fluidized Bed Combustion Process Evaluation: Phase II -
Pressurized Fluidized Bed Coal Combustion Development.
EPA-650/2-75-027c, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Keith, L.H., and W.A. Telliard. 1979. Priority Pollutants
I - A Perspective View. Environmental Science and
Technology, 13(4):416-423.
Kenson, R.E., and Bartlett, P.T. 1976. Technical Manual for
the Measurement of Fugitive Emissions: Roof Monitor
Sampling Method for Industrial Fugitive Emissions. U.S.
Environmental Protection Agency, Office of Research and
Development, Washington, DC.
Kingsbury, G.L., and R.C. Sims. 1978. Preview of the MEG's
Supplement. RTI/1430/27-01S, Research Triangle Institute,
Research Triangle Park, North Carolina.
Kingsbury, G.L., and J.B. White. 1979. Multimedia
Environmental Goals. RTI/1710/04-01S, Research Triangle
Institute, Research Triangle Park, North Carolina.
Klein, D.H., A.W. Andren, J.A. Carter, J.F. Emery, C. Feldman,
W. Fulkerson, W.S. Lyon, W.C. Ogle, Y. Talmi, R.I. Van
Hook, and N. Bolton. 1975. Pathways of Thirty-Seven
Trace Elements Through Coal-Fired Power Plant.
Environmental Science and Technology, 9(10):973.
Koch, R.C., W.G. Biggs, P.H. Hwang, L. Leichter,
K.E. Pickering, E.R. Sawdey, and J.L. Swift. 1977. Power
Plant Stack Plumes in Complex Terrain, An Appraisal of
Current Research. EPA-600/7-77-020, U.S. Environmental
Protection Agency, Environmental Sciences Research
Laboratory, Research Triangle Park, North Carolina.
Kolnsberg, H. J. 1976. Technical Manual for Measurement of
Fugitive Emissions: Upwind/Downwind Sampling Method for
Industrial Emissions. EPA 600/2-76/089a, U.S.
Environmental Protection Agency, Office of Research and
Development, Washington, DC.
Kolnsberg, H. J., P. W. Kalika, R. E. Kenson, and W. A.
Marrone. 1976. Technical Manual for Measurement of
247
-------
Fugitive Emissions: Quasi-Stack Sampling Method for
Industrial Fugitive Emissions. EPA-600/2-76-089c, U.S.
Environmental Protection Agency, Office of Research and
Development, Washington, DC.
Kolnsberg, H. J., and R. L. Severance. 1979. Development of
a Fugitive Assessment Sampling Train for Particulate and
Organic Fugitive Emissions. 72nd Annual Meeting of the
Air Pollution Control Association, Cincinnati, Ohio.
Landwehr, J.M. 1974. Water Quality Indices - Construction
and Analysis. Ph.D. dissertation, University of Michigan,
Ann Arbor, Michigan.
Lawrence-Berkeley Laboratory. Environmental Instrumentation
Group. 1976. Instrumentation for Environmental
Monitoring, Volumes 1 and 2. Lawrence Berkeley
Laboratory, University of California, Berkeley.
Lee, M.L., D.K. Later, D.M. Rollins, D.J. Eatough, and L. D.
Hansen. 1979. Dimethyl and Monomethyl Sulfate: Presence
in Coal Fly Ash and Airborne Particulate Matter.
Submitted to Thermal Chemical Institute, Brigham Young
University, Provo, Utah.
Lentzen, D.E., D.E. Wagoner, E.D. Estes, and W.F. Gutknecht.
1978. IERL Procedures Manual: Level 1 Environmental
Assessment (Second Edition). EPA-600/7-78-201, U.S.
Environmental Protection Agency, Industrial Environmental
Research Laboratory, Research Triangle Park, North
Carolina.
Liu, K.T., G.W. Felton, A.E. Weller, and H. Nack. 1979.
Multi- Solid Fluid Bed Combustor - Initial Pilot Plant
Results. AIChE 86th National Meeting, Houston, Texas.
Lucke, V.H., and K.R. Murphy. 1977. Environmental Analysis
of the General Electric PFB Combined Cycle Power Plant
Design. Proceedings of the Fifth International Conference
on Fluidized Bed Combustion, Volume 2. The Mitre
Corporation, Washington, DC.
Ludwig, C.B., and M. Griggs. 1976. Application of Remote
Techniques in Stationary Source Air Emission Monitoring..
EPA 340/1-76-005, U.S. Environmental Protection Agency,
2ffice of Enforcement, Washington, DC.
Ludwig, F.L., J.H. Kealoha, and E. Shelar. 1977. Selecting
Sites for Monitoring Total Suspended Particulates.
Research Institute, Menlo Park, California.
EPA-450/3-/7-018, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
248
-------
Ludwig, F.L., and E. Shelar. 1978. Site Selection for the
Monitoring of Photochemical Air Pollutants.
EPA-450/3-78- 013, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Lyons, W.A. 1977. Mesoscale Air Pollution Transport in
Southeast Wisconsin. EPA-600/4-77-010, U.S.
Environmental Protection Agency, Environmental Monitoring
Series, Research Triangle Park, North Carolina.
McKenzie, E.G. 1978. Burning Coal in Fluidized Beds.
Chemical Engineering, 85(18): 116.
Merryman, E.L., A. Levy, G.W. Felton, K.T. Liu, J.M. Allen,
and H. Nack.
1977. Method for Analyzing Emissions from Atmospheric
Fluidized-Bed Combustor. EPA-6007-77-034, U.S.
Environmental Protection Agency, Research
Triangle Park, North Carolina.
1978. Method for Analyzing Emissions from Atmospheric
Fluidized-Bed Combustor. In Compilation of
Level 1 Environmental Assessment Data, eds.
N.H. Gaskins and F.W. Sexton.
EPA-600/2-78-211, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
Millan, M.M., and R.M. Hoff. 1976. The COSPEC Remote Sensor,
II-Electronic Set-up Procedures. ARQT-6-76, Environment
Canada, Atmospheric Environment Service, Downsview,
Ontario, Canada.
Mills, M.T., and F.A. Record. 1975. Comprehensive Analysis
of Time-Concentration Relationships and Validation of a
Single-Source Dispersion Model. EPA-450/3-75-083, U.S.
Environmental Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle Park, North
Carolina.
Mills, M.T., and R.W. Stern. 1975. Model Validation and
Time- Concentration Analysis of Three Power Plants.
EPA-450/3-76- 002, U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards, Research
Triangle Park, North Carolina.
Mitchell, R.E. 1976. Nitrogen Oxide Formation from
Chemically- Bound Nitrogen During the Combustion of
Fossil Fuels. SAND 76—8227, Sandia Laboratories,
Livermore, California.
Moffat, A.J., and M. M. Millan. 1969. The Applications of
Optical Correlation Techniques to the Remote Sensing of
249
-------
S02 Plumes Using Sky Light. Symposium on Advances in
Instrumentation for Air Pollution Control Administration
of the Department of HEW, Cincinnati, Ohio.
Morgan, G. B. 1978. Energy Resource Development: the
Monitoring Components. Environmental Science and
Technology,•12(1):34-43.
Murthy, K.S., J.M. Allen, D.A. Sharp, and K.M. Duke. 1977.
Multimedia Pollutant Emissions Data for Fluidized-Bed
Combustion of Coal. Proceedings of the Fifth
International Conference on Fluidized Bed Combustion,
Volumes 1 and 2. The Mitre Corporation, Washington, DC.
Murthy, K.S., and D.B. Henschel. 1978. Environmental
Assessment of Fluidized-Bed Combustion: Status and
Results. AIChE 71st Annual Meeting, Miami Beach,
Florida.
National Academy of Engineering. 1973. Water Quality
Criteria, 1972. U.S. Environmental Protection Agency,
Washington, DC.
National Academy of Science. 1972. Particulate Polycyclic
Organic Matter. National Academy of Sciences,
Washington, DC.
National Institute for Occupational Safety and Health.
1974. NIOSH Manual of Analytical Methods. NIOSH
75-121, U.S. Government Printing Office,
Washington, DC.
1976. Registry of Toxic Effects of Chemical
Substances. Department of Health, Education,
and Welfare Publication, U.S. Government
Printing Office, Washington, DC.
National Research Council. 1977. Environmental Monitoring.
National Academy of Sciences, Washington, DC.
Newby, R.A., and D.L. Keairns. 1977. Initial Assessment of
Alternative S02 Sorbents for Fluidized Bed Combustion
Power Plants. Proceedings of the Fifth International
Conference on Fluidized Bed Combustion, Volume 2. The'
Mitre Corporation, Washington, DC.
Newby, R.A., N.H. Ulerich, D.M. Bachovchin, C.H. Peterson,
N.D. Rohatgi, E.J. Vidt, W.G. Vaux, and D.L. Keairns.
1979. In The Babcock & Wilcox Company, Fluidized Bed
Combustion Development Facility and Commercial Utility
250
-------
AFBG Design Assessment. Technical Progress Report No. 8,
Electric Power Research Institute, Palo Alto, California.
Newby, R.A., N.H. Ulerich, E.P. O'Neill, D.F. Ciliberti, and
D.L. Keairns. 1978. Effect of 803 Emission
Requirements on Fluidized-Bed Combustion Systems:
Preliminary Technical/ Economical Assessment.
EPA-600/7-78-163, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Nuclear Regulatory Commission. 1972. Regulatory Guides,
Safety Guide 23, On-Site Meteorological Programs. U.S.
Nuclear Regulatory Commission, Washington, DC.
NUS Corporation. 1970. Design of Water Quality Surveillance
Systems, Phase I Systems Analysis Framework. Federal
Water Quality Administration, Washington, DC.
Ott, W.R., 1978. Water Quality Indices: A Survey of Indices
Used in the United States. EPA 600/4-78-005, U.S.
Environmental Protection Agency, Office of Monitoring and
Technical Support.
Pasquill, F, 1974. Atmospheric Diffusion. New York:
Halstead Press.
Patterson, W.C., and R. Griffin. 1978. Fluidized-Bed Energy
Technology Coming To a Boil. New York: Inform, Inc.
Peeler, J.W. 1979. Continuous Opacity and Particulate
Emissions Monitoring in the Federal Republic of Germany:
Selected Papers from Current Literature. Entropy
Environmentalists, Inc., Research Triangle Park, North
Carolina.
Peltier, W. 1978. Methods for Measuring the Acute Toxicity
of Effluents to Aquatic Organisms. EPA 600/4-78-012,
U.S. Environmental Protection Agency, Surveillance and
Analysis Division, Athens, Georgia.
Pereira, F.J., J.M. Beer, B.M. Gibbs, and A.B. Hedley.
1976. NOX Emissions from Fluidized Bed Coal
Combustors. 15th International Symposium on Combustion,
Pittsburgh, Pennsylvania.
Petrie, T.W., and G.L. Kingsbury. 1978. Ranking
Environmental Impact of Trace Elements in Coal by Use of
Multimedia Environmental Goals. RTI/1603/23/01S,
Research Triangle Institute, Research Triangle Park,
North Carolina.
251
-------
Pielou, E.G. 1966. Shannon's Formula as a Measure of
Specific Diversity, Its Use and Misuse. American
Naturalist, 100: 463-465.
Pillai, K.K., and A.G. Roberts. 1977. Emission
Characteristics of a Pressurized Fluidized Bed Coal
Combustor. : Draft. U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Piperno, E. 1975. Trace Element Emissions: Aspects of
Environmental Toxicology. In Trace Elements in Fuel, ed.
S.P. Babu. Advances in Chemistry, Series 141, American
Chemical Society, Washington, DC.
Price, J., J. Wiersema, and J. Hartfield. 1979. New Source
Review for Acceptability of Public Exposure to Chemical
Emissions. 72nd Annual Meeting of the Air Pollution
Control Association, Cincinnati, Ohio.
Reznik, R.B., E.G. Eimutis, J.L. Delaney, S.R. Archer, J.C.
Ochsner, W.R. McCurley, and T.W. Hughes. 1978. Source
Assessment: Prioritization of Stationary Water Pollution
Sources. EPA-600/2-78-004q, U.S. Environmental
Protection Agency, Research Triangle Park, North
Carolina.
Robertson, D.E., and R. Carpenter. 1974. Neutron Activation
Techniques for the Measurement of Trace Metals in
Environmental Samples. Technical Information Center,
U.S. Atomic Energy Commission, Oak Ridge, Tennessee.
Ruch, R.R., H.J. Gluskoter, and N.F. Shimp. 1974.
Occurrence and Distribution of Potentially Volatile Trace
Elements in Coal. EPA-650/2-74-054, U.S. Environmental
Protection Agency, Research Triangle Park, North
Carolina.
Rupp, E.M., D.C. Parzyck, P.J. Walsh, R.S. Booth, R.J.
Raridon, and B.L. Whitfield. 1978. Composite Hazard
Index for Assessing Limiting Exposures to Environmental
Pollutants: Application Through a Case Study.
Environmental Science and Technology, 12(7):802-806.
Ryan, L.E., R.G. Beimer, and R.F. Maddalone. 1979. Level-2
Chemical Analysis of Fluidized-Bed Combustor Samples.
EPA- 600/7-79-063b, U.S. Environmental Protection Agency,
Office of Energy, Minerals and Industry, Research
Triangle Park, North Carolina.
Sanders, T.G. 1979. Design of Water Quality Monitoring
Systems. Colorado State University, Boulder, Colorado.
252
-------
Schalit, L.M.f and K.J. Wolfe. 1978. SAM/1A: A Rapid
Screening Method for Environmental Assessment of Fossil
Energy Process Effluents. EPA-600/7-78-015, U.S.
Environmental Protection Agency, Office of Research and
Development.
Schultz, H., E.A. Hattman, and W.B. Booher. 1975. The Fate
of Some Trace Elements During Coal Pretreatment and
Combustion. In Trace Elements in Fuel, ed. S.P. Babu.
Advances in Chemistry, Series 141, American Chemical
Society, Washington, D.C.
Scott, R.D. 1969. The Macro-Invertebrate Biotic Index - A
Water Quality Measurement and Natural Continuous Stream
Monitor for the Miami River Basin. The Miami Conservancy
District, Dayton, Ohio.
Slawson, G.C. Jr., ed. 1979. Groundwater Quality Monitoring
of Western Oil Shale Development: Identification and
Priority Ranking of Potential Pollution Sources.
EPA-600/7-79023, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
Smith, F. 1978. Level 1 Quality Assurance. Research
Triangle Institute, Research Triangle Park, North
Carolina. (EPA- 600/7-78-151.)
Sperling, R.B. 1975. Evaluation of the Correlation
Spectrometer as an Area S02Monitor. PB-249.113,
National Technical Information Service.
Sperling, R.B., M.A. Peache, and W.M. Vaughan. 1979.
Accuracy of Remotely Sensed S02 Mass Emission Rates.
EPA-600/2-79- 094, U.S. Environmental Protection Agency,
Emissions Measurement and Characterization Division,
Office of Research and Development, Research Triangle
Park, North Carolina.
Steele, T.D., and T.R. Dyar. 1974. Harmonic Analysis of
Water Temperatures of Georgia Streams. Transactions of
the American Geophysical Union, 55(12):1114.
Stevens, R.K., and T.J. Dzubay.
1978a. Dichotomous Sampler - A Practical Approach to
Aerosol Fractionation and Collection.
EPA-600/2- 78-112, U.S. Environmental
Protection Agency, Industrial Environmental
Research Laboratory, Research Triangle Park,
North Carolina.
253
-------
1979b. Sampling and Analysis of Atmospheric Sulfates
and Related Species. Atmospheric Environment,
55-68.
Stone, R., and R. Kahle. 1977. Environmental Assessment of
Solid Residues from Fluidized-Bed Fuel Processing. U.S.
Environmental Protection Agency, Industrial Environmental
Research Laboratory, Research Triangle Park, North
Carolina.
Stoner, I.D. 1976. Water Quality Indices for Specific Water
Uses. U.S. Geological Survey, Oklahoma City, Oklahoma.
Sugarek, R.L., and T.G. Sipes. 1978. Controlling Sulfur
Dioxide Emission from Coal-Fired Steam-Electric
Generators: Water Pollution Impact, Volume 2, Technical
Discussion. U.S. Environmental Protection Agency,
Industrial Environmental Research Laboratory, Research
Triangle Park, North Carolina.
Sun, C.C., and J.T. McAdams. 1979. Assessment of RCRA/EP
Test Results on FBC Residue. Westinghouse R&D Center,
Pittsburgh, Pennsylvania.
Suprenant, N., R. Hall, S. Slater, T. Susa, M. Sussman, and
C. Young. 1976. Preliminary Emissions Assessment of
Conventional Stationary Combustion Systems, Volume 2 -
Final Report. GCA/Technology Division, Bedford,
Massachusetts. Prepared for U.S. Environmental
Protection Agency. (Publication Number
EPA-600/2-76-046b.)
Swartz, R.C. 1978. Techniques for Sampling and Analyzing
the Marine Macrobenthos. Corvallis Environmental
Research Laboratory, Newport, Oregon. (EPA-600/3-78-030.)
Swift, W.M., S.H.D. Lee, J.C. Montagna, G.W. Smith,
I. Johnson, G.J. Vogel, and A.A. Jonke. 1977. Plans and
Studies on Flue Gas Cleaning and Particulate Monitoring
in PFBC. Proceedings of the Fifth International
Conference on Fluidized Bed Combustion, Volume 2. The
Mitre Corporation, Washington, DC.
Symader, W., and W. Thomas. 1978. Interpretation of Average
Heavy Metal Pollution in Flowing Waters and Sediment by
Means of Hierarchical Grouping Analysis Using Two
Different Error Indices. Cantena, 5(2):133-144.
Todd, O.K., R.M. Tinlin, K.D. Schmidt, and L.G. Everett.
1976. Monitoring Groundwater Quality: Monitoring
254
-------
Methodology. EPA-600/4-76-026, U.S. Environmental
Protection Agency, Office of Research and Development.
TSCA Interagency Testing Committee (CEQ). 1977. Initial
Report of the TSCA Interagency Testing Committee to the
Administrator, Environmental Protection Agency. Federal
Register, Vol. 42, No. 197.
Turner, B. 1974. Workbook of Atmospheric Dispersion
Estimates. Publication Number AP-27, U.S. Environmental
Protection Agency, Office of Air Programs.
U.S. Environmental Protection Agency..
1973. Handbook for Monitoring Industrial Wastewater.
U.S. Environmental Protection Agency,
Technology Transfer, Washington, DC.
1976a. Guidelines for Evaluating Supplementary Control
Systems. EPA-450/2-76-003, U.S. Environmental
Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle Park,
North Carolina.
1976b. A Technique for Supplementary Control System
Reliability Analysis and Upgrading.
EPA-450/2-76- 015, U.S. Environmental
Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle Park,
North Carolina.
1976c. Quality Criteria for Water. U.S. Environmental
Protection Agency, Office of Water and
Hazardous Materials, Washington, DC.
1977a. Proposed National Interim Secondary Drinking
Water Regulations. Environment
Reporter - Current Developments,
7(49):1884-1887.
1977b. Sampling and Analysis Procedures for Screening
of Industrial Effluents for Priority
Pollutants. U.S. Environmental Protection
Agency, Environmental Monitoring and Support
Laboratory, Cincinnati, Ohio.
1978a. Ambient Monitoring Guidelines for Prevention of
Significant Deterioration (PSD).
EPA-450/2-78-019, U.S. Environmental Protection
Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, North
Carolina.
255
-------
1978b. Guideline on Air Quality Models. EPA-450/2-78-
027, U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards,
Research Triangle Park, North Carolina.
1978c. National Interim Primary Drinking Water
Regulations. Code of Federal Regulations,
Title 40, Part 141.11.
1979a. Methods for Chemical Analysis of Water and
Wastes. EPA-600/4-79-020, U.S. Environmental
protection Agency, Technology Transfer,
Washington, DC.
1979b. Ohio: Approval and Promulgation of
Implementation Plans - Proposed Rule. Federal
Register, Vol. 44, No. 114:33711-33712.
1979c. Regulations on National Primary and Secondary
Ambient Air Quality Standards. Federal
Register, Vol. 44, No. 28: 8221.
1979d. Water Quality Criteria - Request for Comments.
Federal Register, Vol. 44, No. 52: 15926-12981,
U. S. Geological Survey. 1976. Selected Methods of the U. S,
Geological Survey for Analysis of Wastewaters. Report
No. 76-177, U. S. Geological Survey, Reston, Virginia.
Veldman, D.J. 1967. FORTRAN Programming for the Behavorial
Sciences. New York: Holt, Rinehart, and Winston.
Walsh, P. J., G. G. Killough, and P. S. Rohwer. 1978.
Composite Hazard Index for Assessing Limiting Exposures
to Environmental Pollutants: Formulation and
Derivation. Environmental Science and Technology,
12:799-807.
Ward, R.C. 1979. Regulatory Water Quality Monitoring: A
Systems Perspective. Water Resources Bulletin,
15(2) :369- 380.
Weber,'C. I. 1973. Biological Field and Laboratory Methods
for Measuring the Quality of Surface Waters and
Effluents. EPA-670/4-73-001, U.S. Environmental
Protection Agency, Washington, DC.
Weber, E. 1970. Determination of the Life Time of S02 by
Simultaneous C02 and S02 Monitoring. Second
International Clean Air Congress of the International
Union of Air Pollution Prevention Associations,
Washington DC.
256
-------
Weber, J.B. 1977. The Pesticide Scorecard. Environmental
Science and Technology, 11:756-761.
White, H.J. 1968. Characteristics of Particulate Matter.
Industrial Water and Wastes Magazine, 48-56.
Wilhm, J.L.
1967. Comparison of Some Diversity Indices Applied to
Populations of Benthic Macroinvertebrates in a
Stream Receiving Organic Wastes. Journal of
the Water Pollution Control Federation,
39(10):1673-1683.
1970. Range of Diversity Index in Benthic
Macroinvertebrate Populations. Journal of the
Water Pollution Control Federation,
42(5) -.R221-224.
Yang, R.T., M.S. Shen, and M. Steinberg. 1978.
Fluidized-Bed Combustion of Coal with Lime Additives:
Catalytic Sulfation of Lime with Iron Compounds and Coal
Ash. Proceedings of the Fifth International Conference on
Fluidized Bed Combustion, Volume 2. The Mitre
Corporation, Washington, DC.
Yaverbaum, L. 1977. Fluidized Bed Combustion of Coal and
Waste Materials. Noyes Data Corporation, Park Ridge, New
Jersey.
257
-------
258
-------
APPENDIX A
GUIDELINES FOR DEVELOPING A QUALITY ASSURANCE
PROGRAM FOR AMBIENT MONITORING ASSOCIATED
WITH ENERGY TECHNOLOGIES
INTRODUCTION
New types of energy technologies such as FBC of coal,
in-situ oil shale conversion, and coal gasification or lique-
faction require different monitoring strategies. These
strategies, which are employed to prevent adverse effects on
the environment, may include ambient source-oriented moni-
toring to determine the environmental concentration and
fate of potentially dangerous byproducts (Dalton«Dalton'
Newport, 1978). The data derived from ambient source-oriented
monitoring may be used by the EPA to establish standards and
regulations, provide dose estimates of toxic materials and
possible health effects, and support enforcement activities.
For these uses, the data generated must be accurate, precise,
and legally and scientifically defensible.
Much environmental monitoring information is currently
available from a wide range of federal, state, and local
agencies, corporations, individuals, and special interest
groups. However, in some cases, data for which considerable
time, money and effort had been expended have been found
virtually useless because of the lack of validity, accuracy or
completeness. Many of these inadequacies are directly attri-
butable to the lack of a defined and implementable quality
assurance program. To correct this problem, the EPA now re-
quires a quality assurance program as an integral part of any
research or monitoring activity.
Quality assurance is a system for integrating the quality
planning, quality assessment and quality improvement efforts
of an organization. The system is designed to assure the out-
put of a quality product—in this case, data that are valid,
accurate, precise, complete, and comparable to data generated
on other projects or collected at different times on the same
subject.
Quality assurance includes both the operational activ-
ities associated with production of a quality product (quality
control) and the administrative activities required to support
quality control and promote implementation of standards of
259
-------
quality. Quality control includes routine checks of normal
operating procedures such as periodic calibration of instru-
ments, performing duplicate analyses, analysis of split
samples and spiked samples, examination of data for valida-
tion, and maintenance of equipment. Administrative activities
include designing the quality assurance program, establishing
performance standards, and periodically verifying performance
of quality control activities.
The latter typically includes administrative evaluation
of quality control and associated activities and an indepen-
dent audit of the entire quality assurance program. The in-
dependent audit is often performed by a representative of a
regulatory agency. Although the independent auditor is not a
member of the organization conducting the monitoring, the
auditing activity is considered part of the quality assurance
system.
OBJECTIVES
The objectives of a quality assurance program are to:
(Smith and Wagoner, 1974)
• Minimize systematic and random variability in the
measurement and data collection processes.
Provide for prompt detection and correction of
conditions that would contribute to the collection
of poor quality data.
Collect and supply information to describe the
quality of the data.
Evaluate the overall adequacy of the data-gathering1
process as it affects data quality.
To accomplish these objectives, a quality assurance pro-
gram must have several components. These required components
are divided into two general groups, each of which includes
several elements. The quality assurance activities associated
with each element may involve both administrative and opera-
tional activities.
The elements are:
• Organization
Quality assurance plan
Quality assurance coordination
Training of personnel
Procurement and facilities development
Documentation control
- Auditing
260
-------
Documentation of Operations and Procedures
Sample site selection
Sample collection
Analytical methods
Data reporting
Data validation
Test and instrument calibration
Preventive maintenance
ORGANIZATION
Quality Assurance plan
To produce a quality assurance program that satisfies the
3 of data users, management, and the operators, a plan
needs of data users, management, and the operators, a plan
should be provided by management. A quality assurance plan
outlines the quality objectives of the organization and the
lines of responsibility and manner of interactions between
management and operations personnel. The activities required
to achieve the quality objectives of the organization should
be listed. A quality assurance coordinator should be desig-
nated and authorized to implement the quality assurance plan.
Management must visibly support the program at all levels of
the organization, and commit the resources necessary to imple-
ment the quality assurance activities.
The quality assurance plan must:
Outline the administrative structure of the labora-
tory, including an organizational chart.
Define the accountability of all personnel whose
actions have an impact upon data quality.
• Define of the duties of the quality assurance coor-
dinator with respect to production of quality data,
including clear lines of authority.
• Outline the plan to train personnel who produce the
data.
Describe how and by whom equipment will be procured.
Establish facilities suitable for data gathering
and analysis activities.
Quality Assurance Coordination
The overall responsibility for coordination and imple-
mentation should be assigned to one person, the quality
261
-------
assurance coordinator. He will serve as the advisor to man-
agement in developing the quality assurance program and as
liaison between operations supervisors and management for all
actions having a bearing on the quality of the data produced
by the organization. In no case should the coordinator be
subordinate to supervisors having direct control over sampling
and analytical operations; rather, the supervisors and ana-
lysts should actively assist the coordinator's efforts in
implementing the program on a daily basis. Figure A-l il-
lustrates the activities associated with a quality assurance
program and the possible interactions of the coordinator with
the analytical and supervisory personnel.
A major function of the quality assurance coordinator is
to provide evaluation reports to all levels of the organiza-
tion. The reports may include basic statistics describing the
validity of the data, information on trends, corrective
actions which are required or were taken, and evaluation of
performance of the laboratory. The frequency of the reports
and their content depend on the organizational level to which
they are being directed. Reports to a laboratory supervisor
may be on a weekly basis, and include the specific information
required to identify any problems and suggest corrective
action. Reports to a manager having responsibility for pro-
duction as well as monitoring may be necessary only semi-an-
nually and include information on trends and evaluation of
performance.
Another function of the quality assurance coordinator is
to assure that standards and quality control samples are
properly obtained or prepared and analyzed in a manner repre-
sentative of normal operating conditions. A third function is
to insert controls to assure that the data-gathering activ-
ities meet the stated requirements for accuracy and precision.
The quality assurance coordinator for a facility con-
ducting environmental monitoring maintains contact with per-
sonnel of state and federal regulatory agencies to assure that
changes in requirements and regulations are implemented on
schedule. The coordinator may also discuss laboratory evalua-
tion with the regulatory agencies and coordinate intraand
inter-laboratory auditing of performance of the laboratory.
Training of Personnel
The knowledge of the operators and the techniques that
they use to perform their tasks will define or control the
ultimate quality of the data. Therefore, training of the
operators is important. Training goals serve to assure that
the output is accurate and precise, that each operator per-
forms a given task in the same manner, that operator bias is
reduced and controlled, and that the operator is a manager of
262
-------
Statistical
Analysis of Data'
Supervisor
and
duality Assurance
Coordinator
Operator
and
Supervisor,
Procurement
Qua)ity Control
Figure A-1. Quality assurance elements and responsibilities
(The Quality Assurance Wheel).
(EPA. 1978)
263
-------
the measurement system rather than a part of it. Controlling
operator bias so that it is constant for all operators allows
different operators to obtain the same information for a given
sample and measurement system.
To meet training goals, the following elements are re-
quired: ,
Develop a programmed training procedure that in-
cludes not only the mechanics of the procedures but
also the performance standards.
Define the reasons for and benefits of standards of
quality; explain why high quality is essential and
what constitutes acceptable quality.
• Determine the proficiency of the operator and ac-
ceptability of the operator's output through the
use of control samples and quality standards.
Verify the accuracy of the operator's output and
correct defects.
Following the steps outlined above will provide necessary
direction to the personnel, accentuate the need for quality
control as an integral part of the data collection process,
and provide a basis for routine quality control and systematic
checks on the output of the operators. The mechanics of the
training process might include: 1) observing the procedure
performed by an experienced analyst, 2) reviewing the theo-
retical and operational steps involved in a procedure, and 3)
performing "hands-on" analysis under direct supervision, fol-
lowed by independent operation. Accuracy of the data obtained
in the last phase should be immediately verified and the
operator's understanding of the theory and techniques involved
evaluated.
The training program should include both old and new per-
sonnel, with periodic assessments of additional training
needs. The program might include formal classroom training as
well as the on-the-job training outlined above. The formal
training might involve general theory and applications taught
at colleges and technical schools, or more specific training.
such as short courses and workshops offered by the EPA and
other groups.
The training courses offered by the EPA cover a wide
spectrum of techniques including sampling, analysis, data
reduction, and statistics applicable for both air and water
monitoring. Information on the courses available can be ob-
tained from:
264
-------
National Training and Operational Technology Center
U.S. Environmental Protection Agency
26 West St. Clair
Cincinnati, Ohio 45268
Registrar, Air Pollution Training Institute
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Procurement and Facilities Development
Procurement of equipment and materials and development of
a suitable facility are important functions, and will involve
management, the quality assurance coordinator, and the opera-
tions supervisors. This phase of the program will assure that
the operational tasks can be performed in an efficient, effec-
tive manner which is consistent with the quality goals of the
organization. The prime criterion for acquisition of equip-
ment is that it must be suitable to perform the intended
tasks. In general, the equipment needs are dictated by the
parameters being measured; the methods used to measure the
parameters; and the sensitivity, reliability, and accuracy re-
quired.
The quality assurance plan should: 1) describe the
criteria to be used in selection and acceptance of equipment
and material used in the monitoring program, 2) define re-
sponsibility for ordering equipment, maintaining the equipment
inventory, and determining whether the equipment meets the ac-
ceptance criteria, and 3) specify the degree of control that
the operations supervisors and the quality assurance coor-
dinator have over the selection of equipment. It is the
responsibility of the quality assurance coordinator to see
that the criteria for equipment acquisition are met and are
consistent with the quality goals of the organization.
Documentation Control
Documentation control is the process which describes the
route and mechanism for disseminating information through the
organization. All procedures for collecting, analyzing, and
distributing data and information must be thoroughly docu- .
mented and readily available. The quality assurance plan
should also detail the mechanism for establishing and revising
procedures. The details to be covered in the plan concerning
revisions or additions to established documents include:
The lines of responsibility for making the revi-
sions.
A distribution plan to assure effective implementa-
tion of the revisions or additions.
265
-------
• The mechanism to be used in validating the revi-
sions or additions.
Auditing Procedures
Each monitoring system must be audited to evaluate the
quality assurance program and the quality of data produced by
the monitoring program. Two types of audits can be performed,
performance audits and system audits. Performance audits are
quantitative assessments of data quality; system audits are
qualitative assessments of data quality.
Performance audits are checks by an independent auditor
to evaluate the quality of data produced by the total sampling
and analysis system. The following are examples of per-
formance audit procedures commonly used:
• Observation of sampling and transportation tech-
niques and procedures, and use of calibrated meters
or reference standards to check the performance of
sampling equipment.
• Observation of actual laboratory analytical tech-
niques, and providing standards or reference
samples for analysis.
Observation of data-processing methods, spot check-
ing calculations and data-validation procedures by
insertion of a set of dummy data into the pro-
cessing system, and reviewing the validated data.
Performance audits may be conducted by several levels of
personnel. These may include checks of:
An analyst's output by another analyst or operator
• The output of one or more analysts by a supervisor
• The output of the data production process by the
quality assurance coordinator
The output of the data production process by an
independent auditor
The quality assurance plan should incorporate all levels
of auditing, and describe the frequency at which the audits
are to be performed. In general, the auditor will be working
from a checklist to assure that all pertinent factors are con-
sidered. Items on the checklist may be simple, e.g., checking
a balance with approved weights; or more complicated, e.g.,
checking continuous air pollutant monitors for each variable
which may affect the output. The checklist should identify
266
-------
the critical variables for each measurement or parameter, and
the control limits for passing the audit.
A systems audit is an on-site inspection and review of
the quality assurance program as it is applied to the total
measurement system. The quality assurance plan is the basis
for the systems audit. The questions to be answered in the
system audit include:
Is the plan sufficiently detailed, and is it opera-
tional?
• Are sample collection and analytical procedures
adequately described and available for use by ap-
propriate personnel, and are they being followed?
• Are instrument and test calibration procedures
available to appropriate personnel and' are they
being followed?
Are calibration results available and do they
establish traceability to standards of higher
quality (National Bureau of Standards or equiva-
lent)?
• How frequently are control charts and quality con-
trol data reviewed?
Are data quality problems handled quickly and ef-
ficiently?
• Does the laboratory adhere to the preventive
maintenance schedule outlined in the quality as-
surance plan?
• Are the results of interlaboratory tests and other
quality assurance analyses examined and used for
improvement of operations?
A usable system audit checklist and suggested method for
grading quality assurance program performance is available
(Buchanan, 1976) .
An essential part of a quality assurance program for any
monitoring activity is periodic performance and system
audits. These audits are to be conducted by an organization
or agency independent of the organization conducting the moni-
toring. Under present EPA guidelines, individual states
assume the roles of administrator and inspector to assure that
laboratories within their jurisdiction are performing ade-
quately. It is therefore likely that a state agency will at
least partially fulfill the requirements of an independent
267
-------
auditor. The state agencies often perform the two types of
audits simultaneously.
It is recommended that an independent audit of monitoring
and quality assurance programs associated with energy produc-
tion and conversion facilities be performed at least once a
year. If the state agency which would be responsible for the
audits cannot conduct yearly audits, the organization con-
ducting the monitoring should contract an outside firm ap-
proved by the appropriate regulatory agency to conduct the
audits.
DOCUMENTATION OF OPERATION AND PROCEDURES
The following discussion reviews the major operational
elements of an ambient monitoring program. The major elements
of an extractive monitoring system are shown in Figure A-2.
Errors in any of the elements can lead to erroneous data.
Quality control efforts and methods, e.g., procedures documen-
tation, should serve to eliminate sources of error. Some of
the important factors of each element which may affect the
quality of the data are discussed. Thorough documentation of
sampling, analysis, data processing, maintenance and all other
procedures is essential. Many methods and guidelines are
available for reference and methods selection. The documenta-
tion should describe how various methods are used in the moni-
toring program.
Sample Site Selection Procedures
Selection of representative sample sites is a major
factor in assuring that ambient monitoring data properly
reflect the environmental impacts of pollutants from traceable
sources. Efficient and cost-effective monitoring requires the
selection of the fewest number of sample sites needed to ade-
quately characterize ambient conditions. Determining the num-
ber and locations of the sampling sites requires a survey of
the affected area. Among the points to be considered in the
survey are:
• The parameters to be monitored
Topography of the area
• Meteorological conditions and variability
Location and types of emissions
Location, volume of flow and direction of flow of
surface and ground waters
• Population density of the affected area
Types of downstream water, air and land uses.
To select sample sites and design the monitoring program,
it is necessary to obtain detailed information concerning the
above criteria. The selection of the number, location and
268
-------
VO
Site Selection
Error
o
f
w
Samp] ing
Error
o
s
^
/2 2222
Measurement
Method Error
0
j
m
i
Reference
Sample
Error
o
~ \ / ° + o +o + o +o
t V f s m r d
r
w
Data Handl ing
Error
o
d
(Recording, Processing
Presentation)
o =
Figure A-2. Major components of system errors in an extractive monitoring system.
(Total error Is the sum of the Individual terms.)
(Environmental Science and Technology. 1078)
-------
type of sampling stations is a complex problem; each sampling
network must be developed individually. In general, simple
and homogenous environments require relatively few sample
sites, whereas complex and heterogeneous environments require
a greater number to adequately characterize ambient condi-
tions.
For ambient surface water monitoring, the number and
location of sampling sites may be affected by the number of
known and potential discharge points associated with any
facility to be monitored. In general, at least one upstream
and one downstream site relative to a discharge are neces-
sary. The upstream sample site should be located to represent
ambient stream conditions. Selection of downstream sample
sites is based on the location of the mixing zone of the
effluent stream in the receiving water. In some cases, the
downstream site for a source may constitute an upstream site
for the next downstream source. Thus, information may be
required on the rate of flow, density, and temperatures of
both the effluent and the receiving stream. Cross-sectional
transect profile analysis should be performed to determine the
specific location of sample sites if a source is point or dis-
crete non-point in nature and if grab samples are to be col-
lected.
Groundwater monitoring may be required for in situ pro-
cesses and other situations in which an aquifer is likely to
be contaminated by pollutant-bearing surface runoff. Ground-
water should be sampled from monitoring wells designed and
drilled for that purpose. However, existing wells are often
used with satisfactory results. The location of the wells
should be determined based upon a hydrogeologic analysis of
the site. This may include drilling test wells and performing
standard tests to determine water quality, flow direction, and
aquifer characteristics. Details of the tests and determina-
tion of groundwater flow characteristics in an aquifer are
documented (Walton, 1970; Todd, 1967; Davis and Dewiest,
1966). It is critical to locate water quality monitoring
wells on the basis of the hydrologic analysis to assure that
the pollutant pathway is being monitored.
The meteorologic conditions of an area affect the dis-
persion of air pollutants and consequently the location of
sampling equipment needed to monitor ambient air quality. The
National Weather Service currently obtains meteorological data
which may be adequate to support ambient monitoring programs.
Additional meteorological data might be required for areas
where existing meteorological data are insufficient or un-
representative of local conditions. The need to collect
site-specific data may arise from peculiarities in local emis-
sion sources and topographical features such as hills,
valleys, trees, lakes, and urban heat islands.
270
-------
For certain monitoring purposes/ the placement of probes
or sampling equipment is of prime importance for obtaining the
desired data. The Ambient Monitoring Guidelines for Preven-
tion of Significant Deterioration (PSD)(EPA, May 1978a) speci-
fies probe placement criteria in an effort to minimize effects
9f topographical features and direct sources of pollutants
upon measurement of ambient conditions. Table A-l summarizes
the specific probe siting guidelines for the PSD ambient moni-
toring program.
Two quality assurance steps must be observed with regard
to the selection of sampling locations:
Data records that justify the selection of each
sample site should be maintained and open for in-
spection by appropriate auditing personnel or
agencies. The data should include detailed maps
and photographs of each sampling site selected.
Following preliminary and/or final sample site
selection, an independent auditor should verify the
adequacy of the selected sample locations in satis-
fying the objectives of the ambient monitoring
program. The independent auditor should be a
recognized authority in the field of sample site
selection, and may be either a private consultant
or a reprentative of an appropriate regulatory
agency.
Sample Collection Procedures
Generally, each major sampling category (air, surface
water or groundwater) will have a separate protocol. The
procedures must specify:
Types of samples collected and the techniques used
to collect them
Numbers of samples to be collected and the fre-
quency of sampling
Amounts or volumes of samples to be collected and
methods of preservation
Means of sample transfer and storage and holding
time requirements
• Methods for permanent identification of samples
Procedure to assure chain of custody
271
-------
TABLE A-l. SUMMARX OF PROBE SITING CRITERIA
(EPA, May, 1978a)
Pollutant
Height above
ground,
Dieters
Distance from supporting
structure, meters
Vertical
Horizontal*
Other spacing criteria
TSP
2-15
T—- 15
-^r
a. >0 meters from trees
b. Distance from sampler to obstacle, such aa
buildings, must be at least twice the height
the obstacle protrudes above the sampler
c. Unrestricted airflow in 3 of 4 cardinal wind
directions
d. Ho furnace or incineration flues should be
nearby**
e. Spacing from roads varies with height-of
monitor ^
a.Probe on roof or intermediate height on a
building must be <0.8 of the mean height of
the building in the neighborhood
b. >20 meters from trees
c. Distance from sampler to obstacle, such as
buildings, must be at least twice the height
the obstacle protrudes above the sampler
d. Unrestricted airflow in 3 of 4 cardinal wind
directions
e. No furnace or incineration flues should be
nearby**
to
-O
10
CO
3+1/2
a. >35 meters from roadway
b. Unrestricted air flow in 3 of 4 cardinal wind
directions
3-15
>1
a, >0 meters from trees
b. Distance from sampler to obstacle, such as
buildings, must be at least twice the height
the obstacle protrudes above the sampler
c. Unrestricted airflow in 3 of 4 cardinal wind
directions
d. Spacing from roads varies with roadway traffic
NO,
3-15
>1
>1
a. >20 meters from trees
b. Distance from sampler to obstacle, such as
buildings, must be at least twice the height
the obstacle protrudes above the sampler
c. Unrestricted airflow in 3 of 4 cardinal wind
directions
d. Spacing from roads varies with roadway traffic
When probe is located on rooftop, this separation distance is in reference to walls, parapets, or penthouses
located on the roof.
Distance is dependent on height of furnace or incineration flue, type of waste or fuel burned, and quality of
fuel (sulfur and ash content). This is to avoid undue influences from minor pollutant sources.
-------
Procedures for verification of data from continuous
monitoring instruments
The primary purpose of a sample system is to collect
representative samples. Grab samples from environmental media
generally provide neither average concentration data nor
variations in concentration, and should be avoided where pos-
sible. Automated sampling or monitoring equipment is usually
necessary to quantify variations in environmental quality
caused by variations in source effluent quality. In some
types of monitoring programs, however, information derived
from grab samples may be sufficient to satisfy the program ob-
jectives. Samples collected by automated equipment may be
collected in relation to time/ or in proportion to flow.
The quality assurance points which must be documented in-
clude:
Methods for calibrating the timing and volume con-
trol sequences
The appropriate analytical quality control and
quality assurance points if the sampler is also an
analyzer
Variations allowed for the environment in which the
sampler is located and methods for controlling them
The nature, number and frequency of fluid samples
to be submitted by the sampling personnel to the
laboratory including the responsibility for and
methods of preparation of such samples
Procedures for maintaining all records
Acceptance criteria for evaluating the results of
blind sample analyses, and the variation in sampler
operation
The volume of sample to be collected depends on both the
parameter being measured, and the measurement technique. For
example, measurement of metals by flame atomic absorption may
require a liter of sample, whereas flameless methods might
require only 20 ml of sample.
The preservation techniques to be used also depend on the
nature of the sample and the parameters to be measured. The
requirement may range from simple refrigeration to addition of
several reagents. For example, phenol samples require the
addition of copper sulfate and phosphoric acid and subsequent
refrigeration.
273
-------
The "chain of custody" is a specified regimen of sample
identification, handling, and transfer procedures which must
be observed if data validity is to be assured. Where it is
known that samples are being collected as part of an enforce-
ment action or other legal proceeding in which data and
information could be presented in court, rigorous adherence to
chain of custody procedures is required. The chain of custody
procedures to be 'employed should be those developed by the
National Enforcement Investigation Center (EPA, May 1978b).
The main points of the chain of custody procedures are:
Complete documentation of sample and data posses-
sion is required from collection to final utiliza-
tion. A sample is in your custody if:
It is in your physical possession,
It is in your view after being in your pos-
session,
It was in your possession and you locked it
in a secure area.
The smallest number of people possible should
handle the samples in the field, during transfer,
and during analysis.
Each person handling the samples or data is re-
sponsible for their care and custody until properly
transferred or dispatched.
Sample identification tags and the chain of custody
form must be attached to the sample at the time of
collection. The information on the tags should:
State sample number, time sample was taken,
source of sample, type of sample, preserva-
tives used and analyses required
State the name of the sample collector and
any witnesses
Be filled out in permanent (waterproof) ink
Blank samples with and without preservatives may be
necessary as a check on possible contamination.
Maintenance of log books is required. The infor-
mation recorded shall be that which is necessary to
establish the validity and chain of custody of the
samples and subsequent data. It includes all
sample data, field measurements, calibrations per-
formed, and any other pertinent information.
• The chain of custody record shall be signed and
dated by each person who handles the sample, indi-
cating the time it came into and left his posses-
sion.
274
-------
Deviations from the methods specified in the qual-
ity assurance plan must be noted and justified.
It is highly recommended that the NEIC chain of custody
procedures be employed for routine sampling and analysis even
if it is believed that the data will not be used in a legal
proceeding.
Analytical Procedures
All analytical procedures used to measure the concentra-
tion of contaminants in an ambient monitoring program shall be
thoroughly documented. The quality assurance requirements for
documentation include:
• Documentation of analytical method
Description of methods for recording and main-
taining analytical data
Description of the quality control measures to be
observed for each parameter measured
The universe of potential pollutants from the energy
technologies lists over 1,000 elements and compounds (Murthy
and Henschel, 1978). For some of the major pollutants and
those for which standards are in effect, the EPA has published
or accepted official methods. For the many other possible
pollutants, however, no official method exists, although
numerous tentative procedures are available. Some official
and accepted methods and sources are listed in the Biblio-
graphy at the end of this Appendix.
'. If an official or tentative EPA method exists to measure
the concentration of a parameter, that method shall be used.
For those parameters for which no official or tentative method
exists, one will have to be developed, and the following in-
formation describing its suitability must be available:
A detailed, step-by-step description of the tech-
niques used for the analysis
A list describing all specialized or unusual equip-
ment used in the analysis
The sensitivity of the analysis and the limits of
detection for the parameter
• The accuracy and precision which should be ob-
tainable under routine operating conditions
• A description of any interferences and how they
would affect the accuracy, precision or sensitivity
of the analysis, with particular attention given to
the interferences likely to be found in typical
samples
275
-------
A list of the references used to develop the tech-
niques.
Each laboratory performing analyses is to keep a written
account of each method as it is actually performed. Any ir-
regularities or differences in the method used as compared to
an official method must be noted and justified to the extent
that the accuracy,i precision or sensitivity of the test is af-
fected. Criteria for comparison of new or alternate test
procedures may be found in Guidelines Establishing Test Proce-
dures for the Analysis of Pollutants (Federal Register,
December 1976).
All analytical or experimental data are to be recorded
permanently. Bound logbooks are preferred for recording and
permanently storing data which is manually produced. Computer
storage is often preferred for automated monitoring systems
because of the volume of data produced. The data should be
recorded in such a fashion that the work performed is clear to
anyone wishing to check the results or the manner in which
they were obtained. At a minimum, the information in the log-
book must include:
• The date the analysis took place
• . Complete description or identification of the
sample being analyzed
• Parameter being analyzed
• Analytical method used
Results of analytical standards or calibrations
Calculations
• Volume of sample used
• Dilutions or concentrations required
Any difficulties or problems encountered during the
analysis and a description of how they were dealt
with (If the difficulties would affect the
accuracy, precision or recovery of the parameter,
the steps performed to assess the degree of dif-
ficulty must be described.)
• The results of duplicate, replicate or check
samples must be noted
In general, the analytical records must be retained a
minimum of three years and must be made available to the
agencies or contractors responsible for auditing the results.'
The quality control measures required during analysis depend
to a degree on the method being used to measure pollutant
concentrations. For instrumental methods, the instrument
should be calibrated with a minimum of three standards. The
concentration range of the standards must cover the expected
concentration of the parameters being measured. Some instru-
mental air monitoring methods require daily or weekly zero and
span calibrations and quarterly six-point linearity checks.
276
-------
For instrumental and non-instrumental methods of analysis,
determination of precision and accuracy for all parameters is
required.
Precision is the reproducibility among replicate observa-
tions. The precision may be based on measurement of duplicate
samples, or on span checks for instrumental measurement of
gaseous air pollutants. The measurement of precision must
cover the range of concentrations expected for each parameter,
the variety of interferences likely to be encountered, and the
different sample types analyzed. The sample types might in-
clude contaminated and uncontaminated surface and ground
waters, and contaminated and uncontaminated air samples.
Precision is stated as the standard deviation of the replicate
observations, and must be presented in terms of the concen-
tration of the parameter for each type of sample measured.
Accuracy, the difference between observed values and
known or actual values, can be determined by adding known
amounts of a contaminant to a sample, and calculating the
percentage recovery for that parameter. In some cases,
accuracy can be measured by direct analysis of a standard
reference material or by analysis of a synthetic sample that
contains known amounts of a contaminant in a matrix approxi-
mating that of samples. Accuracy should be determined at
different levels of pollutant concentrations, and on samples
that cover the full range of expected sample types and inter-
ferences. Table A-2 summarizes some of the quality control
samples available from the EPA.
Round-robin studies, in which many laboratories analyze a
particular parameter by a specified method, are often used to
determine accuracy and precision of analytical methods. This
type of analysis serves two functions: 1) determines which of
the methods tested for a given parameter provides the highest
degree of accuracy and precision, and 2) determines the ac-
curacy and precision attainable under routine operating condi-
tions for a given method. The accuracy and precision state-
ments from round-robin studies are generally regarded as the
standard of performance for an analytical method.
Quality control charts are a method of graphically
presenting the precision and accuracy data and comparing them
against established limits of precision and accuracy for a
given test and set of operating conditions. The control
charts can readily identify trends in an analysis and deter-
mine when an analysis is out of control. Quality control
charts are recommended as the method for recording and inter-
preting precision and accuracy data generated in monitoring
programs associated with energy production and conversion
facilities.
277
-------
TABLE A-2. QUALITY CONTROL SAMPLES AVAILABLE FROM THE EPA ENVIRONMENTAL
MONITORING AND SUPPORT LABORATORY, CINCINNATI
Nutrient Analyses - Two Levels
Nitrate-N Ammonia-N
Kjeldahl-N Total Phosphorous
Orthophosphorous
Demand Analyses
BOD
COD
TOG
£J Mineral Analyses -
- Two Levels
Trace Metals Analyses
Aluminum
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Copper
Iron
Three Levels
Lead
Mercury
Manganese
Nickel
Selenium
Vanadium
Zinc
Two Levels
Sulfate
Chloride
Fluoride
PH
«> Sodium
Potassium
Calcium
Magnesium
Alkalinity/acidity
Total Hardness
Total Dissolved Solids
Specific Conductance
Mercury Analyses - Three Levels
Organic
Inorganic
Volatile Organics
Nine Compounds
Non-Filterable Residue
Nitriloacetic Acid - Four Levels
Polychlorinated Biphenyls
Arochlor 1254
Arochlor 1016
Linear Alkalate Sulfonate
-------
Additional information on quality control methods and the
use of statistics in controlling analytical processes are
found in many documents (EPA, 1979; EPA, 1976; Federal
Register, 1978).
Data Reporting Procedures
The reporting of data occurs at several levels within and
outside a given organization. The methods of reporting and
the content of the report required for each level should be
detailed in the quality assurance plan. The analysts or
operators who collect and produce data may report to:
Immediate supervisors
Operations managers
Quality assurance coordinator
Regulatory agencies
Regional or national data banks
The format and type of data reported should be geared to
the needs of each level. In general, preprinted report forms
should be used since they present the data in a consistent
format and also serve as an instant reminder of the data
necessary to satisfy the user's needs. Quality assurance for
data reporting includes consistent, routine, and systematic
checking of data at all levels to assure that the data being
reported are the same as those obtained in the data-gathering
activities. Among the steps required are:
Checks of accuracy and precision of the data trans-
cribing process where information is taken from
strip charts or other continuous monitoring out-
puts.
• Systematic interaction among analysts and super-
visors to assure that the data leaving the labora-
tory are the same data obtained from analysis of
the samples.
Duplication of portions of raw or partially pro-
cessed data transmitted over telephone or other
telemetry lines, as a check on the precision of
transmission.
• Determination of the accuracy of transmission by
checking the transmitted data against the data ob-
tained by the sensor.
Checks of data put onto computer files or into
regional or national data banks for accuracy by
retrieval and rechecking, preferably against the
laboratory data. Data entered into a computer file
279
-------
such as STORET must be checked for accuracy each
time a data set is entered. All data points must
be rechecked, unless it is demonstrated that the
accuracy attained warrants a reduction, but in no
case should less than twenty percent of the data be
rechecked.
In addition, results of the quality control tests or
other statements concerning uncertainties in the results
should be included to allow the data users to make valid judg-
ments on the utility of the data. Routine quality control
data are generally submitted to the appropriate state agency,
or applicable permit granting authority. The frequency of
reporting quality control data to state regulatory agencies
will depend on the nature and requirements of the monitoring
program. For ambient monitoring associated with energy con-
version and production facilities, it is recommended that
quality control data for ambient monitoring programs be re-
ported to the appropriate state or permit granting agencies at
least once every six months. Results of interlaboratory and
collaborative quality control/assurance programs are to be
reported to the agency coordinating the program.
Data Validation
Data validation is a quality control process in which
data are screened and accepted or rejected based on a set of
specific criteria. The criteria which may be used to judge
whether data are to be accepted or rejected include:
Special conditions unique to the measurement pro-
cess.
• Specifications for calibration of tests and instru-
ments.
Results of statistical tests which can indicate
whether or not the analysis was in control.
• Relationship of pollutants (e.g., high concentra-
tions of sulfur dioxide and ozone cannot coexist,
and simultaneous concentrations should be
immediately suspect).
• Time-related variations expected for a given para-
meter. .
The data validation process normally operates at the
analyst-supervisor level, and includes routine examination of
data for extreme values. Extreme values may be located by
comparing current data to prior data obtained at a particular
sampling site. The data used to locate extreme values may be
the maximum and minimum concentrations for prior data from a
given sampling site, and the criteria may also be specific for
maximum and minimum concentrations for a given time period
such as season, day or hour. Data which are from one to
280
-------
several orders of magnitude higher than the normal range of
concentration are easily noted as possible spurious values.
However, if a spurious value is only marginally higher or
lower than the norm, spotting it could require specific know-
ledge of the environment from which that sample was taken. In
either case, factors which could affect the concentration of
that parameter should be checked. These may include:
Sampling technique; was bottom sediment or floating
scum picked up with the sample?
Instrument operating conditions; was the proper
amount of sample used, was the instrument in the
proper operating mode?
Calculations; were the proper factors used in
calculation of concentration?
Additional information, e.g., plant operating data
may indicate that the plant's waste treatment sys-
tem discharged high amounts of organic carbon dur-
ing the time period in question.
If analytical errors are found, the data may be ruled in-
valid. If no analytical errors are found the data may still
be ruled invalid if no specific cause of the spurious value
can be found. The actions to be taken if data are ruled
invalid may include:
Rejection of data
• Repetition of measurement
Use of data with special notation
Data collected during a time when accuracy, precision, or
control chart data indicate that the analyses are out of con-
trol are often ruled invalid. The data may still be usable,
providing that the variations applicable to the data can be
determined. These data should be reported with a cautionary
note describing the accuracy or precision of that set.
Whenever a suspicious value is noted, it should not be
accepted or rejected without determining whether there is
cause to do either. Each questionable value should be
checked, and records maintained of such values; if values are
shown to be invalid, the reasons for invalidation must be
documented. Corrective action taken to prevent recurrence
should also be noted. The records are," among other things, a
useful source of information for judging the quality of future
data.
The percentage of data which should be subjected to the
validation process depends on the degree of reliability of the
analytical equipment, and the percentage of past production of
invalid data. As a general rule, ten percent of the data pro-
duced in the monitoring program should be subjected to the
281
-------
validation process. Some air pollution quality assurance
documents recommend that a minimum of seven percent of the
data be validated. ;
Calibration of Tests and Instruments
The methods to. be used for calibration of the tests and
instruments used ini an ambient monitoring program should be
thoroughly documented for each test or instrument used. They
should list the specific standards and equipment required and
the methods used to calibrate the tests. The methods should
also provide specific instructions for obtaining and recording
test data. The frequency and extent of calibration for tests
and equipment required depends on several factors:
Types of equipment or tests being used.
Type of environment in which the tests are being
performed.
• Manufacturers' recommendations.
Level of accuracy required by the monitoring pro-
gram.
Applicable state or federal regulations.
The standards used to calibrate the tests and equipment
must be documented as being traceable to standards provided by
the National Bureau of Standards (NBS) or other primary stan-
dards. Measurements have traceability to designated standards
only if rigorous evidence is produced on a continuing basis to
show that the total measurement uncertainty relative to the
specified standards is quantified (Belanger and Kieffer,
1979).
A specific protocol has been established for determining
the true concentration of working standards for gaseous pol-
lutants. It is to be used when calibrating an air pollution
monitor with standards other than those supplied by the
National Bureau of Standards (EPA, June 1978).
A certified primary standard obtainable from a commercial
source shall be used for calibration when NBS standard
reference materials are unavailable. The material used as a
primary standard must have a known property that is stable and
that can be accurately measured or derived from established
chemical or physical constants (EPA, 1976). Transfer stan-
dards are materials used for routine calibration of field or
laboratory equipment. These standards are calibrated against
higher-level materials such as NBS standard reference
materials or primary standards. Transfer standards used for
calibration of tests and equipment shall be 4 to 10 times more
accurate than the field or laboratory equipment being cali-
brated. For example, if a thermometer used to determine tem-
peratures in the field must have an accuracy of jf 2'C, it must
282
-------
be calibrated against a thermometer with an accuracy of at
least + 0.5'C. Figure A-3 summarizes a coupled system of
analytical methods and reference materials. The function of
each component is to transfer traceability to the next higher
level. This assures measurement compatibility in the system.
Table A-3 lists the standard reference materials avail-
able from the National Bureau of Standards. Many other
reference materials and standards, such as for pesticides, are
available from the EPA.
A detailed calibration plan must be in effect for each
facility providing ambient monitoring data. The plan should
include:
Specified calibration intervals for each test and
analytical instrument, and designations of satis-
factory sources of calibration standards.
A list of the required calibration standards for
each test including the identification numbers for
each standard.
The mechanism used to establish traceability of the
standards used to NBS standard reference materials
or other primary standard.
The operating conditions to be used during the
calibration procedure for each test or instrument.
• A description of the calibration record system,
including the data to be recorded and the manner in
which records are to be maintained and distributed.
The intervals at which calibration of tests and equipment
is performed can determine whether results from a monitoring
program meet accuracy standards. In some cases, calibration
intervals are specified by state and federal agencies; in the
absence of regulations, a preliminary servicing schedule
should be established, based on the operating time of the
equipment. The establishment of calibration intervals is
based upon required accuracy, degree of usage, and stability
of the test or equipment. After evaluation of preliminary
calibration results, the time period can be either lengthened
or shortened as necessary, to provide positive assurance that
the calibration interval is sufficient to maintain the
accuracy of the system.
Preventive Maintenance
A preventive maintenance program increases or sustains
the reliability of the measurement system. The schedule and
activities associated with preventive maintenance must be
clearly defined for each measurement system and its support
283
-------
•Method and Instrument
Development
•Method Evaluation
•Proficiency Testing
•Regulatory Requirements
Qua Ii ty
Control Materials
"True Value"
Definitive Method
c
P r ima ry
V Reference Materials
Reference Methods
R Secondary
reference Materials
Field Methods
Figure A-3. Idealized accuracy-based measurement network,
1979)
284
-------
TABLE A-3. ENVIRONMENTAL STANDARD REFERENCE MATERIALS AVAILABLE
FROM NATIONAL BUREAU OF STANDARDS
(National Bureau of Standards, 1978)
SRM
No.
1609
1658
1659
1660
1661
1662
1663
1664
1665a
1666a
1667a
1668a
1669a
1673a
1674a
1675a
1677b
1678b
1679a
1680a
1681a
1683a
1684a
Type
Analyzed Gases
Oxygen in N2/ 20.95 Mol. %
Methane in air, 0.95 ppm
Methane in air, 9.5 ppm
Methane (4 ppm) and Propane (1 ppm) in air
Sulfur dioxide in N , 480 ppm
Sulfur dioxide in N-, 942 ppm
Sulfur dioxide in N2/ 1497 ppm
Sulfur dioxide in N_, 2521 ppm
Propane in air, 2.8 ppm
Propane in air, 9.5 ppm
Propane in air, 48 ppm
Propane in air, 95 ppm
Propane in air, 475 ppm
Carbon dioxide in N2, 0.95%
Carbon dioxide in N2/ 7.2%
Carbon dioxide in N2, 14.2%
Carbon monoxide in N_, 9.74 ppm
Carbon monoxide in N2, 47.1 ppm
Carbon monoxide in N-, 94.7 ppm
Carbon monoxide in N2, 484 ppm
Carbon monoxide in N2/ 947 ppm
Nitric oxide in N2/ 50 ppm
Nitric oxide in N , 100 ppm
Unit
100 ml
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
Price
($)
120
283
283
283
356
356
356
356
282
282
282
282
282
282
282
282
*
305
305
305
305
*
*
* To be renewed
(Continued)
285
-------
TABLE A-3 (Continued)
SRM
No . Type
1685a Nitric oxide in N , 250 ppm
ft
1686a Nitric oxide in N2/ 500 ppm
1687a Nitric oxide in N2, 1000 ppm
2612 Carbon monoxide in air, 9.5 ppm
2613 Carbon monoxide in air, 18.0 ppm
2614 Carbon monoxide in air, 42.7 ppm
2619 Carbon dioxide in N_, 0.5 mole %
2620 Carbon dioxide in N», 1.0 mole %
2621 Carbon dioxide in N_, 1.5 mole %
2622 Carbon dioxide in N-, 2.0 mole %
2623 Carbon dioxide in N_, 2.5 mole %
2624 Carbon dioxide in N , 3.0 mole %
2625 Carbon dioxide in N_, 3.5 mole %
2626 Carbon dioxide in N_, 4.0 mole %
Other Environmental SPM's
1569 Brewers Yeast
1570 Spinach, trace elements, botanical
1571 Orchard leaves, trace element, botanical
1573 Tomato leaves, trace elements, botanical
1575 Pine needles, trace elements, botanical
1577 Bovine liver
1579 Powdered lead -base paint
1621 Sulfur in residual fuel oil, 1.05 wt. %
1623 Sulfur in residual fuel oil, 0.268 wt. %
1624 Sulfur in distillate fuel oil, 0.211 wt. %
1625 Sulfur dioxide permeation tube, 10 cm
1626 Sulfur dioxide permeation tube, 5 cm
1627 Sulfur dioxide permeation tube, 2 cm
1629 Nitrogen dioxide permeation device
1630 Trace mercury in coal
1632 Trace elements in coal
(Continued)
Price
Unit ($)
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
cyl
50g
60g
75g
70g
70g
50g
35g
100 ml
100 ml
100 ml
ea
ea
ea
ea
50g
75g
327
327
327
270
270
270
353
353
353
353
353
353
353
353
63
77
78
77
77
98
41
40
40
40
85
85
85
88
53
77
286,
-------
TABLE A-3 (Continued)
SRM
NO.
1633
1634
1636
1637
1638
1641
2661
2662
2663
2664
2665
2666
2667
2671
2672
2675
2676
Type
Trace elements in coal fly ash
Trace elements in fuel oil
Lead in reference fuel, four concentrations
Lead in reference fuel, three concentrations
Lead in reference fuel, one concentration
Mercury in water, 1.49 ug/ml
Benzene on charcoal
m-Xylene on charcoal
p-Dioxane on charcoal
1, 2-Dichloroethane on Charcoal
Chloroform on charcoal
Trichloroethylene on Charcoal
Carbon Tetrachlorine on Charcoal
Freeze-dried urine certified for fluorine
Freeze-dried urine certified for mercury
Beryllium on filter media
Metals on filter media
Unit
75g
100 ml
set (12)
set (12)
set (12)
120 ml
set (8)
set (8)
set (8)
set (8)
set (8)
set (8) •.
set (8)
set (2)
set (2)
set (3)
set (2)
Price
(S)
77
77
61
61
61
62
87
87
87
87
87
87
87
82
82
82
*
* To be renewed
287
-------
equipment. The criteria for determining the need for preven-
tive maintenance, and the schedule on which maintenance should
be performed include:
Type of environment in which the equipment is being
used.
Manufacturers' recommendations as found in the in-
struction manuals for each instrument or equipment.
Specific mandates listed in the official or ref-
erence method applicable to the measurement of a
given parameter.
• Operational history of the specific instrument or
equipment.
Preventive maintenance steps, their frequency, and the
derived data should be documented. A record of all routine
and nonroutine servicing of equipment shall be maintained.
The preventive maintenance and other maintenance records
should be used to:
Revise maintenance schedules to maximize time of
operation for each instrument.
Maximize quality of data output.
Minimize loss of data due to equipment failure.
Guide to development of an adequate parts inventory
and schedule of replacement.
Equipment that is used on a continuous basis, such as air
pollution monitors and automatic water monitors and sampling
equipment, commonly requires daily service or performance
checks. The amount of servicing is dictated by the nature of
the equipment, the environmental conditions that can be main-
tained for the monitor, and the degree of accuracy required of
the monitor. Equipment used in the laboratory under con-
trolled environmental conditions may require only periodic
servicing. Service and performance checks, however, should be
routine steps in each analytical procedure. It is recommended
that, where available, service contracts from the manufac-
turers be obtained for at least the major analytical equip-
ment. This will assure that the functional capabilities of
the equipment are maintained.
SUMMARY
A quality assurance program is a system of activities for
integrating the quality planning, quality assessment and qual-
ity improvement efforts of an organization. The objectives of
a quality assurance program are to:
Minimize systematic and random variability in the
measurement and data collection process.
288
-------
Provide for prompt detection and correction of con-
ditions which would contribute to the production of
poor quality data.
Collect and supply information necessary to de-
scribe the quality of the data.
Evaluate the overall adequacy of the data-gathering
process insofar as it affects data quality.
Elements of a quality assurance program fall into two
general groups: organization, and documentation of operations
and procedures. The elements within the organization group
are the administrative activities necessary to manage the pro-
gram, provide the necessary facilities, train the personnel,
and audit the adequacy of the program. The elements of the
second group are the operational activities required for a
monitoring program, such as sample site selection, sampling,
analysis, preventive maintenance and data reporting.
A primary management responsibility is to designate a
person to administer the quality assurance activities of the
organization and coordinate these activities with appropriate
regulatory agencies. The quality assurance coordinator works
with both management and supervisory personnel to develop and
implement training programs, determine propriety of methods,
select equipment, develop facilities, and determine how the
quality of monitoring data should be measured.
The quality assurance plan should document the develop-
ment and implementation responsibilities of each worker.
Documentation of the operational procedures of an ambient
monitoring program are required to assure that:
The sampling network design and sampling techniques
provide representative samples of the ambient
medium.
Sample handling and storage maintain the integrity
and representativeness of samples and meet custody
requirements.
• The analytical methods are applicable and any
modifications are properly justified.
External and internal quality control checks are
performed, and accuracy and precision can be
demonstrated to be within acceptable limits.
Data are reported in a format appropriate for each
user.
289
-------
• Calibration of tests and instruments are performed
at regular intervals and the criteria for cali-
bration are known.
• Preventive maintenance schedules are known and fol-
lowed and documentation of acceptability of in-
struments is available.
Documentation1 of these quality assurance activities is
required to: (1) unify the various elements and personnel
involved in the program, (2) outline the responsibilities and
time frame for performing the various tasks, (3) describe the
action required to eliminate any identified deficiencies, (4)
ensure that valid data are being generated by the monitoring
program and that this validity can be adequately demonstrated
at a later date.
290
-------
REFERENCES
Buchanan, J. 1976. Guidelines for Demonstration Project
Quality Assurance Programs. Industrial Environmental
Research Laboratory, U.S. Environmental Protection
Agency. Research Triangle Park, North Carolina.
Dalton»Dalton'Newport. 1978. Guidelines for
Multimedia Environmental Monitoring of DOE Fossil Energy
R&D Facilities, FE-2495-6A01. Dalton«Dalton*
Newport, Cleveland, Ohio.
Davis and Dewiest. 1966. Hydrogeology. New York: John
Wiley and Sons.
Environmental Science and Technology. 1978. QA Report':
Federal Environmental Monitoring - Will the Bubble
Burst? Environmental Science and Technology,
12(12):1264-1269.
Federal Register.
December 1 Guidelines Establishing Test
1976 Procedures for the Analysis of Pollu-
tants.
August 7 Part II: Air Quality Surveillance
1978 and Data Reporting, Proposed Regulatory
Revisions.
Murthy, K.S., and D.B. Henschel. 1978. Environmental
Assessment of Fluidized Bed Combustion: Status and Re-
sults. Presented at Environmental Assessment of Solid
Fossil Fuel Symposium, 71st Annual AI CheE Meeting,
Miami Beach, Fla., November 1978.
National Bureau of Standards. 1978. Catalog of NBS
Standard Reference Materials. NBS Special Publication
260. National Bureau of Standards, Washington, D.C.
Smith, F., and D.E. Wagoner. 1974. Guidelines for
Development of a Quality Assurance Program. Vol. IV:
Determination of Particulate Emission from Stationary
Sources. National Environmental Research Center, U.S.
EPA Research Triangle Park, North Carolina.
291
-------
Todd, D. 1967. Groundwater Hydrology. New York:
Wiley and Sons, Inc.
John
U.S. Environmental Protection Agency.
1976 Quality Assurance Handbook for Air Pollution
Measurement Systems, Vol. I Principles. Of-
fice of Research and Development. U.S. EPA,
Research Triangle Park, North Carolina.
May
1978a
May
1978b
June
1978
1979
Ambient Monitoring Guidelines for
Prevention of Significant Deterioration. Of-
fice of Air Quality Planning and Standards,
U.S. EPA Research Triangle Park, North
Carolina.
NEIC Policies and Procedures Manual,
National Enforcement Investigation Center,
U.S. EPA Denver, Colorado.
Traceability Protocol for Establishing True
Concentrations of Gases for Calibration and
Audits of Air Pollution Analyzers (Protocol
No. 2) Environmental Monitoring and Support
Laboratory, U.S. EPA Research Triangle Park,
North Carolina.
Handbook for Analytical Quality Control in
Water and Wastewater Laboratories. Methods
Development and Quality Assurance Laboratory,
U.S. EPA. Cincinnati, Ohio.
Uriano, G.A. 1979. The Use of Standard Reference
Materials For Quality Assurance of Environmental Mea-
surements. Proceedings of the National Conference on
Quality Assurance of Environmental Measurements, Denver,
Colorado, November 1978.
Walton, W.C. 1970. Groundwater Resources Evaluation.
, New York: McGraw Hill, Inc.
292
-------
BIBLIOGRAPHY
OFFICIAL METHODS AND SOURCES OF METHODS
FOR SAMPLE COLLECTION, ANALYSIS AND QUALITY ASSURANCE
Annual Book of ASTM Standards. (Part 31) Water. 1975.
American Society for Testing and Materials,
Philadelphia, Pennsylvania.
Biological Field and Laboratory Methods for Measuring the
Quality of Surface Waters and Effluents. 1973. Office
of Research and Development. U.S. Environmental Protec-
tion Agency, Cincinnati, Ohio
Manual of Methods for Chemical Analysis of Water and Wastes.
1979. Office of Technology Transfer, U.S. Environmental
Protection Agency, Washington, D.C.
Methods of Air Sampling and Analysis. 1977. 2nd Edition.
American Public Health Association, Washington, D.C.
NIOSH Manual of Analytical Methods. 1977. 2nd Edition,
Vol. 1-3. National Institute of Occupational Safety
and Health, U.S. DREW, Cincinnati, Ohio.
Quality Assurance Handbook for Air Pollution Measurement
Systems. 1977. Vol. II: Ambient Air Specific
Methods. Environmental Monitoring and Support Labora-
tory, U.S. EPA Research Triangle Park, North Carolina.
Standard Methods for the Examination of Water and Wastewater.
1975. 14th Edition. American Public Health
Association, Washington, D.C.
In addition to the primary sources of analytical
methods, there are numerous other sources of information
usable in establishing and operating an ambient environmental
monitoring program. These sources contain information on site
selection analysis, sample collection, statistics and quality
assurance. They include:
ASTM Standards on Precision and Accuracy for Various
Applications. 1977. American Society for Testing and
Materials, Philadelphia, Pennsylvania.
293
-------
Ambient Monitoring Guidelines for Prevention of Significant
Deterioration. 1978. Office of Air Quality Planning
and Standards, U.S. EPA Research Triangle Park, North
Carolina.
Brown, E.M., and M.W. Skougstad. 1970. Methods for Col-
lection and Analysis of Water Samples for Dissolved
Minerals and Gases. U.S. Geological Survey Techniques,
Water Resources Inc. Book 5.
i
Buchanan, J., and F. Smith, 1976. Guidelines for Development
of a Quality Assurance Program. Vol. XVI - Method for
the Determination of Nitrogen Dioxide in the
Atmosphere. Environmental Monitoring and Support Labo-
ratory, U.S. EPA, Research Triangle Park, North
Carolina.
Duncan, A.J. 1965. Quality Control and Industrial Statis-
tics. 3rd Edition. Homewood, Illinois: Richard D.
Irwin Inc.
Grubles, F.E. 1969. Procedures for Detecting Outlying
Observations in Samples. Technometrics, 11-1:1-21.
Guidelines for Siting and Exposure of Meteorological Instru-
ments for Environmental Purposes. 1976. Meteoro-
logical arid Assessment Division, U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina.
Hamilton, C.E., (Ed). 1978. Manual on Water. American
Society for Testing and Matials, STP 442A, Philadelphia,
Pennsylvania.
Handbook for Analytical Quality Control in Water and Waste-
water Laboratories. 1979. Technology Transfer, U.S.
Environmental Protection Agency, Washington, D.C.
Handbook for Monitoring Industrial Wastewater. 1973.
Technology Transfer, U.S. Environmental Protection
Agency, Washington, D.C.
Kolnsberg, H.J. 1976. Technical Manual for Measurement
of Fugitive Emissions, Upwind/Downwind Sampling Method
for Industrial Emission. Office of Research and Devel-
opment. U.S. Environmental Protection Agency,
Washington, D.C.
Mandel, John. 1964. The Statistical Analysis of Data. New
York: Interscience Publishers.
Minimal Requirements for a Water Quality Assurance
Program. 1975. Environmental Monitoring and Support
294
-------
Laboratory, U.S. Environmental Protection Agency,
Cincinnati, Ohio.
Quality Assurance Handbook for Air Pollution Measurement
Systems. 1976. Vol. 1, Principles. Environmental
Monitoring and Support Laboratory, U.S. Environmental
Protection Agency, Research Triangle Park, North
Carolina.
Rainwater, F.H., and L.L. Thatcher. 1960. Methods of
Collection and Analysis of Water Samples. U.S. Geologi-
cal Survey, Water Supply Paper No. 1454.
Smith, F., and A.C. Nelson, Jr. 1973. Guidelines for
Development of a Quality Assurance Program. Reference
Method for Continuous Measurement of Carbon Monoxide in
the Atmosphere. Office of Research and Monitoring Qual-
ity Assurance, U.S. EPA, Research Triangle Park, North
Carolina.
Smith, F., and A.C. Nelson, Jr. 1973. Guidelines for
Development of a Quality Assurance Program: Reference
Method for Determination of Suspended Particulate.s in
the Atmosphere (High Volume Method). Quality Assurance
and Environmental Monitoring Laboratory, U.S. EPA,
Research Triangle Park, North Carolina.
Smith, F., and A.C. Nelson, Jr. 1973. Guidelines for
Development of a Quality Assurance Program: Reference
Method for the Determination of Sulfur Dioxide in the
Atmosphere. Quality Assurance and Environmental Moni-
toring Laboratory, U.S. EPA, Research Triangle Park,
North Carolina.
Wastewater Sampling Methodologies and Techniques. 1974.
Office of Technology Transfer, U.S. Environmental Pro-
tection Agency, Washington, D.C.
Water Pollution Assessment. 1974. Automatic sampling
and Measurement. American Society for Testing and Mate-
rials, STP582, Philadelphia, Pennsylvania.
Wood, W.W. Guidelines for Collection and Field Analysis
of Groundwater Samples for Selected Unstable Consti-
tuents. U.S. Geological Survey Techniques, Water Re-
sources Inv.
295
-------
296
-------
APPENDIX B
LIST OF LAWS, REGULATIONS, AND EXECUTIVE ORDERS
PERTAINING TO ENVIRONMENT AND SAFETY AND
HEALTH REQUIREMENTS
LAWS
Environment
Cultural Resources Protection
Historic Sites Act of 1935
National Historic Preservation Act of 1966, As
Amended
Archaeological and Historic Preservation Act of 1974
Archaeological Resources Protection Act of 1979
Water Quality and Water Resources
Safe Drinking Water Act of 1974
Federal Water Pollution Control Act of 1972, As
Amended
Federal Water Project Recreation Act of 1965, As
Amended
Water Resources Research Program Act of 1964
Rivers and Harbors Act of 1899
Wild and Scenic Rivers Act of 1968
Water Resources Planning Act of 1965
Soil and Water Resources Conservation Act of 1977
Floodplains, Wetlands, and Coastal Zones
Coastal Zone Management Act of 1972, As Amended
Estuary Protection Act of 1968
National Flood Insurance Act of 1968, As Amended
Wildlife
Endangered Species Act of 1973, As Amended
Pish and Wildlife Coordination Act of 1934, As
Amended
Fish and Wildlife Conservation at Small Watershed
Projects
Public Land, Open Space, and Recreation
National Trails System Act of 1968, As Amended
Wilderness Act of 1964
National Forest Management Act of 1976
Mineral Leasing Act Amendments of 1973
Land and Water Conservation Act of 1965, As Amended
Forest and Rangeland Renewable Resources Planning
Act of 1974, As Amended
297
-------
Oceanography and Marine Environment
Deepwater Port Act of 1974
Marine Protection, Research and Sanctuaries Act of
1972, As Amended
Marine Mammal Protection Act of 1972, As Amended
Fishery Conservation and Management Act of 1976
Air Quality and Noise
Clean Air Act of 1963, As Amended
National:Emission Standards Act of 1967
Noise Pollution and Abatement Act of 1970
Noise Control Act of 1972
Energy
Energy Supply and Environmental Coordination Act of
1974
Trans-Alaska Pipeline Act of 1973
Federal Non-Nuclear Energy Research and Development
Act of 1974
Energy Policy and Conservation Act of 1975
Surface Mining Control and Reclamation Act of 1977
Federal Power Act of 1935, As Amended
The Outer Continental Shelf Lands Act of 1953, As
Amended
Alaska Natural Gas Transportation Act of 1976
The Atomic Energy Act of 1954, As Amended
The Natural Gas Act of 1938, As Amended
Hazardous and Solid Wastes
Resources Conservation and Recovery Act of 1976
Other Environmental Laws
National Environmental Policy Act of 1969
Demonstration Cities and Metropolitan Development
Act of 1966
Safety and Health
Occupational Safety and Health Act of 1970
Toxic Substances Control Act of 1976
Ports and Waterways Safety Act of 1972
Federal Coal Mine Health and Safety Act of 1969, As
Amended
Hazardous Materials Transportation Act
Property
Federal Land Policy and Management Act of 1976
Antiquities Act of 1906
REGULATIONS
Environment
Council on Environmental Quality's Regulations for the
Implementation of the National Environmental Policy Act
Advisory Council on Historic Preservation; Procedures for
Historic Preservation
298
-------
EXECUTIVE ORDERS
Environment
11514, Protection and Enhancement of Environmental Quality
11523, Establishing the National Industrial Pollution
Control Council
11574, Administration of the Refuse Act Permit Program
11738, Providing for Administration of the Clean Air Act
and the Federal Water Pollution Control Act with Respect
to Federal Contracts, Grants, or Loans
11742, Delegating to the Secretary of State Certain
Functions Under the Federal Water Pollution Control Act
with"Respect to Negotiations of International Agreements
11747, Delegating Authority of the President Under the
Water Resources Planning Act
11987, Exotic Organisms
11988, Floodplain Management
11990, Protection of Wetlands
12113, Providing for Independent Water Project Review
12114, Environmental Effects Abroad of Major Federal
Actions
12129, Critical Energy Facility Program
Safety and Health
11807, Occupational Safety and Health Programs For
Federal Employees
Property
IT593, Protection and Enhancement of the Cultural En-
vironment
11643, Environmental Safeguards on Activities for Animal
Damage Control in Federal Lands
11989, Off-Road Vehicles on Public Lands
12088, Federal Compliance with Pollution Control Standards
OTHER RELEVANT EXECUTIVE ORDERS AND GUIDANCE
11490, Assigning Emergency Preparedness Functions to Fed-
eral Departments and Agencies
12044, Improving Government Regulations
OTHER GUIDANCE DOCUMENTS
Environment
Water Resources Council, Guidance for Floodplain Man-
agement
Water Resources Council, Principles and Standards
Council on Environmental Quality Memorandum on Prime and
Unique Farmlands
OMB Circular A-95-Early Coordination
OMB Circular A-109-Major System Acquisition
299
-------
r
TECHNICAL REPORT DATA
//•/CAW read Instructions on the reverie before completing)
1. REPORT NO.
EPA-600/4-81-019
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
MONITORING STRATEGIES FOR FLUIDIZED BED COMBUSTION
COAL PLANTS v
8. REPORT DATE
March 1981
8. PERFORMING ORGANIZATION COOt
7. AUTHOR(S)
A.B. Garlauskas, C.E. Hina, T.T. Blair, M.J. Kangas,
and C.L. Cornett
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Dalton'Dalton'Newport, Inc.
3605 Warrensville Center Road
Shaker Heights, Ohio 44122
11. CONTRACT/GRANT NO.
68-03-2755
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency—Las Vegas, NV
Office of Research and Development
Environmental Monitoring Systems Laboratory
Las Vegas, Nevada 89114
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA-600/07
15. SUPPLEMENTARY NOTES
Richard Bateman, Project Officer, EMSL-LV
16. ABSTRACT
Air and water monitoring strategies for commercial-size Fluidized Bed Combustion
(FBC) coal plants are presented*
This is one of five reports developing air and water monitoring strategies for
advanced coal combustion (FBC), coal conversion (coal gasification and liquifaction),
and oil shale conversion commercial-scale plants. The objective of the five-part
project is to assure that, when any of the coal or oil shale technologies become
commercialized, appropriate and cost-effective monitoring requirements are in effect to
protect the public health and the environment without delaying commercialization.
In this report, air quality monitoring strategies are presented for compliance
audits, and for monitoring fugitive emissions. Water quality monitoring strategies are
presented for leachates from fly ash and other solid wastes and from coal and sorbent
material storage piles, and for pollutants generated by cooling towers, ponds, or
canals.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO THE PUBLIC
19. SECURITY CLASS (Thit Rtport)
UNCLASSIFIED
21. NO. Of PAGES
315
20. SECURITY CLASS (Thispagtj
UNCLAS3TFTEn
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS COITION is OBSOLETE
------- |