EPA-600/2-76-209
July 1976
Environmental Piotecticr. Technology Ss.ies
PERFORMANCE OF EMISSION CONTROL
DEVICES ON BOILERS FIRING
MUNICIPAL SOLID WASTE AND OIL
industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield. Virginia 22161.
-------
EPA-600/2-76-209
July 1976
PERFORMANCE OF EMISSION CONTROL
DEVICES ON BOILERS FIRING
MUNICIPAL SOLID WASTE AND OIL
by
J.B. Galeski and M. P. Schrag
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Contract No. 68-02-1324, Task 40
Program Element No. EHB533
EPA Task Officer: James D. Kilgroe
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
Existing data on particulate emissions from oil-fired electric utility
boilers and from waterwall (steam generating) incinerators firing either re-
fuse or refuse-plus-coal/oil auxiliary fuel were used to estimate particulate
flue gas loadings for combined firing of shredded municipal refuse (MSW) and
oil. Estimates of control device performance were made for several planned
oil-MSW resource recovery systems. On the basis of these estimates, installed
particulate emission controls, designed for coal, are predicted to be signifi-
cantly less efficient for control of particulate emissions from combined fir-
ing of oil-MSW. Anticipated control difficulties result mostly from relatively
high particulate loadings, high flue gas volumes, fine particulates, relatively
low particle density, and relatively high fractions of carbonaceous, low-
resistivity particulate.
iii
-------
CONTENTS
Abstract iii
List of Figures viii
List of Tables. .............. ........ x
Acknowledgments ....................... xiii
Section
1 Summary ............................ 1
Information Acquisition. .................. 1
Control Performance and Cost Correlations. ......... 2
Analytical Model Development ................ 2
Case Studies 3
Recommendations. ...................... 3
Electrostatic Frecipitator Control. . . . 4
Cyclone Control 4
Novel Control Devices ....... 4
2 Introduction. ......................... 5
3 Particulate Emission Data Acquisition and Evaluation. ..... 6
Mass Emissions Data. 7
Uncontrolled Particulate Emissions from Oil-Fired
Electric Utility Boilers. ... 8
Uncontrolled Particulate Emissions from Waterwall
Incinerators. .............. .. 16
Estimate of Refuse Fly Ash from Suspension Firing
of MSW 16
-------
CONTENTS (Continued)
Section
Estimated Particulate Emissions from Combined Firing
of Oil and Municipal Solid Wastes 22
Flue Gas Volume 24
Fly Ash Density 26
Fly Ash Fusion Temperature. 29
Particle Size Distributions 30
Fly Ash Resistivity •• 33
4 Cost and Effectiveness of Particulate Emission Controls on
MSW-Oil Fired Boilers 38
Inertial Collectors ........ .... 38
Wet Scrubbers ............ 41
Electrostatic Precipitators 42
Control Costs for Electrostatic Precipitator Control . 44
Electrostatic Precipitator Performance Model ..... 44
5 Case Studies 43
District of Columbia. .................. 48
Program Status 51
Refuse Derived Fuel (RDF) Preparation and Facilities . 51
Characteristics of Test Boiler ............ 53
Installed Air Pollution Control Equipment. ...... 53
District of Columbia Emission Regulations. ...... 56
Estimated Performance of Installed Air Pollution
Control Equipment. ........ 50
Cost of Air Pollution Control. ............ 53
Wilmington, Delaware (New Castle County). ... 62
Project Status ............... 62
Refuse Derived Fuel Preparation and Facilities .... 62
Characteristics of Test Boilers. ..... 6?
Installed Air Pollution Control Equipment. ...... 65
Delaware Emission Regulations 65
Estimated Performance of Air Pollution Control
Equipment. ..................... 69
Cost of Air Pollution Control 69
vi
-------
CONTENTS (Concluded)
Section Page
New York City 72
Project Status 72
Refuse Fuel Preparation Facilities. .... 73
Boiler System Descriptions 73
Installed Air Pollution Control Equipment ....... 73
New York City Particulate Air Emission Regulations. . . 73
Estimated Performance of Installed Air Pollution
Controls 73
Cost of Emission Control. ............... 73
State of Connecticut (Bridgeport). . ..... 74
Project Status. 74
Boiler System Descriptions. .............. 75
Refuse Preparation Facilities ............. 75
Installed Air Pollution Control Equipment ....... 75
Connecticut Air Emission Regulations. ......... 75
Estimated Performance of Air Pollution Control
Equipment ...................... 76
Cost of Emission Control. 76
6 Recommendations ............ ... 77
Electrostatic Precipitator Control 77
Cyclone Control 78
Scrubber Control 78
References. ............................. 80
Appendix A - Particulate Emissions from Oil-Fired Electric
Utility Boilers 87
Appendix B - Particulate Emissions Data for Waterwall Incinerators. . 97
Appendix C - Electrostatic Precipitator Performance Model 104
vii
-------
FIGURES
No,
1 Solids Burden Plotted Against Excess Oxygen for Different
Boiler Loads and Fuel Oil Types. ...... 13
2 Uncontrolled Electric Utility Emission Versus Capacity (no
additives employed). .................... 14
3 Controlled and Uncontrolled Particulate Emissions for Residual
Oil Burning Base Loaded Power Plant Boilers Operating at
^ 70 MW (no additives employed) 15
4 Dimensionless Plot Showing Fractional Increase in Total Fly
Ash as a Function of Percent MSW for Various Values of
Fuel Characterization Parameter (C). . . . . ........ 19
5 Theoretical Gas Flow Rates for Combined Firing of Oil and
Municipal Solid Waste (MSW) in a 75 MW Power Plant 28
6 Weibull Parameter Interpolation of Andersen Impactor Data for
Harrisburg Municipal Incinerator •••........... 34
7 Fractional Efficiency Data for Cyclone Collection of Fly Ash
from Coal- and Oil-Fired Electric Utility Boilers. ..... 4Q
8 Total Installed Cost of Wet Scrubbers t 43
9 Total Installed Cost for Electrostatic Precipitators 45
10 Process Flow for Preparation of RDF at District of Columbia. . 52
11 Effect of MSW Fly Ash Fraction (fr) on Calculated Particulate
Emissions (uncontrolled) from Combined Firing of MSW and
No. 6 Residual Oil (from Tables 1 and 2) 59
viii
-------
FIGURES (Concluded)
No. Page
12 ESP Efficiency (predicted) for Combined Firing of Oil and
MSW at Pepco Benning Station No. 26. 60
13 Estimated Particulate Emissions (controlled) for Combined
Firing of Oil and MSW at Pepco Benning Station No. 26 61
14 Schematic Representation of Materials Recovery Process, New
Castle, Delaware ....................... 63
15 ESP Efficiency (predicted) for Combined Firing of Oil and
MSW at Delmarva Edgemoor Station No. 4 (150 MW). 70
16 Estimated Particulate Emissions (controlled) for Combined
Firing of Oil and MSW at Delmarva Edgemoor Station
No. 4 71
C-l Block Diagram of ESP Performance Model 108
C-2 Current Density as a Function of Resistivity ..... 113
C-3 Comparison Between the Voltage Versus Current Characteristics
for Cold-Side and Hot-Side Precipitators 114
IX
-------
No.
TABLES
Page
1 Estimated Total Particulate and Percentage of Refuse Fly Ash
in Fly Ash Composite from Combined Firing of Oil and
Municipal Solid Wastes (MSW) •• 23
2 Controlled Emissions for Combined Firing of Refuse with Oil
or Methane. ......«••••••••••••••••••• 25
3 Theoretical Gas Flow Rates for a 75 MW Power Plant at 130° C,
1 ATM 27
4 Particle Size Distribution for Oil-Fired Boilers. ........ 31
5 Particle Size Distribution of Refuse Fly Ash ..... 32
6 Incremental Particle Size Data for Coal, Oil, and Refuse Fly
Ash Determined Using Two Parameter Weibull Distribution .... 35
7 Summary of Effectiveness of Various Control Systems in Use on
Oil-Fired Electric Generating Plants. ............. 39
8 Design and Cost Data for Electrostatic Precipitators Designed
for Collection of Waterwall Incinerator Fly Ash 46
9 Estimated Materials Balance ••••••••........... 49
10 Target Analysis of Refuse Derived Fuel at SWRC-1 50
11 Pepco Benning Station Boiler No. 26 Design Ratings 54
12 Characteristics of Electrostatic Precipitator 55
13 Projected Analysis of Refuse Fuel-New Castle County, Delaware . 64
-------
TABLES (Concluded)
No.
L4 Delmarva Edgemoor Station Boiler Design Ratings. ....... 66
15 Design Data for Cyclone Collector/Delmarva Edgemoor No. 3. . . 67
16 Characteristics of Electrostatic Precipitator on Delmarva
Edgemoor Unit No. 4 o . . . „ 68
C-l Data Inventory for Electrostatic Precipitator Performance
Model 107
xi
-------
ACKNOWLEDGMENTS
This report was prepared for IERL-RTP under Contract No. 68-02-1324. The
work was performed by Dr. Jim Galeski, Associate Environmental Engineer, and
Mr. M. P. Schrag, Head, Environmental Systems Section, with the assistance
of Mr. Joe Shum, Assistant Environmental Engineer.
Approved for:
MIDWEST RESEARCH. INSTITUTE
. >fe*/t--0^-rv-
-------
SECTION 1
SUMMARY
Studies on the control of particulates from the combined boiler firing
of oil and municipal solid wastes (MSW) were conducted for the Industrial En-
vironmental Research Laboratory, Research Triangle Park (IERL-RTP). Objectives
of this project were to develop quantitative emission forecasts and recommend
control strategies for several planned oil-MSW combined firing tests. Oil-MSW
tests included planned EPA demonstrations and industry tests which are to be
carried out in utility boilers.
The program was divided into five major areas of activity: (a) an in-
formation search to acquire particulate emissions data for oil-fired utility
boilers and municipal waterwall incinerators equipped with appropriate control
devices; (b) development of correlations for fractional efficiency and control
costs as functions of major design, control, and operating variables; (c) de-
velopment of appropriate analytical models for control performance; (d) case
studies to develop preliminary emission forecasts for plants where combined
MSW-oil firing tests are planned; and (e) development of recommendations for
future work.
INFORMATION ACQUISITION
Data acquisition included literature searches and contacts with govern-
ment and private industry sources. Literature sources included: Air Pollu-
tion Abstracts, 1970 to 1974 (all entries); NAPCA Abstracts, 1970 to 1974 (all
entries); Applied Science and Technology Index, 1958 to 1973 (all entries);
and Engineering Index, 1965 to 1971. In addition to these sources, an APTIC
Search was ordered and a key word index search was run on holdings in the Bay
Area Air Pollution Library.
Data inventories were acquired for: (a) emissions from oil-fired electric
utility boilers; (b) control device performance for oil-fired utility boilers;
(c) emissions from waterwall incinerators; and (d) control device performance
for waterwall incinerators.
-------
CONTROL PERFORMANCE AND COST CORRELATIONS
Control system performance data were compiled for oil-fired utility
boilers and for refuse firing systems (waterwall incinerators, refractory
wall incinerators, and combined firing systems).
Data correlations for oil-fired boiler emissions were incorporated into
a simplified model developed to predict uncontrolled particulate emissions
as a function of percent fuel ash, percent fuel sulfur, fuel firing rate, per-
cent excess oxygen (or air), and fuel heating value. A range of uncontrolled
particulate emissions was defined on the basis of previous analyses of partic-
ulate levels generated in combined firing of coal and MSW. For a given boiler,
refuse fuel, and method of firing, variations in total fly ash were interpreted
in terms of a dimensionless "fuel characterization factor" comprised of heat-
ing values, ash contents, and fly ash-to-total ash ratios for both the con-
ventional fuel and the auxiliary MSW fuel.
For ESP control, a modified form of the Deutsch equation was used to de-
scribe variations in collection efficiency resulting from changes in flue gas
volume. The effect of particle size distribution was determined by calculating
a separate collection efficiency for each discrete particle size range and
by applying appropriate Cunningham "slip correction" factors. Empirical correla-
tions were used to determine limitations in current density resulting from in-
creased particle resistivity and/or increased flue gas temperature.
Installed cost data were acquired for new ESP installations, for retro-
fitting ESP units designed for coal, and for wet scrubbers. Detailed cost and
design data were acquired for two new ESP units designed specifically for col-
lection of refuse fly ash.
ANALYTICAL MODEL DEVELOPMENT
Analytical model development efforts included the development of predic-
tive models for particle size, flue gas volume, total fly ash, and control
device performance under combined firing conditions. Installed control systems
on boilers in which combined firing tests were planned included ESP's and
cyclones. A semiquantitative predictive model was developed for ESP control.
An analytical procedure was developed for calculating flue gas volumes as
a function of MSW fraction, fuel moisture, fuel sulfur, higher heating values,
and elemental compositions (C, 0, H, N, and S). Calculations for representa-
tive oil and MSW fuels indicated a significant increase in flue gas volumes
corresponding to increasing MSW heat input. For each percent of MSW heat input
the calculations indicate approximately a 1% increase in flue gas volume. It
-------
is anticipated that this increase will have a significant effect on fractional
control efficiencies.
Control performance evaluation methodology was developed for ESP control
on the basis of previous performance studies. Use of the model was limited
somewhat by the available data for refuse properties, particle size distribu-
tion, fly ash density, and resistivity for the MSW fuel. In general, however,
the problem of fly ash collection when oil and MSW are fired in the same
chamber does not appear to have any characteristics which have not previously
been encountered in designing electrostatic precipitators for oil-fired units.
On the basis of the present study, it is estimated that for most medium ef-
ficiency ESP units designed for coal fly ash (95.0 to 98.0% design efficiency)
the efficiency is expected to drop to 60 to 707., for oil, 70 to 85% for oil
plus refuse, depending on refuse composition and MSW heat input.
One multicyclone' control system was evaluated in the study because of a
planned oil-MSW combined firing test at Delmarva Power and Light Company's
Edgemoor Station No. 3. It was found that for cyclone control, the theoreti-
cal basis for pressure drop and collection efficiency had not been well de-
veloped because of complexities of flow fields. Experimentally determined
fractional efficiency data for cyclone control systems acquired in the litera-
ture study were used to adjust performance for changes in particle size dis-
tribution under combined firing conditions.
CASE STUDIES
Planned combined oil-MSW systems which were examined in detail included
(a) District of Columbia, (b) Wilmington, Delaware (New Castle County), (c)
New York City, and (d) Bridgeport, Connecticut. On the basis of control per-
formance estimates, the performance of installed control systems designed for
coal appears to be questionable for meeting applicable regulations for partic-
ulate emissions, except at very low MSW heat input and/or boiler load. For
ESP control, major expected control difficulties result from: high flue gas
dust loading and flue gas volume, low average particle density, fine particles,
and large percentages of carbonaceous low resistivity particulate.
RECOMMENDATIONS
In view of the rapid projected growth rate of MSW fuel utilization in com-
bustion systems and increasing public awareness of associated air emissions,
in-depth evaluation of several existing and novel control systems appears justi-
fied.
The following recommendations are directed toward major deficiencies in
the control technology defined in the present study.
-------
Electrostatic precipitator control
In adapting existing theoretical studies to the development of a practi-
cal performance model to predict ESP performance for combined firing applica-
tions, there was found to be no quantitative information on the effects of:
1. Fly ash density;
2. Re-entrainment; and
3, Sneakage or bypassing.
Analysis of the effects of fly ash density, re-entrainment, bypassing,
and other factors is recommended, based both on analysis and a survey of
available data including contacts with equipment manufacturing firms.
Cyclone control
The theoretical prediction of cyclone pressure drop and collection ef-
ficiency is not possible because of complexities of flow fields. Other factors,
such as the tendency of cyclone collectors to plug when in service on oil ash,
and the performance decline in the corrosive atmosphere of incineration flue
gases, need to be examined.
As in the case of ESP control, additional work on this type of control
system specific to the application of combined fossil fuel-MSW combustion is
recommended. This would include contacts with vendors, literature review, and
analysis beyond the scope of the present study to develop guidelines for use
in combined firing applications.
Novel Control Devices
High performance scrubbers and wet electrostatic precipitators have utility
for collection of particulates, gaseous pollutants (SOX, NOX, and others) and
potentially hazardous trace metals. There were no scrubber or wet ESP units in-
stalled on the boilers evaluated in the present study. An in-depth study of
these control methods appears justified.
-------
SECTION 2
INTRODUCTION
Resource recovery systems involving combined firing of shredded municipal
solid wastes (MSW) with fossil fuels, such as oil or coal, are relatively new.
Consequently, there is little available information from which to predict the
performance of emission control systems. The present study was directed toward
the analysis of control performance when oil and MSW are fired concurrently
in a conventional electrical utility boiler.
The first demonstration plant to process raw municipal waste for use as a
supplementary fuel in power plant boilers—the St. Louis-Union Electric Refuse
Fuel System--is presently demonstrating the potential and problems of coal-MSW
firing.
Oil-MSW firing is potentially more attractive in terms of long range fuel
conservation, if a number of operating problems can be resolved, and if partic-
ulate emissions can be successfully controlled.
Oil-fired units have not utilized or needed very efficient control of
particulates in the past. Particulate control from oil-fired steam generators
normally involves using mechanical collectors which are used primarily during
soot blowing. In combined firing, the resulting (controlled) particulate levels
may be much higher than the existing standard for oil-fired boilers. In addition,
the performance of those systems with high efficiency collection is in ques-
tions. It is not known precisely how the various high efficiency particulate
control devices will behave for combined firing of refuse and oil.
The objective of the present study was to acquire the data base required
for development of semiempirical control performance and cost models for the
control of emissions from combined firing of oil and MSW and, using this in-
formation, to analyze the performance of installed control systems for planned
oil-MSW combined firing systems.
-------
SECTION 3
PARTICULATE EMISSION DATA ACQUISITION AND EVALUATION
A review of information pertinent to concurrent firing of oil and MSW
was made using both literature searches and telephone contacts with private
sources knowledgeable with (a) waterwall incinerators and emission controls,
and (b) combined firing applications. The objective of this part of the study
was (1) to determine the properties of particulate and flue gas from firing
oil and refuse which are pertinent to emission control using cyclones, electro-
static precipitators, and high performance wet scrubbers, and (2) to estimate
the composite fly ash and flue gas properties for combined firing. Pertinent
variables for the control systems considered were determined to be:
1. Total fly ash per weight of fuel burned. Relative amounts of fly ash
from burning oil and refuse fuels separately are needed to determine total
fly ash emitted in combined firing and to estimate the physical properties
of the composite fly ash which are needed for control device sizing (resis-
tivity, bulk density, etc.).
2. Flue gas volumes per weight of fuel burned. Excess air levels and
fuel composition data (moisture, ash, ultimate analysis) chiefly determine
the flue gas volume.
3. Particulate density. Although fly-ash density has not been considered
in previous combined firing studies, discussions with equipment vendors indi-
cate that this may be an important factor in ESP sizing in that a decrease in
density requires a proportional decrease in flue gas velocity through the pre-
cipitator. There is unfortunately no quantitative treatment available to de-
scribe the influence of reduced particle density on collection efficiency.
On the basis of classical analysis, electrical or "capture" forces are in-
dependent of particle density, while aerodynamic or "drag" forces are in-
versely proportional to particle density for laminar flow. For particles of a
-------
given diameter, the rate of acceleration away from the collection electrode
during rapping cycles will also be inversely proportional to particle diameter.
Other key design parameters (e.g., particle resistivity, dielectric constant,
etc.) may also vary systematically with particle density.
4. Ash fusion temperature. This factor is important with regard to pos-
sible plugging of air passages and slagging of boiler tubes. It also indirectly
influences excess air required.
5. Particulate size distribution. Particle size distribution data for
each fuel burned separately are used, with the mass emissions data, to estimate
the particle size distribution of the composite fly ash.
6. Particulate resistivity. Particulate resistivity depends on many fac-
tors including flue gas moisture level, fuel sulfur, fuel metals active as
oxidation catalysts (principally vanadium), basic chemical constituents, and
mass fraction of particulate from each fuel source. In firing of oil alone,
high concentrations of metals active as oxidation catalysts, principally vana-
dium, are believed to be responsible for "acid smuts" which are agglomerates
of oil ash and sulfuric acid. In combined firing of oil-MSW, it is anticipated
that HoSOA will be less of a problem than it sometimes is in oil-fired units.
The higher particulate levels expected in combined firing should serve to col-
lect most of the sulfates formed, resulting in a "well conditioned" fly ash
with reduced resistivity. In the combined firing of oil with coal, none of the
problems associated with firing of oil alone are encountered, indicating that
the coal ash tends to adsorb excess moisture and sulfur compounds..!/
Data which were found for these properties for each fuel type, and meth-
ods of estimation of composite properties needed for control system sizing,
cost, and performance estimation are described in the following subsections.
MASS EMISSIONS DATA
Data inventories were compiled for particulate emissions from oil-fired
electric utility boiler,s and waterwall incinerators, as summarized in abbre-
viated form in Appendices A and B, respectively. The objective of the data
acquisition was to use these data to estimate the range of particulate emis-
sions which is probable when oil and MSW are fired concurrently in a utility
boiler. More precise estimates of particulate emissions require detailed in-
formation about the fuel-oil ash content, refuse ash content, and type of
boiler.
-------
Uncontrolled Particulate Emissions from Oil-Fired Electric Utility Boilers
Particulate from oil firing (uncontrolled) was found to vary from 0.0055
to 0.87 g/106 joules of heat input (approximately 0.013 to 2.02 lb/10 Btu)
with an average value for 29 electric utility boilers of 0.0632 g/10 joules
(0.147 lb/106 Btu). In calculating the average, duplicate tests for the same
boiler were averaged. A more representative value for uncontrolled particulate
from oil firing is obtained by discarding extreme values. This yields a range
of 0.01 to 0.154 g/106 joules (0.023 to 0.36 lb/106 Btu).
As indicated by the above range of values, uncontrolled particulate emis-
sions from oil firing can vary considerably. The lower value is approximately
equivalent to 0.0237% by weight of oil-fired, which is in the same range as
the ash content of No. 6 residual oil (0.002 to 0.3% by weight).!/ The upper
value is approximately equivalent to 3.6% by weight of oil fired, or approxi-
mately one order-of-magnitude higher than the expected ash content. It has
previously been noted that particulate fly ash from oil firing can range from
approximately the ash content of the oil up to 10 to 15 times the ash content,
the higher range of values being attributed to poor combustion.—' The combus-
tible portion of fly ash normally ranges from 60 to 90%..£/
In order to determine the expected range of uncontrolled particulate for
a given installation, the influence of several factors must be quantitatively
defined. The quantity, type and size of uncontrolled particulate emissions
from oil-fired combustion operations are thought to depend mainly upon the fol-
lowing f actor s:^i-i.'
Overall fuel consumption rate
Ash and sulfur content of fuel
Use of mineral fuel additives
Degree of atomization (type and oil viscosity)
Windbox air admittance
Burner tilt
Excess air
Boiler load
Flue gas recirculation
Age and condition of boiler
A review of the available literature indicated that in most cases, quan-
titative relationships describing the dependence of particulate emissions on
these factors either did not exist or correlations developed were very un-
certain. The reason for this is probably that the number of variables involved
in a given plant are very large, and since these change from plant to plant,
it is impractical to obtain sufficient data on the effect of a single variable
while maintaining the others constant. Unfortunately, it is not always correct
8
-------
to assume that each factor exerts an independent influence on the measured
emissions. It would be preferable to use statistical methods (e.g., linear
regression analysis) to describe the influence of a single variable under con-
ditions in which other variables are also changing, but unfortunately this
has not been done.
As an illustration, consider the influence of excess air under variable
boiler load. Some boilers are designed specifically for peak power generation
and operation under variable load conditions. More frequently, when a convec-
tive superheater is used, decreasing the boiler load causes a decrease in
steam temperature.—' Under conditions of reduced boiler load, increased excess
air may be required to maintain both boiler stability and a constant steam
temperature. > ' An empirical equation has been developed by MaartmanZ.' to
describe these variations which also includes the variation in dust concentra-
tion with the sulfur content of the oil:
a
G - r (1)
0P x Q
Q
where G = dust concentration:mg/Nm
S = sulphur content of the oil:percent
02 = excess oxygen:percent
Q = capacity of the boiler:tons of steam per hour
Based on measurements made in England, Germany, and Sweden, it is thought that
the values of the parameters a, p, and 6 are:
0 <: a <: 1 (2)
9 = 1.0 (3)
0 £ 6 £ 1 <4>
The literature and data inventories compiled during this study suggest
the following order-of-magnitude variations in uncontrolled particulate emis-
sions with design and operating conditions:
Overall Fuel Consumption Rate; There is an apparent increase in partic-
ulate emissions (in grams per 10° joules of heat input) by a factor of two as
-------
capacity decreases from 600 to 100 MW.-/ The lower emissions for larger units
probably result from better control, improved design, and better condition
of the burners in the newer and larger oil-fired boilers.
Ash Content of Fuel; There is no quantitative information on the varia-
tion in particulate emissions with ash loading, although the average for No.
6 residual oil is estimated from available literature to be 0.1% by weight.
The approximation made in this study is that greater or lesser concentrations
of ash in the fuel oil will add or subtract from the total particulate in a
1:1 proportion.
Sulfur Content of Fuel; Sulfur content in No. 6 residual oil may be-
tween o73and^%7^Aleast squares linear curve fit of available data indi-
cates that uncontrolled particulate emissions increase by about 25% over the
range 1.0 to 2.5% sulfur by weight^./ The scatter in the available data is
again quite large, probably for reasons of nonlinear influence of design and
operating conditions, as discussed previously.
Mineral Fuel Additives; The typical recommended application of fuel oil
additives is in the proportion of 1 kg of additive per 1,000 kg of fuel oil.—'
The solid additive used is usually dispersed in slurry form in a light oil
(e.g., No. 2 fuel oil) so the concentration of solids is usually less than
0.5 kg/1,000 kg of fuel oil, or 0.05% by weight. This is significant in com-
parison with normal fuel ash and should be included in calculating total fuel
ash.
Degree of Atomization; The effectiveness of atomization is influenced
mainly by the type of atomization (mechanical, steam, or air), the condition
of the burners, and the viscosity of the fuel oil. The viscosity is determined
by the intrinsic viscosity of the fuel and the temperature to which the oil
is heated prior to atomization. Studies have been made to determine the effect
of these variables on particulate emissions;.?-' however, there are little data
and trends are insufficiently clear to permit a quantitative interpretation.
Windbox Air Admittance; Varying the settings on the main and auxiliary
air dampers can cause pronounced effects on fly ash,!/ but there are insuf-
ficient data for a quantitative prediction.
Burner Tilt; There is evidence that burner tilt can influence fly ash
loading under certain conditions.!/ There are not sufficient data to permit
a quantitative prediction at this time.
Flue Gas Recirculation; Uncontrolled fly ash emission is believed to
increase significantly when more flue gas is recirculated into the firebox.
10
-------
This increase is believed to be due to & cooling of the flame and combustion
gases.—' There are insufficient data for a quantitative correlation.
Excess Air and Boiler Load Level; These are most significant operating
parameters, which can exert an order -of-magnitude influence on uncontrolled
particulate emissions. As discussed previously, excess air is frequently
increased with decreasing load level to maintain a constant steam temperature.
At a constant load level, it was observed in one test that as the oxygen con-
centration in the stack gas decreased from 4 to 2% (corresponding to a decrease
in excess air of approximately 22 to 10%) the particulate loading increased
from 0.0086 to 0.060 g/106 joules .I/
On the basis of the preceding discussion, the following empirical rela-
tionship was developed to predict the variation in fly ash from firing No. 6
residual oil as a function of: excess air, expressed as percent oxygen; boiler
load; fuel ash; fuel sulfur content; and fuel heating value.
g (1 + A - A)(l + CM(M - H))(l + CS(S - 5)) H/H (5)
o2/o2
where g = uncontrolled particulate emissions, g/10" joules
g = mean (uncontrolled) particulate emissions for electric utility
boilers firing No. 6 residual oil, g/106 joules
A = ash content of fuel, percent by weight
A = average ash content of No. 6 residual oil-fired, percent by weight
M = boiler load, MW
M = average load of boilers firing No. 6 residual oil, MW
S = sulfur content of fuel, percent by weight
S = average sulfur content of No. 6 residual oil-fired, percent by
weight
H = higher heating value of fuel, joules/kg
H = average higher heating value of No. 6 residual oil-fired,
joules/kg
11
-------
02 = excess oxygen, percent by volume
•J32 = average excess oxygen for boilers firing No. 6 residual oil,
percent by volume
Qyr»Cc = proportionality constants
The functional dependence of total particulates on excess air (percent
oxygen) used in Eq. (5) is derived from the analysis of MaartmannZ/ the re-
sults of which were presented in the preceding discussion as Eqs. (1) through
(4). The linear variation between total particulates and boiler load is based
on regression analysis (Eq. (1), Ref. 4). The linear variation between total
particulates and fuel sulfur is also based on regression analysis (Figure 6,
Ref. 4). The linear variation of total particulates with fuel ash (including
additives) is an assumption, as stated in the preceding discussion. The in-
verse linear variation of total particulates with fuel heating value is pre-
sented without attempting to justify the approximation.
The overall range of validity of Eq. (5) for estimating oil emissions is
unknown. For the purposes of the present study, this equation is considered
to be sufficiently valid, if somewhat conservative. Comparisons between mea-
sured and predicted total particulate are presented in Figures 1, 2, and 3 as
functions of percent oxygen, boiler load, and percent fuel sulfur, respec-
tively. Several values of "g were used in Figures 1 to 3 ranging from "g = 0.0632
g/106 joules (average from the data inventory in Appendix A) to "g = 0.00632 g/
10 joules. Other numerical values used were as follows:
A = 0.1% by weight2./
C^ = 5.324 x lO-4 (adapted from Ref. 4, Eq. (1))
H = 263.87 MW (calculated from the data inventory in Ref. 4)
Cg = 0.0670 (determined from Ref. 4, Figure 6)
S = 1.15% by weight (determined from Ref. 4, Figure 6)
I = 4.233 x 107 joules/kg!/
0 = 2.8942% by volume (calculated from an average excess air level
of 15.0%-X)
In Figure 1, conversion of total particulate to milligrams per normal
cubic meter was made by assuming an average heating value of 4.233 x 107
joules/kg and a volume equivalent of 13.4 Ifa3/kg of oil.
12
-------
400
-O
300
CO
E
Z
CD
-§ 200
00
in
T3
~O
OO
100
0
O
= 0.0632g/106j
;
•g = 0.0063 g/106j
= 0.0126g/106j
0.2% wt of Fuel
D 32mth* Oil D
O 24mth Oil D
A32mth Oil C
V 24mth Oi I C
• 16mth Oil C
— Equation 5
*mth = Metric Tons Per Hour
I I I I I I I I I I I
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Vol % O2 in Flue Gas
Figure 1. Solids burden plotted against excess oxygen for
different boiler loads and fuel oil types.!/
13
-------
-o
o
0.12
0.11
0.10
0.09
§ 0.08
'a
j= 0.07
D
U
E
«
Figure 2. Uncontrolled electric utility emission versus capacity (no additives employed)."
-------
Ui
•o
o
c
o
E
UJ
o
3
^O
I
JU
_Q
S
-------
Conversion of excess air to volume percent oxygen can be accomplished
using standard procedures,I/ provided that the heating value is known. For a
higher heating value of 4.233 x 107 joules/kg, the following expression is
approximately correct for excess air levels between 0 and 40%.
, 3
02 ^ 0.2 EA - 3.1338 x UT5EA (6)
where EA = excess air, percent by volume
02 — as previously defined.
It should be emphasized that there are expected to be rather wide varia-
tions in emissions from boilers firing No. 6 residual oil, as a result of
factors not included in the data correlations, as previously discussed. There-
fore, Eqs. (1) through (6) should be used only to give an order-of-magnitude
estimate of the uncontrolled emissions.
Uncontrolled Particulate Emissions from Waterwall Incinerators
Uncontrolled particulate emissions from waterwall incinerators ranged
from 0.95 to 9.64 g/106 joules (approximately 2.21 to 22.4 lb/106 Btu). The
average value for seven boilers and 22 tests was 2.56 g/10 joules (5.95 lb/
10 Btu). Stack test data for particulate emissions from several waterwall
incinerator plants in the United States and West Germany were included, as
tabulated in Appendix B.
In calculating the average, tests for the same boiler were not averaged,
since tests were usually made under conditions of variable excess air, re-
fuse heating value, ash and moisture content, and boiler load.
Possibly because of the limited data available at the time of this study,
there are no clear trends in uncontrolled particulate emissions with refuse
fuel heating value, ash content, or excess air level. Because of the limited
data available, absence of clear trends, and the fact that the waterwall in-
cinerator plants listed fired unshredded refuse, an empirically based representa-
tion of the available stack test data was not considered justified. At the
time of the study, an in-depth investigation of emissions from waterwall in-
cinerator plants was being conducted for EPA, Office of Solid Waste Management
Programs,I2/ and data from this study, when completed, should be used to sup-
plement the data included in the present report.
Estimate of Refuse Fly Ash from Suspension Firing of MSW
There is no accepted design method for calculating particulate fly ash
from combined suspension firing of refuse and conventional fuels. Control
16
-------
device sizing and performance evaluation is therefore somewhat a stochastic
procedure. Estimates for the fraction of refuse ash which ultimately becomes
fly ash (here denoted "fr") range from about 0.13 up to 0.50, according to
various sources. Estimates determined from the analysis of actual test data,
described in this section, range from 0.13 to 0.35.
The approach for estimation of particulate loading favored by some in-
vestigators2J-L/ is to use a ratio of fly ash to bottom ash characteristic
of the fuel, the firing method, and the boiler system. This approach is of
particular value to those concerned with MSW fuel preparation since it allows
the fly ash to be related directly to MSW ash content and other fuel properties.
Assumptions implicitly made are that the refuse fly ash does not depend
on the type of fuel fired, e.g., oil or coal, and that the presence of both
fuels in the combustion chamber does not influence bottom ash or carbon burn-
out in the fly ash. This assumption implies that fly ash from the two fuels
can be added in linear fashion to yield a total fly ash particulate:—'
g = rg, + (1 - r)g_ (7)
where
g =
r =
g =
g =
total particulate, g/10 joules
fraction of MSW heat input
observed fly ash particulate from refuse burning, g/10
joules
observed fly ash particulate from burning No. 6 residual
oil, g/10° joules.
If it is further assumed that the carbon content of fly ash is negligible
by comparison with the total fly ash, Eq. (7) can be rewritten in terms of
refuse fly ash fraction f and fuel properties in the following form:
gx= 10'
X f r X. f (1 - r)
Ar r Ac c
H
H
(8)
17
-------
where g = grams of fly ash per 10 joules of composite fuel
H = average heating value of refuse, joules/kg
H = average heating value of auxiliary fuel, joules/kg
c
X = weight fraction of ash in refuse
Ar
X = weight fraction of ash in auxiliary fuel
Ac
f = fraction of refuse ash which becomes fly ash
r
f = fraction of ash in auxiliary fuel which becomes fly ash
c
r = fraction of refuse heat input
Equation (8) is used to estimate the total fly ash loading for a given
boiler, firing method, and fuel properties. Rearranging Eq« (8), the follow-
ing dimensionless equation results, which relates the fractional increase in
fly ash particulate at a given refuse heat input (r) to a dimensionless ratio
of fuel properties (c):
-------
20. Or-
°- 0.10 0.20
r = %MSW Heat Input
* See Equation 9 for Definition of \l/, C
Figure 4. Dimensionless plot showing fractional increase in total fly ash
as a function of percent MSW for various values of fuel char-
acterization parameter (C).
19
-------
fr = 0.15 (assumed)
XAr = 0.223H/
Hr = 11,572,000 joules/kgii/
f = 2.72 (assumed value, see p. 8)
XAc =
H = 42,330,000 joules/kg^
For these fuel properties, C depends on the assumed fly ash fraction as
follows:
£c £
0.15 45.9
0.30 91.8
0.50 153.1
Previous experience »' indicates that variations primarily in MSW fuel
properties may also cause variations in the ratios XAr/X.c and Hc/Hr. Changes
in these ratios will also influence the rate of fly ash generation as de-
scribed by Eq. (9).
Methods which have been used for estimating the fraction of refuse ash which
ultimately becomes fly ash (fr) are described in the following paragraphs.
Mass balance data; St. Louis/Union Electric refuse firing demonstra-
tion;!^,^/ Analysis of recent mass balance data from the St. Louis/Union
Electric refuse firing demonstration indicates that from 39.9 to 98.5% of the
refuse ash was ultimately discharged from the boiler in the form of sluice
solids (bottom ash). The average bottom ash fraction was 64.7% of the total
refuse ash.M/ By difference, the average fly ash fraction from the St. Louis
tests was equal to 35.3% of the total refuse ash.
The analysis based on bottom ash assumes that, in addition to the assump-
tions stated for Eqs. (7) and (8), a constant fraction (8.7%) of the ash present
in the coal is collected in the sluice solids as bottom ash.
20
-------
Particulate emissions data: National Center for Resource Recovery (NCCRl
analysis of St. Louis/Union Electric refuse firing demonstration:^* 15/ The
NCKR method is based on analysis of the differences in fly ash emission rates
when firing coal-plus-refuse and firing coal only. The assumption made is that
the flue gas volume is constant at a given boiler load and is independent of
the MSW fraction. This assumption is somewhat at odds with the overall con-
clusions of the St. Louis/Union Electric Air Pollution Test Report,.!!/ but is
consistent with the test data chosen for the example calculation. Notation and
dimensions used by NCKR have been altered as required for consistency.
Following Gershman, the particulate resulting from the MSW portion of the
fuel in combined firing is given bytift/
g'r = g'r+c - 8'c U-r) (10)
O
where g'r+c = total- flY ash from firin8 coal-plus-refuse, g/Nm
g« = fly ash from refuse portion of fuel, g/Nm
g' = fly ash from firing coal-only, g/Nm3
The fraction of refuse ash which becomes fly ash (fr) is determined from the
following equation.^/
_ g.r x 103 x V
XAr Wr
•3
where V = flue gas volumetric flow rate, Nm /min
Wr = MSW mass feed rate into the boiler, kg/min
fr> 8* » XAr as Previously defined.
The NGRR methodology yields fr = 0.21 for the data chosen:
r = 0, 0.1
Kt = 4.35 Nm3 (1.9 gr/dscf) at 100 MW, r = 0.1
O JLm\ "C
gt = 3.84 g/Nm3 (1.7 gr/dscf) at 100 MW, r = 0
21
-------
XAr = 0.21
Wr = 151.2 kg/min (20,000 Ib/hr)
V = 7,702 Ito3/min (272,000 dscfm)
Calculated fly ash fraction (fr) based on waterwall incinerator emissions
data; The average value of 2.56 g/10^ joules obtained from the survey of emis-
sions data for waterwall incinerators (Appendix B) corresponds to a value of
fr = 0.13 for an average heating value Hr of 11,572,000 joules/kg (4,975 Btu/
Ib) and a total ash content of 22.3%.
Calculated fly ash fraction (fr) based on St. Louis/Union Electric air
emissions data at 140 MW, 10% MSW; On the basis of test data from the St.
Louis/Union Electric refuse firing demonstration, in which shredded MSW was
fired in suspension with coal, particulate loading increased with increasing
MSW heat input in approximately the same ratio as increasing gas volumes..!!/
That is, there was no apparent net increase in flue gas grain loading. At 10%
MSW heat input and 140 MW, the theoretical increase in flue gas volume was
approximately 5.2%. Using this information and average fuel properties, the
value of fr was determined from Eq. (8) to be 0.16. Average fuel properties
used were as follows:—'
XAc = 0.0703
Hc = 29,226,000 joules/kg (12,565 Btu/lb)
fc = 0.85
XAr = 0.223
Hr = 11,572,000 joules/kg (4,975 Btu/lb)
The value of fr = 0.16 calculated from the St. Louis/Union Electric air
pollution test resultsJLi/ agrees favorably with the average value for waterwall
incinerators (0.13), when adjusted to a total ash content characteristic of
shredded refuse.
Estimated Particulate Emissions from Combined Firing of Oil and Municipal
Solid Wastes~~~—
The estimated range of uncontrolled particulate stack emissions from com-
binded firing of oil and MSW calculated from Eq. (8) are summarized in Table
1. An independent estimate of particulate, SOX, and NOX emissions made in
22
-------
to
CO
Table 1. ESTIMATED TOTAL PARTICULATE AND PERCENTAGE OF REFUSE FLY ASH IN FLY ASH COMPOSITE FROM
COMBINED FIRING OF OIL AND MUNICIPAL SOLID WASTES (MSW)-^
Flue gas particulate
% MSW
heat input
0.0
1.0
5.0
10.0
15.0
20.0
25.0
(g/106 ioules)
fr = 0.1
0.063
0,082
0.156
0..250
0.343
0.436
0.529
f r = 0. 15
0.063
0.092
0.205
0.346
0.487
0.629
0.770
fr = 0.3
0.063
0.120
t).349
0.635
0.921
1.21
1.49
fr = 0.5
0.063
0.159
0.542
1.02
1.50
1.98
2.46
fr = 0.1
0.0
23.5
61.1
77.2
84.3
88.4
91.0
% Refuse
fr = 0.15
0.0
38.1
76.2
87.1
91.5
93.8
95.3
fly ash
fr = 0.3
0.0
48.0
82.8
91.0
94.2
95.8
96.8
fr = 0.5
0.0
60.6
88.9
94.4
96.4
97.4
98.1
&l Estimate of uncontrolled fly ash emissions made using Eq. (8). Estimate based on average emission
rate for oil-fired boilers (Appendix A) and average MSW properties listed in Ref. 11: heating
value (Hr) of 11,572,000 joules/kg; ash weight fraction (X^r) of 0.223; and various fly ash
fractions (fr) as shown.
-------
an earlier study by another team of investigators at Battelle—' predicted
uncontrolled particulate from oil-refuse firing intermediate between values
listed in Table 1 corresponding to fr = 0.15 and fr = 0.30. The results of
this study are compared with the present estimates in Table 2. The Battelle
emissions model for combined fuel firing was also based on an assumed linear
addition of emissions from each fuel proportioned by relative heating value
as described in Eq. (8).!Z/
With regard to the physical properties of the fly ash particulate, it
is evident from Table 1 that both oil and refuse particulate will have an in-
fluence on the physical properties of the composite fly ash. This conclusion
is emphasized because it appears to be opposite to the general consensus of
several EPA and design engineers.
FLUE GAS VOLUME
Flue gas volume for combined firing of different fuels depends on several
factors, the most significant of which appear to be:
1. Boiler efficiency.
2. Excess air level.
3. Theoretical combustion air based on fuel composition and heating
value.
Since the total volume of flue gas is of primary importance in predicting con-
trol device performance, a significant part of the present study was devoted
to analysis of previous attempts to predict the flue gas volume for combined
firing applications.
It is anticipated that boiler efficiency will decrease somewhat with in-
creasing substitution of refuse derived fuel (for fuel oil). The estimated
magnitude of major losses at 20% MSW heat input are summarized as follows:
1. Heating of additional excess air for combustion, 1.0%0
2. Incomplete fuel combustion, 2.0%.
3. Heating of additional fuel moisture (25% moisture in MSW), 4.0%.
4. Increased "dry" gas losses, due to increased flue gas volume,
1.0%.
24
-------
Table 2. CONTROLLED EMISSIONS FOR COMBINED FIRING OF REFUSE WITH OIL OR METHANE
Cn
a/
Particulates"- Allowable
Heat input
from oil
or methane
(%)
90 (oil)
80 (oil)
70 (oil)
90 (methane)
80 (methane)
70 (methane)
Refuse
heat
input
(%)
10
20
30
10
20
30
(K/106
95%
Collection
efficiency
0.017(0.022)-^
0. 031(0. 044)^
0.046(0. 065 >r,
- (0.022)^.
- (0.043)-r'
- (0.065F-
joules) sulfur content
99% of oil for
Collection no SOX control
efficiency (%)— '
0.003(0.004^ °*77W
0.006(0.009)^, °«84r,
n / D /
0.009 (0.013 P. 0.92-
- (0.004P'
- (0.009)f
- (Q.013)^7
NOX ,.
(g/106 ioulesK
Oa2b/
°*12b/
0.099J^
°-099b1/
0.009s
a/ Values from Table 1 used in calculating uncontrolled particulate correspond to fr = 0.15.
y Data from Ref. 16, Table 24.
£/ If the allowable SOX emission is 0.344 g/10 joules (0.8 lb/10 Btu).
d/ Assuming a 60% reduction in NOX by particulate control system (for nontangentially-fired units).
-------
The above approximate values are estimates based on information in Refs. 12
and 18. A precise determination of combustion efficiency requires detailed
information on refuse fuel composition, ash, moisture and heating value, and
excess air level. The estimate for refuse fuel combustion efficiency (907= for
MSW) is based on data for suspension firing of MSW and Illinois No. 6 coal
in a tangentially fired boiler.!!/ It is not documented at this point whether
there is any improvement in combustion efficiency when the refuse fuel is
shredded to a finer average particle size.il/ Union Electric engineers claim
that fine shredding of refuse does result in better heat recovery. They in-
terpret a higher heat recovery with refuse ground to less than 3.18 cm (1.25
in.) when compared to refuse less than 7.6 cm (3 in.) in size based upon MW
generated per megagram of refuse. The St. Louis study, however, reports that
the combustion efficiency did not improve with fine ground refuse when based
on the two data points for fine-ground refuse,!!' Determination of improvements
in combustion efficiency as a function of refuse fuel particle size requires
additional study.
There is evidence that excess air may need to be higher when firing refuse
with oil than when firing oil alone. The reasons for this are related to cor-
rosion and "slagging," or deposition of fused ash on boiler tubes. Excess air
levels used in waterwall incinerator plants are typically 70 to 1007, where
corrosion is a problem (see Appendix B). There does not appear to be any method
for determining the excess air level a priori. There is some indication that
an excess air level of 15% is sufficient for firing up to 107o refuse heat in-
putiZHrJi' and that excess air should be increased to 257= for 207= MSW heat in-
put .I/
Theoretical air required for combustion was calculated from fuel composi-
tion and heating values using standard procedures (see Ref. 5, p. 4-10). Fuel
nitrogen was assumed to be converted to nitric oxide (NO). When dealing with
refuse derived fuels, it was found to be important to correct "dry basis"
ultimate analyses for fuel moisture actually present. When refuse derived fuels
are substituted for No. 6 residual oil, theoretical flue gas volumes will in-
crease by as much as 17= for each 17= of MSW heat input at the same excess air
level, depending on moisture, heating value, composition, and MSW substitution.
Representative gas volumes for a 75 MW boiler are listed in Table 3. Variations
due to MSW fuel moisture and refuse substitution rate are illustrated in Figure
5 .
FLY ASH DENSITY
It is estimated that fly ash from combustion of municipal solid wastes
will be comprised of a significant fraction of very light particles having
a density on the order of half that of fly ash from pulverized coal.^./ The
26
-------
Table 3. THEORETICAL GAS FLOW RATES FOR A 75 MW
POWER PLANT AT 130°C, 1 ATM
Power
output
(MW)
40
50
60
70
80
aj Basis
(1)
(2)
(3)
Fuel
moisture
(% wt. wet)
Oil Refuse
0.354 10
0.354 30
0.354 50
0.354 10
0.354 30
0.354 50
0.354 10
0.354 30
0.354 50
0.354 10
0.354 30
0.354 50
0.354 10
0.354 30
0.354 50
for calculation:
Ideal combustion to C02
157o excess air for oil
3 a/
Exhaust volume flow rates (m /min)~"
Oil 5% R 10% R
3,445 3,804 3,892
3,445 3,822 3,943
3,445 3,858 4,071
4,307 4,755 4,865
4,307 4,777 4,929
4,307 4,822 5,089
5,168 5,706 5,838
5,168 5,733 5,915
5,168 5,787 6,107
6,029 6,657 6,811
6,029 6,688 6,901
6,029 6,751 7,125
6,891 7,608 7,784
6,891 7,643 7,886
6,891 7,716 8,143
, H20, NO, S025
only; 25% excess air for oil plus
20% R
4,145
4,394
y
5,181
5,493
y
6,217
6,591
y
7,253
7,690
y
8,289
8,788
y
MSW;
Fuel properties used in the combustion calculations determined
as follows: data for
No. 6 residual oil from Ref. 20,
using
average values for refuse properties from Ref. 11.
b/ Refuse
fraction exceeds limits
for necessary boiler heat input.
27
-------
300
CO
Solid Lines Represent Theoretical Gas Flowrate for No. 6 Residual Oil
and 30% Moisture Refuse Assuming 15% Excess Air (Data for Coal
Based on 10% Moisture Coal and 40% Excess Air)
U_
u
CO
o
200
O
_j
u_
CO
<
o
100
30% Moisture
10% Moisture
REFUSE ENERGY, PERCENT
Figure 5. Theoretical gas flow rates for combined firing of oil and
municipal solid waste (MSW) in a 75 MW power plant.
-------
particle density overall ranges from 1.8 to 3.8 g/cc23'24/ compared to an
average particle density of 2.3 g/cc for fly ash from coal fired boilersjl/
The reason for the wide range in densities is that several mechanisms
for particulate formation all contribute significantly to the total partic-
ulate formation. The combustible fraction consists of entrained char ("black-
birds"), soot produced by the thermal cracking of pyrolysis products, and
"white smoke" which is produced by condensation of pyrolysis gases.,24/ It is
the low density, low resistivity carbonaceous particles which are the most
troublesome to control, especially by electrostatic precipitation, since they
tend to lose their charge on contact with the collection electrode and become
reentrained.25/
Particulate from fuel oil combustion also is not composed primarily of
mineral particulate. The size distribution is characteristically bimodal, the
larger particles being skeletons of burn-out fuel particles, called cenospheres,
which are hollow, black, coke-like spherical particles. The smaller particles
formed by the condensation of vapors are of regular shape and usually have a
maximum dimension of about 1.0 \im* The average density is about 2.5 g/cc,J-3/
but densities as low as 1.22 g/cc have been reported.—' In combined firing
of oil and refuse, there will probably be a sufficiently large proportion of
low density (<« 1.2 g/cc), low resistivity, predominately carbonaceous partic-
ulate that a major particulate control problem is expected.
FLY ASH FUSION TEMPERATURE
In previous studies of ash fusion levels, using both test coupons and
tetrahedral cones, the minimum melting range for ash from refuse incineration
was 454 to 1093°G (850 to 2000°F).^I/ Higher levels of lead and zinc were found
than had been anticipated, presumably from pigments, solders, and galvanized
coatings. The melting range correlated with total concentration of basic oxides
NaoO and KoO. The minimum range occurred at a level between 30 to 4070 of the
basic oxides by weight, excluding zinc, lead, and sulfur. Unfortunately, no
relationship could be established between the presence of liquid state and
chemical composition, a much more sophisticated procedure apparently being
required. It is possible that small portions of a liquid phase could form at
low temperature and remain undetected due to the wetting of the larger quantity
of dry material.—'
With few exceptions, the ingredients found in refuse deposits acceler-
ate gas side corrosion above 510°C (950°F). In the presence of a reducing con-
dition the threshold temperature could be reduced to as low as 315°C (600°F).
In an oxidizing atmosphere containing HCl, it was found that corrosion could
take place in a "dry" unfused, powdery ash.ZL/
29
-------
PARTICLE SIZE DISTRIBUTIONS
Particle size distribution data for participate from firing No. 6 resid-
ual oil in electric utility boilers and from incineration of municipal solid
wastes in waterwall and conventional incinerators is summarized in Tables 4
and 5, respectively.
It is widely believed that particulate from oil firing is extremely fine,
with the usual estimate being 90% less than 1 u-m. However, a search of avail-
able particle size data did not confirm this. The test data which showed the
largest proportion of submicron particulate was for an air atomized, tangen-
tially fired boiler which had originally been designed to fire coal and had
been retrofitted for oil firing. Based on Andersen impactor data,2§./ 13.3%
of the fly ash by weight was determined to be less than 1 urn in diameter.
Because of the chemical nature of the oil ash, measurement of particle
size is extremely difficult. Compared to coal ash, the solids emitted by fuel
oil combustion are more hygroscopic.—^' When allowed to cool, the oil ash par-
ticles absorb moisture and tend to agglomerate into larger particles during
storage, transportation, and handling. Therefore, ex situ size determination
methods such as Bahco and electron microscopy are subject to considerable error.
In situ measurements made using heated impactors are probably better, but here
too, there is a chance for error resulting from inelastic collisions between
particles and impactor walls. This could result in a distribution erroneously
weighted toward the larger particle sizes.
Fly ash particulate from incineration of municipal solid wastes is typi-
cally about 10% smaller than 1 urn in diameter.r_t' Andersen impactor data from
Harrisburg Municipal (waterwall) Incinerator indicate that in some cases the
particulate may be somewhat finer than this—about 20 to 30% less than 1 um.32/
The latter size distribution was used in the present study for particulate
from firing municipal solid wastes. Although use of the Harrisburg size distri-
bution data seems conservative, it is considered representative by other in-
vestigators, based on analysis of stack test data from the St. Louis/Union
Electric refuse firing demonstration.^./
When firing oil with other fuels in the same combustion chamber, it is
anticipated that the particle size distribution for the combined fuels will
be determined by the fuel having the higher ash content. That is, in combined
firing of oil and refuse, the assumption is made that the particle size dis-
tribution of particulate fly ash will be the same as the size distribution
for firing refuse alone. This assumption is justified by: (a) the observed
chemical nature of the oil ash, which tends to cause it to agglomerate with
other ash particles in the flue gas, (b) the high percentage of refuse ash
at MSW levels of 10 to 20% (Table 1), and (c) the observation that in combined
30
-------
Table 4. PARTICLE SIZE DISTRIBUTION FOR OIL-FIRED BOILERS
Diameter Wt % less than stated diameter
(tan) Ref. 26-^ Ref. 28^ Ref. 28£/ Ref. 28^
1,000 100.0 -
250 97.8
150 92.7
45 50.0 ...
30 39.6 65.3 77.8 73.0
9.2 13.2 52.3 55.6 46.0
5.5 6.8 38.3 25.7 26.3
3.3 3.2 29.0 10.5 15.9
2.0 1.4 21.6 6.6 9.0
1.0 0.3 13.3 1.3 5.7
0.3 0.0 4.9 0.65 2.3
0.1
a/ Bahco Data, specific gravity of oil ash = 1.22 (data inter-
polated graphically).
b/ Andersen impactor data, air atomizing burners on a tangen-
tially fired boiler.
c/ Andersen impactor data, mechanically atomizing burners on a
tangentially fired boiler.
d./ Andersen impactor data, steam atomizing burners on a tangen-
tially fired boiler.
31
-------
Table 5. PARTICLE SIZE DISTRIBUTION OF REFUSE FLY ASH
Diameter
(urn) Ref. 24^
1,000
250
150
45
30
20
15
10
5
1
.,
95
70
65
53
52
33
8
a/ Average of six
b/ Taken from Ref.
£/ Average
of data
Wt % less than
^ Ref. 29-' Ref
75
65
40
37
34
-
30
-
stated diameter
. 30^ Ref. 31^
••
-
-
45
41
36
-
30
23
13
recently reported studies (see
29, Figure 2.
for U.S. Incinerators. Taken
^
87
71
28
-
-
-
_
-
Ref. 24,
from Ref.
Ref. 32£/
—
81
74
59
55
51
48
45
40
30
p. A-8).
30,
Figure A-18.
d/ Based on sieve analysis of precipitator catch for tests of
Harrisburg waterwall incinerator. Average of three runs with
various heating values.
_e/ Based on two-parameter Weibull distribution parametric repre-
sentation of Andersen impactor data for Harrisburg Municipal
Incinerator (average of six test measurements). Distribution of
particulate larger than 14 ^m was estimated graphically.
32
-------
firing of coal and oil, none of the difficulties encountered in collecting
fly ash from combustion of oil alone are observed.!/
It is helpful to use analytical techniques in interpreting particle size
data because of the need to interpolate between data points to provide narrow
size increments for control device performance evaluation. In the present study,
a two parameter Weibull distribution3-!/ was fitted, using a least squares tech-
nique, to the particle size data (Figure 6). Because of the different mechanisms
operative for particulate formation, the portion of the size distribution less
than 1 urn was curve fit separately. The Weibull parameter distribution can be
expressed as follows:
/ \b
F(R(j)) = 1 - e" (^) (12)
V 9/
where F(R(j)) = the weight fraction of particulate having diameters less
than R(j)
6 , b = independent parameters
Weibull parameters for the refuse portion of the fuel are as follows:
b
(< 1 p-m) 1.48 2.91
(> 1 urn) 92.1 0.23
Incremental size data determined using Weibull parameters to fit selected
size data for fly ash from pulverized coal, No. 6 residual oil, and MSW are
shown for comparison purposes in Table 6. Note that for incineration of un-
shredded municipal solid wastes, a two parameter Weibull curve fit predicts
that a significant fraction of the particulate will be larger than 1,000 ym.
Approximately 20 to 30% of the fly ash is predicted to be less than 1.0 \aa. in
diameter.
FLY ASH RESISTIVITY
Resistivity values for fly ash particulate from municipal solid waste
incineration range from approximately 106 to 5 x 10 ohm/cm. The resistivity
maximum occurs between 149 and 204° C (300 to 400°F),25/ Bulk resistivity mea-
surements were made of integrated hopper-catch samples from two German water-
wall incinerators; the bulk resistivity of fly ash from firing refuse only at
33
-------
1.0
I
o
5
I
s
c
o
™
u
2
6.1
o.oi
0.1
8
A A
O 5/8/73
D 5/8/73
5/8/73
1:08 - 1:28 p.m.
5:03 - 5:23 p.m.
6:45 - 7:05 p.m.
O 5/9/73 10:35-10:50 a.m.
O 5/9/73 1:00 - 1:12 p.m.
V 5/9/73 3:45 - 3:59 p.m.
• Sieve Analysis of ESP Catch
(Average of 3 Samples) 31/
-—• ~— Graphical Extrapolation
—- —— Weibull Parameter Extrapolation
I I I I I I I I
10
Aerodynamic Diameter, ftm
100
1000
Figure 6. Weibull parameter interpolation of Andersen impactor data for
Harrisburg Municipal Incinerator.10?31?32/
-------
Table 6. INCREMENTAL PARTICLE SIZE DATA FOR COAL, OIL, AND REFUSE
FLY ASH DETERMINED USING TWO PARAMETER WEIBULL DISTRIBUTION
Diameter
range
(utn)
100-1,000
12-100
8-12
4.5-8
2.65-4.5
0.975-2.65
Q.70-0.975
0.27-0.70
0.10-0.27
0.01-0.10
< 0.01
MMET
Selected size distribution used in analyses
No. 6
Pulverized residual Municipal
oil£/ solid wastes!/
316.2
34.6
9.8
6.0
3.45
1.61
0.83
0.43
0.16
0.03
coa
0.6333
0.1000
0.0778
0.0622
0.0956
0.0144
0.0064
0.0102
0.0000
0.0000
1.0000
0.0749
0.4115
0.0824
0.1024
0.0773
0.1212
0.0309
0.0546
0.0256
0.0166
0.002.6.
1.0000
0.3150
0.2202
0.0307
0.0415
0.0358
0.0982
0.1508
0.1007
0.0067
0.0004
0.0000
1.0000
a/
c/
d/
MMD =v^-^
10% ash, Illinois No. 6 coal; pulverized coal fired in tangen-
tially fired boilerJ Andersen Impactor data (see Ref. 34).
Andersen Impactor data for No. 6 residual oil fired in a tangen-
tially fired boiler with air atomization (see Ref. 28). Data
interpolated using two parameter Weibull least squares fit
(Ref. 33).
Average of six tests using Andersen Mark III Impactor for parti-
cle size distribution of unshredded municipal solid wastes
fired in Harrisburg Municipal Incinerator (Ref. 32). Data inter-
polated using two parameter Weibull least squares fit (Ref. 33).
Distribution of particulate larger than 14 (o,m was estimated
graphically (Figure 6).
35
-------
Munich North Block 2 was 2 x 1Q9 ohm/cm at 160° C (320°F).|2/ Bulk resistivity
of hopper ash from firing refuse at Dusseldorf was 6 x 10 ohm/cm at 222° C
(432°F).—/ It is probable that these low values may have been influenced by
the selective collection of low resistivity ash by the electrostatic precipi-
tators. In situ resistivity measurements were attempted during acceptance
tests of the Harrisburg Municipal (waterwall) Incinerator by Southern Research
Institute. However, due to the high flue gas temperatures, in situ resistiv-
ity measurements could not be performed. Bulk samples were collected at the
inlet and outlet of the precipitator using a coarse cyclone, fine cyclone
and back-up filters in series. Three inlet samples and one outlet sample were
collected. Resistivity measurements were performed according to the ASME Power
Test Code No. 28 at a temperature of 218° G (425°F). The resistivity of the
coarse material was 3 x 108 ohm/cm and the resistivity of the fine material
was 5 x 106 ohm/cm.2i/
Because of higher flue gas temperatures characteristic of both conven-
tional refractory and waterwalled incinerator plants, it is unfortunately not
possible to compare ESP collection performance for these installations with
performance of precipitators on electric utility boilers. Design methods also
differ somewhat between these two types of installations, with precipitators
of European design being used on all waterwall incinerators presently equipped
with ESP control.iZ.' Data on the change in migration velocity with temperature
indicate that the precipitation rate parameter decreases by about 25 to 50%
as the flue gas temperature decreases from the range characteristic of water-
wall incinerators (200 to 250°C) to the lower temperature range encountered
in oil-fired electric utility boilers (150 to 2nnPn).25,27/
Somewhat in contrast to U.S. manufactured equipment, the European pre-
cipitator design approach aims at a more conservative migration velocity by
using somewhat lower field intensities. Thus, the size of the precipitator
must be increased commensurately. As a result of this, lower gas velocities,
3 to 4 ft/sec, are employed in combined-fired applications. U.S. designers
would typically specify gas velocities of 4 to 4-1/2 ft/sec for refractory
O *? /
walled incinerators.^1./
Gas velocity is selected for a discrete particle size distribution so
that re-entrainment problems are minimized. For comparison, gas velocities
used in design of coal fly ash precipitators are typically 6 to 8 ft/sec^./
while lower velocities on the order of 4 to 5 ft/sec are recommended for oil
ashjti35/
With respect to particulate emissions from oil-fired boilers, stack gas
temperature and sulfur content of the oil affect the resistivity of the non-
combustible portion of these solids; however, the balance of these solids are
composed of highly conductive combustible carbonaceous solids. As a result of
36
-------
these carbonaceous solids, the resistivity of the particulate emissions is
usually less than that for coal, 107 to 109 ohm/cm for oil versus 109 to 10
ohm/cm for coal.—' Difficulties which have been reported in collection of
oil ash result from the low concentration of fly ash having a relatively low
bulk density and a relatively fine particle size. As in the collection of in-
cinerator ash, there is frequently a high concentration of carbon in the oil
ash, which sometimes causes a resistivity so low that ESP collection becomes
difficult. In some cases, these solids are so conductive that they do not re-
tain a charge and subsequently prevent the field from becoming saturated.
Another problem that has been encountered is that these solids, upon deposi-
tion on collecting curtain surfaces, sometimes lose their charge to the cur-
tain and become re-entrained in the gas stream*—' At least part of the dif-
ficulty with ESP collection of oil fly ash could also result from the breakup
of agglomerates. Collection efficiency is improved through the employment of
high voltage, large collection curtains, lower superficial gas velocity and
high retention times. There is adequate evidence to indicate that, for an ESP
unit suitably modified efficiencies of 90+% are possible for collection of oil
ash. It is anticipated that high efficiencies should also be possible for refuse
ash, with similar modifications as for oil fly ash.
37
-------
SECTION 4
COST AND EFFECTIVENESS OF PARTICULATE EMISSION CONTROLS ON
MSW-OIL FIRED BOILERS
This section includes a discussion of control performance (fractional
efficiency) as well as cost data for estimation of control device cost and
performance for meeting specific particulate emission levels when firing oil
and municipal solid wastes (MSW) in a single combustion chamber. Control sys-
tems considered included cyclones, electrostatic precipitators, and high per-
formance wet scrubbers.
INERTIAL COLLECTORS
The efficiency of a multicyclone is dependent upon the size and density
of the particulates in the gas stream. On coal-fired cyclone furnaces the
efficiency usually ranges from 30 to 40%, while on a pulverized unit it ranges
from 65 to 75%.—' These differences can be attributed to the fact that the
mean particle diameter of the emissions from a cyclone furnace is usually lower
than that of the emissions from a pulverized unit. The estimated range of ef-
ficiency of cyclone collectors installed on oil-fired boilers is 82.5 to 90%
as determined from a summary of National Emissions Inventory System data (NEDS)
summarized in Table 7.
estimates that a maximum efficiency of 40% might be obtained for
small oil-fired boilers and that this efficiency would decrease as the boiler
size increased. Though they are not efficient in the reduction of fine partic-
ulate emissions, mechanical collectors could help reduce acid smut emissions
since smut is composed of agglomerated solids and is usually large in diam-
eter. I/
The range of overall collection efficiences for oil-fired boilers, based
on stack test data is 75 to 90%.3?7?26?35/ Fractional efficiency data for coal
fly ash and performance guarantees for oil ash are shown in Figure 7.
38
-------
Table 7. SUMMARY OF EFFECTIVENESS OF VARIOUS CONTROL SYSTEMS IN
USE ON OIL-FIRED ELECTRIC GENERATING PLANTS*/
Cont ro1 method
Wet scrubber - low efficiency
Cyclone - medium efficiency
Cyclone - low efficiency
ESP - high efficiency^/
ESP - medium efficiencyW
ESP - low efficiency^/
Gravity collector - high
efficiency
Cyclone (L)/gravity collector
(L)b/
ESP (M)/eyelone (M)b/
Cyclone (M)/ESP (M)^/
Cyclone (L)/ESP (M)^/
Cyclone (M)/ESP (H)£/
Gravity collector (H)/ESP (L)k/
No.
installed
1
28
11
7
11
28
3
2
6
2
4
1
Range of
efficiency
Average
efficiency
82.5-90
20-85
90-97
35-97
30-95
85.7-87.2
85
95
96-98.5
93
99.2
80
80
85.8
52.5
93.4
68.1
62.0
86.6
85
95
96.5
93
99.2
80
a/ Summarized from NEDS inventory listing, Ref. 4.
b/ Controls—ESP = electrostatic precipitator, M = medium efficiency,
" H - high efficiency, L = low efficiency.
39
-------
X
o
QJ
c
o
o
U
O 6 in
A 10 in
D 10 in
V 4in
2.5 in. AP (Coal)
2.5 in. AP (Coal)
AP (Coal)
dia.
dia.
dia., 2.5 in.
AP (Coal)
6 in. dia., 6.0 in. AP
(Efficiency Guarantee for Oil 3/)
10 15
Particle Diameter, fj,m
L-/V-
20
40
Figure 7. Fractional efficiency data for cyclone collection of
fly ash from coal- and oil-fired electric utility boilers.
40
-------
It should be emphasized that there are operating problems with the use
of multicyclone collectors for control of oil-fired boilers. If the tempera-
ture of the combustion gas is below the 803 dew point, the hygroscopicity and
corrosiveness of the oil ash can cause centrifugal collector operational and
maintenance problems. Build up of cement-like ash on tube and hopper surfaces
results in increased pressure drop as well as corrosion and cleaning problems .I/
There are several European waterwall incinerators equipped with cyclone
control, but no fractional control efficiency data were found for these plants.
The reported efficiency of a cyclone installed at Nashville Municipal (water-
wall) Incinerator is 57.7%.—' This unit is used as a precleaner for a scrub-
ber. Collection efficiency of ash from refractory walled incinerators by
cylones declines rapidly for dust smaller than 20 p,m.^ft/ Theoretical performance
models for cyclone collectors have been developed,.3!/ but the application of
theory to practical design problems has not been achieved.
WET SCRUBBERS
When applied to 170 MW coal-fired boilers, high efficiency wet scrubbers
have demonstrated an average 96% particulate removal efficiency at a pressure
drop of 37.4 mm Hg (20 in. I^O).^/ At. Boston Edison's Mystic Station, a Chemico-
Basic (Magnesia Slurry) scrubbing system installed on Boiler No. 6 (156 MW)
has achieved particulate removal efficiencies as high as/69^.-£§/ The Chemico-
Basic system employs a single stage Venturi scrubber. There have been operating
problems with the system installed at Mystic No. 6. Collection efficiency de-
creases rapidly with decreasing particle size. Negative efficiencies were ob-
served at the Mystic MgOx system for particle diameters below 1.5. pm, presumably
resulting from entrainment of scrubber solids..3-9-' The lower stages of the im-
pactor at the scrubber outlet were also found to be wet, indicating poor per-
on /
formance of the mist eliminators.—'
The estimated average efficiencies of scrubbers installed on oil-fired
electric generating stations (medium efficiency scrubbers) is 85.8% (Table 7).
The only waterwall incinerator plant presently equipped with a scrubber for
particulate control is Nashville's (Nashville Thermal, Transfer Corporation)
Riverside No. 2 Unit. A low energy wet scrubber with a Venturi rod insert is
used to reduce particulate emissions from 2.31 g/Nm3 (1.01 gr/dscf at 12% C02)
to 0.39 g/Nm3 (0.169 gr/dscf at 12% C02).—/ (This emission level is in viola-
tion of New Source Performance Standards for Incinerators.) Also at NTTS, there
are plans for installation of a high energy two-phase scrubber designed by
Chemico-Aronetics.3!/ This scrubber is similar in design to the Chemico-Aronetics
"Adtec" Scrubber.19_/ Based on the pilot plant studies, estimated particulate
emission levels are 0.049 g/Nm3 (0.0214 gr/dscf at 12% C02).3i/ This corre-
sponds to an overall efficiency for particulate removal of 98.0%. There are
no cost-performance data available at this time from Chemico. Reasons given
41
-------
were that (a) present data are based only on a pilot plant installation, and
(b) there are not enough data available at this time to determine whether con-
ventional construction materials will be sufficient to withstand the corrosive
incinerator flue gas .-Li/
The cost and performance of Venturi scrubbers is highly dependent upon
allowable pressure drop and economic tradeoffs between installed and operating
costs. Typical installed costs (1975 basis) for high energy 7ffturi scrubbers
for fly ash removal are $48 to $96/m3 ($1.35 to $2.70/acfm).—' The variation
in installed cost with gas volume and efficiency is shown in Figure 8.
ELECTROSTATIC PRECIPITATORS
When applied to cyclone-fired, coal-burning boilers, electrostatic pre-
cipitators have demonstrated particulate collection efficiencies ranging from
65 to 99.5%. On general pulverized coal boilers, control efficiencies are usu-
ally between 80 and 99.5%.—' As discussed, there are several difficulties in
adapting fly ash precipitator, designed for coal-fired boilers to handle fly
ash from firing oil. For control purposes, major differences are: (a) lower
ash density, (b) a higher concentration of carbonaceous, low resistivity partic-
ulate, (c) a higher percentage of submicron particulate, (d) a hygroscopic
fly ash which tends to agglomerate, and (e) a reduced volumetric ash loading.
As a result of these factors, when a coal-fired boiler with an electrostatic
precipitator is converted to oil with the precipitator unmodified, particulate
control efficiency is usually reduced. A typical collection efficiency for
an unmodified unit is reportedly about 45%.—' If the precipitator is modified,
however, control efficiencies approaching 90% can be realized.—'
Enlargement of collection electrodes can be used to minimize re-entrainment
of low density, low resistivity fly ashft/ which tends to easily lose its charge
when it comes into contact with the collection electrode. Reduction in gas
velocity, increased rapping intensity, and decreased frequency are also recom-
mended to minimize re^-entrainment.^,/ Because the average particle size is
smaller, a lower operating voltage, higher current, and longer gas treatment
path are recommended.-!' Other modifications are required because of ash han-
dling problems. The hygroscopicity of the particulate matter causes a solids
buildup on high tension electrodes, insulators, and collection curtains. When
allowed to cool, these solids absorb moisture, become difficult to remove and
cause arcing and shorts. By locating the precipitator on the hot side of the
air preheater, solids accumulation is reduced on high tension wires and col-
lection curtains. Build up on insulator bushings can be prevented by using
hot air ventilation. Hopper plugging can be avoided by either heating the hop-
per or employing a wet bottom system^.'
42
-------
o
U
-a
-------
The average collection efficiency for high efficiency electrostatic pre-
cipitators on oil ash is estimated, based on NEDS inventory data (Table 7)
to be 93.4%. The average collection efficiency of ESP installions on water-
wall incinerators is approximately 957o (see Appendix B).
Control Costs for Electrostatic Precipitator Control
Installed control costs for new high efficiency, multiple field electro-
static precipitators typically range from $26.5 to $123.6/m ($0.75 to $3.50/
ACFM), based on data for 1975.—/ Retrofitting for collection of oil fly ash
costs an additional $87 to $131/m3 ($2.50 to $3.75/acfm) (see Figure 9); a
new installation designed for oil costs $218 to $350/m3 ($6.25 to $10/acfm).-/
An alternative approach would be to install ESP collectors specifically
designed for refuse fly ash. This approach would be applicable for new installa-
tions, but would probably not be suitable for existing units which may need
to be converted back to coal at some later date. Installed cost and design
data for ESP units installed on existing waterwall incinerators are summarized
in Table 8.
Electrostatic Precipitator Performance Model
As previously discussed, combined firing of oil and MSW will cause de-
partures from fly ash properties and ESP operating conditions. Specifically,
changes may occur in dust loading, flue gas volume, particulate density, size
and resistivity. Because ESP controls are installed on most units planned for
combined oil-MSW firing studies, a considerable effort in the study was directed
toward adaptation of available information to develop an analytical model for
prediction of ESP performance. The model, which is intended to provide the
capability for rapidly estimating ESP performance for combined-firing conditions,
is described in detail in Appendix C.
44
-------
0
v>
o
(J
a
(A
"jjj
"o
Gas Volume, 104M3/min
2
Gas Volume, 105 ACFM
* Adjusted to 1st Quarter (January) 1976 Using Chemical
Engineering (CE) Plant Cost Index.43/
Figure 9. Total installed cost for electrostatic precipitators.—/
-------
Table 8. DESIGN AND COST DATA FOR ELECTROSTATIC PRECIPITATORS DESIGNED
FOR COLLECTION OF WATERWALL INCINERATOR FLY ASH
Installation A - 150 TPD Incinerator;
A. Sizing and performance data
Gas volume = 1,130 m3/min (40,000 ACFM)
Temperature = 290°C (560°F)
Efficiency = 97.5% by weight
Inlet loading = 4.6 g/Nm3 (2.0 gr/scf) at 12% C02
Residual = 0.1 g/Nm3 (0.05 gr/scf) at 12% C02
(Mass. Code)
Maximum excess air = 100%
Migration velocity (w) = 9.03 cm/sec
Gas velocity = 1.10 m/sec (maximum) (3.70 fps)
Precipitator size - One (1) precipitator
14 Gas passages Two (2) fields
4.6 m (15 ft) field height 25 cm (10 in.) passage spacing
Field length = 2.85 m (9.36 ft) x 2 = 5.70 m (18.72 ft)
Precipitator will remove 97.5% by weight of the incoming dry solid par-
ticulate provided the inlet dust is at least 4.6 g/Nm3 (2.0 gr/scf) at
12% C02. If the inlet dust load is less than 4.6 g/Nm3 (2.0 gr/scf) at
12% C02, the unit is guaranteed to have a maximum outlet emission of
0.1 g/Nm3 (0.05 gr/scf) at 12% C02.
B. Prices (budgetary)
1. Basic E/P with electrics, support
steel, access and dust valves $253,460
2. Thermal insulation for entire
precipitator including installation $ 34,125
3. Erection of E/P and auxiliaries
excluding L.V. wiring $ 93,160
Total $380,745
46
-------
Table 8. (Concluded)
Installation B - 750 TPD Incinerator
A. Sizing and performance data
Gas volume = 5,660 m3/min (200,000 acfm)
Temperature = 220°G (428°F)
Inlet loading = 4.6 g/Nm3 (2.0 gr/scf) at 12% G02
Residual = 0.1 g/Nm3 (0.05 gr/scf) at 12% C02
Efficiency = 97.5% by weight
Maximum excess air = 100%
Migration velocity (w) — = 9-.O em/sec.
Gas velocity = 0.15 m/sec (4.25 fps) (maximum)
Precipitator size - One (1) precipitator
37 Gas passages Two (2) fields
7.6 m (25 ft) Field height 25 cm (10 in.) Gas passage spacing
Field length = 3.328 m (10.92 ft) x 2 = 6.657 m (21.84 ft)
Precipitator will remove 97.5% by weight of the incoming dry solid par-
ticulate provided the inlet dust is at least 4.6 g/Nm3 (2.0 gr/scf) at
12% G02» If the inlet dust load is less than 4.6 g/Nm3 (2.0 gr/scf) at
12% CC>2, the unit is guaranteed to have a maximum outlet emission of
0.1 g/Nm3 (0.05 gr/scf) at 12% G02.
B. Prices (budgetary)
1. Basic E/P with electrics, support
steel, access and dust valves $420,130
2. Thermal insulation for entire
precipitator including installation $ 93,225
3. Erection of E/P and auxiliaries
excluding L.V. wiring $258,990
Total $772,345
Source: Wheelabrator Frye, Inc., Air Pollution Control Division, September
26, 1975.
47
-------
SECTION 5
CASE STUDIES
DISTRICT OF COLUMBIA
The Department of Environmental Services, District of Columbia, recently
initiated a combined-firing test program originally planned as an EPA re-
source recovery demonstration program which would have involved concurrent
firing of oil and coal with shredded municipal solid wastes (MSW). The planned
test boiler was the Potomac Electric Power Company (Pepco) Benning Station
Boiler No. 26. The planned refuse processing capacity was 22.7 MT/hr of raw
refuse, or approximately 16.3 MT/hr of refuse derived fuel (RDF) at a 75/25
air classifier cut; considered sufficient to remove 90 to 957o of the aluminum
cans for aluminum recovery in the heavy fraction, and also yield a fuel of
lower ash and higher burn-out than was prepared in St. Louis (see Tables 9 and
10). This refuse capacity is equivalent to 20 to 2570 of the heat input to
Benning Station No. 26 (75 MW) at 100% load. The actual steady-state refuse
capacity when firing oil would have been determined in the test. Three tests
each on coal only and oil only were planned to establish steady-state baseline
conditions. Following these preliminary tests, one to three steady-state tests
each with oil-plus-refuse and coal-plus-refuse were planned. Benning Station
Boiler No. 26 is equipped with ESP control; emission tests on Benning Station
No. 26 (firing oil) have recently been performed by York Research Corporation.^'
Notable characteristics of the planned combined firing tests at Pepco were:
(a) the refuse preparation facility would have employed primary and secondary
shredders, and it was planned to test the combustion characteristics of a wide
range of particle sizes; and (b) air emissions monitoring of NOX, SOX» Hg, HCl,
and particulates were to have been supplemented by an extensive program, con-
ducted by a separate subcontractor, to determine air emissions of trace heavy
metals, POM's, PCB's, and small organic moieties. Statistical methods were to
have been used for refuse sampling. Refuse analyses would have been done by
the same laboratory as for the St. Louis/Union Electric test program. Corro-
sion would have been monitored by probes, waste measurements, and test specimens
48
-------
Table 9. ESTIMATED MATERIALS BALANCE*
Estimated
% Composition
43.0
10.0
0.5
7.0
1.0
14.0
12.0
5.0
7.5
100
-p-
sO
Operations
Receiving
Shredder
Air Classifier
Lights
Heavies
Dust (Loss)
Pneumatic Feeder
Paper
10.75
10.75
10.75
9.93
.82
9.93
Glass
2.50
2.50
2.50
.49
2.01
.49
Non-
Fe
.13
.13
.13
.01
.13
-
.01
Fe
1.75
1.75
1.75
.03
1.72
-
.03
Al
.25
.25
.25
.04
.21
-
.04
Yard
Waste
3.50
3.50
3.50
2.89
.61
2.89
Food
Waste
3.00
3.00
3.00
1.98
1.02
1.98
Rags &
Wood
1.25
1.25
1.25
.50
.75
.50
Ash
Rock
1.87
1.87
1.87
1.07
.80
1.07
Total
(tons/
hr)
25
25
25
16.94
7.87
0.16
17.13
*Table based on feed rate of 25 tons per hour
Source: Department of Environmental Services of the District of Columbia, "Utilization of a Refuse-Derived
Fuel as Supplementary Fuel in an Oil and Coal Fired Electric Utility Boiler", Proposed to
U.S. Environmental Protection Agency fora Research, Development and Demonstration Grant,
April 1, 1975.
-------
Table 10. TARGET ANALYSIS OF REFUSE DERIVED FUEL AT SWRG-1
(Washington, D.C.)£/
Higher heating value 12,800-14,000 joules/g (11,600)1>/
(5,500-6,000 Btu/lb)
Moisture 20-25% (30%).k/
Ash 15-20% (30%)£/
Sulfur 0.3% (0.4%)^
Chlorine 0.6% (1.0%)^/
Particle size £3.8 emir/
(£ 1.5 in.)
Bulk density 32-116 kg/m3-/
(2-11 lb/ft3)
ai/ Dry weight basis.
J>/ Extreme value for acceptance by Pepco.
£/ The maximum specified dimension of each particle is 95%/weight less than
10 cm.
d/ The pneumatic delivery system as designed by Rader Pneumatics can deliver
RDF as low as 32 kg/m3 (2 lb/ft3) only at the highest flow rate of
400 m3/hr (14,000 GFH).
50
-------
to determine any incremental effects from burning RDF compared to the conven-
tional fuel. Slagging effects would have been monitored both visually and dur-
ing operations, by monitoring manometers indicating draft loss. The combined
firing test program and estimated emissions are described in detail in the
following subsections.
Program Status
The District of Columbia Department of Environmental Service submitted a
joint proposal with Pepco and National Center for Resource Recovery (NCRR) to
EPA, Office of Solid Waste Management Programs on April 1, 1975. The program
is now inactive as an EPA resource recovery demonstration program.
Refuse Derived Fuel (RDF) Preparation and Facilities
Shredding of MSW--
The DBS Solid Waste Reduction Center No. 1 (SWRC-1) is equipped with a
tipping floor and steel pan pit conveyor feeding a Williams 780, 2.684 x 10^
joules/hr (1,000 hp) horizontal hammermill. Discharge is onto a Jeffrey vibrat-
ing oscillating pan feeder conveyor and then to a rubber belt conveyor. Full
rated capacity of the shredder system is 25 tons/hr; the limitation of capacity
is the design of the conveying system, which was originally for oversized bulky
wastes (OBW), and not any limitation of the shredder. The feeder was to have
been modified for larger capacity.
Air Classification—
Modifications were planned to install a 25 ton/hr Triple/S "Vibrolutriator"
air classifier to SWRC-1. This classifier is the same as specified in Ames and
Chicago. A 75/25 split was expected in the air classifier between light and
heavy fractions, giving a maximum delivery of approximately 16.3 MT/hr of fuel.
The objective for the split was to recover from 90 to 9570 of the aluminum cans
while also dropping items such as wood, textiles, and heavy food wastes. Achiev-
ing these objectives would have yielded RDF of lower ash and higher burn-out
than was prepared in St. Louis.
Secondary Shredding--
Detailed engineering plans were approved by the district government for
the installation of a cyclone and associated blower and rotary valve (all owned
by NCRR) as the arrangement for de-entrainment of the air classifier light
fraction. The plans called for the cyclone to discharge through a rotary valve
which, in turn, would have discharged to either a positive displacement feeder
or to the second shredder.
The secondary shredder proposed for this work was a verticle type, such
as the Heil 42F. The proposed arrangement (Figure 10) was such that the
51
-------
Feed
Williams 780 Hammermill
Triple S Dynamics Air Class.
Frac.
1
Fraction
RDF Weigh
System
Cyclone
1
I
Shuttle
Conveyor
*Tower
PflAi 1 ^wc4*
r neu • oysr *
to Pit or
Conveyor
Secondary
Shredder
I ,
Pneumatic Delivery System to PEPCO Surge Bin
Figure 10. Process flow for preparation of RDF at
District of Columbia.
52
-------
secondary shredder could have been by-passed, or not, to deliver RDF to the
boiler. This arrangement would have permitted testing of a wide range of
particle sizes of RDF. It was considered impractical to obtain all sizes
using the secondary shredder only. A second planned method for achieving
variable particle size was to change the grate of the Williamson 780 shred-
der. A combination of both methods would have extended the particle size
range from 95% minus 10 cm to something under 5 cm. Particles smaller than
this would have been produced in the second shredder.
Characteristics of Test Boiler
Benning Station Boiler No. 26 is a tangentially fired boiler designed by
Combustion Engineering for a nominal capacity of 75 MW. The boiler system is
broadly similar to that used in St. Louis except with a smaller capacity.
Another difference is that this is a dry ash handling system as contrasted to
the Union Electric wet system. Boiler ratings for firing coal and oil are
listed in Table 11.
Boiler No. 26 was designed to burn either 100% coal or oil at the maximum
capacity rating (MCR). However, it may have been necessary to remove one level
of oil guns to accommodate RDF burners, which would have reduced capability to
67%.
The boiler modification to accept RDF would have consisted of removing one
level of oil guns and replacing them with (two) locally controlled, tiltable
refuse burners. If two-corner burning were found unacceptable, the system would
have been modified to permit burning at four corners. Combustion Engineering
has also investigated the possiblity of locating the refuse nozzles in the wind-
box without removing one level of oil firing to allow refuse-oil firing at the
MCR.
Installed Air Pollution Control Equipment
Boiler No. 26 employs a mechanical-electrical precipitator system which
was initially installed by Aerotec Corporation. The electrical section was
restored in 1968 by Research-Cottrell to original Aerotec specifications.—'
Design specifications are 99% for both collectors in series when burning 1%
sulfur coal at a flue gas volume of 9,344 m3/min (330,000 acfm), 165° C
(330°F).ft6_/ When firing oil in Boiler No. 26, the mechanical collector, which
has a tendency to plug under these conditions, is normally bypassed. When
the mechanical collector is bypassed, the design efficiency, under conditions
just described (i.e., for 1% sulfur coal) drops to 90.8%.-£/ Design parameters
for the electrostatic precipitator are summarized in Table 12.
53
-------
45 /
Table 11. PEPCO BENNING STATION BOILER NO. 26 DESIGN RATINGS—
Boiler rating
Steam: 306,000 kg/hr
(675,000 Ib/hr)
Temperature: 538°G
(1,000°F)
2
Pressure: 1,062 newtons/cm
(1,525 psig)
Coal firing
Heat in: 8.5565 x 1011 joules/hr
(811,000,000 Btu/hr)
Boiler efficiency: 89%
Heat out: 7.6153 x 10 joules/hr
(721,790,000 Btu/hr)
Oil firing
Heat in: 8.6933 x 10 joules/hr
(823,961,180 Btu/hr)
Boiler efficiency: 87.6%
7.6153 x 1011
(721,790,000 Btu/hr)
Heat out: 7.6153 x 10 joules/hr
54
-------
Table 12. CHARACTERISTICS OF ELECTROSTATIC PRECIPITATOR
Plate area—1,235.6 m2 (13,300 ft2)/section
Plate-to-plate spacing
(a) Inlet*/
(b) ChitletS/
Corona wire diameter—
233 _i
Specific collection area—264.5 m /10 m -min
(80.6 ft2/!,000 acfm)
Migration velocity~15 .0 cm/sec
(29.6 ft/min)
Design voltage—45 kv
oc/
Current density—16.8 nanoamps/cnr"
Electrical sets—two in parallel
Design efficiency—90.8% burning coal with 1% sulfur at approximately 75 MW
and 9,344 m3/min (330,000 acfm) into the precipitator
165°C (with mechanical collector bypassed)
a/ Data not available; assumed value was 25.4 cm (10 in.).
b/ Data not available.
~c/ Average for particulate emission tests made when firing No. 6 residual
oil..4!/
55
-------
The electrostatic precipitator is not presently expected to meet design
specification when firing coal. Therefore, the above specifications must be
used with caution. Efficiency measurements with coal and using both the me-
chanical and electrical sections have been made and indicate an average ef-
ficiency of only 96.5%.—/ Recent efficiency tests with No. 6 residual oil
reflect an efficiency of roughly 60%.li/ When burning coal, flue gas tem-
perature is typically 204°C (4009F, which is higher than that for which the
ESP unit was designed. Increasing temperature increases flue gas volume and
frequently decreases ESP collection efficiency for this temperature range.
District of Columbia Emission Regulations
Administering Agency:
Bureau of Air and Water Quality Control
Department of Environmental Services
25 K Street, N.E.
Washington, D.C. 20002
Fuel-Burning Particulate Emission:
For installations using more than 3,500,000 Btu/hr total input, the
particulate emission limitation shall decrease as the rate of heat input
increases as summarized below:
H E
(106 Btu/hr) (lb/106 Btu)
3.5 0.13
10 0.101
100 0.059
1,000 0.034
£ 10,000 0.02
H = total heat input in millions of Btu/hr
E = maximum emission in pounds of particulate matter per million Btu heat
input
H £ 3.5; E = 0.13; 3.5 < H < 10,000; E = 0.17455 H" 0.23522 H ;> 10,000;
E = 0.02
56
-------
Sulfur Oxides:
No person shall purchase, sell, offer for sale, store, transport, use,
cause the use of, or permit the use of, fuel oil which contains more
than 17o sulfur by weight in the District, if such fuel oil is to be
burned in the District.
On and after July 1, 1975, the sulfur content of such fuel oil shall not
exceed 0.5% by weight.
Nitrogen Oxides:
Emission limits for nitrogen oxide in fossil fuel fired steam generating
units of more than 100,000,000 Btu/hr heat input are as follows:
1. 0.20 Ib per million Btu heat input (0.36 g per million cal.),
maximum 2 hr average, expressed as N02»,when gaseous fossil fuel is
burned.
2. 0.30 Ib per million Btu heat input (0.54 g per million cal.),
maximum 2 hr average, expressed as N02»,when liquid fossil fuel is
burned.
3. 0.70 Ib per million Btu heat input (1.26 g per million cal.),
maximum 20 hr average, expressed as N02» when solid fossil fuel
(except lignite) is burned.
4. When different fossil fuels are burned simultaneously in any
combination the applicable standard (Ib NOX per Kr Btu) shall be
determined by proration, according to the following fomula:
x (0.20) + v (0.30) + z (0.70)
x + y + z
where x = the percent of total heat input derived from gaseous fos-
sil fuel
y = the percent of total heat input derived from liquid fossil
fuel
z = the percent of total heat input derived from solid fossil
fuel
57
-------
Estimated Performance of Installed Air Pollution Control Equipment
On the basis of information related to the planned test program, the per-
formance of the installed electrostatic precipitator was calculated for the
tests which would have involved combined firing of oil and MSW. ESP design
and performance data used in the calculation were summarized in Table 12. The
target composition data for the RDF fuel in Table 10 were used; actual com-
position data were not available. The ESP analytical model is described in
Appendix C. Dust loadings were estimated using Eq. (8). An average value of
0.0632 g/10 joules was used for total particulate from the oil portion of
the fuel based on data in Appendix A. The predicted variation in flue gas dust
loading with RDF heat input is shown in Figure 11, in comparison with an earlier
estimate by another group of investigators.—'
The predicted dust loading shown in Figure 11, for fr = 0.15 corresponds
approximately to average emission value for oil fired boilers, and combined
suspension firing of MSW with coal. The curve for fr = 0.50 corresponds ap-
proximately to the highest estimate for the refuse fly ash fraction..6-'
The calculated electrostatic precipitator performance is shown in Figure
12. Efficiencies range between 77 and 86%, depending on boiler load and
percentage of refuse fired. In making the calculation it was assumed:
1. The ESP unit would be put into proper operating condition.
2. The cyclone precleaner would be bypassed, as is done for oil
firing.
On the basis of data for flue gas dust loading and ESP performance, in
Figures 11 and 12, the particulate emissions under two different boiler loads
and with varying refuse heat input fraction were calculated. These results
are shown in Figure 13. As is evident from Figure 13, it is predicted that
particulate emissions will exceed New Source Performance Standards for
Fossil Fuel-Fired Steam Generators, except at low MSW and/or low boiler load.
Referring to District of Columbia Emission Regulations, the particulate emis-
sion standards for the District of Columbia are lower than new source standards
decreasing with fuel consumption rate. At the MCR, the District of Columbia
standard is 0.0161 g/106 joules (0.037 lb/106 Btu), about one-third the federal
new source standard.
Cost of Air Pollution Control
The cost of new or modified control systems to meet existing particulate
emission standards when Benning Station No. 26 is burning oil and MSW was not
determined.
58
-------
See Eq. (8) for definition of f .
Figure 11,
10 20
% MSW Heat Input
Effect of MSW fly ash fraction (fr) on calculated particulate
emissions (uncontrolled) from combined firing of MSW and
No. 6 residual oil (from Tables 1 and 2).
59
-------
.99
u
-------
J)
~2
•i-
o
U^
Ju
O
c
O
D
U
\
New Source Performance Standard for
Fossil Fuel-Fired Steam Generators
D.C. Air Quality Regulation @50 MW (Reduced Load)
D.C. Air Quality Regulation @75 MW (MCR)
10 20
% MSW Heat Input
30
Figure 13. Estimated particulate emissions (controlled) for combined
firing of oil and MSW at Pepco Benning Station No. 26.
61
-------
Pepco estimated the required control efficiency to be 99% when firing oil and
refuse at 10% MSW, based on 1.16 x 107 joules/kg, 7,936 kg/hr (5,000 Btu/lb,
8.75 tons/hr).^/ Based on a lower assumed inlet loading corresponding to
fr = 0.15 in Figures 11 and 13, the required control efficiency at 10% MSW
is 95.3%.
WILMINGTON, DELAWARE (New Castle County)
The State of Delaware, Division of Natural Resources, was awarded a |9
million resource recovery demonstration grant from EPA in October 1972.—'
Delmarva Power and Light Company has agreed to participate in the program,
and will modify either Edgemoor Station Units 3 or 4 to fire refuse. Delaware
Division of Natural Resource will construct a new processing plant in New
Castle County. There are few details available with regard to test schedule
and emission tests because these plans have not yet been finalized between
the State of Delaware and Delmarva Power and Light Company. If Edgemoor Sta-
tion Boiler 4 (160 MW) is the test boiler the Deinarva plan is to fire 5% MSW
with oil at 242 ton/day of MSW.—'
Project Status
207
A feasibility study was completed by Combustion Engineering in May 1974.—
Delaware plans to issue an RFP for boiler modifications within FY 76. Delmarva
estimates it will require approximately 1 year to make boiler modifications;
the Delaware schedule is presently to begin tests by 1979.—'
Refuse Derived Fuel Preparation and Facilities
Presently, the State of Delaware is operating the 500 ton/day facility
shown schematically in Figure 14 in New Castle County. The process uses air
classification, magnetic separation, screening, rising current, heavy media,
and electrostatic separation as well as optical methods for separating munic-
ipal solid waste into paper, ferrous and nonferrous metals, glass, and organic
fractions.^/ Regular markets have already been developed for glass, paper,
and metals.—' Typical analytical properties of the refuse fuel are listed in
Table 13.
Characteristics of Test Boilers
Edgemoor Station Boiler No. 3 is a tangentially fired boiler with a nominal
rating of 75 MW. Edgemoor Station Boiler No. 4 is a tangentially fired boiler
of the same design, except with a capacity of 160 MW. Both boilers are designed
for firing either fuel oil (No. 2 or No. 6 - residual) or pulverized coal. Both
boilers are equipped for flue gas recirculation. The only modification required
62
-------
f
)r
1 ' 1
Rolls
1
t
Elec.
Separator
Glass
Magnetics
o
Other
Nonferrous
—i i—
Aluminum
Flint/*~-\ Mixed Color
(Glass!
Figure 14. Schematic representation of materials recovery process,
New Castle, Delaware.51/
63
-------
Table 13. PROJECTED ANALYSIS OF REFUSE FUEL-
NEW CASTLE COUNTY, DELAWARE^/
(As fired - percent by weight)
Moisture 25.0
Ash 15.0
Sulfur 0.15
Chlorine 0.40
High heating value
joules/kg 1.314 x 10
(Btu/lb) (5,650)
High heating value (Dry, ash free)
joules/kg 2.093 x 10
(Btu/lb) (9,000)
Bulk density
kg/m3 64-176
) (4-11)
Ultimate (As fired - percent by weight)
Carbon 31.90
Hydrogen 4.70
Nitrogen 0.40
Oxygen 22.85
Sulfur 0.15
Moisture 25.00
Ash 15.00
Total 100.00
64
-------
for firing MSW will be to add one new nozzle in each corner for firing refuse.
Design ratings for Edgemoor Station Boilers Nos. 3 and 4 are summarized in
Table 14.
Installed Air Pollution Control Equipment
Edgemoor Station Boiler No. 3 is equipped with a multicyclone collector
with conventional reverse flow, designed by Western Precipitator. The design
efficiency is 83% at 7,044.6 m3/min and 149°C (248,775 acfm and 300°F). Design
ratings are summarized in Table 15.
Edgemoor Station Boiler No. 4 is equipped with a two stage ESP collector
designed by Research-Cottrell. The design efficiency (for coal or oil) is
95.0% at 12,560 m3/min and 135° C (440,000 acfm and 275°F). Design ratings are
summarized in Table 16.
20 /
Delaware Emission Regulations——'
Administering Agency:
Department of Natural Resources and Environmental Control
Air Resources Section
Tatnall Building
Dover, Delaware 19901
Particulates:
Emissions from any fuel-burning equipment shall not exceed 0.3 lb/10
Btu heat input.
Sulfur Oxides:
Sulfur content of distillate oil used in fuel-burning equipment is limited
to 0.3% by weight. Sulfur content of other fuels used in fuel-burning
equipment is limited to 1.0% by weight in New Castle County and to 2%
in Kent and Sussex counties. However, if between July 1, 1973 and October
1, 1974 the national secondary ambient air-quality standard for the
Metropolitan Philadelphia Interstate AQCR is exceeded due to an air-
containment source located in New Castle County, then the enforcement
agency may, after January 1, 1975, reduce the maximum allowable sulfur
content of fuel to a level not lower than 0.3% by weight. High-sulfur
fuel can be used if a state-approved S02-removal system is employed.
65
-------
Table 14. DELMARVA EDGEMOOR STATION BOILER DESIGN RATINGS
Unit 3 Unit 4
a/
Boiler manufacturer" C-E G-E
Type of firing^' , Tangential Tangential
Turbo generator size— , 75 MW 150 MW
Design fuel consumption" 20.4 m /hr 40.5 m /hr
, (128.34 Bbls(oil)/hr) (254.53 Bbls(oil)/hr)
Steam flow, kg/hr (coal)- , 260,800 484,765
Steam pressure, newtons/m 1.04 x 10 1.29 x 10^
b/ (1,500 psig) (1,850 psig)
Steam temperature, °C~ 538 538
b/ (1000°F) (1000°F)
Furnace volume, m^~ 1,319 2,574
b/ (46,600 ft ) (90,885 ft )
Efficiency, % (pulverized coal)- 89.33 89.99
Flue gas flow rate (100% MCR>2/ 7,045 m3/min 12,805 m3/min
/ (248,775 acfm) (452,187 acfm)
Exit gas temperature— 149° C 135° C
a c/ (300°F) (275°F)
Flue gas cleaning equipmentr~*— MCAX E
aj Information furnished by Delmarva Power and Light Company, December 30, 1975.
_b/ Information taken from G-E Power Systems Study.—/
£/ MCAX = multiple cyclones-conventional reverse flow with axial inlet;
E = electrostatic precipitator.
66
-------
Table 15. DESIGN DATA FOR CYCLONE COLLECTOR/DELMARVA EDGEMOOR NO. 3~
Manufacturer/Model No.--Western Precipitator/P37754A
Description—Multiple cyclones—conventional reverse flow; axial inlet
No. of sections
(a) Series—1
(b) Parallel—6
Tube arrangement—14 x 20.3 cm (8 in.)
Pressure drop—4.95 mm Hg (2.65 in HO)
Design efficiency—83% at 7,044.6 m3/min and 149°C (248,775 acfm and 300°F)
a/ Source: D» B. McClenathan, Delmarva Power and Light Company (Ref. 51).
67
-------
Table 16. CHARACTERISTICS OF ELECTROSTATIC PRECIPITATOR ON DELMARVA
EDGEMOOR UNIT NO. 4£/
Plate area—2,274.3 m2 (24,480 ft2)
Plate-to-plate spacing
(a) Inlet—22.86 cm (9 in.)
(b) Outlet—22.86 cm (9 in.)
Corona wire diameter—0.28 cm (0.109 in.)
233 -1
Specific collection area—182.6 m /10 m -min
(55.7 ft2/!,000 acfm)
Migration velocity—14.92 cm/sec
(29.4 ft/min)
Operating voltage--45 kv avg., 70 kv peak
?b/
Current density—70 nanoamps/cm •"
Electrical sets—two in parallel and two in series
Design efficiency—95.0% burning either coal or oil, with 2.7% sulfur coal,
at approximately 150 MW, flue gas volume of 12,459.5
m3/min (440,000 acfm) at 135°C (275°F)
a/ Data from Ref. 51.
bf Data from Appendix C, Figure C-3.
68
-------
Nitrogen Oxides:
After Janaury 1, 1975, emissions from fuel-burning equipment in New Castle
County rated 500 x 106 Btu/hr fuel input and greater will be limited to
0.2 (gas fuels) or 0.3 (other fuel) pound NOX (calculated as N02)/106
Btu heat input.
However, NOX laws do not apply to fuel-burning equipment when the heat
produced in the equipment is used for some purpose other than steam
production.
Estimated Performance of Air Pollution Control Equipment
Edgemoor Station No. 3--
The C-E Power Systems estimate for 10% MSW heat input is based on a col-
lection efficiency of 70% with an inlet dust loading of 0.275 g/106 joules
(0.64 lb/106 Btu).^P_/ This yields a net discharge rate of 0.0825 g/106 joules,
which is nearly twice the New Source Performance Standard for Fossil Fuel-Fired
Steam Generators, but within the Delaware emission regulation for fuel burning
equipment. MRI estimates a net discharge rate of 0.0939 g/106 joules, which
is only slightly higher than C-E, based on an efficiency of 70% and an inlet
dust loading shown in Figure 11. At 5% MSW, MRI estimates a net discharge rate
of 0.0564 g/106 joules, about 30% above the New Source Standard.
As discussed previously, there is not presently an adequate method for
modeling the performance of the multicyclone collector. However, the design
efficiency of 83% corrected for particle size (Table 6) yields an efficiency
of 67.5% based on the fractional efficiency curve in Figure 7. This is within
10% of the C-E estimate.20/
Edgemoor Station No. 4--
Estimated ESP performance for the unit on Edgemoor No. 4 is shown in Figure
15 as a function of flue gas volume. The estimate is based on the particle size
for refuse (Table 6) and the fuel properties listed in Table 13. The ESP per-
formance model is described in Appendix C.
On the basis of calculations made, the estimated range of emissions at
5% MSW is 0.034 to 0.077 g/106 joules, which is within the Delaware Emission
Regulation for Fuel Burning Equipment. The predicted emissions for Edgemoor
Station Boiler No. 4 are shown in Figure 16 as a function of percent MSW heat
input and refuse fly ash fraction fr at the MCR of 150 MW.
Cost of Air Pollution Control
On the basis of calculations made in this report, it is unlikely that
particulate emissions when Delmarva Edgemoor Station No. 4 is burning No. 6
69
-------
Flue Gas Volume, 103 M3/min
10 15
0% MSW
20% MSW-
!200
300 400 500
Flue Gas Volume, 1Q3 ACFM
600
700
Figure 15. ESP efficiency (predicted) for combined firing of oil and
MSW at Delmarva Edgemoor Station No. 4 (150 MW).
70
-------
0. r-
_0>
~o
"c
O
u
3
O_
O
O
c
O
3
U
O
a.
Delaware Emission Regulation
for Fuel Burning Equipment
New Source Performance Standard for
Fossil Fuel-Fired Steam Generators
10 20
% MSW Heat Input
Figure 16. Estimated particulate emissions (controlled) for combined
firing of oil and MSW at Delmarva Edgemoor Station No. 4.
71
-------
residual oil and 5% MSW will exceed applicable Delaware standards. Therefore,
no control modifications are indicated, provided MSW specifications are met.
Estimated emissions from Edgemoor No. 3 will probably exceed the Delaware
standard of 0.129 g/106 joules (0.3 lb/106 Btu) at 57, MSW unless control modi-
fications are made. The cost of such control modifications has not been de-
termined.
NEW YORK CITY
On March 6, 1975, the City's Board of Estimate awarded a $340,000 fea-
sibility and preliminary design contract for the firing of 1,000 tons/day of
solid RDF with oil in the Consolidated Edison No. 20 boiler at Arthur Kill
in Staten Island to Horner and Shifrin of St. Louis and Laramore, Douglass,
and Popham Engineering Consultants of New York City. They will also devise
an equitable formula for apportioning capital costs and determining the dollar
value of refuse to Con Ed, which is also a party to the contract.ll/
Originally the City assumed the cost of the contract. But a $50,000
federal grant was awarded for the project, signifying recognition of refuse
derived energy in the Project Independence strategy. Con Ed purchases most of
its oil from foreign sources. Firing of RDF in this pilot project will cut
this dependence by approximately 1,400 barrels of oil per day.ll/
A notable characteristic of this program is its size; Arthur Kill No.
20 has a net generating capacity of 325 MW and planned MSW heat input is 20%.—'
The Arthur Kill project will be the first of approximately 10 to 20 resource
recovery projects in New York City.ll/ A master plan, due in December 1975,
will recommend a construction timetable for specific processes at specific
sites.ll/ Arthur Kill No. 30 (500 MW, tangentially fired) is also included
in the master plan and will also fire MSW with No. 6 residual oil*2.'
Project Status
The Arthur Kill No. 20 project feasibility study was originally due to
be completed by Laramore, Douglass, and Popham Engineering Consultants (New
York City) by September 1, 1975. New York City granted an extension for com-
pletion by December 1. The major difficulties apparently result from the in-
adequacies of existing air pollution control systems. The electrostatic pre-
cipitator installed on Arthur Kill No. 20 is not presently operational, and the
electrostatic precipitator on Arthur Kill No. 30 has never met design specifica-
tions when firing No. 6 residual oil.ll/
72
-------
Refuse Fuel Preparation Facilities
Detailed plans for the refuse shredding system are not available at this
time. A new installation will be built at an estimated cost of $12 mi 11 ion.I5./
Boiler System Descriptions
Arthur Kill No. 20 is a front-wall fired boiler designed by Foster Wheeler
for a net generating capacity of 325 MW. This unit was originally designed
for coal and was retrofitted to fire oil.li/ Modifications to fire MSW include
additional burners for firing refuse and installation of a flue gas recircula-
tion system.—'
Arthur Kill No. 30 is a tangentially fired boiler designed by Combustion
Engineering for a nominal capacity of 500+ MW.—/ This unit will also be con-
verted for flue gas recirculation.
Installed Air Pollution Control Equipment
Both Arthur Kills Nos. 20 and 30 are presently equipped with Research-
Cottrell ESP systems. However, the ESP on Arthur Kill No. 20 is not operational
and the ESP on Arthur Kill No. 30 has not met design specifications when firing
No, 6 oil.—' There is no test data available for either unit when firing No.
6 oil.
New York City Particulate Air Emission Regulations
Total air emissions are rigidly controlled by a total allocation which
regulates air emissions from a given piece of fuel burning equipment to present
levels.—' A proposed emission standard for particulate is 0.043 g/10 joules
(0.1 lb/106 Btu) for fuel oil or refuse and fuel oil.—'
Estimated Performance of Installed Air Pollution Controls
Since the ESP unit on Arthur Kill No. 20 is not presently in operating
condition, the control efficiency is zero. New York City officials are appar-
ently aware of deficiencies in existing control systems for both Arthur Kills
Nos. 20 and 30. Because of the present early status of the Arthur Kill No. 30
project, estimation of control efficiency for combined firing is not considered
justified until city officials decide on which control system will actually be
used when oil and MSW are fired concurrently in either boiler.
Cost of Emission Control
Laramore, Douglass, and Popham Engineering Consultants, estimate the in-
stalled (erected) cost of a new ESP control system with 99% efficiency for
73
-------
Arthur Kill No. 20 to be $12 million.!/ There is apparently some difficulty
in getting the control manufacturer to guarantee efficiency for this applica-
tion, however.—/ There is no cost estimate available for Arthur Kill No. 30.
STATE OF CONNECTICUT (Bridgeport)
The Connecticut Resource Recovery Authority (CRRA) is currently planning
three resource recovery systems within the state. CRRA was formed following
a study sponsored by the State of Connecticut Department of Environmental
Protection. Funding is obtained both from the State and outside agencies.
Three programs presently planned include Bridgeport (United Illuminating
Company, Bridgeport Harbor Stations Nos. 1 and 2), Central Connecticut (Devon
Station), and South Central ConnecticutjLP-/ The South Central Connecticut pro-
gram will involve construction of boilers designed specifically for combined
firing of refuse and oil.—'
The Bridgeport project will be the first CRRA project undertaken. Orig-
inal plans called for a pilot scale test, with oil and MSW fired in Bridgeport
Harbor No. 1 (82 MW), for a period of about 2 months, followed by modification
of Bridgeport Harbor No. 2 (160 MW) to fire refuse. The objective of the pilot
scale test is to determine required modifications to the larger unit.
Notable aspects of these planned tests are that Bridgeport Harbor Nos.
1 and 2 are both pressurized, cyclone-fired boilers. If this type of system
can be successfully used to burn refuse, flue gas dust loading can be reduced
by about 50% of that for front-wall and tangentially-fired suspension boilers.
Project Status
Engineering feasibility studies have been completed by Gibbs and Hill
and C-E Power Systems.^!/ A contract was signed with Garrett Research Corpora-
tion (now Occidental Research) to supply the refuse fuel. Site preparation
began in the fall of 1975. Construction was planned to begin in the spring of
1976.—' Design changes are still being made.12./ Because of a recently formed
joint venture between Occidental Petroleum and Combustion Equipment Associates,—'
there is some consideration being given to use of the CEA Ecco II process.^2/
The pilot scale oil-MSW combined firing test was originally planned to
begin in January 1976,i°>/ but this test was recently delayed because of dif-
ficulties in locating a source of refuse fuel and difficulties in funding.^2/
CRRA has recently applied for a $900,000 ERDA grant.59/
74
-------
Boiler System Descriptions
Bridgeport Harbor No. 1 is a pressurized, cyclone-fired boiler designed
by Babcock and Wilcox for a net output of 82 MW. The unit has two cyclone
burners.
Bridgeport Harbor No. 2 is a pressurized, cyclone-fired boiler designed
by Babcock and Wilcox for a net output of 170 MW, of similar design to Bridge-
port No. 1, except with five cyclone burners.
Refuse Preparation Facilities
The design of the refuse preparation facility has not yet been finalized.
Installed Air Pollution Control Equipment
Both Bridgeport Harbors Nos. 1 and 2 are equipped with electrostatic pre-
cipitators designed by Research-Cottrell.
Connecticut Air Emission Regulations
Administering Agency:
Department of Environmental Protection
State Office Building
Hartford, Connecticut 06115
Existing and new fuel burning equipment must comply with the following
regulations
Particulates:
Emissions are restricted to 0.1 lb/106 Btu heat input. The heat-input
value is the equipment manufacturer or designer's guaranteed maximum in-
put, whichever is greater.
Sulfur Oxides:
Fuels are restricted to a maximum sulfur content of 0.5% by weight (dry
basis). Under fuel-shortage conditions, variances can be obtained for
burning higher sulfur fuels on a temporary basis. High sulfur fuels also
can be burned if state-approved stack-gas cleaning equipment is capable
of limiting total sulfur-compound emissions to the ambient air to 0.55
Ib S02 (equivalent)/106 Btu gross heat input, and if waste discharges
from the stack gas cleaning system into State waters are approved by
State authorities.
75
-------
Nitrogen oxides:
Emissions from fuel burning equipment rated above 250 x 10 Btu/hr heat
input are limited to 0.2 (gas), 0.3 (oil), or 0.7 (coal) pound NOX
(expressed as N02)/106 Btu heat input.
Estimated Performance of Air Pollution Control Equipment
The Connecticut Resource Recovery Authority is not optimistic about the
performance of ESP units on either boiler. As with most units designed for
coal, the efficiency is expected to drop to 60 to 70% for oil, 70 to 85% for
oil plus refuse, depending on refuse composition and MSW heat input.
Cost of Emission Control
Because the test program is not yet final, and refuse fuel characteristics
and heat input are not known, it is not possible to make significant estimates
of emissions at this time. However, we do know that CRRA and United Illuminating
Company are not presently planning replacement or modification of either ESP
unit. At 70% efficiency, particulate emissions could well be as high as 0.052
g/106 joules (0.121 lb/106 Btu), at 10% MSW heat input, even allowing for a
50% reduction in fly ash because of cyclone burner characteristics. This would
exceed state regulations of 0.043 g/106 joules (0.1 lb/106 Btu) applicable to
new and existing fuel burning equipment.
76
-------
SECTION 6
RECOMMENDATIONS
Since the present study has emphasized the application of existing air
pollution control methodology for particulate air pollutants from power boilers
firing municipal solid wastes and auxiliary fuel oil, it seems appropriate to
point out some deficiencies in the control technology, as applied to this as-
pect of resource recovery. An attempt should be made to resolve these diffi-
culties in future studies. Difficulties encountered in the present study are
discussed according to the general types of control systems considered.
ELECTROSTATIC PRECIPITATOR CONTROL
In adopting existing theoretical studies to the development of a practi-
cal performance model to predict ESP performance for combined firing applica-
tions, there was found to be no quantitative information on the effects of:
1. Fly ash density;
2. Re-entrainment; and
3. Sneakage or bypassing.
This is somewhat surprising, in consideration of the influence of these ef-
fects both on equipment cost and collection efficiency. For example, one source
recommended a reduction in flow velocity directly proportional to a decrease
in fly ash density.12/ Referring to Figure 8, this would increase installed
cost of a new ESP unit in approximately the same proportion as the decrease
in flow velocity. There are sufficient data on the resistivity of oil ash and
refuse fly ash, both of which are relatively high in carbon compared to coal
fly ash, to conclude that a significant proportion of the ash will be low in
resistivity compared to the average. In other words, there is probably a much
larger resistivity range for oil and/or refuse firing than for a given coal.
The low resistivity fraction is comprised primarily of carbon. Such low resis-
tivity particulate tends to lose its charge easily on contact with the collec-
tion electrode, and be re-entrained. In the case of oil ash, a modification
77
-------
of the shape of the collecting electrodes is recommended to prevent re-
entrainment. However, there is no method available to quantatively relate
the fraction re-entrained with ash properties (resistivity, shape, and
density) and system parameters (velocity and electrode geometry). There has
been some work done on bypassing, or sneakage, but this effect has not yet
been quantitatively defined in terms of system variables.
What is needed is an order-of-magnitude analysis of the effects of fly
ash density, re-entrainment, and bypassing based on both approximate analysis
and a survey of available data including contacts with equipment manufactur-
ing firms. None of the above effects were included in the model used to esti-
mate ESP performance in the present study. This is not too important in terms
of the present objectives, since these effects result in a control performance
which is lower than predicted. However, in consideration of the preceding dis-
cussion, it is perhaps not too surprising that one source^./ who attempted to
acquire cost and performance information for a new ESP unit for combined oil-
MSW firing was unable to obtain a performance guarantee.
CYCLONE CONTROL
Because of the relatively low cost of cyclone control, this type of sys-
tem would seem to be useful, in conjunction with ESP control, to collect low
density, weakly charged particulate if placed after the ESP. The function of
the cyclone used in this fashion would be to collect the coarse fraction of
ESP re-entrainment losses. There is at least one such intallation on an oil-
fired boiler.l/
The theoretical prediction of cyclone pressure drop and collection effi-
ciency still is not possible because of complexities of flow fields. Other
factors, such as the tendency of cyclone collectors to plug when in service
on oil ash, and the performance decline in the corrosive atmosphere of in-
cineration flue gases, need to be examined.
As in the case of ESP control, there needs to be additional work done
on this type of control system specific to the application of combined fossil
fuel-MSW combustion. This would include contacts with vendors, literature
review, and analysis beyond the scope of the present study to develop guide-
lines for use in combined firing applications.
SCRUBBER CONTROL
High performance scrubbers and possibly wet electrostatic precipitators
have utility for collection of particulates, gaseous pollutants (SO , NOX,
and others) and potentially hazardous trace metals. There were no scrubber
or wet ESP units installed on the boilers evaluated in the present study.
78
-------
However, there is at least one high performance Chemico Arotec scrubber planned
for installation at Nashville.36/ The manufacturer was contacted regarding the
feasibility of using scrubbers for emission control on boilers where MSW and
fossil fuels are being fired. At the time the contact was made, Chemico was
very cautious about this application, saying that some reports based on pilot
plant data may have been premature. There is a general reluctance on the part
of the gas cleaning industry to recommend scrubbers for service on incinerator
flue gases. The problem, as in cyclone control, is with corrosivity. One other
equipment vendor was contacted regarding the possible application of wet elec-
trostatic precipitators for combined MSW-fossil fuel firing. The manufacturer
was optimistic regarding this application, but cost and design data have not
been received.
Both wet scrubbers and a wet ESP unit have recently been evaluated on
a pilot plant scale as part of the evaluation of EPA's "Landgard" Demonstra-
tion Project in Baltimore, Maryland. The Landgard system is a pyrolysis reac-
tor which fires approximately 7.1 gal. of No. 2 fuel oil per ton of MSW.
Maryland particulate emission regulations for this facility are 0.013 g/Nnr
(0.03 gr/dscf). Particulate emissions measured in shakedown runs in the spring
of 1975 reportedly were in the vicinity of 0.069 g/Nm3 (0.2 gr/dscf)J>°_/ In
the summer of 1975, two Teller Crossflow Nucleating scrubbers were evaluated
adiabatically and in condensing modes. These systems could not achieve state
standards but could achieve the federal standards of 0.18 g/Nm3 (0.08 gr/
dscf).£2/ A Micropuls wet ESP test could meet the state code; however, tests
were not conclusive.££/ An expanded control evaluation program is now planned
which will include several other control systems.6£/
Because of the increasing public awareness of problems resulting from
gaseous and trace metal pollutants, an in-depth study of these control methods
appears justified.
79
-------
REFERENCES
1. Darby, K., and C. Whitehead. The Performance of Electrostatic Pre-
cipitators in Relation to Low Sulfur Fuels. In: Proceedings of the
Second International Clean Air Congress, Washington, D.C., 1970. pp.
911-922.
2. Smith, W. S. Atmospheric Emissions from Fuel Oil Combustion, U.S.
Public Health Service Publication No. 999-AP-2, November 1962.
3. Burdock, J. L. Centrifugal Collectors Control Particulate Emissions
from Oil-Fired Boilers. Tappi, 56(6):78-82, June 1973.
4. Sahagian, J., R. Dennis, and N. Supernant. Particulate Emission Control
Systems for Oil-Fired Boilers. EPA-450/3-74-063, December 1974.
5. Steam, Its Generation and Use. Babcock and Wilcox Company, New York,
1963.
6. Tamlyn, W., Laramore, Douglass, and Popham, Inc. Private Communication,
December 12, 1975.
7. Maartmann, S. Collection of Dust from Oil-Fired Boilers in Multi-
cyclones and Electrostatic Precipitators, In: International Clean
Air Congress, London, October 4-7, 1966.
8. Pershing, D. W., et al. Effectiveness of Selected Fuel Additives in
Controlling Pollution Emissions from Residual Oil-Fired Boilers. EPA-
650/2-73-031, October 1973.
9. Bunz, P., H. P. Niepenberg, and L. K. Rendle. Influences of Fuel Oil
Characteristics and Combustion Conditions on Flue-Gas Properties in
Water-Tube Boilers. J. Inst. Fuel, p. 406, September 1967.
10. Rigo, H. G. Systems Technology Corporation, Private Communication,
October 16, 1975.
80
-------
11. Shannon, L. J., et al. St. Louis/Union Electric Refuse Firing
Demonstration Air Pollution Test Report. EPA 650/2-75-037, April
1975.
12. Fiscus, D. E., P. G. Gorman, and J. D. Kilgroe. Bottom Ash Generation
in a Coal-Fired Power Plant when Refuse-Derived Supplementary Fuel is
Used. Presented at the ASME Solid Waste Processing Conference, Boston,
Massachusetts, May 23-26, 1976.
13. Gorman, P. G. Midwest Research Institute, Private Communication,
May 5, 1976.
14. Gershman, H. National Center for Resource Recovery, Private Communica-
tion, January 14, 1976.
15. Kilgroe, J. D., L. J. Shannon, and M. P. Schrag. Emissions from the
Suspension Firing of Municipal Solid Waste and Pulverized Coal. Paper
Presented at the 68th Annual APCA Meeting, Boston, Massachusetts, June
15-20, 1975.
16. Hall, E. H., et al. Refuse Combustion in Fossil Fuel Fired Steam
Generators. Final Report, Contract No. 68-02-0611, Task 9 to Office
of Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Battelle-Columbus Laboratory, September 23, 1974.
17. Hall, E« H. Battelle-Columbus Laboratory, Private Communication,
August 29, 1975.
18. Blackwood, T. R., and W. H. Hedley. Efficiencies in Power Generation.
EPA-650/2-74-021, March 1974.
19. Mullin, J. F. Combustion Engineering, Inc., Private Communication,
August 30, 1975.
20. Mullin, J. F. Combustion Engineering, Inc. Solid Waste Study Report
for Delmarva Power and Light Company's Edgemoor Power Plant. OES No.
11828, May 31, 1974.
21. Schweiger,R. W. Power from Waste. Power, pp. 5-21, February 1975.
22. Stabenow, G. Ovitron Corporation, Private Communication, October 7,
1975.
81
-------
23. Vandegrift, A. E., et al. Particulate Pollutant System Study Volume
III - Handbook of Emission Properties. EPA Contract No. CPA 22-69-104,
May 1971.
24. Funkhouser, J. T., et al. Manual Methods for Sampling and Analysis of
Particulate Emissions from Municipal Incinerators. EPA-650/2-73-023,
September 1973.
25. Oglesby, S., and G. Nichols. A Manual of Electrostatic Precipitator
Technology, Part II - Application Areas. National Air Pollution Con-
trol Administration, Contract No. CPA 22-69-73, August 25, 1970.
26. Perry, R. E. A Mechanical Collector Performance Test Report on an Oil-
Fired Power Boiler. Combustion, pp. 24-28, May 1972.
27. Roberts, R. M., et al. Systems Evaluation of Refuse as a Low Sulfur
Fuel, Volume II - Appendices, Report to Environmental Protection Agency,
Contract No. CPA-22-69-22, Envirogenics Company, November 1971.
28. McGarry, F. J., and C. J. Gregory. A Comparison of the Size Distri-
bution of Particulates Emitted from Air, Mechanical, and Steam Atomized
Oil-Fired Boilers. J. Air Poll. Contr. Assoc., 22(8):636, August 1972.
29. Day and Zimmerman. Special Studies for Incinerators. Report Prepared
for the Government of the District of Columbia, Department of Sanitary
Engineering, U.S. Department of Health, Education, and Welfare,
Cincinnati, 1968.
30. Shannon, L. J., et al. Fine Particulate Emission Inventory and Control
Survey. EPA Contract No. 68-02-1324, Task No. 1, Midwest Research In-
stitute, January 1974.
31. Schulz, E. J., et al. Harrisburg Municipal Incinerator Evaluation.
Final Report to Harrisburg Incinerator Authority, Battelle-Columbus
Laboratory, December 17, 1973.
32. McCain, J. D., and G. B. Nichols. Letter Report to R. C. Lorentz,
Environmental Protection Agency, Control Systems Laboratory, Southern
Research Institute, February 8, 1974.
33. Rigo, H. G. Predicting Emissions from the Use of Refuse Derived Fuel.
Paper Presented at the Midwest Section Meeting, Air Pollution Control
Association, September 1975.
82
-------
34. McCain, J. D., A. B. Spencer, and W. B. Smith. Precipitator Operation
as Part of Midwest Refuse Firing Demonstration Project. Coal Fire Test.
Preliminary Report to Midwest Research Institute, Southern Research In-
stitute, December 20, 1974.
35. Bagwell, F. A., and R. G. Velte. New Developments in Dust Collecting
Equipment for Electric Utilities, j. Air Poll. Contr. Assoc., 21(12):
781, December 1971.
36. Enforcement Proceedings. United States Environmental Protection Agency
Region IV in the Matter of Nashville Thermal Transfer Corporation,
April 7, 1975.
37. Rao, A. K., M. P. Schrag, and L. J. Shannon. Particulate Removal from
Gas Streams at High Temperature and/or High Pressure. EPA Contract No.
68-02-1324, Task 30, Draft of Task Report, Midwest Research Institute,
May 23, 1975.
38. Koehler, G. R., and E. J. Dober. New England S02 Control Project Final
Results. In: Proceedings of the Symposium on Flue Gas Desulfurization,
Atlanta, November 1974, EPA-650/2-74-126-b, December 1974.
39. Statnick, R. M., and D. C. Drehmel. Fine Particle Control Using Sulfur
Oxide Scrubbers. Presented at the 67th Annual Meeting, Air Pollution
Control Association, Denver, Colorado, June 9-13, 1974.
40. McCain, J. D. Evaluation of Aronetics Two-Phase Jet Scrubber. EPA-650/2-
74-129, December 1974.
41. Whitwell, J. A. Chemico, Private Communication, October 10, 1975.
42. Power, p. 41, November 1975.
43. Economic Indicators. Chemical Engineering, 83(9):7, 1976.
44. Allard, N. Potomac Electric Power Company, Private Communication,
December 9, 1975.
45. Department of Environmental Services, District of Columbia, Utilization
of a Refuse-Derived Fuel as a Supplementary Fuel in an Oil- and Coal-
Fired Electric Utility Boiler. Proposal to U.S. Environmental Protection
Agency for a Research, Development, and Demonstration Grant, April 1, 1975.
83
-------
46. Weiss, M, Potomac Electric Power Company, Private Communication, December
8, 1975.
47. Alter, H. A. National Center for Resource Recovery, Private Communica-
tion, August 4, 1975.
48. Gershnian, H. National Center for Resource Recovery, Private Communica-
tion, September 12, 1975.
49. Hopper, R. E. A Nationwide Survey of Resource Recovery Activities.
U.S. Environmental Protection Agency, Office of Solid Waste Manage-
ment Programs, Publication SW-142, January 1975.
50. Cook, F. Delmarva Power and Light Company, Private Communication,
August 22, 1975.
51. McClenathan, D. B. Delmarva Power and Light Company, Private Communica-
tion, December 8, 1975.
52. Bisselle, C., et al. Urban Trash Hethanatipn Background for a Proof-
of-Concept Experiment. Mitre Corporation, NSF/RANN Contract NSF-C 938,
February 1975.
53. Low, R. A. Energy Conversion in New York. In: Proceedings First Inter-
national Conference on Conversion of Refuse to Energy, Montreux,
Switzerland, November 3-5, 1975.
54. O'Reilly, L. J. City of New York, Environmental Protection Agency,
Private Communication, August 4, 1975.
55. O'Reilly, L. J. City of New York, Environmental Protection Agency,
Private Communication, December 18, 1975.
56. Busch, R. Connecticut Resource Recovery Authority, Private Communica-
tion, August 7, 1975.
57. Finney, C. Garrett Research Corporation, Private Communication, August
22, 1975.
58. Chemical Engineering, December 8, 1975.
59. Busch, R, Connecticut Resource Recovery Authority, Private Communication,
November 21, 1975.
84
-------
60. Sussman, D. EPA Office of Solid Waste Management Programs, Private
Communication, May 6, 1976.
61. Govan, F. A. Relationship of Particulate Emissions Versus Partial to
Full Load Operations for Utility-Sized Boilers. Presented at Third
Annual Industrial Air Pollution Control Conference, Knoxville, Tennessee,
March 29-30, 1973.
62. Zurn Industries, Inc. A Mechanical Collector Performance Test Report
on an Oil-Fired Power Boiler, November 1971.
63. Peterson, I. J. Experience with the Operation of Electrostatic Precipi-
tators on Oil-Fired Boilers. Presented at 13th Annual New England Sec-
tion APCA Meeting, 1969.
64. Test Reports Provided by Source Sampling Section, Division of Air
Resources, New York State Department of Environmental Conservation.
65. Test Reports Provided by Air Compliance Section, United States Environ-
mental Protection Agency, Region I.
66. Allen, R. N. Chicago Northwest Incinerator Test Number 71-CI-ll.
Resources Research, Inc., EPA Contract No. CPA 70-81, September 1971.
67. Gooch, J. P., J. R. McDonald, and S. Oglesby. A Mathematical Model of
Electrostatic Precipitation. EPA 650/2-75-073, April 1975.
85
-------
APPENDIX A
PARTIGULATE EMISSIONS FROM OIL-FIRED ELECTRIC UTILITY BOILERS
87
-------
PARTIGULATE EMISSIONS FROM OIL-FIRED BOILERS
Boiler
Identification
Participate loading
Undisclosed
Description
C-E tangentially
fired, 250 MW
oo
00
Franklin Station,
Rochester, Minn.
Boston Ed.,
Mystic No. 6
Boston Ed.,
Mystic No. 6
Hartford Elect.,
Middletown No. 3
C-E Type VP-14-W
G-E tangentially
fired, 156 MW
G-E tangentially
fired, 156 MW
B and W 5 cyclone
universal pressure
unit, 240 MW
Boston Ed.,
Mystic No. 3—
e/
°L Load
!(>(#
8*r,
\%l
&
Jjjy
100^
94
92
97
95
80
80
80
51
51
&"F~
J2'
™
Of
-
D/
_/£'
—
% Excess
air
15.8
15.5
20.8
27.3
38.2
15.4
18.5
24.0
.
-
-
-
i
-
_
-
-
-
-
-
_
(e/10"
Inlet
-
..
0.1033
0.1256
0.2327 -
0.1015
0.0727
0.1269
0.0538
0.4010s'
0.7622s,
0.8676-,
0.6752s,
0.4113s'
0.0820s'
0.0361^
0.0293s'
0.0784
0.15
0.0490
0.0965
-
-
_
ioulesj
Cntlet
0.0085
0.0076
0.0061
0.0052
0.0210
0.0133
0.0172
0.0331
0.0310
0.0361
0.0478
0.0293
0.0501-'
0.0752s'
0.0633s'
0.0482s/
0.0418s
0.0104^'
0.0127s'
0.0086s
0.0486
0.0645
0.0142
0.0637
0.1050
0.0663
0.0663
Control
method^/ Reference
ESP
61
M
62
37
38
ESP
ESP
63
(Table 2)
-------
oo
VO
Boiler
Identification
Consolidated
Edison, Ravenswood No. 30~
Consolidated
» .
Description
Astoria, No. 5Cr
No.
No.
L. I. Lighting Company
Northport, Suffolk County
Unit No. 3
L. I. Lighting Company
Port Jefferson, Suffolk County
Unit No. 3
L. I. Lighting Company
Barrett Station No. 2
Island Park, Nassau County
375 MW
C-E
type
'
C-E tangentially
fired boiler
% Load
Excess
air
Particulate loading
(g/106 loules)
98.1
98.1
98.1
98.1
96.8
96.8
•t
-
-
_
-
-
_
11
11
11
11
8.7
8.5
31.1
33.6
30.4
42.7
41.5
41.6
39.7
folet
0.0087
0.0070
0.0095
0.010
0.011
Outlet
0.0073
0.0034
0.0052
0.0052
0.0052
0.0044,,
0.0049-*'
0.0150-*'
0.0085s
0.0057
0.0042
0.0116
0.0120
0.0125
0.0198
0.0261
0.0207
0.0197
Control .
method"
ESP
ESP
ESP
ESP
ESP
ESP
Reference
4
(Table 2)
4
(Table 2)
4
(Table 2)
4
(Table 2)
64
64
64
-------
Boiler
Ident i ficat i on
Description
Particulate loading
% Excess (g/10 joules)
air Inlet Outlet
Control.
Reference
Niagara Mohawk Power Corporation C-E tangentially
Albany Steam Station, Glenmont, fired boiler
New York
64
No. 1-
(1974 test series)
No. 2^7
(1974 test series)
No. 3*'
(1974 test series)
h/
No. 4—
(1974 test series)
No. 1—
(1973 test series)
No. 2-
(1973 test series)
No. 3-
(1973 test series)
No. 4—
(1973 test series)
Boston Edison Company
New Boston Station, Unit No. 1
Boston toison Company
Ednar Station, No. 10
87
96
95
31.9
104
128
0.0619
0.0821-
0.0576 .
0.0791-
0.0662 .
0.0765-
0.0464 .
0.0692^
0.0367
0.0367
0.0411
0.0355
0.0318
0.046CP
Unknown^7
Unknown^'
65
65
-------
Particulate loading
Boiler
Identification
Boston Edison, Edgar Station,
Unit No. 9
Boston Edison Company
L-Street, Unit No. 76
Boston Edison Company
Kneeland Street, Unit No. 2
Boston Edison Company
Mystic Station, Unit No. 5
.b.e/
Undisclosed
Description
% Load
95
95
96
96
98
87
85
% Excess
air
112
122
304
181
206
75
77
(e/10° ioules)
Inlet Outlet
0.0219
0.0275^
0.0189^
0.0241 .
0.0813d'
0.0529 ,
0.0641-'
Control ,
, ^a/
method~
Unknown"
Unknown"
Unknown1"1
Unknown1^
Reference
65
65
65
65
0.0585
0.0641
0.0370
0.0568
0.0404
0.0809
0.0559
0.0568
0.0336
0.0426
0.0757
0.0805
0.0676
0.0688
0.0318
0.0611
0.0645
0.0417
0.0602
0.0413
0.0426
ESP
(Table 2)
-------
Boiler
Identification
Description
Undisclosed"
Particulate loading
% Excess (g/106 joules) Control ,
air Inlet Outlet method""
0.0090 ESP
Reference
(Table 2)
Undisclosed
Hartford Elect., Middletown
No. 2£/
Undisclosed—*~"
vo
0.0965
0.0954
0.1507
0.0275
0.0288
0.0301
0.0245
0.0288
0.0357
0.0267
0.0211
ESP
ESP+C
(Table 2)
4
(Table 2)
Undisclosed
Undisclosed""
0.0352
0.0489
0.0393
0.0623
0.0522
0.0065
0.0095
0. 0043
0.0047
0.0151
0.0086
0.0056
0.0043
0.0396
0. 0366
0.0181
0.0146
0.0207
0.0198
0.0224
0.0344
(Table 2)
(Table 2)
-------
Particulate loading
Boiler
Identification
vo
u>
Undisclosed
Boston Edison
L. Street, No. 68
No. 74
No. 75
No. 76
Boston Edison
N. Boston, No. 1
No. 2
Boston Edison
Kneeland- Street, No. 1
No. 2
No. 4
Description
% Load
% Excess (e/10° loules)
air Inlet
M ••
•* Wt
0.0168
0.0827
0.0290
0. 0665
0.0654
0.0228
0.0281
0.0137
0.0133
0.0258
0.0211
0.0055
0.0166
0.0052
Outlet
0. 0026
0.0013
0.0026
0.0026
0.0168
0.0827
0. 0290
0.0665
0.0654
0.0228
0.0281
0.0137
0.0133
0.0258
0.0211
0..0055
0.0166
0.0052
Control .
method" Reference
ESP
None
None
None
None
None
None
None
None
None
(Table 2)
(Table 12)
(Table 12)
(Table 12)
(Table 12)
(Table 12)
(Table 12)
(Table 12)
(Table 12)
(Table 12)
-------
Boiler
Identification
Boston Edison
Minot Street, No. 6
No. 7
Boston Edison No. 9
Edgar Station
Description
% Load
Excess
air
Particulate loading
(e/106 loules)
vo
No. 10
Inlet
0. 0094
0.0110
0.0258
0.0564
0.0585
0.0254
0.0416
0.0375
0.0355
0.0400
0.0331
0.0272
0.0185
0.0222
0.0223
0.0238
0.0217
0.0495
0.0204
0.0170
0.0293
0. 0356
0.0290
0.0244
0. 0336
0.0296
0.0244
0.0302
0.0176
0.0271
0.0213
Outlet
0. 0094
0.0110
0.0258
0.0564
0.0585
0.0254
0.0416
0.0375
0.0355
0.0400
0.0331
0.0272
0.0185
0.0222
0.0223
0.0238
0.0217
0.0495
0.0204
0.0170
0.0293
0,0356
0.0290
0.0244
0.0336
0.0296
0. 0244
0.0302
0.0176
0.0271
0.0213
Control .
method~
None
None
None
Reference
(Table 12)
(Table 12)
(Table 12)
None
(Table 12)
-------
Particulate loading
Boiler
Identification
No. 11
Boston Edison Company
Mystic Station No. 3
No. 5
Boston Edison Company
Mystic No. 6
Description
Braintree Elect. Potter
Station No. 1
No. 2
7, Excess (a/loo
air Inlet
0.1149
0.0478
0.0430
0.0209
0.0719
0.1050
0.0200
0.0455
0.1484
0.1024
0.0784
0.0490
0.0113
0.0402
0.0278
0.0175
0.0527
0.0564
0.0456
0.1196
0.0739
0.1213
0.0468
0.0234
0.0321
0.0306
0.0988
0.0398
0.0524
ioules )
Outlet
0.1149
0. 0478
0.0430
0.0209
0.0719
0.1050
0.0200
0.0455 /
0.1484-f,
0.1024f;
0.07847,
0.04902
0.0113
0.0402
0.0278
0.0175
0.0527
0.0564
0.0456b/
0.1196- ,
0.0739s/.
b/
0. 1213r'
0.0469s
0.0234-^
D/
0.0321-f,
0.0306-
0.0988-^
0.0398s
^,^.^-
Control ,
method— _ Reference
None 4
(Table 12)
None 4
(Table 12)
4
(Table 13)
None 4
(Table 12)
None 4
(Table 12)
(Table 13)
None 4
(Table 13)
None 4
(Table 13)
-------
Particulate loading
Boiler % Excess (g/lO0 loules) Control .
Identification Description % Load air folet Outlet method^ Reference
No. 4 - 0.0434 0.0434-( None 4
0.0565 0.0565s (Table 13)
o.oeos o.oeoak/
_a/ ESP = electrostatic precipitator, C = cyclone, M = multicyclone, V — venturi scrubber.
b/ Fuel Additives; First Test Series (Undisclosed Boiler), MgO in liquid carrier used to control SO^ emissions and maintain
a soft tube scale. Fuel-to-additive ratio was 2,300:1. Franklin Station. Calgon Velvamag (No. 2 fuel oil containing 8.6
Ib b%0/gal.). One gallon (14.3 Ib/gal) used per 4,000 gal. No. 6 oil). Mddleton No. 3. CH-22 Fuel Oil Additive (additives
not specified for other tests).
_c/ Grain loading in grams per million joules recalculated from values reported in gr/scf assuming a ratio of 9,917 million
joules/MW-Hr (9.40 million Btu/MW-Hr).
d/ Test measurements made during soot blowing.
_e/ Boiler originally designed for coal; retrofitted to fire oil.
fj Control system designed for oil.
_g/ Control system not described.
-------
APPENDIX B
PARTIGULATE EMISSIONS DATA FOR WATERWALL INCINERATORS
97
-------
HARRISBURG MUNICIPAL INCINERATOR
vD
00
Test No.
Refuse
metric tons/hiT"
a/
HHV, joules/kg x 10
% moisturti "ll"
Auxiliary fuel, kg/hr
3
Steam, 10 kg/hr
Fly ash, kg/hr
ESP inlet
ESP outlet
% control efficiency
6a.b/
Residue, metric tons/hiT"*""
c/
Ratio: fly ash/residue""
Excess air, %
o«d/
Flue gas temp., C^
Fly ash f/
kg/metric ton"T
g/106 joules-
c/
II
15.69
58.0
8.70
26.7
37.10
259.91
5.40
97.68
3.29
0.079
153.8
225.0
16.57
1.90
(tests
12
15.69
58.0
8.70
26.7
37.10
339.11
8.16
97.59
3.29
0.103
94.8
225.0
21.61
2.42
made May 1973 X2i/
il
15.88
54.2
9.30
25.1
48.94
172.50
7.30
95.77
4.28
0.040
93.0
207.8
10.86
1.17
Unit No.
i£
15.88
54.2
9.30
25.1
48.94
183.75
10.07
94.52
4.28
0.043
71.5
207.8
11.57
1.24
1
15
14.61
54.2
9.30
25.1
40.87
308.04
5.67
98.16
3.59
0.086
71.5
203.9
21.00
2.26
16.
14.61
54.2
9.30
25.1
40.87
240.63
7.98
96.68
3.59
0.067
73.0
203.9
16.47
1.77
iz
12.97
35.9
15.10
15.30
51.71
265.67
14.11
94.69
3.48
0.076
76.2
227.8
20.48
1.37
18
12.97
35.9
15.10
15.30
51.71
-
3.48
-
-
227.8
-
-------
HARRISBURG MUNICIPAL INCINERATOR
vO
Test No.
Refuse .
metric tons/hr^
HHV, joules/kg x 10
Auxiliary fuel, kg/hr
3
Steam, 10 kg/hr
Fly ash, kg/hr
ESP inlet
ESP outlet
% control efficiency
Residue, metric tons/hr" A
c.1
Ratio: fly ash/residue"
Excess air, %
o d/
Flue gas temp., G~
Fly ash
kg/metric tori"
g/106 joules^'
f/
(tests
made May 1973)li/
Unit No. 2
I
13.88
58.0
8.70
26.7
38.42
203.12
9.12
95.51
2.29
0.089
52.3
221.7
14.63
1.69
2
13.88
58.0
8.70
26.7
38.42
170.28
8.62
94.94
2.29
0.074
131.1
221.7
12.27
1.41
.3
14.06
54.2
9.30
25.1
48.63
200.81
10.84
94.60
3.16
0.064
109.8
209.4
14.28
1.54
4
14.06
54.2
9.30
25.1
48.63
252.88
10.25
95.95
3.16
0.080
83.5
209.4
17.99
1.93
3.
14.70
54.2
9.30
25.1
44.82
195.77
8.66
95.58
4.33
0.045
74.2
225.0
13.28
1.43
j6
14.70
54.2
9.30
25.1
H
44.82
288.40
10.16
96.48
4.33
0.067
74.5
225.0
19.62
2.11
1
11.97
35.9
15.10
15.30
49.62
174.09
10.80
93.80
2.51
0.069
119.5
235.6
14.54
0.95
&
11.97
35.9
15.10
15.30
49.62
185.43
12.70
93.15
2.51
0.074
87.0
235.6
15.49
1.03
-------
CHICAGO NORTHWEST INCINERATOR TESTS MAY 1971—^
Test No. PE-X PE-1 PE-2/PD-2 PE-3/PD-3 PE-4/PD-4
Refuse
metric tons/hr -
% ash -
HHV, 10 joules/kg -
% moisture - - - - ~
Auxiliary fuel, kg/hr -
9
Steam, 10 joules/hr -
Fly ash, kg/hr
ESP inlet - - 92.99 226.80 215.46
ESP outlet - - 10.61 8.21 7.03
% efficiency - - 88.6 96.4 96.7
Residue, metric tons/hr -
Ratio: fly ash/residue -
Excess air, % 136 136 130 78.1 78.1
Flue gas temp., °G 252.2 225.6 179.4 181.1 180.0
Fly ash ^/
kg/metric ton*" -
g/106 joules^ -
-------
27/
STUTTGART (W. GERMANY) WATERWALL INCINERATOR—
Test No.
Refuse
metric tons/hr
% ash
% moisture .
HHV, 106 joules/kg^
Auxiliary fuel, kg/hr (oil)
9
Steam, 10 joules/hr
Fly ash, kg/hr
ESP inlet
ESP outlet
% efficiency
c/
Residue, metric tons/hr^
Ratio: fly ash/residue
Excess air, %
Flue gas temp., C
Fly ash f ,
kg/metric tori"1
1/106 ioulea^7'
Stuttgart
&
24.10
25.9
30.5
7.624
5,484
342.1
532.1
-
-
5.10
0.104
25.5
190.6
22.08
2.90
Unit 28
5.
21.04
28.5
41.0
6.780
2,951
209.1
294.4
-
-
4.98
0.059
40.9
190.6
13.99
2.06
Stuttgart
A
22.33
31.3
38.4
6.836
6,108
331.9
681.3
-
-
5.49
0.124
30.3
183.3
30.51
4.46
Unit 29
1
21.54
30.6
37.6
6.918
2,774
203.3
686.8
•M
-
5.18
0.133
53.0
182.2
31.88
4.61
-------
DUSSELDQRF (W. GERMANY) WATERWALL INCINERATOR AND MUNICH
o
ro
(W. GERMANY) WATERWALL INCINERATOR?!/
Test No.
Refuse
metric tons/hr
% ash
% moisture ,
HHV, 106 joules/kg^
Auxiliary fuel, kg/hr
g
Steam, 10 joules/hr
Fly ash, kg/hr
ESP inlet
ESP outlet
% efficiency
Residue, metric tons/hiT"
Ratio: fly ash/residue
Excess air, %
Flue gas temp. , C
Fly ash ~,
kg/metric ton""
a/106 ioule&£'
Dusseldorf
Jk
10.52
33.7
32.4
7.227
0.0
42.7
499.0
-
M
2.76
0.181
112.0
235.0
47.43
6.56
Munich North I
6
26.10
30.0
44.4
6.413
0.0
97.1
1,613.9
-
M
8.67
0.186
131.3
157.2
61.84
9.64
Munich North II
A
45.50
36.8
28.0
7.501
0.0
246.4
648.2
-
-
13.61
0.048
44.5
164.4
14.25
1.90
-------
a/ Based on daily average (composite) measurements.
b/ Average values for low, medium, and high Btu runs.
c/ Excluding metals in residue*
jd/ Measured at ESP inlet.
ej Recalculated from lower heating value of refuse fuel.
f/ Based on refuse portion of fuel.
103
-------
APPENDIX G
ELECTROSTATIC PRECIPITATOR PERFORMANCE MODEL
104
-------
Concurrent firing of oil and MSW will cause departures from ESP operating
parameters which apply to firing oil alone. The ESP model is used to cal-
culate the magnitude of resulting changes in performance.
As discussed previously, the most significant departures anticipated when
refuse is fired concurrently with oil are in flue gas volume, moisture and
composition, particulate density, size distribution, resistivity, fusion
temperature, and dust loading. The calculation of resulting ESP perfor-
mance, as described below, is based on the simplifying assumptions that:
1. The electric field is unaffected by the changes in particulate properties,
It is well known that the introduction of a significant number of fine dust
particles into an electrostatic precipitator significantly influences the
voltage-current characteristics of the interelectrode space. Qualitatively,
the effect is seen by a decrease current for a given voltage compared to
a dust-free situation.^2.' If the dust loading increases as expected for
combined oil-MSW firing, this approximation will yield a higher limit value
for the corona current, and therefore an upper limit value for collection
efficiency.
2. Charging time is negligible.
This is the assumption (implicitly made) when the saturation charge is used
to describe the instantaneous charge on each particle. Order-of-magnitude
calculations indicate that a particle residence time of less than 1 sec is
required to achieve 90% of saturation for a 0.18 p-m particle. For larger
particles, the required residence time to achieve charge saturation will
decrease.£l/ Therefore, this approximation will yield an upper limit to the
calculated collection efficiency.
3. Precipitator performance is not current-limited.
It is felt that this will be a reasonable assumption for gas volume and
dust loadings expected.
4. The Deutsch-Anderson equation is applicable in integral form for a
discrete particle size range.
This is the usual approximation which is made when this equation is used
to describe the average performance of an electrostatic precipitator in
design applications.
In addition to the approximation described, several nonideal effects which
are known to exist in full-scale electrostatic precipitators are not dealt
with explicitly in the present model.
105
-------
The factors of major importance are:
1. Gas velocity distribution,
2. Gas sneakage, and
3. Rapping reentrainment.
These departures from ideality will reduce the collection efficiency that
may be achieved for a precipitator operating with a given specific collecting
area.67/
In the application of the model to actual conditions for combined firing,
the particle size distribution, resistivity, dielectric constant, flue gas
volume, precipitator design specifications, and at least one set of perfor-
mance data must be known or estimated as summarized in Table C-l« The test
data need not correspond to conditions for combined firing. A computerized
model, recently developed by Southern Research Institute under EPA sponsor-
ship, was used as the basis of the analytical model.xZ/ Electric field cal-
culations are omitted, and a system-dependent parameter is calculated in-
stead from known performance data. This is a one-parameter "fit" in which
effective migration velocities are determined for each particle size range
and particle resistivity under a fixed set of design or test conditions.
This step was done using trial-and-error, or iterative procedures. A block
diagram illustrating the computational procedure is shown in Figure G-l.
The computation procedure is divided into two parts. Initially, precipi-
tator design or sizing data and at least one set of performance data are
used to determine the distribution of migration velocities for the known
performance data over the range of particle sizes. In the second calcu-
lation stage, migration velocities are a'djusted for changes in flue gas
temperature and viscosity, particulate resistivity and relative dielectric
constant, operating current and voltage. ESP performance is then deter-
mined for anticipated combined-firing test conditions. The computation
procedure is described in detail in the remainder of this section.
1. Determine migration velocities for each particle size based on known
performance data.
Step 1 - Calculate the saturation charge on the median diameter particle
within the jtn discrete particle size range, based on performance
data. A "modified" saturation charge expression, developed by
Southern Research Institute^.' is used:
(R(j) + Xm)2(1.2E)l + 2 -=- - (C-l)
K
106
-------
Table C-l. DATA INVENTORY FOR ELECTROSTATIC
PRECIPITATOR PERFORMANCE MODEL
Precipitator design data
Plate area
Plate-to-plate spacing, inlet
Plate-to-plate spacing, outlet
Length of electrical sections
Corona wire diameter
Number of Series electrical sections
Number of Parallel electrical sections
Performance data
Average efficiency
Flue gas volume
Flue gas temperature and composition
Precipitation rate parameter
Operating voltage
Current versus resistivity curve (if known)—'
Voltage versus current curve (if known)—'
Particle size distribution (if known)
Particle resistivity
• Test Conditions
Percent MSW on a Btu basis
MSW composition, ash, and moisture
MSW heating value
Oil composition, ash, and moisture
Oil heating value
Percent excess air
Boiler efficiency curve (if known)
a/ A conservative estimate of the allowable current density as a function
of resistivity is given in Figure C-2.
b/ A reasonable approximation of the average electrical conditions in the
precipitator is given in Figure C-3 by the curves labeled "typical."
107
-------
New Trial
Value of K
No
Data Inputs
(ESP Design Data
& Performance Test Data)
i
Calculated
Saturation Charges
1
Calculate
Cunningham Factors
I
Calculate
Effective Migration
Velocity
I
Calculate Migration
Velocity Distribution
I
Calculate Fractional
Efficiencies
I
Calculate
Average
Efficiency (fj)
i
Calculated
Agrees with
Measured
Yes
Data I nputs
(Combined Firing)
I
Estimate
ESP Current,
Voltage
Calculate
Flue Gas Volume
I
Calculate Adjusted
Migration Velocity
Distribution
Estimate Particle
Size Distribution
Calculate Fractional
Efficiencies
Calculate
Average
Efficience (17)
Figure C-l. Block diagram of ESP performance model.
108
-------
where £Q = permittivity of free space
= 8.85 x 10~l-2 cou!2/newton-m2
R(j) = mass median particle radius, corresponding to the jth
discrete particle size range, m
Xm = an adjustable parameter = mX (G_2)
where X = ion mean free path, m
m = number of mean free paths
EQ = average charging field, volts/m
X = relative dielectric constant of the particle
The mean free path X is calculated using the following
expression which is valid for the range of temperatures
of interest for fly ash precipitators and at one atmos-
phere:
X = 1.9176 x 10"10 (T°K)
Step 2 - Calculate the Cunningham correction factor, or slip correction
factor for each discrete particle size range, j , using performance
data:
C(j) = 1 + AX/R(j) (C-3)
where A = 1.257 + 0.400 exp (-1.10 R(j)A)
Step 3 - Calculate the effective or length averaged migration velocity for
the different particle size ranges from the Deutsch equation using
performance data.
we =2- ln/-J^-\
-------
Step 4 - From the test or design data used for Steps 1 to 3, estimate the
effective migration velocity, w (j), for each discrete particle
size range in the distribution. The distribution of migration
velocities for each discrete size range we(j) can be expressed
in terms of the overall effective (length averaged) migration ve-
locity w as follows:
° * q; C/K
where K = complex function of particle size distribution, precipi-
tator geometry, and operating condition, having dimension
of length (m)
q' = q'OO is the value of modified saturation charge corre-
S 3
spending to a radius of size "K"
C = C(K) is the value of the Cunningham correction factor cor-
responding to a radius of size "K"
The system-dependent parameter K is determined using an
iterative procedure described in the next two computational
steps. A convenient starting trial value is the mass median
radius of the particle size distribution.
From classical theory, the migration velocity is given by:
w (j) . q(1) EP C(1) (c.6)
eU' 6TrR(j) u
where Ep = electric field near the collection electrode, volts/m
u = gas viscosity, kg/m-sec
C(j), R(j) = as previously defined
The direct solution ofEq. (C-6) would require the solution of two simulta-
neous second-order partial differential equations in order to calculate the
field adjacent to the collection electrode Ep. The alternative approach
used here is to assume that the ratio of instantaneous charges on particles
in the jtn size range is approximately the same as the saturation charge
ratio. Use of the modified saturation charge to describe the instantaneous
charge will yield an upper limit value of the migration velocity for each
particle size range, which in turn will yield an upper limit value for the
collection efficiency.
110
-------
Step 5 - Calculate the fractional efficiency for each discrete particle
size range and each stage or series electrical section. The
Deutsch equation is used in the following form:
= 1 - exp(-we(j)A/Q) (G_7)
where T](j) = the fractional collection efficiency
we(j), A, Q = as previously defined
For multiple stage precipitators, the overall fractional
efficiency corresponding to a given size range j is ob-
tained as follows:
Tl(j) = E TKi,j) (C-8)
i
Step 6 - Calculate the overall fractional efficiency.
Tl = 1 - ZXj exp(-wp(j)A/Q) (C-9)
where Xj = the mass fraction of the jth discrete particle size
range
If the fractional efficiency calculated in Eq. (C-9)differs significantly
from the test efficiency, a new trial value of K is used, and Steps 4
through 6 repeated until convergence is obtained.
2. Determine fractional efficiency for combined firing test conditions.
The major inputs from Part 1, computation Steps 1 through 6, are migration
velocities w (j) for each discrete particle size range j based on known
performance data. These data, which implicitly contain system dependencies,
and test data summarized in Table C-l are used to calculate the electro-
static precipitator performance for combined firing according to the fol-
lowing procedure.
Step 7 - Calculate adjusted value of average corona current density.
Current and voltage relationships in a precipitator are governed by elec-
trode geometry and by the mobility of the charge carriers. Electrical con-
ditions are limited by either breakdown of the gas in the interelectrode
space or by breakdown in the collected dust layer.
Ill
-------
At present there is no theoretical basis for predicting either the cur«.ent-
resistivity behavior or the maximum allowable corona current. Ideally,
therefore, an experimentally determined current versus resistivity curve
should be used.
If current-resistivity data are not available the curve in Figure C-2 may
be used to estimate the current density for a given value of dust resistiv-
ity. Figure C-2 was obtained from the literature, and is based on the ob-
servation that critical current densities in full-scale precipitators can
be reduced from the theoretical dust breakdown values by a factor of about
10. The use of this curve should give a conservative estimate of the allow-
able current density as a function of resistivity.ilZ/
Field experience has shown that current density for cold-side precipitators
is limited to around 50 to 70 nA/cm (1 x 10"' A = 1 nA) due to electrical
breakdown of the gases in the interelectrode region.
Step 8 - Calculate adjusted value of voltage corresponding to the maximum
allowable corona current.
As in Step 7, the voltage-current relationships for an electrostatic pre-
cipitator are governed by the mechanical design of the collector system,
the size and concentration of dust particles in the gas stream, the pres-
ence of a dust layer on the collection electrode, and the temperature and
composition of the gas stream. Therefore, it is preferable to determine
the operating voltage from an experimentally determined current-voltage
curve (see Table C-l).
Lacking this experimental data, a reasonable approximation to the average
electrical conditions in the precipitator is given in Figure C-3 by the
curves labeled "typical."£Z/
Step 9 - Calculate adjusted flue gas volume. The flue gas volume for a
given combined firing application is calculated by standard pro-
cedures based on data inputs summarized in Table C-l "test condi-
tions." In general, flue gas volume will increase significantly
.with increasing MSW fraction for combined firing of oil and MSW:
(a) at 10% MSW on a Btu basis, assuming 25% excess air and 5,650
Btu/lb (HHV) for the MSW portion, the theoretical gas volume will
increase by 5.770; and (b) on the same basis, the theoretical in-
crease in flue gas volume at 20% MSW is 18.0%.
Step 10 - Estimate the particle size distribution for combined firing.
112
-------
CN
U
Q.
O
O
O
c
z
LU
Q
I—
z
LU
10
Source: Ref. 67
10" 1012
RESISTIVITY, ohm/cm
1013
Figure G-2. Current density as a function of resistivity.
113
-------
70
60
50
cs
4
Q.
40
CO
z
30
(J
•370°C
150°C
Outlet
Typical
(150°C)
Inlet
Outlet
(Typical
(370°C)
i
0
Source: Ref. 67
30 40
APPLIED VOLTAGE, kilovolts
50
60
Figure C-3. Comparison between the voltage versus
current characteristics for cold-side and hot-side
precipitators. Corona wire radius = 0.277 cm
(0.109 in.), plate spacing = 22.86 cm (9 in.).
114
-------
The particle size distribution for combined firing is presently based on a
linear combination of Weibull (or Rosin-Ramler) parameters. !!/ The particle
size distribution for firing refuse alone must be estimated. Weibull param-
eters are calculated separately for submicron particulate and particulate
larger than 1 urn. The two-parameter Weibull distribution function is fit
by a least squares technique in the following form:
F(R(J)) = 1 - e- (c-10)
\ 9 '
where F(R(j)) = the weight fraction of particulate having diameters less
than R(j)
8 , b = independent parameters
Weibull parameters for the refuse portion of the fuel are as follows:
9 (um) b
(< 1 um) 1.48 2.91
(> 1 urn) 92.1 0.23
The particle size distribution and ash content of the fossil-fuel portion
must be known independently to determine the particle size distribution for
the composite fly ash.
Step 11 - Calculate fractional efficiencies. When ESP current and voltage,
flue gas volume, migration velocities, and particle size distri-
bution have been adjusted for combined firing conditions, frac-
tional efficiencies for each discrete particle size range are
calculated using Eq. (C-7).
Step 12 - Calculate effective, length averaged (total) efficiency. Total
efficiency T| for the ESP under combined- firing conditions is
calculated using Eq. (C-9).
The ESP performance calculation, as described in Appendix C, can be per-
formed using a moderately sophisticated calculator, such as Hewlett-Packard
9810-A, or equivalent.
115
-------
1 TECHNICAL REPORT DATA "
1 (Please read Instructions on the reverse before completing)
J1. REPORT NO. 2.
EPA-600/2-76-209
J4. TITLE AND SUBTITLE
Performance of Emission Control Devices on Boilers
Firing Municipal Solid Waste and Oil
7. AUTHOR(S)
u . B. Galeski and M. P. Schrag
9. PERFORMING OR3ANIZATION NAME AND ADDRESS
Midwest Research Institute
K25 Volker Boulevard
Kansas City, Missouri 64110
112. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
July 1976
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
EHB533
11. CONTRACT/GRANT NO.
68-02-1324, Task 40
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 7/75-6/76
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES Task officer for this report is J.D. Kilgroe , Mail Drop 61,
Ext 2851.
|16. ABSTRACT
The report gives results of estimating particulate flue gas loadings for
combined firing of shredded municipal waste (MSW) and oil, using existing data on
particulate emissions from oil-fired electric utility boilers and from waterwall
(steam generating) incinerators firing either waste or waste-plus-coal/oil auxiliary
fuel. Control device performance was estimated for several planned oil/MSW
resource recovery systems. On the basis of these estimates, installed particulate
emission controls, designed for coal, are predicted to be significantly less efficient
for control of particulate emissions from combined firing of oil/MSW. Anticipated
control difficulties result mostly from relatively high particulate loadings, high flue
gas volumes, fine particulates, relatively low particle density, and relatively high
fractions of carbonaceous low-resistivity particulate.
J17. KEY WORDS AND DOCUMENT ANALYSIS
la. DESCRIPTORS
Air Pollution
Boilers
Fuels
I Wastes
Fuel Oil
Dust
18. DISTRIBUTION STATEMENT
Unlimited
b.lDENTIFIERS/OPEN ENDED TERMS
Air Pollution Control
Stationary Sources
Combined Fuel
Municipal Solid Waste
Oil/Municipal Solid Wash
Particulate
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
c. COSATI Field/Group
13B
13A
21D
i
11G
21. NO. OF PAGES
128
22. PRICE
EPA Farm 2220-1 (9-73)
116
------- |