United States
                   Environmental Protection
                   Agency	
Air and Energy
Engineering Laboratory
Research Triangle Park NC 27711
                   Research and Development
 EPA/600/S7-88/014  Jan. 1989
&EFA          Project  Summary
                   Ohio/Kentucky/TVA Coal-Fired
                   Utility SO2  and  NOX Retrofit
                   Study

                   T. E. Emmel, S. D. Piccot, and B. A. Laseke
                     This document summarizes initial
                   results from an ongoing National
                   Acid Precipitation Assessment Pro-
                   gram (NAPAP) study, the objective of
                   which Is to significantly improve
                   engineering cost estimates  for
                   retrofit  of  the  following  control
                   technologies at  the 1980 "top 200"
                   SOj-emlttlng   coal-fired  power
                   plants  in  the  31 eastern  states:
                   lime/limestone  FGD,  lime  spray
                   drying  FGD, coal  switching and
                   cleaning, furnace sorbent injection
                   with humldification (LIMB), duct
                   sorbent  Injection, low NOX  burners,
                   overfire air, natural  gas  reburn, and
                   selective catalytic reduction. Retrofit
                   cost factors   and costs  were
                   developed  for 12 coal-fired power
                   plants: 5 In Ohio, and 7  in Kentucky
                   and the TVA system (Tennessee,
                   parts of Alabama,  and Kentucky).
                   Activities included:  selecting plants
                   with boilers representative of the top
                   200 population;  conducting plant
                   visits and  collecting  site  specific
                   data;  developing  boiler/control-
                   specific retrofit difficulty factors; and
                   developing  boiler/plant-specific
                   cost and performance estimates.
                   Results from this effort are being
                   used to develop simplified proce-
                   dures to estimate the retrofit costs
                   for a number of the remaining  top
                   200 plants which  are not visited.
                     This Project Summary was devel-
                   oped by EPA's Air  and  Energy
                   Engineering Research  Laboratory,
                   Research Triangle Park,  NC, to  an-
                   nounce key findings of the research
                   project that is fully documented in a
                   separate report  of  the same title
(see Project Report ordering  infor-
mation at back).

Introduction
The National Acid Precipitation Assess-
ment Program (NAPAP) is responsible
for developing cost and performance in-
formation  on various methods  for
reducing the emissions  of acid rain
precursors. Coal-fired utility  boilers  are
major emitte'rs of  S02  and  NOX. How-
ever, estimating  the  cost and per-
formance of SC>2 and NOX controls for
coal-fired power plants is difficult due to
differences in  plant  layout  and  boiler
design.
  This report documents the initial results
of an ongoing study conducted  under
NAPAP, the objective  of which is  to
significantly improve the accuracy  of
engineering cost estimates used  to
evaluate the economic effects of applying
SOg and NOX controls to existing coal-
fired utility boilers.
  This  report  presents the SOg/NOx
control technology cost and performance
estimates  developed for  12 coat-fired
utility plants in Ohio,  Kentucky, and  the
Tennessee Valley  Authority  (TVA)
system. The following procedures were
used to develop  the cost performance
estimates:  select  plants with boilers
representative of the population, conduct
plant visits  and collect site-specific data,
develop boiler/control specific  retrofit
difficulty factors,  and develop  boiler/
plant-specific cost and  performance
estimates for the SOg and NOX controls
selected for evaluation. This performance
and cost estimating process is dynamic:
it incorporates recommendations  from a
technical advisory group and partici-

-------
pating  utilities,  as well as  experience
from site  visits and  performance/cost
evaluations.

Control Technologies
Evaluated
  The  commercial and developmental
SC>2 and  NOX control technologies
selected for inclusion in the program are
listed in Table  1. Evaluated  qualitatively
without cost  estimates were FBC  and
CG. Additional technologies were at first
considered - chemical coal cleaning,
pressurized  FBC,   and  advanced
S02/NOX (combined)  control devices -
were not included in  this  study due to
their general  inapplicability to retrofit
situations and early development status.
  All of the  estimates were developed
using  computer-based  simulation
models.  This  approach  was  taken
because  of the large  quantity  of data
which needed to be accessed, used  in
computation, and stored.  Although  a
number of mainframe- and personal-
computer-based  models  were evalu-
ated, the Integrated Air Pollution Control
Systems (IAPCS) cost model was  finally
selected for use in this study because of
its versatility.  The IAPCS  model has
been upgraded to include  the  tech-
nologies being evaluated in this program.

Plants Visited
  Four criteria  were used to select  the
plants to be  visited: (1)  plant selection
focused  on  boiler types  and  sizes
accounting  for the majority  of 802
emissions;  (2)  within  the limits of cri-
terion,  the  selected plants were repre-
sentative examples of the diversity of the
boiler population; (3)  due to costs,  the
number of plants  selected for detailed
study were limited to less than 30; and
(4) the plants selected contained multiple
boilers of diverse types.
  Boilers in the top 200 S02-emitting
plants were categorized  by generating
capacity, coal percent  sulfur, firing type,
age, and  capacity  factor. The top 200
plants  were ranked according to their
diversity,  and  the 30 highest-scoring
plants were evaluated  by boiler category
to see if a realistic, proportional sample
had  been achieved successfully. Plants
near the bottom of the  list (with dispro-
                             portional, extreme ratings) were repla<
                             to ensure the representa-tiveness of
                             sample. From  this list of 30 plants,
                             were chosen for  evaluation:  5 in  Ol
                             and 7 in  Kentucky and the TVA syst
                             (Tennessee,  parts of Ala-bama,  &
                             Kentucky).
                               The  Ohio and  Kentucky plants  wi
                             selected  for evaluation first because
                             the  opportunity to conduct the progr
                             jointly with Ohio and Kentucky State A
                             Rain (STAR) programs. In addition to c
                             TVA plant in Kentucky, three TVA pla
                             outside  of  Kentucky were  includ
                             through  TVA's  participation  in  t
                             Kentucky  STAR program. These  w<
                             considered to  be  representative of  1
                             top 200 SOg emitting plants. Table 2 li
                             the  plant/boiler characteristics for  the
                             12 plants.
                               Prior to the  plant site visit,  a  pi;
                             profile  was completed  using  sources
                             public  information; a  primary  referen
                             with the Department of Energy's (DOE
                             Energy Information Agency (EIA)  Fo
                             767. The  plant  profile included
                             information  and  data  needed  1
                             performance/cost  analyses.  The  pl<
         Table 1.  Emission Control Technologies Selected
                                                                              Development Status
                                           Species Controlled
                  Control Technology
    SO?
/VCX
                                                                Commercial
  Limited
Commercial
Experience
    Near
 Commerical
Demonstration
          Lime/limestone (ULS) flue gas
          desulfurization (FGD)                   X

          Additive enhanced ULS FGD            X

          Ume spray drying (LSD) FGD*            X

          Physical coal cleaning (PCC)            X

          Coal switching and blending (CS/B)       X

          Low-NOx combustion (LNC)

          Furnace sorbent injection (FSI) with
          humidification (UMB)                   X

          Duct spray drying (DSD)                X

          Natural gas rebuming (NGRp            X

          Selective catalytic reduction (SCR)

          Fluidized bed combustion (FBC) or
          coal gasification (CG) retrofit?           X
                           X

                           X

                           X

                           X

                           X
              X

              X
                                                X

                                                X

                                                X
          ^Commerical on low-sulfur coals, demonstrated at pilot scale on high sulfur coals.
          hFor wet bottom boilers and other boilers where LNC is not applicable.
          cEvaluated qualitatively as combined life extension and SOX/NOX control option. No costs were developed.

-------
Table 2. Boiler Characteristics of Plants
Boiler
Plant (State) No.
Burger (OH)
Conesville (OH)
Muskingum River (OH)
Sammis (OH)
J. M. Stuart (OH)
E. W. Brown (KY)
Elmer Smith (KY)
Big Sandy (KY)
Paradise (KY)
Johnsonville (TN)
Cumberland (TN)
Colbert (AL)
1-4
5-6
7-8
1-2
3
4
5-6
1-2
3-4
5
1-4
5
6-7
1-4
1
2
3
1
2
1
2
1-2
3
1-4
5-6
7-10
1-2
1-4
5
Evaluated Using Detailed Procedures
Net Dependable
Generating Capacity
Capacity Factor,
Per Unit, MW Age, yr Percent
30
45
75
125
165
720
375
205
215
585
180
300
600
559
100
156
410
151
265
260
800
704
1150
125
147
173
1300
200
550
40-43
37
32
28-30
25
14
9-11
33-34
29-30
19
25-28
20
16-18
13-17
30
24
16
23
13
24
18
24
17
35-36
34-35
28-29
14
32
22
29-30
31
38-56
32-26
37
58
45-48
63
55
61
44-57
53
44-51
54-66
50
65
65
30
50
63
63
30
30
39
30
41
60
50
50
Coal
Sulfur,
Percent
3.4
3.4
3.4
2.9
2.9
2.9
4.1
4.4
4.4
4.4
0.9
2.4
2.4
1.2
1.9
1.9
1.9
2.9
2.9
1.2
1.2
2.9
2.9
1.7
1.7
1.7
2.9
2.3
2.3
Firing Type
Roof
Roof
Wall
Cyclone
Wall
Tangential
Tangential
Wall (Wet Botom)
Cyclone
Wall
Wall
Wall
Wall
Wall (Cell Burner)
Wall
Tangential
Tangential
Cyclone
Tangential
Wall
Wall
Cyclone
Cyclone
Tangential
Tangential
Wall
Wall
Wall
Wall
profile data were verified and completed
using information obtained during the 1-
day site visit.

Summary of Performance and
Cost Estimates
  Using the  data  and  information
obtained from  the  plant  visits, site-
specific  cost estimates were developed
using the IAPCS cost model. These cost
estimates reflect site-specific  retrofit
costs because retrofit factors,  scope
(cost) adders,  and  performance esti-
mates were developed and input to the
model.  Figure 1 shows the methodology
used to  develop the retrofit costs using
the IAPCS cost model.
  For all technologies,  retrofit factors and
scope adders were developed using the
Electric  Power Research Institute (EPRI)
report,  "Retrofit  FGD Cost Estimating
Guidelines." Retrofit factors are  process
area multipliers which  adjust the  cost
model  to  reflect the following  location
and retrofit effects:

• Location - regional material and labor
  costs, foundation and  support structure
  costs related to soil conditions and
  seismic  zone,  and freeze protection
  costs.
• Retrofit  -  access/congestion,  under-
  ground  obstructions,  and distance
  between process areas.

Scope  adders are additional costs that
are included  in the cost  of retrofit but not
in the cost model algorithm bases. These
cost  adders  include: a  new chimney
liner, draft control  modifications, equip-
ment demolition  and replacement, and
particulate control modifications.
  For CS, fuel cost differentials were
developed from  the cost of currently
used coals using data from FERC Form
423.  Two  fuel price differentials  (FPDs)
were evaluated: the current low to high
sulfur coal FPD and the current FPD plus
$15  a ton.  The  $15  a ton fuel  cost
addition was assumed  to  span  the
potential fuel price premium that would
result if  extensive CS  occurred  due to
acid  rain  legislation.  For PCC,  the
incremental fuel cost was determined by
assuming  that  the  plant coal  had
properties similar to one of the six coals
contained in the IAPCS cost  model  PCC
module.
  Performance estimates were  devel-
oped  for the spray drying technologies
and  the low  NOX combustion tech-
nologies. LSD, FGD and DSD  SOX per-
formance estimates  were  developed
based on  flue gas temperature  and
particulate control  type:  ESP  or fabric
filter.  Low NOX burner and over-fire  air
performance estimates were developed
by evaluating  the  furnace heat  release
rates versus flue gas residence time and
coal properties.

-------
  The full report describes in  detail the
procedures used to develop the  retrofit
factors,  scope adders, and LNC control
NOX reduction estimates. However,  a
brief  summary of  the  cost results
arranged by control technology follows.
The costs presented are levelized  annual
costs based on August 1987 dollars. The
capital  recovery and operating cost
levelization factors are  based on the
1986  EPRI   Technical  Assessment
Guidelines  report and are: 0.18  capital
recovery factor and 1.45 operating  cost
levelization  factor.

FGD Cost Estimates
  Figure 2  and Table  3  summarize the
plant level  cost estimates developed for
                                      conventional  L/LS FGD. Figure 2 shows
                                      the annual cost versus SOX reduction for
                                      the  12  plants  evaluated.  Table  3
                                      summarizes  the  plant  level  average
                                      annual unit cost of control and retrofit
                                      difficulty range for each boiler/plant.
                                        Note that L/LS FGO was applied only
                                      to units  6 and 7 at the Sammis  Plant.
                                      This was due  to the extreme retrofit
                                      difficulty for unit  6. The retrofit difficulty
                                      and  annual cost  of  control would be
                                      much greater for units 1-5 than for unit
                                      6, making it unlikely  that conventional
                                      FGD would be  applied to units 1-5. The
                                      plants  having the  lowest  unit  cost  of
                                      control  were  Cumberland  and  Muskin-
                                      gum River. The  Cumberland  units  are
                                      large (1300 MW each), fire a 2.9% high
                              Plant Visits and El A - 767 Data
                                    PM Control &
                                    Disposal Data
                                                      Boiler Design &
                                                      Operating Data
 Plot Plans &
Other Drawings
                                                 Demolition
                                                  Relocation
                                                                   Performance
                                                                    Estimates
Process/PM Control
  Retrofit Factors
                        Cost Adder
                      Retrofit Factors
                                                                      sulfur coal, and  have moderate retro I
                                                                      difficulty.  The  Muskingum  River  Plai
                                                                      units have low unit cost of  control eve
                                                                      though  units  1-4  have  high  retrof
                                                                      difficulty factors;  these units  have  hig
                                                                      capacity factors (40  to 60%) and burn
                                                                      very high 4.4% sulfur coal.
                                                                        Figure 3 and Table 4 summarize  th
                                                                      plant level cost estimates developed fc
                                                                      lime spray drying. Two control  option
                                                                      were considered for the  retrofit of thi
                                                                      technology: (1) reuse of the  existing ES
                                                                      and (2) a new fabric filter. For units wher
                                                                      the SCA of the existing ESP was  sma
                                                                      (<43.3 m2/act. m3/sec) or the addition (
                                                                      new plate area was impractical (e.g., ro(
                                                                      mounted  ESPs),  reuse of  the existin
                                                                      ESP was not considered.  In such case:
                     Integrated Air Pollution Control System Cost Model
   Figure  1.    Methodology using IAPCS cost model.

-------
         Table 3. Summary of ULS FGD Cost and Performance Estimates
                                                                  Levelized Annual Costs
Plant
Burger
Sammis
Muskingum River
Conesville
J. M. Stuart
E. W. Brown
Elmer Smith
Big Sandy
Paradise
Cumberland
Johnsonville
Colbert
DUII&I
No.
1-6
7
a
6
7
1-2
3-4
5
1
2
3
4
1
2
3
4
1
2
3
1
2
1
2
3
1
2
1-10
1-4
5
r\ju neirum
Difficulty
1.30
1.26
1.29
3.01
1.86
1.70
1.66
1.54
1.54
1.54
1.51
1.55
1.43
1.45
1.44
1.44
1.70
1.60
1.41
1.66
1.26
1.50
1.54
1.35
1.46
1.40
1.48
1.54
1.38
$/ton S02
1,218
1,112
845
1,397
1,122
671
704
496
1,961
1,779
1,526
703
1,429
1,449
1,469
1,283
2,306
1,526
1,035
1,906
867
2,018
1,416
956
598
575
1,213
1,222
807
mills/kWh
32.2
29.4
22.3
24.2
19.5
22.8
23.9
26.9
40.2
36.5
31.3
14.4
13.1
13.3
13.5
11.8
30.2
20.0
13.5
42.5
19.4
16.4
11.5
18.0
12.6
12.1
16.7
22.2
14.7
0^/2 ncuwoi/u",
tonslyr
14,500
13,300
19,500
46,200
39,900
76,600
71,300
106,800
7,100
8,100
10,900
74,500
25,200
24,900
24,300
29,700
5,700
11,600
30,000
8,900
26,000
11,600
35,800
63,300
141,000
141,000
60,000
69,200
44,200
a  new  fabric filter was required  for
participate control after the spray drying
reactor. However, reuse of the ESP was
considered impractical for many  units.
Boilers where LSD with ESP reuse (LSD
+  ESP) and  LSD with new fabric filters
(LSD  + FF) were applied are identified
on Figure 3 and Table 4. Note  that the
cost of retrofitting  new fabric filters
results in  a high  retrofit  difficulty  factor
and  a  high cost of  control.  For  the
Sammis plant, only unit 7 was evaluated
because retrofit of LSD  FGD would be
very costly for units 1-6.

Coal Switching and  Cleaning
  Figure 4 and  Table  5  summarize the
plant level cost  estimates developed for
coal switching  (CS).  For these  tech-
nologies a number of plants  and units
were not considered  applicable for CS
for the following  reasons: the  units
already  burn a low sulfur coal; the plant
'eceives coal by conveyor from local
mines and the construction of truck, rail,
and barge  receiving  facilities would be
very costly;  and the  units  have wet
bottom boilers which can burn only coals
having  special ash fusion properties. As
Figure  4 shows, the unit cost of control
for CS is very dependent upon the fuel
cost differential. The impact of paniculate
control and coal  handling upgrades  are
generally small by comparison.
  Table 6  summarizes the  plant level
cost of physical coal cleaning (PCC). A
number of plants were  not evaluated for
PCC  because the  coal  already  is
extensively cleaned, and the  IAPCS coal
cleaning  costs are based on cleaning
run-of-mine coals.  As  Table 6  shows,
the  unit cost and  the  amount  of 862
reduction obtained by PCC are both low.

Sorbent Injection Cost  and
Performance Estimates
  Two  sorbent injection technologies in
active research and  development were
evaluated  in this study: furnace sorbent
injection (FSI)  with humidification (LIMB)
and duct  spray drying (DSD). Figure 5
and Table 7 summarize the  plant level
cost estimates developed  for  these
technologies.  Not  all  boilers  were
considered good  candidates for  these
technologies for the following reasons:

• LIMB and DSD with ESP reuse were
  not  considered  practical for  boilers
  having an ESP SCA of <43.3 m2/act.
  m3/sec,  and

• DSD with   ESP   reuse  was  not
  considered if the duct residence time
  from the injection point after the  air
  heater to the ESP inlet was  less  than 2
  sec (< 100 ft-30.5 m-of duct length).

For boilers  where  ESP reuse was  not
considered practical, DSD with new fabric
filter was evaluated. The costs presented
for  FSI assume 70%  SOx control  and
35% sorbent utilization.

-------
   200.

is
«
1
o
Levelizec
s>.
D




/
/


//
f<
r/3


V
'/'
' /



/
'



/




/


5


Key




Plant
3 Burger
4 Sammis
5 Muskingum River
6 Conesville
7 Stuart
8 Brown
               40       80       120      160     200
                       S02 Reduction, thousand tons/year
                                                240
                                                         280

(B
X
O
ed Cost, mil
§
• 1
Leve/iz
§




/
/*


13
\A


14
/




/




'



/


/


Key
'„



Plant
9 Elmer Smith
1 0 Big Sandy
1 1 Paradise
12 Cumberland
13 Johnsonville
14 Colbert
Figure 2.
   40       80       120      160     200
              S02 Reduction, thousand tons/year
L/LS FGD cost versus S02 reduction.
                                                            240
                                                                     280

-------
         Table 4. Summary of ULS FGD Cost and Performance Estimates
                                                                   Levelized Annual Costs
Plant
Burger


Sammis
Muskingum River


Conesville



J. M. Stuart



E. W. Brown


Bmer Smith

Big Sandy

Paradise
Cumberland

Johnsonville
Colbert

Diwer
No.
1-6
7
8
7
1-2
3-4
5
1
2
3
4
1
2
3
4
1
2
3
1
2
1
2
3
1
2
1-10
1-4
5
r\ju noiruiH
Difficulty
1.50*
1.19
1.19
-1.93
2.01*
1.96*
1.70
1.63
1.63
1.60
2.07*
1.70
1.72
1.71
1.70
2.00*
1.69
1.25
1.88*
1.38
1.88*
1.75*
1.71
1.66
1.32
1.82*
1.80*
1.30
$/ton SO2
1,463
954
704
1,012
612
650
501
1,649
1,479
1,260
751
1,634
1,678
1,709
1,422
2,016
1,194
814
1,923
846
1,953
1,507
947
624
555
1,350
1,171
784
mills/kWh
37.1
17.9
13.2
12.5
19.1
21.2
11.0
28.7
25.7
21.9
14.8
8.8
9.1
9.2
7.7
25.3
13.2
8.5
41.4
13.9
14.8
11.8
15.1
10.3
9.2
17.9
19.7
10.1
tons/yr
13,900
9,400
13,900
28,400
70,200
68,500
69,000
6,100
6,800
9,300
71.600
14,900
14,700
14,400
17,600
5,500
9,800
24,000
8,600
19,100
10,900
34,400
53,700
110,000
110,000
57,500
64,000
31,300
         *This retrofit difficulty includes fabric filter; all others assume reuse of ESP.
  As Figure 5 shows, the cost of FSI at
Big Sandy and Johnsonville is very high
as a result of the need to retrofit new
fabric filters.

Retrofit of Fluidized Bed
Combustion  and Coal
Gasification
  The retrofit potential of FBC or CG with
reuse of the existing steam turbine and
other  plant  facilities was qualitatively
assessed  for each  boiler  using the
following criteria: boiler size, boiler heat
rate, boiler capacity  factor, boiler  age,
particulate  control  performance,  and
S02/NOX emission levels. The following
boilers were found to qualify as potential
candidates, based on the boiler age and
size  criteria:  Burger boilers  1-8,
Sammis boilers  1-4, Muskingum  River
boilers 1-4, Conesville  boilers  1-3,
Smith  boiler 1, Big  Sandy boiler 1,
Johnsonville boilers 1-10, and  Colbert
boilers 1-4.  However,  other  criteria are
also important and are discussed in more
detail for each plant in the full report.

Low  NOX Combustion
  Figure 6 and Table 8  summarize the
plant   level  cost  and  performance
estimates  for  application of low  NOX
burners   on dry-bottom  wall-fired
boilers, over-fire air on tangential-fired
boilers, and  natural gas reburn on other
boilers (wet bottom and  roof fired).  As
Figure 6 and Table 8 show, the unit cost
of low  NOX  burners and  over-fire air is
low (<$400/ton).  However,  for  plants/
boilers where NCR is  applied,  the  unit
costs  are  much higher ($800 to $1200
per ton). This is due to the high cost of
natural gas relative to  coal  ($2  per
million Btu fuel price differential).
Selective Catalytic Reduction
Cost Estimates
  Figure  7 and Table 9 summarize the
plant level cost estimates for  application
of SCR.  Except for  Sammis units  1-6,
tail-end systems were assumed (reactor
downstream  of  particulate control or
scrubbers).  For Sammis  units  1-6,
space limitations require that  a hot-side,
high-dust system configuration be used
after the  economizer and before the air
heater.  Use  of  the tail-end  system
minimizes unit downtime, which reduces
the uncertainty of estimating the  cost of
replacement power, and  maximizes the
catalyst  life  (cost trade-off with  energy
penalty  associated  with  flue  gas
reheating).
  The cost estimates for SCR presented
in this report  are  based on  1  year of
catalyst  life. German and  Japanese ex-
perience  indicates that a  3-  to  5-year

-------
    750
  <8
  SL
  x
    100
  8
  o
  •0
  91
  5  50
Legend
• LSD + ESP
x LSD + FF

/
/

/'.
'" /

6
u


/

/
/"
5
/»




Plant
3 Burger
4 Sammis
5 Muskingum River
6 Conesville
7 Stuart
8 Brown
                 40
80       120
                                              160
                                                        200
                                                                   240
                         SOi Reduction, thousand tons/year
I
'lized Cost, million
3 i

-------
         Table 5. Summary of Coal Switching Unit Cost Estimates
                                                           Levelized Annual Cost
                 Plant
Muskingum River


Conesville



J. M. Stuart




E. W. Brown



Elmer Smith


Big Sandy


Paradise

Cumberland

Johnsonville0

Colbert
                                    No.
                                   1-2
                                   1-4
                                    5
                                                  Baseline FP&>
                            Baseline FPD + $i5/ton
$/ton SO2    mills/kWh   $!ton SO2
                                        mills/kWh
                                                   176
                                                   458
                                                   392

                                                   763
                                                   764
                                                   769
                                                   736
                                                   838
                                                   777

                                                   395
                                                   238

                                                 2,235
                                                 2,226
                                                   189
         182
         146
               5.4


               7.2
               6.1

               2.5
               2.5
               2.5
               2.4


               6.2
               5.7

               6.7
               4.0

               9.6
               9.5
                                                              3.0
               2.2
               1.8
  523


1,158
1,073

3,862
3,865
3,874
3,811


2,117
2,056

  981
  802

4,313
4,304
                                809
  693
  657
                                           16.0
                                           18 1
                                           16.8

                                           12.7
                                           12 7
                                           127
                                           12.5
                                          15.6
                                          15.1

                                          16.6
                                          136

                                          18.5
                                          18.4
                                      13.0
8.6
8.1
                                                          tons/yr
Burger


Sammis


1-6
7
8
5
6
7
571
480
459
584
585
591
12.9
10.8
103
7.1
7.1
72
1,044
945
913
1.467
1,471
1,484
234
21.2
20.5
178
17.8
18.0
12.900
11.300
16.600
16,800
32.300
27,900
                96,300


                 8.300
                56,900

                 9,000
                 9,000
                 8,700
                10,700


                 6,500
                16,900

                 6,700
                19,600

                 6,100
                18,100
                                                                                                     106,900
                                                                                            11,800
                                                                                            30,100
         *FPD  = fuel price differential.
         bCoal switching was not evaluated for wet bottom boilers.
         c Coal switching was not evaluated for boiler at this plant, since boilers have small roof-mounted ESPs. Adding additional
           plate area is not possible, and the use of 803 conditioning is likely to be marginally beneficient
  For CS and  PCC,  the major  retrofit
factors,  excluding  fuel price  differential,
were  participate control  upgrade  costs
and  boiler performance  impacts. The
latter was not assessed because detailed
coal  analyses were not  available. As  a
result,  CS was  not  evaluated for wet
bottom boilers,  since boiler performance
impacts are likely to be significant.
  For the sorbent injection technologies,
FSI and DSD, particulate control upgrade
costs  would  have  the greatest impact.
Additionally,  for DSD,  sufficient duct
residence time must be available  to
ensure good droplet drying.
  For LNC and NGR, boiler  type and
configuration  are  important.  Low  NOX
burners  were   applied  only  to dry-
bottom  wall-fired  boilers. Overfire air
was  applied  only to tangential-fired
units. NGR  was  applied to wet  bottom
boilers and  other miscellaneous  boiler
types. Boiler heat  release rates  and
residence  times  in  different furnace
zones would have significant effects on
NOX removal efficiency  for  LNC  and
NGR.
  SCR costs would  be greatly affected
by  access  and  congestion  near  the
economizer  area   for   high  dust
applications  and  the  chimney  area  for
tail-end  applications  and  by flue  gas
ducting  distances.  For  high  dust
systems,  boiler  downtime costs  and
catalyst life would be  significant cost/
performance factors.  For tail-end sys-
tems, the energy  penalty for preheat is
balanced by increased catalyst  life  and
reduced catalyst costs.
                                        As discussed earlier, the objective  of
                                      the program is to  improve significantly
                                      the cost/performance estimates used  to
                                      evaluate  impacts of potential  acid  rain
                                      legislation. The information presented  in
                                      this  document  is  very useful  in  that
                                      regard. However, cost comparisons have
                                      not been  made  for retrofit options  for
                                      specific   boilers  at each plant.  For
                                      example,  costs  have bee  projected  for
                                      L/LS FGD and  the other  SO2  control
                                      technologies (LSD  FGD, CS,  PCC,  FSI,
                                      and DSD), but  no comparison  has been
                                      made regarding the best option.
                                        From  the utility  company's  perspec-
                                      tive, a decision concerning which  retrofit
                                      control to  apply to a given  boiler is very
                                      complex  and  would be based  on the
                                      following criteria:

-------
 OS
 £
 X
 
-------
Table 6. Summary of Coal Cleaning Unit Costs Estimates
        Plant
Boiler
 No.
                                                   Levelized Annual Cost
                                                       Baseline FPDa
                                               $lton SO2
                 mills/kWh
               SOZ Reduction,
                    tons/yr
 Burger


 Sammis


 Muskingum Riverb

 Conesvillet>
 J. M. Stuart



 £. W. Brown


 Elmer Smith

 Paradise0

 Cumberland13
 Johnsonville0

 Colbert
 1-6
  7
  8

  5
  6
  7
  2
  3

  1
  2
 1-4
  5
 600
 397
 363

 381
 382
 3S6
7.546
1,549
1,556
1,502


 389
 375

 843
 527
 676
 595
5.7
3.7
3.4

2.5
2.5
2.6
                                         3.2
                                         3.2
                                         3.2
                                         3.1
3.2
3.1

5.9
3.7
3.7
3.3
 5,200
 4,700
 6,900

 9,200
17,700
15,300
                     5,700
                     5,600
                     5,500
                     6,700
 7,400
19,100

 2,800
 8,100
                                                                                           5,300
                                                                                          13,400
aFPD = fuel price differential.
b Physical coal cleaning (PCC) was not evaluated because coal presently is washed to reduce sulfur and ash
 prior to firing. Insufficient data were available to evaluate the cost/performance of advanced coal cleaning.
c PCC not evaluated, since boilers have small roof-mounted ESPs. Adding additional plate area is not possible,
 and the use of SOS conditioing is likely to be marginally beneficient.
                                                  11

-------
13
.§
 • x




/





/


12
^


Key
Plant
9 Elmer Smith
10 Big Sandy
1 1 Paradise
12 Cumberland
13 Johnsonville
14 Colbert
1
                   40          80           120         160
                         SO2 Reduction, thousand tons/year
                                                                     200
Figure 5.    LIMB (70% removal) and DSD costs versus SOi reduction
                                                            12

-------
TaWe 7. Summary of Sorbent Injection
Boiler
Plant No.
Burger


Sammis






Muskingum




Conesville



J. M. Stuart



E. W. Brown


Elmer Smith

Big Sandy

Paradise
Cumberland
Johnsonville
Colbert

1-6
7
8
1
2
3
4
5
6
7
1
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
1
2
1
2
3
1-2
1-10
1-4
5
Cost and Performance
Technology
DSD-FF
LIMB
LIMB
LIMB
LIMB
LIMB
LIMB
LIMB
LIMB
LIMB
DSD-FF
DSD-FF
DSD-FF
DSD-FF
LIMB
LIMB
LIMB
LIMB
DSD-FF
LIMB
LIMB
LIMB
LIMB
DSD-FF
LIMB
LIMB
DSD-FF
LIMB
DSD-FF
DSD-FF
LIMB
LIMB
DSD-FF
DSD-FF
LIMB
Estimates
Levelized
$/ton S02
1,284
602
519
736
725
693
622
540
476
499
581
583
628
640
364
597
832
754
652
655
688
607
577
7,874
773
563
7,583
588
7,684
7,302
673
469
7,747
585
567
Annual Costs
mills/kWh
25.4
12.4
10.6
3.7
3.6
3.5
3.1
7.3
6.4
6.7
14.3
14,4
16,0
6,3
9,6
14.3
13.3
12.0
10.0
4.7
4.8
4.3
4.1
17.8
7.3
5.7
26.6
10.2
10.0
7.9
9.0
7.7
11.8
8.2
8.0
SO2 Reduction,
tons/yr
10,900
10,300
15,200
3,500
3,500
3,800
4,500
18,600
35,900
31,000
27,800
27,700
27,100
26,400
83,100
5,600
6,300
8,500
55,900
79,600
79,400
78,900
23,700
4,300
9,000
23,300
6,700
20,200
8,600
26,900
49,300
279,400
44,900
53,700
34,400
13

-------
 .i
 a
 1
                               Legend
                          • LNC
                          xNGR
                              ,13
                                           14
                                           f
                                                                Key
                                                              Plant
                                                         3 Burger
                                                         4 Sammis
                                                         7 Stuart
                                                         S Brown
                                                         9 Elmer Smith
                                                        10 Big Sandy
                                                        12 Cumberland
                                                        13 Johnsonvitle
                                                        14 Colbert
                           6      8     10     12     14
                          /VO« Reduction, thousand tons/year
                                                            16
18
                                                                         20
    40
  | 30
 \

  o
8  20

-------

1^

s
i
M 0
? UOI//IUJ 'I
elized Cos
8 S
i i
^ "-
01
~4

0




\y
/f6 /
re




X

) 2C


^
X^


) 3C

', X
/y^>
^

7


Key
Plant
3 Burger
4 Sammis
5 Musk ing urn River
6 Conesville
7 Stuart
8 Brown
) 40 Si
                         N0f Reduction, thousand tons/year
    770-
    150'
 V
 I
 O
 O
   120.
    90.
    60
    30'
               j
                        '13
                                       14
          Key
                                                                 Plant
 9  Elmer Smith
10  Big Sandy
1 1  Paradise
12  Cumberland
13  Johnsonville
14  Colbert
10            20             30

      NO, Reduction, thousand tons/year
                                                               40
                        50
Figure 7.    SCR costs versus NO, reduction.
                                                            15

-------
Table 8. Summary of Low NOX Combustion Cost and Performance Estimates
                                                      Levelized Annual Costs
NOX Reductions
Plant
Burger





Sammis





Muskingum River



Conesville


J. M. Stuart



£. W. Brown


Elmer Smith

Big Sandy

Paradise

Cumberland

Johnsonville

uoiier
No.
1-2
3-4
5
6
7
8
1
2
3
4
6
7
1-2
3
4
5
1
2
3
1
2
3
4
1
2
3
1
2
1
2
1-2
3
1
2
1-6
7-10
LNC Type*
NGR
NRG
NGR
NGR
LNB
LNB
LNB
LNB
LNB
LNB
LMB
LNB
NGR
NGR
NGR
LNB
NGR
NGR
LNB
LNB
LNB
LNB
LNB
LNB
OFA
OFA
NGR
OFA
LNB
LNB
NGR
NGR
OFA
OFA
LNB
LNB
$/ton NOX
2,797
2,765
2,631
2,536
478
325
527
515
482
407
194
224
1,237
1,147
1,150
203
1,389
1,359
473
197
199
204
167
456
183
103
1,192
152
281
144
1,200
1,182
200
200
370
427
mills/kWh
6.0
5.9
5.4
5.4
1.0
0.7
0.8
0.8
0.7
0.6
0.3
0.4
4.6
4.7
4.7
0.3
5.2
5.1
1.0
0.3
0.3
0.3
0.3
1.0
0.1
0.1
5.2
0.1
0.4
0.2
4.7
4.7
0.2
0.2
0.3
1.0
Efficiency, %
50
50
50
50
50
50
35
35
35
35
40
40
50
50
50
30
50
50
50
35
35
35
35
50
25
20
50
25
35
35
50
50
20
20
25
50
tons/yr
160
170
260
300
1,070
1,570
1,000
1,000
1,100
1,400
4,600
4,000
4,200
4,300
4,200
4,400
1,300
1,500
1,100
4,400
4,400
4,200
5,200
1,000
700
1,800
1,700
1,000
2,400
6,930
8,000
13,300
6,000
6,100
300
1,200
a/JVC Types:  Low NOX burners (LNB) for dry-bottom dwall-fired boilers and  overfire (OFA) for tangential-fired boilers;
             and natural gas reburning (NGR) used on all other boilers.
                                                      16

-------
Table 9. Summary of SCR Unit Cost Estimates
Boiler
Plant No. System Type
Burger


Sammis






Muskingum River




Conesville


J. M. Stuart



£ W. Brown


Bmer Smith

Big Sandy

Paradise


Cumberland
Johnsonville
Colbert


1-6
7
a
1
2
3
4
5
6
7
1
2
3
4
5
1
2
3
1
2
3
4
1
2
3
1
2
1
2
1
2
3
1-2
1-10
1-3
4
5
Tail-End
Tail-End
Tail-End
High-Dust
High-Dust
High-Dust
High-Dust
High-Dust
High-Dust
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Tail-End
Levelized Annual Costs3
$/ton A/OX
10,940
7,272
5,004
5,048
4,940
4,632
3,937
4,094
3,992
2,242
2,352
2,363
2,585
2,567
3,302
5,705
5,071
8,067
3,497
3,527
3,600
2,977
6,548
5,846
5,704
5,272
5,847
4,161
3,474
3,612
3,694
3,304
3,346
2,428
4,728
4,864
4,004
mills/kWh
38.6
24.8
17.1
17.4
17.0
16.0
13.6
14.2
13.8
7.8
14.1
14.1
16.8
16.7
12.2
34.
30.2
27.3
12.8
12.9
13.2
10.9
22.8
14.6
14.2
36.9
16.6
15.0
12.5
22.9
23.4
20.9
12.1
6.4
17.1
17.6
14.5
NOX Reductions
Efficiency, %
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
80
tons/yr
1,900
1,700
2,500
2,400
2,400
2,600
3,100
4,800
9,200
8,000
6,800
6,700
6,900
6,700
11,600
2,100
2,300
1,800
10,000
10,000
9,700
11,900
1,500
2,200
5,700
2,800
3,300
5,200
15,900
12,900
12,900
21,300
24,100
11,400
3,500
3,500
8,800
"Costs are based on  1-year catalyst life. Costs based on 3-year catalyst life would be 40-50%  less.
                                                       17

-------
              Table 10.     Retrofit Factors Affecting Cost/Performance
                                        X


                                        X
X


X
                                                                                    X


                                                                                    X
                                                                 Additional
      Control         Access and      Ducting       Particulate        Boiler          Boiler
    Technology       Congestion      Distance        Control          Type        Configuration

Lime/Limestone
Flue Gas
Desulfurization

Lime Spray
Drying

Coal Switching/
Blending

Physical Coal
Cleaning

Furnace Sorbent
Injection with
Humidification
(LIMB)

Duct Spray
Drying

Low A/0X
Combustion

Natural Gas
Returning

Selective
Catalytic
Reduction
X


X


X




X


X
T. E. Emmel and S. D. Piccot, are with Radian Corp., Research Triangle Park, NC
 27709; and B. A. Laseke is with PEI Associates, Inc., Cincinnati, OH 45246..
Julian W. Jones is the EPA Project Officer (see below).
The complete report, entitled "Ohio/Kentucky/TVA Coal-Fired Utility S02 and NOX
 Retrofit Study," (Order No.  PB 88-244 447IAS; Cost: $49.95,  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:
        Air and Energy Engineering Laboratory
        U.S. Environmental Protection Agency
        Research Triangle Park, NC 27711
                                                           18

-------
    o
    3*.
    n
                                                                              co  -
                                                                              »  <

                                                                              |c
                                                                              q =

                                                                              2 I
                                                                              -  e

                                                                              0~
                                                                              vt
                                                                              N>
                                                                              CD
                                                                              00

-------