Environmental Protection Technology Series
S02 REDUCTION IN
NON-UTILITY COMBUSTION SOURCES -
TECHNICAL AND ECONOMIC
COMPARISON OF ALTERNATIVES
IndistrlaS Envirannental 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 policies of the Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
-------
SO2 REDUCTION
IN NON-UTILITY COMBUSTION SOURCES
TECHNICAL AND ECONOMIC COMPARISON OF ALTERNATIVES
by
P.S.K. Choi, E.L. Kropp, W.E. Ballantyne,
M.Y. Anastas, A.A. Putnam, D.W. Hissong, andT'.J. Thomas
Battelle-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
Contract No. 68-02-1323, Task 13
ROAPNo. 21ACX-083
Program Element No. 1AB013
EPA Task Officer: C. J. Chatlynne
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
October 1975
-------
ABSTRACT
An analysis of non-utility combustion (NUC) sources was conducted
for various size classes and fuel types with respect to the significance of
sulfur dioxide emissions. Technical and economic comparison of various
sulfur dioxide control alternatives was made for the important size classes
and fuel types. Categories of alternatives included in the study are:
physical cleaning of coal, coal gasification, coal liquefaction, fluidized-
bed combustion of coal, and flue gas desulfurization. For small size
classes of NUC sources, applicabilities of package sorption system were
reviewed.
iii
-------
ACKNOWLEDGMENTS
Many individuals contributed their advice and assistance to this
study. In particular, the present Project Officer, C0 J. Chatlynne, and
Ro Do Stern of the Environmental Protection Agency and G. S0 Haselberger,
now with the Federal Energy Administration, deserve mention.
The contributions of Mr. Paul Spaite, consultant to Battelle-
Columbus, to the overall study and in review of the drafts is gratefully
acknowledged.
Several staff members at Battelle-Columbus also contributed to this
study, including: G. R. Smithson, Jr0, R. B. Engdahl, B. C. Kim, H. S.
Rosenberg, J. B. Brown, Jr., F. A» Creswick, J. E. Flinn, and J. M. Allen.
iv
-------
TABLE OF CONTENTS
Page
INTRODUCTION 1
MANAGEMENT SUMMARY 2
CONCLUSIONS 5
RECOMMENDATIONS 9
PART I. INDUSTRIAL AND COMMERCIAL BOILER DATA BASE
DESCRIPTION OF DATA SOURCES . 12
DESCRIPTION OF PARAMETERS NECESSARY TO CHARACTERIZE BOILER
POPULATION I6
Use Category 16
Size 17
Fuel Type 19
Annual Load Factor 19
Stack Temperature 20
Fuel Sulfur Content 24
Flue Gas Flow Rate 26
BOILER POPULATION CHARACTERIZATION 32
PART II. NON-UTILITY COMBUSTION SOURCE CONTROL ALTERNATIVES
CONTROL ALTERNATIVES 60
CLEAN FUELS 61
Supply Projections 61
Utilization and Applicability 63
Costs 65
PHYSICAL CLEANING OF COAL . 70
Sulfur in Coal 70
v
-------
TABLE OF CONTENTS
(Continued)
Page
Cleanability of U.S. Coals 70
Technology of Coal Cleaning 71
Environmental Impacts and Control 72
Applicability to NUC Sources 73
Capital and Annualized Costs 74
COAL GASIFICATION 78
Gasification Processes 78
Fuel Gas Desulfurization and Sulfur Recovery 80
Applicability to NUC Sources 83
Model Plant Calculation 86
COAL LIQUEFACTION 91
Process Description 91
Environmental Problems 93
Applicability to NUC Sources 93
Model Plant Calculations 94
FLUIDIZED-BED COMBUSTION 101
FBC Technology and Environmental Emissions .... 101
Applicability to NUC Sources 103
Model Plant Calculation 105
FLUE GAS DESULFURIZATION (FGD) PROCESSES 109
Process Descriptions 109
Applicability to NUC Sources Ill
Model Plant Calculation 112
VI
-------
TABLE OF CONTENTS
(Continued)
Page
EVALUATION OF ALTERNATIVES ...................... H9
Approach ............................ 119
Evaluation Criteria. ... ..... ...... ........ 119
Alternative Evaluation ..................... 120
Evaluation Result ........................ 123
COST OF ALTERNATIVES ......; ..... . ............ 129
PART III. PACKAGEABILITY OF SORPTION PROCESSES
SURVEY OF EXISTING PACKAGE SORPTION SYSTEMS ........ ......
SURVEY OF SORBENT MATERIALS ....... ............. ... 149
DESCRIPTION OF SORPTION PROCESSES. .......... ........ 162
SORPTION PROCESS EVALUATION ...................... 169
Approach ............................ 169
Evaluation Criteria ....... ................ 169
Process Evaluation ................ . ...... 171
COST OF SORPTION PROCESSES ...................... 177
REFERENCES .............................. 181
APPENDIX A
ACCOUNTING METHOD ........................... A-l
APPENDIX B
DESCRIPTION OF FLUE GAS DESULFURIZATION PROCESSES ........... B-l
APPENDIX C
ESTIMATED COSTS OF CENTRAL REGENERATION AND ACID PRODUCTION PLANT. . . C-l
vii
-------
CONVERTING UNITS OF MEASURE
EPA policy is to express all measurements in metric units. When
implementing this practice will result in undue cost or lack of clarity,
conversion factors are provided for the nonmetric units used in the report.
Generally, this report used British unit of measure. For conversion to the
metric system, use the following conversion factors.
TABLE OF CONVERSION FACTORS
Multiply
English Unit
by
Conversion
To Obtain
Metric Unit
acres
acre-feet
barrel, oil
British Thermal Unit
British Thermal Unit/pound
cubic feet/minute
cubic feet/second
cubic feet
cubic feet
cubic inches
degree Fahrenheit
feet
gallon
gallon/minute
horsepower
inches
inches of mercury
pounds
million gallons/day
mile
pound/square inch (gauge)
square feet
square inches
tons (short)
yard
0.405
1233.5
158.97
0.252
0.555
0.028
1.7
0.028
28.32
16.39
0.555(°F-32)
0.3048
3.785
0.0631
0.7457
2.54
0.03342
0.454
3785
1.609
(0.06805 psig+l)00
0.0929
6.452
0.907
0.9144
hectares
cubic meters
liters
kilogram-calories
kilogram calories/kilogram
cubic meters/minute
cubic meters/minute
cubic meters
liters
cubic centimeters
degree Cencigrade
meters
liters
liters/second
kilowatts
centimeters
atmospheres
kilograms
cubic meters/day
kilometer
atmospheres (absolute)
square meters
square centimeters
metric tons (1000 kilograms)
meters
(a) Actual conversion, not a multiplier.
viii
-------
FINAL REPORT
on
S02 REDUCTION IN NON-UTILITY COMBUSTION
SOURCES--TECHNICAL AND ECONOMIC
COMPARISON OF ALTERNATIVES
to
ENVIRONMENTAL PROTECTION AGENCY
by
P.S.K. Choi, E. L. Kropp, W. E. Ballantyne,
M. Y. Anastas, A. A. Putnam, D. W. Hissong,
and T. J. Thomas
from
BATTELLE
Columbus Laboratories
October 1, 1975
INTRODUCTION
The objective of the tasks (EPA Contract No. 68-02-1323, Tasks 13
and 19) was to analyze available small industrial and commercial boiler data
and to evaluate various alternatives for the reduction of SO. from the non-
utility combustion (NUC) sources. This study covered the review of the
existing boiler data obtained from various sources such as the National
Emissions Data System (NEDS), the Walden Survey, current EPA-related programs,
and American Boiler Manufacturers Association data. A methodology was
developed for estimation of missing data and various boiler subgroups were
evaluated with respect to the significance of SO- emissions. The control
alternatives under consideration included clean fuels, processed fuels,
combustion modification, and flue gas desulfurization (FGD). Additional
emphasis was placed on a technologically promising segment of flue gas
desulfurization; the segment being package sorption.
The report is thus broken into three parts: Part I describes
the acquisition and analysis of boiler data, Part II discusses in general
the four control alternatives studied, and Part III provides in-depth
background information to assist EPA in preparing development/demonstration
studies on package sorption systems for the abatement of SO emissions from
X
small industrial and commercial boilers.
-------
MANAGEMENT SUMMARY
This study was performed for the Environmental Protection Agency
(EPA) under Contract No. 68-02-1323, Tasks 13 and 19. The objective of
Task 13 was to analyze available small industrial and commercial boiler
data and to evaluate various alternatives for the reduction of sulfur dioxide
emissions from the non-utility combustion (NUC) sources. The objective of
Task 19 was to develop background information which could assist EPA in
preparing development/demonstration studies on package sorption systems for
the abatement of sulfur dioxide from the NUC sources under consideration.
This report consists of three parts,, The acquisition and analysis
of the boiler data are described in Part !„ Data on existing industrial and
commercial boilers were obtained, for the most part, from the National
Emissions Data System (NEDS)', the Walden survey, current EPA-related programs,
and American Boiler Manufacturers Association (ABMA) compilations. Where gaps
in the data base existed, methodology was developed for estimation of key data.
Various boiler classes were also evaluated with respect to the significance of
sulfur dioxide emissions.
Various control alternatives were reviewed, analyzed, and evaluated.
and the results are included in Part II. The alternatives under consideration
included clean fuels, processed fuels, combustion modification, and flue gas
desulfurization (FGD). A set of evaluation criteria was established and
employed to examine the technical feasibility of each alternative. The control
costs were estimated for the boiler classes of environmental concern in order
to examine the economic feasibility of the alternatives. In the evaluation
study, attempts were made to provide the general technical and economic infor-
mation for the selected boiler classes using the average parameter values of
the acquired boiler data. Specific case studies were not made.
Part III of this report covers the survey and evaluation of existing
and potential package sorption devices and technologies applicable to the NUC
sources identified in Part I. The sorption processes considered in this study
included both liquid and solid phase sorption processes which were classified
into two categories--throwaway and regenerable processes. . ,
-------
The results of the boiler data analysis indicated that the para-
meters that fix the annual load factor are the use category and type of
fuel fired. The annual load factor is higher for coal and industrial use
than for oil and commercial use. The majority of the flue gas temperatures
are between 400°and 450° F, There is a tendency for the temperature to
be higher for high-sulfur oil-firing units and for coal-firing units of
smaller size. The important boiler size classes to consider for reduction
of area source emissions are middle to large size coal units (both
stoker and pulverized) and almost all classes burning high-sulfur oil
(including sizes as small as 2 x 10° Btu/hr).
It is expected that the NUC sources under consideration will
be forced to use dirty fuels (i.e., high-sulfur fuels) due to the insuffi-
cient supply of clean fuels. Processed fuels such as high-Btu synthetic
natural gas (SNG) and coal liquefaction products produced on a large scale
are economically favorable over application of FGD processes for the small
size classes of the NUC sources. The primary concern for the processed
fuels, however, is the availability of the processes in the near future.
Technically, all FGD processes are feasible for application to
the NUC sources under consideration, although the high oxygen concentration
in the flue gas may possibly cause difficulties in processes where oxida-
tion is undesirable. The FGD processes are economically favorable over
other alternatives for the NUC sources in large size classes. Most
important of all, many FGD processes are commercially available although
the application to the NUC sources remains to be demonstrated.
Among various FGD processes, regenerable processes with regenera-
tion performed at a central facility,such as the MgO process with central
regeneration, should be the most attractive choice where there is no
existing capability for regeneration within the NUC source physical plant.
Such a system is relatively simple in operation, small in size, and low in
capital and annualized costs. The sorption capacity of the sorbent,
however, should be high so that the size of the sorption unit and the
quantity of sorbent to be used may be minimal.
-------
Throwaway processes should be more favored than integrated regen-
erable processes in controlling NUC sources without an in-house capability
for regeneration if there is land available for the sludge disposal. This
is mainly because the former generally is simpler in operation, smaller in
size, and lower in annualized cost than the latter.
Currently no package sorption system is available for the control
of SO-, The package unit concept for various FGD processes is not deemed
feasible for control of small and large sizes of NUC sources. This is
mainly because for small size NUC sources, the economics are very favorable
toward processed fuels, and, thus, the FGD processes are not economically
feasible as compared with processed fuels. For large size NUC sources, the
FGD processes should become economically favored over other alternatives;
however, the size of the system would become too big to be handled as a
package unit. The concept of a packaged unit may be feasible for control of
medium size NUC sources, i.e., boiler sizes up to 30,000 Ib/hr, if the
packaged unit can be manufactured for a low cost and installed relatively
easily, and the fixed capital charge portion of the annualized cost can be
kept to a small fraction of the capital cost.
-------
CONCLUSIONS
(1) The annual load factor was computed for each boiler for
which data were available by dividing the total fuel heat input rate per
year by the design firing rate and the number of.hours per year. A con-
sideration of the load factors led to the conclusion that, for the range
f 8
of design capacities from 10 Btu/hr through 5 x 10 Btu/hr, the parameters
that fix the annual load factor are the use category and type of fuel
fired (with sulfur content not important).
(2) NEDS data on flue gas flow rate for individual boilers
can be justified; however, it is not possible to make a definitive state-
ment about any average flue gas flow rates due to the many different types
of combustion equipment using different amounts of excess air.
(3) An assessment of reported stack exit temperatures leads .
to the conclusion that a majority of temperatures are between 400°F and
450°F. There is a tendency for the temperature to be higher for high-
sulfur residual-oil-firing units and for coal-fired units of smaller size.
(4) Typical sulfur contents in fuel ranged from nil in natural
gas to a high of almost 3.5 percent in stoker-fired coal units. Coal-
fired units showed a slight increase in fuel sulfur content as boiler size
increased, but all other fuels fired tended to remain at a constant
sulfur level.
(5) Sulfur oxide emissions were calculated using NEDS analysis
and EPA emission factors. Both potential and actual (adjusted by annual
load factor) emissions were calculated. Conclusions reached are that the
important boiler size classes to consider for reduction of non-utility
combustion source emissions are middle to large size (10 x 10^ to 500 x 10^
Btu/hr) coal units (both stoker and pulverized) and almost all classes
burning high-sulfur residual oil (including sizes as small as 2 x 10^
Btu/hr).
(6) The ABMA data file can be used successfully to supplement
the NEDS file in the analysis of boiler data.
(7) The use of clean fuels, such as natural gas, distillate fuel
oil, low-sulfur residual oil, and low-sulfur coal are the best control
-------
alternatives for the non-utility combustion sources under consideration if
the clean fuels are available and can be obtained for a reasonable price.
The present uncertainties in the energy area, however, make projections of
fuel supply and cost very difficult and subject to considerable speculation.
Particularly important are the questions of whether the price of natural
gas will be deregulated and to what extent the U. S. will continue to import
foreign oils.
(8) Processed fuels produced on a large scale are economically
favored over application of FGD processes to boilers in the small size
ranges of the non-utility combustion sources. The capital cost requirement
to the boiler system is low for the processed fuels. The annualized cost
is also low, mainly due to the small fixed capital charges. The primary
concern associated with processed fuels, however, is the availability of the
processes in the near future.
(9) Technically, all FGD processes are feasible for application
to commercial and industrial boilers, although the high oxygen concentra-
tion in the flue gas may possibly cause difficulties in processes where
oxidation is undesirable such as the double alkali and Wellman-Lord pro-
cesses. The FGD processes are economically favorable over other alternatives
for the non-utility combustion sources in large size classes. Most important
here, many FGD processes are commercially available, although the application
to the non-utility combustion sources remains to be demonstrated.
(10) Physical cleaning of coal is very attractive in its
economics; however, its application is limited to certain types of coal.
(11) A coal gasification process for low-Btu gas generation should
be a retrofit system to a group of non-utility combustion sources. The
application in general is limited to coal-fired boilers due to non-compatible
combustion chamber configurations of gas- and oil-fired boilers.
(12) A coal gasification process for high-Btu synthetic natural
gas manufacture will be a viable alternative for the non-utility combustion
sources except for the large size boilers. The process will not be avail-
able until 1980 to 1983.
(13) The solvent refined coal process is a viable alternative
for control of non-utility combustion sources. The solid product will be
-------
used for coal-fired boilers and the liquid (-product for oil-fired boilers.
For oil-fired boilers, however, the H-coal process will be.more favorable
economically since heating of the product from the H-coal process in its
utilization is required to a lesser extent. The liquefaction process,
however, will not be available until 1981 to 1983.. •
(14) Among various FGD processes, regenerable processes with
regeneration performed at a central facility, such as the MgO process with
central regeneration, should be the most attractive choice where there
is no existing capability for regeneration within the non-utility combustion
source physical plant. Such a system is relatively-simple in operation,
small in size, and low in capital and annualized costs. The sorption capacity
of the sorbent, however, should be high, so that the size of the sorption unit
and the quantity of sorbent to be used may be minimal.
(15) Throwaway processes could be more favorable than integrated
regenerable processes in controlling non-utility combustion sources where
there is no in-house capability for regeneration and no close central
regeneration capability. This is mainly because the former generally is
simpler in operation, smaller in size, and lower in annualized cost than
the latter. In addition, the operation of a throwaway process becomes
more reliable as more information is developed for the underlying chemistry
of the process. Although land is required for sludge disposal, the solid
waste generated by the processes is small in quantity because of the small
system size.
(16) Integrated regenerable processes may be more favorable in
cases of industrial boilers where there are captive uses for sulfur com-
pounds for in-house capability for regeneration. For example, the pulp
and paper industry can regenerate spent sodium sulfite or ammonium scrubbing
solution in its manufacturing operation. The chemical industry frequently
has captive uses for sulfuric acid or sulfate salts. The petroleum indus-
try has sources of hydrogen sulfide that could be used to produce sulfur
from SC>2 emissions. An anlysis would be needed for the specific individual
case to determine the technical and economical feasibilities of the various
FGD processes.
-------
(17) Currently no package sorption system is available for
the control of S02. The package unit concept for various FGD processes
is not deemed feasible for control of small and large sizes of non-utility
combustion sources. This is mainly because for small size non-utility com-
bustion sources, the economics are very favorable towards processed fuels,
and, thus, the FGD processes are not economically feasible as compared
with processed fuels. For large size non-utility combustion sources, the
FGD processes become economically favorable over other alternatives;
however, the size of the system would become too big to be handled as a
package unit. The concept of a packaged unit may be feasible for control
of medium size non-utility combustion sources, i.e., boiler sizes up to
30,000 Ib/hr, if the packaged unit can be manufactured for a low cost
and installed relatively easily, and if the fixed-capital portion of the
annualized cost can be kept to be a small fraction of the capital cost.
-------
RECOMMENDATIONS
(1) The existing NEDS-ABMA data files should be kept current
to maintain trend analysis capabilities.
(2) Non-utility combustion source SOX emissions appear to be
1 7 ' 8
concentrated in coal-fired units of between 2 x 10 and 5 x 10 Btu/hr size
class and in high sulfur residual oil-fired units of sizes between 2 x 10
and 5 x 10 Btu/hr; thus, SOX control technique analyses should be performed
for these size range units.
(3) Clean fuel development activities should be accelerated
to promote control of small non-utility combustion sources more economically
in the near future. Priorities should be given to high-Btu coal gasifica-
tion, solvent refined coal, and H-coal processes. Development of physical
and chemical coal cleaning processes should be continued so that the sulfur
removal efficiency may be improved to a high degree (i.e., greater than 70
percent of total sulfur) and the application of these processes can be
broadened to many types of high-sulfur coal.
(4) Conceptual and demonstration studies of FGD processes for
application to the non-utility combustion sources are recommended under
the following categories:
(a) Regenerable processes with central regeneration
facility
(b) Throwaway processes.
The selection of the regenerable process will be based on the magnitude
of sorbent capacity, ease of transportation of spent and regenerated
sorbent material, and simplicity of sorption operation. Magnesium oxide
and sodium sulfite-based regenerable processes will be appropriate due to
the high sorption capacity. The selection of the throwaway process will
be based on availability of raw material, reliability of operation, and
simplicity of the system. Limestone based simple wet scrubbing and double
alkali processes are recommended.
(5) A study is recommended to identify sorbents which have a
high sorption capacity for SO-. A literature survey will be necessary to
-------
10
review various sorbent materials. The result of this study will be
useful to select sorbents for regenerable processes with central regenera-
tion facilities.
(6) No study is recommended in the immediate future for the
concept of packaged sorption units for non-utility combustion sources.
The study should be carried out when the results of the conceptual and
demonstration studies of the selected FGD processes mentioned above are
available.
-------
PART I
INDUSTRIAL AND COMMERCIAL BOILER DATA BASE
-------
12
DESCRIPTION OF DATA SOURCES
The primary sources utilized in acquiring industrial and commercial
boiler data were the National Emission Data System (NEDS) and American
Boiler Manufacturers Association (ABMA) data. Data were accumulated on all
sizes of commercial, industrial, and utility boilers; utilities were included
since utility boiler information was contained in the data files analyzed.
The NEDS file contains pollutant source information gathered
primarily in 1972 and 1973. The information on file includes the following
parameters:
Control Equipment
Estimated Control Efficiency, percent
Percent Annual Throughput
Normal Operating Time, Hours
Emission Estimate, tons/yr
Percent Space Heat
Allowable Emissions, tons/yr
Compliance Status
Control Regulations
Source Classification Code (SCC)
Fuel, Process, Solid Waste
Operating Rate
Maximum Design Rate
Sulfur Content, percent
Ash Content, percent
Heat Content, 10 Btu/scc, i.e.,
106 Btu/106 scf for gas, 106 Btu/
103 gal for oil, and 10° Btu/ton
for coal
Boiler Design Capacity
These data are on file for all sources inventoried from 1972 to mid-1974.
The year for which the data apply is recorded and varies from 1968 through
1974.
State
County
ACQR
Plant ID Number
City
UTM Zone
Year of Record
Establishment Name and
Address
Person to Contact
Owner
Point ID
Standard Industrial
Classification (SIC)
IPP Process
UTM Coordinates
Stack Data
-------
13
The American Boiler Manufacturers Association (ABMA) records data
on all boilers sold by its members. The data on watertube boilers are kept
in computer-card form, one card prepared for each boiler sold. Records
are also kept for firetube (Scotch) boilers but the records are not nearly
as detailed as are the watertube records.
The watertube computer file was initiated in 1965 and is currently
updated each month. Information contained in the file is as follows:
Capacity per Unit:
Boiler capacity reported in thousands pounds of steam
per hour as the maximum capacity on the base fuel.
Primary Fuel:
Bituminous Coal Waste Heat
Oil Waste Heat, Auxiliary firing
Natural Gas Lignite
Wood Bark, or Solid Wood Raw Municipal, Unsorted
Bagasse Raw Municipal, Non-combustible
Removed
Black Liquor Raw Municipal, Sorted & Sized
Other Fuels Other Industrial Waste
Alternate or Auxiliary Fuel
Firing Method:
Pulverized Coal Gas Turbine or Engine Exhaust
Spreader Stoker Other Non-combustible Waste Gas
Underfeed Stoker Combustible Waste Gas
Overfeed Stoker Non-solid Fuel Firing
Other Fuel Firing
Packaged or Field Assembled:
Pressure vessel completely shop assembled.
Pressure vessel shop assembled and shipped as two,
three, four, five, or six major modules. .
Packaged design shipped knocked-down.
Field assembled, bottom supported.
Field assembled, top supported
Standard Industrial Classification Number, 2 digit.
Domestic or Export.
Stationary or Marine Boiler.
* 1000 Btu/hr = 1 Ib steam/hr.
-------
14
Type Code:
Watertube
Wasteheat, watertube, bare tube
Wasteheat, watertube, extended surface
Draft Conditions:
Pressurized furnace
Balanced draft
Steam or Hot Water
Capacity per Unit (MW):
For utility boilers, the manufacturers rating in MW of
the generator is to be used with the boiler as indicated.
If the boiler is a non-generating unit, the designation
"9999" is indicated.
Design Pressure, psig
Operating Pressure, psig
Saturated Steam or Hot Water Outlet Temperature
Saturated Steam
Hot Water Outlet Temperature, °F
Superheat, °F
Steam Temperature at First Reheater Outlet, °F
Steam Temperature at Second Reheater Outlet, F.
The aforementioned data are recorded for each watertube boiler sold;
however, there are no data recorded regarding a projected installation date
or specific location for the boiler.
Firetube (Scotch) boiler data are recorded mainly in the form of
number of boilers sold each month. Number sold is recorded for low-pressure
steam, high-pressure steam and hot-water boilers. Fuel (gas, oil, or combi-
nation gas and oil) is recorded but not for a specific boiler, e.g., of the
total number of LP, HP, and HW boilers sold, it is possible to determine
that a given percentage of all boilers burned oil but not that a given per-
centage of only LP boilers burned oil. None of the firetube data are
currently computerized.
-------
15
The two main differences between the files are (1) the ABMA data
concerns boilers sold while the NEDS data concerns boilers in the field, and
(2) The ABMA file is a boiler information file whereas NEDS is a "source"
registration file which contains information on emission sources but not
primarily on boilers. Because boiler information is not stressed heavily
in NEDS, the accuracy of the boiler information is unknown. The data are
probably good for both large boilers and large companies since large com-
panies tend to supply more knowledgeable people to fill out emission inven-
tory questionnaires and data on large boilers are usually more fully
documented. However, as the boiler sizes get smaller and the company size
gets smaller, the information file probably tends to become more and more
suspect. Other sources of boiler data, such as the Walden Survey and
various trade journals were reviewed in addition to using the two computer
files to characterize the boiler population.
-------
16
DESCRIPTION OF PARAMETERS NECESSARY TO
CHARACTERIZE BOILER POPULATION
In order to estimate SO control costs or even decide on a feasi-
ble control technology, one must first be able to quantify the pertinent
characteristics of the S0_ source. Several boiler parameters become impor-
tant in this quantification process, thus the need to define classes of
boilers by these parameters.
In general, one would like to maintain as few classes of parameters
as possible in any analysis to reduce the amount of material that must be
comprehended, and to increase the count number in each class and, thus, the
accuracy of any deductions about the parameter. On the other hand, any
factor which might cause a decided change in pertinent characteristics of
a class must be used to split the class. For the present study, as a
result of both the viewpoint just presented and the results presented in
Reference (2), the following parameters have been considered.
Use Category
Three use categories were considered: commercial, industrial,
and utility. While it has been common to use arbitrary design capacity for
a dividing line between use categories, this study has shown that the annual
load factor is more dependent on use category than on design size, or even
fuel. If one were to calculate a load factor for several given size classes
of boilers, irrespective of use category, one would see an apparent increase
in load factor with increasing design size. The NEDS data, however, when
subdivided by use categories, shows that the load factor is independent of
design size but dependent on use category. The commercial boilers have a
lower load factor over almost all design size ranges than do industrial
boilers, which in turn have a lower load factor than utility boilers. An
average load factor over .an active boiler class will appear to increase
with size simply due to the mix (number) of boilers in the classes.
-------
17
Size
In Reference (2), in classifying data as a function of boiler size,
the total population of boilers was divided into 5 size ranges (in 10^ Btu/hr
steam) as follows: 10+ - 20, 20+ - 50, 50+ - 100, 100+ - 200, 200+ - 500.
To maintain consistency in form, the lower limit in each range, including
the lowest range, starts slightly above the division, e.g., 10,001, 20,001,
etc. Divisions at these points were found to be consistent with the liter-
ature. In the present study, this range is extended to cover the entire
range of NEDS data. The assumption is made that ABMA data on Ib/hr of steam
can be converted to Btu/hr by use of a factor of 1000.
Table 1 shows the total classification used to cover the NEDS data.
In addition to the nominal ranges, as listed, the geometric mean size is
presented. Class 1* is a result of the NEDS reporting format, i.e., the
smallest size able to be recorded is 1 x 10 Btu/hr and only boilers in the
size range .5 x 10 - 1.4 x 10 Btu/hr may be included in the NEDS as a
1 x 10 Btu/hr boiler. It was possible to identify numbers of boilers and
total installed capacity for each fuel type in Class 1* but the fuel proper-
ties, i.e., sulfur and ash content, etc., were assumed to be the same as for
Class 1.
-------
18
TABLE 1. SIZE RANGES USED TO CLASSIFY NEDS DATA
Size
Class
Number
!*<•>
l(b)
2
3
4
5
6
7
8
9
10
11
12
13
Nominal
Range of
Capacity,
10° Btu/hr
1
1+ - 2
2+-5
5+ - 10
10+ - 20 x 10
20+ - 50 x 10
50+ - 100 x 10
100+ - 200 x 102
200+ - 500 x 102
500* - 1000 x 102
1000+ - 2000 x 103
2000+ - 5000 x 103
5000+ - 10,000 x 103
10,000+ - 20,000 x 104
Log Mean
Size,"0-*
106 Btu/hr
.837
1.936
3.674
7.563
14.640
32 . 140
71.210
141.900
316.700
707.600
1415.000
3163.000
7072.000
14140.000
(a) 1* does not refer to size range, but to a specific
set of data with minimum recordable capacity.
(b) Because of the incremental values of 1, 2, 3,
this set of data contains only data of labeled
capacity 2 x 10^ Btu/hr.
(c) Square root of the product of maximum size in range
(to nearest .1 x 10° Btu/hr) and minimum size in
range (to nearest .1 x 10 Btu/hr).
-------
19
Fuel Type
The fuel divisions of main interest to this project are coal,
residual oil, distillate oil, and gas. Because of the differences in
firing methods and thus in the amount of excess air, however, the coal
types are split into pulverized fuel and cyclone in one classification,
and all other systems (stoker types) in the second classification. An
examination of the data has also shown that non-natural-gas data appear to
be highly scattered. Therefore, the gas data are divided into two classes,
natural gas (including LPG) and non-natural gas. This results in six
categories as shown in Table 2.
TABLE 2. COMPILATION OF FUEL TYPES
Pulverized coal and cyclone fired coal
Stoker and other coal not in item above
Residual oil
Distilled oil
Natural gas and LPG
Other gases not in item above
Annual Load Factor
The annual load factor was computed for each boiler for which NEDS
data were available by dividing the total fuel heat input rate per year by
the design firing rate and the number of hours per year. In several instances,
this value was greater than unity, often by a considerable amount. This
indicated some error in these data, so the value of 1.05 was arbitrarily
assigned. This will result in some overestimation, but neglect of these
units will result in too low a value since some boilers are fired at above
the design rate. Comparison of the portion of the data developed for this
study with the data of Reference (4) indicates only about 5 percent differ-
ence in annual load factor for each class. For each category defined by size
-------
20
class, use category, fuel, and sulfur content, an average annual load factor
was calculated.
A consideration of the annual load factor values for each category,
along with the number of boilers and total capacity involved in defining
each value, led to the conclusion that, in most instances, the annual load
factor was not a function of design firing rate of the boiler, within the
accuracy to which it could be determined. Figure 1 and Table 3 present
some of the results of the individual class calculations and show that the
load factor is independent of design size class for most cases. In cases
of extremely high or low load factors, there are few boilers in the class
from which to calculate an average so that any anomalies in the data will
have a greater than normal effect on the average.
Average load factor values were calculated for all fuel types and
size classes based on the NEDS. These averages are shown in Table 4. The
same average load factor results were obtained by either number weighting
or total design-firing-rate weighting the annual load factor for each cater
gory. In the interest of accurate cost calculations for Parts II and III
of this report, and because they were already calculated, it was decided to
use specific load factor values from Table 3 instead of the average values
from Table 4 for the cost equations. For future work, however, the con-
clusion formed was that the values from Table 4 will be sufficient for
costing purposes.
In summary, it was found that for most practical purposes, for
the range of design capacities considered herein, from 10 Btu/hr through
2 x 10 Btu/hr, the parameters that fix the annual load factor are the
category of use, and the type of fuel (with sulfur content not important).
Stack Temperature
The stack temperatures shown in Table 5 and Figure 2 are the
average temperatures from NEDS for size classes 1 through 8 (non-utility
combustion source classes). They are defined as "the temperature of the
exhaust stream at the stack exit, in degrees Fahrenheit, under normal oper-
ating conditions. If measured temperatures are not available, an estimate
to the nearest 50°F should be made."
-------
1.0
0.9
0.8
0.7
•o
o 0.6
c 0.5
c
§, 0.4
o
i_
QJ
< 0.3
0.2
O.I
0.0
Commercial boilers, high sulfur coal fired
Commercial boilers, high sulfur oil fired
Industrial boilers, high sulfur coal fired
Industrial boilers, high sulfur oil fired
1
I
4 5
Boiler Size Class
8
FIGURE 1. AVERAGE ANNUAL LOAD FACTOR FOR HIGH SULFUR COAL FIRED AND HIGH SULFUR
OIL FIRED BOILERS IN EIGHT NON-UTILITY COMBUSTION SOURCE SIZE CATEGORIES
-------
22
TABLE 3. ANNUAL AVERAGE LOAD FACTOR FOR HIGH SULFUR COAL AND
OIL-FIRED BOILERS IN 8 AREA SOURCE SIZE CLASSES
Load
Size Class Fuel Type Commercial
1 High Sulfur Coal .381
2
3
4
5
6
.371
.752
.393
.417
.292
7 .228
8 y -381
1 High Sulfur Oil .459
2
3
4
5
6
7
.350
.302
.279
.276
.259
.223
8 W .149
Factor
Industrial
.339
.415
.475
.599
.497
.417
.428
.411
.523
.372
.357
.351
.263
.338
.383
.430
-------
23
TABLE 4. ANNUAL LOAD FACTORS AS A FUNCTION OF
USE AND FUEL TYPE, BASED ON NEDS DATA
Commercial
(e)
Industrial
Utility
(b)
(a)
Stoker coal
Pulverized coal
Residual oil
Distillate oil
(c)
Natural gas
Other than natural gas
(d)
0.305
0.424
0.245
0.206
0.318
0.426
0.524
0.368
0.330
0.518
0.630
(8)
(g)
(h)
0.479
0.423
0.429LS/0.647HS(i)
0.474
(k)
(a) Stoker coal includes all coal firing except pulverized coal and cyclone.
(b) Pulverized coal includes cyclone.
(c) Natural gas includes natural gas and LPG. All low sulfur.
(d) Other than natural gas includes all gases except natural gas and
LPG. Almost all low sulfur.
4-6 8
(e) Load factor values valid for sizes in range 1 x 10 to 5 x 10 Btu/hr.
Values ranged from 0.01 to 0.16 for the few boiler data above this
design size range.
Q
(f) Up through 5 x 10 Btu/hr. For the few boilers above, use 0.268.
Q
(g) Up through 5 x 10 Btu/hr. For boilers above, use 0.580.
Q
(h) No data above 5 x 10 Btu/hr.
8
(i) Use industrial value up to 5 x 10 Btu/hr.
(j) All low sulfur.
(k) Very sparse population; suggest using industrial value.
-------
24
TABLE 5 . AVERAGE TEMPERATURE OF STACK GASES IN °F AS
A FUNCTION OF FUEL AND NON-UTILITY COMBUSTION
SOURCE SIZE CATEGORY
Size Class(i;>
Fuel
Coal
Residual
Distillate
Gas
1
535
403
338
555
411
354
377
2
489
482
370
413
501
387
420
3
460
423
360
431
418
343
442
500
4
439
464
392
452
390
320
430
350
5
470
495
416
454
429
503
457
721
6
438
450
427
453
444
346
460
560
7
406
410
418
430
491
463
405
545
8
410
392
421
406
378
435
850
(1) Upper number is for low sulfur, and lower number is
for high sulfur.
An observation of the data, together with a consideration of the
number of items involved in each point, indicates that the majority of
temperatures are between 400°F and 450°F. There is a tendency for the
temperature to be higher for the high-sulfur residual than for the low-
sulfur residual; this is probably the result of a deliberate effort to
avoid acid condensation in the stack. The high values for high-sulfur gas
are associated with a small number of special gases, not natural gas. For
coal-fired units of smaller sizes (i.e., Classes 1-5), the temperature tends
to fall in the 450-500°F range rather than 400-450°F range.
Fuel Sulfur Content
Following categorization of a boiler with respect to size and
fuel type, the influence of fuel sulfur content (weight percent) was
examined. Two categories were selected, less than or equal to 1.0 percent
sulfur (defined as low sulfur) and greater than 1.0 percent sulfur (defined
-------
25
850 r
eoo
750
700
650
600
a>
a.
E
-------
26
as high sulfur). Since the sulfur content was to be used for the purpose
of calculating control costs on boilers, it was decided to use the modal
(most frequently occurring value) sulfur content rather than an average.
This would assure that costs would be calculated for fuel with sulfur
content more nearly like that of the fuels received by a boiler owner.
The mode was thus calculated for each category defined by size, fuel type,
and low or high sulfur group. There appears to be little relationship
between sulfur content modes in each class, i.e., sulfur content seems to
remain independent of use category. The only discernible trend is a slight
tendency of sulfur content to increase with boiler size increases.
Sulfur content, in this study, is a means to an end (control
costs); therefore, it was decided that further effort on the analysis of
the parameter, i.e., graphing it, would not be fruitful. Table 6 presents
the modal fuel sulfur content for each of 13 boiler size classes, six fuel
types (with low and high sulfur considered separately), and three use
categories.
Flue Gas Flow Rate
Initially, an attempt was made to analyze the flue-gas flow rates,
as reported in NEDS, with respect to design firing rate. Unexpectedly,
analysis of the distribution of flow rates with respect to design firing
rate indicated that there was little effect of design firing rate on flue
gas rates. Therefore, the flow rate data were split into classes defined
by boiler fuel type and use category and then normalized with respect to
design firing rate.
The object of all of the data manipulation was to determine if
one could define a "typical" flue gas flow rate given boiler parameters such
as size, use category, and fuel type. Further analysis of the data following
normalizing and different class divisions indicated that one could not define
a "typical" flue gas flow rate due to the wide variation in reported flow
rates for a given class of boilers. It was decided that an investigation
was necessary to determine the reason for the wide variation in the reported
NEDS flue gas flow rates. Upon examination of several references in .con-
junction with the NEDS data, it was found that at least six factors influence
the range of reported flow rates. These factors include (1) stoichiometric
-------
TABLE 6. MODAL FUEL SULFUR CONTENT, WEIGHT PERCENT
Fuel
Stoker Coal,
Low Sulfur
High Sulfur
Pulverized Coal,
Low Sulfur
High Sulfur
Residual Oil,
Low Sulfur
High Sulfur
Distillate Oil,
Low Sulfur
High Sulfur
Non-Natural Gas,
Low .Sulfur
High Sulfur
Natural Gas ,
Low Sulfur
High Sulfur
Class
0: 5+-10 x 105 Btu/hr
Com. Ind. Utility
0.64 0.75 0
2.00 1.01. 2.5
0.65 0.65 0
0 3.50 0
1.00 0.96 0.75
2.29 2.41 2.5
0.26 0.23 0.1
1.01 1.01 0
0 0 0
00 0
00 0
00 0
Class
1: l+-2 x 106 Btu/hr
Com. Ind. Utility
0.64 0.75 0
2.00 1,01 2.5
0>65 0.65 0
0 3.50 0
1.00 0.96 0.75
2.29 2.41 2.5
0.26 0.23 0.1
l.Ol 1.01 0
00 0
00 0
00 0
00 0
Class
2: 2+-5 x 106 Btu/hr
Com. Ind. Utility
0.65 1*00 0
3.40 3.28 3.25
0.65 0
0 4.00 0
0.99 0.96 0.95
l.oi 2.21 2.5
0.25 0.24 '0.1
1.01 1.01 0
00 0
0.0 0
00 0
00 0
Class
3: 5+-10 x 106 Btu/hr
Com. Ind. Utility
0.65 0.98 0
3.22 2.50 0
0.65 0.85 0
00 0
0.98 0.97 0.94
2.23 2.29 2.5
0.25 0.24 0.1
1.01 1.01 0
00 0
0 2.50 0
00 0
0 1. 0
(1) Top number is low sulfur value, bottom number is high sulfur value.
-------
TABLE 6. MODAL FUEL SULFUR CONTENT, WEIGHT PERCENT (Continued)
Fuel
Stoker Coal,
Low Sulfur
High Sulfur
Pulverized Coal,
Low Sulfur
High Sulfur
Residual Oil,
Low Sulfur
High Sulfur
Distillate Oil,
Low Sulfur
High Sulfur
Non-Natural Gas,
Low Sulfur
High Sulfur
Natural Gas,
Low Sulfur
High Sulfur
Class
4: l+-2 x 10'Btu/hr
Com. Ind. Utility
0.64 0.91 0.54
2.86 2.46 1.01
0.91 0.1
1.01 3.43 2.5
0.99 1.00 0.45
2.27 2.28 2.47
0.25 0.22 0.2
1.01 1.01 0
00 0
00 0
00 0
0 3.50 0
Class I01*88
5: 2+-5 x 10'Btti/htj 6: 5-10 x 107R*-,,/vJ
Com. Ind. Utility
0.65 0.75 0.65
2.58 2.47 2.43
0.83 0.75 0.75
3.50 2.13 1.01
0.99 0.99 0.45
2.06 2.37 2.46
0.25 0.22 0.15
1.01 2.31 0
00 0
0 1.01 0
00 0
0 4.00 0
Com. Ind. Utility
0.96 0.95 0.96
2.54 2.56 3.14
0.83 0.93 0.85
1.01 2.53 3.45
0.99 1.00 0.55
2.31 2.39 2.41
0 . 27 0 0 . 23
1.01 1.01 2.5
00 0
0 1.01 0
00 0
0 1.01 0
Class
7: 1+-2 x 108 R*,,/V,T.
Com. Ind. Utility
0.45 0.98 0.99
3.29 2.85 2.0
0.7i 0.93 0.93
2.0:) 2.45 2.39
0.93 0.99 0.75
2.2J 2.38 2.44
0.25 0.15 0.25
1.01 1.01 0
00 0
0 1.01 0
00 0
0 1.01 0
to
00
-------
TABLE 6. MODAL FUEL SULFUR CONTENT, WEIGHT PERCENT (Continued)
Fuel
Stoker Coal,
Low Sulfur
High Sulfur
Pulverized Coal,
Low Sulfur
High Sulfur
Residual Oil,
Low Sulfur
High Sulfur
Distillate Oil,
Low Sulfur
High Sulfur
Non-Natural Gas,
Low Sulfur
High Sulfur
Natural Gas,
Low Sulfur
High Sulfur
Class
8:
Com.
0.73
3.25
—
0
0.96
1.01
0.25
0
0
0
0
0
2+-5x
Ind.
0.65
1.01
0.76
2.22
0.44
2.43
0.15
0
0
3.50
0
0
108 Btu/hr
Utility
0.1
3.0
0.92
2.96
0.85
2.34
0.24
0
0
1.01
0
0
Class
9:
Com.
0
0
0
0
0.90
0
0.20
0
0
0
0
0
5*-10
Ind.
0.35
2.33
0.86
3.29
0.97
2.40
0.15
0
0
1.01
0
0
0
X 10 Btutar
Utility
0.75
3.67
0.92
2.39
0.85
2.26
0.25
2.50
0
0
0
0
Class
10:
Com.
0
0
0
0
0
0
0
0
0
0
0
0
+
Ind.
0.75
3.5
0.85
3.33
0.45
2.5
0
0
o -
1.01
0
0
X 10'Rt-n/hr-
Utility
0
4.0
0.93
2.64
0.44
1.01
0.95
0
0
0
0
0
-------
TABLE 6. MODAL FUEL SULFUR CONTENT, WEIGHT PERCENT (Continued)
Fuel
Stoker Coal,
Low Sulfur
High Sulfur
Pulverized Coal,
Low Sulfur
High Sulfur
Residual Oil,
Low Sulfur
High Sulfur
Distillate Oil,
Low Sulfur
High Sulfur
Non-Natural Gas ,
Low Sulfur
High Sulfur
Natural Gas,
Low Sulfur
High Sulfur
Class
11: 2+-5 x 109
Com
0
0
0
0
0
0
0
0
0
0
0
0
Ind.
0
3.5
0
1.01
0.45
2.5
0
0
0
0
0
0
Btu/hr
Utility
0
4.0
0.86
2.69
0.99
2.28
0.35
0
0
0
0
0
Class
12: 5+-10 x
Com.
0
0
0
0
0
0
0
0
0
0
0
0
Ind.
0
0
0
5.5
0
0
0
0
0
0
0
0
109 Btu/hr
Utility
0
0.56
2.76
0
2.5
0.65
0
0
0
0
0
Class
13: 1+-2 x
Com.
0
0
0
0
0
0
0
0
0
0
0
0
Ind.
0
0
0
0
0
0
0
0
0
0
0
o •
1010 Btu/hr
Utility
0
0
0
4.5
0
0
0.25
0
0
0
0
0
CO
o
-------
31
firing rate, (2) excess air, (3) leakage, (4) corrections for stack temper-
ature, (5) the boiler load factor, and (6) the fan flow safety margin.
The combination of these variables can result in differences as high as a
factor of six between flow rates reported for a given boiler size class.
Thus, it can be shown that reasonable design assumptions based on the
literature lead to justification of the data as reported in the NEDS.
In order to prepare SCL control alternative cost calculations as
in Parts II and III of this report, it was decided that flue gas flow rate
data would be supplied by utilizing an alternative source of actual stack
data. This was accomplished by a stack test data literature review for
the specific sizes of boilers to be controlled. . The size of boiler to be
controlled was based on the SCL emissions calculations as described in
the next section.
The flue gas flow rates were based on averages from Reference (5)
and were as shown in Table 7.
TABLE 7. AVERAGE FLUE GAS FLOW RATE FOR VARIOUS
BOILER SIZES AND FUELS
Fuel
High Sulfur Oil
High Sulfur Oil
High Sulfur Oil
High Sulfur Pulverized Coal
High Sulfur Pulverized Coal
High Sulfur Stoker Coal
High Sulfur Stoker Coal
Size (106 Btu/hr)
2
20
250
20 ,,
250
- ; 20
. 250
Flow Rate (scfm)
450
4430
63333
5300
62500
6660
68750
-------
32
BOILER POPULATION CHARACTERIZATION
Table 8 presents data on the number, total installed capacity,
and average hourly fuel use of boilers in each class defined by size, fuel
type, and use category. Since the NEDS contains information on all sizes
of boilers in the national inventory, the pertinent information for
boilers other than non-utility combustion sources is included in the table.
The information presented in the table is adjusted for all available data
sources, including NEDS, ABMA. sales data, and the boiler literature review
(per Reference 6).
In dealing with SO control alternative analysis, the average
X
hourly fuel use is the most important parameter shown. Average hourly fuel
use is shown as the heat input equivalent of the amount of fuel used per
hour by the total number of boilers in each class. It takes into account
the annual load factor of each boiler class and is the parameter through
which one may get an estimate of the actual amount of SO emissions by each
X
boiler class in a year.
It is for the above reason that Figures 3 through 10 are shown.
These figures are a graphical representation of average hourly fuel use by
boiler size class and use category for each type of fuel considered in
this study. In viewing these graphs, it is readily apparent that the
boilers burning natural gas and by-product gas use more fuel per hour
(heat equivalent) than other types of boilers. Oil fuel is second in
quantity used, and coal is third.
By using the low/high sulfur content (Table 7) of each fuel and
splitting the totals in Table 8 into their low/high sulfur components
using a ratio from the NEDS data, it was possible to calculate S0_ emissions
from each class of boiler. These emissions were calculated using EPA
emissions factors from Reference (8) and can be considered to be potential
emissions. They are emissions which would result if each boiler in the
country were fired at 100 percent of its design capacity for 24 hours a
day for an entire year. In order to obtain an estimate of actual S02
-------
TABLE 8. ESTIMATED NUMBER, TOTAL INSTALLED DESIGN CAPACITY.(10 BTU/HR),
' AND AVERAGE HOURLY FUEL USE (106 BTU/HR) AS A FUNCTION OF
DESIGN CAPACITY RANGE, FUEL TYPE, AND USE
Stoker coal
Pulverized coal
Residual oil
Distillate oil
By-Product Gas
Natural gas
Sua
Class sum
Class
1*: 0.5-1.0x10 Btu/hr '
Ccr... Ir.c. •-'::.
4,y;0 I./!-.- 1
4,910 1,734 1
1,495 7^7 0
0 2S1 0
0 281 0
0 1-7 0
5,261 3,067 4
5,261 -8,0(-7 4
1,289 2, 969 1
19,992 14,169 2
19,992 14,169 2
4,118 4,676 0
0 421 0
6 421 0
0 ?65 ' 0
11,644 15,853 1
11,644 15,853 1
3,703 8,212 0
41,807 40,545 8
41,807 40,545 8
10. 60S 17,016 1
82,360
82,360
27,625
Class
1: 1+ -2xl06 Btu/hr
Co™. • Ir.d. Ut.-
3,387 1,967 1
6,774 3,934 2
2,066 1,676 1
109 765 0
218 1,536 0
92 802 0
3,496 8,959 7
6,992 17,918 14
1,713 . 6,594 5
13,547 15,733 4
27,094 31,466 8
5,581 10,384 1
0 874 0
0 1,748 0
0 1,101 0
7,866 17,590 2
15,732 35,180 4
5,003 18,223 2
28,405 45,888 14
56,810 91,782 28
14,455 38,780 9
74,307
148,620
53,244
Class
2- 7+ -5*10&BtU/hl
Cor.. i;id. L't.
4,998 3,<>46 6
20,166 17,473 26
6,151 7,443 12
0 689 0
0 2,563 0
0 1,346 0
7,891 23,109 7
30,437 92,313 32
7,457 33,971 12
11,461 16,789 14
43,401 65,320 56
8,941 21,556 8
0 877 0
0 3,064 0
0 1,933 0
6,137 27,118 9
24,174 106,591 35
7,687 55,214 17
30,467 72,528 36
118,178 287,333 151
30,236 121,463" 49
103,051
405,662
151,743
Class
Com. Inii. ut.
3,917 3,264 3
32,699 25,789 17
9,973 10,986 8
27 326 0
190 2,584 0
81 1,354 0
3,972 13,983 12
32,345 111,806 82
7,925 41,145 30
3,047 4,758 5
22,824- 37,514 36
4,702 12,380 5
0 245 0
0 2,176 0
0 1,371 0
2,992 12,976 11
24,157 102,013 78
7,682 52,843 37
13,955 35,582 31
112,215 281,882 213
30^363 120,079 80
49,568
394,310
' 150,522
Class
4:i+ -7xin7Btu/hr
Co.-a. • Ir.d. Ut.
1,373 2,004 14
21,441 32.411 2-U
6,540 13?P37 117
63 428 2
908 6,931 28
385 3,632 12
2,620 7,453 17
40,564 114,311 2S2
9,938 42,066 104
856 1,764 5
11,879 26,879 90
2,447 8,870 12
0 84 0
0 1,315 • 6
0 828 0
981 6,493 33
15,355 100,021 489
. 4,883 -51,811 232
5,898 18,226 71
90,147 281,868 1,133
24,193 121,014 477
24,195
373,148
145,684
Top number is number of boilers, middle number is total installed design capacity, and the bottom number is the
product of annual load factor and total installed design capacity.
-------
TABLE 8. ESTIMATED NUMBER, TOTAL INSTALLED DESIGN CAPACITY (10 BTU/HR),
AND AVERAGE HOURLY FUEL USE (10$ BTU/HR) AS A FUNCTION OF
DESIGN CAPACITY RANGE, FUEL TYPE, AND USE (Continued)
i
i
Stcker coai
Pulverized coal
Residual oil
Distillate oil
By- Product Gaa
Natural gas
Sua
Class sum .
Class _
5*'2 -5xlo'Btu/hr
Cc-.. In-.:. t-t.
1,033 2,379 39
35. --7 91,037 1,482
1C, £11 3r-,io3 710
100 36S 14
3,289 14,100 592
1,395 7,33S 250
1,926 5,058 17
65,042 169,137 282
1,593 62,242 104
442 858 5
13,533 25,063 90
2,793 9.261 12
0 174 0
0 7,038 0
0 4,447 0
684 5.000 56
21,721 276,fc42 1,950
6,967 149,404 924
4,205 14,037 131
139,057 507,137 4,396
23.559 271,907 2,000
18,373
730,590
297,466
Class . 7
6: 5 -10xlo'Btu/hr
Com. Ind. Ut.
336 1,573 67
24,609 115,979 5,546
7,506 49,194 2,657
33 294 19
2,518 24,679 1,549
1,068 12,932 655
621 1,785 82
44,009 131,806 5,497
10,782 48,505 2,023
121 203 14
8,879 15,992 1,070
1,829 5.2C.4 139
0 170 0
0 12,124 0
0 7,638 0
224 2,158 43
17,691 156,194 6,161
5,626 80,908 2,920
1,335 6,183 265
98,706 456,774 19,823
26,811 204.441 8,394
7,783
575,303
239,646
Cla78:V-2xl06Btu/hr
Con. Ind. Ut.
47 643 117
6,088 92,010 17,701
1,857 39,196 829
16 409 81
2,229 60,155 12,517
945 31,521 5,295
151 743 168
20,914 106,429 25,930
5,124 39,166 9,542
41 114 43
6,149 16,388 6,239
1,267 5,408 811
0 137 0
0 18,108 0
0 11,408 0
100 918 181
13,982 136,580 27,733
4,319 70,798 13,145
355 2,964 590
49,362 429,670 90,118
13,512 197,497 29,622
3,909
569,150
240,631
Class A c.
8: 2 -5xlO°Btu/hr
Com. Ind. L't.
13 157 38
3,831 45,085 16,503
1,168 19,205 7,905
0 220 302
0 66,569 103,233
0 34,832 43,683
41 285 243
13,009 .86,684 74,605
3,187 31,900 27,455
17 29 67
5,833 8,717 20,949
1,202 2,877 2,723
0 60 4
0 19,181 1,496
0 12,089 942
36 324 33
12,099 98,373 108,281
3,835 50,957 51,325
107 1,075 707
34,772 324,609 325,118
9,392 151,911 134,033
1,889
634,499
295,341
Class + s
9: 5 -10x10 Bi-M/h
Com. . Ind. Ct.
0 15 5
0 11.071 3,005
0 2,967 1,439
0 51 235
0 33,022 168,744
0 19,152 71,379
4 44 163
4,209 30,273 111,307
1,031 11,140 57,165
3 3 33
1,907 2,492 21,735
393 822 2,826
0 26 0
0 16,554 0
0 10,429 0
5 75 241
2,794 48,744 169,221
883 25,249 8,021
12 214 677
8,910 142,161 474,012
2,312 69,759 140,230
903
625, C33
212,901
i
u>
* Top number is number of boilers, middle number is total installed design capacity
and the bottom number is the product of annual load factor and total installed design capacity.
-------
TABLE 8. ESTIMATED NUMBER, TOTAL INSTALLED DESIGN CAPACITY (10 BTU/HR),
AND AVERAGE HOURLY FUEL USE (106 BTU/HR) AS A FUNCTION OF
DESIGN CAPACITY RANGE, FUEL TYPE, AND USE (Continued)
i
i
Stoker coal
Pulverized coal
Residual oil
Distillate oil
By-Product Gas
Natural gas
Sum
. Class sum
Class
iO*:l+-2xl09Btu/hr
Co-.. inc. Ut.
U 5 2
0 6,831 3,123
0 l.S3% 1,496
0 10 266
0 16,019 370,857
0 9.^91 159,410
0 12 83
0 17,450 114,113
C 6,422 57,910
- 0 0 3
0 0 4,129
0 0 537
0 .4 0
0 5,604 - 0
0 3,531 0
4 .. 32 - '174
6,301 46,063 242,621
2,004 23,561 115,010
4 63 528
6,301 91,987 -734,837
2,004 44,941 354,404
395-
833,125
401,349
Class
11: 2+-5xl03'Btu/hr
Con. inri. tt.
022
0 7,600 6,132
0 2,037 2,937
0 5 125
0 11,995 378,536
0 6,928 160,121
0 1 43
0 3,020 128,886
0 1,111 70,068
01 1
0 2,274 2,854
0 750 371
0 0 . 0
,0 0 0
000
1 11 80
2,626 28,908 247,992
• 835 14,474 117,543
1 20 251
2,626 53,797 764,400
835 25,300 351,045
272
820,823
377,180
Class
12: s'-lOxlO^tu/hr
Com. Ind. Ut.
000
000
0 00
0 1 53
0 5,068 327,022
0 2,939 138,330
0.0 5
0 0 26,906
0 0 17,408
0 0.1
0 ,0 5,184
0 : 0 674
000
000
000
0 5 17
0 38,964 101,915
0 20,183 48,308
0 6 76
0 44,032 461,027
0 23,122 204,720
82
505,059
227,842
Class Class
13: l+-2:<101(Btu/hi; 14. > 2xl010 Btu/hr
Com. Ind. Ut.
0 0 U
000
000
002
0 0 22,466
0 0 9,503
0 0 0
0 " 0 0
000
001
0 0 13,070
0 0 1,699
000
0 00
000
1 4 1
10,350 61,550 10,931
3,291 31,883 5,191
144
10,350 61,550 46,517
3,291 31,883 16,393
9
. 118,417
31,567
Coa. Ind. Ut.
U 0 U
0 O'O
0.0 0
0 0 4
00-
00
0 00
000
000
000
0 0 0
000
000
000
0 0 0
1 2 1
1 25
8
Top number is number of boilers, middle number is total installed design capacity,
and the bottom number is the product of annual load factor and total installed design capacity.
-------
36
1000
100
a
ffi
"Q
0>
Ifl
"5
I
o>
o
67 8
Size Class
10
13
FIGURE 3. ANNUAL HOURLY FUEL USE BY SIZE CLASS AND USE CATEGORY
FOR STOKER COAL FUEL
-------
37
1000
1*1 2 3 45 6 7 8 9 10 II 12 13
Size Class
FIGURE 4. ANNUAL HOURLY FUEL USAGE (10 Btu/hr) BY SIZE CLASS AND
USE CATEGORY FOR PULVERIZED COAL FUEL
-------
38
1000
100 —
D
m
"o
w
01
o
m
O
I . 2 . 3 4- .5 ..,• 6^- -.7 ,5 8 9' 10-
10 —
I '12 13
FIGURE 5. ANNUAL HOURLY FUEL USE BY SIZE CLASS AND USE CATEGORY
FOR RESIDUAL OIL FUEL • ; ' .
-------
39
1000
100
CO
Ol
o
0)
Ol
o
.g
_ ti
•
O
I
H-
o
81 10
_j
I 111
II 11
I I 11
1
5678
Size Class
10 II 12 13
FIGURE 6. ANNUAL HOURLY FUEL USE BY SIZE CLASS AND USE CATEGORY
FOR DISTILLATE OIL FUEL
-------
40
7-^8910, II 12 .13
FIGURE 7. ANNUAL HOURLY FUEL USE BY SIZE CLASS AND USE CATEGORY
FOR BY-PRODUCT GAS FUEL
-------
41
1000
4 . 5 . . 6 7 89 10 :f;ll 12 13
I*
FIGURE 8. ANNUAL HOURLY FUEL USE BY SIZE CLASS AND USE CATEGORY
FOR NATURAL GAS FUEL
-------
42
1000
100
3
£
"
' 3
O
8.
S
01
o
o
10
2 .
4
5 :6 7 8
Size Class
10
12
13
FIGURE 9. ANNUAL HOURLY FUEL USE BY SIZE CLASS AND USE CATEGORY
FOR THE TOTAL OF ALL FUELS PER SIZE CLASS
-------
43
1000
i too
o
0>
(U
. o
0)
in
D
—
-
-
™
-
#
2 3
I
4
i
i
5
t
6
7
8
9
10
II
12
13
Size Class
FIGURE 10. ANNUAL HOURLY FUEL USE BY SIZE CLASS FOR ALL USE
CATEGORIES AND ALL FUELS IN EACH CLASS
-------
44
emissions, one can adjust the boiler design capacity in each class by its
appropriate annual load factor. The figures in Table 8 were adjusted by
load factor in calculating the average hourly fuel use. Applying EPA
emission factors to the average hourly fuel use (from Table 8) values will
thus yield an estimate of actual SO- emission for each class. Table 9
(from Reference 6) shows the potential and actual S0« emissions for each
;. *
class of boiler as calculated iri the aforementioned manner. Once again,
in addressing SO- control alternative analysis, the estimate of actual SO-
emissions is the important parameter; Figures 11;^through 20 are graphical
illustrations of estimated actual SOt, emissions for each size class, use
category, and fuel. By-product gas -and high sulfur natural gas are not
shown since their emissions were negligible.
. The graphs readily show that preponderance of non-utility SO-
..• ' ' *
emissions appears to come from high-sulfur coal-fired units in classes 5
through 8 (10 x 10 Btu/hr to 500 x 10 Btu/hr) and from high-sulfur oil-
fired units in classes 1 through 8 (1 x 10 Btu/hr thrpugh .500 x 10 Btu/hr),
-------
TABLE 9. SULFUR DIOXIDE EMISSIONS (TONS/YR) BY BOILER SIZE CLASS
Fuel
Stoker Coal,
Low Sulfur
Stoker Coal,
High Sulfur
Pulverized Coal,
Lou fulfur
Pulverized Coal,
High Sulfur
Residual Oil,
Low Sulfur
Residual Oil,
High Sulfur
Distillate Oil,
Lou Sulfur
Distillate Oil,
High Sulfur
By-Product Gas
Lou Sulfur
By-Product Gas :
High Sulfur
Natural Gas,
Low Sulfur
Natural Gas,
High Sulfur
Class 1*:
Cora.
13,993.0
4.269.3
40.155.0
12,250.0
0
0
0
0
12,299.0
2,394.8
39,770.0
9,744.3
39,001.0
5,355.6
12,555.3
2,586.3
0
0
0
0
30.6
9.7
0
0
5+-lO x io5 Btu/hr
Ind. Utility.
5,027.5 0
2,509.3 0
3,497.2 15.6
1,489.1 0
944 . 8 0
494.3 0
3,118.0 0
1,087.4 0
27,378.0 0
10,080.0 0
78,760.0 39.2
28,987.0 9.9
17,713.0 0.9
5,845.5 0
4,301.3 0
1,419.6 0
9.8 0
6.1 0
0 0
0 0
40.9 0
21.6 0
0 0
0 0
Class 1:
Com.
19,305.3
5,887.9
55,398.0
16,896.1
821.2
346. 5
0
0
16,345.4
3,182.6
52,855.4
12,949.5
35,237.4
7,258.4
17,015.5
3,504.9
0
0
0
0
41.4
13.2
0
0
l+-2 x IO6
Ind.
8,143.8
4,065.8
12,746.1
5,430.2
5,164.4
2,696.5
17,043.5
8,899.0
60,836.2
22,388.1
174.938.2
64,379.7
39,336.0
12,981.2
9,552.4
3,152.6
40.5
25.5
0
0
92.5
47.9
0
0
Btu/hr
Utility
0
0
31.3
15.6
0
0
0
0
40.9
14.7
19.6
6.9
3.4
.4
0
0
0
o-
0
0
0
0
0
0
Class 2:
Com.
55,398.3
16,897.5
169.298.4
51.638.9
0
0
0
0
92,259.0
22,603.5
85,125.3
2,055.4
49,072.0
9,834.9
26.593.3
5,478.3
0 .
0
0
0
63.6
20.2
0
0
2+- 5 x IO6
Ind.
57,938.
24,681.8
501,124.2
213,466.7
8.578.6
4,496.2
27,816.6
14,580.8
354.549.5
130,473.5
1.050,932.6
386,741.1
60,079.7
12.377.0
24,484.6
5,044.3
71.0
44.8
0
0
280.3
145.2
0
0
Rf-ii /hv
Utility
0
0
1,941.9
896.2
0
0 '
0
0
211.4
79.2
52.4
19.8
34.0
4.9
0
0
0
0
0
0
0
0
0
0
Tlaco 3:
Com.
186,993.2
57,032.1
154,839.3
47,224.6
735.0
313.4
0
0
63,543.9
15.569.0
1,074,148.4
263,184.3
33,491.2
6,899.6
12.684.5
2,613.4
0
0
0
0
63.5
20.2
0
0
5"*"- 10 x IO6
Ind.
74,815.7
31.871.2
313.632.4
133.605.8
11,718.2
6,140.3
0
0
302,675.7
11,135.7
1,245,368.0
458,300.0
92,901.1
30,658.3
31,656.6
10,447.0
41.8
26.4
214.3
135.0
768.3
139.0
0
0
Btu/hr
Utility
0
0
0
0
0
0
0
0
544.5
199.3
165.5
60.4
15.7
2.2
0
0
0
0
0
0
0
0
0
0
.e-
Ln
(1)
1* indicates either model fuel sulfur content of zero or no installed capacity in the category,
-------
TABLE 9. ESTIMATED SULFUR DIOXIDE EMISSIONS (TONS/YR) BY BOILER SIZE CLASS (Continued)
ruel
Stoker Coal,
Low Sulfur
Stoker Coal,
High Sulfur
Pulverized Coalt
Low Sulfur
Pulverized Coal,
High Sulfur
Residual Oil,
Low Sulfur
Residual Oil,
High Sulfur
Distillate. Oil,
Lou Sulfur
Distillate Oil,
High Sulfur
By-Product Cos ,
Low Sulfur
By-Product Gas
High Sulfur
Natural Gas ,
Low Sulfur
Natural Gas,
High Sulfur
Class It:
Com.
49,113.2
14,980.6
206,802.6
64,669.7
0
0
5,727.
2,428.3
79,171.9
19,396.6
316,506.0
77,542.9
9,435.5
1,943.6
15,086.6
3,107.8
0
0
0
0
40.4
12.8
I
1+-2 x 107
Ind.
101,075.7
43,058.3
561,292.9
238,886.4
36,825.1
19,297.3
32,045.2
16.791.9
295,755.5
108,100.6
839,968.0
309,105.
38,537.8
12,717.4
8,811.6
2,907.6
30.5
19.2
0
0
263.
1,226.3
0
0
Btu/hr Class
Utility Com.
8,488.7 87,819.6
4,074.9 26,784.0
1,980.1 330,131.8
948.8 100,687.8
9.3 14,272.5
40.0 6,053.5
222.3 22,297.2
95.3 9,457.4
455.7 107,708.6
168.9 2,637.9
3,320.4 414,158.7
1,225.1 10,143.8
83.2 11,580.6
11.1 2,385.7
0 11,001.8
0 2,266.2
0 0
0 0
0 0
0 0
1.3 57.1
6.1 18.3
0 0
0 0
5: 2+-5 x 107 Btu/hr
Ind.
187,626.8
79,928.7
1,429,050.3
608,772.8
27,462.6
14,389.6
179,053.5
93,818.9
333,207.6
122,619.7
1,300,827.
478,701.8
26,111.3
8,616.9
48.802.5
16,105.4
106.3
67.0
572.3
360.6
728.
392.9
0
0
Utility
3,601.2
759.1
24,912.2
11,934.8
1,459.4
616.0
1,993.6
841.8
290.9
107.3
1,949.6
719.0
0
0
0
0
0
• 0
0
0
5.2
2.4
0
0
Class 6:
Com.
66,312.8
19,920.6
257,673.4
78,593.9
3,816.2
1,618.5
11,362.4
4,823.3
78,357.5
19,197.2
288,704.7
70,731.1
9.110.2
1,876.7
3,499.1
720.6
0
0
0
. 0
46.5
14.8
0
0
5+-10 x 107
.Ind.
249,368.1
105,772.9
1,502,051.4
637,114.0
37,197.8
19,869.5
314,961.7
165,041.7
321,849.8
11,844.2
1,099,533.
404,632.
33,281.4
" 10,955.
32,331.1
10,642.3
212.6
134.1
681.0
429.0
410.8
212.8
0
0
Rf-ii /hr-
Utillty
25,293.8
12,118.1
52,515.3
25,158.5
1,724.4
779.5
42,726.6
18,068.6
7.562.3
2.783.1
39,487.3
14,532.4
1,438.7
186.9
4,304.5
558.2
0
0
0
0
16.2
7.7
0
'o
rijjQc
Com.
6,334.6
1,932.1
94,402.6
28,795.5
7.356.2
3,118.9
10,351.7
4,472.9
44,111.5
10,880.9
116,918.4
28.646.5
10,141.2
2,089.6
2,017.2
415.6
0
0
0
0
36.8
11.4
0
0
7: l+-2 x 108
Ind.
254.713.7
108.507.3
1,092,888.4
465,567.5
130.612.4
68,440.7
699,'090.0
366,320.
244,067.9
89.897.2
739,378.5
272,092.4
75,331.1
24.859.1
6,358.5
2.098.3
314.6
198.2
1,048.0
660.7
359.
186.
0
0
Btu/hr
"EtiTUy
60,636.2
2,839.5
170,690.9
7,994.5
11.133.0
4,709.9
264.865.6
117,045.8''
64,851.2
23.864.6
130,721.8 .
48,104.4
7,774.1
1,010.5
0
0
0
0
0
0
72.9
34.6
0
0
(1) 1* indicates either model fuel sulfur content of zero or no installed capacity in the category.
-------
TABLE 9. ESTIMATED SULFUR DIOXIDE EMISSIONS (TONS/YR) BY BOILER SIZE CLASS (Continued)
Fuel
Stoker Coal.
Lou Sulfur
Stoker Coal.
High Sulfur
Pulverized Coal,
Lou Sulfur
Pulverized Coal,
High Sulfur
Residual Oil,
Lou Sulfur
Residual Oil,
High Sulfur
Distillate Oil,
Lou Sulfur
Distillate Oil,
Hifth Sulfur
Non-Natural Gas,
Lou Sulfur
By-Product Gas , t
High Sulfur
Natural Gas,
Lev Sulfur
Natural Gas,
High Sulfur
Class 8:
Com.
6,927.6
2,111.9
59.889.0
18,260.1
0
0
0
0
23,647.4
5,793.0
36,086.9
B, 841.0
5, 681.0
1.170.7
0
0
0
0
0
0
71.8
10.0
0
0
2+-5 x l08Btu/hr
Ind.
86,723.6
36,943.9
227,796.1
96,877.0
99,358.1
52,063.5
716,952.4
375,681.0
71,458.5
26,297.0
756,032.4
278,222.2
5,621.2
1,855.3
0
0
412.6
260.1
1,110.9
700.2
258.7
134.0
0
0
Utility
3,199.8
1,532.7
268,163.2
128,451.0
99,744.3
42.191.4
2.896,232.3
1.225,072.4
194,919.3
71,731.2
494,003.2
181,797.6
22,203.6
2,886.1
0
0
0
0
693.2
436. 5
284.8
135.0
0
0
Class
Com.
0
0
0
0
0
0
0
0
25,157.2
6.162.3
0
0
1.689.8
348.2
0
0
• o
0
0
0
7.3
2.3
0
0
9: 5"*"-10 x
Ind.
6,169.6
1,653.5
471,614.2
176.393.5
41.274.4
73,938.1
1.568.542.4
909,683.0
57,840.4
21,784.4
203.247.9
74,792.0
1,709.2
563.8
0
0
0
6
3,835.1
2.416.1
128.0
66.4
0
0
108Btu/hr
Utility
1,892.3
5.694.9
41.765.2
20,000.1
233,517.5
98.778.0
4,439,993.3
1.878,124.8
346,274.4
177,839.5
471,242.3
242.020.2
21.967.0
2,856.1
0
0
0
0
0
0
445.0
21.1
0
0
P. 1 a Q B
Com.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
00
0
0
16.6
5.3
0 .
0
10: i*
Ind.
27,157.6
7,277.9
46,209.7
12,383.4
27,751.7
16,096.0
253.371.6
146,954.9
15,705.8
5.780.0
90.857.1
33,438.0
0
0
0
0
0
0
1,298.3
818.0
121.1
62.7
0
0
"-2 x 10 R»-,,/hT-
Utlllty
0
0
158,391.8
75,873.9
493,300.2
212,041.2
9,026.003.0'
3.879.758.0
205,602.5
104,339.0
234,660.3
119,085.3
17,016.4
2,213.1
0
0
0
0
0
0
638.0
302.5
0
0
(1) 1 indicates either model fuel sulfur content or zero or no installed capacity in the category.
-------
TABLE 9. ESTIMATED SULFUR DIOXIDE EMISSIONS (TONS/YR) BY BOILER SIZE CLASS (Continued)
Fuel
Stoker Coal,
Lou Sulfur
Stoker Coal,
High Sulfur
Pulverised Coal,
Lou Sulfur
Pulverized Coal,
Nigh Sulfur
Residual Oil,
Low Sulfur
Residual Oil,
High Sulfur
Distillate Oil.
Low Sulfur
Distillate Oil,
High Sulfur
By- Product Can ,§
Low Sulfur
By-Product Gas f
High Sulfur
Natural Gas,
Low Sulfur
Natural Gas,
High Sulfur
Class
Com.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
6.9
2.2
0
0
11: 2+- 5 x
Ind.
0
0
177,243.8
47,506.0
0
0
84,826.2
48,993.4
0
0
0
0
0
0
0
0
0
0
0
0
76.0
38.0
0
0
109 Btu/hr
Utility
0
0
157,332.4
75,356.4
1,019,488.0
43,124.5
9,839,853.0
4,162,264.0
407.225.9
221,385.3
716,123.3
389,316.1
4,539.0
590.0
0
0
0
0
0
0
652.2
Class
Com.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
309.1 " "0
0
0
0 | 0
12: 5+-10
Ind.
0
0
0
0
0
0
232.136.0
134,613.0
0
0
0
0
6
0
0
0
fr
0
0
0
102.5
53.1
0
0
x 109Btu/hr
Utility
0
0
0
0
593,083.0
750,874.0
6.370.437.0
2,694,689.0
0
0 0
306,257.5
198,146.6
16,124.9
2,096.5
6
0
0
0
0
0
268.0
127.0
0
0
Class
Coo.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
27.2
8.7
0
0
13:
Ind
0
0
0
0
0
0
0
0
0
0
0
0
0
n
0
0
0
0
0
0
161.
83.
'0
0
lT-2 x 1010BtU/tr
Utility
0
0
0
0
• o
0
765,264.0
323.703.0
0
0
0 '•''••
0
22.804.9
2.964.5
0
0
0
0
0
- o
9 28.9
8 13.7
0
0
•e-
oo
(1) 1 indicates either model fuel sulfur content of zero or no installed capacity
in the category.
-------
49
1000
FIGURE 11. ESTIMATED ACTIF^L SO EMISSIONS BY SIZE CLASS
AND USE CATEGORY FOB LOW SULFUR STOKER COAL FUEL
-------
50
1000
• 67
Size Class
12
13
FIGURE 12. ANNUAL HOURLY FUEL USE (10 Btu/hr) BY SIZE AND
USE CATEGORY FOR HIGH SULFUR STOKER COAL
-------
51
1000
6 7
Size Class
12 13
FIGURE 13. ESTIMATED ACTUAL SO EMISSIONS BY SIZE CLASS AND
USE CATEGORY FOR Loft SULFUR PULVERIZED COAL FUEL
-------
52
1000
6 78
Size Class
10
12
13
FIGURE 14. ANNUAL HOURLY FUEL USE (1(T Btu/hr) BY SIZE AND USE
CATEGORY FOR HIGH SULFUR PULVERIZED COAL FUEL
-------
53
1000
I I
FIGURE 15. ESTIMATED ACTUAL SO EMISSIONS BY SIZE CLASS AND USE
CATEGORY FOR LOW SULFUR RESIDUAL OIL FUEL
-------
54
1000
100
m
O
in
g
O*
CO
I
o
In
U
1
3
"0
5' 6 7 ••
. Size Class
• 9
10
•12
13
,9'-.
FIGURE 16, ANNUAL HOURLY FUEL USE (10 Btu/hr) BY SIZE AND USE
CATEGORY FOR HIGH SULFUR RESIDUAL .OIL FUEL
-------
55
1000
10
o 100
10
c
o
O
(O
0
3
•o
ID
O
E
10
o>
o
•o
c
5 • 6 7 8
Size Class
10
12 13
FIGURE 17. ANNUAL HOURLY FUEL USE (10 Btu/hr) BY SIZE AND USE
CATEGORY FOR LOW SULFUR DISTILLATE OIL FUEL
-------
56
1000
£ioo
in
O
in
in
I
LJ
x
O
o
3
TJ
0)
O
E
10
o<
o
III 1 lilt I I I I I I
I 2 3 4 5 6 789 10
FIGURE 18. ANNUAL HOURLY FUEL USE (109 Btu/hr) BY SIZE AND USE
. CATEGORY FOR HIGH SULFUR DISTILLATE OIL FUEL
-------
57
1000
in
2 100
"o
in
g
u>
u>
3
*-
u
•o
0)
*-
o
E
o<
o
10
•O
c
-o
I 2 3 4 5 6 7 8 9 10 II 12 13
Size Class
FIGURE 19. ANNUAL HOURLY FUEL USE (10 Btu/hr) BY SIZE AND USE
CATEGORY FOR HIGH SULFUR NON-NATURAL GAS FUEL
-------
58
1000
Ul
c
° 100
10
o
-------
PART II
NON-UTILITY COMBUSTION SOURCE CONTROL ALTERNATIVES
-------
60
CONTROL ALTERNATIVES
The control alternatives considered in this study included
(1) Clean Fuels
Natural gas
Low sulfur oil
Low sulfur coal
(2) Processed Fuels
Physical cleaning of coal
Coal gasification
Coal liquefaction
(3) Combustion Modification
Fluidized bed, combustion of coal
(4) FGD Processes
Limestone slurry "
Lime scrubbing
Double alkali
MgO-central regeneration
MgO-integrated
Wellman-Lord
For each of these alternatives the objective was to assess the
Applicability to small industrial and commercial boilers in terms of the
underlying chemistry, operation and maintenance, secondary emissions, raw
material requirement, retrofitability, economics, and extent of development.
-------
61
CLEAN FUELS
Supply Projections
The available supplies of the various clean fuels were estimated
for the years 1975, 1980, and 1985. These clean fuel supply projections
include the natural clean fuels (natural gas, low sulfur coal, and low
sulfur fuel oil) plus desulfurized fuel oil. The quantities used for
transportation and for petrochemical feedstocks and other nonfuel uses
have been excluded. Thus, these are the supplies available to the residential
and commercial, industrial, and utility sectors.
Table 10 shows the clean fuel supply projections. In the upper
portion of the table values are given in the usual units and in the
lower portion they are given in terms of the fuel heating value. These
(9)
projections are based on some previous estimates made by Battelle. The
following explanation gives the basis of the original estimates and the
changes made in this report.
The gaseous fuel supply including natural gas, pipeline imports,
and LNG imports was used as provided by Dupree and West. The only
petroleum products considered applicable to industrial and commercial boilers
were distillate and residual fuel oils. Most distillate fuel oil contains
less than 1 percent sulfur by weight. The Minerals Yearbook 1971 indi-
cated that distillate fuel oil accounted for 17„5 percent of the total
consumption of petroleum products in 1971. This percentage was assumed to
hold constant through 1985, and thus the distillate fuel supply was estimated
using Dupree and West's projection of total petroleum supply.
(12)
According to a study by Hittman Associates, Inc., the U. S.
supply of low sulfur (< 1 weight percent S) residual fuel oil in 1970 was
1.07 x 10 bbl/day. This includes oil from both domestic and foreign
sources., A growth rate of 10 percent per year was estimated through 1980,
the growth rate then decreasing to 5 percent per year. The high initial
growth rate is attributed to the following:
-------
62
TABLE 10. CLEAN FUEL SUPPLY PROJECTIONS
Type of Fuel
Gas
Distillate fuel oil
Residual fuel oil
Coal
Gas
Distillate fuel oil
Residual fuel oil
Coal
TOTAL
Sulfur Content
(weight percent) Units
12
10 scf
<1.0 109 bbl
<1.0 109 bbl
6
<0 . 7 10 tons
10 15 Btu
<1.0 10 Btu
<1.0 1015 Btu
<0.7 1015 Btu
Supply Projection
1975
23.5
1.07
0.63
150
23.5
6.2
3.8
3.6
37.1
1980 -
24.8
1.28
1.01
170
24.8
7.5
6.1
4.2
42.6
for Year
1985
26.0
1.54
1.29
220
26.0
9.0
7.7
5.5
48.2
-------
63
• The fuel demand for the industrial and electrical
sectors will depend heavily on low sulfur resid until
coal conversion and flue gas desulfurization techno-
logies become commercialized;
• South American oil refineries have shown a willingness
to invest in and operate desulfurization plants.
Hittman projects a growth rate of 15 percent per year
for such facilities through 1980„
The supply projection for low sulfur coal was based on two studies
of the distribution of sulfur content of coal - a survey of coal availability
by Hoffman, et al.^ ' and a Bureau of Mines report on coal shipments.*- '
Data based on the latter report are shown in Table 11 . These studies
indicate that coal containing less than 1.0 percent sulfur constitutes about
33 percent of the total coal production on a heating value basis or about
39 percent on a weight basis. To adjust these percentages to the desired
break point of 0.7 percent sulfur, another fact from the Hoffman report was
used, namely, that the recoverable coal reserves containing less than 0.7
percent sulfur are about 67 percent of the reserves containing less than
1.0 percent sulfur. These percentages were used with Dupree and West's
projection of total coal supply to estimate the low sulfur coal supply.
Utilization and Applicability
One approach which could be used to reduce the sulfur oxide
emissions from an industrial or commercial boiler using high sulfur coal
or oil would be to switch to a clean fuel. Natural gas is assumed to be
unavailable for additional use in industrial and commercial boilers.
Residential uses have the first priority for this fuel, and as the supply
situation for natural gas has become very tight, industrial consumption has
been cut back. In most areas of the country, gas companies are not accepting
any new customers. Thus, only the following fuel switching possibilities
are of interest here:
(1) From high sulfur coal to low sulfur coal
(2) From high sulfur coal to low sulfur oil
(3) From high sulfur oil to low sulfur oil.
-------
TABLE 11. SULFUR DISTRIBUTION IN BITUMINOUS AND LIGNITE COAL - BUREAU OF MINES, 1970
Based on Shipments of Coal
Region
Appalachia
Midwest
Near West
Far West
Entire U.S.
Shipments Accounted
for (106 tons)
410.6
143.7
9.5
34.2
598.0
Percent of Coal with Sulfur Content (weight percent)
<0.5
2.3
—
1.8
20.4
2.8
0.5 - 1.0
45.3
4.0
10.4
70.0
36.2
1.0 - 1.3
6.4
1.0
—
8.7
5.1
1.3 - 1.8
11.8
3.0
19.1
0.5
9.2
>1.8
34.2
92.0
68.7
0.4
46.7
-------
65
Note that a switch from high sulfur oil to low sulfur coal is not included
because the alterations required to change a boiler system from a liquid
fuel to a solid fuel would be so extensive as to render this change
impractical.
Costs
The costs associated with fuel switching are of three types:
i
e Operating costs due to the difference in price
between the two fuels involved
o Boiler modification costs, which include an invest-
ment and the associated investment-related annual
costs
o Other effects on the boiler operating cost (excluding
fuel) due to a change in the state of the fuel.
These costs can vary widely depending on the particular features of the
boiler involved, the sources of the fuels involved, and the nature of the
fuel purchase agreements. Some generalizations will be made in the
following discussion.
Table 12 presents some average cost data for fuels sold to
utilities at two time periods, mid-1973 and the end of 1974. Note that
in mid-1973 the difference in cost between high sulfur and low sulfur
coal was about $2.30/ton (9
-------
TABLE 12. COST OF FUELS
Type of Fuel Weight Percent S
Natural gas
Distillate fuel oil
Residual fuel oil <1.0
Residual fuel oil >1.0
Coal <0.7
Coal >0.7
Mid-1973
Fuel Units
43C/103 scf
$5.70/bbl
$4.20/bbl
$3.90/bbl
$12.10/ton
$9. 80 /ton
r, (15)
Cost
$/10° Btu
0.43
0.98
0.70 c
0.65
0.50
0.41
End- 19 74
Fuel Units
61C/103 scf
$12.00/bbl
$12.30/bbl
$10.40/bbl
$28.50/ton
$22.80/ton
„ (16)
Cost '
$/10° Btu
0.61
2.06
2.05
1.73
1.18
0.95
-------
67
In considering boiler modification costs, note that the first
and third fuel switching possibilities listed above involve no modification
cost because no change in the state of the fuel is involved. Only the
switch from high sulfur coal to low sulfur oil requires boiler modifications.
Although no direct, general estimates of this investment are available,
an approximation can be made based on some data given by Schreiber, et
al/17) These data are for the investment required to convert boilers
from coal to liquid Solvent Refined Coal (SRC). This is a change from a
solid to a liquid fuel, although the SRC requires a considerable amount
of heating to keep it in a liquid state, thus increasing the investment
substantially. Figure 21 shows the investment for the conversion from
coal to liquid SRC as a function of boiler capacity. This line represents
the consensus of the data given by Schreiber after adjustment to a general
U.S. location (using location factors given by Schreiber) and to mid-1973
(using the Chemical Engineering Plant Cost Index). The materials/labor
ratio for this investment is about 37/63. The investment for converting
from coal to fuel oil will be less than that shown in Figure 21 , and this
difference will vary widely depending on the properties of the fuel oil.
Heavy residual oils require some heating to maintain flow (although less
than SRC) and the investment for these fuels should be not too much
lower than the values shown in Figure 21 (probably about 10 percent lower).
For distillate fuel oils, which require no heating, the investment would
be much lower than the values shown in Figure 21 (probably at least 50
percent lower).
Associated with the investment for boiler modifications is, of
course, an investment-related annual cost. If one assumes that there is
no maintenance associated with this investment, the annual cost will
include only the local taxes and insurance, depreciation, return on rate
base, and Federal income tax. Fot th6 Utility Financing Method being used
here (.see Appendix A ), these costs amount to annual cost of 14,6 percent
of the investment.
There are other effects on the.boiler operating cost related to
the change of state of the fuel. A survey published by Olmsted' ' showed
the following average operating costs excluding fuel for power plants in
1973.
-------
CO
•o
10
10
10
68
10"
10
I
10
I,
10
8
Btu/hour
Megawatts
10
100
10
10
1000
FIGURE 21. INVESTMENT FOR CONVERTING BOILERS FROM COAL TO LIQUID SRC
-------
69
Operating Cost Excluding Fuel
Fuel Type (mills/kWh) ..
Gas 0.56
Oil 0.95
Coal 0.75
Note that the operating cost excluding fuel is 0-20 mills/kWh higher for an
oil-fired boiler than for a coal-fired boiler. This increase in operating
cost must be included when estimating the cost of switching from coal to oil.
-------
70
PHYSICAL CLEANING OF COAL
The sulfur oxide emission standards for industrial boilers may be
met in a variety of ways. The use of low sulfur, coal is, of course, a
distinct alternative. Assuming an allowable emission factor of 1.2 Ib S0?
g ^
p^r 10 Btu, the allowable sulfur content in a coal of heating value of
12,000 Btu/lb will be about 0.7 percent.
In a number of coal-producing regions in the continental United
States, some coals are amenable to reduction of their sulfur contents by
the so-called "physical" techniques. These techniques utilize differences
in physical properties of the coal and the refuse (including minerals
of sulfur) such as specific gravity.
Sulfur in Coal
Sulfur is present in coal in several forms. In the organic form
it is chemically bonded to the carbon atoms and as such cannot be removed
by physical means. In a number of Eastern United States coals, it represents
(19)
approximately 20 to 85 percent of the total sulfur present. The inorganic
form is present mainly as the chemical species pyrite or marcasite (FeS«) with
relatively small amounts of the sulfates of calcium and iron, usually in the
range of 0.07 to 0.2 percent. The total sulfur content of coal varies from
less than 1.0 to more than 9.0 percent. Also, there is a large variability
in the percentage of physically removable pyritic sulfur, not only in coals
from different regions, but also in coals obtained from the same mine. The
average organic sulfur content of a number of Eastern U.S. coals has been
reported at 51.2 percent^19' of the total sulfur.
Cleanability of U.S. Coals
The coal resources of the United States are found in three belts
or regions, namely, the Appalachians, the Eastern Interior, and the Western
(Rocky Mountain). It has been found that the bituminous coals from the
Northern Appalachian Region show a greater propensity for sulfur reduction
by physical means. The 0 to 0.7 percent sulfur category with total reserves
-------
71
(as of 1968) of 43.8 million tons may "grow" to 9,930 million tons by
crushing higher sulfur coals to a 3/8-inch top size and washing such that
a 90 percent yield is obtained. In the remaining coal-producing regions,
physical cleaning does not materially change the sulfur content. The coals
in the Western Interior are "characteristically high in (organic) sulfur"
while the coals in the Northern and Southern Rocky Mountain regions are
inherently of low sulfur content.
A recent analysis of the sulfur contents of 200 samples from the
Northern Appalachian region showed an average total sulfur content of 3.0
percent. Crushing to a top size of 3/8 inch followed by washing reduced
(21) (22)
this average to 1«,6 percent. A more recent analysis has shown that
in the Northern Appalachian region (Maryland, Ohio, Pennsylvania, and Northern
West Virginia) less than 5 percent of the samples would comply with the EPA
standard of 1.2 lb/10 Btu. Only 35 percent of the samples would comply after
crushing to 14 mesh top size and washing to a point where 50 percent of the
Btu value is recovered. The corresponding figures for Midwest Region coal
(Illinois, Indiana, and Western Kentucky) are 1 percent and 5 percent,
respectively.
Technology of Coal Cleaning
The principal coal preparation processes used today are oriented
toward product standardization and ash and sulfur (pyritic) reduction,,
In a preparation plant the raw coal is typically subjected to
(1) size reduction and screening, (2) separation of coal from ash and
pyrite in a device utilizing differences in specific gravity or surface
properties between them, and (3) dewatering the product coal and refuse.
Size reduction is accomplished in rotary or roll crushers. The
extent of preliminary and subsequent (farther along in processing) size
reductions depends on the type of coal processed and the desired product
characteristics, mainly ash and pyrite and Btu content. It is a well
known fact that more of the impurities are liberated as size of the coal
is reduced. An economic optimum is usually sought as the costs of pre-
paration rise exponentially with the percentage of "fines" to be treated.
-------
72
Screening (either wet or dry) is practiced to separate the various size
fractions resulting front the crushing operations. Separations based on
differences in specific gravity are usually reserved for fractions Mesh
No. 35 or over.
Separation of pyrite and ash from the raw coal may be accomplished
in a variety of devices or processes. Jigging currently handles the largest
percentage of coal cleaned by physical means,
Dewatering operations are carried out in equipment such as fixed
and vibrating screens, thickeners, centrifuges, vacuum fillers, and thermal
dryers.
Environmental Impacts and Control
Air Pollutants. Particulate matter originating from the operation
of thermal dryers, and coal crushers is the most serious pollutant from coal
preparation plants. With increased percentage of fine coal (Mesh No. 35 and
under) being processed, the uncontrolled emissions of particulate matter per ton
of total product have increased because the finer sizes are usually thermally
dried. Uncontrolled emissions of particulate matter may be in the range of
15 to 25 pounds per ton of thermally dried coal, depending upon the type of
(23)
dryer used. The thermally dried coal may constitute two-thirds the total
plant output. The uncontrolled emissions of particulate matter may thus be
(24)
as high as 10 to 17 pounds per ton of clean coal. Federal regulations
limit the emissions from thermal dryers to 0.031 grains per scf. For a
typical dryer, the gaseous discharge is of the order of 37,000 scf per ton
(24)
of coal dried. This translates into an allowable emission level of 0.16
pound of particulate matter per ton of coal dried.
Other pollutants of concern are SO , NO , hydrocarbons, and carbon
X X
monoxide resulting from the combustion of product coal to effect drying by
direct contact of wet coal and the hot gas. These emissions are variable
and are not regulated. The variability results mainly from the variability
in sulfur content of product coal and efficiency of the combustion device
used. If a heat requirement of 230 Btu per pound coal dried is assumed,
-------
73
then the emissions per ton product may be calculated using published
(24)
emission factors on industrial boilers burning bituminous coal. These
emission factors in pounds per ton of coal burned are: 1 for carbon monoxide,
15 for nitrogen oxides (as N07), and 1 for hydrocarbons. Assuming a sulfur
content of 1.5 percent, 12,000 Btu per pound for product coal, and 67 percent
of plant output thermally dried, then the emissions per ton of coal cleaned
may be of the order of 0.8 Ib S00, 0.2 Ib NO (as NO ), 0.013 Ib carbon
t~ X 2.
monoxide, and 0.013 Ib hydrocarbons.
Water PoHution. Large amounts of water circulate in a coal
washing plant. Although modern coal cleaning plants are designed to operate
(19)
on a closed circuit, make-up is needed for water losses in the refuse,
the clean coal, and. in the thermal dryer. This may vary between 15 and 40
gallons per ton of clean coal, depending on age and type of plant. The
characteristics of water circulating in a coal cleaning plant are similar
to those of acid mine drainage water.
Solid Waste. Solid wastes from a coal preparation plant are mainly
the pyrite and ash-forming refuse together with the coal value lost in
preparation. These may amount to 300 Ib per ton of clean coal and an
associated surface moisture of 70 Ib per ton. These quantities vary with
such factors as ash content of coal, product specifications, and type of
washing operation employed. Appropriate methods of disposal are discussed
in Reference 19.
Applicability to NUC Sources
The physically cleaned coal is a solid fuel and thus the application
to industrial boilers is limited to the boilers that burn coal. Although the
cleanability of U. S. coals is somehow limited due to the physical and chem-
ical characteristics of coal, if coal is accessible to cleaning, this alter-
native is very favorable economically to coal-fired industrial boilers which
in general are small in size.
-------
74
Capital and Annualized Costs
The capital costs presented in Figure 22 are mid-1974 for a
modern plant including appropriate emission and effluent control systems
and refuse handling systems. The breakdowns of total capital require-
ment and annual cost are shown in Table 13. The estimations were based
on the Utility Financing Method as modified by the Panhandle Eastern Pipe-
(27)
line Company (see Appendix A) unless specified in the table and the
footnotes. The product coal cost was estimated at $13.58/ton and $13.25/
ton for a 500 tph and 1500 tph capacity plant, respectively.
-------
75
100
10
co
O
a
«
u
100
10,000
Capacity, tph (input)
FIGURE 22. CAPITAL COSTS OF COAL WASHING PLANTS MID- 1974
-------
76
TABLE 13. COSTS FOR PHYSICAL COAL CLEANING
Mid-1973, 4000 hours/yr Operation
Item
Coal Input Rate (tons/hr)
Clean Coal Product Rate (tons/hr)
Capital Requirement (1000$)
(a)
Total bare costv '
Engineering and design
Contractor fees
Subtotal Plant Investment
Project contingency
Total Plant Investment
Interest during construction
Startup
Working capital
Total Capital Requirement
Annual Costs (1000$/yr)
Direct operating labor , ^
Maintenance (5 percent/yr)
Supervision
Admin, and general overhead
Local taxes and insurance
Electricity (IC/kwhrK '
Lime (1.4c/lb)wr
Magnetite (2
-------
77
Footnotes to Table 13
(a) 1973 Total Bare Cost ($) • 60,630 (tons/hr input)0'74.
(b) Operating labor = 6 men/shift for 500 tons/hr, 10 men/shift for
1500 tons/hr.
(c) Maintenance materials/labor ratio = 56/44.
(d) Power consumption = 4000 H.P. for 500 tons/hr.
(e) Materials consumptions:
Lime 2.6 Ib/T dry solids
Magnetite 0.5 Ib/T coal treated by dense medium
Frothing agent 0.17 Ib/T coal treated by froth flotation
Flocculant 3.6 Ib/T dry solids
Water 17 gal/T clean coal
From material balance flow sheet:
T clean coal/T coal input = 0.858
T dry solids/T clean coal = 0.165
T coal by dense medium/T clean coal = 0.783
T coal by froth flotation/T clean coal = 0.217
T coal dried/T clean coal = 0.662
Heat required for dryer = 230 Btu/lb co'al dried.
(f) Based on clean coal heating value of 12,500 Btu/lb.
-------
78
COAL GASIFICATION
Gasification of coal is intended to produce gaseous products for
consumption as fuels and for industrial uses such as chemical synthesis and
reducing gas in iron and steel production. The technology has been prac-
ticed in a commercial scale primarily in Europe for many years to manufacture
town gas, fuel gas, and synthesis gas. Gasification of coal involves the
reaction of coal with air, oxygen, steam, CO^, or mixtures of these gases.
A low-Btu gas is obtained if an air-steam mixture is used directly to gasify
the coal and contains nitrogen as a major component. Intermediate-Btu gas,
which contains a minor amount of nitrogen, is obtained when an oxygen-steam
mixture is used. High-Btu gas, which is similar to natural gas and contains
over 90 percent methane, is obtained by further processing of intermediate-Btu
gas. Low-Btu gas is suitable for use as an energy source near its point of
generation, but it is not economically favorable for long distance transpor-
tation. Intermediate-Btu gas can be used either as an energy source or as a
synthesis gas for the production of chemicals. Fpr analyses in this study,
the low- and intermediate-Btu gases were categorized as low-Btu gas to
distinguish them from methanated high-Btu gas.
Gasification Processes
Low-Btu Gas. In general, the process consists of coal preparation,
gasification, gas cleaning, desulfurization, and compression. Mined coal is
crushed, screened, and then conveyed to the storage bunkers atop coal lock
hoppers. As the coal is fed to gasifiers, it reacts with externally supplied
oxygen and steam, and as the result hydrogen, carbon monoxide, carbon dioxide,
and methane are produced. The steam is the source of hydrogen and the heat
resulting from combustion of coal supplies the heat required for gasification.
In addition to coal gasification, coal devolatilization or carbonization
takes place in the reactor. Gaseous products of devolatilization are rich
in methane and hydrogen and contain tars and oils.
The crude gas is quenched and scrubbed by a wash cooler and then
desulfurized before utilization.
-------
79
Various gasification processes can be used to produce low-Btu fuel
gas. Presently, four gasification systems can be considered commercially
viable in the sense that their technologies are commercially proven and
(28)
their systems are available through commercial dealers. These four
systems include
(1) Lurgi
(2) Koppers-Totzek
(3) Winkler
(4) Wellman-Galusha.
None of these processes are in widespread use in the.United States.at.the
present time mainly due to the past abundance of inexpensive natural gas
and fuel oils which today are becoming increasingly scarce.
Hieh-Btu Gas. Basically, a high-Btu gasification process is
identical to the low-Btu gasification process except for shift reaction
and methanation processes. The scrubbed crude gas is introduced to the
shift reactor. About one-half of the total gas is subjected to shift
conversion. The resulting hot gas is then cooled to facilitate subsequent
purification by an acid gas removal process. The product gas from acid
gas removal process is fed to a fixed-bed methanation reactor. The metha-
nated gas is compressed and dehydrated for pipeline gas.
Various coal gasification processes have been developed for
manufacturing high-Btu synthetic natural gas (SNG). Currently, none of
these have been constructed for commercial operation. The processes in
an advanced stage of development include Lurgi, Synthane, Hygas, and
C0_-Acceptor. The Lurgi process is commercially proven for high-Btu SNG.
The full-scale evaluation of the methanation process in conjunction with
the Lurgi process at Westfield, Scotland, indicates that the methanation
is now commercially feasible.
-------
80
Fuel Gas Desulfurization and Sulfur Recovery
Low-Btu Gas. The hot raw gas from coal gasification contains
many impurities. The primary impurity of environmental concern is sulfur
in the form of hydrogen sulfide (H^S) with very small amounts of carbonyl
sulfide (COS) and carbon disulfide (CS«). In general, the coal gasification
process has an advantage over direct combustion of coal in the control of
sulfurous compounds since the reduced form of hydrogen sulfide is more
easily removed than the oxidized form of sulfur dioxide in flue gas. In
addition, removal processes are commercially available for hydrogen sulfide
while sulfur dioxide diluted in large volumes of flue gas presently cannot
be effectively controlled by commercially-proven processes.
The concentration of hydrogen sulfide in fuel gas depends on the
sulfur content in the coal, the heating value of gas, the heating value of
coal, and gasification efficiency. Assuming that a coal with a heating
value of 12,000 Btu/lb on a moisture- and ash-free basis is gasified with
a gasification efficiency of 70 percent, and that at least 90 percent of
the sulfur in the coal is converted to hydrogen sulfide, the concentration
of H?S in fuel gases with different heating values can be plotted against
sulfur content in coal as shown in Figure 23. Figure 24 shows the emission
of S02 in pounds per 10 Btu fuel gas heat input as a function of H_S con-
centration and fuel gas heating value. It can be seen from Figure 24 that
a concentration of about 700 ppm and 2100 ppm would be the allowable concen-
tration of H_S for the fuel gases with heating values of 100 Btu/scf and
300 Btu/scf, respectively, to meet the Federal standard for combustion
sources of 1.2 Ib SO /10 Btu heat input.
Control of sulfurous emissions can be considered under three broad
(29)
steps :
(1) Desulfurization of fuel gas
(2) Sulfur recovery
(3) Tail gas treatment.
Desulfurization of fuel gas, is usually accomplished by absorption
(29)
into a liquid phase using suitable gas-liquid contacting equipment.
Absorption processes can be divided into three broad categories depending
-------
81
a
a
ex
CO
o
CN
a
15
10
Coal heating value » 12,000 Btu/lb
Gasification efficiency =0.7
2 3
Sulfur in Coal, percent
FIGURE 23. H S IN FUEL GAS AS A FUNCTION OF SULFUR IN COAL
(Parameter: Gas heating value, Btu/scf)
-------
82
15
a
•u
o
w
oo
to
I
O
en
10
Coal heating value = 12,000
Btu/lb
Gasification efficiency !
0.7
New Source Performance standard =1.2 lb/10 Btu
5000 10,000
H0S in Fuel Gas, ppm
15,000
FIGURE 24. S02 EMISSION VERSUS H2S CONCENTRATION IN FUEL GAS
(Parameter: Gas heating value, Btu/scf)
-------
83
upon the type of absorbent used such as physical solvent process, amine
process, and alkaline salt solution process. There are direct conversion
processes, such as the Stretford and Giammarco-Vetrocoke processes, which
are capable of absorption and oxidation of H S to produce sulfur directly.
Commercially important processes which recover elemental sulfur
from sulfur-bearing gases such as fuel gas and regenerated acid gas include
(29)
the Glaus process, the Stretford process, and the Giammarco-Vetrocoke.
All three processes are designed normally to convert hydrogen sulfide in
the feed gas to sulfur.
The tail gas from the Glaus plant contains unconverted H_S and
S09 and lesser quantities of other sulfur constituents, such as COS, CS~,
and elemental sulfur vapor and particles. The tail gas from typical Glaus
plant operations contains about 1 to 2 percent total sulfur. The normal
practice in the past has been to discharge the tail gas directly to atmos-
phere after passing it through an incinerator to convert sulfur compounds
into sulfur dioxide. During the past several years, a number of treatment
processes have appeared for removing the residual sulfur compounds from the
tail gas. Tail gas treatment processes which have received commercial
(29)
acceptance are Beavon, Cleanair, IFP-1, SCOT, and Wellman-Lord.
High-Btu Gas. Like a low-Btu coal gasification process, control
of sulfurous emissions can be considered under the three steps described
above except for the desulfurization of fuel gas. The desulfurization
process for high-Btu gasification not only removes sulfurous components
but also carbon dioxide in the fuel gas. Removal of carbon dioxide is
necessary to avoid undesired corrosion in the pipelines.
Applicability to NUC Sources
Low-Btu Gas. The applicability of low-Btu gas to existing coal-,
oil-, or gas-fired boilers will technically depend on boiler configuration
and operation. For a given heat input rate, the volumetric flow rate of
low-Btu gas is high compared with that of high-Btu gas such as natural gas
and so is the combustion flue gas flow rate. These conditions would create
the following problems:
-------
84
(1) The size of piping or ducts to transport the fuel
will increase as the heating value decreases.
(2) The burners to handle very low-Btu gas will be
different from the existing burners. For oil or
high-Btu gas, a small portion of the throat area is
required for fuel and most of the burner throat area
is for the combustion air. For a low-Btu gas with
high volumetric flow rate, it is necessary to
increase the burner throat size or the number of
burners.
(3) The heat absorption pattern may be different. In
general, heat is transferred from the products of
combustion to the boiler by radiation and convec-
tion. When a low-Btu gas is applied, the rate of
heat transfer depends on the temperature and the
mass flow of the flue gas over the heating surface.
Any change in flue gas temperature or flue gas
quantity in a boiler affects the heat transfer balance.
According to the result of the industrial and commercial boiler
data analysis conducted in this study, industrial boiler subgroups of
environmental concern include small and large industrial boilers burning
high-sulfur coal and resid. Therefore, the application pattern of low-
Btu gas to existing boilers may be considered only for the coal- and
oil-fired boilers; however, it can also be conceived that the application
may be necessary to natural gas-fired boilers due to the shortage of the
supply.
The application of low-Btu gas to a coal-fired boiler would be
more adaptable than to other type boilers. This is mainly because a coal-
fired boiler has a large combustion chamber. Although existing stoker coal
boilers were designed for maximum utilization of radiation, the conversion
to low-Btu gas boiler would not hamper the heat absorption rate because of
the increased convective heat transfer. Reduced heat absorption, however,
may result for a low-Btu gas with a heating value less than 200 Btu/scf.
The pressure loss due to the increased quantity of flue gas may exceed the
design condition for gas with a heating value of less than 200 Btu/scf.
-------
85
The application of low-Btu gas to oil- and gas-fired boilers may
be difficult for a gas with a heating value less than 300 Btu/scf. This
is mainly because the firing chamber is small and, thus, may not be able
to handle a larger volume of combustion gas. Less heat would be absorbed
in the furnace and more heat in the superheater and reheater than the unit
was originally designed for because of the low temperature of the combustion
gas and the increase in loss of sensible heat. The pressure loss would be
very high when burning low-Btu gas since the loss is proportional to the
square of the flue gas quantity. Some extensive structural alterations may
be required to minimize the pressure loss.
The economic applicability of low-Btu gas to existing boilers will
depend on load factor, heat loss, and retrofit difficulty. The load factor
of industrial boilers falls between 0.35 and 0.50 and, thus, will play an
important role in evaluating the economic feasibility. The heat losses from
a low-Btu gas boiler will be higher than those for existing boilers due to
the increased quantity of flue gas with considerable sensible heat. The
retrofit difficulty may be significant for existing boilers with little
extra space around the boiler area.
High-Btu Gas. The application of high-Btu SNG to existing coal-,
oil-, and gas-fired boilers will be similar to that of natural gas to the
boiler systems. The application to natural gas boilers should not present
any problem. The application to oil-fired boilers may need a minor rebal-
ancing of heat absorption in the furnace. The application to coal-fired
boilers may need a moderate adjustment of the heat absorption system,
particularly for the stoker boilers, because the system was originally
designed to maximize the radiation heat transfer due to high firing temper-
ature. In general, the conversion of an oil- or coal-fired boiler to a
high-Btu gas boiler is not difficult both technically and economically,
and, thus, the applicability of this alternative is deemed high.
-------
86
Model Plant Calculation
Low-Btu Gas. In determining the emission factors and capital and
annualized operating costs of low-Btu gasification, a conceptual design
study for a plant producing low-Btu gas for five 250,000 Ib steam/hr indus-
trial boilers was carried out in this study. The results are listed in
Table 14. The sulfur dioxide emissions from the combustion of the"fuel gas
was estimated at 0.5 lb/10 Btu based on the assumption that the fuel
gas is desulfurized by a MDEA absorption process with a removal efficiency
of 93.5 percent. The sulfur dioxide emission from a combined system of a
Claus unit and a Beavon tail gas treatment system was assumed to be about
0.02 lb/106 Btu.
High-Btu Gas. To determine the manufacturing cost of high-Btu
SNG, a conceptual design study was carried out for a typical SNG gasifica-
tion process. The Hygas process was selected mainly because the process
can use both caking and noncaking coals. The Hygas process under consider-
ation in this study is based on the design by IGT to produce 265 x 10
scf/day of pipeline gas using a Pittsburgh seam coal. Since no updated
economic data are available in the open literature, the data presented in
this study (see Table 15) were based largely on the IGT design. The data,
however, were corrected with respect to coal flow rate, sulfur content, and
base year. A scaling factor of 0.8 was assumed for sulfur content of coal.
The total capital requirement was estimated at $401.8 million and net annual
operating cost at $82.8 million. The average gas cost over the life time
of the plant of 20 years was estimated to be $1.60/10 Btu.
-------
87
TABLE 14. ESTIMATED COST OF THE KOPPERS-TOTZEK LOW-BTU GAS
Gasification Process: Koppers-
Totzek Coal Feed Rate: 2,107 tons/day
Plant Size: 118 x 106 scf/day Type of Coal: Eastern Bituminous
Plant Load Factor: 0.9 Sulfur Content of Coal: 3%
Gasification Efficiency: 0.7 Ash Content of Coal: 14%
Gas Heating Value: 300 Btu/scf Coal Heating Value: 12,000 Btu/lb
Item Cost, $10
Capital Requirement
Bare cost3 19.7
Engineering and design 1.0
Contractor's overhead and profit 2.0
Subtotal Plant Investment . 22.7
Project Contingency 3.4
Total Plant. Investment 26.1
!
Interest during construction 4.4
Startup cost 1.9
Working capital 1.9
Total Capital Requirement 34.3
Annual Operating Cost
Labor (b) 0.79
Administrative and general overhead 0.49
Materials and utilities (c) 0.87
Fuel Cost(d) 6.92
Local taxes and insurance 0.70
Gross Operating Cost 9.75
Credit (e) 0.19
Net Operating Cost 9.56
-------
88
TABLE 14. (Continued)
Item Cost, $10
Average Gas Cost
Return on rate base 1.90
Federal income tax 0.63
Depreciation .1.62
Net operating cost 9.56
Average Annual Cost 13.71
Average Gas Cost, $/10 Btu 1.18
(a) The value was obtained from Reference (31). It was corrected with
respect to base year by using the CE plant cost index. This cost
includes gasification, acid gas removal and sulfur recovery, oxygen
plant, and pollution control equipment. This does not include
utilities, off-site facilities, and land.
(b) The value was assumed.
(c) This includes maintenance and operating supplies and direct material
and utility cost excluding cost for coal.
(d) The cost of coal was assumed at $10/ton.
(e) This includes credits for by-product sulfur at $10/long ton and
for reduced operating cost of boiler system.
-------
89
TABLE 15. ESTIMATED COST OF THE HYGAS SNG
Gasification Process: Hygas
Gasification Plant Size:
260 x 106 scf/day
Plant Load Factor: 90 percent
Gas Heating Value: 963 Btu/scf
Gasification Efficiency: 60 percent
Coal: Eastern coal
Coal Feed Rate: 17,517 tons/day
Sulfur Content of Coal: 3 percent
Ash Content of Coal: 14 percent
Heating Value of Coal:
12,000 Btu/lb
Item
Cost, $10
Capital Requirement
(a)
Total bare cost
Engineering and design cost
Contractor's overhead and profit
Subtotal Plant Investment
Project contingency
Total Plant Investment'
Interest during construction
Startup cost
Working capital
Total Capital Requirement
248.6
Included in
total bare cost
24.9
273.5
41.0
314.5
53.1
17.1
17.1
401.8
Annual Operating Cost
Labor(b)
Administrative and general overhead
Materials and utilities(c)
Fuel cost(d)
Local taxes and insurance
Gross Operating Cost
Credits(e)
Net Operating Cost
7.9
4.7
7.0
57.5
8.5
85.6
2.8
82.8
-------
90
TABLE 15. (Continued)
Item Cost, $10
Average Gas Cost
Return on rate base
Federal income tax
Depreciation
Net operating cost
Average Annual Cost 131.3
Average Gas Cost, $/10 Btu 1.60
(a) This includes costs for coal preparation, gasification, shift conver-
sion, acid gas removal and sulfur recovery, oxygen plant, methanation,
pollution control, utilities and off sites, and land. The value was
obtained from References (32) and (33). It was corrected with respect
to coal flow rate and base year by using a scaling factor of 0.9 and
the CE plant cost index, respectively, where needed.
(b) This includes direct operating labor, maintenance labor, and super-
vision. A direct operating labor of 52 men/shift was used as suggested
in References (32) and (33).
(c) This was estimated from References (32) and (34). It was corrected
with respect to coal consumption rate and base year. This does not
include the cost for coal.
(d) The cost of coal was assumed at $10/ton.
(e) This includes the following credits:
Elemental sulfur at $10/long ton $1.5 x 10-
Ammonia at $25/ton $1.1 x 10^
Phenol at $0.02/lb $0.2 x 106
$2.8 x 106
-------
91
COAL LIQUEFACTION
The increased demand for clean fossil fuels has stimulated develop-
ment of methods for converting the nation'.s abundant coal resources into a
low-sulfur, low-ash fuel. One technique, coal liquefaction, utilizes
solvents, heat, and high pressure to "liquefy" the coal to produce an ashless,
low-mineral, low-sulfur, high-Btu fuel.
Process Description
The current U.S. coal liquefaction or extraction processes may be
classified either as a solvent refining or hydrogenation process. The
solvent refined.coal programs of the Office of Coal Research and Southern
Research Institute, and the dissolved coal .variation of the Consolidation
Coal Company synthetic fuel process are in the first group; the H-coal,
USBM, Gulf, and other catalytic processes fall into the second group.
The H-Coal process involves simultaneous catalytic hydrogenation
and dissolution of the coal in a specially designed ebullated reactor.
The reactor product slurry is transferred to a flash drum to separate the
lighter hydrocarbons from the slurry. The slurry is then passed through
hydroclones to separate the recycle solvent. The underflow stream is
filtered to remove the minerals and undissolved carbonaceous matter,
leaving a liquid stream which may be distilled to separate the naphtha
from fuel oil. Hydrocarbon Research states that the ashless liquid
product contains about 0.2 percent sulfur and has a heating value of about
/35\
18,000 Btu/lb.v ' About 18,600 scf of hydrogen is consumed per ton of
coal in processing Illinois No. 6 coal with 5.0 percent sulfur and 9.9
percent ash content. About 2.7 barrels of synthetic crude distillate are
produced per ton of coal processed.
The Synthoil process features a packed-bed reactor operating at
840°F and 2000 to 4000 psig in which coal dissolution and catalytic hydro-
genation occur simultaneously. The effluent gases are separated from the
extract in high-pressure receivers. After pressure let-down, the extract
oil is either centrifuged or filtered to remove mineral and undissolved
organic matter. The product oil is of reasonably low viscosity and flows
-------
92
freely at room temperature. Hydrogen consumption is maintained relatively
low, about 9000 scf/ton of coal, as only enough is added to remove the sulfur.
Gas production is also minimized. The major difficulty with the process is
the high pressure drop with attendant high pumping requirements. Experimental
results indicate that the Synthoil Process is capable of producing three bar-
rels of synthetic oil with a 0.19 percent sulfur content from one ton of 4.6
/O f\
percent sulfur coal. The liquid has a heating value of 15,000 Btu/lb.
The Gulf process utilizes a fixed-bed reactor specifically designed
to minimize catalyst plugging to liquefy and catalytically hydrogenate the
coal. The reactor product passes to a gas-liquid separator where hydrogen is
recovered for recycle. The liquid product goes to solids separation, normally
hydroclones, where the slurry overflow is recycled and the high solids under-
flow is sent to a solids removal process such as filtration of vacuum distil-
lation. Gulf Research states that their catalytic process will generate
about 3-7 barrels of low-sulfur (0.05 to 0.2 percent) synthetic oil from one
ton of bituminous coal. The fuel oil has a heating value of 18,000 Btu/lb.
Solvent refining was initiated with the objective of producing a
low-cost antipollution alternative to residual oil and natural gas. This
process can produce either an ashless, low-sulfur solid product or a liquid
fuel both with a heating value of about 15,900 Btu/lb. It is in the most
advanced state of development of all of the coal liquefaction processes and
was selected as the model process. The process involves adding hydrogen to
the coal-solvent slurry and depolymerizing the coal in the reactor vessel.
The sulfur is removed as hydrogen sulfide in the pressure let-down vessel,
and the liquefied slurry is filtered, distilled, and solidified to produce
the ashless solid product. A 2 tons of coal/hour pilot plant was started in
mid-1974 at Fort Lewis, Washington. To date continuous integrated operation
has not been achieved. In January, 1974, a 6 tons of coal/day pilot plant
was started in Wilsonville, Alabama. It has operated intermittently at less
than rated capacity.
The Consol process was designed to transform high-sulfur Eastern
bituminous coal into a low-sulfur synthetic crude oil, or a fuel oil suit-
able for use in utility plants. The process involves slurrying the coal
with the recycled solvent and heating to. 750 F at about 400 psia; no
hydrogen is added to the reactor, it is strictly a solubilization step.
-------
93
The resulting slurry is passed through hydroclones to concentrate the solids.
The overflow extract is hydrogenated over a Co-Mo catalyst and the solids
are gasified to generate the required hydrogen. Consolidation Coal Company
states that the Consol process is capable of producing 1.5 barrels of 0.2
(38)
percent sulfur fuel oil and 0.5 barrel of naphtha from one ton of coal feed.
Environmental Problems
Environmental problems associated with coal liquefaction may
involve significant health problems. It is well known that sufficient
exposure to a variety of chemicals can cause cancer in man. Since 1900,
it has been recognized that workers handling coal tars, certain aromatic
amines, and some heavy metal compounds have increased incidence of
carcinoma of the skin, bladder, and lungs, respectively. Likewise, other
coal-derived products such as benzo(a)pyrene, dibenz(a)anthracite, 7, 12-
dimethyl-benz(a)-anthracite and 3-methylchol-anthrene are known to be
strong carcinogens. Therefore, prompt attention to conversion of waste
to environmentally acceptable materials, hopefully at an economic advantage,
is very important.
Applicability to NUC Sources
The applicability of liquefaction products to existing coal-, oil-,
and gas-fired small industrial and commercial boilers will technically
depend on boiler configuration and operation. Liquefaction products can
be classified in terms of solid fuel such as solid SRC and liquid fuels
such as liquid SRC, H-coal product, and other process products. According
to the studies conducted by the Bureau of Mines, Combustion Engineering,
and Babcock and Wilcox as reported by Schreiber, et al., ' the solid SRC
appeared similar to a high volatile bituminous coal except for the reduced
sulfur and ash content. The grindability index, however, is high (about 16
percent) as compared with that of nonprocessed coal (about 60 for the
Kentucky No. 11 coal). The liquid SRC was similar to No. 6 fuel oil in
handling and combustion characteristics although the preheating requirement
was greater. The liquid fuels obtained from other processes were quoted to
be similar to a crude oil, fuel oil, or naphtha.
-------
94
According to the results of the industrial and commercial boiler
data analysis conducted in this study, boiler subgroups of environmental
concern are boilers burning high sulfur coal and fuel oil. Therefore, the
application pattern of liquefied fuels to existing small boilers may be con-
sidered only for the coal- and oil-fired boilers. In general, conversions
from oil-fired combustion equipment to coal fired are associated with an
exorbitant cost. Only conversions from coal firing to oil firing are
judged practical. Under these considerations, it was concluded that solid
SRC can be applicable only to coal-fired boilers while liquid SRC, along
with other liquid fuels of liquefaction process can be applicable to both
coal- and oil-fired boilers. The use of liquid SRC, however, would require
heating of all fuel handling equipment in contact with the fuel to above
350 F, resulting in a high boiler modification cost.
Model Plant Calculations
Of the various coal liquefaction processes under development,
the SRC is the most advanced. In addition, it produces a solid form of
fuel which can be readily used in the existing coal-fired boilers. For
these reasons the process was chosen in the model plant calculation of
liquefaction process. The liquid fuel which can be obtained from the
process is difficult to handle in the existing coal- and oil-fired boilers,
however, and is not considered to represent a typical liquid fuel from
liquefaction. The H-coal process, therefore, was also treated in this
analysis to examine the economics of liquid fuel application.
SRC Process
The SRC process is capable of reducing the sulfur and ash con-
tents in the coal to 0.6 percent and 0.05 percent by weight, respectively.
The heating value of the product is.estimated at 15,900 Btu/lb regardless
of the original heating value of the coal. H_S gas is generated -and
utilized in a Glaus reactor to produce elemental sulfur. When the SRC
fuel is consumed in industrial and commercial boilers, the S00 emission
/• £•
is estimated at 0.75 lb/10 Btu.
-------
95
Table 16 shows estimated costs for SRC manufacturing in mid-1973.
(42)
The estimations were largely based on information provided by Battelle
(43)
and M. W. Kellogg Company. ' The Utility Financing Method, as presented
in Appendix A of this report, was employed. The estimated product cost is
$1.04/10 Btu or $33.07/ton of product.
H-Coal Process
The estimated costs for the H-coal liquefaction process are shown
in Table 17 for producing synthetic crude oil (Case 1) and for producing
fuel oil and naphtha (Case 2). The estimations were based largely on the
information provided by Hydrocarbon Research, Inc. The manufacturing cost
6 fi
was estimated at $1.38/10 Btu and $1.34/10 Btu for the production of
synthetic crude oil and naphtha, respectively.
-------
TABLE 16. COSTS FOR SOLVENT REFINED COAL PROCESS
Mid-1973, 8000 hours/yr Operation
Item
Product Rate (tons/day)
Type Coal
Coal Input Rate (tons/day)
Sulfur Production Rate (long tons/day)
^apital^JRequirement (10 $J
Coal preparation
Preheaters/dissolvers
Ash separation
Solvent/aromatics recovery
Sulfur recovery
Product handling
Hydrogen plant
Other
Total Bare Cost
Engineering and design
Contractor fees
Subtotal Plant Investment
Project contingency
Total Plant Investment
Interest during construction
Startup
Working capital
Total Capital Requirement
Quantity
7,236
Eastern
Medium S
11,993
122
8.18
22.17
10.81
16.76
4.44
5.36
8.54
29.19
105.45(a)
5.27
10.55
121.27
18.19
139.46
23.53
10.13
10.13
183.25
7,236
Eastern
High S
12,664
300
8.50
23.03
11.42
21.73
7.29
5.36
8.87
31.85
118.05(a)
5.90
11.81
135.76
20.36
156.12
26.35
10.70
10.70
203.87
7,236
Central
Medium S
13,765
145
9.01
24.42
12.41
18.60
4.89
5.36
9.40
31.49
115.58(a)
5.78
11.56
132.92
19.94
152.86
25.80
11.38
11.38
201.42
7,236
Central
High S
14,391
347
9.30
25.20
12.98
23.89
7.90
5.36
9.70
34.22
128.55(a)
6.43
12.86
147.84
22.18
170.02
28.69
11.92
11-.92
222.55
7,834
13,600
300
7.29
29.15
10.93
21.86
7.29
7.29
7.29
29.16
120.26(b)
6.01
12.03
138.30
20.74
159.04
26.84
11.35
11.35
208.58
-------
TABLE 16. (Continued)
Item
Annual Costs (10 $/yr)
(c)
Direct operating labor
Maintenance (3.5 percent/yr)
Supervision
Administration and general overhead
Local taxes and insurance , ,-.
Catalysts, chemicals, etc.
Water*37 (
Sulfur recovery supplies
Gross Cost Excluding Coal
Aromatics and power credit
Sulfur credit ($10/long ton)
Net Cost Excluding Coal
Coal cost ($10/ton)
Net Annual Operating Cost
Depreciation
Return on rate base
Federal income tax
Total Annual Cost
Product Cost ($/ton product)
Product Cost ($/10 Btu output) 8
Heat Output Rate (10 Btu/yr) 8
2.33
4.88
0.64
2.95
3.77
0.50
0.20
0.33
15.60
-4.53
-0.41
10.66
39.98
50.64
8.66
10.15
3.35
72.80
30.18
0.95
76.7
2.33
5.46
0.68
3.11
4.22
0.50
0.21
0.81
17.32
-5.03
-1.00
11.29
42.21
53.50
9.66
11.26
3.71
78.13
32.39
1.02
76.7
Quantity
2.33
5.35
0.67
3.08
4.13
0.50
0.23
0.39
16.68
-5.20
-0.48
11.00
45.88
56.88
9.50
11.17
3.68
81.23
33.68
1.06
76.7
2.33
5.95
0.71
3.25
4.59
0.50
0.24
0.94
18.51
-5.74
-1.16
11.61
47.97
59.58
10.53
12.31
4.06
86.48
35.85
1.13
76.7
2.33
5.57
0.68
3.15
4.29
0.50
0.23
0.81
17.56
-5.13
-1.00
11.43
45.33
56.76
9.86
11.55
3.81
81.98
31.39
0.99
83.0
vO
-------
98
Footnotes to Table 16
(a) Bare cost of all sections except sulfur recovery based on Battelle
Energy Program report "Liquefaction and Chemical Refining of Coal",
July, 1974. Sulfur recovery section costs based on Shore, et al.,
EPA 650/2-74-098, September, 1974.(43>
(b) Bare cost of all sections based on Shore, et al.
(c) Operating labor = 175 men, consensus of two sources cited in (a).
(d) Based on Battelle Energy Program report.
(e) Based on Battelle analyses of requirements for amine scrubbing units
and Claus plants. Total requirements per long ton of sulfur recovered
by Claus plant are 92.5 kWh electricity, 13,300 Ib steam, 54,000 gal
cooling water, 1.81 Ib monoethanolamine, 800 gal boiler feed water,
and 0.4 Ib activated alumina.
(f) Based on Battelle Energy Program report. Phenol at Ic/lb, cresylic
acids at 0.5c/lb, power at0.6c/kWh, and 0.5 kWh per pound of ash
burned.
(g) Based on 15,900 Btu/lb product.
-------
99
TABLE 17. COSTS FOR H-COAL LIQUEFACTION PROCESS
Mid-1973, 8000 hours/yr Operation
Quantity
Item
Products (bb I/day)
Synthetic crude oil
Fuel oil
Naphtha
By-products
High Btu fuel gas (109 Btu/day)
Ammonia (tons/day)
Sulfur (long tons/day)
Capital Requirement (10 $)
On-site investment
Off-site investment
Initial catalyst charge
Total Bare Cost
Engineering and design
Contractor fees
Subtotal Plant Investment
Project Contingency
Total Plant Investment
Interest during construction
Startup
Working capital
Total Capital Requirement
Annual Costs (10 $/yr) '
Payroll with benefits
Maintenance materials
Maintenance labor
Contracted services
Overhead and other expenses
Local taxes and insurance
Electricity (l
-------
100
TABLE 17. (Continued)
Quantity
Item
Annual Costs (10 $/yr)
Fuel gas credit (900/10 Btu)
Ammonia credit ($33/ton)
Sulfur credit ($10/long ton)
Net Cost Excluding Coal
Coal cost ($10 /ton)
Net Annual Operating Cost
Depreciation
Return on rate base
Federal income tax
Total Annual Cost
Product Cost ($/bbl product)
Product Cost ($/10 Btu output) ^
12 (c)
Heat Output Rate (10 Btu/yr)
Case 1
-26.97
-2.26
-3.26
22.69
96.14
118.83
22.84
26.48
8.73
176.88
7.87
1.38
127.9
Case 2
-10.23
-1.67
-2.36
31.78
92.59
124.37
18.69
22.24
7.33
172.63
8.20
1.34
128.8
(a) Total plant investment based on C. A. Johnson, et al.,
"Present Status of the H-Coal Process," Hydrocarbon
Research, Inc., 1973.(35>
(b) Operating requirements and/or costs based on paper by
C. A. Johnson, et al.
(c) Based on following densities and heating values:
Density Heating Value
Product ("API) (Btu/lb)
Synthetic crude oil 25.2 18,000
Fuel oil -3.1 16,700
Naphtha 50.0 18,700
Heating values from J. B. Maxwell, Data Book on
Hydrocarbons, p 180, 1950.
-------
101
FLUIDIZED-BED COMBUSTION
One of the potentially viable techniques for SO control is
X
fluidized-bed combustion (FBC) of high-sulfur fossil fuel. Winkler
invented the fluidized-bed combustion concept in 1921 for use in coal
gasification long before fluidized-bed technology came into general use
in the 1940's for catalytic cracking in petroleum refining. Not until
early in the 1960's, however, did fluidized-bed combustion as a boiler
firing technique receive attention, first in Europe as a method of
utilizing anthracite fines^ ' and lignite/ ' In the early 60's,
experimental programs on fluidized-bed combustion were undertaken by the
National Coal Board (NCB)^ ' and the British Coal Utilization Research
(47 48)
Association (BCURA) ' for the main purpose of reducing capital costs
of power stations. In the United States, research programs were begun in
the mid-60's by Pope, Evans, and Robbins (PER)^ ' to develop packaged
industrial boilers.
FBC Technology and Environmental Emissions
Figure 25 shows a simplified fluidized-bed combustion boiler
concept. The combustion air passes through a bed of lime (or limestone),
coal, and ash particles at such velocity (2-15 ft/sec) as to suspend all
particles in the bed and to set all particles in a homogeneous fluid
motion. In this state, the particles are separated from each other by an
envelope of the fluidizing gas and present an extended surface for combus-
tion. In addition, the randomly moving particles remain in the fluidized
bed long enough for efficient combustion.
Fluidized-bed combustion has a high volumetric heat release rate
of 500,000 Btu/hr-ft , as compared to 20,000 Btu/hr-ft3 in a pulverized-
coal-fired boiler. Also, the rapid movement of the solid particles passing
over tubes immersed in the bed results in a high rate of heat transfer.
Thus, smaller boilers with less tube surface should be possible for
fluidized-bed combustion systems, allowing a 250,000 Ib steam/hr industrial
coal-fired boiler to be shop fabricated.
-------
102
Water
Walls
Baffle -*t-
Tubes •*
Evaporator
Section
Air
Lime
Primary
Cyclone
Secondary
Paniculate Removal
•'.'.••.'•••.'»•.'. *• .' '•'(•.•• '..•
Coal
(IA/1
Exhaust
Heat Recovery
Section
Water
Walls
I
Ash; Particulates
Sulfate. Ash
Preheater, Superheater
or Reheater Section
— Distributor Plate
Pressure:
Coal Size:
Air Flow:
Temperature:
1 - 25 atm
pf - 1/4 in.
2-15 ft/sec '
1400 - 1900°F
Surface: Water Walls. Horizontal, and
Vertical Tubes in Bed
Sulfur removal: CaO + S02 +
FIGURE 25. FLUIDIZED-BED COMBUSTION BOILER1
(44)
-------
103
The pollutants from the process include residual emission of S0_
and by-product solid wastes. As a sorbent material, dolomite is more
efficient than limestone. For a mole ratio of Ca/S of 3, dolomite was
able to remove 95 to 99 percent while limestone was able to remove 75 to
85 percent of the total sulfur. This might be attributed to the fact that
dolomite becomes more porous than limestone when carbonated. The by-product
solid wastes in general contain ash, CaSO., CaO, unburned carbon, and MgO.
The discharge rate depends on the Ca/S ratio and sorbent material employed.
For a Ca/S ratio of 3, about 948 Ib/ton of coal and 726 Ib/ton of coal
would be generated for using dolomite and limestone, respectively.
The fluidized-bed temperatures (1400°F to 1900°F) are selected to
achieve the maximum SO. capture by the lime or limestone (over 90 percent
removal). At these low temperatures, NO emissions are reduced (250-600 ppm)
and clinkering problems are minimized. Experimental evidence indicates that
the reaction of NO with CO to form N is promoted by CaO. Therefore,
X eL
when combustion was carried out in two stages, one under reducing conditions
(oxygen deficient) and one under oxidizing conditions (oxygen sufficient),
NO emissions were reduced to 70 ppm. Pressurized operation also favors NO
X (52) X
reduction (50-200 ppm NO at 5 atm).v '
X
Applicability to NUC Sources
In FBC, the high volumetric heat-release rate and heat-transfer
rate in the fluid-bed combustion system permits the design of compact
boilers. Thus, an industrial boiler of up to 300,000 Ib/hr steam capacity
could be shop-fabricated and transported by rail. Design of FBC units for
operation under pressure will have the effect of further reducing the size
of the fluidized-bed boiler. It also enables a portion of the power to be
generated by a gas-turbine yielding a higher overall thermal efficiency
for the total plant. Thus, pressurized operation has advantages for
larger boiler systems when it is used for the production of electrical
power. This value is considerably decreased when the objective is process
steam. The pressurized concept requires much more expensive components
such as pressure units and particle control system, as compared to the
-------
104
atmospheric pressure design and this becomes a disadvantage for small plants.
Therefore, a small pressurized fluidized-bed industrial boiler of capacity
250,000 Ib/hr steam will not be economical compared to an atmospheric pres-
sure unit.
Conceptually, FBC is a design for a new boiler system; most of
the energy resulting from combustion is extracted by steam coils in the
fluidized bed and only a small portion of the energy is carried ovet as
sensible heat by the combustion flue gas. Steam coils are placed in the
bed to control the bed temperature at 1400 F to 1800°F which is the optimum
temperature for the reaction of CaO and S0_.
Two conceptual approaches may be considered for the application
to the existing industrial and commercial boilers. One of them is to
install a fluidized-bed combustion unit prior to the existing boiler unit.
Steam is generated both in the fluidized bed and the existing boiler unit.
The existing boiler unit, however, is used as a heat exchanger to recover
the sensible heat from the combustion gas. The net effect is that the
existing facilities will be derated except for coal handling and storage
facilities.
The other approach is to operate a FBC unit under reducing condi-
tions wherein only a fraction of the stoichiometrit: amount of air necessary
for complete combustion is employed in the fluidized bed. The temperature
of the bed is maintained at about 1600°F with a minimal amount of heat
withdrawal through steam coils. The unburned carbon will be recycled to
the carbon burnup cell where oxidizing conditions will be maintained to
achieve complete combustion. A low-Btu gas (approximately 150 Btu/scf)
resulting from the fluidized-bed combustion and a hot flue gas (at about
1600 F) resulting from the carbon burnup cell combustion are used in the
existing boiler for steam generation. This type of combustion is similar
to a low-Btu gasification process. This concept was not considered in this
section of the study.
In retrofitting a FBC unit to an existing boiler system, modifi-
cations of the boiler configuration would be necessary. The pressure drop
across the existing boiler system would be increased due to the increase
in the flue gas flow rate. The pressure drop, however, may not exceed
-------
105
a certain level, i.e., 2 psi, due to the limited strength of the boiler
wall and supporting system. Therefore, some of the baffles and steam coils
should be removed to reduce the pressure drop. Moreover, the air heater or
economizer may also have to be removed and, consequently, the boiler capacity
and efficiency would be decreased.
Model Plant Calculation
In this study, two alternatives were analyzed to examine economic
and technical feasibility. The first was to replace an existing boiler
system with a new FBC boiler system and the second, to retrofit a FBC sys-
tem to an existing boiler system. In the latter case, the FBC unit was
assumed to be operated using excess combustion air, i.e., 150 percent.
About 55 percent of the total steam was assumed to be generated from the
existing boiler system and the remainder from the FBC unit.
The basis of model plant calculation is as follows:
3
• Steam generation capacity: 250 x 10 Ib/hr
• Sulfur content of coal: 3 percent by weight
0 Ash content of coal: 14 percent by weight
• Heating value of coal: 12,000 Btu/lb
• Fluidized-bed temperature: 1600°F
• Operating pressure: atmospheric
« Ca/S ratio: 3 by mole
• Sulfur removal scheme: once through.
The thermal efficiency of the adiabatic FBC unit may be assumed to be 90
percent excluding sensible heat loss from effluent gas stream. ' The
thermal efficiency of the existing boiler system when retrofitted with a
FBC unit would be reduced to about 50 to 60 percent due to the boiler
modification and the increased sensible heat loss from the effluent gas
stream. Table 18 shows the operating conditions of the FBC systems under
consideration. Table 19 shows estimated costs of two different FBC appli-
cations under consideration. Although the capital cost of the retrofit
system is lower than that of the new system, the operating cost is higher
due to the low thermal efficiency of the retrofit system.
-------
TABLE _ 18. OPERATING CONDITIONS OF FBC SYSTEMS
Boiler Capacity = 250,000 Ib steam/hr
Boiler Load Factor = 45 percent
Item
Excess Air, percent
Boiler Modification
New FBC Boiler
10
None
Retrofit System
150
The pressure drop may increase b
y about
Fraction of Steam Produced by
Existing Boiler
Existing Boiler Efficiency, percent
Overall Efficiency, percent
Overall Steam Generation Capacity,
103 Ib/hr
Coal Requirement, 10^ tons/yr
Limestone (or Dolomite) Requirement,
103 tons/yr
3
Solid Wastes, 10 tons/yr
Power Requirement, kW
0.0
N.A.
83
250
*49.5
13.5 (24.6)'
18.0 (23.5)'
300
3 times. To reduce gas flow resistance,
baffles, if any, would be removed. If
necessary, the air heater, economizer,
or some of the steam coils would be
removed. A bigger capacity fan should
be employed.
0.55
55
71
250
57.8
15.8 (28.7)'
21.0 (27.4)'
500
* When dolomite is used instead of limestone.
-------
107
TABLE 19. ESTIMATED COSTS OF FBC ALTERNATIVES
FOR INDUSTRIAL BOILER APPLICATION
Item
Capital Requirement. $106
FBC System^ ,,,
Boiler Modification^ '
Total Bare Cost
Engineering and Design
Contrac tor's Overhead
and Profit
Subtotal Plant Investment
Project Contigency
Total Plant Investment
New FBC
Boiler
1
I
0
0
1
0
2
(c)
Interest During Construction 0
Start-up Cost
Working Capital
Capital Requirement
Retrofit
Total Capital Requirement
Operating Cost, $10
Labor
Administrative and General
Overhead . .
Materials and Utilities^6'
Solid Waste Disposal (f)
Additional Fuel Cost*'8'
Local Taxes and Insurance
Gross Operating Cost
Credit
Net Operating Cost
Annualized Control Cost, $106
Return on Rate Base
Federal Income Tax
Depreciation
Net Operating Cost
Average Annual Cost
Annualized Control Cost,
$/lb S removed
$/106 Btu Output
0
0
2
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.64
0
.64
.08
.16
.88
.28
.16
.10
.08
.08
.42
.42
.06
.04
.15
.09
.01
.06
.41
.41
.13
.04
.12
.41
.70
.29
.71
(0
(0
(2
(2
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
*
.10)
.10)
.46)
.46)
.22)
.12)
.51)
.51)
.13)
.04)
.12)
.51)
.80)
.28)
.81)
Retrofit
FBC Bpiler
1
0
1
0
0
1
0
1
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.02
.10
.12
.06
.11
.29
.19
.48
.07
.11
.11
.77
.18
.95
.05
.03
.18
.11
.10
.04
.51
.51
.11
.04
.18
.51
.84
.30
.85
(0
(0
(1
(0
(2
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
.14)
.14)
.83)
.18)
.01)
.27)
.15)
.64)
.64)
.11)
.04)
.19)
.64)
.98)
.30)
.99)
* Values in parentheses represent when dolomite is used instead of
limestone.
-------
108
Footnotes for Table 19
(a) A new FBC boiler system includes main FBC combustion unit, carbon
burnup cell, superheater, economizer, air plenum, casing enclosure,
ducts, structural supports, platforms, boiler trim, forced draft
fan, auxiliary equipment, etc. It excludes coal preparation
facilities and off-site facilities. The retrofitted FBC systems
does not include carbon burnup cell, economizer, superheat, and
air heater. Installation of steam coils in the retrofit FBC unit
depends on the fraction of total steam to be generated from the
unit. The bare cost of FBC system was estimated based on the infor-
mation from References (53) and (54).
(b) Boiler modification cost includes costs for removing baffles and
some of the steam coils, and cost for fan replacement. In the case
of installing a new FBC boiler system, the dismantling cost of the
existing boiler was assumed to be equal to the salvage value of the
existing boiler system.
(c) Interest during construction was obtained by Interest During
Construction = Total Plant Investment x Interest Rate (0.09) x
Effective Construction Period (0.5 year).
(d) This includes the direct operating labor, maintenance, and
supervision.
(e) This includes maintenance and operating supplies, limestone, or
dolomite (at $7/ton) and power. The cost forA fuel was not included.
(f) The solid waste disposal cost was assumed at $5/ton0
(g) The basis of boiler thermal efficiency was assumed to be 85 percent.
The cost of coal was assumed at $10/ton.
-------
109
FLUE GAS DESULFURIZATION (FGD) PROCESSES
Five processes are considered for post-combustion control of SO
from NUC sources: limestone, lime, double alkali, MgO (with both integrated
and centralized regeneration), and Wellman-Lord,, Below are brief process
descriptions; more complete descriptions are included in Appendix Bo
Process Descriptions
Limestone Slurry
The process considered here has been developed by Peabody Engin-
eering. Flue gas is first cleaned of particulate matter in an electro-
static precipitator (ESP) or equivalent. SO. is then reacted with CaCCL in
a spray tower absorber where 70 to 90 percent of the sulfite is oxidized to
sulfate. After vacuum filtration, the resulting 70 percent solids cake of
CaSOo/CaSO, is transported by truck to a landfill area. SO^ removal efficiency
is from 70 to 90 percent. Other than the CaSO~/CaSO/ (three pounds on a dry
basis per pound of S02 removed) there is no waste produced. A full-scale
(175MW) unit at Detroit Edison's St. Glair No. 6 is presently undergoing
start-up.
Lime Scrubbing
A. Bo Bahco Ventilation, Enkoping, Sweden has developed an industrial
sized lime scrubbing process that is being marketed in the States by Research-
Cottrell, Inc., Bound Brook, New Jersey. The Bahco process uses lime slurry in
a two-stage venturi scrubber to remove particles and S0? from flue gas. ' '
Both CaSO, and CaSO, are produced and are removed from the process in the form
of a sludge stream which is thickened and filtered. SO- removal ranges from
70 to 90 percent depending on the S0~ concentration in the flue gas. Other than
CaSO_/CaSO, sludge (2.5 Ib per pound of SO- removed on a dry basis) there are
no waste streams. Currently, 19 commercial units have been installed in Japan
and Sweden. Start-up of a 20 MW coal fired unit in the U. S. is expected in
early 1976.
-------
110
Double Alkali
FMC is one of a number of developers of processes .that scrub with
a Na?SO_ buffer solution and then react the clear solution with lime or
limestone to precipitate CaSCL. ' The purpose of separating scrubbing
from precipitation has been to eliminate scaling difficulties. Since the
sludge removed from the system contains four to five percent Na-SO-r and
Na_SO,, soda ash must be added to the system to replace these sodium losses.
Removal efficiencies have been 99 percent for flyash and 90 percent for S0?.
Start-up of a 45 MW unit is expected shortly.
MgO Process
Chemical Construction Corporation, New York, New York, has developed
a regenerable FGD process that has eliminated sludge disposal problems„
Flue gas passes through an ESP and contacts a finely divided slurry of MgO in
a venturi scrubber. SO reacts with the MgO to form hydrated MgSO- and a
small amount of MgSO,. The MgSO., and MgSO, are centrifuged and dried. Sub-
sequently, the dried product is taken to a regeneration facility where it is
calcined forming MgO and driving off S0_ which can be used to produce high
grade sulfuric acid. Ninety percent S09 removal efficiency has been demon-
strated. In addition to make-up MgO, fuel oil is required to fire the calciner
and the drier. The process has been demonstrated on a 155 MW oil fired boiler
(Boston Edison's Mystic Station) and on a 190 MW coal fired boiler (Potomic
Electric Power's Dickerson No. 4).
Wellman-Lord
In this process, flue gas which has been cleaned of particulate
matter is contacted with a slurry of Na SO.,, NaHSO_, and Na0SO. . SO reacts
( 63^
with the sulfite to form bisulfite. In addition, some oxidation of
sulfite takes place. A ten percent slip stream is sent to an evaporator
-------
Ill
where the bisulfate is decomposed, regenerating the sulfite and evolving SQ^
which can be used in the production of H_SO or elemental sulfur. A certain
amount of sulfate and thiosulfate is produced in the bisulfate decomposition,,
These species are treated in a purge treatment system,, NaOH or Na-CCL are
added to make up for sodium ion lost in the purge. SO- removal has been 90
percent. In addition to sodium make-up, steam is required for evaporator
operation. A large number of units are presently operating in Japan on acid
plants and oil-fired boilers. A coal-fired demonstration at Northern Indiana
Public Service Corporation's Gary, Indiana power plant is scheduled for
start-up in January 1976.
Applicability to NUC Sources
Although most of the flue gas desulfurization (FGD) systems applicable
to utility boilers can be used on NUC sources, the application to NUC sources
differs somewhat from its application to utility boilers. Many NUC sources
release stack gas at 400 to 500°F, so there is greater potential for gas reheat
by heat recovery from the incoming gas. In general the NUC sources require
larger excess air than the utility boilers. Because of larger requirements
for excess air, SO. concentration will be lower and oxygen concentration will
be higher. This can possibly cause difficulties in processes where oxidation
is undesirable such as the double alkali and Wellman-Lord processes. Higher
excess air also means that a larger quantity of flue gas must be handled for
a given quantity
-------
112
frequently has captive uses for sulfuric acid or sulfate salts. The petroleum
industry has sources of hydrogen sulfide that could be used to produce sulfur
from S0_ emissions. Such special cases will be more abundant in industrial
applications than in utility applications.
The turndown ratio for an NUC source is such that the boiler is
shut down and started up many times during a year. Therefore, an FGD system
must have load following capability. This type of operation can be achieved
by a high degree of automation and by providing a large surge volume after
the scrubber so that the regeneration or waste disposal system can continue
to operate when the boiler and scrubber are shut down. Of course, remote
regeneration or waste disposal facilities tied to several boilers would
automatically provide the surge volume.
The relatively small size of NUC sources may offer some unique
situations for the disposal of purge streams or waste products from FGD systems,
The volume of the purge stream or the tonnage of the waste products can be
two orders of magnitude less than for a typical utility boiler. Water
authorities may allow the purge stream to be discharged to city sewer system
when combined with other waste streams in the plant. The waste products such
as calcium sulfite/sulfate sludge may be trucked feo a nearby sanitary landfill
for disposal.
Industrial plants in general are built with a higher ratio of equity/
debt as compared with utilities; thus, taxes are higher. Also, nonregulated
industry requires higher return on investment because of the risk involved.
The combined effect results in a higher annualized capital charge. Moreover,
the relative impact of capital costs is increased because of the small scale
of operation. In addition, the annualized cost per unit of heat output can
be high because of the low load factor for NUC sources.
Model Plant Calculation
To determine the control cost of each of the FGD processes described
above, a conceptual cost study was carried out for a 250,000 Ib steam/hour
-------
113
boiler. The coal was assumed to contain 3 percent of sulfur, 14 percent of
ash, and 12,000 Btu/lb of heating value. The labor, materials, and utility
requirements for each process are shown in Table 20 and the estimated control
costs are shown in Table 21. The Utility Financing Method, as presented in
Appendix A of this report, was employed for the estimations.
-------
TABLE 2a LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR FGD PROCESSES
Capacity: 250,000 Ib steam/hr (65,600 scfm flue gas)
Coal: 3% sulfur; 14% ash; 12,000 Btu/lb heating value
Item
Utility
Power, kW
Steam, Ib/hr
Water, gal/min
Fuel Oil, gal/hr
Materials
Lime, ton/hr
Limestone, ton/hr
Soda Ash, ton/hr
Natural Gas, scf/hr
MgO, ton/hr
Coke, ton/hr
Maintenance, % TPI*/yr
Labor
Operation, man/shift
Maintenance, % TPI/yr
Peabody
Limestone
550
4,700
80
0
0
1.5
0
. 0
0
0
2.0
1
1.0
Bahco .
Lime
470
4,700
32
0
0.74
0
0
0
0
0
1.5
0.5
1.5
FMC
Double
Alkali
560
6,000
39
0
0.91
0
0.14
0
0
0
1.0
0.5
1.0
MgO
MgO Central
Integrated Regeneration
500
0
344
71
0
0
0
0
0.009
0.008
3.5
2
3.5
350
0
28
39
0 .
0
0
0
0.009
0
2.0
1
2.0
•Wellman-
Lord
800
9,300
270
0
0
0
0.037
6,300
0
0
2.0
2
2.0
* TPI indicates total plant investment.
-------
115
TABLE 21. ESTIMATED CONTROL COST FOR FGD PROCESSES
Boiler Capacity: 250,000 Ib steam/hr (65,600 scfm flue gas)
Load Factor: 45 percent
Coal: 3% sulfur; 14% ash; 12,000 Btu/lb heating value
Cost (mid-1973), 103 dollars
Peabody
Item Limestone
Capital Requirement
Bare cost
Engineering and design
Contractor's overhead and profit
Subtotal Plant Investment
Project Contingency
Total Plant Investment
Interest during construction
Startup cost
Working capital
Capital Requirement
Capital requirement for boiler
retrofit and modification ^ °'
Total Capital Requirement
Annual^ Operating Cost
Labor
Administrative and general overhead
Materials and utilities
Solid waste disposal or
central regeneration
Additional fuel requirement
Local taxes and insurance
Gross Operating Cost
Credit
Net Operating Cost
Annualized Control Cost
Return on rate base
Federal income tax
Depreciation
Net operating cost
Average Annual Cost
Annualized Control Cost
$/lb S removed
$/10 Btu output
l,452(a)
—
138
1,590
239
1,829
82
54
54
2,019
606
2,625
30
18
125(d)
46(e)
0
49
268
0
268
141
46
257
268
712
0.34(«
0.79
FMC
Bahco Double
Lime Alkali
1,962(1)
—
..
1,962
264
2,226
100
71
71
2,520(s) 2,468
756 494
3,276 2,962
30(h)' 23(m)
129(D 180(n)
fi.,U) 77(o)
O J / /
0 0
62 60
302 354
0 0
302 354
175 159
58 53
322 290
302 354
857 856
0.35(k) 0.33^
0.87 0.87
MgO-
Integrated
2,481(q)
—
--
2,481
..
2,481
112
80
80
2,753
826
3,579
102(r)
168(s)
0
0
67
398
69(t)
329
192
63
350
329
934
0.36
0.95
MgO-
Central
Regeneration
l,428(u)
--
—
1,428
__
1,428
64
34
34
1,560
468
2,028
38
61(w>
u(x)
0
39
172
0
. 172
108
36
200
172
516
0.20
0.52
Wellman-
Lord
l,995(y)
—
199
2,194
329
2,523
114
62
62
2,761
828
3,589
69(Z)
132(aa)
10(bb)
56
308
13
295
192
63
• 352
295
902
0.35
0.92
-------
116
Footnotes to Table 21
(a) The exponent scale-up factor was assumed to be 0,5.
(b) The retrofit factor was assumed at 1.3 for Peabody limestone, Bahco
lime, MgO-integrated, MgO-central regeneration, and Wellman-Lord
processes. The retrofit factor was assumed at 1.2 for FMC double
alkali process.
(c) Direct operating labor = 1 man/shift.
Maintenance =1/3 man/shift.
(d) Limestone: $131,000
Power: 48,000
Steam: 20,000
Water: 22,000
Maintenance: 37,000 (2 percent of total plant investment)
Supply: 20 » OOP (3 percent of labor)
$278,000 at full load
(e) Solid waste generation = 4 Ib/lb of SO- removed.
(f) Coal required at full load = 1007 x 10 tons/year;
Sulfur removed at 80 percent efficiency = 2,577 tons/year.
.
(g) ($2.3 million) V' = $2'01 million» where l^'1 and 164-7 are
CE plant cost indexes for 1973 and 1974,
respectively.
($2.01 million) (i) ' = $2. 52 million.
This represents the capital requirement.
(h) Direct operation $ 22,000
Maintenance: 35,000
Supervision: 9,000
$ 66,000 at full load
(i) Operation supplies: $ 20,000
Maintenance: 35,000
Power: 41,000
Water: 8,000
Reheat steam: 20,000
Lime : 165.000
$286,000 at full load
(j) Sludge generated = 5.1 Ib/lb of SO removed.
(k) Sulfur removed = 2.738 tons/year.
-------
117
Footnotes to Table 21 (continued)
(1)
(m)
(n)
(o)
(P)
(q)
(r)
(s)
The bare cost of a 45 MW systenr ' = $2.791 x 10 „ The bare cost
includes engineering and design and contractor's overhead and profit,
The exponent scale-up factor =0.6.
Direct operation:
Maintenance:
Supervision:
Lime:
Soda ash:
Power:
Steam:
Water:
Operating supplies:
Maintenance:
$ 22,000
22,000 (1 percent of total plant investment)
7.000 .
$ 52,000 at full load
$200,000 .
61,000 ,
49,000
43,000
10,000
16,000
22.000 (1 percent of total plant investment)
$401,000 at full load
Solid wastes generated
Sulfur removed = 2.899 tons/year at full load.
5.85 Ib/lb of SO- removed.
The bare cost for a 200 MW boiler in 1972 = $11.476 x 10- (estimated
from Reference 61). The bare cost for a 25 MW system in mid-1973 =-
(11.476 x 106)
- 0.532 x 1C (for ESP) -
$2.481 x 10. . This cost includes costs for engineering and design,
contractor's overhead and profit, and project contingency.
Direct operation:
Maintenance:
Supervision:
Power:
Fuel oil:
Process water:
MgO:
Coke:
Maintenance:
Operating supplies:
$110,000
87,000 (3.5 percent of TPI)
30.000
$227,000 at full load
$ 44,000
66,000
90,000
16,000
2,000
87,000 (3.5 percent of TPI)
68.000
$373,000 at full load
-------
118
Footnotes to Table 21 (continued)
(t) SO. removed at full load = 5.798 tons/year;
S2S04 (98 percent) = (1.33) (5.798) =7.711 tons/year.
(u) Bare cost for an integrated MgO system = $11.476 x 10 [from Footnote (q)]
Bare cost for the scrubbing system only =
= $1.428 x 10 (total plant investment).
(v) Direct operation: $ 44,000
Maintenance: 29,000 (2 percent of TPI)
Supervision: 11.000
$ 84,000 at full load
(w) Maintenance: $ 29,000
Operating supplies: 25,000
Power: 26,000
Fuel oil: 47,000
Water: 9,000
$136,000 at full load
(x) The regeneration cost was estimated at $8.53/ton sulfur removed (see
Appendix C) . -* •
(y) Total bare cost of a Wellman-Lord system handling a flue gas of
294,000 scfm = $5.856 x .10° (Battelle's estimate in 1974). Total bare
cost of a Wellman-Lord system installed on a 250,000 Ib/hr capacity boiler
(5.856 x 106) CPJ' = $1.995 x 106 (in mid-1973).
(z) Direct labor and
supervision: $113,000
Maintenance: 41.000
$154,000 at full load
(aa) Power: $ 20,000
Process water: 7,000
Cooling water: 32,000
Steam: 71,000
NaOH: 75,000
Natural gas: 50,000
Maintenance: 41,000
$296,000 at full load
(bb) Waste disposal cost = $22,800 at full load.
-------
119
EVALUATION OF ALTERNATIVES
Approach
The evaluation of the potential role of the various alternatives
under consideration in the control of the emissions from small industrial
and commercial boilers requires consideration of a number of diverse
factors which must be related and compared in a meaningful fashion. The
approach involves the following steps:
(1) Development of evaluation criteria
(2) Evaluation of each alternative with
respect to each criterion.
The conversion of the evaluation to a rating scale would be desired for
the rating of the alternatives based on the aggregate points. However,
the procedure involves subjective judgments which would influence the out-
come significantly. The quantitative analysis of the evaluation, therefore,
was not conducted in this study.
Evaluation Criteria
A set of six criteria is employed in the evaluation of the
alternatives as follows:
(1) Pollutant emissions
(2) Retrofitability
(3) Operation maintenance
(4) Capital requirement
(5) Annualized cost
(6) Availability.
The alternatives under consideration have differing potential
for minimizing air pollutant emissions and generating new pollutant emissions.
The variability is expressed in terms of residual and secondary emissions
which result from the application of an alternative.
-------
120
The application of the alternatives to the existing industrial
boiler systems should be made relatively easily. The variability is
evaluated with respect to space requirement and boiler modification.
Alternatives employed are to be operated and maintained as
trouble-free as possible. Operation-maintenance is evaluated with respect
to material handling, technical expertise, the number of moving par±s,
plugging and scaling possibilities, corrosion and erosion possibilities,
and operating temperature and pressure.
Capital requirement indicates the amount of capital required to
incorporate an alternative process. The contribution of capital cost to
annual operating cost is included in the annualized cost.
Annualized cost consists of return on rate base, Federal income
tax, depreciation, and net annual operating cost.
In view of the urgency of related environmental problems., the
availability of given alternatives is an important criterion in the
evaluation. Factors such as raw material availability, developmental
status, year of commercialization, and growth rate are components of the
availability consideration.
Other factors such as by-product were not established as separate
criteria since the criterion would not be a significant factor for small
boilers. Besides, the factor is incorporated in the annualized cost as
credit.
Alternative Evaluation
The next step in the procedure was to develop an evaluation of
each alternative with respect to each of the six criteria. The evaluation
was carried out based on the boiler operation viewpoint. That is, in the
determination of pollutant emissions of the alternatives, the quantity was
limited to the emissions resulting from the combustion of the fuel in
boiler,but did not include the emissions from the fuel conversion process
unless the process was assumed to be retrofitted to the boiler system.
Similarly, in the evaluation with respect to operation-maintenance,
-------
121
only the difficulties that might take place at the site of boiler operation
were considered for evaluation. Therefore, the alternatives of fuel substi-
tution were considered to be free of operation-maintenance problems.
A quantitative evaluation was employed wherever possible, other-
wise qualitative categories for evaluation were developed. The evaluation
of alternatives with respect to capital requirement and annualized .cost
were not included in this section but are discussed extensively in-the next
section since the relative costs depended on the size and operating charac-
teristics of the boilers.
Pollutant Emission
The residual emission of sulfur dioxide was evaluated in terms
of pounds of sulfur dioxide per million Btu steam output. The secondary
emissions resulting from the sulfur dioxide control process were expressed
in terms of the quantity of pollutants per million Btu of steam output.
available.
Retrofitability
The retrofitability was evaluated on the basis of space requirement
and need for boiler modification. The space requirement was categorized by
four groups as follows:
Category 1 - No space requirement
Category 2 - Low space requirement
Category 3 - Moderate space requirement
Category 4 - High space requirement.
The need for boiler modification was evaluated with respect to four categories,
Category 1 - Need for no modification
Category 2 - Need for low modification
Category 3 - Need for moderate modification
Category 4 - Need for high modification.
-------
122
Operation-Maintenance
The operational and maintenance difficulties were.assessed in terms
of technical expertise; characteristics of material handling; possibilities
of plugging, scaling, erosion, and corrosion; number of moving parts, and
operating pressures and temperatures.
The degree of technical expertise required is based on the process
complexity, control sensitivity, and operating conditions. It was expressed
in terms of technical knowledge equivalent to one of either technician or
engineer.
The material handling was evaluated in terms of gases, liquids,
solids, and slurry. The handling of solids or slurry is more difficult than
that of liquids or gases.
Some sorption processes involve solid or slurry streams that are
more susceptible to scaling and/or plugging than others. Scaling and
plugging can precipitate equipment failure and result in operation dis-
ruptions. The potential was evaluated with respect to three categories
as follows: v
Category 1 - Minimal possibility
Category 2 - Moderate possibility
Category 3 - High possibility.
In SO- sorption processes, corrosion is caused primarily by the
presence of dilute sulfuric acid and/or chlorine ions. Erosion is caused
by the abrasive nature of liquids and solids. Both corrosion and erosion
were evaluated with respect to three categories.
Category 1 - Minimal possibility.
Category 2 - Moderate possibility.
Category 3 - High possibility.
The number of moving parts was the summation of all of the major
pieces of equipment containing moving parts. This included conveyors,
rotary drum filters, pumps, blowers and mixers. This factor was categorized
in terms of low, moderate, and high.
-------
123
Operating pressures and temperatures influence the reliability
of process operation to some extent. A high operating pressure or temperature
is more conducive to failure than a low one.
Availability
The availability was evaluated on the basis of raw material avail-
ability, development status, and the year of commercialization and growth
rate (i.e., the rate of implementation). The raw material availability was
evaluated on the basis of two categories defined as follows:
Category 1 - Materials readily available and in
surplus generally through the United
States
Category 2 - Materials either in short supply or
available only to specific areas.
The development status was classified into five categories—conceptual,
bench, pilot, prototype, and commercial. The commercialization was
evaluated in terms of the estimated year of commercial availability as
applied to industrial boiler systems. A major factor influencing the rate
of implementation is the complexity of the alternative. A highly complex
process, requiring a longer lead time for fabrication of components and
construction, and being more capital intensive will result in a lower
implementation rate. With these considerations, the alternatives were
evaluated with respect to three categories defined as follows:
Category 1 - Low degree of complexity
Category 2 -r Intermediate degree of complexity
Category 3 - Highly complex process.
Evaluation Result
The results of the evaluation based on the criteria discussed
above are shown in Tables 22 and 23 for coal- and oil-fired boilers,
-------
TABLE 22. ALTERNATIVE EVALUATION MATRIX FOR COAL-FIRED BOILER
Pollutant Emission
Alternative
Physical Cleaning
Coal Gasification
Coal Liquefaction
Coal Liquefaction
(Low- Btu)
(Solid SCR)
(H-Coal)
Fluldized Bed Combustion
Limestone Slurry
(Peabody)
Lime Scrubbing (Bahco)
Double Alkali (FMC)
MgO (Integrated)
MgO (Central Regeneration)
Wei Iman- Lord
S02
(lb/106 Btu)
NA*
0.52
None
0.75 •
0.22
1.00
1.00
0.75
0.50
0.50
0.50
0.50
Others
(lb/106 Btu)
None
None
None
None
solid waste
(42.6)
sludge
tie.o)
sludge
(21.7)
6 ludge
(26.3)
None
None
Purge Stream
(2.5)
Retrofitabilitv
Space
Requirement
None
High
None
None
None
Low
Medium
Medium
Low
High
Low
High
Boiler
Modification
None
Moderate
Low
Moderate
High
None
None
None
None
None
None
Technical
Expertise
Operation - Maintenance
Material Plugging
Handling Scaling
Erosion
Corrosion
Moving
Parts
Operating
Condition .
Engr
Liquid
Minimum
Moderate
High
High Temp
Normal Operation ^
"• UU1.UU. UF
Tech
Tech
Tech
Tech
Engr
Tech
Engr
Solid
Slurry
Slurry
Liquid
Slurry
Slurry
Liquid
Moderate
High
High
Minimum
Moderate
Moderate
Minimum
Moderate
Moderate
Moderate
Low
High
Moderate
Moderate
Low
Moderate
Moderate
Moderate
High
Moderate
High
High Temp
Normal
Normal
Normal
High Temp
High Temp
High Temp
* NA = Not Applicable.
-------
TABLE 22. ALTERNATIVE EVALUATION MATRIX FOR COAL-FIRED BOILER (Continued)
Availability
Alternative
Physical Cleaning
Coal Gasification (Low-Btu)
Coal Gasification (High-Btu)
Coal Liquefaction (Solid SCR)
Coal Liquefaction (Liquid SRC)
Coal Liquefaction (H-Coal)
Fluidized Bed Combustion
Lines tone Slurry (Peabody)
Lime Scrubbing (Bahco)
Double Alkali (FMC)
MgO (Integrated)
MgO (Central Regeneration)
Wei Iman- Lord
Raw Material
Availability
Available
Available
Available
Available
Available
Available
Available
Available
Available
Available
Available
Available
Questionable
Developmental
Status
Commercial
Commercial
Prototype
Prototype
Prototype
Pilot
Conceptual
(Proven Tech)
Commercial
Commercial
Commercial
Commercial
Commercial
Commercial
Commercialization
Present
Present
1980-1983
1981
1981
1983
1978-1980
Present
Present
Present
Present
Present
Present
Growth
Rate
High
Medium
Low
Low
Low
Low
High
High
High
High
Medium
High
Medium
-------
TABLE 23. ALTERNATIVE EVALUATION MATRIX FOR OIL-FIRED BOILER
Alternative
Coal Gasification
(High-Btu)
Coal Liquefaction
(Liquid SRC)
Coal Liquefaction
(H-Coal)
Limestone Slurry
(Peabody)
Lime Scrubbing
(Bahco)
Double Alkali
(FMC)
MgO (Integrated)
Mgo (Central
Regeneration)
Pollution
S02
(lb/106 Btu)
None
0.75
0.22
0.52
0.39
0.26
0.26
0.26
Emission
Other
(lb/106 Btu)
None
None
None
sludge
(0.83)
sludge
(11.2)
sludge
(13.6)
None
None
Retrofitability
Space Boiler
Requirement Modification
None Moderate
None Moderate
None None
Medium None
Medium None
Low None
High None
Low None
Operation - Maintenance
Technical Material Plugging Erosion Moving
Expertise Handling Scaling Corrosion Parts
^ Normal Opera ti on (H^atl "g of Fu**-l R^qiii red)
Tech Slurry High Moderate Moderate
Tech Slurry High Moderate Moderate
Tech Liquid Minimum Low Moderate
Engr Slurry Moderate High High
Tech Slurry Moderate Moderate Moderate
Operating
Condition
Normal
Normal
Normal
High Temp
High Temp
Wellman-Lord 0.26 Purge Stream High
(0.7)
None
Engr Liquid Minimum Moderate High High Temp
-------
TABLE 23. ALTERNATIVE EVALUATION MATRIX FOR OIL-FIRED BOILER (Continued)
Availability
Developmental Growth
Alternative Availability Status Commercilization Rate
Coal Gasification Available
(High-Btu)
Prototype
1980-1983
Low
Coal Liquefaction
(Liquid SRC)
Available Prototype
1981
Low
Coal Liquefaction
(H-Coal)
Available Pilot
1983
Low
Limestone Slurry
(Peabody)
Available Commercial
Present
High
Lime Scrubbing
(Bahco)
Available Commercial
Present
High
Double Alkali
(FMC)
Available Commercial
Present
High
MgO (Integrated) Available Commercial
Mgo (Central
Regeneration)
Available Commercial
Present
Present
Medium
High
Wellman-Lord
Questionable Commercial
Present
Medium
-------
128
respectively. Oil-fired boilers in general have fewer alternatives than
coal-fired boilers due to the unique boiler configurations which do not
lend themselves to solid fuels and low-Btu gases. The FGD processes
employed for the evaluation were considered commercially available for
the application to commercial and industrial boilers because of the small
size. The type of fuel tested was not taken into consideration here.
-------
129
COST OF ALTERNATIVES
This section is concerned with the control cost of the alternatives
under consideration when applied to NUC sources such as small commercial and
industrial boilers. According to the results of the boiler data analysis
conducted in this study, the area source SO- emissions appear to be concen-
trated in high sulfur coal-fired boilers of between 10,000 and 500-,000 Ib
steam/hr size class and in high sulfur oil-fired boilers of sizes between
1,000 and 500,000 Ib steam/hr. For the purpose of conducting the control
cost analysis, two different boiler size subgroups for coal-fired boilers
and three different boiler size subgroups for oil-fired boilers were selected.
The characteristics of the selected boiler subgroups are shown in Tables 24
and 25 for coal- and oil-fired boilers, respectively. For convenience of
analysis, the nominal standard properties of the fuels employed in the cost
estimation were assumed as shown in Table 26.
The total capital requirements of the alternatives consisted of
costs for on-site installed facilities and costs for retrofit and boiler
modification. It was a battery limit cost otherwise specified and included
costs for equipment, materials, installation, engineering and design, and
startup. Credits for any existing facilities were incorporated in the
estimation. The base year for the cost estimates was mid-1973 and the Utility
(27)
Financing Methodv ' presented in Appendix A of this report was employed in
the estimation of the related costs.
The annualized cost in general consisted of capital charges, main-
tenance, labor, utilities, raw materials, by-product credits, and additional
costs or credits due to the use of the control alternatives.
The estimation of control cost for the processed fuel alternatives
such as coal cleaning, gasification, and liquefaction processes was somewhat
different in procedure from that for retrofit control systems such as FGD
processes. Tables 27 through 32 show the estimated control cost of the
processed fuel alternatives as applied to the selected boiler subgroups
described above.
The control cost of the FGD processes was estimated based on the
same format used in the FGD sections. The following assumptions were
-------
TABLE 24. CHARACTERISTICS OF SELECTED COAL-FIRED COMMERCIAL AND INDUSTRIAL BOILERS
Commercial (10
Item
Load factor
Excess combustion air, percent
Pulverized coal
Stoker
Flue gas flow rate, scfm
Pulverized coal
Stoker
Boiler efficiency,** percent
20
0.42
60*
100*
5,300*
6,700*
85
Ib steam/hr)
250
0.31
50
100
62,500
68,800
35
Industrial (10
20
0.55
60*
100*
5,300*
6,700*
85
3
Ib steam/hr)
250
0.45
50
100
62,500
68,800
85
* Battelle's estimate based on pertinent information.
** Nominal value.
-------
TABLE 25. CHARACTERISTICS OF SELECTED OIL-FIRED COMMERCIAL AND INDUSTRIAL BOILERS
Item
Load factor
Excess combustion air, percent
Flue gas flow rate, scfm
Boiler efficiency,* percent
Commercial
2
0.40 0
36
450 4
85
<103
20
.28
33
,430
85
Ib steam/hr)
250
0.19
52
63,300
85
Industrial
2
0.45
36
450
85
(103 Ib
20
0.36
33
4,430
85
steam/hr)
250
0.41
52
63,300
85
* Nominal value.
-------
132
TABLE 26. NOMINAL STANDARD PROPERTIES OF FUEL
Property
Coal
Oil
Sulfur content, weight percent
Ash content, weight percent
Heating value
3
14
2.3
12,000 Btu/lb 6 x 10 Btu/bbl
-------
TABLE 27. CONTROL COST ANALYSIS FOR PHYSICAL COAL CLEANING
Coal-Fired Boiler, 103 Ib/hr Steam
Item
Boiler Load Factor (7.)
Flue Gas Flow Rate (scfm)
Total Capital Requirement for
Boiler Modification (103 $)
Annual Operating Cost (103 $/yr)
Fuel Cost(a)
Less Base Case Fuel Cost
Investment-Related Cost
Effect on Boiler Operating
Cost
Total
Sulfur Removed (106 lb/yr)(b)
Control Cost ($/lb S Removed)
Control Cost ($/106 Btu Output)
20(C)
42
6,000
47.0
-36.1
10.9
0.078
20(1)
55
6,000
61.6
-47.2
14.4
0.102
250(C)
31
65,600
434
-333
101
0.719
1 A 1
250(1)
45
65,600
630
-483
147
1.043
Oil-Fired Boiler. 103 Ib/hr Steam
2(C) 2(1) 20(C) 20(1) 250(C) 250(1)
40 45 28 36 19 41
500 500 4,400 4,400 63,300 63,300
U)
u>
(a) 54.3 c/10 Btu, based on analysis of 500 ton/hr plant.
(b) Based on cleaned coal heating value of 12,500 Btu/lb and sulfur content of 2 weight percent.
(c) C and I indicate commercial and industrial boilers, respectively.
-------
134
TABLE 28. CONTROL COST ANALYSIS FOR KOPPERS-TOTZEK
COAL GASIFICATION
Coal-Fired Boiler,
Item
Boiler Load Factor (%)
Flue Gas Flow Rate (scfm)
Total Capital Requirement- for, .
Boiler Modification (10 $) ^a'
0
Annual Operating Cost (10 $/yr)
Fuel Cost(b)
Less Base Case Fuel Cost
(c)
Investment -Related Cost
Effect on Boiler Operating Cost
Total
Sulfur Removed (106 Ib/yr) ^e)
Control Cost ($/lb S Removed)
Control Cost ($/10 Btu Output)
20 (C)
42
6,000
29.7
102.2
-36.1
4.3
-14.0
56.4
0.197
0.286
0.77
20(1)
55
6,000
29,7
133.8
-47.2
4.3
-18.3
72.6
0.258
0.281
0.75
103 Ib/hr steam(f)
250(C)
31
65,600
153
942
-333
22.3
-129
502.3
1.82
0.276
0.74
250(1)
45
65,600
153
1,368
-483
22.3
-187
720.3
2.64
0.273
0,73
(a) Based on R. Schreiber, et al., EPA-650/2-74-123, November, 1974, Section 6.
(b) $1.18/106 Btu.
(c) 14.55 percent of investment per year, based on utility financing method
with no maintenance.
(d) Based on average 1973 difference in operating cost (excluding fuel) between
gas-fired and coal-fired boilers (-0.19 mills/kWh). Values from Electrical
World, November 1, 1973.
(e) Difference between sulfur emission for base case and sulfur emission at
boiler and coal gasification plant.
(f) C and I indicate commercial and industrial boilers, respectively.
-------
TABLE 29. CONTROL COST ANALYSIS FOR HYGAS COAL GASIFICATION
Coal-Fired Boiler, 103 Ib/hr Steam Oil-Fired Boiler, 10 Ib/hr
Steam
Item 20(C) 20(1) 250(C) 250(1) 2(C) 2(1) 20(C) 20(1) 250(C) 250(1)
Boiler Load Factor (%) 42 55 31 45 40 45 28 36
Flue Gas Flow Rate (scfm) 6,000 6,000 65,600 65,600 500 500 4,400 4,400 63
Total Capital Requirement for. .
Boiler Modification (1Q3 $)(a; 24.5 24.5 123 123 5.95 5.95 11.0 11.0
Annual Operating Cost (103 $/yr)
Fuel Cost(b) 138.5 181.4 1,278 1,855 13.19 14.84 92.3 118.7
Less Base Case Fuel Cost -36.1 -47.2 - 333 -483 -4.12 -4.64 -28.9 -37.1
Investment-Related Cost(c^ 3.6 3.6 18 18 0.87 0.87 1.6 1.6
Effect on Boiler Operating -14.0 -18.3 - 129 -187 -2.73 -3.07 -19.1 -24.6
Cost
-------
TABLE 30. CONTROL COST ANALYSIS FOR SOLVENT REFINED COAL (SOLID)
Coal-Fired Boiler, 103 Ib/hr Stear/6^ Oil-Fired Boiler, 103 Ib/hr Steam
Item
Boiler Load Factor (%)
Flue Gas
Flow Rate (scfm)
20(C)
42
6,000
20(1)
55
6,000
250(C)
31
65,600
250(1)
45
65,600
2(C)
40
500
2(1)
45
500
20(C)
28
4,400
20(1)
36
4,400
250(C)
19
63,300
250(1)
41
63,300
Total Capital Requirement for
Boiler Modification (103$)(a) 9.9 9.9 48.8 48.8
Annual Operating Cost (103 $/yr)
Fuel Costv
Less Base Case Fuel Cost
(c)
Investment-Related Cost
Effect on Boiler Operating
Cost
Total
Sulfur Removed (106 lb/yr)^
Control Cost ($/lb S Removed)
Control Cost ($/10 Btu/Output)
89.9
-36.1
1.4
WM^^^BKBK
55.2
0.189
0.292
0.750
117.7
-47.2
1.4
«^^^««v.-
71.9
0.248
0.290
0.746
829
-333
7
•^_«^_
503
1.75
0.288
0.741
1,204
- 483
7
^MMB^^B
728
2.53
0.287
0^739
(a) Based on R. Schreiber, et al., EPA-650/2-74-123, November, 1974, pp 6-22 & 23.
(b) $1.04/10 Btu, based on average of four analyses for 7,236 ton/day plants.
(c) 14.557, of investment per year, based on utility financing method with nomaintenance.
(d) Based on solvent refined coal with 15,900 Btu/lb heating value and 0.5 weight percent sulfur.
(e) C and I indicate commercial and industrial boilers, respectively.
-------
TABLE 31. CONTROL COST ANALYSIS FOR SOLVENT REFINED COAL (LIQUID)
— — • i ••^•^•^••^^^^^•^••j a,, •!— .•!— •^•J^^-^^— •^••^•-•^•^••^-•^••»^^^^— ^^— •^-^^^a^^j— ^^•••••..•••••••^•••^•.•^•^•••-••.^-•^^•••^•-
•3 (f)
Coal-Fired Boiler, 10 Ib/hr Steam Oil-Fired
Item 20(C) 20(1) 250(C) 250(1) 2(C) 2(1)
Boiler Load Factor (7») 42 55 31 45 40 45
Flue Gas Flow Rate (scfm) 6,000 6,000 65,600 65,600 500 500
Total Capital Requirement for
Boiler Modification (103 $)(a) 262 262 1,477 1,477 127 127
Annual Operating Cost (103 $/yr)
Fuel Cost(b) ^ 89.9 117.7 829 1,204 8.56 9.63
Less Base Case Fuel Cost -36.1 -47.2 -333 - 483 -4.12 -4.64
Investment-Related Cost(c) 38.1 38.1 215 215 18.48 18.46
Effect on Boiler Operating
Cost 15.8 20.7 146 212 0.11 0.12
Boiler,
20(C)
28
4,400
275
59.9
-28.9
40.0
0.7
Total 107.7 129.3 857 1,148 23.03 23.59 71.7
Sulfur Removed (106 lb/yr)(e) 0.189 0.248 1.75 2.53 0.0085 0.0095 0.0593
Control Cost ($/lb S Removed) 0.57 0.52 0.49 0.45 2.72 2.48
Control Cost ($/106 Btu Output) 1.46 1.34 1.26 1.16 3.29 2.99
(a) Based on R. Schreiber, et al., EPA-650/2-74-123, November, 1974, Section 6.
(b) $1.04/10 Btu, based on average of four analyses for 7,236 ton/day plants.
1.21
1.46
ii ^^— ••.^•a
, 103 Ib/hr
Steam
20(1) 250(C)
36
4,400 63
275 1
77.1
-37.1
40.0
1.0
81.0
0.0762 0
1.06
1.28
19
,300
,139
508
-245
166
6
435
.503
0.86
1.05
250(1)
41
63,300
1,139
1,097
- 528
166
14
749
1.084
0.69
0.83
(c) 14.55 percent of investment per year, based on utility financing method with no maintenance.
(d) Includes cost of additional SRC to heat up and melt the SRC (199 Btu/lb) plus for
coal-fired the average 1973 difference in operating cost (excluding fuel) between
boilers (0.20 mills/kWh from Electrical World, Nov. 1, 1973).
(e) Based on SRC with 15,900 Btu/lb heating value and 0.5 weight percent sulfur.
(f) C and I indicate commercial and industrial boilers, respectively.
boilers originally
oil-fired and coal-fired
-------
TABLE 32. CONTROL COST ANALYSIS FOR H-COAL LIQUEFACTION
Coal-Fired Boiler, 103 Ib/hr Steam(f^
Item
Boiler Load Factor (%)
Flue Gas Flow Rate (scfm)
Total Capital Requirement
Rn-Mer- MoHi f 1 rat-inn ("10
20(C)
42
6,000
fo? >^
<;Ua) „ A
20(1)
55
6,000
S7.i
250(C)
31
65,600
7QS
250(1)
45
65,600
7QS
2(C)
40
500
Oil-Fired Boiler, 103 Ib/hr Steam
2(1)
45
500
20(C)
28
4,400
20(1)
36
4,400
250(C)
19
63,300
250(1)
41
63,300
Annual Operating Cost (103 E/yr)
Fuel Costvu' 119.7
Less Base Case Fuel Cost -36.1
Investment-Related Cost^ 7.6
Effect on Boiler Operating 14.7
Total 105.9
Sulfur Removed (106 lb/yr)(e) 0.207
Control Cost ($/lb S Removed) 0.512
Con
(a)
(b)
(c)
(d)
trol Cost ($/10 Btu Output) 1.44
156.7
-47.2
7.6
19.3
136.4
0.271
0.503
1.42
1,105
-333
43
136
951
1.91
0.498
1.40
1
-
1
,603
483
43
197
,360
2.77
0
.491
1.38
Based on R. Schreiber, et al., EPA-650/2-74-123, November,
$1.38/10 Btu, based on analysis
14.55 percent of investment per
Based on average 1973 difference
of plant
producing
57
year, based on utility
in operating cost
,466
-
0.
-
^
1974,
bbl/D
financing
11.40
4.12
7.28
0101
12.83
- 4.64
8.19
0.0114
79.8
-28.9
50.9
0.0710
102.6 677
- 37.1 -245
65.5 432
0.0913 0.602
IU
nm
1,461
- 528
933
1.299
Section 6.
synthetic crude oil»
method
with
no maintenance.
(excluding fuel) between oil-fired
and coal-fired
boilers (0.20 mills/kWh from Electrical World, Nov. 1, 1973).
(e) Based on feedstock with 18,000 Btu/lb heating value and 0.2 weight percent sulfur.
(f) C and I indicate commercial and industrial boilers, respectively.
LO
CD
-------
139
employed:
(1) The existing coal-fired boiler system has been
equipped with a dust collecting system.
(2) Elemental sulfur and sulfuric acid were assumed to be the
by-products resulting from the Wellman-Lord and MgO
processes, respectively.
(3) The sludge generated from the FGD processes was
assumed to be filtered and disposed of in landfills.*
(4) Flue gases from FGD processes would be reheated if
necessary using an indirect steam reheat system.
(5) The retrofit factor was assumed to be 1.2 for the
double alkali system and 103 for the other processes.
(6) The costs for oil-fired boilers were estimated from
those for coal-fired with adjustments made with
respect to flue gas flow rate and sulfur input.
The summary of the costs of the FGD processes along with other
alternatives is shown in Tables 33 and 34 for coal- and oil-fired boilers,
respectively.
Physical cleaning of coal appears very attractive in its economics;
however, its application is limited to certain types of coal.
Processed fuels produced on a large scale are economically favored
over application of FGD processes to boilers in the small size classes. The
capital requirement to the boiler system is low and the annualized cost also
is low mainly due to the small fixed capital charges.
The FGD processes are economically favorable over other alternatives
for the NUC sources in the large size classes. Among the various FGD
processes, the MgO process with regeneration performed at a central station
appears most attractive economically. Throwaway processes in general are
more attractive than integrated regenerable processes if land is available
for sludge disposal.
-------
TABLE 33 . CAPITAL REQUIREMENT AND ANNUALIZED CONTROL COST OF ALTERNATIVES FOR COAL-FIRED BOILERS
C: Commercial Boiler . I: Industrial Boiler
Capital Requirement. $10
Alternative
Physical cleaning of
coal
Coal gasification
(low Btu)
Coal gasification
(high Btu)
Solvent refined coal
(solid)
Solvent refined coal
(liquid)
H-coal
Fluldlzed-bed combustion
(coal)
Wet lirestone scrubbing
(Peabody)
Line scrubbing (Bahco)
Double alkali (FMC)
XgO (Integrated)
MgO (central regeneration)
WelliLan-Lord
20 (C)
0
29.7
24.5
9.9
262
52.4
443
616
780
697
858
482
757
20 (I)
0
29.7
24.5
9.9
262
52.4
452
624
780
702
871
490
765
250 (C)
0
153
123
48.8
1,477
295
1.892
2,588
3,276
2,918
3.524
2,007
3,548
250 (I)
0
153
123
48.8
1,477
295
1,970
2,625
3,276
2,962
3,579
2,028
3,589
Annual! zed Control Cnst
Steam Output. 3/10° Btu
20 (C)
0.77
1.25
0.75
1.46
1.44
2.16
2.04
2.50
2.27
3.32
1.69
2.76
20 (I)
0.75
1.24
0.75
1.34
1.42
1.90
1.71
2.04
1.86
2.81
1-41
2.29
250 (C)
0.74
1.23
0.74
1.26
1.40
1.09
0.94
1.15
1.11
1.24
0.69
1.21
250(1)
0.73
1.22
0.74
1.16
1.38
0.93
0.79
0.87
0.87
0.95
0.52
0.92
20 (C)
0.29:
0.28
0.29
0.57
0.51
0.63
0.87
1.00
0.86
1.25
0.64
1.04
Sulfur Removal, $/lb
20 (I)
0.28
0.27
0.29
0.52
0.50
0.56
0.73
0.82
0.70
1.06
0.53
0.87
250 (C)
0.28
0.27
0.29
0.49
0.50
0.32
0.40
0.46
0.42 .
0.47
0.26
0.46
s
250 (I)
0.27
0.27
0.29
0.45
0.49
0.27
0.34
0.35
0.33
0.36
0.20
0.35
-------
TABLE 34. CAPITAL REQUIREMENT AND ANNUALIZED CONTROL COST OF ALTERNATIVES FOR OIL-FIRED BOILERS
C: Commercial Boiler I: Industrial Boiler
Annualized Control Cost
Capital Requirement. $10^
Alternative
Coal gasification
(high Btu)
Solvent refined coal
(liquid)
Wet limestone scrubbing
(Peabody)
Lfcne scrubbing (Banco)
Double alkali (FMC)
MzO (integrated)
KgO (central regeneration)
Wellman-Lord
2(C)
6.0
127
139
173
154
111
2(1)
6.0
127
139
173
154
111
20(C)
11
275
498
637
538
607
393
602
20(1)
11
275
503
637
540
612
398
607
250(C)
70
1,139
2,456
3,156
2,662
2,948
1,925
2,929
250(1)
70
1,139
2,497
3,156
2,708
, 3,021
1,956
2,986
2(C)
1.03
3.29
5.32
7.53
6.85
4.78
Pv
2(1)
1.01
2.99
4.91
6.89
6.34
4.46
Steam Output, S/106 Btu
20(C)
0.94
1.46
2.26
2.85
2.47
3.30
1.90
3.08
20(1)
0.93
1.28
1.89
2.30
2.01
2.79
1.60
2.58
250(C)
0.93
1.05
1.29
1.60
1.42
1.60
1.00
1.56
250(1)
0.92
0.83
0.69
0.84
0.77
0.89
0.53
0.86
By
2<;c) 2(1)
0.23 0.22
2.72 2.48
4.40 4.06
5.84 5.34
5.04 4.47
3.52 3.29
Sulfur Removal,
20(C)
0.21
1.21
1.87
2.21
1.82
2.43
1.40
2.27
20(1)
0.21
1.06
1.56
1.78
1.48
2.05
1.18
1.90
$/lb S
250(C)
0.21
0.86
1.06
1.24
1.04
1.17
0.73
1.14
250(1)
0.2
0.6
0.57
1.72
0.57
0.65
0.39
0.63
,_,
M
(U
3
CL
M
NJ
-------
PART III
PACKAGEABILITY OF SORPTION PROCESSES
-------
144
SURVEY OF EXISTING PACKAGE SORPTION SYSTEMS
A "package system" is loosely defined here as a complete, compact,
factory assembled, and easily transportable unit or combination of units,
and/or a unit which can be assembled on site from a few prefabricated and
readily transportable components. Such systems are characterized by:
(1) small size, (2) complete integral components, (3) simple and easy install-
ation, (4) ready availability, (5) ease of operation and maintenance, and
(6) low capital and operating costs.
Package systems in general have been widely used in many areas such
as household appliances, wastewater treatment plants for small establishments,
and solvent recovery and air purification from exhaust and vent gases.
In order to obtain information on existing package sorption systems,
if any, applicable to removal of SO from small commercial and industrial
X
boilers, a literature search was conducted extensively using information
systems such as National Technical Information Service (NTIS), Stack Gas
Control Coordination Center (an organization within Battelie-Columbus),
Battelle's Energy Information Center base, Engineering Index System, Atomic
Energy Commission (AEC) system, and Battelle-Columbus libraries. In addition,
manufacturers of 'sorbent materials and engineering and fabrication companies
producing sorption systems were contacted to acquire the field information on
existing package sorption systems for SO control.
X
The survey results indicate that for the prupose of recovering
solvents and purifying circulating air, activated carbon is used extensively
mainly because it adsorbs all types of organic vapors and mists regardless of
variation in concentration and humidity. Factory assembled solvent recovery
systems that are essentially off-the-shelf units are commercially available
up to 10,000 scfm flow capacity. Three mechanically different types of
systems, fixed bed, moving bed, and fluidized bed, are being utilized.
However, as Tables 35 and 36 indicate, currently there is no package sorption
system commercially available for SO control. This is attributed to the fact
X
that the marketability of such a system is low since industry or a potential
buyer would not invest its capital where there is no return on investment
or where there is no OSHA requirement. In addition, according to the opinion
-------
TABLE 35. SURVEY OF ENGINEERING AND FABRICATION COMPANIES OF SORPTION SYSTEMS
Company
Sorption
Systems
Package Sorption
System for SO
Remarks
American Air Filter
Company, Inc.
Day and Zimmerman,
Inc.
Mine Safety
Appliances Company
1. Melsheimer
Company, Inc.
Vulcan Cincinnati,
Inc.
Cambridge Filter
Corp.
Vic Manufacturing
Company
Farr Company
Howard S. Caldwell
Company
Detrex Chemical
Indus tries, Inc.
Lime and limestone
slurry
Carbon res orb
system
Carbon based air
purification
Multistage beds for
air purification
Solvent recovery,
stationary carbon
bed
Carbon based air
purification
Solvent recovery
Carbon based air
purification
Carbon based air
purification
Solvent recovery
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Custom design only
Custom design only
Panel and pleated carbon
beds
California Carbon Company
provides regeneration
of carbon
Custom design
Custom design; small packaged
system available
Regeneration at California
Carbon Company
Custom design of compact
system
Ui
-------
TABLE 35. (Continued)
Company
Sorption
Systems
Package Sorption
System for SO
Remarks
SSE, Inc.
Conner Engineering
Corp.
Hoyt Manufacturing
Corp.
Carbon panel air
purification NA
Carbon panel and radial
flow canister air
purification NA
Solvent recovery NA
Package system for air
purification
Standard system available
Cabinet enclosed system
-------
TABLE 36. SURVEY OF SORBENT MANUFACTURERS
Activated Noncarbon Package Sorption
Company Carbon Sorbent System for SO Remarks
X
Borg-Warner Corp. Potassium permangated
impregnated alumina
Barnebey-Cheney Company X
Sude Chemie U.S.
Company X Zinc oxide catalyst
C. H. Dexter Corp. X
American Norit Company,
Inc.. X
Witco Chemical Company,
Inc. X
Aluminum Company of Granular activated
America alumina
Husky Industries X
Davison Chemical Silica gel
Molecular sieve
ICI United States X
Westvaco X
Union Carbide Corp.
at Fostoria, Ohio X
Linde Company Molecular sieve
Calgon Corp. X
NA(a)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Sold the plant to H. E. Burroughs &
Assoc.
Granular carbon
Catalyst for H^S sorption
No interest in SO control
X
Activated carbon for air conditioning
equipment
Powdered carbon for water treatment
Carbon for liquid phase adsorption
Regenerable FGD process
Purasiv S system for sulfuric acid
plant
SO control process similar to
Sulfacid
(a) NA - not available.
-------
148
survey, activated carbon, which has been used extensively in package sorption
systems, would not be adequate for SO control since its sorption capacity
X.
for SO is relatively small (i.e., generally less than 10 percent by weight)
X
and the regeneration requires a special feature for handling sulfurous regen-
eration products. The central regeneration mode of the operation also requires
the transportation of bulky activated carbon, and thus, a dry activated carbon
process is not deemed feasible for a package system to control SO „
X
-------
149
SURVEY OF SORBENT MATERIALS
The potential package sorption systems were,surveyed by reviewing
the applicability of sorbent materials to the removal of SO- from boiler flue
gases and examining the current development of various flue gas desulfurization
(FGD) processes using the sorbent materials.
Sorbent materials which can be used to control SO may be classified
:x J
into solid sorbent, aqueous solution, and organic liquid. The description of
each of the sorbent materials follows.
Solid Sorbents
A number of solid materials react with sulfur dioxide under
suitable conditions and can be utilized to remove it from flue gases. The
most Important requirement for a solid sorbent, in addition to a high
affinity for sulfur dioxide, is that it has a large surface area. The
area can be increased by either granulating the solid or making it very
porous. In granular form the sorbent can be contacted by the flue gas as
a fluidized or gravitating bed or as entrained particles. The gas may
contact the interior of porous solids by diffusion through the pores. Non-
porous sorbents may be modified to give increased exposure to the gas by
impregnating them on more porous inert materials.
Most solid sorbent processes are dry and have the advantage
that the flue gas is not cooled. Unlike aqueous solutions, little, if any,
solid or liquid wastes are generated, eliminating the problem of waste
disposal.
With the exception of limestone, most solid sorbents are expen-
sive and must be regenerated. In addition, solid sorbents tend to lose
their activity due to contamination and plugging of the sorbent pores by
solid impurities in the gas stream.
Dry Limestone. In the dry state and at ordinary temperatures,
neither limestone (CaC00) nor lime (CaO) react well with SO-.' ' At high
temperatures, however, on the order of 1800°F, the CaCO_ is quickly
calcined to CaO and readily reacts with the S0«.
-------
150
Limestone is very abundant and inexpensive. There are, however,
intrinsic disadvantages. That is, the low porosity of the calcined lime-
stone particles and the narrowing of the exterior pores due to the
formation of CaSO, result in a low degree of SO- adsorption and consequently
a low removal efficiency.
Activated Carbon. Activated carbon may be utilized as a 'sorbent
for low temperature (less than 300°F) adsorption of SO. from flue gases.
The SO- diffuses into the pores of the carbon matrix and is catalytically
oxidized to SO- by contacting active carbon sites. The presence of water
in the flue gas converts the SO- to sulfuric acid freeing the active sites
for further adsorption.
The acid may be removed from the carbon structure by washing
with water or alkaline solution, by thermally desorbing with a hot scavenger
gas, or by reducing with a reducing gas such as H S. Water washing will
produce a stream of weak sulfuric acid, while thermal desorption, conducted
at 750°F, utilizes the carbon as a reductant to generate a concentrated
S02 off gas. Reduction with H-S will generate elemental sulfur.
Carbon adsorbent may be manufactured from a variety of carbon-
aceous materials. Charcoal and semicokes prepared from coal, lignite,
and peat are suitable adsorbents. Lignite is more applicable to a dry
process employing thermal regeneration. Upon successive adsorption-
regeneration cycles, the porosity and activity of the lignite tend to
increase.
Two primary drawbacks to all carbon-based systems are the
limited capacity and low gas velocity requirements. The sorbent capacity
is generally only about 2-10 percent sulfur by weight. This necessitates
the use of large quantities of carbon adsorbent. In addition, the rate
of adsorption of SO on carbon is limited to the rate of SO- diffusion
into the pores. Thus, to allow time for the diffusion the gas velocities
must be limited to 1-4 feet per second. Carbon, however, has several
characteristics which make it an attractive sorbent for SO-.
-------
151
(1) Carbon is easily regenerated and is amenable to a
process utilizing separate independent adsorption .
and regeneration options.
(2) Carbon-based processes are generally simple.
(3) The regeneration operations are not complicated by
side reactions.
(4) In dry carbon processes, adsorption may be carried
out at air preheater temperatures (300°F), thus no
flue gas reheat is required.
(5) No waste streams are generated.
(6) In wet carbon processes preceded by adequate particle
removal equipment, the sorbent has a long, almost
indefinite life.
Metal Oxides. The oxides of Ifci, Fe, Cu, and Zn will react strongly
with S0? at elevated temperatures and have been investigated for utilization
as sorbents to remove S02 from flue gases. " ' They have several charac-
teristics that make them applicable to flue gas desulfurization systems.
(1) High affinity for SQ2.
(2) Adsorption process is dry, at air heater temperature of
700°F, eliminating need for flue gas reheat„
(3) No generation of waste streams.
(4) Regenerable.
Of the above sorbents, the Mn and Cu oxides have been found to
be the most promising. According to the studies done by the United States
Bureau of Mines and the TVA, MnO was superior to other oxides of metals
for absorbing S0». The process, however, involves a complicated regenera-
tion process.
Cupric oxide is presently the only metal oxide sorbent being
actively tested in the United States. Sulfur dioxide and oxygen, upon
contacting the cupric oxide, react rapidly with it to form CuSO,. The
optimum operating temperature is about 700°F. Cupric oxide granules do
not have sufficient porosity to function as an effective sorbent. To
increase the porosity, they are impregnated on an activated alumina
acceptor. The optimum copper content of the acceptor appears to be 4-6
percent by weight.
-------
152
The sorbent/acceptor is expensive and must be regenerated.
Moreover, several problems, Intrinsic to the sorbent, dictate the design
and operating conditions of the absorption-regeneration system. At high
temperatures or under thermal stress, the activated alumina is attacked
by SO , forming A1-(SO,)_, which stresses and weakens the alumina
structure. To minimize the thermal stress, the absorption and regeneration
reactions must operate at similar temperatures. This is why thermal
regeneration, which requires a temperature of 1300 F, would not be
acceptable. Regeneration by reduction, however, requires a much lower
temperature. The CuSO, may be contacted by hydrogen or methane at 750 F
to regenerate the copper and drive off a concentrated stream of SO ^ °
Physical movement or transport of the acceptor results in
degradation and disintegration of the matrix structure. Moving bed or
continuous flow systems would not be acceptable as they would subject the
sorbent to excessive movement and vibration. A fixed bed system would be
required. ,
Tests conducted on a Shell pilot plant using reduction regeneration
and fixed, parallel beds indicated an acceptor life of about 1.5 years.
The pore structure is easily contaminated and plugged by parti-
culate matter. To help alleviate this problem the absorber must be
preceded by a highly efficient electrostatic precipitator.
Another problem associated with the regenerable cupric oxide
absorbent is the production of sulfur dioxide gas. It is an intermediate
by-product and requires processing to convert it to a more marketable
form such as sulfur, sulfuric acid, or liquid SO.-
Oxidation Catalyst. The catalytic oxidation process for sulfur
dioxide removal from flue gases is based on the same concept as the contact
process for manufacturing sulfuric acid. The best known catalyst is
vanadium pentoxide impregnated on an alumina support. At the optimum
o
operating temperature of 800-900 F, about 90 percent of the SO^ is converted
to S0» on the catalyst. Following oxidation the gas is passed through an
absorption tower where the S0» is removed by contacting a countercurrent
solution of sulfuric acid.
-------
153
The vanadium pentoxide based process has some unique advantages
(72)
over sorption-regeneration processes.
(1) It is a simple process requiring only two steps.
No regeneration is involved.
(2) No waste side streams are produced.
(3) Except for catalyst.replacement, no raw materials
are consumed.
(A) The temperature of the gas leaving the absorber is
250°Fj no flue gas reheat is necessary.
The catalytic oxidation process, however, is accompanied by some
operational difficulties.
(1) The catalyst is easily plugged by particulate
matter.and requires frequent replacement. The
catalyst is expensive and removal from the con-
verter for cleaning generally results in
attritional losses.
(2) The catalytic converter must be operated at a
high temperature. A retrofit process would
require flue gas preheat facilities and heat
exchangers to minimize energy losses.
Molecular Sieve. Molecular sieves can be utilized as effective
adsorbents for the removal of SO from flue gases. The molecular sieve
most applicable to SO- removal is a crystalline metal aluminosilicate. It
is highly porous and selectively adsorbs polar molecules like SO and
H20. <">
Sulfur dioxide removal with molecular sieves is a relatively
simple process, requiring two steps, adsorption and regeneration. The
removal efficiency is high; for example, in sufficient amounts the molecular
(73)
sieve can remove greater than 99 percent of the SO,, from the flue gas.
i.
However, problems, intrinsic to the nature of the sorbent,
eliminate it as a process applicable for boiler flue gas treatment.
(1) The molecular sieve adsorbs water vapor with the
SO,,. Since water vapor in boiler flue gas streams
is considerably greater in concentration than the
-------
154
SCL, the sieve would be quickly contaminated or
loaded with water. An additional sieve would
have to be installed ahead of the SO^ adsorber to
selectively remove the water vapor.
(2) The sorption capacity of a molecular sieve for S02
is decreased substantially with increases in
temperature. The loading can vary from 6 percent
S02 by weight at 105°F to 13 percent at 50°F.
For this reason, it is mandatory that the adsorp-
tion, temperature be maintained as low as possible.
This is difficult for boiler flue gas streams
which are generally at about 300 F. An acid
resistant heat exchanger would have to precede
the adsorber.
(3) Other problems associated with the process are
the necessity for flue gas reheat and installation
of facilities to process the SO. off gases.
Alkalized Alumina. Alkalized alumina (NaAlO-) is,an effective
absorbent for SO.. At 570°F to 660°F the sulfur dioxide readily reacts
with the alkalized alumina to form sodium sulfate and Al-Oy The alumina
is included to provide increased porosity for the sodium salt and to
assist in the regeneration process. During regeneration the alumina
provides an anion to combine with the Na and reduces the reaction energy
requirements. Regeneration is accomplished by reduction with H. gas at
1350°F. The H.S off-gas can be utilized in a Glaus reactor for the
production of elemental sulfur.
Alkalized alumina has several characteristics which make it an
attractive sorbant for S02 removal systems.
(1) The absorption kinetics favor reaction with the
SO at near flue gas preheater effluent
temperatures. Retrofit is easy and no flue gas
is necessary.
(2) It has a high affinity for £0-. Ninety percent
removal efficiencies are attainable with a
reasonable amount of absorbent.
(3) The sorption capacity is high, about 15-20 percent
sulfur dioxide by weight.
(4) There are no adverse side reactions or waste
streams.
-------
155
The sorbent does, however, have some drawbacks. It is very
costly and requires regenerating. The fragile crystal structure degrades
rapidly from the mechanical recycling operation to and from the absorption
and regeneration vessels and the thermal stress created by the high
regeneration temperatures. Operating costs, resulting from the replacement
of the disintegrated sorbent, are estimated to be very high. '
Another problem is the generation of the H_S off-gas. It
requires processing with SO. in a Claus reactor to produce elemental
sulfur.
Organic Solids. The application of organic solids processes
for the removal of sulfur dioxide has received relatively little attention
as compared to other processes. This is primarily due to the inherent
properties of most solids, i.e., low adsorption capacity and adsorption
rates, poor thermal stability and high cost.
Laboratory studies were conducted on the suitability of organic
ion exchange resins for sulfur dioxide removal. However, the adsorption
rates were slow and could not compare with that of molecular sieves or
carbon. Other studies indicated that successive adsorption-thermal ;
regeneration cycles lead to the decrease of the adsorption capacity of
, (76)
the resin.
A study examining nitrogen containing polymers, incorporated
into melt-spun fibers, for use as adsorbents revealed that styrene-
dimethylpropylmalimide was the most applicable. 'It was subject to
a build up of sulfate, however, and its adsorption capacity decreased
with each adsorption-regeneration cycle.
An investigation by TRW on several organic solids showed that
waste newsprint can adsorb up to 10 percent of its weight in sulfur
dioxide. The adsorption rate and resultant flow velocities were very
slow hov/ever.
Aqueous Solutions
Aquaous solutions have received the most attention for removing
(79 80^
SO. from flue gases. ' They may be classified into three different
categories: slurry solutions, clear solutions, and weak acid solutions.
-------
156
Slurry Solutions. Slurry solutions involve the use of a 5-15
percent lime or limestone slurry to absorb SO.. The spent absorbent,
CaSO and CaSO,, is transferred to a disposal facility and discarded as waste.
The attractive feature of the lime/limestone based scrubbing
system is that the low cost of the absorbent eliminates the need for
regeneration facilities. Alternately, provisions must be made to
dispose of the CaSO- and CaSO waste.
The lime or pulverized limestone slurry contacts the SO- in
the scrubber and reacts to form CaSO- and CaSO,. Although the reaction
products are the same for lime or limestone, lime is a more effective
adsorbent than limestone. This may be because the CaCO- particles
formed from lime are smaller than the limestone particles and hence have
more surface area and greater reactivity. Moreover, the calcium from
lime is already in solution, whereas the calcium from the natural CaCO_
is not, and must undergo dissolution to calcium ion.
Magnesium in the form of dolomitic lime or limestone may be
added to either system to improve the absorption efficiency. The higher
solubility of MgSO- permits higher ion concentrations of absorbents in
the solution. Pilot plant tests have demonstrated that a higher S02
removal and improved operation and maintenance are possible by adding
/oi\
magnesium into the scrubbing solution. v '
One critical problem with lime/limestone slurry scrubbing is
the deposition of solids on surfaces in the scrubber and associated equip-
ment from crystallization of calcium sulfite or calcium sulfate. The
solubility of calcium sulfite is very sensitive to pH variations and
decreases with increasing pH. The liquor entering the scrubber is
saturated with calcium sulfite. As the liquor proceeds through the
scrubber, SO is absorbed, the pH decreases, and CaCO- dissolves and
reacts with SO to form CaSO-. The newly formed calcium sulfite either
supersaturates or oxidizes to CaSO,. Any calcium sulfite in excess of
the saturation equilibrium concentration formed during absorption must
be removed in the hold tank prior to recycling to the scrubber. This
-------
157
is accomplished by raising the pH with the addition of lime or limestone.
If the slurry recycled to the scrubber contains supersaturated sulfite,
the potential for scaling exists.
Calcium sulfate is formed from the oxidation of calcium sulfite.
Unlike calcium sulfite, the sulfate is not precipitated from solution by
the rise in pH that occurs in the hold tank and, consequently, accumulates
until it is supersaturated at which time it crystallizes resulting In
scaling of the scrubber internals. Three of the several methods which
have been employed to alleviate the supersaturation conditions are:
(a) adding seed crystals to the hold tank to maximize crystallization,
(b) providing maximum residence time for the slurry in the hold tank to
enable the sulfate to precipitate, (c) adding magnesium, as MgO, to form
the more soluble MgSO, and reduce the saturation of the sulfate, and
(d) keeping oxidation of sulfite to sulfate at a low level.
An important consideration, with coal fired boilers is the
influence of chloride on scrubbing performance. The chloride enters the
scrubber with the flue gas as HC1 and because it forms no insoluble com-
pounds with calcium, it accumulates in the scrubber slurry. It can only
leave the system with the liquid purge, i.e., filter cake moisture or
clarifier underflow. Since the liquid purge from a closed system is
small, high concentrations, <5000 ppm, of chloride can exist in the slurry.
The addition of chloride lowers the pH of the scrubbing solution.
Since oxidation increases with decreasing pH, an increase in chloride ion
/Q1 \
increases the sulfate content. v ' The chloride ion also increases the
calcium ion concentration resulting in decreased dissolution of calcium
sulfite. Consequently, the calcium sulfite available for reaction with
sulfur dioxide decreases, resulting in reduced scrubbing efficiency and
forcing the utilization of higher liquid to gas ratios.
The generation of huge volumes of calcium sulfite and calcium
sulfate sludge poses an additional problem. The sludge is of little
commercial value and must be disposed of as a waste product. The most
environmentally sound methods for sludge disposal are landfilling of
chemically fixed sludge and disposal of untreated sludge in ponds lined
with an impervious material such as clay, plastic, or rubber. However,
-------
158
calcium sulfite presents a significant land use problem. Sulfites tend
to crystallize into small, thin platelets which settle to a loose bulky
structure that may occlude a relatively large amount of water. The
sludge is difficult to compress and dewater, and conversion to a suitable
landfill presents an expensive and formidable problem. Ponds on the
hand require a large, suitable disposal site close to the plant and may
be not only structurally unstable but aesthetically objectionable.
Other problems associated with wet lime/limestone slurry
scrubbing are
(1) Flue gas reheat is necessary
(2) For limestone systems, limestone grinding facilities
must be installed or provisions must be made to
purchase the limestone in pulverized form
(3) The abrasive slurry solutions may cause pump and
equipment erosion problems.
Clear Solutions. Clear scrubbing solutions involve the use of
alkali absorbents to remove sulfur dioxide from flue gas. The most
common alkali absorbents are sulfites of ammonia, sodium, potassium, and
lithium. The more popular scrubbing solutions are ammonium sulfite and
sodium sulfite. Ammonia is reasonably priced, permits a variety of
regeneration procedures, and in the sulfate form, has commercial value
as a fertilizer. Sodium, although higher priced than ammonia per
equivalent, is nonvolatile, eliminating the problem of absorbent vola-
tilization in the scrubber. Potassium and lithium solutions have also
been used but they are more costly than the ammonia or sodium absorbents.
The sulfur dioxide reacts with alkaline salt in the form of
sulfite to produce the bisulfite. In cases where the starting material
is a metal oxide or hydroxide, its reaction with sulfur dioxide forms
the sulfite first followed by conversion to the bisulfite.
-------
159
Clear solutions have three characteristics which make them
attractive absorbents for SO^.
(1) They have a high affinity for SO and are
generally capable of greater than a 90 percent
removal efficiency.
(2) All of the compounds are highly soluble and
remain in solution minimizing plugging and
scaling problems associated with other wet
processes like lime or limestone.
(3) The sorption capacity is high, resulting in
the use of less amounts of sorbent in the
operation.
Alkali absorbents are not inexpensive and thus should be either
regenerated or transformed into a marketable by-product. Some methods
generally employed to process the spent absorbent are: ^
(1) Direct thermal treatment - The sulfite-bisulfite
chemistry is applicable to an adsorption-thermal
regeneration type system. The bisulfite is rela-
tively unstable and SO- can be desorbed from it
relatively easily. A problem with thermal
regeneration is the large amount of heat consumed.
Most clear solutions contain less than 20 percent
salt by weight and consequently the heat supplied
must be expended in evaporation of water.
(2) Acidification - This process is applicable to
ammonia scrubbing systems. Acidulation of the
ammonium bisulfite and sulfite with H SO,, HNO-,
or H-PO, will produce SO. and either ammonium
sulfate, ammonium nitrate, or ammonium phosphate
as a fertilizer by-product.
(3) Reduction - Direct reduction of the sulfite with
H_S will regenerate the absorbent and form elemental
sulfur. The United States Bureau of Mines Citrate
process bubbles H?S through the spent absorbent
followed by separation of the by-product sulfur
and recovery of the citrate solution for recycle
to the scrubber.(82)
(4) Oxidation - This process also is applicable to
ammonia systems for generating a marketable by-
product. Air may be bubbled through a spent
ammonium sulfite, ammonium bisulfite solution
to oxidize it to ammonium sulfate. The product
may be evaporated, dried, and marketed as fertilizer.
-------
160
(5) Lime or limestone regeneration - This process has
been applied to sulfite buffer systems where lime or
limestone is added to the spent sodium sulfite,
sodium bisulfite solution to regenerate the sodium
sulfite and remove the SO^ as insoluble CaSO_.
The FMC and GM double alkali processes utilize
this method.(83> 84.)' .
There are some characteristic disadvantages of clear solutions :
(1) The sulfite tends to oxidize to the sulfate.
The sulfate is very stable, unreactive, and not
easily reconverted to the sulfite.
(2) The processes using clear solutions are wet
processes and cool the flue gas down to 120 to
140 F. To increase buoyancy and eliminate the
plume, a flue gas reheat system is necessary.
(3) The thermal and acidification-based regeneration
systems require an accompanying sulfur dioxide
processing facility.
Weak Acid Solutions. An aqueous process'not involving the use of
alkali salts or calcium compounds utilizes a ferric sulfate catalyst in
solution to oxidize S02 absorbed as H2S03 to H SQ such as in the Chiyoda
process. The weak acid can be neutralized with limestone to form insoluble
gypsum.
The use of the weak sulfuric acid as a sorbent offers several
(85) -
advantages:
(1) It lends itself to an easy to operate, uncompli-
cated system. Only three processes are involved:
absorption, oxidation, and neutralization.
(2) The ferric sulfate catalyst concentrations are
weak, only 2000 ppm, and the loss with the gypsum
is negligible.
(3) The sorbent will not plug, scale, or erode the
equipment.
However, the absorption capacity of water for sulfur dioxide is
quite low. To effectively absorb the sulfur dioxide, very high liquid
flow rates must te used. Normal liquid to ga's ratios (L/G) are on the
order of 300. This implies that a very large absorber is required.
Furthermore, the sorbent is highly corrosive so all process equipment
must be constructed of expensive stainless steel.
-------
161
Organic Liquids
Organic liquids have found only very limited application as
sulfur dioxide sorbents. Although most amines, alcohols, amides, imides,
ketones, and esters are capable of absorbing sulfur dioxide, only two,
dimethylaniline and xylidene, have been used on limited industrial scale.
Both applications are on smelter gases containing about 4-6 percent- sulfur
dioxide.(86)
One problem in common to most liquid organic sorbents is the loss
of the purified gas stream. Although the sorbent may have a low vapor
pressure, significant losses can still occur if a large volume of gas is
treated. Organic liquids are costly, and thus, the loss can contribute
substantially to the operating costs. Furthermore, the loss of the
organic sorbent to the atmosphere can pose a potential pollution threat.
-------
162
DESCRIPTION OF SORPTION PROCESSES
The candidate sorption processes evaluated .with respect to the
possibility of being a package unit for small,industrial and commercial
boilers included
(1) Throwaway Process .• • '
Limestone slurry (Peabody)
Lime scrubbing (Bahco)
Double alkali (GM and FMC)
(2) Regenerable Process (central regeneration)
MgO process - central regeneration
(3) Regenerable Process (on-site regeneration)
Chemiebau
Foster Wheeler
Westavco
Sulfacid
Chiyoda
Ammonia process (Peabody and Catalytic)
MgO process - integrated
Wellman-Lord
Shell FGD
Citrate (Morrison-Knudsen)
Calsox (Monsanto)
Aqueous carbonate (Atomic International)
The description of the following processes was not included in
this section since they have been included in Part II of this report.
Limestone slurry (Peabody)
Lime scrubbing (Bahco)
Double alkali (FMC)
MgO process (central regeneration)
MgO process (integrated)
Wellman-Lord
Below are brief process descriptions; more complete descriptions are
included in Appendix B.
Double Alkali Process (GM)
General Motors Corporation (GM) has developed one of a number of
double alkali processes that scrub with a Na-SO- buffer solution and then
(87^
react the clear solution with lime or limestone to precipitate CaSO~.
Like FMC Double Alkali processes, the purpose of separating scrubbing from
precipitation is to eliminate scaling difficulties. Unlike the FMC process,
-------
163
however, the GM process employs a weaker solution and thus oxidation may be
more significant than in the FMC process. The removal efficiencies of SC^
and particulate matter are about 90 percent and 83 percent, respectively.
A full-scale facility on a combined coal-fired combustion source equivalent
to 400,000 Ib steam/hr capacity in Parma, Ohio, has been in operation since
March, 1974.
Chemiebau Process
Reinluft GmbH, Essen, Germany, developed this process and Common-
wealth Associates, Jackson, Michigan, has acquired the Western hemisphere
licensing rights. The process employs moving beds of lump char to remove
SO, from flue gas. The deactivated adsorbent is thermally regenerated,
/QQ 8Q^
producing a concentrated 20 percent SO- gas stream. ' Since char is
lost due to mechanical attrition and reaction with SO.,, the makeup char
must be added to the system at a rate of 20 Ib/lb of SO- removed to replace
these losses, A heat requirement of 5,000 to 6,000 Btu/lb of SO- removed
has been estimated for the thermal regeneration of the deactivated char.
The removal efficiency of S0» ranges from 85 to 95 percent. To date no
Chemiebau process has been sold in the United States.
Foster Wheeler (FW)
The FW process for S0? removal is a combination of char adsorption
and regeneration processes developed by Bergbau Forschung, GmbH and elemental
sulfur conversion process developed by Foster Wheeler, who currently markets
the process in the United States. After treatment for removal of particulate
matter, the flue gas is introduced into the adsorber which contains vertical
(90 91)
parallel louver beds through which the char flows. ' The flue gas
passes through the adsorber bed in a cross flow. The deactivated char is
regenerated thermally by mixing with hot sand in a fluidized bed and S09 gas
is liberated. The concentrated SO- gas stream is directed to the Foster
-------
164
Wheeler off-gas treatment system where SO reacts with crushed coal and is
reduced to elemental sulfur. The process is capable of removing 86-95 percent
of the S0_, 90-95 percent of the particulate matter, and 40-60 percent of
the NO . About 0.14 Ib of char/lb of S09 removed is required to make up the
X ' ^
losses primarily due to the production of CO- in the regeneration. Fuel is
required to heat the sand to 1500°F. About 0»5 Ib of elemental sulfur is
produced per pound of SO. removed. A test program is underway for a demon-
stration unit on a 47.5 MW coal-fired boiler at Gulf Power Company's Scholz
Steam Plant.
Westavco Process
Westavco, Charleston Heights, South Carolina, has developed and
markets the process. After treatment for removal of particulate matter, the
flue gas is introduced to an activated carbon fluidized bed where SO- is
removed through catalized oxidation to SO- and a subsequent hydrolysis to
sulfuric acid which remains adsorbed in the carbon particle. The acid loaded
carbon is contacted by a stream of H9S which reduces the sulfuric acid to
(92 93)
elemental sulfur. ' The system is capable of removing 90 percent S09
from flue gas. Some carbon is lost due to mechanical attrition, normally
less than 1 percent per cycle. Fuel oil is consumed in the H-S generator/
sulfur stripper to raise the temperature up to 1200 F. Hydrogen must be
produced from a coal gasifier to generate H^S. Westavco has recently
3
completed pilot plant tests on a 20,000 ft /hr flue gas stream from an
oil-fired boiler and they are interested in evaluating the system on a
coal-fired boiler in an increased scale, i.e., 15 MW.
Sulfacid Process
Lurgi of Frankfurt, West Germany, has developed the process and
the Rust Engineering Company, Birmingham, Alabama, markets the system in
the United States. Stack gas is pretreated to adjust the temperature,
humidity, and fly ash content and the conditioned gas flows upward at low
velocity through a bed of carbon-based catalyst of 1 or 2 feet deep,, S09,
oxygen, and water are adsorbed on the impregnated carbon where sulfuric acid
-------
165
is formed. ' The acid is washed from the bed by a continuous spray
of water as it is formed. A SO removal efficiency of 90 percent can be
achieved by the process„ About 10 to 30 percent sulfuric acid is produced
as the by-product. If there is no market available for the acid, it should
be neutralized with limestone for disposal. The process has been commercially
applied to the treatment of chemical plant waste streams for several years in
Europe. A plant to handle 20,000 to 30,000 cfm of sulfuric acid tail gas
will be built in Pittsburgh, Pennsylvania, for the United States St.eel Company,,
Chiyoda Process
The Chiyoda Chemical Engineering and Construction Company, Japan,
developed the "Thoroughbred 101" sulfur dioxide process. Following treatment
by an ESP and a venturi type wet prescrubber the flue gas flows upward through
a packed bed absorber and contacts a countercurrent 2 to 5 percent sulfuric
acid solution, containing about 2,000 ppm of ferric sulfate»^ ' ' The
solution absorbs SO^ in the flue gas and flows to the oxidizer tower, where,
in the presence of the ferric ion catalyst, air injected into the liquor
oxidizes the H-SO, to H2SO,. A portion of the liquor leaving the oxidizer
returns to the absorber and the remainder passes to the gypsum production
steps where the liquor is neutralized with lime or pulverized limestone to
form insoluble gypsum which then is separated from the mother liquor by
centrifuging. The process is capable of a 95 percent SO. removal efficiency.
Ten commercial plants have been installed on Glaus plants and oil-fired
boilers in Japan. In the United States, a demonstration program for a 23-MW
pilot plant on Gulf Power Company's coal-fired boiler at Scholz plant in
Sneads, Florida, is under way.
Ammonia Scrubbing Process (Peabody)
The Peabody Engineering Company, Stamford, Connecticut, has developed
the process. After pretreatment to adjust fly ash content, temperature, and
hymidity, the flue gas is introduced into an abosrber where the gas is
-------
166
scrubbed countercurrently with a mixture of ammonia, ammonium sulfite,
(97 98)
ammonium bisulfite, and ammonium sulfate.N ' SCL in the gas reacts
with ammonia and ammonium sulfite in the solution to form ammonium bisulfite.
Spent liquor is pumped to the neutralizer where ammonia is added to convert
the bisulfite to sulfite to reduce sulfur dioxide loss during oxidation. The
neutralized solution is subsequently treated in an oxidizer where the sulfite
is reacted with air to form the sulfate which is pumped to a double.-effect
vacuum evaporator crystallizer. Greater than a 90 percent removal efficiency
of SO. can be obtained with the process and ammonium sulfate is the by-product,
Although the individual step of the process has been utilized in the ammonia
industry, the process has not been tested for flue gases from coal-fired
boilers.
Ammonia Scrubbing Process
(Catalytic, Inc.)
Catalytic, Inc., a subsidiary of Air Products and Chemicals, Inc.,
developed the process. ' The absorption steps of the process are
similar to those for Peabody ammonia scrubbing process. However, unlike
Peabody ammonia scrubbing process, this process utilizes the Institut Fran-
cais du Petrole (IFP) sulfur reducing process to regenerate the spent
ammonium salts and produce sulfur. In 1970 IFP began marketing its Glaus
tail gas treating process which involves ammonia scrubbing coupled with the
reduction-regeneration system. To date, seven installations have been
constructed and all are currently operating. The complete IFP S0« removal
system is being installed on a 35 MW utility boiler in France.
Shell Flue Gas Desulfurization
Process (Shell FGD)
Shell International Petroleum developed the process in the early
1960's and Universal Oil Products (UOP) purchased the licensing rights for
the United States. ' After treatment for removal of fly ash, the
flue gas is passed through an adsorber which is a fixed bed of elemental
copper supported on an alumina structure with open channels along the side.
-------
167
The elemental copper reacts with oxygen and S02 to form CuSO^. The deacti-
vated bed is arranged for regeneration by passing a stream of hydrogen
through the bed. The conversion of CuSO^ to Cu, H20, and S02 takes place.
A cyclic operation can be arranged with two or more identical adsorbers
for continuous processing of flue gas. The process can remove 90 percent
of the SCL from flue gas and hydrogen consumption is about 0.1 Ib/lb of
SCX, removed. To date, no commercial plant has been built in the United
States.
Citrate Process
The Uo S. Bureau of Mines developed the process and the Morrison-
Knudsen Company, Inc., Boise, Idaho, and Peabody Engineering Company, Stamford,
(QO\
Connecticut, independently offer the process on a commercial basis.v '
Basically the process involves absorption of SCL by a solution of sodium
citrate, citric acid, and sodium thiosulfate followed by reacting the absorbed
SO- with H-S to precipitate elemental sulfur and regenerate the citrate
solution. The major difference of the Morrison-Knudsen process and the Peabody
process is the flotation method for elemental sulfur separation. SO- removal
efficiency ranges from 95 to 99 percent. Small scale pilot plant testing
(1000 to 2000 cfm) is under way in Kellogg, Idaho, and Terre Haute, Indiana,
by the Morrison-Knudsen and Pfizer-McKee-Peabody, respectively.
Calsox Process
The Monsanto Company of St. Louis, Missouri, developed the process
which utilizes a dilute ethanolamine solution (0.5 weight 7») to absorb S0_
from flue gas/ ' Spent liquor is regenerated by using lime and calcium
sulfite and calcium sulfate sludges are formed. The process can achieve a
90 percent S0_ removal efficiency. The loss of high cost ethanolamine is
the main disadvantage of the process. A 3000 cfm pilot plant was operated
at a boiler owned by the Indianapolis Power and Light Company. To date no
large-scale commercial plant is available.
-------
168
Aqueous Carbonate Process
The Atomic International Division of Rockwell International
Corporation developed the process in the 1970's., The flue gas is introduced
into the spray tank dryer where it co-currently contacts an atomized mist
of a 4 to 20 weight percent Na_CO- solution. The SO- in the gas reacts
with the sodium carbonate to form sodium sulfite and sodium sulfates The
particulate matter is separated from the flue gas and either disposed of in
the open loop operation or regenerated in the closed loop operation. Open
loop operation is less costly in a small-scale operation, but the soluble
Na«SO~ and Na.SO, salts present a disposal problem. In closed loop operation,
the scrubbing and regeneration systems are independent and can be uncoupled
and operated separately. The regeneration system involves three steps:
reduction of sodium sulfite and sulfate to sodium sulfide with either coke
or coal, reformation of Na CO., from the sodium sulfide by dissolving in
water and treating with a CO--rich gas, and conversion of H^S resulting from
the second step to elemental sulfur by a Glaus process. The process can
achieve greater than 90 percent S0? removal. Although both scrubbing system
and the regeneration system have been tested separately, the complete aqueous
carbonate process has not been tested in a single installation.
-------
169
SORPTION PROCESS EVALUATION
Approach
The evaluation of the potential role of the stack gas cleaning
process as a form of package unit in the control of the emissions from
small stationary combustion sources requires consideration of a number of
diverse factors which must be related and compared in a meaningful fashion.
The approach taken to this evaluation involves the following steps.
(1) Development of evaluation criteria
(2) Evaluation of each stack gas cleaning process with
respect to each criterion.
The conversion of the evaluation to a rating scale would be desired for
the rating of the sorption processes based on the aggregate points. How-
ever, the procedure involves subjective judgements which would influence
the outcome significantly. The quantitative analysis of the evaluation,
therefore, was not conducted in this evaluation.
Evaluation Criteria
The characteristics of FGD sorption processes that are important
in determining the packageability of such systems for small-scale nonutility
combustion sources include:
(1) Size
(2) Installation
(3) Raw material availability
(4) Operation - maintenance
(5) Residual and secondary emissions
(6) By-products
(7) Capital requirement
(8) Annual cost
(9) Process availability.
-------
170
The overall size of the process is to be small enough for retrofit
to the existing boiler system. The sulfur dioxide sorption processes under
consideration have different configurations in design; and thus, the overall
size varies from one to another. This variability is expressed in terms of
size and number of components and space requirement in the installation.
The installation of a sorption process should be relatively easy
so that nonspecialized personnel may be able to put the system in-place
within a short period of time. This variability is evaluated with respect
to type of equipment and number of component.
Materials and supplies consumed by sorption processes should
be readily obtainable throughout the United States. This variability is
expressed in terms of the availability.
Sorption processes installed in small boilers are to be operated
by non-specialized personnel and maintained as trouble-free as possible.
Operational complexity is evaluated with respect to materials handling,
control requirement, and number of process steps, and the maintenance problem
is projected on the basis of the number of moving parts such as pumps,
valves, dampers, centrifuges, and vacuum filters, plugging and scaling
possibilities, and corrosion and erosion possibilities.
The sorption processes under consideration have differing po-
tentials for minimizing air pollutant emissions and generating new pollutant
emissions. This variability is expressed in terms of residual and secon-
dary emissions which result from the application of a sorption process.
In each case, cross-media emissions (i.e., air, water, and land) are
considered.
By-products resulting from the sorption processes may not be
significant considering the size of unit operation and the possible type
of by-products. If markets are not available for by-products, the by-
products would become wastes to be disposed of.
Capital requirement indicates the amount of capital required
to install a sorption process. The contribution of capital cost to annual
operating cost is incorporated in the annual cost.
-------
171
Annual cost consists of return on rate base, Federal income
tax, depreciation, and net annual operating cost. This indicates total
expense of sorption process resulting from capital requirement and net
operating cost.
In view of the urgency of related environmental problems, the
availability of a given sorption process is an important criterion in the
evaluation of its applicability. The factors of type of fuel tested, de-
velopment status, and operation assessment are components of the availability
consideration.
Process Evaluation
The next step in the evaluation procedure was to develop an
evaluation of each process with respect to each of the nine criteria.
A quantitative evaluation was employed wherever possible, otherwise
qualitative categories for evaluation were developed. The evaluations
with respect to capital requirement and annualized cost were not included
in this section but discussed extensively in the next section. The re-
sults of this evaluation are summarized in Table 37. This summary includes
18 sorption processes.
Size and Space Requirement
The important factors in this criterion include size and number
of packages in the shipment and space requirement in the installation.
It is difficult to obtain accurate information for the sorption pro-
cesses under consideration since many of the processes have not been
commercialized yet and, consequently, little experience is available in
the shipment and installation of the process systems. Therefore, in this
study, the possible size of packaged systems were characterized based on
the size of larger components and the number of all components excluding
off-site facilities such as storage tanks, grinding machines, and addi-
tional facilities necessary to process the by-product into a more marketable
form, i.e., an acid plant.
-------
TABLE 37. SOBPTION PROCESS EVALUATION HATRIX
Operation- Maintenance
Size and Space Requirement
Sorption
Process
Limestone slurry
(Peabody)
Lime scrubbing
(Bahco)
Double Alkali
(GM)
Double Alkali
(FMC)
MgO Process
(central
regeneration)
Chemiebau
Foster Wheeler
Westvaco
Sulfacid
Chiyoda
Ammonia
(Peabody)
Ammonia
(Catalytic, Inc.)
MgO Process
(Integrated)
Wellman-Lord
Shell FGD
Citrate
Calsox
Aqueous Carbonate
Size of
Vessel
Medium
Medium
Med ium
Small
Medium
Large
Large
Large
Large
Large
Medium
Medium
Medium
Medium
Medium
Medium
Med ium
Medium
Number of
Component
Moderate
Moderate
Moderate
Moderate
Moderate
Low
Moderate
Moderate
Low
Moderate
Moderate
High
High
High
High
High
Moderate
High
System
Area
Moderate
Moderate
Moderate
Small
Small
Moderate
Moderate
Moderate
Moderate
Large
Moderate
Large
Large
Large
Large
Large
Moderate
Large
Ease of
Installation
Moderate
Moderate
Moderate
Moderate
Easy
Moderate
Difficult
Difficult
Easy
Moderate
Moderate
Difficult
Difficult
Difficult
Difficult
Difficult
Moderate
Difficult
Raw Material Number o
Availability Processe
Available
Available
Available
Available
Available
Questionable
Available
Questionable
Available
Available
Questionable
Questionable
Available
Questionable
Questionable
Questionable
Available
Available
5"
4
6
5
5
5
4
4
3
7
8
7
9
10
8
8
6
6
f Technical
s Expertise
Technician
Technician
Technician
Technician
Technician
Engineer
Engineer
Engineer
Technician
Technician
Engineer
Engineer
Engineer
Engineer
Engineer
Engineer
Technician
Engineer
Material
Handling
Slurry
Slurry
Liquid
Liquid
Slurry
Solid
Solid
Solid
Liquid
Liquid
Liquid
Liquid
Slurry
Liquid
Ga£
Liquid
Liquid
Solid
Plugging
Scaling
High
High
Moderate
Minimum
Moderate
Minimum
Minimum
Minimum
Minimum
Minimum
Minimum
Minimum
Moderate
Minimum
Minimum
Minimum
Moderate
Minimum
Erosion
Corrosion
Moderate
Moderate
Low
Low
Moderate
Low
Low
Low
Moderate
Moderate
Low
Low
High
Moderate
Moderate
Moderate
Low
Moderate
Number of
Moving Operating
Parts Temperature
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Low
High
High
High
High
High
High
High
Medium
High
.Low
Low
Low
Low
High
High
High
High
Low
Low
Low
High
High
High
High
High
Low
High ,
-------
TABLE 37. SORPTION PROCESS EVALUATION MATRIX (Continued)
Emissions
Sorption
Process
Limestone slurry
(Peabody)
Lime scrubbing
(Bahco)
Double Alkali
(GM)
Double Alkali
(FMC)
MgO Process
(central
regeneration)
Chemiebau
Foster Wheeler
Westvaco
Sulfacid
Chiyoda
Ammonia
(Peabody)
Ammonia
(Catalytic, Inc.)
MgO Process
(Integrated)
Wellman-Lord
Shell FGD
Citrate
Calsox
Aqueous Carbonate
S02 Removal
Efficiency, Secondary
percent Emissions
70-90
70-95
90
90
80-90
80-95
85-95
90
80-90
85-95
90
90
80-90
90
75-90
85-99
90
90
Sludge
Sludge
Sludge
Sludge
None
None
None
None
Sludge
Chloride
purge
None
None
None
Sulfate
purge
Sulfate
purge
Sulfate
purge
Sludge
None
By-
products
None
None
None
None
None
Sulfur
Activated
Carbon
Sulfur
Sulfur
None
Gypsum
Ammonium
Sulfate
Sulfur
98% H2S04
Sulfur
Sulfur
Sulfur
None
Sulfur
Process Availability
Type of
Fuel
Tested
Coal
Oil
Coal
Coal
Coal
Coal
Coal
Oil
H2S04 Plant
Oil
HoSO, Plant
V 4
Oil
Coal
Oil
Oil
Coal
Coal
Synthetic
flue gas
Development
Status
Commercial
Commercial
Commercial
Commercial
Commercial
Pilot
Prototype
Pilot
Prototype
Commercial
Commercial
Pilot
Commercial
Commercial
Commercial
Pilot
Pilot
Pilot
Operation
Assessment
Poor- fair
Fair
Fair
Fair
Fair
Poor
Good
Good
Good
Good
Fair
Fair
Fair
Good
Good
Good
Good
Fair
-------
174
In assessing the space requirement a significant factor is the
area devoted to waste disposal facilities such as ponds or pits. These
facilities can contribute substantially to the total system area, and the
area is frequently larger than the required area for the sorption process
itself. In this study all processes requiring a pond for sludge or liquid
disposal were modified by adding surge or neutralizing tanks with filters
for the transformation into a cake.
Installation
The ease of installation depends on type and number of equipment
components and process flow in the overall operation. This criterion was
qualitatively evaluated with respect to three categories defined as follows
Category 1 - Relatively easy to install, simple and un-
complicated process, and moderate field
fabrication required.
Category 2 - Moderately difficult to install, more involved
process, and more field fabrication than in
Category 1.
Category 3 - Difficult to install, complex process with
more process steps and extensive piping, and
considerable field fabrication required.
Raw Material Availability
The availability of raw material required was evaluated on the
basis of two categories defined as follows:
Category 1 - Materials readily available and in surplus
generally throughout the United States, for
example, lime, limestone, magnesium oxide,
lignite, and soda ash.
Category 2 - Materials either in short supply or available
only to specific areas, for example, hydrogen,
ammonia, and methane.
-------
175
Operation - Maintenance
The operational and maintenance difficulties were assessed in
terms of number of process steps, technical expertise, characteristics of
materials handling, possibilities of plugging, scaling, erosion, and
corrosion, number of moving parts, and operating temperatures.
The number of process steps is based on the number of operation
or process steps that require monitoring. This, in general, includes
adsorption, regeneration, evaporation, stripping, thickening, centrifuging,
heating, drying, and other related processes.
The degree of technical expertise required is based on the process
complexity, control sensitivity, and operating conditions. It was expressed
in terms of technical knowledge equivalent to either technician or engineer.
The material handling was evaluated in terms of gases, liquids,
solids, and slurry. The handling of solids or slurry is more difficult
than that of liquids or gases in transportation and processing.
Some sdrption processes involve solid or slurry streams that
are more susceptible to scaling and/or plugging than others. Scaling and
plussing can precipitate equipment failure and result in operational dis-
ruptions. The potential was evaluated with respect to three categories
as follows:
Category 1 - Minimal possibility
Category 2 - Moderate possibility
Category 3 - High possibility.
In SCL sorption processes, corrosion is caused primarily by
the presence of dilute sulfuric acid and/or chlorine ions. Erosion is
caused by the abrasive nature of liquids and solids. Both corrosion and
erosion were evaluated with respect to three categories:
Category 1 - Minimal possibility
Category 2 - Moderate possibility
Category 3 - High possibility.
The number of moving parts was the summation of all of the
major pieces of equipment containing moving parts. This included con-
veryors, rotary drum filters, pumps, blowers, mixers, etc. This factor
was categorized in terms of low, moderate, and high.
-------
176
Operating temperatures influence the reliability of process
operation to some extent. A high operating temperature is more conclu-
sive to failure than a low one.
Residual and Secondary Emissions
The residual emission of sulfur dioxide was evaluated on .the
basis of the removal efficiency. The control of fly ash emission from
the existing control system was not taken into consideration in this study
except for the dry limestone injection process. The secondary emissions
resulting from the sulfur dioxide control process were expressed in terms
of pollutant and the quantity per pound of SCL removed.
By-Products
The by-products were evaluated with respect to marketability
and ease of handling. The following three categories were employed:
Category 1 - Moderate marketability and easy to handle:
elemental sulfur activated carbon
Category 2 - Moderate marketability and relatively diffi-
cult to handle: sulfuric acid
Category 3 - Poor marketability and easy to handle: gypsum,
ammonium sulfate
Process Availability
The process availability was evaluated on the basis of type of
fuel tested, development status, and operation assessment. The develop-
ment status described the state of development of each system as a nonpackage
unit. It was classified into four categories - bench, pilot, prototype,
and commercial. The operation assessment indicating the degree of successful
operation was classified into three categories - good, fair, and poor.
-------
177
COST OF SORPTION PROCESSES
The capital requirement and annualized cost of the various sorp-
tion processes under consideration were estimated for the boiler subgroups
of environmental concern described in Part II of this report. Since the
concept of the package unit may not be feasible for the large size- class
(i.e., the boiler size of 250,000 Ib/hr) probably due to the size limita-
tion, the NUC source class was excluded in the cost estimation.
The capital requirement employed in this study included costs
for equipment, material, installation, engineering and design, and startup
(battery limit cost). Although the capital requirement for a package unit
was preferred in this study, it was very difficult to obtain such informa-
tion because of lack of data. The base year for the estimation was mid-1973
and the Utility Financing Method listed in Appendix A was employed to esti-
mate the related costs.
The annualized cost included fixed capital charges, labor, utili-
ties, raw material, and by-product credit. The cost was estimated based on
the same format used in Part II of this report. The following assumptions
were made:
(1) The existing coal-fired boiler system has been
equipped with a flyash collecting system. The
Shell FGD System would need a more efficient
ESP system.
(2) The sludge generated from the throwaway processes
would be filtered and disposed of in landfills.
(3) Flue gases from FGD processes would be reheated,
if necessary, us.ing an indirect steam reheat
system.
(4) The retrofit factor was assumed to be 1.2 for the
double alkali systems, 1.4 for Shell FGD systems,
and 1.3 for other sorption processes.
(5) The costs for oil-fired boilers were estimated from
those for coal-fired boilers with adjustments made
with respect to flue gas flow rate and sulfur input.
-------
178
The summary of the estimations is shown in Tables 38 and 39 for
coal- and oil-fired boilers, respectively. Among the various sorption
processes, the MgO process with regeneration performed at a central facility
appears most attractive economically. Both the capital requirement and
annualized cost are relatively low compared with those for other sorption
processes. Throwaway processes in general are low in capital requirement
and annualized cost than regenerable processes with on-site regeneration
facilities.
-------
TABLE 38. CAPITAL REQUIREMENT AND ANNUALIZED COST OF SORPTION
PROCESSES FOR COAL-FIRED BOILERS
C: Commercial Boiler
I: Industrial Boiler
Annualized Cost
Capital Requirement,
$103
Sorption Process
Limestone slurry (Peabody)
Lime scrubbing (Bahco)
Double alkali (GM)
Double alkali (FMC)
MgO (central regeneration)
Chemiebau
Foster Wheeler
Westvaco
Sulfacid
Chiyoda
Ammonia scrubbing (Peabody)
Ammonia scrubbing
(Catalytic, Inc.)
MgO (integrated)
Wellman-Lord
Shell FGD
Citrate
Calsox
Aqueous carbonate
20 (C)
616
780
754
697
482
971
810
1,136
1,002
879
759
1,117
858
757
924
806
604
1,300
20 (I)
624
780
761
702
490
979
818
1,152
1,013
887
770
1,125
871
765
929
816
609
1,300
Steam Output,
$/106 Btu
20 (C)
2.04
2.50
2.54
2.27
1.69
3.23
2.84
4.00
3.17
2.95
2.49
3.91
3.32
2.76
3.18
2.60
2.27
4.40
20 (I)
1.71
2.04
2.10
1.86
1.41
2.64
2.37
3.34
2.59
2.49
2.02
3.23
2.81
2.29
2.52
2.10
1.84
3.61
Sulfur Removal,
$/lb S
20 (C)
0.87
1.00
0.96
0.86
0.64
1.38
1.07
1.51
1.20
1.11
0.94
1.48
1.25
1.04
1.20
0.98
0.86
1.66
20 (I)
0.73
0.82
0.79
0.70
0.53
1.12
0.89
1.26
0.98
0.93
0.76
1.22
1.06
0.87
0.93
0.79
0.70
1.36
-------
TABLE 39. CAPITAL REQUIREMENT AND ANNUALIZED CONTROL COST OF
SORPTION PROCESSES FOR OIL-FIRED BOILERS
C: Commercial Boiler
I: Industrial Boiler
Annualized Cost
Capital Requirement^
Sorption Process 2 (C)
Limestone slurry 139
(Peabody)
Lime scrubbing 173
(Bahco)
Double alkali (GM) 162
Double alkali (FMC) 154
MgO (central 111
regeneration)
Chemiebau --
Foster Wheeler
Westvaco
Sulfacid 222
Chiyoda . 190
Ammonia scrubbing
(Peabody)
Ammonia scrubbing
(Catalytic, Inc.)
MgO (integrated)
Wellman-Lord
Shell FGD
Citrate
Calsox 122
Aqueous carbonate
2 (I)
139
173
162
154
111
--
--
222
190
--
--
--
--
--
--
123
™-
20 (C)
498
637
580
538
393
676
607
751
803
677
527
735
607
602
876
620
591
791
$103
20 (I)
503
637
584
540
398
681
612
759
806
683
530
740
612
607
879
625
594
791
Steam Output^ $/106
2 (C)
5.32
7.53
6.56
6.85
4.78
--
—
7.86
7.86
--
--
—
--
--
--
5.44
— —
2 (I)
4.91
6.89
6.09
6.34
4.46
—
--
7.20
7.31
.
—
--
--
--
--
5.03
— "
20 (C)
2.26
2.85
2.75
2.47
1.90
3.24
3.06
3.89
3.61
3.24
2.53
3.75
3.30
3.08
4.34
3.14
3.08
4.00
Btu
20
1.
2.
2.
2.
1.
2.
2.
3.
2.
2.
2.
3.
2.
2.
3.
2.
2.
3.
(I)
89
30
27
01
60
68
52
28
92
68
14
16
79
58
46
60
49
30
Sulfur Removal
2 (C) 2 (I) 20
4.40 4.06 1.
5.84 5.34 2.
4.83 4.48 2.
5.04 4.47 1.
3.52 3.29 1.
2.
2.
2.
5.79 5.30 2.
5.79 5.38 2.
1.
2.
2.
2.
3.
2.
4.01 3.70 2.
2.
, $/lb S
(C)
87
21
03
82
40
68
25
87
66
39
86
76
43
27
20
31
27
94
20 (I)
1.56
1.78
1.67
1.48
1.18
2.21.
1.86
2.42
2.15
1.97
1.58
2.32
2.05
1..90
2.54
1.91
1.83
2.43
00
o
(a) The blank indicates that the process is too complicated to be applied to the size class.
-------
181
REFERENCES
( 1) Ehrenfeld, J. R., Bernstein, R. H., Carr, K., Goldish, J. C., Orner,
R. G., and Parks, 1., "Systematic Study of Air Pollution from Inter-
mediate Size Fossil-Fuel Combustion Equipment," Walden Research
Corporation Report to EPA (Contract No. CPA-22-69-85), July, 1971.
( 2) Barrett, R. E., Putnam, A. A,, Blosser, E. R., Jones, P. W., "Assess-
ment of Industrial Boiler Toxic and Hazardous Emissions Control Needs,"
Battelle-Columbus report to EPA (Contract No. 68-02-1329 fTask 8]),
October 16, 1974.
( 3) Locklin, D. W., Krause, H. H., Putnam, A. A,, Kropp, E. L., Reid, W.
T., and Duffy, M. A., "Design Trends and Operating Problems in Com-
bustion Modification of Industrial Boilers," Final Report to EPA,
Grant No. 802402, April, 1974.
( 4) Paddock, R. E., and McMann, D. C., "Distributions of Industrial and
Commercial—Institutional External Combustion Boilers," Research
Triangle Institute Report to EPA (Contract No. 68-02-1323, Task 5),
February, 1975, EPA-650/2-75-021.
( 5) "Field Testing: Application of Combustion Modifications to Control
Pollutant Emissions from Industrial Boilers," EPA Publication No,
650/2-74-078-a, October, 1974.
( 6) Putnam, A. A., Kropp, E. L., and Barrett, R. E., "Evaluation of
National Boiler Inventory," Draft Final Report to EPA (Contract No,
68-02-1223), May, 1975.
( 7) Barrett, R. E., Putnam, A. A., Blosser, R. R., and Jones, P. W.,
"Assessment of Industrial Boiler Toxic and Hazardous Emissions
Control Needs," Final Report to U. S. EPA, Contract No. 68-02-1223,
October, 1974.
( 8) "Compilation of Air Pollutant Emission Factors," EPA Publication No.
AP-42, 1973.
( 9) Hall, E., Choi, P., and Kropp, E., "Assessment of the Potential of
Clean Fuels and Energy Technology and Recommendations of Technology
Development Priorities," EPA Contract No. 68-01-2114, January 3, 1974.
(10) Dupree, W. G. and West, J. A., "United States Energy Through the
Year 2000," U. S, Department of the Interior, December, 1972.
(11) U. S. Bureau of Mines, Minerals Yearbook 1971, Volume I, Metals,
Minerals, and Fuels, Washington, D. C., 1973.
(12) Hittman Associates, Inc., "Study of the Future Supply of Low Sulfur
Oil for Electrical Utilities," report to EPA, February, 1972.
-------
182
REFERENCES
(Continued)
(13) Hoffman, L., Lysy, F. J., Morris, J. P., and Yeager, K8 E., "Survey of
Coal Availabilities by Sulfur Content," report to EPA by Mitre Corpor-
ation, May, 1972o
(14) U. S. Bureau of Mines, "Bituminous Coal and Lignite Shipments from Coal
Producing Districts by Ranges of Sulfur Content; Calendar Year 1970,"
June, 1973.
(15) Electrical Week, July 2, 1973, pp. 9-10.
(16) Electrical Week, December 23, 1974, pp. 10-12,
(17) Schreiber, R., Davis, A,, Delacy, Jo, Chang, Y., and Lockwood, H0,
"Boiler Modification Cost Survey for Sulfur Oxides Control by Fuel
Substitution," EPA 650/2-74-123, November, 1974.
(18) Olmsted, L. M., "18th Steam Station Cost Survey," Electrical World,
November 1, 1973.
(19) "Coal Preparation," Leonard, J. W., Editor; Third Edition, 1968, AIME,
New York.
(20) Hoffman, L., Truett, J. B., and Aresco, S. J., "An Interpretative
Compilation of EPA Studies Related to Coal Quality and Cleanability,"
a report by Mitre Corporation to EPA, ORD, NERC-RTP, CSL, Research
Triangle Park, North Carolina, Contract 68-02-1352, PB 232011, May,
1974.
(21) Diverbrouch, A. W., "Coal Preparation 1973," Mining Congress Journal,
60_ (2), 65-67, February, 1974.
(22) Diverbrouch, A0 W.,and Jacobsen, P. S., "Coal Cleaning - State of the
Art," paper presented at the Coal Utilization Symposium (SO^ Emission
Control), Louisville, Kentucky, 1-10, October, 1974, pp. 22-24.
(23) Anonymous, "Compilation of Air Pollutant Emission Factors," 2nd Edition,
USEPA, Publication No. AP-42, April, 1973.
(24) "Standards of Performance for New Stationary Sources, Coal Preparation
Plants," Federal Register, 39. (207), Part II, 37924, October 24, 1974.
(25) Hardison, L. C., "Air Pollution Control Technology and Costs in Mine
Selected Sources," IGCI report to EPA, Durham, North Carolina, Contract
68-02-0301, Volume 2, September 30, 1972.
(26) Hurst, E., Lively Manufacturing and Construction, Beckley, West Virginia,
personal communication, June, 1974.
(27) "The Supply--Technical Advisory Task Force--Synthetic Gas--Coal," Final
Report prepared by Synthetic Gas-Coal Task Force for the Supply-Technical
Advisory Committee, National Gas Survey, Federal Power Commission,
Washington, DeC., April, 1973.
-------
183
REFERENCES
(Continued)
(28) "Comparative Study of Commercial Coal Gasification Processes - Koppers-
Totzek, Lurgi, and Winkler," sponsored by Indian Government, reproduced
by Koppers, 1969.
(29) Kim, B. C., Genco, J. M., Oxley, J. H., and Choi, P., "Development of
Information for Standards of Performance for the Fossil Fuel Conversion
Industry," Final Report for EPA by Battelle-Columbus, October 11, 1974.
(30) Kim, B. C., Genco, J. M., and Choi, P., "Development of Cost Data for
Coal Gasification Processes and Emission Control Systems," Final Report
for EPA by Battelle-Columbus, September 12, 1974.
(31) Farnsworth, J. F.} Leonard, H., Mitsak, D. M., and Wintrell, R., "Utility
Gas by the K-T Process," paper presented at meeting of EPRI, Monterey,
California, April, 1974.
(32) Tsaros, C. L», Knabel, S0 J., and Sheridan, L0 A., "Process Design and
Cost Estimate for Production of 265 Million scf/day of Pipeline Gas
by the Hydrogasification of Coal," a report prepared for OCR, Department
of the Interior, PB 176982, October, 1965.
(33) Glaser, F, Hershaft, A., and Shaw, R., "Estimation from Processes
Producing Clean Fuels," Draft Report prepared by Booz-Alien-Hamilton,
Inc., for EPA, Contract 68-02-1358, March, 1974.
(34) Second Supplement to Application of El Paso Natural Gas Company for a
Certificate of Public Convenience and Necessity, proposed by Stearns-
Roger, Inc., FPC Docket No. CP-73-131, October 8, 1973.
(35) Johnson, C. A., Chervenak, M. C., Johanson, E. S., Stotler, H. H.,
Winter, 0., and Wolk, R. H., "Present Status of the H-Coal Process,"
Hydrocarbon Research, Inc., 1973.
(36) Yavorsky, P. M., "Synthoil Process Converts Coal into Clean Fuel Oil,"
Clean Fuel from Coal Symposium, IGT, Chicago, Illinois, September 10-14,
1973.
(37) Chun, S. Wo, "Gulf Catalytic Coal Liquids Process," presented at NSF
and OCR Workshop on Materials Problems and Research Opportunities in
Coal Conversion, Columbus, Ohio, April 16-18, 1974.
(38) Consolidation Coal Company, "Summary Report on Project Gasoline,"
OCR, Research and Development Report, (39), Volume I.
(39) "Report on Combustion Trials on Spencer Low-Ash Coal," U. S. Department
of the Interior, Bureau of Mines, Pittsburgh, Pennsylvania, January,
1965.
(40) Frey, D. J., "De-Ashed Coal Combustion Study," Combustion Engineering,
Inc., October, 1964.
-------
184
REFERENCES
(Continued)
(41) Sage, W0 L., "Combustion Tests on a Specially Processed Low-Ash
Low-Sulfur Coal," Babcock and Wilcox Company, July, 1964.
(42) Battelle Energy Program Report, "Liquefaction and Chemical Refining
of Coal," July, 1974.
(43) Shore, D., O'Donnel, J. J0, and Chan, F. K., "Evaluation of R&D
Investment Alternatives for SO Air Pollution Control Processes,"
Final Report prepared by M0 W. Kellogg Company for EPA, Contract
No. 68-02-1308, EPA 650/2-74-098, September, 1974.
(44) Godel, A. A., "Ignifluid, A New System of Combustion," Combustion
Engineering Association Document 7593, pp 1-22, 1963.
(45) Novotny, P., "Fluid-Bed Combustion of High-Ash Coals," S.N.T.L.
Technical Digest (Prague), (12), pp 883-891, 1965.
(46) McLaren, J. and Williams, D. F., "Combustion Efficiency, Sulfur
Retention and Heat Transfer in Pilot Plant Fluidized-Bed Combustors,"
Combustion, 41. (11), pp 21-26 (1970).
(47) Wright, S. J., Ketley, H. C., and Hickman, R. G., "The Combustion of
Coal in Fluidized Beds for Firing Shell Boilers," Journal of the
Institute of Fuel, XLII (341), pp 235-240, June, 19690
(48) Hoy, H. R. and Roberts, A, G., "Power Generation via Combined Gas/
Steam Cycles and Fluid Bed Combustion of Coal," Gas and Oil Power,
July-August, 1969.
(49) Bishop, J. W., Robinson, E. B., Ehrlich, S., Jain, L. K., and Chen,
P. M., "Status of the Direct Contact Heat Transferring Fluidized Bed
Boiler," Paper 68-WA/FU-4, presented at Winter Annual Meeting, ASME,
New York, New York, December 1-5, 1968.
(50) "Energy Conversion from Coal Utilizing CPU-400 Technology," Research
and Development Report No. 94, Interim Report No. 1 prepared by
Combustion Power Company for Office of Coal Research, Contract No.
14-32-001-1536, September, 1974.
(51) Hoke, Ro C., Shaw, H., and Skopp, A., "A Regenerative Limestone Process
for Fluidized Bed Coal Combustion and Desulfurization," Proceedings
of the Third International Conference on Fluidized Bed Combustion,
EPA, October, 1972, pp 93-116.
(52) Wright, S. J., "The Reduction of Emissions of Sulfur Oxides and
Nitrogen Oxides by Additions of Limestone or Dolomite during the
Combustion of Coal in Fluidized Beds," Proceedings Third International
Conference on Fluidized Bed Combustion, EPA, October, 1972, pp 135-154.
-------
185
REFERENCES
(Continued)
(53) Archer, D. H., Keairns, D. L., Hamm, J. R., Newby, R. A., Yang, W. C.,
Handman, L. M., and Elikan, L., "Evaluation of the Fluidized Bed Com-
bustion Process, Volume II, Technical Evaluation," a report by Westing-
house Research Labs to EPA, Office of Air Programs, Research Triangle
Park, North Carolina, PB 212960, November, 1971.
(54) Gordon, J. So, Glenn, R. D., Ehrlich, S., Ederer, R., Bishop, J. W.,
and Scott, A. K., "Study of the Characterization and Control of Air
Pollutants from a Fluidized-Bed Boiler - The S02 Acceptor Process,"
Final Report prepared by Pipe, Evans, and Robbins, Inc., for EPA,
Contract No. CPA 70-10, July, 1972.
(55) "SO- and Fly Ash Removal Scrubbing Systems," a brochure furnished by
Peabody Engineering Systems, September 12, 1974.
(56) McKenna, J. D. and Atkins, R. S., "The R-C/Bahco System for Removal
of Sulfur Oxides and Fly Ash from Flue Gases," a paper presented at
Second International Lime/Limestone-Wet Scrubbing Symposium, New
Orleans, Louisiana, November 8-12, 1971.
(57) Atkins, R. S., "Process Experience of the R-C/Bahco Sulfur Dioxide
Removal System," in Pollution Control and Energy Needs, edited by
Jimeson, R. M. and Spindt, R. S., Advances in Chemistry Series No.
127, 1973.
(58) Brady, Jack D., "Sulfur Dioxide Removal Using Soluble Sulfites," a
paper presented at Rocky Mountain States Section Air Pollution
Control Association, Colorado Springs, Colorado, April 30, 1974.
(59) Brochure furnished by FMC on sulfur dioxide and fly ash control.
(60) Koehler, George R., "Operational Performance of the Chemico Basic
Magnesium Oxide System at the Boston Edison Company, Part 1," a
paper presented at the Flue Gas Desulfurization Symposium, New
Orleans, Louisiana, May 14-17, 1973.
(61) TVA, "Sulfur Oxide Removal from Stack Gas, Magnesia Scrubbing,
Regeneration: Production of Concentrated Sulfuric Acid," Contract
No. TV-29233A, May, 1973„
(62) Shah, I. S. and Quigley, Ce P., "Magnesium Base S02 Recovery Process:
A Prototype Installation at Boston Edison Company and Essex Chemical
Company," AIChE Symposium Series, 67_ (126), pp 139-146, 1972.
(63) Paper furnished by the Davy Powergas Company on operating costs for
the Wellman-Lord Process.
(64) "Sulfur Dioxide Removal from Power Plant Stack Gas by Limestone or
Lime Dry Process," Final Report prepared by TVA for NAPCA, PB 178972,
1968. '
-------
186
REFERENCES
(Continued)
(65) "Applicability of Inorganic Solids Other than Oxides to the Development
of New Processes for Removing S0? from Flue Gases'," Final Phase I Report
prepared by FMC for NAPCA, Contract No. PH 22-68-57, PB 184751, June,
1969.
(66) Friedman, L. D., "Applicability of Inorganic Solids Other than Oxides
to the Development of New Processes for Removing SO from Flue Gases,"
Final Phase II Report prepared by FMC for NAPCA, Contract No. CPA .
22-69-92, PB 203496, December, 1970.
(67) Slack, A. V., "Sulfur Dioxide Removal from Waste Gases," Noyes Data
Corporation, Park Ridge, New Jersey, 1971.
(68) Thomas, A0 D., Jr., Davis, D. L0, Parsons, T., Schroeder, G. D., and
DeBerry, D., "Applicability of Metal Oxides to the Development of New
Processes for Removing S09 from Flue Gases, Volume I," Final Report
prepared by Tracer Company for NAPCA, Contract No« PH 86-68-68,
PB 185562, July 31, 1969.
(69) Thomas, A. D., Jr., Davis, D. L., Parsons, T., Schroeder, G. D., and
DeBerry, D., "Applicability of Metal Oxides to the Development of New
Processes for Removing S0» from Flue Gases, Volume II," Final Report
prepared by Tracer Company for NAPCA, Contract No. PH 86-68-68,
PB 185563, July 31, 1969.
(70) "Economic Evaluation of Metal Oxide Processes for SO^ Removal from
Power Plant Flue Gases, Phase 3, Final Report, Cost Sensitivity Study
of Major Process Parameters," Final Report prepared by M. W. Kellogg
Company for NAPCA, Contract No. PH 86-68-86, PB 200882, March 31, 1970.
(71) Opferkuch, R. E0, Mehta, S. M., Constam, A. H., Zanders, D. L., and
Strop, H. R., "Applicability of Catalytic Oxidation to the Development
of New Processes for Removing S09 from Flue Gases - Volume I - Literature
Review," Final Report prepared by Monsanto Research Corporation for
NAPCA, Contract No. PH 22-68-12, PB 198808, August, 19700
(72) Miller, W. E., "The Cat-Ox Demonstration Program," presented at the
Flue Gas Desulfurization Symposium, Atlanta, Georgia, November 4-7, 1974,
(73) Collins, J. J0, Fornoff, L. L., Manchanda, K. D0, "The Purasive Process
for Removing Acid Plant Tail Gas," Chemical Engineering Progress, 70 (6),
June, 1971.
(74) Dibbs, H. P., "Methods for Removal of Sulphur Dioxide from Waste Gases,"
Mine Branch Information Circular, Canadian Department of Mines and
Resources, 1971.
(75) Cole, R. and Shulman, H. L., "Adsorbing Sulfur Dioxide on Dry^ Ion
Exchange Resins," Ind. Eng. Chem., 52, 859, 1960.
-------
187
REFERENCES
(Continued)
(76) Pinaev, V. A. and Muromtseva, L. A., "Sorption of Sulphur Dioxide by
Synthetic Resins," Zh. Prikl. Khim., 41., 2092, 1968.
(77) Fuest, R. W. and Harvey, M0 P., "Development of Regenerable Fibers
for Removal of Sulfur Dioxide from Waste Gases," U. S0 Clearinghouse
Fed. Sci. Tech. Inform., Report No. PB 185093, 1968.
(78) Meyers, R. A., Grunt, A., and Gardner, M., "Applicability of Organic
Solids to the Development of New Techniques for Removing Oxides of
Sulfur from Flue Gases," U. S0 Clearinghouse Fed. Sci. Tech. Inform.,
Report No. PB 187645, 1969.
(79) Gressingh, L. E., Graefe, A. F., Miller, F. E., and Barber, H.,
"Applicability of Aqueous Solutions to the Removal of S0_ from Flue
Gases, Volume I," Final Report prepared by Envirogenics Company for
NAPCA, Contract No. PH 86-68-77, PB 196780, October, 1970.
(80) Graefe, A. F., Gressingh, L. E., and Miller, F. E., "The Development
of New and/or Improved Aqueous Processes for Removing SO- from Flue
Gases, Volume II," Final Report prepared by Envirogenics Company for
NAPCA, Contract No. PH 86-68-77, PB 196781, October, 1970.
(81) Borgwardt, R. H., "EPA/RTP Pilot Studies Related to Unsatuirated
Operation of Lime and Limestone Scrubbers," presented at EPA Flue
Gas Desulfurization Symposium, November 4, 1974.
(82) McKinney, W. A., Nissen, D. A., Rosenbaum, J. B., U. S. Bureau of
Mines, Salt Lake City Metallurgy Center, "Pilot Plant Testing of
the Citrate Process for S0» Emission Control," presented at the EPA
Flue Gas Desulfurization Symposium, Atlanta, Georgia, November 4-7,
1974.
(83) Brochure furnished by FMC on sulfur dioxide and fly ash control.
(84) Phillips, R., "Operating Experiences with a Commercial Dual-Alkali
S02 Removal System," report presented at 67th Annual Meeting at the
Air Pollution Control Association, Denver, Colorado, June 9-13, 1974.
(85) Noguchi, Masaaki, "Status Report on Chiyoda Thoroughbred 101 Process,"
presented at Flue Gas Desulfurization Symposium, Atlanta, Georgia,
November 4-7, 1974.
(86) Battelle's Pacific Northwest Laboratories, "Applicability of Organic
Liquids to the Development of New Processes for Removing Sulfur Dioxide
from Flue Gases," Final Phase I Report for NAPCA, Contract No.
PH 22-68-19, PB 183513, March, 1969.
-------
188
REFERENCES
(Continued)
.(87) Dingo, T. T., and Piasecki, E. J., "Initial Operating Experiences
With A Dual-Alkali S0_ Removal System," a paper presented at EPA
Symposium on Flue Gas Desulfurization, Atlanta, Georgia, November
4-7, 1974.
(88) Brochure furnished by Commonwealth Associates on the Chemiebau
Process.
(89) Private communications between A0 C. Kelsall of Commonwealth Asso-
ciates, Inc., and P. Choi of Battelie-Columbus, December 27, 1974.
(90) Bischoff, W. W0, "FW-BF Dry Adsorption System for Flue Gas Cleanup,"
presented at 1973 Flue Gas Desulfurization Symposium, New Orleans,
Louisiana, December, 1973„
(91) Private communication between W. Bischoff and E0 Beckman of Foster
Wheeler and P0 Choi of Battelie-Columbus, December 30, 1974.
(92) "SO. Recovery Process," brochure furnished by Westavco.
(93) "Westavco S0_ Process 15-MW Design and Cost," draft report furnished
by Westavco, August 23, 1974.
(94) The Rust Engineering Company, "Stack Gas Desulfurization by the
Sulfacid Process," a brochure supplied by the Rust Engineering Company,
September, 1974.
(95) Private communication between B. D. Trusty, The Rust Engineering
Company, and P. Choi, Battelie-Columbus, September 11 and October 14,
1974.
(96) "Chiyoda Process Applied to Tail Gas Processing Claus Plants," report
of the Edison Electric Institute Study Program on S02 Removal Processes
in Japanese Plants, March, 1973.
(97) Private communication between Ab Saleem, Technical Director, Air
Pollution Control Division, Peabody Engineering Systems, and W«
Ballantyne, Battelle-Columbus, October 24, 1974«
(98) Tennessee Valley Authority, "Sulfur Oxide Removal from Power Plant
Stackgas - Ammonia Scrubbing - Production of Ammonium Sulfate and
Use as an Intermediate in Phosphate Fertilizer Manufacture,"
Conceptual Design and Cost Study Series #3, Contract No. TV-29233A,
September, 1970.
(99) "Reliable S02 Removal," a brochure furnished by Catalytic, Inc.
-------
189 and 190
REFERENCES
(Continued)
(100) Letter from Joseph R0 Polek of Catalytic, Inc, to W. E. Ballantyne
of Battelie-Columbus, November 6, 1974.
(101) Pohlenz, J. B., "The Shell Flue Gas Desulfurization Process," a
paper presented at the Flue Gas Desulfurization Symposium, Atlanta,
Georgia, November 4-7, 1974.
(102) Private communication between J. Bo Pohlenz of Union Oil Products
and P. Choi of Battelle-Columbus, January 6, 1975.
(103) Dantzenberg, F. M., Naber, J« E., and Van Ginneken, A. J. J., "The
Shell Flue Gas Desulfurization Process," a paper presented at the
68th National Meeting of AIChE, Houston, Texas, February 28 -
March 4, 1971.
(104) Personal communication with R. E. Barnard,and Richard league,
Monsanto Enviro-Chem Systems, Inc., St. Louis, Missouri, September
24, 1974.
(105) Botts, W. V., and Gehri, D. Co, "Regenerative Aqueous Carbonate
Process (ACP) for Utility and Industria
presented at 167th American Chemical So
Los Angeles, California, April 4, 1974.
Process (ACP) for Utility and Industrial SO- Removal Applications,"
presented at 167th American Chemical Society National Meeting,
-------
APPENDIX A
ACCOUNTING METHOD
-------
A-2
APPENDIX A
ACCOUNTING METHOD
An accounting method was derived from the Utility Financing
Method as modified by the Panhandle Eastern Pipeline Company. The
accounting method was then employed uniformly for all cost estimations.
A description of the method follows.
Total Plant Investment and
Total Capital Requirement
Total Bare Cost
The total bare cost includes major equipment costs, direct con-
struction labor costs, undistributed costs such as costs for construction
facilities and services, and other plant costs such as for utilities and
off-site facilities. This cost is used as a basis for other cost estimations.
Engineering and Design Cost
This cost generally is assumed at 5 percent of the total bare
cost. In some available data, this cost was included in the total bare
cost.
Contractor's Overhead and Profits
This cost is assumed at 10 percent of the total bare cost. In
some available data, this cost was included in the total bare cost.
Subtotal Plant Investment
This is the summation of the total bare cost, engineering and
design cost, and contractor's overhead and profits.
-------
A-3
Project Contingency
This represents the degree of uncertainty in the overall con-
struction cost estimate. This is assumed at 15 percent of the subtotal
plant investment,
Total Plant Investment (TPI)
This is the summation of the subtotal plant investment and
project contingency.
Interest During Construction (IDC)
This is obtained from the following equation.
IDC = (interest rate)(TPI)(average construction period in year).
Startup Cost
This cost is assumed at 20 percent of the gross annual operating
cost.
Working Capital (WKC)
This cost is also assumed at 20 percent of the gross annual
operating cost.
Total Capital Requirement (TCR)
The total capital requirement is given as the summation of the
total plant investment, interest during construction, startup cost, and
working capital.
-------
A-4
Net Annual Operating Cost
Direct Material and Utilities
This cost includes costs for raw materials consumed in the
process and utilities such as power, fuel, process steam, process ajid
cooling water, and compressed air.
Maintenance and Operating Supplies
This cost includes costs of supplies for operating and main-
tenance. The cost is assumed to be the summation of 30 percent of direct
operating labor and 1.5 percent of total plant investment, if not specified,
Direct Operating Labor (POL)
This cost is obtained using the following equation.
DOL = (man-hour required/hr)($5/man-hour)(8,304 hrs/yr).
Maintenance Labor
The annual maintenance labor cost is given as 1.5 percent of
the total plant investment, if not specified.
Supervision
The cost for supervision in general is assumed at 15 percent of
the summation of direct operating labor and maintenance labor costs.
-------
A-5
Administration and General Overhead
This cost is assumed at 60 percent of the total labor cost
including supervision.
Local Taxes and Insurance
The annual local taxes and insurance are estimated at 2.7 percent
of the total plant investment.
Gross Operating Cost
This is the summation of all annual operating costs listed above.
By-Product Credit
This credit comes from the sale of by-products. The by-product
credit is subtracted from the gross operating cost to obtain the net
operating cost.
Net Annual Operating Cost (AQC)
/
The net annual operating cost is obtained by subtracting the
total by-product credit from the gross annual operaring cost. The
escalation of net annual operating cost was not considered in this study.
Annualized Coat
The control cost represents the average cost during the life of
the plant. The basis for the calculation is:
• 20-year plant life
• 5 percent per year straight line depreciation on
total capital requirement excluding working capital
-------
A-6
• 75 percent/25 percent debt/equity ratio
• 9 percent per year interest on debt
• 15 percent per year return on equity after tax
• 48 percent Federal income tax rate.
For retrofit processes such as fluidized combustion and FGD processes, the
life of the system was assumed at 10 years and accordingly, 10 percent per
year straight line depreciation was used. The following procedures are
used to calculate the average annual revenue requirement and product (or
control) cost (for the case of 5 percent straight line depreciation).
Average Annual Depreciation (D) = 0.05 (TCR-WKC).
Average Return on Rate Base (RRB) = 0.0525 (TCR + WKC).
Average Federal Income Tax (FIT) = 0.01731 (TCR + WKC).
Average Annual Revenue Requirement (ARR) = RRB + FIT + D
+ Net AOC.
-------
A-7 and A-8
Unit Cost
Coal, $/ton
Limestone, $/ton (pulverized limestone)
Lime, $/ton
NaOH, $/ton
NaCO_, $/ton
MgO, $/ton
Ammonia, $/ton
Fuel oil, $/bbl (high S)
No. 2 fuel oil, $/bbl
Carbon dioxide, $/ton
H2SO, (98 percent), $/ton"|
H2S04 (80 percent), $/ton|By-pr°duCt Credlt
Elemental sulfur, $/ton
Labor, engineer, $/hr
technician, $/hr
Electricity, $/kwhr
o
Process water, $/10 gal
o
Cooling water, $/10 gal
Steam, $/60 Btu
Ferric sulfate, $/ton
Coke, $/ton
Oxygen, $/ton
Sand, $/ton
Activated carbon, $/ton
10
7 (10)
25
150
50
140
150
3
4.5
60
20
10
10
7.5
5
0.01
0.5
0.1
0.5
50
40
10
5
800 (for Westvaco)
300 (for FW)
60 (for by-product sale)
-------
APPENDIX B
DESCRIPTION OF FLUE GAS
DESULFURIZATION PROCESSES
-------
B-2
APPENDIX B
DESCRIPTION OF FLUE GAS
DESULFURIZATION PROCESSES
Limestone Slurry (Peabody Engineering)
Developer /Manufacturer
Peabody Engineering Systems, Stamford, Connecticut, has developed,
and is currently marketing a limestone based S02 removal system for indus-
trial boilers.
Process Description
The Peabody process utilizes a lime, limestone, or soda ash
slurry in a spray tower absorber to remove S02 and particulates from flue
gas. For the limestone system, the spent slurry is thickened and filtered
and the filter cake, containing predominantly CaSC<4, is trucked away. The
overflow is returned to the absorber to complete the loop.
The flue gas, after passing through a multiple cyclone separator
or ESP, is passed through an ID fan and into the quench system where it is
cooled by a contacting spray of limestone slurry (see Figure B-l). It
flows through the absorber, a three or five bank spray tower, where the S02
reacts with the limestone slurry to form CaS03 and CaSO^. In the spray
tower, 70 to 90 percent of the CaS03 is oxidized to CaSO^. After passing
through an impingement tray and a mist eliminator the gas exits the
scrubber, and flows through a reheater, and is vented to the stack. The
system operates with a liquid-to-gas ratio of 75-100 gal/1000 ft3. Pressure
drop through the tower is about 4 inches of water.
S02 + H20 = H2S03
H2S03 + CaC03 = CaS03 + H20 + C02
CaS03 + 02 = CaSC-4
CaS03 + i- H20 = CaS03 • i- H20
CaS04
-------
B-3
FLUE 9AS
SLURRY
BLEED
PEABOOY/LURGI
RADIAL VENTURI
FIGURE B-I. PEABODY LIMESTONE SCRUBBING PROCESS
-------
B-4
The spent slurry drains from the bottom of the tower into the
recycle tank. Limestone is added to the recycle tank to control the pH.
Slurry overflows from the tank to a sump tank from where it is pumped to
a thickener. The CaS04 crystals are preferred over 63803 cyrstals because
of their better settling and dewatering characteristics. This is why the
high oxidation rate in the spray tower is desired. The thickener under-
flow is pumped to a vacuum filter where a 70 percent solids cake is* formed,
The cake is conveyed to a truck and transported .to a landfill area. Thickener
overflow and the filtrate are returned to the recycle tank to maintain closed
loop operation.
Removal Efficiencies
The quench-tower system is capable of a 70 to 90 percent S02 and
95 percent fly ash removal efficiency. For higher fly ash removal effi-
ciencies, 99 percent, a Peabody Radial Flow Venturi may be substituted for
the quench system.
Was tes
The only waste stream is the filter cake containing predominantly
CaSO,. About 4 lb of filter cake (70 percent solids) is generated per
pound of 862 removed.
By-Products
None.
Materials of Construction
All slurry pumps and flow lines are rubber lined. The reaction
tank and scrubber are lined with fiberglass reinforced polyester. The
absorber interface tray is constructed of 316 L stainless steel.
-------
B-5
Raw Material and Heat Requirements
The only raw material required is pulverized limestone. Heat
is required to reheat the flue gas.
Advantages and Disadvantages
Advantages: (1) Simple operation
(2) Low pressure drop
(3) Low cost absorbent (limestone)
(4) Good turndown ratio
(5) Comparatively low operating cost.
Disadvantages: (1) Waste sludge generation
(2) Stack gas reheat may be required
(3) High liquid to gas ratio.
Development Status
In 1973 Peabody designed and built a 1-MW pilot plant S02
scrubbing facility at Detroit Edison's River Rouge Station, Detroit. It
has reportedly been operating satisfactorily since startup in February,
1974. A full-scale facility is presently undergoing start-up at Detroit
Edison's St. Glair Unit No. 6, a 175-MW coal-fired boiler.
Capital and Operating Costs
The 1974 capital cost for a Peabody limestone SO2 scrubbing
system was estimated at $0.8 million and $1.4 million for coal-fired
50,000 and 150,000 Ib steam/hr boiler systems, respectively. The cost
includes the engineering and design cost of the system including the
reheater. Table B-l shows the labor, material, and utility require-
ments of the Peabody system installed on a 150,000 Ib steam/hr boiler
system.
-------
B-6
TABLE B-l. LABOR, MATERIALS, AND UTILITY REQUIREMENTS FORA
PEABODY LIMESTONE SCRUBBING PROCESS^)
Basis: 150,000 Ib steam/hr coal boiler
3 percent sulfur in coal
75 percent load factor
6,500 hours/yr operation
80 percent removal efficiency
92,000 acfm of flue gas at 450°F
Item Quantity
Utility
Power 330 kW
Steam 2,800 Ib/hr
Water 50 gpm
Material
Limestone 0.9 tons/hr
Labor
Direct operation 1 man/shift
Maintenance-labor 1/3 man shift
Material 2 percent of capital cost/yr
-------
B-7
Lime Scrubbing (Bahco)
Developer/Manufacturer
The Bahco S02 removal process was developed by A. B. Bahco
Ventilation, Enkoping, Sweden. The process is being marketed in the
United States by Research-Cottrell, Inc., Bound Brook, New Jersey.
Process Description
The Bahco process used a lime slurry in a two-stage venturi
scrubber to remove particulate matter and SO2 from flue gas ,5 * As can
be seen in Figure B- 2 , flue gas is introduced into the bottom stage of
the scrubber where it contacts hydrated lime slurry and passes into the
first venturi. Sulfur dioxide reacts with the lime slurry to form cal-
cium sulfite and calcium sulfate.
Ca(OH)2 + S02 + H20 -. CaS03«2H20
CaS03-2H20 4- j. 02 -» CaSO^ 2H20
The atomized droplets are separated from the gas at the top of the venturi
by a centrifugal force drop collector. The gas is directed to the second
venturi and the collected liquid is returned to the first stage contact
zone. A concentration regulator continuously withdraws a fraction of
return stream and feeds it to a thickener or a drum filter. After con-
tacting the second stage drop collector, the gas is sent through a reheater
where it is heated to 175°F and expelled to the stack. The collected
liquid is also returned to the first stage contact zone where any over-
flow is pumped to the dissolving tank and ultimately returned to the
second stage impingement zone. The level in the first stage is regulated
-------
B-8
FIGURE B-2 . BAHCO S02 SCRUBBER
(56)
1. Flue Gas Inlet
2. First Scrubber Stage
3. Guide Vane
4. Second Scrubber Stage
5. Guide Vanes
6. Clean Gas Outlet
7. Storage Bin
8. Screw Feeder
9. Dissolving Tank
10. Water Supply
11. Tubes
12. Level Tank
13. Tubes
14. Tubes
15. Sludge Mill
16. Tubes
17. Concentration Regulator
18. Settling Tank
19. Sludge Outlet
-------
B-9
by controlling the level in the dissolving tank. The sludge stream leaving
the thickener is concentrated in a drum filter.
Removal Efficiency
The sulfur dioxide removal efficiency ranges from 70 percent
to 99 percent depending on sulfur dioxide concentration in flue gas.
Raw Materials
Materials required in Banco process include lime and process
water.
By-Product
No by-product is obtained from the Bahco process.
Wastes
Waste emissions from the Bahco process include sulfur dioxide
residual emission and sludge resulting from the process. About 5.1 Ibs of
calcium sulfate sludge (50 percent solids) are produced per pound of SO
removed.
Advantages
(1) Simple process
(2) Simultaneous removal of fly ash and sulfur dioxide
(3) High S02 removal efficiency
(4) High reliability.
-------
B-10
Disadvantages
(1) High lime cost
(2) Waste sludge generation
(3) High power requirement
(4) Reheating requirement for flue gas.
Development Status
In 1964, A. B. Bahco Ventilation of Sweden initiated invest!- ;.
gations of sulfur dioxide control using alkaline base scrubbing reagents.
In 1966, they installed a 1,400 scfm pilot unit on the boiler of their
central heating plant. In 1969, their first commercial unit was installed
on an oil-fired boiler producing 75,000 Ibs/hour of steam. In 1970,
Bahco licensed their process technology rights in Japan to Marubeni with.
Tsukishima Kikai as a sublicensee. In August, 1971, Research-Cottrell
acquired the rights to license the Bahco system in the United States and
Canada. Currently, 19 commercial units have been installed in Japan and
Sweden, and one unit will be installed on a coal-fired boiler in the
United States in 1975. Table B-2 summarizes the location and service
of these installations.
Capital and Operating Costs
The installed cost for a Bahco system using carbon steel and
treating the flue gas characterized in Table B-3, was estimated at $0.8
million in 1971. This did not include any unique installation costs
such as interconnecting duct work, utility connections, remote
instrumentation. Total installed system costs are often significantly
higher than installed costs. For example, for a Bahco system to
be installed on a 18-MW coal-fired stoker boiler at Rickenbacker Air
Force Base, Columbus, Ohio, the total installed cost was quoted as
-------
TABLE B-2 . INSTALLATIONS OF THE BAHCO S02 REMOVAL SYSTEM
(57 )
Company
Soders jukhuset
Daishowa Selshi Co.
Daishowa Seishi Co.
Osaka City
Hiroshima City
Yahagi Iron Works
Taio Paper Co.
Central Glass Co.
Stora Kopparberg
Kanegafuchi Chemical
Daishowa Seishi Co.
Rickenbacker Air
Force Base^
Location
Stockholm, Sweden
Suzukawa, Japan
Yos h inaga , Japan
Osaka, Japan
Hiroshima, Japan
Nagoya, Japan
lyomishima, Japan
Sakai, Japan
Gryeksbo, Sweden
Takasago City, Japan
Yoshinaga, Japan
Columbus, Ohio,
United States
No.
of
Units
3
1
5
1
1
1
1
1 3
1
2
2
1
Unit
Capacity,
scfm
at 32 F
17,700
14,700
44,200
10,000
10,000
48,300
83,000
31,300
17,700
79,500
66,400
45,000
Service
Oil-fired boiler
Oil-fired boiler
Oil-fired boiler
Secondary sludge
incinera-tor
Inle±
Scrubbing S02
Reagent Cone., ppm
Ca(OH)2 800-1500
NaOH 900-1200
NaOH 900-1000
NaOH
Secondary sludge NaOH
incinerator
Sintering plant Ca(OH)2 2500-4000
waste carbide sludge
Oil-fired boiler
Glass furnace
Black liquor boiler
Oil-fired boiler
SCA-Billerud recovery
boiler
Coal-fired boiler
NaOH 1000-1500
NaOH 1200
(25% S02)
CaO and 1000-6000
CaC03 dust
NaOH
NaOH
Ca(OH)2
so2
Removal
Efficiency,
percent
97-99
97-99
97.5
90-95
98
98
70
(a) Start-up late 1975.
-------
B-12
TABLE B-3. FLUE GAS CHARACTERISTICS FOR BAHCO PROCESS^56)
Item Value
Flow rate 137,000 acfm
Temperature 350°F
SO- concentration 2,000 ppm
Fly ash loading 2.4 gr/scf
-------
B-13
about $2.3 million including costs for dust collector, sludge pond, etc.,
in 1974. The system will be based on stainless steel and the construction
will begin in early 1975. The exponential scale-up factor for equipment
cost was quoted as 0.6-0.7.
Operating costs consist of labor, material, and utilities. The
requirements for treating the flue gas characterized in Table B-3 are
shown in Table B-4.
-------
B-14
TABLE B-4 . LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR BAHCO PROCESS '
Item Quantity
Utilities
Power 600 kW
Makeup water 40 gpra
Material
Lime 0.94 tons/hr
Labor
Direct operation 0.5 man/shift
Maintenance* 3 percent of total plant
investment/yr
* This includes labor and material.
-------
B-15
.Double Alkali Process (FMC)
Developer/Manufacturer
The FMC/Link-Belt Double Alkali SO Absorption System was
developed by FMC Corporation.
Process Description
In an effort to overcome th,e scaling difficulties with direct
calcium slurry scrubbing, FMC employs a "concentrated" sulfite/bisulfite
buffer solution to remove particulate matter and S0_ from flue gas. The
(58}
spent solution adds lime or limestone outside of the scrubber system.
A simplified flow sheet of the FMC/Link-Belt Process is shown
(59)
in Figure B- 3 • The flue gas enters the scrubber at about 300°F
after passing through a cyclone dust collector and a forced-draft fan.
The scrubber is a dual throat variable-flow venturi where both fly ash
and SO are removed by contacting with a 20 weight ;% solution of Na~SO ,
NaHSO , and Na SO,. Sulfur dioxide reacts with the sodium sulfite to
form sodium bisulfite; some sodium sulfite oxidizes to sodium sulfate.
S03 + S02 + H20 + 2HSO~
so!: + 1/2 o0 +SOT
3 2 4
n o
The scrubber normally operates with a liquid-to-gas ratio of 10 gal/10 ft ,
In the cyclone, the water droplets separate from the gas and
descend to the bottom. The flue gas exits through the top of the scrubber,
is reheated to 200°F and is sent to the stack. The spent slurry is pumped
from the bottom of the scrubber to the recirculation tank. A regeneration
stream is withdrawn from the recycle stream at a rate equal to the rate
of S02 collected in the scrubber. It is transferred to the lime reactor
where lime is added to form insoluble CaSO and regenerate the Na SO .
SO^ + Ca(OH)2 •»• CaS034- + 20H~
20H- + 2HS03 -> 2SO~ + 2H 0
-------
120 PSIG STEAM ^
VENTURI
TO STACK
REHEATER
LIME
STORAGE
BIN
CONDENSATE
SODA ASH
STORAGE
BIN
LIME
REACTOR
ROTARY
FILTER
FLYASH
and
CoS03
EXHAUST
FILTRATE
RECEIVER
CYCLONE
SCRUBBER SYSTEM
FILTER
VACUUM
PUMP
RECYCLE STREAM
HICKENER
UNDERFLOW
PUMP
RECIRC.
PUMP
30 PSIG .
PLANT WATER
FILTRATE
RETURN
PUMP
SPARE
PUMP
w
REGENERATION STREAM
SURGE TANK
FIGURE B-3. FMC/LINK-BELT ALKALINE ABSORPTION PROCESS
FOR SULFUR DIOXIDE CONTROL (59)
-------
B-17 '
Shipping and Construction
For small units, i.e., less than 20 MW, the venturi scrubber
and vacuum filter may be prefabricated in the shop. The large size of
the storage tanks and thickener, however, requires that they be shipped
to the construction site in sections and field fabricated.
By-Product
No marketable by-product.
Raw Materials and Heat Requirements
Lime and soda ash are the required raw materials. Steam is
necessary to reheat the flue gas from 120 to 200°F.
Advantages
(1) High S02 and particle removal efficiency with a
liquid-to-gas ratio
(2) No severe plugging or scaling problems
(3) High reliability (minimal sulfite oxidation and greater
pH flexibility in scrubber).
Disadvantages
(1) Waste sludge generation
(2) Flue gas reheat required
(3) High lime cost
(4) High power requirement.
-------
B-18
Complete lime reaction and formation of calcium siilfite-is insured by
maintaining a high sulfite concentration. The mixture of calcium sulfite,
sodium sulfite, sodium sulfate and fly ash is transferred to the thickener.
The thickener overflow, containing soluble sodium sulfite and sodium sul-
fate is returned to the recirculation tank. The thickener underflow,
containing 25 to 30 percent solids, is transferred to a rotary vacuum
filter. The resultant filter cake contains about 55 percent solids,
composed primarily of CaSO^ and fly ash with about 4 to 5 percent t^SO-j
and Na2SO,. Most of the sodium salts are recovered from the cake by
washing and are returned with the filtrate to the recirculation tank. The
sodium losses are made up by the addition of soda ash to the recirculation
tank. :
Removal Efficiencies .••••'
The fly ash and S02 removal efficiencies are 99 and 90 percent,
respectively. .
Was tes
The only waste emission is the CaSO^ and fly ash filter cake.
About 5.85 lb of solid waste (55 .percent solids) is generated per pound
of S(>2 removed.
Materials of Construction
The scrubber is constructed.of,316L stainless steel. To
prevent fly ash abrasion, slurry lines and pumps are rubber lined.
Direct steam tube reheat requires Hastelloy G heating tubes. Indirect
reheating of outside air and blending with the flue gas may be accomplished
with carbon steel tubes.
-------
B-19
Development Status
The chemistry of this process has been tested at FMC's 20,000
acftn barium sulfate reduction kiln in Modesto, California. A packed bed
absorber is used instead of the venturi to remove S02 with inlet concen-
trations of up to 8,000 ppm. FMC reports that since startup in December,
(590
1971, the system has operated troublefree for over 22,000 hours.
In 1971, a semi-trailer was fitted with a 3,500 acfm pilot plant
and several runs have been made on different coal-fired boilers. It is
still operating and is a demonstration device for marketing purposes.
Construction has begun on the only large industrial unit at
Caterpillar Tractor's 45-MW boiler at Mossville, Illinois. Startup is
expected soon.
Capital and Operating Costs
"~ —^—• „ ^—f— - _^w«-«^«
The capital cost for a coal-fired boiler (capacity equivalent
to 45 MW) was estimated at $3.057 million in 1973-(59^ The labor,
material, and utility requirements of the system are shown in Table B-5.
-------
B-20
TABLE B-5. LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR FMC
DOUBLE ALKALI PROCESS' (Capacity, 45 MW)
Item Quantity
Utility .
Power 1 MW
Steam 17,600 Ib/hr
Water 70 gpm
Material
Lime (92 percent) 1.64 tons/hr
Soda ash 0.25 tons/hr
Labor
Direct operation 1/2 man shift
Maintenance 2 percent of capital cost
* This includes labor and material.
-------
B-21
MgQ Process
Developer /Manufacturer
Chemical Construction Corporation, New York, New York, developed
the regenerable magnesium oxide-sulfur dioxide removal process.
Process Description
The flue gas at 300°F is passed through an electrostatic pre-
cipitator and is introduced into the top stage of a two-stage venturi
scrubber/absorber where the fly ash and particulate matter are removed by
contacting a water spray (see Figure B-4). The gas passes into the
lower stage where it contacts a 12 percent solid slurry of MgO, MgSC^,
and MgSO^. The S02 i n the flue gas reacts with the MgO to form MgS03 ;
some oxidation to MgSO* occurs, normally 15-20 percent.
Absorption Main Reactions :
MgO + S02 + 3H20 -»-
MgO + S02 + 6H20 -»•
Side Reactions :
MgS03 + S02 + H20 + Mg(HS03)2
Mg(HS03)2 + MgO -»• 2MgS03 -fc H20
MgO + S03 + 7H20 ->•
MgS0 + 0 H- 7H0
3 3
The normal liquid to gas ratio is about 33 gal/10 ft . At design operating
conditions, the total pressure drop through the scrubber is about 6 in
water.
The purified gas exits the scrubber at 125° F, is reheated, and
vented to the stack.
The water-fly ash slurry drains from the venturi section of
the scrubber into a split stream; one stream is recycled to the scrubber,
-------
@KEMICO
I
fO
M(O FROM ACID PLANT
tygSCfe TO ACID PLANT
FIGURE B-4. MgO SCRUBBING HIOCESS FLOW SHEET
(60)
-------
B-23
the other flows to a thickener. The thickener bottom, containing princi-
pally fly ash with some insoluble impurities, is pumped to a settling
pond; the thickener overflow is returned to the top of the scrubber.
The magnesium salt slurry drains into the sump from where a
slipstream is withdrawn and pumped to a centrifuge. In the centrifuge,
the solids are separated and the mother liquor is returned to the
scrubber. Makeup MgO slurry is added to the mother liquor stream. . Nor-
mally about 50 percent of the slurry slipstream are centrifuged. An 85
percent solids centrifuge cake, containing MgO, MgS04f7H20, MgSO-j'Sl^O, and
MgSO-'SH.O, is passed to a rotary dryer to dehydrate the crystals.
Dryer System
7H20 •*• MgS04 + 7H20
6H20 •*• MgS03 + 6H20
3H20 •*• MgS03 + 3H20
The dryer is direct fired and operates at about 700°F. The anhydrous
solids are conveyed to a silo for storage. They will be trucked to the
separate regeneration-acid plant.
At the regeneration facility, the dry product, containing about
85 percent MgSOo and 15 percent MgSO,, is fed to a direct-fired rotary
calciner (see Figure B-5). At 1700°F the MgSO. is converted to MgO
and S02. Crushed green petroleum coke is added to reduce the MgSO^. to
MgO.
Calciner
MgS03 -»• MgO + S02
MgS04 + \ C + MgO + S02 + \ C02
The 15 percent SO gas stream is used for the production of H SO,. The
regenerated MgO is returned to the scrubber facility as makeup.
Removal Efficiencies
The MgO process can remove 90 percent of the SO . Coupled with
an ESP, it can reduce the particle emission by greater than 99 percent.
-------
B-24
CONVEYOR
KB 80s
SILO
MgSOj
ELEVATOR
SOj GAS CLEANING
CONCENTRATED SOj GAS
SULFURIC ACID PLANT
CONVEYOR
STORAGE
MgO RETURN
FIGURE B-5. MgO REGENERATION PROCESS FLOW SHEET
(60)
-------
B-25
Wastes
The only waste stream is the fly ash slurry.
*.
Raw Material and Heat Requirements
Makeup magnesium oxide and coke are required in the scrubbing
and regeneration processes. If an acid plant is included in the overall
operation, makeup catalyst also is necessary. Fuel oil is required to
heat the dryer and calciner.
By-Product
The regenerated SC^ gas is utilized in the manufacture of I^SOA
About 1.33 Ib of sulfuric acid (98 percent) is produced per pound of S02
removed.
Advantages
(1) High SOo removal efficiency, 90 percent
(2) Absorption and regeneration steps can be separated
(3) Minimal solid wastes disposal problem.
Disadvantages
(1) Stack gas reheat may be required
(2) High energy requirements for drying and calcining
(3) High cost of MgO absorbent
(4) Absorption facility must be located near a regenera-
tion facility.
Development Status
After several pilot plant studies, a commercial size system
was constructed on Boston Edison's oil-fired 155-MW boiler in 1971. It
-------
B-26
operated on and off for 2 years. The spent magnesium salt was trans-
ported to the Essex Chemical Company's acid plant in Rumford, Rhode
Island, for regeneration.
To gain experience with a coal-fired facility, an MgO scrubbing
system was constructed at Potomac Electric Power Company's (PEPCO's)
Dickerson Unit 3, 190-MW boiler to treat half the flue gas, 295,000 acfm.
The system was placed in operation in September, 1973. Initial ope'ration,
debugging, and modifications have been made since then and the system is
now being analyzed for optimum operation. The SCL removal efficiency has
been in excess of 90 percent. PEPCO's estimate for the capital cost is in
excess of $100/kW in 1974 dollars. The operating cost is not available
due to lack of pertinent data.
Capital and Operating Cost
TVA estimated the capital cost of an MgO system for a
200-MW existing boiler burning 3.5 percent sulfur coal with on-site
regeneration facilities at $13.1 x 106 in 1972. This included $4.58 x 106
for the calcining and sulfuric acid manufacturing plants. The labor,
material, and utility requirements are shown in Table B-6.
( 62^
Shah and Quigley estimated the labor, material, and utility
requirements for an MgO system excluding the regeneration facilities and
the results are shown in Table B-7. They also estimated the capital
cost for a central regeneration station including calcination and sulfuric
acid plants (capacity, 1000 tpd of sulfuric acid) at $8.2 x 10 in 1972.
Table B-8 shows the labor, material, and utility requirements of the
regeneration station.
-------
B-27
TABLE B-6. LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR
INTEGRATED MgO SYSTEM<61)
Basis: 200 MW existing boiler
Coal fired, 554,200 tons/yr
3.5 percent sulfur in coal
Operation, 7,000 hours/yr
H_SO,, 46,600 tons/yr
Item
Quantity
Coal fired
Utility
Power, MW
Fuel oil, gal/hr
Process water, gpm
Material
MgO, tons/hr
Coke, tons/hr
Labor
Direct operation, men/shift
Maintenance
4
452
2,200
0.07
0.05
7 percent of capital cost
-------
B-28
TABLE B-7. LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR
MgO SCRUBBING PROCESS (excluding calcination
and acid production)^)
Basis: 600-MW oil-fired boiler
2.5 percent sulfur oil
Load factor, 65 percent
Removal efficiency, 90 percent
Fuel consumption, 4,500,000 bbl/yr
Item Quantity
Utility
Power 5.4 MW
Fuel oil 14.3 bbl/hr
Water 430 gpm
Material
Makeup MgO 0.074 tons/hr
Labor
Direct operation 2.3 men/shift
Maintenance 4 percent of capital cost/yr
-------
B-29
TABLE B-8. LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR,
CENTRAL REGENERATION AND ACID PRODUCTION PLANT"62'
Basis: Capacity, 1000 tons./day of H SO
Load factor, 100 percent
Operation, 330 days/yr
Item Quantity
Utility
Power 2 MW
Boiler feed water 88 gpm
Process water 28 gpm
Cooling water 417 gpm
Fuel oil 31.7 bbl/hr
Material
Coke 0.1 tons/hr(a)
Labor
Direct operation 8.3 men/shift
Maintenance 4 percent of capital cost
(a) This value seems small compared with that obtained
from TVA study
-------
B-30
Wellman-Lord Process
Developer/Manufacturer
The Wellman-Lord'S02 removal process was developed by Davy
Powergas in the late 1960's. The process is marketed internationally by
Davy Powergas.
Process Description
The Wellman-Lord process employs a sodium sulfite scrubbing
solution to remove SC^ from flue gas. Thermal regeneration is utilized
to recover the sulfite and produce a by-product stream,of 95 percent S02 •
Flue gas, at approximately 270°F is compressed by a booster fan
(see Figure B-6) and passed through a venturi scrubber to remove the
fly ash. The scrubbing liquid, containing the fly ash, flows into a recir-
culation sump from where it is withdrawn and recycled to the venturi. A
purge stream is continuously withdrawn from the sump and transported to
the fly ash solids handling area. The flue gas proceeds into a fire
stage absorber where it contacts a countercurrent 27 percent solution
of Na2S03, NaHS03, and Na^O^.
The SO reacts with S0~ to form HSO~; some oxidation of
S0~ to SOT occurs.
3 4
S02 + S0= + H20 -*• 2HSO~
S0= + 1/2 02 -*• SOj
The purified gas flows through a chevron mist eliminator, out of the
absorber, and to the stack. The flue gas, at 125°F, is generally not
reheated.
The spent scrubber solution flows from the bottom stage of the
absorber and separates into two streams. One stream, about 10 percent of
the total flow is sent to the purge treatment system for sodium sulfate
removal, the remainder is sent to a surge tank. From the surge tank, the
absorber slurry is heated in a heat exchanger and introduced to a
-------
FIGURE B-6. WELLMAN-LORD S02 RECOVERY PROCESS
-------
B-32
double-effect evaporator. Sodium bisulfite decomposes to sodium sulfite
releasing water and sulfur dioxide:
2HSO •»• SO + SO + HO
•J O £m L*
A disproportionation reaction takes place in the evaporator at high
temperatures:
6Na+ + 6HSO~ •*• 2Na_SO. + Na.S00, + 2S00 + 3H.O
3 24 223 2. 2
Sodium sulfate and thiosulfate (Na_S 0 ) formed in this reaction are
removed from the solution in the purge treatment system.
The overhead vapors, containing S09 and HO, are cooled and sent
to the SO- stripper. The stripper overhead vapor is cooled to reduce the
water content to 5 percent by weight. It is then heated, compressed and
sent to an acid or sulfur processing plant. Condensate from the stripper
is used to slurry the Na_SO. crystals in the dissolving tank. Either
NaOH or Na^CO. is added to the dissolving tank to make up for the sodium
lost in the purge stream.
The purge stream is first cooled in a heat exchanger followed
by additional cooling in a chilled vessel. It flows to an ethylene glycol
refrigerated crystallizer where the leas soluble Na2SO^ precipitates.
The crystallizer bottoms are transferred to a thickener. The thickener
underflow is sent to a centrifuge, and the thickener overflow is returned
to the crystallizer.
Wet cake from the centrifuge drops into a jacketed dryer
where any sodium pyrosulfite decomposes to sodium sulfite and SOot
Na2S2°5 "* Na2S03 * S02
S02 vapor is vented to the flue gas handling system. The
dried cake is transported to a storage bin.
Mother liquor from the centrifuge flows to a purge tank and
is returned to the absorber product surge tank. Normally about 50
percent of the Na2SO^ formed during abosrption and regeneration is
removed in the purge treatment system.
-------
B-33
Removal Efficiencies
The Wellman-Lord system is capable of removing 90 percent of
the SC>2 and 99 percent of the fly ash from flue gas. This system has no
NOX removal capability.
Raw Material and Heat Requirements
Raw materials required in Wellman-Lord process include soda
ash and water. Steam is required for the evaporator operation.
Materials of Construction
The absorber is made of a tile-lined carbon steel with 316 L
stainless steel internals. All flow lines and surge tanks are rubber
lined. The slurry pumps, heat exchangers, evaporator, stripper, and
purge system process equipment are 316 L stainless steel.
By-Product
A 95 percent S02 off-gas stream is produced. It may be utilized
in the production of sulfuric acid or elemental sulfur.
Wastes
There are two waste streams, the fly ash slurry from the venturi
and the purge stream. The rate of purging is estimated to be 0.6 Ib
solids content (32.5 percent)/Ib of sulfur removed.
Advantages
(1) Small waste generation
(2) High particle and SC>2 removal efficiency
-------
B-34
(3) Proven reliability and performance in commercial
installations (for oil-fired boiler only)
(4) No plugging or scaling in scrubber
(5) Low liquid to gas ratio, normally about 10 gal/10-*ft ,
Disadvantages
(1) High heat and energy requirements
(2) Complicated process
(3) 862 processing plant required
(4) Not a compact system
(5) Stack gas reheat may be required
(6) Not demonstrated on coal-fired boiler.
Development Status
This process has been applied almost exclusively to acid plant
tail gases and oil-fired boilers as shown in Tables B-9 and B-10. The
first commercial coal-fired power plant installations are under construc-
tion at Northern Indiana Public Service Company's 115-MW boiler in Gary,
Indiana, and Public Service Company of New Mexico's 700-MW boiler in
Fruitland, New Mexico.
Capital and Operating Costs
The mid-1974 capital cost estimate for a 500-MW coal-fired
boiler with facilities to reduce the off-gas S02 to elemental sulfur is
$33 x 10 . The cost for the same unit with an acid plant is
$30 x 10 . The labor, material, and utility requirements are ;
shown in Table B-ll.
-------
B-35
TABLE B-9. COMMERCIAL APPLICATIONS OF WELLMAN-LORD PROCESS
(63)
Company and Location
Olin Corporation
Paulsboro, NJ
Toa Nenryo
Kawasaki, Japan
Japan Synthetic Rubber Co.
Chiba, Japan
Standard Oil of California
El Segundo, CA
Allied Chemical Corporation
Calumet, IL
Olin Corporation
Curtis Bay, MD
Sumitomo Chiba Chemical Co.
Sodegaura, Japan
Japanese Synthetic Rubber
Yokkaichi, Japan
Kashima Oil Company
Kaahima, Japan
Chubu Electric Company
Nagoya, Japan
Start-Up Date
July, 1970
August, 1971
August, 1971
September, 1972
November, 1972
May, 1973
November, 1973
December, 1973
February, 1974
May, 1973
Type and Size of Plant
Sulfuric Acid Plant
700 tpd
Sulfur Recovery Plants
2 & 150 tpd each
2 Oil Fired Boilers
(70 MW equivalent)
Sulfur Recovery Plants
3 9 150 tpd each
3 Sulfuric Acid Plants
Total cap. 500 tpd
3 Sulfuric Acid Plants
Total cap. 1000 tpd
Oil-Fired Boiler
(125 MM equivalent)
Oil-Fired Boiler
(140 MW equivalent)
Sulfur Recovery Plants
2 @ 90 tpd each
Oil-Fired Boiler
220-MW power plant
.Operational
Time, years
4.0
3.0
3.0
1.11
1.9
1.2
0.9
0.8
0.7
1.4
-------
B-36
TABLE B-10. WELLMAN-LORD S02 RECOVERY SYSTEMS EITHER
UNDER DESIGN OR CONSTRUCTION
Company and Location
Confidential Client
Kawasaki, Japan
Kashlma Mutual Power
Kashima, Japan
Kuraray
Okayama, Japan
Mitsubishi Chemical
Mitzushima, Japan
Northern Indiana Public Service Co.
Gary, IN
Public Service Co. of New Mexico
Fruitland, NM
Standard Oil Co. of California
Richmond, CA
Standard Oil Co. of California
El Segundo, CA
Toa Nenryo Kogyo K.K.
Hatsushitna, Japan
Toyo Rayon
Nagoya, Japan
Toa Nenryo Kogyo K.K.
Type of Plant
Steam boiler
Oil-fired power plant
Oil-fired boiler
Oil-fired boiler
Coal -fired power plant
Coal -fired power plant
Claus plants
Glaus plant
Claus plant
Oil-fired boiler
Claus plant
Flue Gas Rate
Through Units, SCF
435,000
590,000
248,000
373,000
310,000
1,800,000
30,000
30,000
10,000
218,000
34,000
Arita, Japan
-------
TABLE B-ll. LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR
A WELLMAN-LORD PROCESS
Basis: 500-Mtf boiler
3.5 percent sulfur coal
100 percent load factor, 330 days/yr
Item
Quantity
Elemental Sulfur
Sulfuric Acid
Utilities
Power, MW
Steam, Ib/hr
Water, gpm
Material
Soda ash, tons/hr
Natural gas, 10 scf/hr
Labor
Direct operation, man/shift
Maintenance
OJ
16.0 ,
1.68 x 10-
4,900
0.66
114
4 percent of capital/hr
16.4 ,
1.88 x 10-
7,000
0.66
4 percent of capital/yr
-------
B-38,
Double Alkali Process (GM)
Developer/Manufacturer
GM developed the dilute double' alkali process installed at
their Parma, Ohio, auto plant.
Process Description
This process employs a dilute sulf ite buffered scrubbing solution
with lime regeneration. Calcium sulfate plugging is minimized by
softening with sodium carbonate. ' -
First the flue gas is treated for fly ash removal and. then intro-
duced into the bottom of a scrubber where it is contacted with a slurry
spray and cooled to prevent wet-dry interphase scaling (see Figure B-7).
The gas flows upward through three trays where SCL in the gas is removed by
reacting with a 0.1 molar caustic soda slurry at a liquor to ,gas ratio of
3 ' • ; .• .
20 gal/10 Ib. Each tray contains a series of floating bubble caps which
rise and fall with changing gas flow. The following reactions occur.
20H~ + SO = SOl5 + HO
2HSO~
S03 + 1/2 02 =
The normal pH of the scrubbing slurry is 5.0. System pressure
drop is about 7.5 inches water. The clean gas passes through a mesh mist
eliminator and is vented to atmosphere. Scrubbing slurry is pumped from
the bottom of the absorber where 75 percent is recycled to the top of the
absorber and the remainder is pumped to a mix tank. In the first mix tank
CaC03 converts bisulfite to sulf ite and CaSO- precipitates.
HSO~ + CaC03 = CaS03 + C02 + S03 + H20
In the second mix tank, lime is added to causticize Na SO- and Na,SOA.
; tL J *• *f
-------
Primary
Clarifier
Softener
Clarifier
Scrubber
Feed
VO
FIGURE B-7. GM DOUBLE ALKALI PROCESS FLOW SHEET
'(87)
-------
B-40
Ca(OH)2 + S03 = 20H + CaSO-
Ca(OH) + SO" = 2QH~ + CaSO,
/ 4 4
Because of the formation of Na.SO,, the regeneration section has to be
capable of regenerating both sulfate and sulfite. The sulfate reaction,
however, is quite difficult because of the relative solubility of the
product, calcium sulfate. In addition, the sodium sulfate cannot be
= - ++
caustic!zed in the presence of high SO or OH because Ca levels are
held below CaSO, solubility product. Thus, to provide effective sulfate
regeneration, the system must be operated at a dilute OH concentration
(<0.14M) while maintaining sufficient levels of sulfate (>0.4M) to effect
calcium sulfate precipitation.
The solution is pumped to the reactor-clarlfier where the
calcium salts precipitate out of solution. The clarifier underflow is
dewatered in a vacuum filter to about 50 percent moisture. The filter
cake is high in CaSO, with some CaSO,,, fly ash, and small amounts of
sodium contaminants.
Clarifier overflow, containing about 800 ppm of calcium ion,
and saturation amounts of sulfate, sulfite, and hydroxide ions, is pumped
to a second reactor clarifier. Soda ash is introduced to soften the
solution by precipitating CaCO_ which is recycled to the first mix tank.
COg + Ca(OH)2 = 20H~ + CaC03
The soda ash also serves to replace sodium lost to the filter cake. The
regenerated slurry, containing about 250 ppm of calcium, is recycled to
the scrubber.
Removal Efficiencies
The S0_ removal efficiency varies between 88 and 92 percent.
The system is capable of reducing particles from an inlet loading of 0.3
gr/scf to 0.05 gr/scf.
-------
B-41
Materials of Construction
The scrubbers are constructed of 316L stainless steel.
Wastes
The only waste stream is the filter cake. It contains primarily
CaSO, with fly ash, CaSO., and some sodium salts.
By-Product
None.
Raw Materials and Heat Requirements
As GM does not reheat the flue gas, no heat is required for the
process. Soda ash and lime are consumed as raw materials.
Advantages
There is insufficient information on the operating performance
of this process to quantitatively appraise it. Listed below are the
conclusions drawn from information obtained from GM performance reports
and other similar processes.
(1) High S02 removal efficiency
(2) No severe plugging or scaling problems.
Disadvantages
(1) High oxidation rate results in difficulty in
controlling pH of incoming slurry
(2) Waste sludge disposal requirement
(3) High lime and soda ash cost
(4) Dilute OH ion concentration requires circulating
large quantities of slurry.
-------
B-42
Development Status
Following a pilot plant operation in 1969, GM completed a full-
scale double alkali S02 removal facility on a 400,000 Ib steam/hour
capacity coal-fired boiler in Parma, Ohio. It was started up in March,
1974 and a 1-year in-depth evaluation of the total system is now in
progress. GM intends to apply a similar system to GM's other industrial
boilers.
Capital and Operating Costs
GM estimated the cost of the double alkali system for a coal-
fired boiler equivalent to 32 MW at $3.5 x 106 in mid-1973. (87) The
labor, material, and utility requirements were estimated as shown in Table
B-12.
-------
B-43
TABLE B-12. LABOR, MATERIAL, AND UTILITY ,BEQUTREMENTS
FOR GM DOUBLE ALKALI PROCESS *• '
Item Quantity
Utility
Power 400 kW
Steam 2,700 Ib/hr
Water 21.4 gpm
Material
Lime 0.23 tons/hr
Soda ash 0.05 tons/hr
Carbon dioxide 0.007 tons/hr
Polymer 0.03 Ib/hr , v
Supplies 1 percent of capital cost/yr
Labor
Direct operation 1.4 men/shift
Maintenance 0.5 man/shift
(a) This value was assumed.
(b) The operating load factor was 0.47.
-------
B-44
Chemlebau Process
Developer /Manufacturer
The Chemlebau process was developed by Reinluft GmbH, Essen,
Germany, and acquired by Chemlebau - Dr. A. Zieren, GmbH, Cologn, Germany,
in 1967. Commonwealth Associates, Jackson, Michigan, has acquired 'the
western hemisphere licensing rights.
Process Description
The process employs moving beds of lump char to remove SO. from
(88) ' 2
flue gas. ' The absorbent is thermally regenerated, producing a con-
centrated 20 percent SO- gas stream. A schematic process flow sheet is
shown in Figure B-8.
The flue gas, following treatment by a mechanical dust collector,
is introduced into the bottom of the adsorber. The adsorber is a steel
shell containing vertical shafts through which walnut-size lumps of char-
coal flow. Baffles and louvers are installed to direct the gas cross
counter current against the downward moving beds of char. At an optimum
operating temperature of 250 to 300°F, the S02 is adsorbed by the dry char
and catalytically converted to SO.. The SO, reacts with water to form
H0SO. which condenses within the char.
24
S0 + 1/2 0 + S0
The purified gas exits froia the top of the adsorber, passes through an
ID fan and is vented to the stack.
Conveyor belts and bucket elevators transport the acid-laden
char from the bottom of the adsorber into the top of the desorber. The
desorber unit is similar in structure to the adsorber; char flows down
through vertical shafts where it is heated to about 700 °F by a cross-
countercurrent flow of inert scavenger gas. Sulfuric acid dissociates
into S03 and water, and the S03 reacts with carbon to form SO. and CO..
-------
B-45
GAS TO ATMOSPHERE
ADSORBENT
MAKE-UP
STORAGE
AS
LOW EFFICIENCY
DUST COLLECTOR
r
i
i
i
i
t>
--•i
rj
1
.
1
^1
1
DESORBER
,.<_ — .-
**i
1
i
ACTIVATED
CARBON
HEATER
S02
PRODUCT GAS
RECIRCULATING
BLOWER
FIGURE B-8. CHEMIEBAU FLOW SHEET
(88)
-------
B-46
2S03 + C ->• C02
The scavenging gas, containing about 20 percent SO., passes from
the desorber into a circulating fan. At the fan discharge, a portion of
the gas is withdrawn to a sulfur recovery system at a rate equivalent to
the amount of SO. generated in the desorber. The S0? product recovery
stream is reheated and returned to the desorber. Regenerated adsorbent
is,discharged from the bottom of the desorber, screened to remove the
fine particles, and returned to the adsorber. Makeup char, stored in the
char bin, is added to the adsorber to replace the char lost due to
mechanical attrition and reaction with SO..
Removal Efficiencies
. The Chemiebau system is capable of a 95 percent S0? removal
efficiency. Fly ash can also be removed by the system and has no adverse
effect on the adsorbent activity. Although the specific information is
not available, the system has appreciable NO and halogen removal
X
capability.
Wastes
There are no waste streams.
Materials of Construction
The adsorber and desorber are constructed of carbon steel. The
conveyor system from the desorber must be constructed of materials capable
of handling the hot regenerated char particles.
-------
B-47
By-Product
In addition to the 20 percent SO- off-stream produced, the char
removed in the screening process is very active and can be marketed as
activated carbon. About 0.6 Ib of activated carbon would result per
pound of SO. removed.
Raw Material and Heat Requirements
Adsorbent char is required in the adsorber at a rate of about
20 Ib per pound of S0? removed. About 4 percent of the char would be lost
per cycle due to mechanical attrition and reaction with S0_, and, thus,
the makeup char must be added for the loss. Low cost lignite serves as
the adsorbent char.
Direct fired heat is required to reheat the scavenger gas from
500° to 750° F for adsorbent regeneration. Assuming an 80 percent
heater efficiency, the heat requirement would be 5,000-6,000 Btu/lb of
S00 removed.
L
Advantages
(1) No flue gas reheat
(2) Capable of removing NO and halogens
A
(3) No waste streams
(4) Good turndown capability
(5) Possible credit for activated by-product.
Disadvantages
(1) High char attrition rate
(2) Low space velocities require massive adsorber and
desorber vessels
(3) Requirement of concentrated S02 processing system.
-------
B-48
Disadvantages
(1) High char attrition rate
(2) Low space velocities require massive adsorber and desorber
vessels
(3) Requirement of concentrated S0_ processing system.
Development Status
The most recent installation was a 10-MW pilot plant test facility
at Kellerman Power Station, Lunen, Germany, 1966-68. Commonwealth Asso-
ciates, Jackson, Michigan, is the Chemiebau licenser for the Western Hemi-
sphere. To date no Chemibau processes have been sold in the United States.
Capital and Operating Costs
Figure B-9 shows the estimated capital costs for the Chemiebau
/QQ\
system in November, 1973. The estimates were based on field erection
costs of two 50 percent capacity trains ready to run with initial loading
of adsorbents. The cost included the mechanical conveying equipment, a
char storage bin, an 80 percent efficiency mechanical dust collector, and
the equipment to reduce the S0_ to elemental sulfur. The operating labor,
material, and utility requirements for a Chemiebau process including
elemental sulfur production system installed on a 100 MW equivalent boiler
are shown in Table B-13. A credit may be taken into consideration for a
by-product activated carbon produced at a rate of 2.16 tons/hr.
-------
B-49
$/kW
POWER PLANT CAPABILITY IN MEGAWATTS
$/kW
80
70
60
50
40
30
20
10
0
12«SULFUR IN COAL
100 200 3UO . 4UO 500 GOO
POWER PLAMT CAPABILITY IN MEGAWATTS
7UO
800
FIGURE B-9. CAPITAL COST FOR CHEMIEBAU PROCESS - INCLUDING 30 DAYS CHAR SUPPLY
AND SULFUR REDUCTION - TWO 50 PERCENT TRAINS AT A NEW INSTALLATION(88)
-------
B-50
1 TABLE B-13. LABOR, MATERIAL, AND UTILITY REQUIREMENTS
FOR CHEMIEBAU PROCESS(89>
Basis: Coal-fired boiler (100 MW equivalent)
4.5 percent sulfur in coal (SO' cone., 3,600 ppm)
95 percent removal efficiency
85 percent load factor
Heating value, 12,000 Btu/lb
Item Quantity
Utility
Power 1.3 MW
Fuel oil 1,850 Ib/hr
.Raw Material
Lignite 2.99 tons/hr
Labor
Direct operation 1.5 men/shift
Maintenance 4.5 percent of capital cost
-------
8-51
Foster Wheeler (FW)
Developer /Manufacturer
The FW process for SO removal is a combination of char adsorption
and regeneration processes developed by Bergbau Forschung, GmbH, and
elemental sulfur conversion process developed by Foster Wheeler. Poster
Wheeler currently markets the process in the United States.
Process Description
After treatment for removal of particulate matter, the flue gas,
at about 300°F, is introduced into the adsorber (see Figure B-10) . The
adsorber contains vertical parallel louver beds through which the char
flows. The char moves in a plug flow fashion with the flow rate controlled
by a vibratory feeder located at the bottom of each bed, and the flue gas
passes through the adsorber bed in a cross flow. A portion of SO adsorbed
in the char is converted to H-SO, by the following reaction.
S0 + 1/2 0 + S0
The acid laden char pellets flow from the bottom of the adsorber where
they are screened for fly ash and are conveyed to the top of the regenera-
tion unit.
In the regenerator the char is heated by mixing with hot sand
at 1500°F. The sand serves as an inert heat transfer media. The following
reactions take place.
+ C + 2S02 + C02
S02 gas is liberated and the regenerated char pellets are discharged from
the bottom of the vessel to the separator where they are separated from the
sand, cooled, and recycled to the adsorber.
-------
CLEANED FLUE
GAS TO. STACK.
I
Ol
SULRUR
-------
B-53
The concentrated SO. gas stream is directed to the Foster Wheeler
off-gas treatment system where SCL reacts with crushed coal and is reduced
to elemental sulfur.
C + S02 -> CO + S
The resulting gases enter a condenser where the sulfur is condensed and
stored in a heated tank. The remaining gases are recycled to the adsorber
to capture any remaining S0_.
Removal Efficiencies
The system is capable of removing 86-95 percent of the S02, 90-95
percent of the particulate matter, and 40-60 percent of the N0x>
Raw Materials and Heat Requirements
About 0.14 Ib char is lost per pound of SO. adsorbed. These
losses are primarily due to the production of CO. in the regenerator
(about 90 percent) with the remainder (about 10 percent) attributed to
mechanical attrition. Twice annually, the entire char system is replaced
by new char.
Heat is required to heat the sand to 1500°F for the regeneration
process.
Wastes
None.
Advantages
(1) No flue gas reheat
(2) Production of marketable sulfur by-product
(3) Good turndown capability
(4) Capable of removing NO^
-------
B-54
(5) No waste streams
(6) Coal is the reducing agent for sulfur production. .
Disadvantages
(1) Char must be replaced twice annually
(2) Solids handling equipment presents possible
maintenance problems
(3) Low space velocities require massive adsorber
and regenerator vessels.
By-Product
About 0.5 Ib of elemental sulfur is produced per pound of S0»
removed.
Development Status
Bergbau-Fbrschung started up a demonstration unit in early 1974
at the Kellerman Power Plant in Lunder, West Germany. The unit processes
flue gas equivalent to about 35 MW as a slip stream from a 350-MW coal-
fired boiler. The off gas is treated in a Glaus reactor. Foster Wheeler
is installing a demonstration unit at Gulf Power Company's Scholz Steam
Plant on a 47.5-MW coal-fired boiler. The adsorption section is designed
for half load (50 percent of flue gas flow) and the regeneration section
is designed for full load so that higher sulfur coal can be tested. The
1-year test program is under way.
Capital and Operating Costs
The capital cost for a turnkey installation on a 500-MW boiler
system, burning 3 percent sulfur coal was estimated at $55 to $70/kW in
mid-1974. (91) The labor, material, and utility requirements for a 20-MW
equivalent system are shown in Table B-14.
-------
B-55
TABLE B-14. LABOR, MATERIAL, AND UTILITY REQURIEMENTS FOR FW PROCESS
Basis: 20-MW coal-fired boiler
3 percent sulfur in coal
100 percent load factor
Heating value, 12,000 Btu/lb
90 percent removal efficiency
(91)
Item
Quantity
Utility
Power
Fuel oil
Material
Makeup char
Char
500 kW
500 Ib/hr
126 Ib/hr
Labor
Direct operation
Maintenance
1 man/shift
4 percent of capital cost
-------
B-56
Westvaco Process
Developer/Manufacturer
Westvaco, Charleston Heights, South Carolina, developed and
markets the Westvaco S0? process.
Process Description
The Westvaco SO. process uses activated carbon to adsorb dilute
SO- from flue gas. Upon regeneration of the carbon, H_S is utilized to
reduce the entrained H^SO, to elemental sulfur.
Following treatment for particle removal, the flue gas is intro-
duced to a five-stage activated carbon fluidized-bed adsorption unit (see
Figure B-ll). SO^ is removed through catalyzed oxidation to SO- and a
subsequent hydrolysis to sulfuric acid which remains adsorbed in the
carbon particles.
S02 + 1/2 02 +'H20 -> H2S04
Sufficient water vapor and oxygen are present normally in the flue gas for
the reaction. The purified gas exits through the top of the adsorber and
is vented to the stack.
The acid loaded carbon is transferred mechanically to the sulfur
generator where, at about 300°F, it is contacted by a stream of H^S. The
H-S reduces the H SO, to elemental sulfur. In general, hydrogen sulfide
is not available in all industrial boilers, and therefore, the following
reaction is added.
activated
carbon
3H2 + AS -» 3H2S + S
The reaction takes place in the H^S generator/sulfur stripper at temperatures
near 1000°F. The hydrogen required may be possibly supplied through a
gasifier utilizing coal or other fossil fuels. The mixture of H S and
sulfur vapor leaves the H^S generator/sulfur stripper and passes to the sulfur
-------
SPENT GAS
TO BOILER
H2504 *" s * *
RECYCLE
SULFUR
CONDENSER
,. SULFUR
PRODUCT
FIGURE B-ll. WESTVACO SO, REMOVAL PROCESS
-------
B-58
condenser where liquid sulfur at 270°F is separated from the H S. The
sulfur is filtered to remove dust, solidified, and stored. H S is recycled
to the sulfur generator. The regenerated carbon is cooled to 300°F and
returned to the absorber.
Removal Efficiency
The. system is capable of removing 90 percent S07 from flue gas.
By-Product
Elemental sulfur (99.7 percent pure) is the by-product. A 15-
MW
(93)
produce about 419 Ibs/hr of sulfur.
boiler, generating about 30,000 scfm of 3,300 ppm SO- flue gas would
/ ft A \ 4b
Wastes
None.
Heat and Material Requirements
Some carbon is lost due to mechanical attrition, normally less
than 1 percent per cycle. A 15-MW installation, circulating 6,600 Ibs
carbon/hr would require about 25 Ibs carbon/hr of makeup. Coal would be
consumed in the gasifier to generate the H_ reduction stream. About 716
Ibs of coal/hr would be required for the 15-MW system. Sixty three gallons/hr
of No. 2 fuel oil would be consumed in the H^S generator/sulfur stripper.
Advantages
(1) Production of marketable by-product, elemental sulfur
(2) High S02 removal efficiency (90 percent)
(3) No stack gas reheat
(4) No waste stream generated.
-------
B-59
Disadvantages
(1) Extensive solids handling
(2) Complicated process
(3) Requirements of hydrogen and fuel oil in the process.
Development Status
The Westvaco process was originally designed for Claus plant
tail gas applications. Westvaco has recently completed pilot plant tests
2
on a 20,000 ft /hr flue gas stream from an oil-fired boiler. They are
interested in evaluating their system on a coal-fired boiler and have
designed a 15-MW prototype unit. They are actively pursuing paths for the
installation of such a unit.
Capital and Operating Costs
The capital cost for a 15-MW battery limit installation was
estimated at $2.4 x 10 by Westvaco in August, 1974. ^93^ The cost included
construction expense, contractor's fee, and contingency. The labor,
material, and utility requirements for the installation are shown in
Table B-15. (93)
-------
B-60
TABLE B-15. LABOR, MATERIAL, AND .UTILITY REQUIREMENTS
FOR WESTVACO PROCESS1- '
Basis: 15-MW coal-fired boiler
Flue gas flow rate: 30,000 scfm
S0? concentration, 3,230 ppm
Item Quantity
Utility
Power 670 kW
Fuel oil (No. 2) 63 gal/hr
Steam 12,000 Ibs/hr
Cooling water 500 gpm
Material
Activated carbon 25 Ibs/hr
Coal 0.358 tons/hr
Labor
(a)
Direct operation 2 men/shift ,, •>
Maintenance 4 percent of capital cost
(a) Engineer, 1 man/shift; technician, 1 man/shift.
(b) This value was assumed.
-------
B-61
Sulfacid Process
Developer/Manufacturer
The Sulfacid process was developed by Lurgi of Frankfurt, West
Germany, and is licensed in the United States to the Rust Engineering
Company, Birmingham, Alabama.
Process Description
The Sulfacid process utilizes an impregnated carbon bed to adsorb
and oxidize sulfur dioxide, and water to reactivate the bed. ' As can
be seen in Figure B-12, stack gas is pretreated to adjust the temperature,
humidity, and particle content in a humidifying chamber or a venturi
scrubber. The conditioned gas with a temperature of 120°F to 175°F, a
dew point of about 120°F, and particle loading of less than 0.007 grain/scf
flows upward at low velocity through a bed of carbon-based catalyst of 1
to 2 feet deep. Sulfur dioxide, oxygen, and water are adsorbed on the
impregnated carbon where sulfuric acid is formed by the reaction
S02 + 1/2 02 + H20 + H2S04.
The acid is washed from the bed by a continuous spray of water as it is
formed. The product acid flows continuously from the reactor as a 10
to 15 percent solution. This solution can be used to quench high
temperature gas streams, thereby increasing its concentration up to 20
to 30 percent, if desired.
Removal Efficiency
A sulfur dioxide removal efficiency of 90 percent can be achieved
by the process. Sulfur dioxide removal efficiency is a function of catalyst
depth, making the system amenable to efficiency upgrading, if necessary,
after installation.
-------
TO STACK
CLEAN GAS
RAW GAS
WATER
STEAM
INJECTION WATER
MIXING
CHAMBER
T
CONDENSATE
td
i
FIGURE B-12. SULFACID PROCESS
-------
B-63
Material and Heat Requirements
For the flue gas characterized in Table B-16, about 55.2 gal/
min and 42.5 gal/min of process water are required to generate 12.5
percent and 25 percent acids by weight, respectively. The material balance
was based on a 90 percent sulfur dioxide removal efficiency. Heat is
required to reheat the flue gas. About 1,100 Btu are consumed per 1000
scfm of flue gas flow rate.
By-Product
The Sulfacid system handling the flue gas characterized in
Table III-13 produces about 152.4 ton/day of 12.5 percent sulfuric acid
or 76.2 ton/day of 25 percent sulfuric acid. If there is no market
available for the acid, it should be neutralized with limestone for
disposal.
Wastes
Residual sulfur dioxide emission is the only emission resulting
from the process. If the by-product sulfuric acid should be neutralized
by limestone or lime for disposal, the resulting waste sludge for the
flue gas in Table B-16 would be about 52.9 ton/day (solids content of
50 percent).
Advantages
The following advantages of the process have been reported.
(1) Simple regeneration
(2) Low operating cost (simple operation)
(3) High reliability (less moving parts)
(A) Adjustable removal efficiency
(5) Simple in overall process.
-------
B-64
TABLE B-16. FLUE GAS CHARACTERISTICS FOR SULFACID PROCESS
Item Value
Flow rate 37,900 scf/min
Temperature 510°F
SO concentration 3,000 ppm
Water vapor content 7.3% by volume
-------
B-65
Disadvantages
(94 95)
The following disadvantages of the process have been reported. '
(1) Generation of low concentration by-product acid to be
used or disposed of
(2) High water consumption
(3) Possible corrosion
(4) Potential cold, wet plume problem.
Development Status
The Sulfacid process has been commercially applied to the treat-
ment of chemical plant waste streams for several years in Europe. One
system in West Germany has been installed on a titanium dioxide recovery
process with a flue gas flow rate of about 20,000 cfm. Another system in
Holland has been installed on a sulfuric acid plant with a flue gas
rate of 20,000 cfm. No major problems have been encountered regarding
catalytic activity of impregnated carbon, and general operation and
maintenance for 7 years. A plant to handle 20,000 to 30,000 cfm of sulfuric
acid plant tail gas will be built in Pittsburgh, Pennsylvania for the
United States Steel Company. In general, the Sulfacid process lends itself
more to sulfuric acid plant tail gases because of the high volume of low
concentration sulfuric acid by-product.
Capital and Operating Costs
The total capital cost for a Sulfacid system handling a flue
( 95")
gas of 72,000 acfm was quoted as about $2 million. ' This excluded the
cost for a reheat system and assumed a low fly ash concentration in the
flue gas. The adjusted cost, including the cost for a reheater was given
as $2.1 million. The corresponding cost for a Sulfacid system handling
( 95^
a flue gas of 7,200 acfm was estimated at $520,000. The exponential
scale factor for total capital cost was estimated to be 0.6 and the con-
/QC\
struction time required to be about 6 months for custom design. The
labor, material, and utility requirements of the system handling of the
flue gas listed in Table B-16 are shown in Table B-17.
-------
B-66
TABLE B-17. LABOR, MATERIALS, AND UTILITY REQUIREMENTS
FOR SULFACID PROCESS
Item
Quantity
Utilities
Power
Process water
Steam (for reheat)
Material
(c)
Limestone
Makeup carbon
Labor
Direct operating
Maintenance
170 kW(a)
55.2 gpm for 12.5 percent
by-product acid
42.5 gpm for 25 percent
by-product acid. »
9,500 lb/hrw
25.2 tons/day
nil
(d)
1 man/shifte -
1 percent of capital cost
(a) The value was derived from Reference 94.
(b) This was obtained from Reference 94.
(c) It was assumed that limestone is used to neutralize
the by-product acid.
(d) This was obtained based on the use of 10 percent
excess limestone with a purity of 85 percent.
(e) The value was obtained from the Rust Engineering
Company. '
-------
B-67
Chiyoda Process
Developer/Manufacturer
The Chiyoda Chemical Engineering and Construction Company, Japan,
developed the "Thoroughbred 101" sulfur dioxide removal process. They
have designed and manufactured the systems for Glaus plants and oil-fired
boilers in Japan, and the process is marketed in the United States through
their Seattle, Washington office.
Process Description
Following treatment by an ESP for fly ash removal, the flue gas
is introduced into a venturi-type prescrubber where the gas is cooled and
any remaining particulates are removed (see Figure B-13). the gas passes
through a packed bed absorber where it flows upward contacting a counter-
current 2 to 5 percent sulfuric acid solution; containing about 2,000 ppm
of ferric sulfate. The SOL is absorbed by the acid Solution
S02 + H20 -> H2S03 .
Pressure drop through the venturi and scrubber is about 9 inches water and
the liquid-to-gas ratio in the scrubber'is about 300 gal/1000 scf. After
passing through a demister, the flue gas is reheated and vented to the
stack.
The scrubber effluent solution flows to the oxidizer tower,
where, in the presence of the ferric ion catalyst, air injected into the
liquor oxidizes the H SO to H SO .
H2S03 + 1/2 02 - H2S04
A portion of the liquor leaving the oxidizer returns to the absorber,
and the remainder passes to gypsum production. First; the liquor is
neutralized with lime or pulverized limestone to form insoluble gypsum.
H2S04 + CaC03 + H20 ->
-------
Cleaned gas
Reheater Absorber Oxidizer
Prescrubberj Filter
Sludge
Crystallizer
Limestone
Centrifuge
Air
Purge
to treatment
i
c^
co
FIGURE B-13. PROCESS FLOW DIAGRAM OF CHIYODA PROCESS
(96)
-------
B-69
Then, gypsum crystals are withdrawn from the bottom of the crystallizer
and centrifuged. Crystallizer overflow and mother liquor from the centri-
fuge are sent to a clarifier. The clarifier underflow is returned to the
crystallizer and the overflow is recycled to the absorber. The centrifuged
gypsum contains about 5 to 20 percent free water. It may be trucked away
and used in wallboard production.
Chloride ion accumulation in the scrubber corrodes the stainless
steel and a purge stream must be withdrawn to maintain the chloride content
below 200 ppm.
Removal Efficiency
The process is capable of a 95 percent SCL removal efficiency.
Raw Material and Heat Requirements
Heat is required to reheat the flue gas. Ferric sulfate is lost
with the gypsum and purge stream and must be replaced. Limestone is
consumed in the crystallizer.
By-Product
About 2.44 Ibs of gypsum containing 85 to 90 percent solids
(CaSO,*2H90, 97 percent; limestone, 0.6 percent; others, 2.4 percent) is
produced per pound of S0« removed.
Wastes
The fly ash filter cake and purge stream are emitted from the
process.
Advantages
(1) Relatively simple process
-------
B-70
(2) Proven performance on oil-fired boilers, large and small
(3) Production of gypsum instead of sludge
(4) Good SO- removal efficiency
(5) , No plugging or scaling in scrubber.
Disadvantages
(1) Flue gas reheat
(2) Poor market for gypsum in the U.S.
(3) Corrosion problem due to chloride ion accumulation
(4) High liquid to gas ratio and high pumping requirement
(5) Large absorber required. f
Development Status
Since 1972, 10 commercial Chiyoda Thoroughbred processes have
been installed on Glaus plants and oil-fired boilers in Japan. A 23-MW
pilot plant on Gulf Power Company's coal-fired Scholz plant in Sneads,
Florida will be completed in late 1974.
Capital and Operating Costs
The capital cost for a 250-MW Chiyoda process was estimated at
$80 to .$100/kW. ' For a Chiyoda process installed on a 30-MW boiler
burning high sulfur resid (3.5 percent sulfur), the labor, material, and
(97)
utility requirements are shown in Table B-18.
-------
B-71
TABLE B-18. LABOR, MATEKlAL, AND UTILITY MqUlKEMlSWTS AND
BY-PRODUCT PRODUCTION FOR CHIYODA PROCESS^?)
Basis: Oil-fired boiler (30 MW equivalent)
Sulfur content of oil, 3.5 percent
Flue gas, 53,000 scfm
SO- concentration, 1,450 ppm
Gas temperature, 340 F
Item
Quantity
Utility
Power
Steam
Water
Material
Limestone
Labor
Direct operation
Maintenance
By-Product
Gypsum (10 percent free water)
760 kW
11,900 Ib/hr
25 gpmu;
0.53 tons/hr
1 man/shift
3.5 percent of capital cost
0.95 tons/hr
(a)
(a) This value was assumed.
-------
B-72
Ammonia Scrubbing Process (Peabody)
Developer/Manufacturer
The Peabody Engineering Company, Stamford, Connecticut, designs
and constructs ammonia-based sulfur dioxide removal systems.
Process Description
Small industrial boiler facilities are not equipped to process
the by-products, sulfur dioxide, and ammonium sulfite generated by thermal
stripping or acidification of the spent ammonium scrubbing slurry. The
following process is a proposed method for ammonium scrubbing with the
generation of a marketable, easily handled by-product ammonium sulfate
fertilizer. *• ' The complete integrated process has not been tested, but
the individual process operations (absorption, oxidation, and evaporation)
are straightforward and are in common use in the ammonia industry.
The flue gas is first treated with an electrostatic precipitator
or mechanical separator to remove the fly ash and particulate matter. It
is cooled from 300 to 170°F in a water-cooled heat exchanger and introduced
into the bottom of a four-stage absorber (see Figure B-14). First the
remaining fly ash is scrubbed with water and the solids slurry is withdrawn
and transported to a settling pond. The gas flows through the next three
trays where it is scrubbed countercurrently with a mixture of ammonia,
ammonium sulfate, ammonium bisulfite, and ammonium sulfate. The sulfur
dioxide in the gas reacts with ammonia and ammonium sulfite in the solution
to form ammonium bisulfite; oxidation of ammonium sulfite to ammonium
sulfate also occurs.
NH.. + H00 + S00 = NH.HSO,
o 2 2 4 3
-------
Water
Cooler •
•^ «ii
4
Rehe
$
v
Si
To
Stack
250 F
120 F
Flue
Cas
300 F
Scrubber
170 F
NH4HS03 ftnd
(HHA)2S03
Oxldlrer
Double-Efface Evaporator
Air
Tank Heutrallzer
Thickener
__,^ Aah Slurry
"^ To Pond
Centrifuge
(HH4)2S04
Crystals
FIGURE B-14. AMMONIA SCRUBBING WITH OXIDATION TO AMMONIUM SULFATE
-------
B-74
Peabody impingement-type trays are used, with each tray having a
separate circulation system. The flue gas passes upward through a
mist eliminator to capture any vaporized ammonia, and exits through the
top of the scrubber. It is reheated from 120 to 250°F by passing through
a water-heated heat exchanger; cooled water from the heat exchanger is .
recycled to the flue gas precooler heat exchanger. The heated effluent
gas is vented to the stack.
Spent absorbent slurry containing about 50 percent ammonium
salts is withdrawn from each tray and pumped to a surge tank. From the
surge tank the slurry is pumped to the neutralizer where ammonia is added
to convert the bisulfite to sulfite to minimize sulfur dioxide loss during
oxidation.
Water is also added to prevent ammonium sulfate crystallization in the
oxidizer. The neutralized solution is introduced to the oxidizer where
the sulfite is reacted with air at 100 psig and 185°F to produce the
sulfate.
(NH4)2S03 + 1/2 02 + (NH4)2S04
Heat of reaction is removed by circulating the solution through a water-
cooled heat exchanger to maintain the temperature at 185°F. Temperature
control is necessary to maintain the: solubility of oxygen. A slip stream,
with about 40 percent solids, is withdrawn from the cooling stream and
pumped to a double-effect vacuum evaporator crystallizer. The crystals are
separated in a centrifuge and the liquid phase returned to the evaporator.
The ammonium sulfate cake (70 percent solids) may be conveyed, to a storage
bin or dried in a spray dryer.
Removal Efficiency
Greater than a 90 percent removal efficiency of sulfur dioxide
can be obtained with ammonia scrubbing. The mechanical separator or low-
efficiency ESP coupled with the scrubber can remove up to 99.5 percent of
-------
B-75
fly ash and particulate matter. The system has no NO removal capability.
A
Raw Material and Heat Requirements
Ammonia and process steam are required. Utilizing a water re-
circulated gas cooling and heating system, the heat requirements for
reheating the flue gas can be eliminated. Steam, however, will be required
in the double-effect evaporator.
By-Product
The usable by-product from the process is ammonium sulfate.
Wastes
The only solid waste is the fly ash removed from the scrubber.
If the scrubber system is not equipped with a highly effective demister,
the stack gas would contain considerable amounts of ammonium sulfite
compound.
Advantages
. (1) Good SO removal efficiency, 90 percent or greater
(2) No scaling or plugging problems in scrubber
(3) Production of marketable by-product, ammonium sulfate
fertilizer
(4) No regeneration of absorbent required.
Disadvantages
(1) High cost of ammonia absorbent and questionable
availability
(2) Unstable market for ammonium sulfate fertilizer
(3) Flue gas reheat system required
(4) Loss of ammonia to stack and resultant "blue" plume
requires investment in costly Brinks mist eliminator.
-------
B-76
Development Status
The process has not been tested for, flue gases from coal-fired
boilers. However, the individual process operations have been utilized
in the ammonia industry for several years. The absorption unit is manu-
factured and marketed by Peabody Engineering and is used in several of
their ammonium sulfur dioxide removal systems. The Japan Engineering
Consulting Company (JECCO) developed the oxidation system; it has been
proven in large installations in Japan. Peabody Engineering
designed and constructed ammonium scrubbing systems primarily for paper
mills where the regenerated ammonium sulfite is used in the cooking
liquor (see Table B-19).
Capital and Operating Costs
The capital cost information was obtained from the result of
(99)
the TVA study. The estimated cost for a 200-MW coal-fired boiler
system was $5.089 x 10 in 1969. The labor, material, and utility
requirements are shown in Table B-20.
-------
TABLE B-19. PEABODY AMMONIA SCRUBBING IN COMMERCIAL OPERATION
Plant/
Location
Conserv. Chemicals,
Barton, Florida
Boise Cascade,
Salem, Oregon
Rayonier Quebec,
Port-Cartier,
Quebec
USS Agri-Chemicals,
Ft. Meade, Florida
Type of
Gas
Sulfuric acid
Tail gas
Flue gas from
black liquor
Flue gas from
black liquor
Sulfuric acid
Tail gas
Flow Rate,
acfm
108,000
122,000
400,000
174,000
Gas S02
Temperature, Inlet
F ppm
195 2,025
450 8.000
470 —
220 —
SO
Outlet
ppm
250
800
200
^M
Remarks
Scrubber effluent is used
in fertilizer production.
Startup in spring, 1974.
System produces (NH,)_SO
pulp cooking liquor.
Startup in 1972.
System produces (NlO-SO.
pulp cooking liquor.
Startup expected soon.
Contract signed.
DO
I
-------
B-78
TABLE B-20. LABOR, MATERIAL, AND UTILITY REQUIREMENTS FOR
PEABODY AMMONIA SCRUBBING PROCESS <")
Basis: 200-MW boiler
3.5 percent sulfur in coal
554,400 tons/yr coal burned
7,000 hrs/yr operation
Item Quantity
Utility
Power 3.27 MW
Steam 10,900 Ib/hr
Water 1,848 gpm
Material
Ammonia 1.58 tons/hr
Labor
Direct operation 1.45 man/shift
Maintenance 20 percent of capital cost/yr
-------
B-79
Ammonia Scrubbing Process (Catalytic. Inc.)
Developer/Manufacturer
Catalytic, Inc., a subsidiary of Air Products and Chemicals,
Inc. , developed the ammonia-based S0? scrubbing system. The system
utilizes the Institut Francais du Petrole (IFF) reducing process to"
regenerate the spent ammonium salts and produce sulfur. The integrated
process is marketed in the United States by Catalytic.
Process Description
Following treatment by an ESP for fly ash removal, the flue gas
is cooled and water saturated by passing through a venturi scrubber (see
Figure B-15). The water-fly ash slurry is thickened, neutralized with
lime, and either filtered or pumped to a disposal facility. From the ven-
turi the flue gas flows upward through the absorber where it contacts a
14 mole percent slurry -of (NH^SCy (NH^SO^, and NH^HSCy The following
reactions occur:
There is normally about 10 percent oxidation of ammonium sulfite to
ammonium sulfate.
The adsorber is a cylindrical column containing 3 or 4 floating
cap trays, each with an individual circulation system. Ammonia concen-
tration, pH, and liquor to gas ratio (normally about 5 gal/1000 cf) in
the column are controlled to eliminate the "blue" plume from the scrubbing
systems. The purified gas flows from a mist eliminator through a blower
and to the stack. About 5 ppm of ammonium sulfite compounds exit with
the flue gas,
-------
A"
Flue gas
from ESP
Venturl
To Stack
Fuel oil
Methane
NH_
1
nmonla
^^
Si
scor£
irge
ige
^^
^^
i
\
3
t
:NH^
^ t.
^2S04 Re
Air
Air
CO gas
Water
Condenser
Evaporator
Sulfate
Reduction
Sulfur
Claus
Reactor
Generator
Sulfur
Condenser
V
Sulfur
storage pit
oo
o
FIGURE B-15. FLOW PROCESS DIAGRAM --CATALYTIC AMMONIA SCRUBBING PROCESS<100>
-------
B-81
The spent scrubbing slurry is pumped from the bottom of the
absorber to a surge tank. From the surge tank it is pumped to an . .
evaporator where at 300 F and 35 psi the less stable sulfite and bisulfite
are converted to SO^ and NH_.; ..... • -:
The SCL and NH , gases are transferred to the
-------
B-82
By-Product
Elemental sulfur is produced from the process.
Wastes
A fly ash slurry waste stream is generated in the venturi pre-
scrubber.
Raw Material and Heat Requirements
Ammonia makeup is required to replace losses to the flue gas and
in the regeneration system. Fuel oil is necessary to generate the reducing
gas in the manufacture of H S. Steam heat is required in the evaporator
and sulfur generating systems.
Advantages
(l) Good SO- removal efficiency
(2) No waste sludge disposal problem
(3) Generation of marketable by-product sulfur
(4) No plugging or scaling.
Disadvantages
(1) High cost of ammonia
(2) High capital cost
(3) Entire system not proven on utility boiler.
Development Status
In 1970 IFP began marketing its Claus tail gas treating process.
It involves ammonia scrubbing coupled with the reduction-regeneration
-------
B-83
system. To date, seven installations have been constructed ; and all are
currently operating. The complete IFF SO. removal system is being
installed in a 35-MW utility boiler in France. •
In 1972 Catalytic, Inc., TVA, and EPA jointly evaluated the
feasibility of Catalytic, Inc. 's ammonia scrubbing system at .TVA' s Colbert
Station pilot plant. Catalytic guarantees the continuous., supply of .
ammonia to their customers and is actively engaged : in marketing their
ammonia-based scrubbing system coupled with the IFF regeneration process.
Capital and Operating Costs
,-. The capital cost for a 20-MW coal-fire.d boiler (3.5 percent
sulfur in coal) producing a flue gas of 42,000 scfm (2,500 ppm SCO was
estimated at $3 million in November, 1974. ^g cost was for the
complete battery limit system including engineering, royalties, site
development, construction, etc.
The annual operating cost for the 20-MW system was estimated at
$550,000 in November, 1974, not including the credit for by-product sulfur.
The detailed breakdown of labor, material, and utility requirements are
not available1.
-------
B-84
Shell Flue Gas Desulfurization Process (Shell FGD)
Developer/Manufacturer
Shell International Petroleum developed the process in the early
1960's. Universal Oil Products (UOP) purchased the licensing rights for
the U.S. from Shell in 1971 and is marketing the process to chemical
plants and utility boilers.
Process Description
The Shell FGD process employs a copper oxide-alumina adsorbent In a
fixed-bed reactor to adsorb SO- from the flue gas. The spent adsorbent
is regenerated with a stream of 30 percent H_ in steam to generate an off-
gas stream which can be processed to yield liquid S0_, sulfuric acid, or
elemental sulfur.
After passing through the economizer and a participate removal
system, the flue gas is passed through a blower and into the adsorber (see
Figure B-16). The adsorber is a fixed bed in which the flue eas flows
through open channels along side and in contact with the adsorbent material.
The absorbent is elemental copper supported on an alumina structure and
contained in unit cells.
Upon contact with the flue gas, the copper and any cuprous
sulfide (Cu_S) contaminants are oxidized to CuO and CuSO..
2 4
Cu + 1/2 02 •*• CuO
Cu0S + 5/2 00 .-»• CuO + CuSO.
22 4
The existence of Cu.S is undesirable as one-half of the copper is converted
to CuSO^ and is unavailable to participate in the acceptance reaction. The
SO reacts with the CuO to form CuSO,.
CuO + 1/2 00 + SO,, ->• CuSO.
22 4
When the adsorber becomes loaded with sulfur, it is arranged for regeneration.
A stream of hydrogen is pass*
Cu, tLO, and SO,, takes place,
A stream of hydrogen is passed through the bed and the conversion of CuSO, to
4
-------
ABSORBER
OFF-GAS
EXCESS
•TOPPED
WATER
REGENERATION CAS
ACCEPTANCE TIME: 120 MIN.
BOILER
FEED WATER
SO, TO
CtAUSUMIT
REGENERATION
OFF-GAS.400DC
STIAM
I
oo
FIGURE B-16. SIMPLIFIED PROCESS FLOW SCHEME OF SHELL FGD UNIT^102'
-------
B-86
CuO + H2 -»• Cu + I.
, + 2En ->• Cu + SO,
Two or more identical adsorbers are applied in cyclic operation to provide
for continuous processing of flue gas. The off gas S0_ stream is further
concentrated to 90 percent by removing the H^O and inerts and may be either
liquefied or processed in a Claus reactor to produce elemental sulfur.
Removal Efficiencies
The Shell FGD process is capable of removing 90 percent of the
SO. from flue gas. The process may also remove NO . It .has no particulate
fc X
removal capability.
Raw Material and Heat Requirements
About 0.1 Ib of H. is required in the regeneration process for
every pound of SO- recovered/ ' Heat is required to raise the temperature
of the flue gas to the optimum reaction temperature of 700 F. With the
recovery of some portion of the added thermal energy, the net reheat require-
ment is approximately 1 percent of the fuel input to the boiler.
By-Product
The by-product is elemental sulfur.
Waste
The only waste is stripping water containing 20 ppm S, of which
75 percent is present as sulfate.
Advantages
(1) Dry process, no flue gas reheat
(2) No waste sludge generation
-------
B-87
(3) Reliable operation
(4) Low utility requirement .
Disadvantages
(1) Requires SO™ processing plant
(2) Expensive for small systems
(3) Requires H. reduction gas
(4) Requires "hot" ESP
(5) Requires expensive reheat for retrofit.
Development Status
In the early 1960's Shell developed the Shell Flue Gas Desulfuri-
zation process. In 1967 a 400-600 scfm side stream was withdrawn from the
flue gas of an oil-fired boiler at the Shell refinery near Rotterdam,
Netherlands to evaluate reactor design, catalyst type, and operating
parameters. The system operated for approximately 20,000 hours. In mid-
1973, in Japan, a commercial-size system was constructed to process the
combined flow of flue gas from an oil-fired boiler and a Glaus plant tail
gas stream, a total flow of about 90,000 scfm. It is reportedly operating
satisfactorily.
Following short tests on a coal-fired boiler in Rotterdam to
assess the effects of fly ash on acceptor life, a processing system was
installed on a 1400 scfm slip stream at Tampa Electric Company's Unit No. 1
coal-fired boiler. The purpose of the tests are to evaluate acceptor life
under adsorption-regeneration cycling. The test module consists of only
one reactor; bottled hydrogen is used to regenerate the acceptor and the
regenerated off gas is vented to the stack. Testing began summer of 1974
and is expected to be completed by spring, 1975.
Capital and Operating Costs
The capital cost of a Shell FGD system for a small coal-fired
boiler (i.e., capacities less than 40 .MW) was estimated at about $100/kW
-------
B-88
capacity in 1974. This included costs for the elemental sulfur reduction
system. The operating labor, material, and utility requirements for a
Shell FGD system installed on an oil-fired boiler (30 MW) are shown in
Table B-21. This included costs for the elemental sulfur recovery
process operation.
-------
B-89
TABLE B-21. LABOR, MATERIAL, AND UTILITY REQUIRE-
MENTS FOR SHELL FGD PROCESS(104)
Basis: 30-MW oil-fired boiler
Sulfur content, 2.85 percent
88 percent S02 removal efficiency
Flue gas flow rate, 59,000 scfm
Onstream time, 8,000 hrs/yr
Item Quantity
Utility > • •
Power - 110 kW
Material
Catalyst $4/hr(a)
Hydrogen 200 Ib/hr
Labor
Direct operation 0.5 men/shift
Maintenance 5 percent of capital cost
(a) 1972 cost.
-------
B-90
Citrate Process
Developer/Manufacturer
The U.S. Bureau of Mines developed the Citrate Process for re-
moving SCL from industrial waste gas. The Morrison-Knudsen Company, Inc.,
Boise, Idaho, and Peabody Engineering Company, independently offers the
process on a commercial basis.
Process Description
The Citrate Process involves absorption of SCL by a solution
of sodium citrate, citric acid, and sodium thiosulfate followed by re-
acting the absorbed SCL with H.S to precipitate elemental sulfur and
regenerate the citrate solution.
Following treatment by an ESP or cyclone to remove fly ash, the
flue gas is passed through a humidifier (see Figure B-17). The humidifier
is a fiberglass-lined tower containing a section packed with 1-inch saddles
and a stainless steel mist eliminator. Water is used to cool the gas to
about 140°F and eliminate H_SO, mist and any remaining fly ash. The cleaned
and cooled gas stream passes upward through the packed absorber where it
contacts counter-currently a citrate solution (pH, 4.5). The SCL is re-
moved by the following reactions.
S02 + H20 = HS03" + H+
H+ + HCit" = H2Cit
As the absorption of SCL is pH-dependent, decreasing with de-
creasing pH, the citrate functions as a buffering agent. The spent citrate
solution flows to the sulfur reactor where H?S gas is bubbled through the
solution and reacts with the SO- in the aqueous solution. Although the
chemistry of the reaction is complex, the overall reaction is as follows.
S0 + 2HS - 3S + 2H0
-------
GAS CLEANING
AND
COOLING
Cleaned and
Flue
gas-
cooled gas
^ A ^
H20-
S02 ABSORPTION I SULFUR PRECIPITATION
I AND
I SOLUTION REGENERATION
To atmosphere
H2S GENERATION
Steam
to
vo
FIGURE B-17. GENERALIZED CITRATE PRCTCESS FLOWSHEET^105)
-------
B-92
The H-S can be generated by reacting sulfur with methane and
steam:
CH4 + 4 S + 2 H20 -* C02 + 4 H2S.
The l-37o solids slurry overflows to the effluent surge tank
from where it is pumped to the conditioner tank. In the conditioner tank,
the sulfur separates from the citrate slurry by floating to the surface.
The regenerated citrate solution is pumped to a. feed tank.
The sulfur product is withdrawn from the storage bin and pumped
through a heat exchanger where the sulfur is melted at 275°F. The molten
sulfur and citrate solution pass into a closed settler tank, where at 35
psi the molten sulfur settles out and is removed from the bottom. The
citrate solution is returned to the absorber feed tank.
Removal Efficiency
The process is capable of up to a 99% SCL removal efficiency.
Raw Material and Heat Requirements
Methane, citrate, and kerosene (only for the U.S. Bureau of
Mine Process) are consumed in the process. About 36,000 scf of methane
are required for process heating and production of H_S per ton of ele-
mental sulfur recovered. About 8.4 Ibs of citrate are lost per ton of
sulfur recovered. And about 90 Ibs of kerosene are lost per ton of
sulfur recovered due to volatization from the hot sulfur slurry in the
conditioner and skimmer. Various lower volatility oils can be used to
minimize the evaporation loss.
-------
B-93
By-Product
A 99.6-plus percent sulfur is generated.
Wastes
The humidifier slurry, containing fly ash and H.SO , is a waste
stream. It would require neutralizing with lime, thickening, and trans-
porting to a landfill. About 0.02 Ibs of sludge (50% solid) are generated
per pound of SCL removed.
Advantages
(1) Good S0_ removal efficiency, 90-plus percent
(2) No scaling or plugging in scrubber
(3) No major waste sludge generation
(4) Produces marketable, elemental sulfur.
Disadvantages
(1) Flue gas reheat required
(2) Requires H_S generation gas,
Development Status
In 1968, the Bureau of Mines began investigating the citrate
process at their Metallurgy Research Center in Salt Lake City. In
November, 1970, a 300 cfm pilot plant went on stream at the San Manuel
smelter in Arizona. It ran intermittently for 6 months. In February,
1974, the Bureau of Mines put on stream a 1000 scfm pilot plant processing
off-gas from a lead smelter in Kellogg, Idaho. The Morrison-Knudsen
Company, Boise, Idaho, constructed the facility. With a 5000 ppm influent
SO concentration, the system demonstrated up to a 99% SO removal efficiency.
-------
B-94
H_S regeneration gas was supplied from a storage tank. In September, 1974,
construction was completed on the H-S generation plant and it is presently
being tested.
On March 15, 1974, the construction of Pfizer-McKee-Peabody
citrate process pilot plant was completed at Terre Haute, Indiana. .The
scrubber treats 2000 scfm from a coal-fired boiler with an inlet SO con-
centration of 1000 ppm. A venturi s.crubber was used in lieu of tne
humidifier and an impingement plate scrubbing tower replaced the packed
adsorber to permit higher gas velocities. The sulfur separation was based
on the flotation principle, but no hydrocarbon addition was made. Between
March and September 1, the system was operated for 2330 hours. The av-
erage SO- removal efficiency was greater than 95%.
Capital and Operating Costs
The U.S. Bureau of Mines, Salt Lake City, Utah, estimated the .
capital cost of a citrate process applied to a 1000 MW coal-fired boiler
burning 3 percent sulfur coal at $36.39 x 10 in May, 1974. The
annual operating cost was estimated at $13.87 x 10 /yr.
-------
B-95
Calsox Process
Developer/Manufacturer
The Monsanto Company of St. Louis, Missouri, developed the
Calsox Process. They are actively engaged in marketing the process for
SCL control to utility lockers.
Process Description
The flue gas, after passing through a forced draft fan and a hu-
midifier/cooler, is introduced to the absorber (see Figure B-18). The
absorber is of cross-flow design, although a vertical tower could be used,
where the gas contacts a 0.5 weight percent enthanolamine water solution.
The enthanolamine has a high affinity for SCL and readily absorbs it.
RNH2OH + S02 - RNHSOO + H20
The purified gas flows from the absorber through a mist eliminator and to
a reheater before being vented to the stack. Pressure drop across the ab-
sorber is normally 2-5 inches of water.
The solution from the absorption system goes to a two-step pre-
cipitation system. First it is mixed with makeup absorbent and the liquid
from the thickener. The calcium ions in the latter lead to precipitation
of CaSCL. The resulting stream goes to a clarifier, where the filtrate
from the cake filtration is also added. The clear liquid from the clari-
fier is returned to the absorption system. The concentration of soluble
calcium in this liquid is low enough to avoid scaling in the absorption
system.
In the second stage, the clarifier bottoms are mixed with lime
to complete the precipitation of CaSCL. The resulting stream goes to the
thickener and then to the filter. The result is a cake which.contains
about 50 percent water, the dry portion being composed of 85-90 percent
CaSO , 10-15 percent CaSO,, and about 0.5 percent ethanolamine. The CaSO,,
-------
FROM
ID FAN
I ABSORBENT
MAKEUP
TO ATMOSPHERE
AIR
FORCED
DRAFT
FAN
FLUE
GAS
REHEATER
HUMIDIFIER I ABSORBER JDEMISTER
COOLER
SOLIDS—'
HANDLING
SYSTEM
FILTER BELT
RECYCLE
REACTOR
REGENERA
TION
REACTOR
THICKENER
POND WATER
CALSOX
SLURRY
TO POND
i
VO
ON
FIGURE B-18. CALSOX PROCESS
(106)
-------
B-97
an unwanted compound because of its high solubility in water, results .from
oxidation of the sulfite. The loss of ethanolamine in the cake is about
70 Ib/hr from a 125-MW boiler. The two precipitation steps provide a
countercurrent flow effect.
Removal Efficiencies
The Calsox Process can achieve a 90 percent SCL removal efficiency
and can remove about 80 percent of the fly ash exiting the ESP. The process
has essentially no NO removal capability.
X
Material and Heat Requirements
Lime is consumed in the precipitation reaction and heat is re-
quired to reheat the flue gas from the 120°F mist eliminator exit temperature
to the desired 170-190°F stack gas temperature. Moreover, some ethanolamine
is lost to the flue gas and filter cake.
Based on a 1:1 stoichiometry, a 125-MW facility treating flue
gas with an S09 concentration of 3,000 ppm would consume 3.6 tons of lime
(CaO) per hour. No information was available on the flue gas reheat re-
quirements . About 2 pounds per hour of ethanolamine is lost to evaporation
in the absorber; this combined with the amount lost to the filter cake
yields a total loss of 72 Ibs/hr for a 125~MW scrubbing facility. The cost
of ethanolamine is 18£/lb.
By-Product
No marketable by-product is generated by the Calsox Process.
Wastes
The only waste stream is the .CaSO , CaSO, filter cake. A 125-
MW boiler scrubbing facility would generate about 14 tons per hour of wet
cake.
-------
B-98
Advantages
(1) High SO removal efficiency
(2) No plugging or scaling in scrubber.
Disadvantages
(1) Flue gas reheat necessary
(2) Waste stream generation
(3) High cost of ethanolamine.
Development Status
A 2000 scfm (3000 acfm) pilot plant was operated during the
period from February through October of 1973 at a boiler owned by the
Indianapolis Power and Light Company. The target of 90 percent SCL re-
moval was met and several improvements in the process were worked out.
A thirty-day period of uninterrupted operation was achieved in May of 1973.
Prior to this (November-December, 1972) a portable, relatively unsophisti-
cated, 5000 acfm pilot plant was operated at a utility boiler for about a
month. No large-scale units have been operated to date. A design for a
125-MW boiler has been submitted to the Indianapolis Power and Light Company
and is now being evaluated.
Capital and Operating Costs
For a 125 MW unit which Monsanto has designed for Indianapolis
Power and Light Company, the investment, calculated to 1976, is estimated
to be 9 million dollars ($72/kW). This is not a minimum cost
plant and includes some extra equipment such as a separator for the
treated flue gas. Monsanto expects the investment to be 40-50 $/kW for
large optimized plants. The operating cost was given as 3 mills/kWh
for the plant.
-------
B-99
Aqueous Carbonate Process
Developer /Manufacturer
The Atomics International (AI) Division of Rockwell International
Corporation developed the Aqueous Carbonate Process in the early 1970' s.
Process Description
The Aqueous Carbonate Process (ACP) employs a solution of sodium
carbonate in a modified spray tower to remove SCL from flue gases. The
spent powder is removed by a cyclone separator or an ESP and may either be
disposed of (open loop) or regenerated (closed loop) to yield elemental
sulfur.
The flue gas, containing SO- and fly ash, is first introduced
into the spray tower where it co- currently contacts an atomized mist of
a 4 to 20 weight percent Na^CO, solution (see Figure B-19) .
Instead of spray nozzles, centrifugal wheels are employed to
circulate the droplets at a high velocity. The system normally operates
with a liquid/gas ratio of about 1/3 gal/1000 scf. The liquid flow rate
is determined by the flue gas temperature and the Na-C0~ concentration is
determined by the SO- concentration in the flue gas.
The S0_ reacts with the sodium carbonate to form sodium sulfite
(Na2SO ), and sodium sulfate (Na SO, ) .
S02 + NaC03 - Na2S03 + C02
The flue gas and particulate matter leave the dryer at about 160 °F,
at least 20 °F above the dew point, as the thermal energy of the flue gases
(at about 300°F) is sufficient to vaporize the water in the spray dryer
without saturating the gas. At these conditions, the flue gas requires
little or no reheat.
-------
FROM EXISTING PLANT UTILITIES
COOLING WATEH PIPING
TYPICAL EACH MACHINE
TERMINAL BOARD
SPRAY MACHINE
(6 REQUIRED)
ELECTROSTATIC
PRECIPITATOR
SOLUTION
STORAGE TANK
TO
REGENERATION
PLANT
• SOLUTION
FEED PUMPS
VACUUM BLOWER
PUMP
CLEAN GAS TO
ATMOSPHERE
ID FAN
o
o
EXISTING FLUE
GAS STACK
FIGURE B-19. SCRUBBER SUBSYSTEM
-------
B-101
The system may be operated.open loop, by simply disposing of
the dry waste products, or closed loop, regenerating the Na CO reactant.
Open loop operation is less costly, but the soluble Na^SO and Na.SO, salts
present a disposal problem. In closed loop operation, the scrubbing and
regeneration systems are independent and can be uncoupled and operated
separately.
The AGP regeneration system involves three chemical steps. First,
the sodium sulfite and sulfate are reduced to sodium sulfide with either
coke or coal at 1700°F. AI has developed a high temperature molten salt
reactor for this reduction step.
Na2S03 + 3/2 C - Na2S + 3/2 C02
Na0SO. + 2C - Na0S + 2C00
24 i 2
Second, the sulfide is dissolved in water and treated with a CO.-rich gas
to reform the Na»CO for recycle to the scrubber and to evolve a gas rich
in hydrogen sulfide. Technology similar to that used in chemical recovery
processes of the pulp and paper industry is used in this step.
Na2S + C02 + H20 -* H2S + 2Na+ + CO^
The third and final step is the conversion of the H«S to elemental sulfur
by a Glaus process. The tail gas from the Glaus plant is recycled to the
scrubber.
In the reducer, part of the coke or coal is burned to provide
the necessary heat. The molten mixture leaving the reducer goes to a
quench tank, where it is dissolved to yield the "green liquor." The off
gas from the reducer is used as the source of C0_ for subsequent carbonation
steps and as a source of process heat.
The green liquor is cooled and filtered to remove any carbon or
flyash present. The conversion of the Na~S in this liquor back to Na-CO,,
involves aqueous chemical processing steps which are also used in several
paper-making technologies. AI is working to develop a conversion process
specifically for the AGP and considers the results obtained so far quite
promising. However, as a backup for this step, they have also evaluated
-------
B-102
the Tampella process, which was developed for the pulp and paper industry
in Finland and is licensed in this country by the Babcock and Wilcox Com-
pany. The Tampella process will meet all the requirements for this step
of the AGP and represents proven technology and hardware. Under a con-
tract from AI, Babcock and Wilcox recently completed an engineering study
of the application of the Tampella process to this step of the AGP. The
results were quite satisfactory, and B6W is now prepared to design and de-
liver the equipment for a full-scale modular unit.
The Tampella process for converting Na~S into Na_CO is included
in the regeneration flowsheet shown in Figure 4. The precarbonator serves
to convert Na?S into NaHS by reaction with the CCL in the reducer off-gas.
Precarbonator: 2Na2S + CC>2 + H20 - 2NaHS + Na2C03
The carbonator produces an NaHCO» slurry by reaction of Na_CO with CCL ,
also obtained from the reducer off-gas. The precarbonated liquor and the
NaHCO» slurry are combined in the stripper, where steam is used to strip
H2S from the liquid phase. The stripped liquor is then pumped to a
crystallizer, where Na~CO *H_0 is produced for recycle to the carbonator.
The final product Na^CO, solution is returned to the scrubber after ap-
propriate dilution. Glaus plant feed gas from this process typically
contains 80-90% HLS after condensation of the water vapor.
Carbonator: Na2c°3 + H2° + G02 ~* 2NaHC03
Stripper: NaHS + NaHCO -» H2S + Na2CO
Crystallizer: Na2C03 + H20 - Na2C03 H^
Removal Efficiencies
The AGP can achieve greater than 90 percent S02 removal. Par-
ticulate removal is very high (> 99.8%) because of the electrostatic
precipitator. The process removes some NCv and NO, the overall NO re-
moval being about 5 percent.
-------
B-103
Material and Heat Requirements
Atomics International maintains that the flue gas normally would
not require reheating. The primary heat requirement for the closed loop
system is for heating the dry carbonate, sulfite, sulfate and fly ash
mixture to 1700°F in the regenerator. Coke is normally used in the re-
generator, as it serves as a reducer and furnishes the necessary heat for
maintaining the reaction. A 330-MW scrubbing facility would consume about
46,200 tons per year of coke for a boiler burning 3.5 wt. percent sulfur
coal (107).
By-Product
The closed loop process generates elemental sulfur by-product,
The 330-MW sample plant would generate about 32,000 tons of sulfur per
(107)
year.
Waste Streams
No liquid waste streams are produced. The open loop version of
the process yields a solid product consisting of Na^SO , Na.SO., fly ash,
and small amounts of Na2CO_. A 125-MW unit processing 3 percent sulfur
coal and having a raw flue gas ash content of 2 grain/scf would product
about 6.4 ton/hour of solid, with about 1.4 ton/hour of this being fly ash,
The closed loop version of the process produces no waste streams, since
the sulfur is recovered in elemental form.
Advantages
(1) No scaling or plugging
(2) Open loop version is a relatively simple
process
(3) High SO- removal efficiency
-------
B-104
(4) No flue gas reheat
(5) Low L/G
(6) No waste sludge disposal problem
(7) Good turndown capability
(8) Low pressure drop
Disadvantages
(1) Sodium salt disposal problem for open loop
version
(2) High coal or coke consumption
(3) Closed loop version requires Claxis plant
Development Status
The complete ACP has not been tested in a single installation,
but both the scrubbing system and the reduction reactor have been tested
separately. The rest of the regeneration system is considered proven
technology.
The scrubbing portion, or open loop version, was tested in 1972,
Both a 5- and 7-foot diameter spray tower were used to assess the system
performance with SCL inlet concentrations varying from 200 to 8,000 ppm
at the Mohave Power Generating Station in Laughlin, Nevada.
Reduction tests were conducted at Santa Susana on a reaction
vessel equivalent to a 3-5 MW size system. The studies demonstrated good
conversion efficiencies and the off-gas compositions obtained have been
suitable for use in the aqueous regeneration steps.
-------
B-105
Capital and Operating Costs
AI has made a detailed analysis of the investment and operating
costs for a 330 MW unit (two 165 MW trains)^ • The total investment
for the regenerative system is about 21 million dollars (63.8 $/kW) and
the total operating cost is 2.8 mills/kWh. This investment includes all
equipment, engineering, management, construction, startup, and shakedown
costs and is based on a starting date of January, 1974. The costs include
no by-product sulfur credit and no reheating cost (since reheating is
considered unnecessary for most applications). The costs do include the
costs of the required Glaus plant. A breakdown of the utility costs is
shown in Table B-22.
-------
B-106
TABLE B-22. BREAKDOWN OF UTILITY COSTS
FOR 330-MW AGP
Utility
Electricity
Natural Gas (or oil)
Coke
Cooling Water
Process Water
Boiler Feed Water
Steam
Makeup Na,CO-
• £* J
Total Utility Cost
Unit Cost
10 mills /kWh
$0.40/Mcf
$20/ton
$0.03/Mgal
$0.25/Mgal
$0.40/Mgal
$0.40/Mlb
$50/ton
mills/kWh
0.230
0.039
0.400
0.0087
0.0136
0.0022
0.0114
0.0076
0.7125
-------
APPENDIX C
ESTIMATED COSTS OF CENTRAL REGENERATION
AND ACID PRODUCTION PLANT
-------
C-2
TABLE 01. CENTRAL REGENERATION AND ACID PLANT (1000 tons/day, 330 days/yr)
Item Cost (mid-1973), $103
Capital Requirement
Bare cost 8,612
Engineering and design
Contractor's overhead and profit —
Subtotal Plant Investment 8,612
Project contingency
Total Plant Investment
Interest during construction
Startup cost
Working capital 1>221
Capital Requirement 11,442
Annual Operating Cost
Labor(b)
Administrative and general overhead
Materials and utilities^.
Transportation of solids^ '
Additional fuel requirement
Local taxes and insurance
Gross Operating Cost
Credit
Net Operating Cost -494
Annualized Cost
Return on rate base 665
Federal income tax 219
Depreciation 511
Net operating cost -494
Average Annual Cost 901
Annualized Control Cost
$/lb,S removed (£) g 53
-------
C-3
Footnotes to Table C-l.
(a) $8.2 x 106 for a 1000 ton/day plant in 1972 (8,200) (-—y^) = 8,612.
This cost is the total plant investment.
(b) Direct operation: $123,000
Maintenance: 172,000
Supervision: 44.000
$339,000
(c) Power: $ 165,000
Boiler feed water: 21,000
Process water: 7,000
Cooling water: 20,000
Fuel oil: 1,129,000
MgO: 462,000
Coke.: 33,000
Maintenance material: 172,000 (2 percent of TPI)
Operating supplies: 102.000
$2,111,000
idled = 3. 3 x I
E transportatic
$3,220,000/yr.
(d) MgS03 to be handled - 3.3 x 105 ton/yr; MgO to be handled - 1.3 x 105
ton/yr; cost of transportation (assumed) = $7/ton; cost of transportation
(e) Sulfur removed - (330,000)(0.98) (||) - 1.056 x 105 tons/yr.
70
-------
C-4
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO.
EPA-600/2-75-073
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
SO2 Reduction in Non-utility Combustion Sources--
Technical and Economic Comparison of Alternatives
5. REPORT DATE
October 1975
6. PERFORMING ORGANIZATION CODE
s.K. Choi, E.L. Kropp, W.E. Ballantyne,
M.Y. Anastas, A.A. Putnam, D.W. Hissong, and
T.J. Thomas
8. PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Batte lie-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ACX-083
11. CONTRACT/GRANT NO.
Contract 68-02-1323, Task.13
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final: 5/74-9/75
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT.
The report gives results of an analysis of non-utility combustion (NUC)
sources for various size classes and fuel types with respect to the significance of
SO2 emissions. Technical and economic comparisons of various SO2 control
alternatives were made for the important size classes and fuel types. Categories
of alternatives included are: physical cleaning of coal, coal gasification, coal
liquefaction, fluidized-bed combustion of coal, and flue gas desulfurization. For
small size classes of NUC sources, applicabilities of package sorption systems
were reviewed.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution
Coal
Combustion
Sulfur Dioxide
Gasification
Liquefaction
Fluidized Bed
Processing
Desulfurization
Sorption
Air Pollution Control
Stationary Sources
Non-utility Sources
Physical Cleaning
Package Sorption
13 B
21D
21B
07B
13H, 07A
07D
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)'
Unclassified
21. NO. OF PAGES
316
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |