xk"'
vi-EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711 ' T\ VN
Research and Development
EPA-600/S2-82-070C Nov. 1982
Project Summary
Acid Rain Mitigation Study:
Volume III. Industrial
Boilers and Processes
J. G. Ball, C. A. Muela, and J. L. Meling
The U.S. EPA has initiated a multi-
phased study of the acid rain problem.
As part of Phase I, Radian Corporation
investigated SOa emissions and con-
trols in the industrial sector.
The scope of this 4-month study
was limited to existing industrial
sources of SO2 emissions in the Acid
Rain Mitigation Study (ARMS) region.
This region includes all of the states
east of the Mississippi River, as well as
Minnesota, Iowa, Missouri, Arkansas,
Louisiana, North Dakota, South Dakota,
Nebraska, Kansas, Oklahoma, and
Texas. The objectives of the study
were to (1) identify and characterize
existing industrial sources of SO2
emissions, (2) identify the control
techniques that can be used to reduce
SO2 emissions from these sources,
and (3) estimate the SO2 emission
reduction potential and the associated
costs in constant 1980 dollars based
on application of these controls.
Because of severe time limitations,
only a portion of the SOa sources were
investigated in detail. Simplifying
assumptions were made about the
balance of the SO2 sources studied.
Time contraints also prevented an
evaluation of the availability of low
sulfur control options (i.e., physically
cleaned coal and low sulfur fuel oil).
In addition, since site visits were not
made, the remaining useful lives of the
sources were not determined and
"average" flue gas desulfurization
unit retrofit factors were estimated.
Each of these considerations signifi-
cantly affects both the potential SO2
emissions reduction and the associated
costs.
The results of the investigations
conducted to meet each study objec-
tive are presented in the report.
Recommendations concerning the
use of these results are also discussed.
This Project Summary was devel-
oped by EPA's Industrial Environ-
mental Research Laboratory. Research
Triangle Park, NC, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
There is a growing concern about the
acidity of precipitation in the north-
eastern United States and Canada.
Many scientists think that acidic precip-
itation kills aquatic and plant life,
damages crop-growing soil, and accel-
erates erosion and damage to buildings.
Although the mechanisms producing
acid rain are not clearly understood,
sulfur dioxide (SOa) and oxides of
nitrogen (NOX) are thought to be the
precursors of the chemicals that cause
acid rain. Large quantities of SOa and
NOX are produced by various combustion
and non-combustion processes in both
the utility and industrial sectors.
Reducing these SOa and IMOx emissions
to the atmosphere should reduce the
potential for acid rain.
Because of this growing concern, the
U.S. EPA initiated a multi-phased study
-------
of the acid rain problem. As part of this
study, Radian Corporation investigated
SO2 emissions and controls in the
industrial sector; Teknekron, Inc. made
a similar study of the utility sector. The
study was to be completed in 4 months
to provide direction for additional
phases. These later phases were
planned to investigate 862 emissions in
more detail than Phase I, to investigate
NOx sources, and to model source/
receptor relationships.
The scope of this study was limited to
existing industrial sources of SO2
emissions in the Acid Rain Mitigation
Study (ARMS) region. This region
includes all of the states east of the
Mississippi River, as well as Minnesota,
Iowa, Missouri, Arkansas, Louisiana,
North Dakota, South Dakota, Nebraska,
Kansas, Oklahoma, and Texas. The
objectives of the study were to (1)
identify and characterize existing indus-
trial sources of SOa emissions, (2)
identify the control techniques that can
be used to reduce SOa emissions from
these sources, and (3) estimate the SOa
emission reduction potential and the
associated costs* based on application
of these controls. Because of severe
time limitations, only a portion of the
SOa sources were investigated in detail.
Simplifying assumptions were made
about the balance of the SC>2 sources
studied.
It should also be noted that, due to
time constraints, the availability of low
sulfur control options (i.e., physically
cleaned coal and low sulfur fuel oil) was
not evaluated. In addition since site
visits were not made, the remaining
useful lives of the sources were not
determined and "average" flue gas
desulfurization unit retrofit factors were
estimated. Each of these considerations
significantly affects both the potential
SOa emissions reduction and the
associated costs.
Summary
The results of the investigations
conducted to meet each study objective
are summarized below. Theassumptions
made and recommendations concerning
the use of these results are also
discussed.
SO2 Emissions
Characterization
The Multistate Atmospheric Power
Production Pollution Study (MAP3S)
was used as a basis for a survey of
industrial plants emitting S02 within
the Acid Rain Mitigation Study (ARMS)
region. This survey identified 257 major
industrial emitters (an emitter may
contain more than one point source) of
S02 ranging in the magnitude of S02
emissions from 5,000 to 200,000 tons
per year. These data show that utility
and industrial plants emitting more than
5,000 tons per year of SC>2 account for
90 percent of the total SO2 emissions.
The total quantity of S02 emitted each
year by these 257 industrial plants is 3.6
million tons. Because of time limitations
in this initial phase, plants with 5,000
tons per year of SO2 emissions were
the smallest emitters evaluated in this
study.
A total of 39 industrial plants, each
emitting more than 20,000 tons of SO2
per year, together account fo about 1.6
million tons of S02. Only these 39 plants
were studied in detail. For these 39
plants, industrial boilers are responsible
for 695,300 tons of SO2 per year. Non-
boiler processes in the metals industry
emit about 543,200 tons of SO2 pe
year; and in other industries, including
chemicals, petroleum refining, anc
cement, non-boiler sources contribute
some 350,000 tones of S02 per year
These data are summarized in Table 1
Due to time and feasibility constraints
S02 emissions from plants emittinc
between 5,000 and 20,000 tons pei
year were not characterized individually
They were evaluated by making £
number of assumptions (1) the approx-
imately 2 million tons of S02 per yeai
emitted by the 218 plants emittinc
between 5,000 and 20,000 per yeai
was assumed to result exclusively f rorr
coal-, oil-, and gas-fired boilers; (2) the
distribution of these boilers was assumec
to be similar to the nationwide distribu-
tion of industrial watertube boilers; anc
(4) the percent sulfur in coal, oil, and gas
was assumed to be the same as thef uel-
weighted average sulfur content of the
boilers in plants emitting more than
20,000 tons per year of S02. The results
of these calculations are presented in
Table 2.
Table 1. Industrial SOz Emissions from 39 Largest SOz Emitters
Emissions*
(103 tons SO2/year)
Process
Boilers
Coat-fired f> 100 MWJ
Coal- fired «100 MWj
Oil-fired
Gas-fired
Per Process
f 39 2.4)
( 86.9)
(172.6)
( 43.4)
Total
6.95.3
Percent of Total
43.1
Metals Industry
(excluding boilers)
Copper smelting (235.9)
Iron and steel (115.3)
Coke metallurgy ( 99.2)
Lead smelting ( 85.4)
Aluminum ( 3.1)
Other f 4.3)
Other Industrial Processes
(excluding boilers)
Cat crackers ( 71.3)
Process heaters ( 60.3)
Cement kilns f 54.2)
Sulfuric acid ( 49.7)
Phosphate fertilizer ( 41.8)
Sulfur plants ( 33.9)
Alfalfa dehydration (21.6)
Flares ( 8.9)
Sulfite pulping ( 7.0)
Sludge conversion ( 1.3)
Unclassified
543.2
33.7
350.0
21.7
25.5
1.5
'All of the cost presented in this study are
expressed in constant 1980 dollars.
1614.0
1OO.O
*Based on 1977 survey.
-------
Table 2. Size Distribution of Boilers Needed to Account for 2 Million* Tons of SO 2 Emitted per Year from Plants Emitting Between
5,000 and 20,000 Tons/Year of S02
A verage
Boiler
Size (MWJ
45
96
220
700 (coal, oil!
1265 fgasj
Total
103 tons
SO2/yr
686
371
172
149
1378
Coal
JO3 tons SOi/
yr/Boiler
16
34
7.8
24.9
Approximate
Number of
Boilers"
429
109
22
6
566
103 tons
SOi/yr
290
135
56
15
496
Oil
103 tons SO2/
yr/Boiler
1.1
23
54
17 1
Approximate
Number of
Boilers'1
264
59
10
1
334
W3 tons
SOi/yr
51
22
17
36
126
Gas
!03 tons SO,/
yr/Boiler
0.6
1 3
29
16.9
Total
Approximate
Number of
Boilers"
85
17
6
2
110
Approximate
Number of
Boilers"
778
185
38
7
2
1010
103 tons
SCVyr
1027
528
245
164
36
2000
a The actual total of 1996 x 10 tons per year was rounded to 2000 x 10 tons per year
" The number of boilers in each size category that would yield the annual SOz emissions derived in Table 3-8 fof the full report), given the fuel quality and average boiler
size and capacity factor discussed in the text.
Applicable SOz Emission
Control Technologies
Both pre- and post-combustion clean-
ing processes have been used to reduce
SO2 emissions from industrial combus-
tors. The pre-combustion processes
most commonly used for coal- and oil-
fired combustors are physical coal
cleaning and oil desulfurization, respec-
tively. Another method of reducing S02
emissions from industrial boilers is to
substitute naturally occurring low
sulfur coal and oil for high sulfur fuels.
The most commonly used post-combus-
tion process is flue gass desulfurization
(FGD). In the metals and process
industries, S02 emissions from non-
combustion sources have also been
controlled by FGD processes. In addition,
sulfuric acid plants are frequently used
to recover the SOz if the flue gas is
sufficiently concentrated to warrant
economic recovery.
Estimated Costs of Applying
SO2 Control Techniques to
Industrial Sources and the
Potential SOz Emissions
Reductions
Preliminary costs* for reducing S02
emissions from different industrial
sources with the control alternatives
identified above were developed. Be-
cause each control alternative cannot
be effectively applied to all of the
industrial sources, the control alterna-
-*The capital investment and annual costs are
significant only to two figures, however, results
are reported to the last whole significant figure to
be consistent with other similar studies In this
study, annual costs are first year costs and
include O&M, overhead, utilities, etc., and
capital-related charges equivalent to 0.13 times
the total capital investment All of the costs
presented in this report are expressed in
constant 1980 dollars
tives were applied only to those source
categories that seem to be logical based
on actual installations. Only one scenario
was evaluated.
The use of physically cleaned coal to
reduce S02 emissions was applied to
boilers having capacities less than 75
MWt (boiler heat input in thermal
megawatts*), and firing coals with 2.0
percent or greater sulfur content. This
control method was adopted for small
coal-fired boilers because an FGD
process would require very high capital
investment/annual costs. The use of
physically cleaned coal would reduce
uncontrolled SOa emissions from these
boilers by 219 x103 tons per year, which
represents about a 30 percent reduction.
It was assumed that physically cleaned
coal would be purchased from a major
coal producer. As a result, each boiler
operator would not incur the costs of
constructing and operating a physical
coal cleaning facility; instead, the
operator would pay the producer a
premium for cleaning the coal. The
cost of physically cleaned coal was
estimated to be $4.28 per ton (1980
dollars) for a total annual premium of
approximately $46 x106 for the applic-
able industrial sources. This cost
represents the cost of preparation and
does not include the costs of raw coal,
taxes, and transportation. The availability
of physically cleaned coal to the SO2
sources was not examined.
Low sulfur fuel oil was assumed to
be used for reducing emissions from all
oil-fired boilers. Uncontrolled S02
emissions from these boilers are
approximately 669 x 103 tons per year.
The use of low sulfur fuel oil (0.8
percent S) reduced the uncontrolled
emissions by 51 percent assuming an
average uncontrolled fuel oil sulfur
content of 1.64 percent. It was assumed
that low sulfur oil would be purchased
from a major producer. The resulting
incremental cost for the low sulfur oil
was taken as $2 50/bbl (based on
hydrode sulfurization costs) for a total
annual premium of $314 x 106 This
represents the incremental cost of the
lower sulfur fuel compared to the fuel
currently being used, the availability of
low sulfur oil to the S02 emission
source was not examined.
FGD systems were assumed to be
used for reducing emissions from all of
the larger coal-fired boilers (those
having capacities greater than 75 MWt).
FGD would reduce S02 emissions from
these boilers by approximately 940 x 103
tons per year. The capital investment cost
for installing these FGD systems would
be about $2,020 x 106* The annual cost
associated with these FGD systems
would be approximately $565 x 106. The
capital investment and annual costs for
industrial FGD systems were derived by
integrating cost studies for small
industrial and large utility systems. This
integration required putting the cost
studies on the same equipment basis,
using the same investment and cost
algorithms, and using the same unit
pricing of labor and raw materials.
FGD systems were also assumed to
be used for most non-boiler sources in
the metals and process industries for all
processes emitting greater than 1000
tons per year of S02 from a single point
source. Retrofitting an FGD process to
the uncontrolled sources in the metals
industry decreased SO2 emissions from
543 x 103 to 62 x 103 tons per year. A
*A boiler with a heat input capacity of 200 x 106
Btu/hr would have a capacity of 58 6 MWt
"The capital investment for an FGD system retrofit
was assumed to be 1.3 times the capital
investment of an FGD system applied to a new
boiler
-------
capital investment of $420 x 106andan
annual cost of about $128 x 106 would
be required to achieve this reduction. In
the process industry, the use of FGD
systems on process heaters, sludge
concentrators, sulfuric acid plants,
catalytic crackers, and elemental sulfur
plants reduced SOz emissions from 350
x 103 to 64 x 103 tons per year, an 82
percent reduction. The capital and
annual costs required to achieve this
reduction would be $580 x 106 and
$178 x 106, respectively. Table 3
summarizes the assumed S02 control
applications. Figures 1 and 2 show the
costs used for FGD applications.
Overall, industrial-sector plants
emitting greater than 5,000 tons per
year of S02 (257 plants) generate
approximately 3.6 x 106 tons of SO2 per
year. Implementation of the control
options described above would reduce
the uncontrolled SO2 emissions by
approximately 63 percent. To achieve
this reduction, industry within the
ARMS region would have to invest
about $3.0 billion for installing the
control processes, as well as incur
annual costs of about $1.2 billion (see
Table 4). Plants emitting more than
20,000 tons per year of S02 (39 plants)
account for 1.6 x 106 tons per year of
uncontrolled S02 emissions. Marginal
cost and investment analysis shows
that investment and cost per ton of SO2
removed is lower for these plants than
for plants emitting 5,000 to 20,000 tons
per year of SO2 (see Tables 4 and 5). The
controls applied to the 39 plants could
reduce uncontrolled SO2 emissions by
75 percent and would require a $1.5
billion capital investment and about
$0.6 billion in annual costs.
The cost effectiveness (in $ ton/SO2
removed) for the various control options
are:
Table 3. Summary of SO 2 Control Applications
Control
Applied
None0
Physically
Cleaned Coal
FGD (Limestone)
Oil Desulfuri-
Source
Category
Boilers
Boilers
Boilers
Boilers
A verage Sulfur Boiler Number of
Type of Level in Size Affected
Fuel Used Fuel (%f (MWJ Boilers
Coal
Coal
Coal
Oil
1.3 All Sizes
3.0 <75
2.4 >75
1.64 All Sizes
31C
449C
193
441
zation (0.8% S Oil)
None
FGDa
FGD
Boilers
Non-boiler
sources in
metals
industry
Non-boiler
sources in
process
industry
Gas
—
—
0.8 All Sizes
— 25-644e
— 17-1000'
160
—
—
^Before controls are applied.
* Boilers in the top 39 S02-emitting plants currently firing 2.0% or lower sulfur coal do
not require controls.
cNine boilers were unclassified as to size and nine boilers were unclassified as to fuel
type. Fourteen of these boilers were found to be at sites that used low S coal
exclusively. Thus the number of low S coal burning boilers was increased from 17 to
31. The other four unclassified size/fuel boilers were arbitrarily assigned to
physically cleaned coal.
6Sulfuric acid plants were used as controls on two smelting operations.
e The capacities of non-boiler sources in the metal industry are equivalent to boilers of
this size range.
' The capacities of non-boiler sources in the process industry are equivalent to boilers
of this size range.
The average cost effectiveness for the
controls evaluated range from about
$210/ton of SO2 for physical coal
cleaning to $915/ton for oil hydro-
desulfurization.
Recommended Use of Study
Results
Preliminary estimates of the SO2
reduction potential from existing indus-
trial sources and the associated control
$/ton S02
Removed for Plants
Emitting More Than:
Application
Boilers
Boilers
Boilers
Nonboiler Sources/
Metals Industry
Nonboiler Sources/
Process Industry
Weighted Average
Boiler Size
MW,
<75
>75
All
Fuel/
Wt. % Sulfur
Coal/3
Coal/2.4
Fuel Qil/1.6
S02 Control
Technology
Physical
Coal Cleaning
FGD
Low Sulfur Oil*
FGD
FGD
5000 tons/
yr SO2
210
601
915
266
622
542
20,000 tons/
yr SO2
210
536
910
266
622
472
costs are provided in this study and will
be useful for planning future work in the
acid rain area. Due to time and budget
constraints, a number of key assump-
tions were made i n order to obtain these
results. Many of these assumptions
were highlighted in this summary.
If control of existing industrial sources
of SO2 is considered to be a reasonable
strategy for attacking the acid rain
problem, then a more detailed evaluation
of the industrial sector should be
undertaken. The scope of work should
include:
1. A more detailed characterization
of the targeted point sources. Sites
should be visited.
2. An evaluation of the availability
and costs of low sulfur fuel oil, low
sulfur coal, and physically cleaned
coal to the various plants.
3. A characterization of source ages
and remaining useful lives.
4. Engineering evaluations to assess
the feasibility and cost of FGD
retrofits.
5. An evaluation of other scenarios
for controlling S02 emissions.
f Costs are based on hydrodesulfurization.
-------
*>
o
%25-
•+*,
-c
^
oa 20-
S>
*-.
!f 11-
i
«)
s
^ 10-
.*-
S-
0
m
*
CO
1
"5
c
c
^
0^
25-
20-
15-
Shop-Fabricated/Packaged
FGD Units
Large shift at 60 MW,
due to differences in
capital charges caused by
shop-fabricated/packaged
units vs. field-fabricated/
erected units
Basis:
3.5 % S Coal - 90% removal
Costs on this figure are
for an FGD system retro-
fitted to an existing boiler.
* Annual costs are expressed
two ways:
• $/106Btu is the annual
costs ($/yearj divided by the
annual boiler heat input
(Bttt/year).
• $/MWt is the annual costs
($/year) divided by the boiler
capacity (MW,).
**7 MWt = 3.412 x 106 Btu/hr.
Field-Fabricated/
Erected FGD Units
0
0
' 300
1000
600 ' 900 ' 1200
Boiler Heat Input Capacity, MW,**
2000 3000 4000
Boiler Heat Input Capacity, 10s Btu/hr
' 1500
5000
Figure 2. Annual costs for a retrofitted limestone FGD system.
-------
Table 4.
Summary of Impact and Cost of Controls for Plants Emitting More Than 5,000 Tons/Year of SOz
S02 Emission Sources" SC>2 Control Processes
Boiler Avg. Sulfur Uncontrolled
Source Fuel Size Level SOz Emissions
Category Type (MWJ 1%) (1O3 ton/yr! Control Applied
SO2 Reduction Controlled Capital Cost
Achieved SOz Emissions lnvestmentg Annual Cost Effectiveness'
1%) (JO3 ton/yr} f$10e} ($10e}°'c ($/ton SO2 Removed!
Boilers
Non-boiler
Sources
/Metals
Industry}
Non-boiler
Sources
(Process
Industry/
Totals
Coal — " 1.3
Coal <75 3.0
Coal >75 2.41
Fuel
Oil All 1 64
Gas All 0 8
— —
— —
83
730
1044
669
169
543
350
3,588
None
Physical Coal
FGD (Limestone}
Low sulfur fuel oil
None
FGD (Limestone}9'"
FGD (Limestonef
—
30
90
51'
—
89
82
63'
83
511
104
326
169
62
64
1,319
—
- 46°
2,020 565
- 314"
— —
420 128
580 178
3,020 1,231
—
210
601
915
—
266
622
542
a Uncontrolled emissions for 1977.
"in this report annual costs include annual O&M. overhead, utilities, etc., and annual capital-related charges equivalent to 0 13 times the total capital investment
c 1980 S.
" Boilers in the top 30 S02 emitter that actually burn <2.0 percent sulfur coat
* The cost directly represents the premium paid by boiler operator for upgraded fuel.
1 This reduction efficiency is set by assuming the use of 0.8 percent sulfur fuel oil down from an average level of 1 64 percent. The costs are based on resid
hydrodesulfurization.
9 Includes two sulfuric acid plants.
"FGD process applied to point sources emitting > 1000 tons per year SOs.
' Weighted average.
TableB.
Summary of Impact and Costs of Controls for Plants Emitting More Than 20,000 Tons/Year of S02
S02 Emissions Sources' SOi Control Processes
Boiler Avg. Sulfur Uncontrolled
Source Fuel Size Level SO2 Emissions Control
Category Type (MWJ 1%) (103 ton/yr} Applied
S02 Reduction Controlled Capital Cost
Achieved SO2 Emissions Investment" Annual Cost Effectiveness"
/%} (1O3 ton/yr} f$10e} ($106f-ti (S/ton SO2 Removed}
Boilers
Non-boiler
Sources
(Metals
Industry/
Non-boiler
Sources
(Process
Industry}
Total
Coal —" 1.3
Coal <75 3.0
Coal >75 2.41
Fuel All 1.64
Oil
Gas 08
— —
— —
83
44
352
173
43
543
350
1,688
None
Physical Coal
Cleaning
FGD (Limestone}
Low sulfur oil
None
FGD (Limestone}3''1
FGD (Limestonef
30
90
51'
—
89
82
75'
83
31
35
84
43
62
64
402
- 3"
477 770
— 81"
— —
420 128
580 178
1,471 560
—
270
536
970
—
266
622
472'
" Uncontrolled emissions for 1977.
" 1980 S.
"In this report annual costs include annual O&M, overhead, utilities, etc., and annual capital-related charges equivalent to 0 13 times the total capital investment
" Boilers in the top 39 S02 emitters that actually burn <2 0 percent sulfur coal.
"The cost directly represents the premium paid by boiler operator for upgraded fuel.
'This reduction efficiency is set by assuming the use of 0.8 percent sulfur fuel oil down from an average level of 1.64 percent. The costs are based on resid
hydrodesulfurization
9 Includes two sulfuric acid plants
"FGD process applied to point sources emitting >1000 tons per year S02
1 Weighted average.
U. S. GOVERNMENT PRINTING OFFICE: 1982/659-095/0547
-------
-------
J. G. Ball, C. A. Muela, andJ. L. Me/ing are with Radian Corporation, Austin, 7X
78759.
William Baasel is the EPA Project. Officer (see below).
The complete report, entitled "Acid Rain Mitigation Study: Volume III. Industrial
Boilers and Processes," (Order No. PB 83-101 337; Cost: $11.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
------- |