5ER&
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 2771 1
EPA-600/7-79 228b
October 1979
Coal Conversion
Control Technology
Volume II. Gaseous
Emissions;
Solid Wastes
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-228b
October 1979
Coal Conversion Control Technology
Volume II. Gaseous Emissions; Solid Wastes
by
LE. Bostwick, M.R. Smith, D.O. Moore, and O.K. Webber
Pullman Kellogg
16200 Park Row, Industrial Park Ten
Houston, Texas 77084
Contract No. 68-02-2198
Program Element No. EHE623A
EPA Project Officer: Robert A. McAllister
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
Information has been gathered on coal conversion process streams.
Available and developing control technology has been evaluated in
view of the requirements of present and proposed federal, state,
regional and international environmental standards. The study
indicates that it appears possible to evolve technology to reduce
each of the components of each process stream to an environmen-
tally acceptable level. The conclusion has also been reached
that such an approach would be costly and difficult of execution.
Because all coal conversion processes are net users of water,
liquid effluents need be treated only for recycling within the
process, thus achieving essentially "zero discharge." Further,
with available technology gaseous emissions can be controlled to
meet present environmental standards, particulates can be con-
trolled or eliminated and disposal of solid wastes can be managed
to avoid deleterious environmental effects.
Volume I focuses on environmental regulations for gaseous, liquid,
and solid wastes, and the control technology for liquid effluents.
Volume II deals with the control technology of gaseous emissions
and solid wastes.
Volume III includes a program for economic analysis of control
technology and includes the appendix.
11
-------
Table of Contents
(Tables of Contents for Volumes I and III
start on pages iv and vi, respectively.) Page
Abstract ii
List of Figures vii
List of Tables xii
9. Environmental Data Acquisition : Control of 523
Gaseous Emissions
Development of Conversion Process Emission 523
Stream Models
Literature Survey and Data Gathering 550
Target Pollutant Residuals 550
Commercial Emission Control Methods 562
Integrated Schenes for Emissions Control £70
Costs for Control of Gaseous Emissions 708
Need for Additional Data, Information 774
and Development
10. Environmental Data Acquisition : Control of Solid 784
Wastes
Literature Survey and Data Gathering 785
Target Pollutant Residuals 787
Dust Control 792
Costs of Dust Control 824
Solid Waste Disposal and Management £30
Cost of Solids Disposal 868
Need for Further Study 885
iii
-------
Table of Contents (Cont.)
Volume I. Gaseous Emissions; Solid Wastes page
Abstract ii
Table of Contents iii
List of Figures vi
List of Tables x
Acknowledgements xv
1. Introduction 1
2. Management Summary 4
Definition of the Problems 4
Establishment of Objectives: Environmental 6
Standards
Liquid Effluent Treatment 10
Gaseous Emission Treatment 16
Solid Waste Control 25
Economic Analysis and Program Emphasis 29
3. Conclusions 30
4. Recommendations 34
For Projection of Future Environmental Goals 34
For Studies of Liquid Effluent Treatment 37
For Studies of Gaseous Emission Control 44
For Solid Wastes Disposal and Management 48
5. Current Technology Background 50
Development of the Data Base 50
Development of Gasification Process Emission 69
Stream Models
Coal Liquefaction Processes and Date Gathering 85
Development of Liquefaction Emission Stream 92
Models
iv
-------
Table of Contents (Cont.)
Volume I (Cont). Gaseous Emissions; Solid[Wastes Page
6. Current Environmental Background : Environmental 101
Regulations
Introduction 101
Objectives of the Survey 101
Basis for Jurisdictional Selection 102
Jurisdictional Selection 104
Method of Information Acquisition 106
Specific Environmental Areas Covered. Comments 107
Summary of Most Stringent Water Quality 112
Standards
Summary of Most Stringent Air Quality 122
Standards
7. Development of Environmental Objectives 163
Comparison of Most Stringent Regulations 164
with MEG Criteria
Recommendations for Projection of Future Goals 173
8. Environmental Data Acquisition : Control of Liquid 179
Effluents
Development of Conversion Process Effluent 179
Stream Models
Literature Survey and Data Gathering 191
Target Pollutant Residuals 194
Development of the Recycle Philosophy 199
Commercial Water Treatment Methods 201
Costs of Water Treatment 387
Integrated Schemes for Wastewater Treatment 462
Efficiency of Wastevrater Treatment Schemes 498
Need for Demonstrating of Commercial Processes 516
Need for Further Study 519
-------
Table of Contents (Cont.)
Volume m. Economic Analysis Page
Abstract ii
Table of Contents iii
11. Program for Economic Analysis of Control Technology 889
Treatment of Liquid Effluents from Coal 890
Conversion
Treatment of Gaseous Emissions from Coal 900
Conversion
Treatment of Solid Wastes from Coal 905
Conversion
Basis for Economic Studies 909
The Capital Cost Model 911
The Operating Cost Model 913
Use of the Cost Models 915
12. Technology Transfer 916
Reports Completed 916
Symposia and Meetings 917
Appendix. Project Bibliography, Pullman Kellogg A-l
Reference File
Arrangement of the Project Bibliography A-2
Subject Index A-15
Accession Number Index A-69
Title Index A-201
Author Index A-238
Corporate Author Index A-305
vi
-------
FIGURES
Page
9-1 Flow diagram for SNG production by Lurgi 525
gasification
9-2 Lurgi process: coal gasification section 527
9-3 Lurgi process: gas liquor treatment section 528
9-4 Lurgi process: shift conversion 529
9-5 Lurgi process: Selexol H2S removal 530
9-6 Lurgi process: Selexol C02 removal 531
9-7 Lurgi process: methanation and drying 532
9-8 Lurgi process: incinerator/power boiler and 533
flue gas desulfurization
9-9 Lurgi process: Claus-Beavon sulfur recovery 534
9-10 Lurgi process: overall feed and product weight 535
balance
9-11 Gaseous emission streams from coal gasification 538
9-12 Coal liquefaction: SRC-II block flow diagram and 539
material balance
9-13 Liquefaction: coal slurrying and dissolving 540
9-14 Liquefaction: dissolver slurry fractionation, 541
filtration and naphtha hydrogenation
9-15 Liquefaction: gas purification, drying and 542
separation
9-16 Liquefaction: process gasifier and hydrogen 543
production
9-17 Liquefaction: fuel gas gasification, purification 544
and sulfur recovery
vii
-------
FIGURES (Cont.)
Page
9-18 Liquefaction: sour water stripping and 545
ammonia recovery
9-19 Liquefaction: sulfur recovery and tailgas treating 547
9-20 Liquefaction: process furnaces and steam 548
and power generation
9-21 Liquefaction: overall feed and product weight balance 548
9-22 Gaseous emission streams from coal liquefaction 549
9-23 Fuel nitrogen conversion as a function of fuel 566
nitrogen content
9-24 Typical hydrotreater unit 570
9-25 Effect of space time on denitrogenation of 572
Anthracene Oil
9-26 Effect of weight hourly space time on 572
denitrogenation of COED oil
9-27 Effect of space time and temperature on denitro- 573
genation
9-28 Effect of fuel nitrogen on NOX emissions 575
9-29 UOP/Shell NOX removal process 581
9-30 Unconverted NO- as a function of catalyst bed 585
length for UOP process
9-31 Performance of Shell reactor at SYS 586
9-32 Typical flow scheme for a fixed bed hydro- 598
desulfurization process
9-33 Flow diagram of the modified Meyers process 601
9-314 Typical schematic and reactions for catalytic 637
conversion processes
viii
-------
FIGURES (Cont.)
Page
9-35 Characteristics of particles and particle 640
dispersoids
9-36 Types of collectors for various constituents 641
9-37 Operating principle of a dry vertical cyclone 642
collector
9-38 Typical frational performance curves for a multi- 642
tube mechanical collector
9-39 Operating principles of a surface type fabric filter 646
9-40 Resistivity of fly ash 649
9-41 Sketch of Ecodyne "Hi-V" drift eliminator system 654
9-42 Fall velocity of water drops as a function of size 657
9-43 Dynamic behavior of cooling tower drift 659
9-44 Natural sea salt concentration in air 661
9-45 Investment vs. emission reduction 667
9-46 Organic abatement operating cost 668
9-47 Sulfur balance: Lurgi gasification base case with 671
low sulfur coal
9-48 Sulfur balance: Lurgi gasification with high sulfur 672
coal
9-49 Sulfur balance : Bi-Gas process with low 674
sulfur coal
9-50 Sulfur balance: Bi-Gas gasification with high sulfur 675
coal
9-51 Sulfur balance : SRC-II liquefaction 676
9-52 Claus process flowsheet 702
9-53 Beavon tail gas treatment process 704
9-54 Installed capital costs of particulate control 713
devices
IX
-------
FIGURES (Cont.)
Page
9-55 Coal preparation capital cost 718
9-56 Deleted
9-57 Deleted
9-58 Meyers process capital cost 724
9-59 Meyers process incremental processing cost 725
9-60 Capital cost comparison: FBC vs. conventional 730
boiler with FGD
9-61 Steam cost comparison: FBC vs. conventional 731
boiler with FGD
9-62 Capital investment for limestone and lime slurry 737
FGD processes
9-63 Capital investment for magnesia slurry and 738
catalytic oxidation FGD processes
9-64 Effect of sulfur content of coal feed on FGD 739
operating cost
9-65 Effect of plant size on FGD operating costs 740
9-66 Capital cost comparison: limestone slurry vs. 743
citrate process without HgS generation
9-67 Operating Cost Comparison: limestone slurry vs. 744
citrate process without H2S generation
9-68 Sulfur dioxide control costs for coal fired boilers 747
9-69 Sulfur dioxide control costs for oil fired boilers 750
9-70 Capital investment for NOY/SOp
control 754
9-71 Operating costs for NOX control for oil fired 757
boilers
9-72 Capital investment for fuel oil hydrotreating 759
9-73 Operating costs for fuel oil hydrotreating 760
-------
FIGURES (Cont.)
Page
9-71* Operating costs for NOX/S02 control
for oil fired boilers 764
9-75 Operating costs for NOX/S02 control for oil
fired boilers 767
9-76 Capital investment for Claus-Beavon sulfur 770
recovery
9-77 Operating costs for Claus-Beavon sulfur recovery 772
10-1 Size consist of crushed coals 796
10-2 Solid wastes disposal area 869
xi
-------
TABLES
Page
9-1 Target Pollutant Residuals: Most Stringent Air 551
Standards
9-2 Conversion of Fuel Nitrogen to NO 565
X-
9-3 NO Reduction with Boiler Modifications 568
a
9-4 UOP/Shell NO Control System Treating Flue Gas 583
from Generation of 500 Megawatts
9-5 Important Hydrodesulfurization Processes 598
9-6 Desulfurization of Coal via the Meyers Process 604
9-7 Methods of Cleaning U.S. Bituminous Coals 608
and Lignite
9-8 Summary of Composite Product Analysis by 610
Region for Crushed and Cleaned Coals
9-9 Characteristics of Commercial FGD Processes 617
9-10 Characteristics of Advanced FGD Processes 618
9-11 Size and Mass Distribution of Drift Particles 656
9-12 Coal Properties 677
9-13 Gaseous Fuels to Incinerator 680
9-14 Liquid Fuels to Incinerator 681
9-15 Incinerator Flue Gas Composition 682
9-16 Possible Future Goals for NO Emissions 690
^C
9-17 Flue Gas Desulfurization with the Citrate Process 698
9-18 Preparation Plant Capital and Operating Costs 715
xii
-------
TABLES' (Cont.)
Page
9-19 Coal Preparation Plant Characteristics 717
9-20 Coal Cleaning with the Meyers Process 721
9-21 Comparison of Investments and Cost of Steam for 727
Single Boiler Added to Coal Fired Plant
9-22 Comparison of Investments and Cost of Steam for 728
Single Boiler Added to Oil Fired Plant
9-23 Comparison of Investments and Cost of Steam for 729
Grassroots Boiler Plants with Backup
9-24 EPA-Sponsored Stack Gas Desulfurization Demonstra- 733
tion Systems
9-25 Flue Gas and Sulfur Dioxide Emission Rates for New 734
Coal Fired Power Plants
9-26 Required Removal Efficiencies in FGD Units 734
9-27 Capital Investment for FGD Units 736
9-28 Operating Costs for FGD Units 736
9-29 Cost Comparison of Limestone Slurry and Citrate 742
FGD
9-30 Control of NO /SO- for Oil Fired Boilers 763
X t*
10-1 Size Distribution of Products from Coal Crushing 794
10-2 Coal Crushing with 1.25 in. Opening 795
10-3 Size Consist of As-Received Coals 797
10-4 Capital and Operating Costs for Wet Dust 826
10-5 Ash and FGD Sludge Production without Sulfur 835
10-6 Volume of Solids from Coal Conversion without 838
FGD
Kill
-------
TABLES (Cont.)
Page
10-7 Capital Costs of Oil-Fluidized Evaporation 847
10-8 Estimated Disposal Site Construction Costs 871
10-9 Installed Costs of Disposal Area Liners 872
10-10 Estimated Capital Costs of Lined Disposal Areas 874
10-11 Estimated Capital Costs of Soil and Clay Lined 876
Disposal Areas
XIV
-------
SECTION 9
ENVIRONMENTAL DATA ACQUISITION: CONTROL OF GASEOUS EMISSIONS
DEVELOPMENT OF CONVERSION PROCESS EMISSION STREAM MODELS
The study of available information on gaseous emissions from coal
conversion processes paralleled the study on liquid effluents,
discussed in Section 8, and utilized the same sources. As the
study progressed it became apparent that the gaseous emissions
from gasification processes operating at low temperatures and
those from gasification processes operating at high temperatures
were, with few exceptions, basically similar and therefore solu-
tions proposed for emission problems in one category would rea-
sonably well fit the other. Although there were much less data
and information on emissions from liquefaction processes, it
appeared that the emissions from the several liquefaction pro-
cesses were sufficiently similar to warrant consideration as a
class, but sufficiently different from the gasification emissions
to warrant separate consideration.
In the EPA planning of the overall investigations into discharges
from coal conversion plant operations, other contractors were to
investigate means of recovering products and byproducts from the
conversion processes and determine compositions and quantities of
the emission streams from the recovery operations. These emis-
sion streams were to have been the starting point for Pullman
Kellogg1 s efforts on application of control technology of such
523
-------
efficiency that the final gas streams released to the atmosphere
would be of a quality to meet or be better than present environ-
mental standards.
Unfortunately, the emission stream information was not developed
by the other contractors, and therefore Pullman Kellogg had no
recourse but to develop process flowsheets and material balances
from available published information in order to estimate the
compositions and quantities of gases from recovery processes.
The time required to develop the flowsheets and material balances
shortened the time available, within the framework of Pullman
Kellogg's contract with EPA, for investigation of emission
control technology.
As a result of these constraints, the investigations take the
form of general descriptions of the control technology to be
applied to a gaseous emission, examples of application of the
technology whenever these were available, evaluation where pos-
sible of means of increasing the control process efficiency and
cost information. The results of the investigations are most
valuable as a means of pointing out areas of strengths and
weaknesses in available control technology and emphasizing the
possibilities for future development of control technology to
meet future environmental goals.
The Lurgi Dry Ash process was selected as the base case for ap-
plication of control technology to gaseous emissions from coal
gasification processes because of its commercial status, the
amount of information available from commercial operation, the
number of studies that had been made on the process, and the
collaboration with -the liquid effluent studies. The flowsheet in
Figure 9-1 is based on the C. F. Braun conceptual designs (1)*
for gasification of western (low sulfur) coal to produce 250
*Item in reference list.
524
-------
un
ro
on
Figure 9-1.- Flow diagram for SNG production by Lurgi gasification.
-------
million standard cubic feet of SNG per day, supplemented in
several sections with information from reports by Cameron Engi-
neers (2), from private communications with C. F. Braun and
Allied Chemical Company concerning the Selexol process and from
work done by Pullman Kellogg operating within a confidentialty
agreement with Allied Chemical but without disclosure of any
confidential information.
Included in Figure 9-1 are a Glaus plant for sulfur recovery, a
Beavon plant for treatment of Glaus plant tail gas and, for the
incinerator/boiler, a flue gas desulfurization unit based on the
U. S. Bureau of Mines citrate process. Details of these sections
of the flowsheet will be discussed in "Integrated Schemes for
Emission Control."
Details of the sections of the Lurgi flow diagram are shown in
the following:
Figure 9-2. Coal gasification section.
Figure 9-3. Gas liquid treatment.
Figure 9-U Shift conversion.
Figure 9-5. Selexol H2S removal.
Figure 9-6. Selexol C02 removal.
Figure 9-7 Methanation and drying.
Figure 9-8 Incinerator/boiler.
Figure 9-9. Claus-Beavon plants.
An overall feed and product weight balance for the plants is
shown in Figure 9-10. Completion of the base case Lurgi process
overall material balance was necessary in order to calculate the
hydrogen sulfide feed to sulfur recovery for the base case for
operation with low sulfur coal and for the alternate case of
Lurgi operation with high sulfur coal. Development of this
information then allowed inquiry into means of reducing the final
526
-------
COAL 22136.
STEAM 22170.0
OXYGEN 4999.4
niisr LOSS i
COAL
PREPARATION
19440.0
COAI.
GASIFICATION
COAL FINES TO
INCINERATOR 26957T
WASTE
HEAT
BOILER
CRUDE GAS TO SHIFT
CONVERSION 42246.
TARRY GAS LIQUOR 3329.4
RECYCLE GAS LIQUOR
357.1
ASH
1390.41,
ASH
QUENCH
^ ASK SLURRY TO DISPOSAL 3016
(FLOWS IN STPD)
Figure 9-2. Lurgi process: coal gasification section.
-------
nTT.Y GAS LIQUOR 12472.7
TARRY GAS LIQUOR 3329.4
EXPANSION GAS
746.0
PROCESS CONDENSATE RECYCLE
1236.0
TAR AND OIL
SEPARATION
RECYCLE GAS LIQUOR
357.1
(Ji
to
00
CLEAN GAS LIQUOP 12316.6
CONTAMINATED GAS LIQUOR 1969^6
PIIENO-
SOLVAN
PROCESS
TAR 1065.6
TAR OIL 583.2
DEPHENOLIZEC
CLEAN GAS
LIQUOR
12525.0
AHMOHI
FROM NH3 SCRUBBER 1.1
>EPI]ENOLIZf~t> CONTAMIMAT-
ED GAS LIQUOR
1626.1
GAS-LIQUOR
STRIPPING
ACID GAS 212.6
WATER
12056.5
ANHYDROUS AMMONIA
257.0
(FLOWS IN STPD)
Figure 9-3. Lurgi process: gas liquor treatment section,
-------
SHIFT BYPASS 19011.1
cn
ro
GASIFICATION 42246.7
OILY GAS LIQUOR 12472.7
SHIFTED GAS TO
REMOVAL 28350.5
NAPHTHA (C£H£ ) 187.5^
(FLOWS IN STPD)
Figure 9-4. Lurgi process: shift conversion.
-------
SHIFTED SAS
28350.
RECYCLE
GAS
GAS TO C0? REMOVAL
27184.4
ABSORBERS
RICH
SOLVENT
FLASH AtlD
RECYCLE
COMPRESSION
STRIPPERS
H S-nTHll fAS.Tn (-TJUIS t!NTT
AND FCD UNIT
1166.1
LEAN SOLVENT
(FLOWS IN STPD)
Figure 9-5. Lurgi process: Selexol H2S removal.
-------
Ul
OJ
rpriM ii
REMOVAL
RECYCLE GAS
C°2
ABSORBERS
r.icn
SOLVENT
FLASH,
CHILLING &
HECYCLE
COMPRESSION
GAS TO SULFUR GUARD
9110.4
VENT CO- TO INCINERATOR 3595.8
co2
STRIPPERS-
LEAN SOLVENT
j FLASH CO., TO INCINERATOR
1 2- - •-- —
(FLOWS IN STPD)
NITROGEN 615.2
15093.4
Figure 9-6. Lurgi process: Selexol CC>2 removal.
-------
U1
LO
to
ZINC OXIDE MAKEUP
GAS FROI1
CO2 REMOVAL
9110.4
SULFUR
GUARD
BED
9110.4
UiiifOSATT
METHANATION
AND
HEAT RECOVERY
STEAM
GENERATION
T TO RECOVERY
COMPRESSION
AND
COOLING
DRYING
CONDENSATE
CUf GilCU A _
1652.0
(FLOWS IN STPD)
Figure 9-7. Lurgi process: methanation and drying
-------
^S FROM SCLC XOL UN IT
777.0
CO
u>
NAPHTHA
COAL FINES
FLASH C02
EXPANSION r,AS
TAR
TAR OIL
VENT CO,
ACID GAS
PHENOL
COMBUSTION AIR
187.5
2695.3
15093.4
746.0
1065.6
583.2
3595.8 ^
212.6 ^
135.1
56539.5 __
INCINERATOPv
AND
POU'EK
BOILEP,
FLUE GAS 30735.2
EXCESS H2S 43.5
FLUE CAS
DEGULFl.R-
IZATIOri
UNIT
STACK GAS 81388.6
SULFUR eo.i
ASH 162.3
_ _ B.^
Figure 9-8. Lurgi process: incinerator/power boiler and flue gas
desulfurization.
-------
VCHT n
17.1
r«0f: SCLCXOC
H?S TO
SCRUB9ER
777
HATtt
0.9 1
ftfWONIA
350.1 SCRUBBER
A
61.
338.9
n
CLAUS
PLAHT
ACUEOUS
AtVIONIA TO
MS It ODOR
STRIPPING
,1.1
T
r^
_f.»« f»B i-tlf
T4lL HM
f L_
1
HtlPUCItlG GAS
1
REACTOR
lEAVOft PLANT
151.5
33.q
AESOOIER
'
OX
IDIZCR
37.0
ftCAVON PLANT '
1
I
. iin'-jn 1 °-9
cco«r.,nn 1 k'«TF» m
1
J
SULFUR
1-5 SULFU, TO SALES .
22
(FLOHS IN $TPD)
Figure 9-9. Lurgi process: Claus-Beavon sulfur recovery.
-------
cn
u>
en
WATER TO AMI ION I A SCRUDDER 0.9
NATURAL GAS OR SNG 2.G
OXYGEN TO GASIFIERS 4999.4 ^
NITROGEN TO CXu STRIPPER 615.2 _.
KTVAM TO GASIFIERS ;pi?n n ^
AIR TO GLAUS AND BEAVON UNITS 146.8
COAL 22136. 7 19441.4 ^
TARS, OILS, NAPHTHA, PHENOL,
GAS STREAMS
2695. 3 ^
COMBUSTION AIR 56039.5 ^
SNG
FROM
COAL
PLANT
INCINERATOR
BOILER
RF.AVON .SOLUTION PURGE
WflTKp VUnM RKAVON L'NI1?
COAL DUST LOSSES
BEAVON VENT
flHC (250 MM rtHi/Dflyl
ANHYDKOUS AMMONIA TO SALPS
CLEAN WATER
DEPIIENOLIZED CONTAM. GAS LIQUOR
METHANATOR CONDENSATK
CLAUS/BEAVON SULFUR
GASIFIER ASH
INCINERATOR ASM
REGEWERABLE FHD SULFUR
STACK CAS
0.9 ^
20.0 ^
1-4 ^
489.4 ..
^45B.4 ^
257.0
12056. 5
1626. 1
3G52.0
26.0 _
1390.4 ^
162. 3
80.1
bl388.6
(FLOWS IN STPD)
Figure 9-10. Lurgi process: overall feed and product weight balance-
-------
emissions of sulfur compounds into the atmosphere to meet
environmental standards.
A similar series of calculations was carried out for the Bi-Gas
high temperature gasification process, using the C. F. Braun
study (1) as a basis, to establish sulfur balances for operation
with high and low sulfur coal. The four sulfur balances are dis-
cussed later in this section of the report in "Integrated Schemes
for Emissions Control". Figure 9-11 illustrates the gaseous
emission streams from coal gasification to which control techno-
logy must be, or has been, applied.
The SRC II process was selected as being representative of the
liquefaction processes, principally because of the availability
of process data and information and also because of the colla-
boration with the liquid effluent studies in this project. The
conceptual engineering design of Ralph M. Parsons (3) was used
as a basis to typify the process streams and the gaseous emission
streams. Figure 9-12 is the flowsheet of the SRC II process
based on the Parsons design to convert 20,000 TPD of dry, high
sulfur, eastern (Illinois, Indiana, Kentucky) coal fed to the
dissolvers into SNG, liquefied petroleum gases, naphtha, and fuel
oil. As in the Lurgi gasification flowsheet, sulfur recovery and
sulfur emissions control systems are included in Figure 9-12,
details of which will be discussed later.
Details of the sections of the liquefaction flow diagram are
shown in the following:
Figure 9-13. Coal slurrying and dissolving.
Figure 9-14. Dissolver slurry fractionation, filtration, and
naphtha hydrogenation.
536
-------
Figure 9-15. Gas purification, drying,and separation.
Figure 9-16. Process gasifier and hydrogen production.
Figure 9-17. Fuel gas gasification, purification,and sulfur
recovery.
Figure 9-18. Sour water stripping and ammonia recovery.
Figure 9-19. Sulfur recovery and tail gas treating.
Figure 9-20. Process furnaces and steam and power genera-
tion.
The overall feed and product weight balance for the plant is
shown in Figure 9-21. The gaseous emission streams from coal
liquefaction to which control technology must be, or has been,
applied are shown in Figure 9-22. A sulfur balance considering
the same high sulfur coal is discussed in"Integrated Schemes for
Emissions Control" in this section of the report.
References
1. Detman, R., "Factored Estimates for Western Coal Commercial
Concepts." October 1976. FE 2240-5. 29^, 295, 296*
2. Sinor, J., "Evaluation of Background Data Relating to New
Source Performance Standards for Lurgi Gasification." 1977.
EPA 600/7-77-057. 552*
3. O'Hara, J., et al. , "Oil/Gas Complex Conceptual Design/Econo-
mic Analysis: Oil and SNG Production." March 1977. FE-
1775-B. 814*
•Pullman Kellogg Reference File number,
537
-------
COflt STORAGE,
PREPARATION,
rattans
DUST (1)
GASIFICATION
CAS
QUENCH
SHIFT
CONVERSION
GAS
COOLING
METKAHATIOH,
COMPRESSION,
DRYING
SNC
OXYGE9
PLANT
NITROGEN (2)
LOCK HOPPER VENT (3)
ASH QUENCH VENT (4)
ASH OtJESCH
GAS-LIQUOR SEPARATION
PHENOL EXTRACTION,
GAS-LIQUOR STRIPPING
NAPHTHA
BYPRODUCT
STORAGE
TANKS
TANK
VENTS (5)
^MISCELLANEOUS LEAKS
EXPANSION GAS,
TA8, 7AH OIL,
PHENOLS, ACID
CAS
FLUE GAS (12)
FLASH t VEST CO.
TAIL
CLAUS
PLANT
ny^
BEAVON
PLANT
VENT r
GAS (I'M
COOLING
TOV.'ER
EVAPORATIOtl & DRIFT (18)
DRYER VENT (19)
Figure 9-11. Gaseous emission streams
from coal gasification.
538
-------
, HATED 104.0
OXYGEN STEAM DUST TO DISPOSAL STEAM. NITROGEN
4497. 2~J £"8538.7
COAL PROCESS2" 26, 267. 9^
10,000.0 fcAblFIER
!"•' i""'0' '"'"'I
I "1
CYCLONE AND° 21,255.5 ,;nIPT 21 22545.0 ACID GAS22 1532.6 2)50.3
DUST FILTER CONVERTER REMOVAL |
«J RECOV. CHAR 4987.6 |
RECOVERED HATER 558.6 ^
CONDENSATE 2198.8
~* SOUR WATER 1497.3
SLAG |
197.0 QUENCH
SLAG SLURRY TO DISPOSAL |
4788.1
SOUR
HATER
532.6
CO ( COAL — _ cn.t. 12 79,467.4
^ 755.9 '
HATER FROM HATER TREATMENT
2835.2
HYDROGEN
4825.3
* 6590.0
^_ COAL " ««-S ' ACIO GAS17_lt npY,«G
VO ' 20.000.0 SLURRY1NG DISSOLVING REMOVAL GEN1C S
TO UNIT 2
1417.9
H2O 24. 1 1
HYDROGEN J^
J"*'3 FROM UMIT 17
(90.8 136.9
4434.5
FROM
™JT FROM
1417.9 ™JT
J- 776.0
ntioniU.Ht.U
COMPEHSATE f
~* 5.4
NAPHTHA FROM UNIT 14^
1 1278.8
N
1
& CRYO- H LNG
EPARATION IFRACTIONATIO
3663.9 ^_ SNG
P ^ 53 5 ' CONDENSATE PURIFICATIO
fillLnin " 3319.1^ TAIL GAS 2? CLEAN EXIT GAS
PLANT TREATMENT 3101.8
eia.o 42.8 j— —
SULFUR 860. 8
. .. — i^STEAM
HTHA
19 -5 C3 LPG 531 .7
N C( i.pf: jni -i
(32.0
3307.3 SNG 3939. J
tt
NAPHTHA TO UNIT 16 ^~~ ] j ] *" i H,O VENT i »
, ,
1 4 , •
^TRIPPING STEAM J {„,„.., . 1 L-MAKEUP HASH OIL
TO BFW 353.3
DUST TO DISPOSAL
284.9 1
„ 1* HOT CHAR ^ ^4 25499.6 SULFUR SULFUR
RECOVERED SOLVBIT t . ™ „.„ , FROM UN1T_^2 _UELJT _BOCE
20,000.0
3504.6 | 19.682.6
i """" "'"'" - •- FUEL CA/' _J r,r.,...f ' __ MS 24| STEAM I POWER OFFGAS
^ AIL. '.T.'.. '• GASIFIER | ^1^-lAJnc- COOLING | GKNrRATIOM 38447 2
stjln SOUR WATER TO UNIT 26
Figure 9-12.- Coal liquefaction: SRC-II block flow diagram and material balance,
-------
Ul
4*
O
COAL FEFD
40000
SOLVENT RECYCLE
FROM FRACTIONATION
4000('
SOLVENT RECYCLE
FROM FILTRATION
RECYCLE HYDROGEN FROM
CRYOT.RNIC SEfMWTION
1117
HYDROGEN FROM
PROCESS GARIFIER
3638.3
SLURRYING
SOUR HATER TO
Nil. STRIPPING
S3J.6
SOUR GAS
S TO Al
7243.5
ACIO RA8 REMOVAL
COAL
DISSOLVING
SLUR
73199
•¥•
J5OUR WATER
~n»T»
CONOENSATE
3589
(FLOWS IN STTD)
FRACTIONATION
Figure 9-13. Liquefaction: coal slurrying and dissolving.
-------
FROM METHANA7ION
Ul
SOHR CAS TO COAL DISSOLVING
~*~ " 379.2
STEM
HOT CIIAR FROM FIJI'L HAS GASIFIER
5001.6
SLURRY FROM DISSQtVER
'7428.4
SOLVENT RECYCLE TO
COAL SLURRYING
40000
SOUR WATER TO NH., STHIP
1417.9 15495.3
SOLVENT RECYCLE TO COAL SLURRYIHG
20000
(FLOWS IN STPD)
Figure 9-14
Liquefaction: dissolver
slurry fractionation, filtration.
and naphtha hydrogenation.
-------
SOUR GAS FPOM
COAL DISSOLVING
7265.6
7243.5
SOUR GAS FROM
NAPHTHA HYDROCnNATION
22.1
TO COAL DISSOLVING
1187
!I2S TO SULFUR
RECOVERY
690.8
FUEL GAS
8.9
H2-R1CH
GAS
ACID GAS
REMOVAL
CONDHNSATE
24.1
6590
39.1
METHANATION
TO NAPHTHA HYDROC3ENATION
33. /
CONDENSATE
5.4
DRYING AND
CRYOGENIC
SEPARATION
COMDENSATE
11.1
3663.9
SNG
PURIFICATION
II .0 VENT
3307.3
ro BFW
353.3
1688.9
LHG
FRACTIONATION
NAPHTHA TO HYDROGENATION
C.LPG
405.7
SNG
632.0
C-LPG
531.7
(FLOWS IN STPD)
Figure 9-15. Liquefaction: gas purification, drying,and separation
-------
STEAI1
en
.&.
u>
TO ATMOSPHERE
'1788.1
-------
D?Y-FJLIE» CAKF t CUM,
8959.1
fUEL GAS
GASIflEP
AIR I
2^31.7 SL.G
5001
SLURRY TO DISPOSAL
~~" 10001.7
.7
SLAG
OUENCH
to CYCLONE »• l,A!, ^ ' LELIHUSIAT1C *" sul CUR PLAN.! fyfL GAS ^,
315B2.1 ~ *""" * 26560.5 ""' 257811.5 "«-inuai«i iv 25*199.5 st;'ov*t. 2513.7
HOT CHAR TO MLTER sou" """ t0
_, CME DRYING_ |^Hj STRIPPING ^ , tu?I TO, BLSPJ1JAU SULFUR
50C1.6 77C 281.9 370.9
~"~7200
(FLCMS IN STPD)
Figure 9-17. Liquefaction: fuel gas gasification, purification and sulfur
recovery.
-------
O1
532.6
FROM SLUJtRY FRACTIONATION
1417.9
FROM NAPHTHA 1IYDROGENATION
PROCESS GASIFIER
SHIFT CONVERTER
1497.3
FROM FUEL GAS COOLING
776
1128
SOUR HATER FEED
4434.5
265.7
ACID CA3>TO SULfUR RECOVERY
AMMONIA
SEPARATION
4168.8
755.9
ANHYDROUS AMHoqiA TQ r*
STRIPPEO HATER TO STEAM SYSTEM
420.8
STRIPPED MATER TO PROCESS CASIFIER
(FLOWS IN STPO)
Figure 9-18. Liquefaction: sour water stripping
and ammonia recovery.
-------
Ul
FROM LIQUEFACTION
690.8
FROM PROCESS GASIFIER _
FROM AMMONIA RECOVERY '
AIR
1786.8
ACID GAS FEED
126.9
2350.3
SULFUR
RECOVERY
3319.1
818
TAIL GAS
TREATING
OFFGAS TO STACK
3101.8
42.8
CONDENSATE TO STEAM SYSTEM
174.7
SULFUR TO STORAGE
860.8
(FLOWS IN STPD)
Figure 9-19. Liquefaction: sulfur recovery
and tail gas treating.
-------
FROll PROCESS GASIFIER ACID GAS REMOVAL SYSTEM ^
18764.6
FROM FUEL GAS GASIFIER SYSTEM
25105.4
19682.6
STEAM AND
POWER
GENERATION
AIR
31444
5422.8
PROCESS
FURNACES
AIR
8106.9
(FLOWS IN STPD)
COMBUSTION GASES TO STACK
69891.2
COMBUSTION GASES TO STACK
13529.7
Figure 9-20. Liquefaction: process furnaces and
steam and power generation.
-------
STEAM 1»S1. 9
HATER 156.8
SOLVENT 105.6 ^
AIR TO SURFUR RECOVERY 1786.8 ,_
COAL 20000.0 __
OXYGEN
MATER
STEAM
NITROGEN
HYDROGEN 1
ACID GAS
10000.0 ___
4497.2 ^_
2575.1
4427.0
986. S
5670.0
AIR 19934.7
MR
39550.9
'
LIQUEFACTION
AMMONIA
SULFUR
RECOVERY
PROCESS
GASIFIES
NAPHTHA 1278.4
FUEL OIL 11309.0
FUEL GJkS * ' n
SHC 3939.3
pRnpksr fiA^ 531.7
BUTAME LPG 405.7
WATER VEST 3.3
CLEAN MATER 9Rf 1
SULFUR 860.8
WASTE WATER 755.9
AMMONIA 90.1
SULFUR PLANT OFFGAS 3101.8 _
FILTER CAKE
ASH
1197.0
DUST 24 . 8
OFF-GASS FROM
PCID GAS REMOVAL
FUEL GAS
GASIFiER
SULFUR
RECOVERY
ASH
DUST
SULFUR
FUEL f AS
PROCESS
FURNACES
STEAM
DOILERS
3001.7
284.9
394.2
cormusTiON GASES 83420.9
TO STACK
Figure 9-21.
Liquefaction: overall feed and
product weight balance.
548
-------
Ul
-pi
NITROGEN (2)
COOLING TOMER DRIFT (31 _
COOLING
TOWER
LEAKS (4)
1
PROCESS
LEAKS
1 BYPRODUCT
STORAGE TANKS
TANK VENTS (5)
DUST (6)
T 1
ELECTROSTATIC
PRECIPITATOR
SULFUR
(STRET
REMOVAL! 1
FORD) 1 ^
STEAM < POKES
GENERATORS
PROCESS
HEATERS
COMBUSTION CASES (11)
COMBUSTION GASES (12)
^-SULFUR
Figure 9-22.- Gaseous emission .streams from coal liquefaction.
-------
LITERATURE SURVEY AND DATA GATHERING
The project literature survey methods have been treated in Section
5 under the headings "Information Procurement, Storage, and
Retrieval" and "Subjects Monitored."
Efforts in the collection of pertinent data and information on
gaseous emissions from conversion processes closely paralleled
the efforts of the water treating group in collection of liquid
effluent data. Many of the same source reports and personal
contacts produced information for both groups. Much additional
information was gathered from Pullman Kellogg reports on control
of gaseous emissions and designs for emission control processes,
although great care was taken to avoid exposure of confidential
information. These sources were supplemented by contacts with
process and equipment vendors.
TARGET POLLUTANT RESIDUALS
As described in Section 6 of this report, the environmental
standards group gathered into a separate report the federal
state, regional,and international regulations, both present and
proposed, concerning permissible levels of contaminants in
emissions to the atmosphere. A summary of the most stringent of
these regulations was developed and was used as a standard for
comparison of the efficiency of processes for emission control.
In TABLE 9-1 are shown the target pollutant residuals that were
used as criteria in evaluation of emission control technology.
550
-------
TABLE 9-1. TARGET POLLUTANT RESIDUALS:
MOST STRINGENT AIR STANDARDS
Visible
Any source
Incinerator
Participates
Coal fired
Oil fired
Gas fired
Combined
fuels
20% Opacity (No. 1 Ringelmann Chart)
10$ Opacity (No. 0.5 Ringelmann Chart)
Heat Input
MM Btu/hr
> 250
< 250
100
10
1
_> 114
< 114
> 2,500
Any Size
Standard,
Ib/MM Btu
0.05
0.02 (_< 2
0.151
0.275
0.5
0.005
0.10
0.10
microns dia.)
All
Mont.
N.M.
N.M.
Colo.
Colo.
Colo.
N.M.
111.
Tex.
111.
E=SSH3 + 0.10 H^
where
E=Allowable particulate emission,
Ib/MM Btu
Ss=Solid fuel particulate emission
standard, Ib/MM Btu
H3=Actual heat input from solid
fuel, MM Btu/hr
Hn=Actual heat input from liquid
fuel, MM Btu/hr.
Fugitive Dust None visible outside property line
Kan.
Nitrogen Oxides
Coal fired
Oil fired
Gas fired
Combined fuels
Heat Input,
MM Btu/hr
_> 250
>. 250
>. 250
Any Size
Standard,
Ib NO?MM Btu
0.45 N.M.
0.30 Most
0.20 Most
E=(0.2X+0.3Y+0.7Z)/(X+Y+Z) Colo.
where
E=lb NO /MM Btu
X,Y,Z=* of total heat input from
gas, liquid and solid fuels,
respectively.
551
-------
TABLE 9-1. TARGET POLLUTANT RESIDUALS:
MOST STRINGENT AIR STANDARDS (CONT)
Hydrocarbon Vapors
Organic materials 15 Ib/day, 3 Ib/hr*
PCR (photoehemieally reactive) 40 Ib/day, 8 lb/hr»
Non-PCR 3,000 Ib/day, 450 lb/hr»
Colo.
Colo.
Colo.
Other Chemicals
Fluoride
Beryllium
Nitric acid mist and/or
vapor
Hydrochloric acid mist
and/or vapor
Mercury
•Unless reduced by 85$
6 ppb(v) as HF, 3 hr. ave.
0.01 yg/m , 24 hr. ave.
70 mg/500 Nm3
210 mg/500 Nm3
7.05 lb/24 hr. from
incineration of wastewater
treatment sludges
Tex.
Tex.
W. Va.
W. Va.
Fed.
Sulfur Dioxide (Fossil fuel fired steam generators)
Coal fired
Oil fired, ASTM
grades 4, 5, 6
Oil fired, ASTM
grades 1, 2
Gas fired
Combined Fuels
Heat Input,
MM Btu/hr
< 250
T 250
<. 115
> 115
< 250
> 250
Any Size
Standard,
Ib/MM Btu
1.2 Okla.
0.2 Wyo.
1.0 111.
0.34 N.M.
0.3 111.
0.3 111.
0.13 Okla.
50 grains H2S/100
SCF exit gas 111.
where
E=lb SOo/hr
Sg=Solid fuel emission
standard, Ib/MM Btu
HS=Actual heat input from
solid fuel, MM Btu/hr
H^sActual heat input from
distillate oil fuel
(ASTM grades 1,2),
MM Btu/hr
Sj-sEmission standard for
residual oil fuel (ASTM
grades 4, 5, 6), Ib/MM Btu
H =Actual heat input from
* residual oil fuel, MM
Btu/hr
552
-------
TABLE 9-1. TARGET POLLUTANT RESIDUALS:
MOST STRINGENT" AIR STANDARDS (CONT)
Carbon Monoxide (Corrected
to 50% excess air)
Fuel combustion
Incinerators
Petroleum processes
Petroleum processes, FCCRU*
200 ppm(v) CO
500 ppm(v) CO
350 ppm(v) CO
350 ppm(v) CO
Odors
Waste Gas Disposal
Gasification Plants
S02 , Gas burning boilers
Sulfur
H2S, COS, CS2
H2 S component
Ammonia
HC1
HCN
Sulfur Recovery Plants
Sulfur dioxide
0.16 Ib S09/MM Btu(LHV)
0.008 Ib S7MM Btu(HHV)
100 ppm(v) in effluent gas
10 ppm(v) in effluent gas
(max.)
25 ppm(v) in effluent gas
5 ppm(v) in effluent gas
10 ppm(v) in effluent gas
Hydrogen sulfide
Petroleum Processing Facilities
Hydrogen sulfide
Fuel gas burning equip.
Mercaptans
Ammonia
0.01 Ib S02/lb sulfur
processed
100 Ib S02/hr (max.)
150 ppm(v) in effluent gas
10 ppm(v) in effluent gas
0.1 grain/500 SCF in fuel
gas
0.25 Ib/hr total mercaptans
25 ppm(v) in effluent gas
111,
111,
111,
111,
Prevented beyond property line All
Flares to be smokeless All
N.M.
N.M.
N.M.
N.M.
N.M.
N.M.
N.M.
0., Ala.
Okla.
Ala.
N.M.
N.M.
N.M.
N.M.
3
Sulfuric acid plants
Other than sulfuric acid
plants
General Processing Facilities
Sulfur dioxide
35 rag/Nm3 of effluent gas
<1,300 T?Y H2S04 usage:
0.1 Ib/hr
>1,300 TPY H2S04 usage
~ 0.5 Ib/T H SO used
2 4
500 ppm(v)
Mo.
111.
111.
Colo,
Ohio
Pa.
»FCCRU= Fluid catalytic cracking regeneration unit
553
-------
TREATMENT OF GASEOUS EMISSIONS AS SUGGESTED BY OTHERS
In the discussion that follows, the treatment methods proposed
for the emission streams shown in Figures 9-11, for gasification,
and 9-22, for liquefaction, in conceptual designs and process
analyses are compared with the target pollutant residuals.
Dust (Streams 1)
Cameron Engineers (3, p.22) estimated particulate emissions to be
0.05 Ib/ton of coal for crushing, screening, and conveying, plus
0.025 to 0.04 Ib/ton of coal for storage and reclaiming, based on
information obtained from the Wyoming Coal Gas Company. In
addition to use of water sprays with a wetting agent at dump
hoppers, transfer points, screens and crushers, and use of dust
collectors in the screening plant, Cameron suggested that
screening and coal fines cleaning operations be enclosed and
vented to wet scrubbers or baghouses, that conveyors be covered,
that free fall of coal onto the storage pile be minimized and
that spontaneous combustion in storage piles be prevented by
avoiding segregation of fines and compaction. These or similar
methods are recommended by Exxon Research and Engineering Company
(1), Battelle (2), by Pullman Kellogg in a client study (4). and
by Parsons (5).
•
A critique of these methods is included in Section 10 of this
report.
Nitrogen (Streams 2)
Cameron and all others proposed that waste nitrogen from the
oxygen plant be vented to atmosphere through a 150 to 300 foot
stack to disperse the nitrogen. It is felt that this method is
adequate to avoid dangerous concentrations of nitrogen at ground
554
-------
level, since its lower molecular weight will compensate for the
low exhaust temperature of about 7°F to yield a resultant density
lower than that of ambient air.
Lock Hopper Vent Gas (Stream 3, Gasification)
Cameron examined several alternate schemes, of which the most de-
sirable were (a) use of crude gas to pressurize the hoppers and a
high pressure compressor to recycle most of the gas and (b) use
of crude gas to pressurize with a low pressure recycle compres-
sor. Residual vent gases were to be incinerated to recover heat-
ing value. The flue gases were to be scrubbed if necessary for
S02 control.
Battelle, Exxon,and Kellogg proposed to use crude gas for pres-
surizing with a recycle compressor. Residual gases were to be
vented to the atmosphere.
The venting of residual lock hopper gas to atmosphere will vio-
late environmental standards. The most practical, economical, and
environmentally satisfactory method is displacement of the resi-
dual gases with an inert gas stream, either CO or nitrogen, to
the incinerator before the lock hopper is opened.
Ash Quench Vent (Streams 4)
Cameron Engineers proposed to quench the high pressure ash lock
vent gases in a direct contact water condenser, since these gases
are mainly steam. Quenching of the ash evolves a large quantity
of steam that carries ash and clinkers. Camerson proposed to
separate the solids in a wet cyclone and then condense the steam
with collection of fine ash particles in a direct contact con-
denser.
555
-------
Alternately, Cameron noted that if the ash lock chamber is not
repressurized before the valve to the gasifier is opened, gas
from the gasifier could flow into the hopper and be emitted with
the other noncondensables from the condenser.
Battelle advocated venting the condenser exit gases to atmosphere
through a 100-to 200-foot stack. Kellogg proposed a final water
spray into the vent gases to remove any remaining dust.
Parsons returned the vapors from the process gasifier quench to
the gasifier. No disposal method is described for vapors from
the fuel gasifier quench.
The control method selected after study of these proposals uses
cyclones to collect cinders, a direct contact condenser to
condense steam/and incineration of the noncondensable gases in
order to oxidize any organic materials prior to discharge to the
atmosphere.
Tank Vents and Miscellaneous Leaks (Streams 5 and 6)
The emissions from tank breathing, leaks, spills, and venting of
tanks during filling can be any of the byproducts and chemicals
that must be stored. Cameron estimated the more important
emissions, based on API design, to be about 12.5 Ib/hr,
consisting of crude phenol 12 percent, tar oil 21 percent,
naphtha 17 percent, ammonia 12 percent, product gases 25 percent
and methanol 13 percent.
Cameron's proposed methods for controlling tank ventings included
a refrigerated vent condenser, scrubbing with a low vapor
pressure solvent, incineration,and adsorption. Exxon presented
similar tank vent figures but did not propose control methods.
Battelle and Kellogg did not mention quantities or means of
control.
556
-------
Parsons proposed refrigerated LPG storage with evaporation
recovery.
Evaluation of these proposals showed that the Cameron combination
of methods appeared to offer the best assurance that environmen-
tal standards will be met. It is felt that the possibility of
hydrocarbon losses could be reduced by using floating roof stor-
age tanks equipped with secondary (wiper) seals.
Acid Gas Removal (Streams 7 and 8)
The C02 streams from acid gas removal in gasification are routed
to vent stacks for dispersion into the atmosphere in the Cameron,
Exxon, and Battelle reports. Pullman Kellogg1 s report recommended
routing the C02-rich gases to the utility boiler for incineration
prior to venting to the atmosphere. Parsons advocates
incineration and venting to atmosphere.
Because the C02-rich gases from both the Selexol and Rectisol
units contain substantial quantities of H2S, CO, CH4 , (^ H 4, and
C2Hg , venting directly to the atmosphere is environmentally
unacceptable. Therefore, incineration prior to venting was
selected as being the most practical and acceptable means of
control.
Expansion Gas and Acid Gas (Streams 9 and 10)
Exxon advocates sending the expansion gases from gas-liquor
separation and acid gases from gas-liquor stripping to incinera-
tion with flue gas desulfurization (FGD) prior to venting to the
atmosphere. These gases are not mentioned in the Battelle and
Pullman Kellogg reports. Expansion and acid gases in liquefac-
tion were routed by Parsons to the Claus sulfur recovery plant.
557
-------
Incineration followed by efficient FGD appears to be practical
and adequate for environmental acceptability.
Incinerator/Boiler Flue Gas (Stream 11) and Flue Gas
Desulfurization (Stream 12)
The Exxon and Battelle reports proposed that the combined flue
gases from the incinerator and the superheater be sent to an FGD
system prior to being vented to the atmosphere, with Battelle
recommending a wet limestone scrubber for FGD. Pullman Kellogg
proposed venting direct to the atmosphere, with the precaution
that the boiler feed gases should be thoroughly mixed to avoid
bypassing to zones of lower temperature with consequent
incomplete incineration. Parsons advocated desulfurization of
the fuel gas prior to combustion and then venting the flue gases
to atmosphere.
Study of the compositions of the streams to the incinerator/-
boiler led to the conclusion that either FGD or desulfurization
of the feed gases before incineration would be adequate to meet
most environmental standards, with final choice being made from
economic studies.
Particulates must be removed from the flue gases by cyclones and
electrostatic precipitators. NOX emissions may be controlled by
boiler modifications or by flue gas treatment.
Sulfur Recovery Tail Gas (Stream 13)
Battelle proposed to recover sulfur in a Glaus plant, incinerate
the tail gas, remove S02 in a wet limestone scrubber, and then
vent the gases to atmosphere through 150-to 300-foot stacks.
Exxon proposed to recover sulfur with the Stretford process
incinerate the tail gas in the superheater boiler and treat in an
558
-------
FGD system before venting to atmosphere. Pullman Kellogg
advocated recovery of sulfur in a Stretford plant, incineration
of the offgases in the utility boiler, and venting to atmosphere.
Cameron proposed to recover sulfur in a Stretford unit, incine-
rate and vent the Stretford tail gas,and vent the lean absorber
and oxidizer offgases directly to atmosphere. Parsons advocated
sulfur recovery in a Claus plant with tail gas treatment in a
Beavon plant, with Beavon tail gas vented to atmosphere.
Cameron proposed an alternate scheme in which the rich H-S stream
to sulfur recovery would be treated in a Claus plant and the lean
H2S stream would be treated in a Stretford plant. Tail gases
from both processes would be incinerated and scrubbed before
release. In a second alternate, Cameron proposed to treat only
the tail gas from the Claus plant, then the offgases from tail
gas treatment and tail gas from the Stretford plant would be
combined, incinerated,and vented.
Evaluation of these various proposed processing schemes led to
the conclusions that:
o Present environmental standards for many states would
probably be met by incineration and venting to the
atomosphere.
o Present most stringent environmental standards would only
be met by use of the Beavon, or similar process, or by
tail gas incineration and FGD.
Evaluation of the entire sulfur handling problem led to the
conclusions that:
o Recovery of sulfur in salable form would probably be more
economically attractive than FGD with a throwaway process,
o Claus plant tail gas can be economically treated to meet
most stringent environmental standards in a Beavon plant,
with additional sulfur recovery.
559
-------
o Because of the availability of an I^S-rich gas stream in
the coal conversion plants, incinerator/boiler flue gases
can be economically treated to environmentally acceptable
levels with the citrate process, with additional recovery
of sulfur.
Accordingly, the Claus/Beavon and the citrate processes were
selected for further study in integrated process schemes.
SNG Dryer Vent (Stream 14)
Exxon proposed to cool the gas leaving methanation, separate the
condensate, remove water and CO from the gas stream in the acid
gas removal system,and then compress the dried gas to pipeline
pressure. Cameron proposed to remove water from the SNG via
glycol dehydration and then recover liquid water from the over-
head of the glycol regenerator column. Parsons advocates glycol
dehydration but vents the water vapor to atmosphere.
The SNG dryer vent gas consists of water vapor with about 8 mole
percent methane and, because of the small quantity involved, can
be vented through stacks without creating environmental problems.
Cooling Tower Evaporation and Drift (Stream 15)
The reports consider only the water content of the evaporation
and drift from the cooling tower system, except where Cameron
notes traces of ammonia and non-methane hydrocarbons.
Since the cooling tower in these reports is a receiver of blow-
downs and wastewaters that may contain volatile contaminants in
appreciable quantities, and since the cooling tower water may
contain substantial quantities of nonvolatile materials that may
560
-------
be noxious or hazardous or cause nuisances, it is concluded that
possible problems may have been dismissed too lightly. According-
ly, in studies of integrated processing schemes for gaseous
emissions the means of reducing cooling tower drift were investi-
gated and for liquid effluents the means of rendering cooling
water constituents nonvolatile and nontoxic were developed.
References
1. Shaw, H., and Magee, E., "Evaluation of Pollution Control in
Fossil Fuel Conversion Processes; Gasification; Section 1.
Lurgi Process." EPA 650-2-74-009. 35*
2. Battelle, "Detailed Environmental Analysis Concerning a Pro-
posed Coal Gasification Plant for Transwestern Coal
Gasification Co. Pacific." Feb 1, 1973. 476*
3. Sinor, J. , "Evaluation of Background Data Relating to New
Source Performance Standards for Lurgi Gasification." June,
1977. 552»
4. Pullman Kellogg, "Engineering Evaluation of a Process to Pro-
duce 250 Billion Btu/Day of Pipeline Quality Gas From
Illinois and Wyoming Coal for Panhandle Eastern Pipeline
Company and Peabody Coal Company." June 20, 1972.
5. O'Hara, J. B., and Hervey, G. H., et al.f "Oil/Gas Complex Concep-
tual Design/Economic Analysis: Oil and SNG Production." R
and D Report No. 114, Interim Report No. 4. March 1977.
814«
•Pullman Kellogg Reference File number
561
-------
COMMERCIAL EMISSION CONTROL METHODS
By definition, commercial emission control methods are those pro-
cesses and equipment that have been operated in full scale appli-
cations. The processes are offered by licensors who are, in most
cases, the developers and operators of the processes. The equip-
ment is offered by vendors who are able to demonstrate commercial
operations.
The emission control methods that are considered to be most im-
portant for application to coal conversion processes, and that
have been investigated, are:
o Processes and techniques for control of nitrogen oxides (NO )
x
o Processes and techniques for control of sulfur dioxide
o Processes for control of hydrogen sulfide
o Techniques for control of particulates
o Control of cooling tower drift
o Other control techniques applied to hydrocarbons, lock hop-
per vent gas, ammonia, ash quench vent gas, and miscellaneous
leaks.
The emission control methods are presented in a general format
that includes discussion of:
o Process description
o Capability, efficiency,and limitations
o Case histories
o Wastes produced
o Cost data (As found in the literature. Later in this
section of the report these costs will be updated to the
same year and on the same basis for comparison purposes.)
o Possible problems
o Possible improvements
o References
562
-------
Processes and Techniques for Control of Nitrogen Oxides
Nitrogen oxides (NO ) are formed during the combustion of coal,
JC
oil, and gas with air. The NO is comprised of about 95 percent
,/t
nitric oxide, NO, and about 5 percent nitrogen dioxide, N02
(l,p.20).* Nitric oxide is colorless while nitrogen dioxide has
a reddish-brown color. Both are toxic and both undergo photo-
chemical reactions with hydrocarbons in the atmosphere to produce
highly active free radicals which create aerosols and toxicants
that typify photochemical smog.
NO is formed by two mechanisms:
j£
o high temperature fixation of oxygen and nitrogen present in
the combustion air (thermal NOX)
o reaction of oxygen with nitrogen contained in the fuel
(fuel NOV)
J\L
In either case, NO is the primary product formed because the
residence time in most stationary combustion units is too short
for a significant amount of NO to be oxidized to N02.
Formation of thermal NO is primarily dependent on temperature,
Jt
being greater at higher temperatures. Its formation is also
affected by oxygen availability. Formation of fuel NOV , on the
X
other hand, is relatively independent of combustion temperature
but dependent on the amount of nitrogen in the fuel and the
availability of free oxygen outside the combustion zone. NOV
•Other oxides of nitrogen formed to a minor degree are N03, N20,
N203, N204, and N205 (2,p.7)
563
-------
will be formed by both mechanisms during the combustion of fossil
fuels although it is difficult to predict the exact contribution
of each. Less than 30 percent of the nitrogen contained in coal
is normally converted to NO (2, p. 9); however, about 40 to 70
Jt
percent of that contained in oil will be converted to
As shown in TABLE 9-2 and Figure 9-23, the percentage of liquid
fuel nitrogen converted to NO is a function of the nitrogen
J\
content of the fuel. Fuels with a relatively high nitrogen
content, 0.4 to 1.0 percent, will exhibit a lower percentage
conversion to NOV than those with the relatively low nitrogen
«C
content of 0.1 to 0.2 percent.
Some NO undergoes decomposition via the following reaction in the
hot zones of the boiler:
2MO(g)-» N2(g)+02(g)
However, once rapid cooling of the gases begins, no further
decomposition occurs. NO is continuously oxidized to NC^ as
follows:
N0(g)+ 1/2 02(g)-^N02(g)
This reaction is slow and the time available is too short for
equilibrium to be reached. Thus, as mentioned above, most of the
NOX leaving the boiler will be in the form of NO.
Currently, the most stringent standard for NC^ emissions from
fuel combustion is that of New Mexico:
NO^. Ib/MM Btu NO^. ppm (v)
Coal 0.45 338
Oil 0.30 225
Gas 0.20 150
564
-------
TABLE 9-2. CONVERSION OF FUEL NITROGEN TO N0y*
Nitrogen Conversion
Fuel Type Content, Vt.% to NOX, %
#2 Oil 0.006-0.02 100»»
#5 Oil 0.10 56-60
#5 Oil 0.13 70
#5 Oil 0.20 41
#5 Oil 0.28 44
#6 Oil 0.27 52
#6 Oil 0.44 43-51
#6 Oil 1.50 45
» From Item 3, p. 96 in reference list
••Fuel nitrogen content too low to determine realistic values.
Near 100$.
565
-------
.Ul
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
I
0.0 0.1 0.2 0.3 0.4 0.5
FUEL NITROGEN (%)
0.6
0.7
Figure 9-23. Fuel nitrogen conversion as a function of fuel nitrogen content.*
*From Item 5, p.62 in reference list
-------
However, projected EPA standards will be considerably lower, as
shown by the following table (4,p.36):
NO , ppm (v)
1980 1985
Coal 200 100
Oil 150 90
Gas 100 50
Control of NOX by Boiler Modifications—
The use of boiler modifications to lower NC^ emissions will
probably receive the initial emphasis since these methods are the
least costly. Design changes which lower the flame temperature
and reduce oxygen availability result in lower NOV formation.
X
These modifications can be one or more of the following:
- Two-stage combustion
- Low excess air firing
- Flue gas recirculation
Several sources have somewhat differing opinions on the effects
of these techniques, as shown in TABLE 9-3.
Therefore, it is concluded that boiler modification can result in
a reduction in NOX emissions of between MO and 50 percent.
The following table (1, p.21) (3, p.95) gives the expected
emissions of NOX for both uncontrolled and controlled boiler
conditions.
567
-------
TABLE 9-3. NO.. REDUCTION WITH BOILER MODIFICATIONS
Staged Combustion
Approximate Average
Low Excess Air
Approximate Average
Combined Staged Combustion
and Low Excess Air
Approximate Average
Flue Gas Recirculation
Ref.»
3
4
6
3
4
6
3
4
6
3
4
6
NO^ Reduction
"Soal
34
40
H
22
20
25
20"
38
40
60
Iff
15
NR
33
Oil
16
40
40
35
20
20
i
30
35
1*0"
17
20
33
i %
Gas
25
55
50
24
20
I
42
50
90
SO"
40
60
W
Approximate Average
•Numbers refer to items in the reference list
Item 3, p.95: ranges in the article have been averaged for this
table.
Item T, p.35: source for the article was EPA report
650/2-74-066.
Item 6, p.III-55.
568
-------
Expected NO Emissions
Coal Oil Gas
Uncontrolled Combustion, ppm 500 280 200
With Boiler Controls, ppm 370 150-210 85-110
NOX Reduction with Boiler
Controls, % 26 25-46 45-57
Therefore, it may be concluded that boiler modifications are
sufficient to meet the most stringent present standards for NOX
emissions (perhaps with coal being a borderline case) . These
modifications also appear to meet the projected 1980 goals for
oil and gas, but not for coal. However, none of the projected
1985 goals can be met solely by boiler modifications.
Control of NOX by Reducing Liquid Fuel Nitrogen Content—
As previously mentioned, reducing the fuel nitrogen content will
result in lower NO emissions. This can be accomplished by
J^
hydrodenitrogenation of the coal-derived liquids such as tars
and tar oils. These liquids are expected to contain about the
same percentage of nitrogen as the feed coal (7,p.124). The
hydrodenitrogenation scheme is shown in Figure 9-24 (8,p.5).
Hydrodenitrogenation is accomplished by hydrogenolysis of
organonitrogen compounds to ammonia and the corresponding
hydrocarbon as shown below (8,p.3):
CwHxNy+l/2(z+3y-x)H2 -*yNH3+CwHz
Petroleum oils are usually hydrotreated at a temperature in the
range of 600-800°F, a pressure in the range of 300-4,000 psig and
a hydrogen to oil ratio in the range of 300 to 15,000 SCF per
barrel. Studies by ARCO (Atlantic Richfield Company) and Wan on
COED oil showed that the degree of nitrogen removal is a function
of hydrogen partial pressure, temperature, space velocity/and
569
-------
HYDROGEN
.n OTT.
\
FUHNACE
REACTOR
^
Y
!
X^^N
HIGH PRESSURE
FLASH
LIGHT I.WDS
PKODUCT OIL
ET;:IPPL-R
COLUMN
Figure 9-24. Typical hydrotreater unit.*
*From Item 8, p.5 in reference list
-------
catalyst particle size. The removal is favored by higher
pressures and temperatures and by lower space velocities and
catalyst particle sizes. The effects of space velocity and
pressure on denitrogenation are shown in Figures 9-25 and 9-26.
A slight improvement in nitrogen removal can be effected by
increasing the hydrogen recirculation ratio. Catalyst pore
diameter appears to have no effect on denitrogenation. These
studies have shown that up to about 80 percent nitrogen removal
can be attained at pressures of 2,0 00 to 3,0 00 psig and
temperatures of 750 to 800°F (8,p.14-21).
Similar studies by Satchell have shown that about 75 percent
nitrogen removal can be attained at a pressure of 1,000 psig and a
temperature of 750°F. Presumably over 80 percent removal could
be achieved at higher pressure and/or for temperature (8, p.140).
Figure 9-27 presents typical data from the investigations.
An example will serve to illustrate the benefits which accrue
when liquid fuel denitrogenation is employed. An oil fired
boiler with 2.7 percent oxygen in the flue gas (corresponding to
about 15 percent excess air) which burns fuel containing 0.4
percent nitrogen will have about 280 ppm NO in the flue gas.
Jt
Assuming 45 percent conversion of the fuel nitrogen to NO , the
Ji
relative contributions of the different NOV mechanisms are:
Jt
ppm (v)
NOX (Fuel at 45 percent Conversion) 207
NOX (Thermal) 73
Total 280
If 80 percent of the fuel nitrogen is removed by the techniques
discussed, the residual nitrogen content will be 0.08 percent;
571
-------
100
en
§ 50
2
U
§
e-i
H
a
6-"
§20
u
o.
10
0.0
TEMPERATURE: 800 F
PRESSURE:
500 PSIG O
1000 PSIG O
2000 PSIG D
_L
100
0.5 1.0
SPACE TIME, HOURS
1.5
2.0
TEMPERATURE: 725~F
PRESSURE: 3000 PSIG
10,000 SCF H,/BBL
DATA SOURCE: (23)
Figure 9-25. Effect of space time on
denitrogenation of Anthracene
Oil.*
*From Item 8 in reference list.
Figure 9-26.
Effect of weight
hourly space time on
denitrogenation of
COED oil.*
-------
1/3
O6
TEMPERATURE, «F
6OO O
6SO O
TOO A
750 ?
PRESSURE HXX5 PSIG
CATALYST SIZE: 8-iOMr£H
CATALYST NALCOMO ^74
CATALYST 030 DEPTH 20 INCHES
CATALYST LOADING NUMBERS
HYDROGEN flAT^ 1300 SCF/BBL
J73 75
SPACE TIME, HOURS
I.S
Figure 9-27.
Effect of space time and temperature
on denitrogenation.*
*From Item 8 in reference list
573
-------
however, about 70 percent of the residual will be converted to
NOV. The resulting NOV emissions become:
X X
ppm (v)
NOX (Fuel at 70 percent Conversion) 65
NOX (Thermal) 73
Total 138
If the original fuel contains twice as much nitrogen (0.8
percent) it is clear that a 90 percent denitrogenation efficiency
is needed to achieve the same results.
These calculations are in good agreement with the data presented
in Figure 9-28. In this example the NOX emissions from an oil
fired boiler were reduced about 50 percent, sufficient to meet
the present most stringent standard and the projected 1980 goal.
However, fuel denitrogenation alone will not be sufficient to
meet the projected 1985 goal of 90 ppm NO for oil fired units.
a
For this goal a combination of boiler modifications, to reduce
N0x emissions by 30 to 40 percent, and fuel denitrogenation might
suffice:
N03[. ppm (v)
0.7 x 138 = 97
0.6 x 138 = 83
Control of NOV (and S00 ) by Flue Gas Denitrification Processes--
X £,
Currently, there are some 48 flue gas denitrification processes,
both dry type and wet type, reported in the literature. A signi-
ficant amount of information is reported on 42 of them.
For a variety of reasons, including simplicity, more favorable
economics, and the fact that most existing power plants in Japan
operate with low sulfur oil as fuel with small amounts of
574
-------
400
350 —
300 —
250 —
200 —
150
100 —
125 MW
POWER
PLANT
(OIL FIRED)
0.1
0.2
FUEL NITROGEN, %
0.3
0.4
Figure 9-28. Effect of fuel nitrogen on
NO emissions.*
Jx
*From Item 2, p. 13 in reference list
575
-------
particulate and SO, flue gas emissions, the dry NO -only removal
£, X
processes seemed the most promising and were developed first.
The initial research for most of the wet processes was begun
later and, hence, the dry processes as a group have been more
extensively tested and are more commercially acceptable.
Although there are many different types of dry and wet processes,
in most cases the dry processes have the following advantages
over the wet processes:
o Lower projected total capital investment and lower annual
revenue requirements
o Simpler process with few equipment requirements
o Higher NO removal efficiency (over 90 percent)
Ji
o More extensive tests in large oil and/or gas fired boilers
o No waste stream generation
However, dry systems also have the following disadvantages:
o More sensitive to inlet particulate levels
o Requirement for ammonia from either an energy-sensitive
source (natural gas) or more expensive coal gasification
methods
o Possible emission of ammonia, ammonium sulfate, and ammonium
bisulfate. Precipitation of the latter two may foul down-
stream equipment
o Relatively higher reaction temperatures are required
(350-400°C), which means that the system must be located in
the power generation cycle before the air preheater or the
temperature must be attained by auxiliary heating after the
preheater
The most critical of these disadvantages, particularly for the
U.S. utility industry with its heavy reliance on coal for power
576
-------
generation, is the sensitivity of-these processes to inlet
particulate levels. However, major research is now underway to
develop methods so that dry systems can handle flue gas with high
particulate loading.
Another disadvantage of the dry, selective catalytic reduction
(SCR) processes is that the ideal catalyst location may be in the
region between the economizer outlet and the air preheater inlet
and, hence, the process is intimately involved in the power
generation cycle. Therefore, if problems of operating these
processes occur, the adverse impacts on the basic utility
operations may be greater.
In addition to the above-mentioned disadvantages, the long-term
supply of ammonia for these dry NO removal processes is a
potential problem. Ammonia is the reducing agent for converting
NO to molecular nitrogen for the SCR processes, which comprise
X
nearly all of the dry processes and about half of all the NO
X
removal processes, and the selective non-catalytic reduction
process. With an NH3:NOX raol ratio of about 1:1, a single 500 MW
coal-fired power plant (600 ppm NOX in the flue gas) would
consume about 5950 tons per year of liquid anhydrous ammonia. In
view of the continuing increase in the world's demand for ammonia
and ammonia-based fertilizers, the availability of ammonia for
larger numbers of these dry NO removal units warrants concern
and further investigation.
The wet N0x removal processes have certain general advantages and
disadvantages as compared with the dry systems. The major
advantages include:
o Simultaneous S02 and NOX removal may be a potential
economic advantage
o Relative insensitivity to flue gas particulates
o Higher S02 removal (over 95 percent)
577
-------
On the other hand, the major disadvantages of these wet systems
include:
o More expensive processes due to the low solubility of NOV
Jv
in aqueous solutions and due to more extensive equipment
requirements
o Formation of nitrates and other potential water pollutants
o Formation of low-demand byproducts
o Flue gas reheat required (however, if a wet S02 removal
system were used in series with a wet removal system for
NO only, the reheat would already have been incorporated
Jt
into the design)
o Only moderate NO removal
J\L
o Application of some processes may be limited to flue gas
with high S02:NOX ratio
The two primary disadvantages of the wet systems are the high
capital and operating costs and the formation of nitrate-
containing wastewater. The formation of nitrate salts in most of
these processes requires removal of these salts from the effluent
by either evaporation or biological treatment.
Another inportant factor which must be considered in the
application of certain wet NOX removal processes is the minimum
SO,:NO ratios required. It may not be feasible to operate wet
£ X
NO removal processes using flue gas from Western U.S. coals,
a *
which characteristically possess low amounts of sulfur, since
sufficient S02:NOX ratios may not be achieved for adequate NO
removal (2,pp. xvi-xviii).
Of the 42 processes screened, the following 8 were selected for
further study (2,p. xxviii):
578
-------
Type of Process (Classification)
UOP Shell Copper Oxide
Dry Simultaneous S02-NOX
(Selective catalytic reduction)
UOP Shell Copper Oxide
Dry NOV only
Jt
(Selective catalytic reduction)
Hitachi Zosen
Dry NOV only
Ji
(Selective catalytic reduction)
Kurabo Knorca
Dry NO only
X
(Selective catalytic reduction)
Moretana Calcium
Wet simultaneous S02~NOX
(Oxidation-absorption-reduction)
Ishikawajima-Harima
Heavy Industries
Wet simultaneous S02~NOX
(Oxidation-absorption-reduction)
Asahi Chemical
Wet simultaneous S02-NOX
(Absorption-reduction)
MON Alkali Permanganate
Wet N02 only
(Absorption-oxidation)
In a 1977 study conducted by Pullman Kellogg for a client, the
overall conclusion was reached that dry NOV removal processes are
a
definitely superior to wet processes. For this report the same
viewpoint is adopted. Two of these dry processes will be
described as illustrative examples of the type.
579
-------
UOP Shell Process — The Shell copper oxide process that is
licensed by Universal Oil Products (UOP) may be used to remove
only NOV from flue gas or to remove both NO and SO .
A A ^
For removal of NO only, as shown in the block flow diagram,
Figure 9-29, the flue gas leaves the boiler economizer, ammonia
is injected and the mixed gas stream enters the reactor. The
fixed bed in the reactor is composed of copper supported on
special aluminum oxide. The copper first oxidizes to CuO and
then reacts with the SO in the flue gas:
S02(g) + 1/2 °2(g) + CuO(s) CuSOi|(s)
The copper sulfate acts as a catalyst for the reaction of NO
with ammonia and the consequent reduction of the NO to nitrogen
•A.
and water:
6NO(g) + 4NH3(g)— »5N2(g) + 6H20(g)
6N02(g) f 8NH3(g)— »7N2(g) + 12H20(g)
The treated flue gas leaves the reactor and flows through the air
heater, particulate removal and flue gas desulfurization equip-
ment and is finally exhausted to atmosphere. The copper in the
reactor bed becomes fully converted to copper sulfate to act as
the catalyst for NO removal. In this form the catalyst has no
A
effect on the S02 in the flue gas and the SOp passes through the
reactor unchanged.
The UOP/Shell Process can be designed for simultaneous NO -SO.-,
A C.
removal by adding a regeneration step, wherein part of the copper
sulfate is converted to CuO by steam-diluted hydrogen rich gas
and returned to the reactor bed while the rich SO stream is sent
to sulfur recovery or other disposal.
580-
-------
U1
CO
NH5
AIR
PARTICIPATE REMOVAL,
FGO, AND /OR
STACK
Figure 9-29. UOP/Shell NO removal process.
(From Item 2 in reference list)
-------
NOX removal of at least 90 percent is expected with beds 4 meters
long. Removal efficiencies of 95 to 97 percent with beds 6
meters long and 99 percent with beds 7 meters long have been
achieved during prototype-scale testing.
All the reactions are exothermic. By locating the flue gas
treatment section upstream of the air heater, the heats of
reaction may be recovered together with the fan compression heat.
The Shell International Petroleum Company began development work
in the early 1960's and in 1967 built a unit of 0.2 to 0.3
megawatt (MW) equivalent capacity for simultaneous NOX-S02
removal at Pernis, The Netherlands to process flue gas from
combustion of high sulfur fuel oil. Successful operation for 4
years and more than 20,000 regeneration cycles proved the
feasibility of the process. In 1973 an NQX-S02 unit of HO MW
equivalent capacity was started at Showa Yokkaichi Sekiyu (SYS),
Japan on a flue gas from an industrial boiler containing 2,500 pp
S02 and has been operating successfully. A pilot plant of 0.6 MW
equivalent capacity has been operating successfully for 3 years
at the Tampa (Florida) Electric Company (TECO) on flue gases from
combustion of coal containing 3.5 percent sulfur.
The UOP Process Division holds the worldwide (except for the Far
East) licensing rights for the Shell process.
Operating parameters, capital investment and operating
requirements reported (2) for a 500 MW installation are shown in
TABLE 9-4.
The NO removal efficiency is reported to be independent of the
X
582
-------
TABLE 9-4. UOP/SHELL N0x CONTROL SYSTEM TREATING
FLUE GAS FROM GENERATION OF 500 MEGAWATTS
Flue gas at 400°C, Ncu.m/hr 1,582,000
SO2 inlet, ppm 2,580
NOV inlet, ppm 634
Jv
NO outlet, ppm 63
Ji
NO removal 90$
Ji
Reactors (4) 4m bed length, 8.7m diameter
Catalyst, kg/reactor 51,480
Pressure drop across system, cm water 23
Capital investment (1977 Midwest U.S.) $15,500,000
Operating Requirements
Ammonia, STPD 21
Power, KWH/day 77,800
Steam, STPD 13.2
Labor, Manhours/day 5
Maintenance, per year $194,000
Heat credit, MBtu/hr 26.5
Operating Cost. mills/KWH 1.4
583
-------
NO concentration in the inlet gas. The NO reaction is first
x x
order and the reaction rate is directly proportional to the NOV
Jv
concentration. Removal rate increases as the reactor bed length
increases.
NO removal data from tests at pilot plants and the prototype
scale unit (40 MW equivalent) at SYS are shown in Figure 9-30.
The acceptor used was completely in the sulfated form. The
results are plotted as a function of bed length, since conversion
fluctuated with time on stream, design of reactor internals and
operating conditions. NOX removal efficiencies of 95 to 97
percent with beds 6 meters long and 99 percent for beds 7 meters
long were obtained during tests.
The parallel-passage reactor has been shown capable of operating
at normal conditions with full particulate loading of 10 grains
or more per standard cubic foot. Initial testing with full
flyash loading to the reactor at the Pernis unit showed no
deterioration in performance. However, at SYS the high V- and
Na-containing flyash gradually fouled the reactor internals
forcing runs to be limited to operating periods of 1 to 2 months.
The effect of the fouling on acceptor performance is shown in
Figure 9-31.
The higher particulate loading with coal fired flue gas at TECO
of 10 grains or more per standard cubic foot caused a decline in
performance after only days of operation. As a result of this
problem, a procedure was developed and implemented at this pilot
plant which provided in situ cleaning of reactor internals during
normal operation. This technique allowed stable performance with
high particulate loadings. Pressure drop across the reactor and
desulfurization were not affected.
Testing at TECO also demonstrated that chlorine and chlorides in
584
-------
100
10
^
\\
^ X
\ V
\
V
V
\
x \
^ ' \
^ \
•iN rv
; •*• v >L v
i ^ Q ^
v*\ ^
^J V
i
'. t
t
CONDITIONS:
... /inn r
1
• •"• 1 UU v_
Cu AS cuSO.
NH /NO 1.1 1.5
O^ NORI-1AL EXPECTED
PERFORMANCE
• PERFORMANCE
AFFECTED BY
rNTimo-rn.!? T7 A rvnr\r>c
OB i
\ \
\ s
\
\
>
\ u
\
\
\
\
\ C
V
>
\
\
k
V
X
\
\
\
i
\
\
\
\
\
\
\
N
\
\
\
\
\
\
L
I
v
\
\
\
\
H
J
o
2
234
BED LENGTH, METERS
Figure 9-30. Unconverted NO as a function of catalyst
bed length forxUOP process.*
*Frora Item 2 in reference list
585
-------
450
400 -
- 300
ui
o
o
u
tiJ
ff
o
u
UJ
K
FLOW * 137,000 Nm'/h
SOt < I260ppmv
NOX * 293 ppmv
REACTOR BED LENGTH • 4 METER
NOX AT NH, /NO-0.0
200
40 60 80 100
ACCEPTANCE TIME, MlN
Figure 9-31. Performance of Shell reactor
at SYS: instantaneous S02
and NO slip.*
*From Item Sin reference listx
586
-------
flue gas had no adverse effect upon the acceptor performance.
The loss of copper was negligible.
The use of an alumina base for the copper sulfate catalyst is
questionable since competing processes appear to be changing to
alternate supports that are supposedly more stable. It is
recognized that UOP refers to their catalyst as being on a
"special" alumina support.
The UOP/Shell process has these advantages:
o Achieves NOX removal efficiencies of 90 percent or more
o Has been applied to flue gas from commercial oil-fired
boilers
o Is a slight modification of a commercially available flue
gas desulfurization system
o Operates with full particulate loading of 7 to 10 or more
grains per standard cubic foot. (Note: After cyclones,
loading should be 1.5 to 2 grains.)
o Claims less than 10 ppm by volume of ammonia in the treated
flue gas. Pilot and prototype models have shown an average
of 1 ppm in the flue gas
o Claims full turndown capability
o Requires no waste disposal for the system since no by-
products are produced
o Although UOP considers information on materials of
construction to be proprietary, statements are made that no
unusual materials are used and that a service life of at
least 15 years can be expected.
The disadvantage that must be cited is the lack of test data on
flue gas from large scale coal fired units.
Hitachi Zosen Process—Hitachi Zosen (Hitachi Shipbuilding and
587
-------
Engineering Compay, Ltd.) has developed an N(^ removal process in
which dry, selective catalytic reaction of NOX with ammonia
occurs. Hitachi Zosen is now developing a catalyst and reactor
design which permits treatment of flue gas with a high
particulate loading. Therefore, the flue gas from a coal fired
boiler may be fed directly to the reactor, upstream of the air
heater, without any particulate removal treatment.
Ammonia is injected into the flue gas ahead of the reactor. In
the reactor NO and NC^ are reduced to nitrogen and water by
reaction with ammonia in the presence of a catalyst. Any excess
ammonia is oxidized to nitrogen and water; therefore, Hitachi
Zosen reports that there is no problem of excess ammonia in the
flue gas leaving the system. The treated flue gas passes to the
air heater for heating the air feeding the boiler. After the
treated flue gas is cooled in the heat exchanger, the gas is sent
to the particulate removal and desulfurization steps, and then
leaves through the stack.
About 0.8 to 1.2 moles of ammonia are fed per mole of NOX . The
reaction temperature is 300 to 400°C. The area velocity (flow
rate of gas per unit surface area of catalyst) is between 7 and
10 N cu.m/hr/sq.m. Catalyst surface area ranges from 550 to 600
sq.m/cu.m. NO removal efficiency is reported to be above 90..
J\,
percent. The process is able to function with inlet particulate
loadings up to 16 gr/N cu.m (about 7 grains per standard cubic
foot). Hitachi Zosen claims that the pressure drop across the
reactor is very low.
The catalyst is manufactured in the shape of units of honeycombs.
These units are welded together for the particular size required.
The flue gas passes parallel to the catalyst surface. The
catalyst composition has not been revealed for proprietary
reasons; however, Hitachi Zosen does state that it is constructed
588
-------
of common material. The expected catalyst life is reported to be
one year.
Hitachi Zosen has been developing an NOY removal process and
A
catalyst since 1970. Five different catalyst series have been
created and are known as NOXNON 100, 200, 300,400, and 500
series. The 100 series is nonselective and for use with CO,
hydrogen, and hydrocarbons as reductants. The other four series
are for use with ammonia as the reducing agent. Series 200 is
for treating "clean" flue gas, that is, gas which does not
contain any significant amount of SOX or particulates. The 300
and 400 series are resistant to SOV. The 500 series is for use
ji
in gas with a considerable particulate loading. There have been
over 21 different pilot plant tests since 1973 in developing
these catalysts. The 500 series is still being tested and the
results to date have been promising, even on gas containing 15
gr/Ncu.m(6.44 grains per standard cubic foot) of particulates and
300 ppm SO . The following pilot plant and bench scale units are
now testing the NOXNON 500 series catalyst:
Source of flue gas Test unit
(plant type) Fuel capacity. Nm3/hr
Iron ore sintering Heavy oil 5,000
Iron ore sintering Heavy oil 200
Power Coal 200
Glass melting furnace Heavy oil 200
A commercial scale NO removal facility was constructed by
X
Hitachi Zosen in the fall of 1975 for a petroleum plant with an
oil fired boiler and flue gas treatment capacity of 440,000
Nm3/hr. Another facility was completed in the fall of 1975 for a
petroleum opeation with 350,000 Nm3 /hr flue gas capacity. Two
more denitrification systems for steel manufacturing plants with
589
-------
71,000 and 750,000 Nm /hr flue gas rates have been completed and
are reported to be operating successfully.
The estimated capital cost for a denitrification unit to treat
flue gas containing 300 ppm NOX on a 250 MW plant is reported to
be about $16/KW. The corresponding operating cost would be
approximately 1.5 mills/KWH. These are assumed to be 1976 costs
at a Japanese site.
The only major raw material required is ammonia. For a 250 MW
plant with 300 ppm N0x in the flue gas, an estimated 5 short
tons/day of ammonia is required for NOX removal. On the same
basis, the energy consumption for denitrification is estimated to
be:
Electricity 400 KW/day
Steam 4.1 Tons/day
No reheat fuel is necessary. The electrical requirement
represents 0.2 percent of the total power output of the plant.
The majority of the steam (2.9 tons/day) is used for ammonia
vaporization, while the remaining 1.2 tons/day is for soot
blowing.
With no particulate scrubbing before the NOX removal system, soot
blowing of the air heater and the reactor are required. For
pilot plants, the reactor has required washing out of deposits
once every 2 months. Hitachi Zosen states that this operation
requires only a short time to complete. The technical support
necessary for operation of this NOX removal system should be
similar to other dry systems.
There is no information available at present on the sensitivity
of the NOX removal efficiency to factors such as inlet gas
composition and various operating conditions. Specifically
.590
-------
there is no comment on any possible adverse effects from chloride
in the flue gas from a coal fired boiler. These effects might
include interference of NOV removal efficiency or deterioration
Jt
of catalyst.
The composition of the catalyst is not known. The exact cost is
not known, though the estimated cost for the initial investment
in catalyst necessary for a 250 MW plant with flue gas containing
300 ppm NOX is given by Hitachi Zosen as $1.7MM (1976 cost and
assuming 300 yen = $1).
There are few available facts on wash frequency and other washing
requirements on the reactor treating the flue gas directly from a
coal fired boiler with no particulate removal. More information
should be accessible as the pilot plant tests progress.
Maintenance, operating and technical support requirements have
not been published. The major material of construction is
reported to be carbon steel.
There are no pollutants removed, other than NOX , by the Hitachi
Zosen process. There is no byproduct from the process.
Excess NH3 at the outlet of the system is a potential problem;
however, Hitachi Zosen claims NH3 leaving the system is very low
and creates no pollution problem. Some type of disposal for the
reactor deposit washing solution would be essential. Also, the
catalyst is disposed of and not reclaimed.
The Hitachi Zosen process has these advantages:
o Achieves NO removal efficiencies of 90 percent or more
H
o Has been applied to flue gas from commercial oil fired
boilers
591
-------
o Operates with full particulate loadings of 7 or more grains
per standard cubic foot
o Probably contains less than 10 ppm by volume of ammonia in
the treated flue gas
o Claims full turndown capability
o Requires no waste disposal for the system since no by-
products are produced
o Claims carbon steel to be the major material of construc-
tion
The disadvantages are principally technical and reflect data gaps
regarding effects of inlet gas composition and variations in
operating conditions.
References--
1. Do, N. Loan, and Hunter, W.D., "NOX Control Technology," Pullman
Kellogg Report No. RD-77-1342, September 1977 (Confidential).
2. Faucett, H.L., Maxwell, J.D., and Burnett, T.A., "Technical As-
sessment of NO Removal Processes for Utility Applications."
Ji
November 1977.
3. Siddiqi, Aziz, Tenini, John W., and Killion, Larry D., "Control
NOX Emissions from Fixed Fireboxes." Hydrocarbon Processing,
October 1976. 578«
4. Ricci, Larry J., "EPA Sets Its Sights On Nixing CPI's NO
Emissions." Chemical Engineering, February 14, 1977.
•Pullman Kellogg Reference File number
592
-------
5. Air Pollution and Its Control, A'IChE Symposium, Series 126,
Volume 68, 1972. 902*
6. Coal Conversion Program, Energy Supply and Environmental
Coordination Act (as amended), Section 2, Volume 1. 847*
7. Hoffert, F.D., Sonng, W.Y., and Stover, S.E., "Summary of Gas
Steam Control Technology for Major Pollutants in Raw
Industrial Fuel Gas." October, 1977. 899*
8. Satchell, Don P., "Development of a Process for Producing an
Ashless Low-Sulfur Fuel From Coal." Volume IV - Product
Studies - Part 6 - "Hydrodenitrogenation of a Coal Derived
Liquid." 232*
593
-------
Processes and Techniques for Control of Sulfur Dioxide
When sulfur is a component of fuels fed to combustion equipment,
its oxidation leads to the formation of sulfur dioxide (SO ) and
sulfur trioxide (S0_). Normally about 98 percent of the sulfur
leaving the equipment will be SO .
In coal, sulfur is present in three forms:
o pyritic sulfur, (FeS2)
o organic sulfur, where the sulfur is chemically bound in the
coal molecules
o sulfate sulfur, FeS04 or CaS04 , normally less than 0.1
percent
In liquid fuels, sulfur is normally present in organic form.
Tars and tar oils produced in low temperature gasification
processes (e.g. Lurgi) normally contain appreciable amounts, of
sulfur although the total sulfur content will amount to only 50
to 80 percent of that of the feed coal (1, pp. 122-123).
Gaseous fuels such as natural gas contain very little sulfur.
However, miscellaneous waste gas streams produced in coal
conversion facilities may contain appreciable amounts of sulfur
as H S, COS, CS , and sulfur vapor.
M ^
Standards for sulfur dioxide emissions have become very stringent
in the last few years. Projected future standards indicate that
allowable maximum emissions of S02 will be even lower. The most
stringent current standards for S02 from fuel burning equipment
are the following:
•594
-------
Lbs SOp/MM Btu
>250 MM Btu/hr <250 MM Btu/hr
Coal Burning 0.20 (Wyoming) 1.20 (Oklahoma)
Oil Burning
Residual Oil 0.34 (N.Mexico) 1.00 (Illinois)
Distillate Oil 0.30 (Illinois) 0.30 (Illinois)
Gas Burning 0.13 (Montana) 0.13 (Montana)
Perhaps a few examples will help to put these standards into
perspective. The lowest sulfur coals in the United States are
generally found in the western states. The sulfur content of
these coals is about 0.5 percent (minimum) and the higher heating
value is about 10,020 Btu/lb. Burning this fuel without control
will result in an SC^ emission of about 1.0 Ib/MM Btu which is
higher by a factor of 5 than the most stringent standard for fuel
burning equipment with a heat input greater than 250 MM Btu per
hour. Therefore, it may be concluded that some type of sulfur
control will be needed whenever coal is burned in these fairly
high capacity ranges.
A typical tar from a gasification facility has a higher heating
value of about 16,500 Btu per pound (1, p.121). Its sulfur
content will be on the order of 60 percent of that of the feed
coal. In order to meet a standard of 0.34 Ib S02/MM Btu, the tar
or tar oil can contain only 0.28 percent sulfur, which corres-
ponds to a coal containing about 0.4? percent sulfur. Therefore,
for this standard, tars and oils from only the lowest sulfur
coals could be burned without controls. However, when meeting a
standard of 1.0 Ib S02/MM Btu, the tar could contain 0.82 percent
sulfur, which corresponds to a coal containing about 1.37 percent
sulfur. Thus quite a few coals would produce tars which could
qualify for combustion without controls.
It is difficult to generalize regarding combustion of gas streams
because of the wide range of heating values and sulfur contents
of these fuels. A typical low Btu gas may have a higher heating
595
-------
heating value of about 200 Btu per standard cubic foot (SCF) (1,
p.24) (2, p.315) or about 3,030 Btu/lb (molecular weight = 25).
To meet a standard of 0.13 lb SO, /MM Btu, such a gas can have a
maximum sulfur content of about 0.02 percent by weight or about
200 ppm by weight. On the other hand, a high Btu gas (1,020
Btu/SCF, HHV) can have a maximum sulfur content of about 0.15
percent by weight or 1500 ppm by weight and still qualify for
combustion in boilers. However it is very unlikely that this
valuable product would be used in such a manner. More likely, a
high Btu gas (SNG) would be delivered to the pipeline and in this
case the maximum sulfur content permitted is 1/4 grain/SCF or
about H ppm (2,p.3).
There is a broad spectrum of waste gas streams from coal con-
version facilities which are potential candidates for combustion.
Each stream, or combination of streams, will require analysis to
determine the degree of sulfur control, if any, that is required
for combustion.
Desulfurization of Liquid Fuels—
Liquid fuels produced as byproducts of coal conversion facilities
will have appreciable sulfur contents. In order to reduce the
sulfur content of these fuels to acceptable levels, hydrodesul-
furization as practiced by petroleum refineries may be an
attractive choice. Some of these processes are listed in TABLE
9-5.
The Gulf HDS process may be regarded as typical of the hydrode-
sulfurization schemes. A simplified flow diagram in is shown in
Figure 9-32.
In this process, the heavy oil passes through a solids removal
section which separates filterable solids, such as iron com-
pounds. The oil feed is mixed with makeup and recycle hydrogen,
596
-------
LICENSOR
TABLE 9-5. IMPORTANT HYDRODESULFURIZATION PROCESSES*
PROCESS SCOPE OF APPLICATION
cn
Chevron Research Co.
Cities Service R&D Co. and
Hydrocarbon Research Inc.
ESSO Research & Engineering Co.
and Union Oil of California
Gulf Research & Development Co.
Institute Francais de Petrole
Standard Oil Co. (Indiana)
Universal Oil Products Co.
VGO Isomax
RDS Isomax
VRDS Isoraax
H-Oil
Go-Fining
Residfining
Gulf HDS Type I and Type II
Gulf HDS Type III
Heavy Gas Oil Gulfining
IFP Vacuum Gas Oil HDS
IFP Resid HDS
Resid Ultraflning
VGO Ultrafining
RCD Isomax
Hydrobon Process
Vacuum gas oil and lighter feedstocks
Whole crude, vacuum gas oil, and vacuum
tower bottoms
Vacuum tower residuum
Residual oil
High boiling virgin and cracked gas oil
Atmospheric tower residuum
Atmospheric tower residuum
Atmospheric tower residuum
Virgin or cracked heavy gas oils
Vacuum gas oil
Atmospheric tower residuum
Vacuum residuum
Vacuum gas oil
Atmospheric reduced crude, deasphalted
vacuum tower residuum
Light and heavy distillates
•From Item 2, p. XI-10, in reference list
-------
ATMOSPHERIC
REDUCED CRUDE
SOLIDS
REMOVAL
HYDROGEN MAKEUP
-^•-*-
i
IIDS
REACTOR
SEPARATION
RECYCLE GAS
PURIFICATION
H-S-RICH
— ^ .—
GAS
FRACTIONATION
SOUR GAS
DISTILLATES
FUEL OIL
Figure 9-32. Typical flow scheme for a fixed
bed hydrodesulfurization process.*
*From Item 2 in reference list
• 598
-------
heated, and reacted over a fixed bed of catalyst at elevated
pressure and temperature with evolution of H S. Specific opera-
ting conditions vary depending on the type of feedstock, the
desired product, and the particular process. For residuum feed,
the pressure may range upward from 71.2 kg/cm2 (1,000 psig) .
Typical reaction temperatures are 399 to 454 degrees C. The
reactor effluent is cooled and recycle gas is separated. Sulfur
is removed from the recycle gas before it joins the reactor feed.
Separator liquids flow to fractionation or stripping.
When distillates are fed to desulfurization processes, the solids
removal step is unnecessary since these distillates are already
purified liquids. The remaining processing steps are essentially
the same as illustrated. However, with lower boiling distillates
the catalyst quantity needed in the reactor is less and the
pressure and temperature conditions are less severe for a given
level of sulfur removal. For distillates, hydrotreating
pressures may range upward from 29 kg/cm2 (400 psig) and the
temperatures may be between 371 and 427 degrees C.
An alternative system to the fixed bed reactor is used in the
H-Oil process which was developed by Cities Service R&D and
Hydrocarbon Research, Inc. This process uses an ebullating bed
of catalyst through which the oil and hydrogen flow. The
ebullating bed (similar to a fluid bed but with liquid as the
dispersing medium) process is competitive with fixed bed systems
primarily when residuums are fed. Advantages of the ebullating
bed reactor over a fixed bed reactor in such a system are:
o Solids contained in the feed pass through the ebullating bed
and do not cause the plugging that occurs with fixed beds.
599
-------
o Catalyst can be added and removed from the ebullating bed
reactor while it is in operation, thus avoiding the
necessity for shutting down when catalyst activity is too
low
o In general, smaller catalyst particles can be used in
ebullating beds. These show higher reactivity for residuum
desulfurization.
A sulfur removal efficiency of about 90 percent can be achieved
with these methods (2,pp. XI-3-13).
Desulfurization of Coal Prior to Combustion. - The Meyers
Process—
The Meyers process, being developed by TRW, Inc., is reported to
be effective in removing pyritic sulfur from coal. In this
process, the pyrites in the coal react with ferric sulfate in a
solution containing ferric and ferrous sulfates and sulfuric
acid. The ferric ion is continuously regenerated by reaction of
oxygen and ferrous ion. The elemental sulfur product is
extracted with an organic solvent. The iron product from the
pyrites is removed as solid ferric and ferrous sulfates. A block
flow diagram of the basic Meyers process is shown in Figure 9-33.
Coal ground to less than 100 mesh is mixed with recycled leach
solution and is then pumped from the mixing vessel to one of lo
reactor vessels where the slurry is contacted with oxygen at
about 150°C. The pyritic sulfur is 90 percent converted to
elemental sulfur and,sulfate according to the following
reactions:
-i- Fe^SO^—^ 3FeS04 + 2S
+ 7Fe2(S04)3 + 8^ 0—f 15FeS04 + 8^ S04
600
-------
COAL
PREPARATION
OXYGEN
REACTION
SULFUR
REMOVAL
FILTRATION
VAPORS
PRODUCT
DRYING
COAL
PRODUCT
LIQUID
VAPORS
^« SOLVEMT
f
IQUID
SULFUR
RECOVERY
SULFUR
PRODUCT
FILTRATION
LIQUID
I ROD EULFATE
FILTRATION
IRON
SULFATE
PRODUCT
Figure 9-33. Flow diagram of the modified Meyers process*
*From Item 3, pp. 4-23 in reference list.
601
-------
4FeS04 + 21^ S04 + % * 2 F^ (S04 )3 +2^0
The net overall reaction for the leaching and regeneration steps
may be considered as:
Fe^ + 2.4 02—* 0.2 F^ (S04 >3 + 0.6 FeS04 + 0.8 S
The reaction slurry passes first to hydrocyclones, where about 60
percent of the liquid is removed and recycled to the reactors,
and then to filters, where the remainder of the leaoh solution is
removed and sent to iron sulfate recovery. The filter cake is
reslurried in recycled naphtha to dissolve most of the elemental
sulfur and then filtered. Water is separated from the filtrate
by decantation and the sulfur-laden solvent flows to sulfur
recovery.
The filter cake, containing about 25 percent moisture and 5
percent solvent, is partially dried under vacuum, where the
sensible heat of the coal is sufficient to drive off the solvent
and reduce the moisture to about 20 percent in the product coal.
Vapors are condensed, water is separated by decantation and both
are reused in the process.
In sulfur recovery, distillation separates the solvent from the
product sulfur. Water and solvent in the overhead vapors are
condensed, separated and recycled.
In iron sulfate recovery the filtrate from the reaction slurry is
heated to about 130°C. Some of the water flashes and is used for
heating in the reaction and sulfur recovery sections. The slurry
of iron sulfates is filtered, the filtrate is recycled to the
reaction section and the cake is sent to disposal.
602
-------
TABLE 9-6 is an example of operation of the process (3), where a
coal feed containing 3.92 percent total sulfur yielded a coal
product containing 0.95 percent total sulfur, amounting to over
75 percent reduction in total sulfur. The feed and product
contained 3.21 percent and 0.17 percent of pyritic sulfur,
respectively, demonstrating pyritic sulfur removal of over 95
percent. Because of the reduction of the ash content of the coal
product, the higher heating value increased by about 5 percent,
to 12,7^7 Btu/lb. The thermal efficiency for this example of the
process is 92.1 percent.
The Meyers process, as reported, is effective in removing pyritic
sulfur from coal and has an attractive potential for treating,
prior to combustion, those coals whose sulfur content is
predominantly in the pyritic form.
It is of interest to note that the total of the forms of sulfur
other than pyritic in the feed and in the product coal are
virtually identical, indicating that the Meyers process does not
significantly affect the other forms.
Desulfurization of Coal Prior to Combustion. Physical
Cleaning—
Physical removal of pyritic sulfur is the most highly developed
method technologically and potentially the lowest in cost.
The coals of the United States have highly variable charac-
teristics by seam and by geographic location. Since coals vary
so widely, coal cleaning processes are typically engineered for
each coal source and designed with respect to the use to be made
of the coal.
Coals are prepared by size reduction and subsequent particle
sorting based upon particle size and density. The level of coal
603
-------
TABLE 9-6. DESULFURIZATION OF COAL VIA THE MEYRS PROCESS
Basis; Processing 200,000 pounds per hour of dry coal
Ultimate Analysis of Feed, by Weight
c
H
N
Cl
68.53?
3.85
1.20
0.08
S-Pyritic
-Sulfate
-Organic
Ash
0
3.21*
0.04
0.67
20.86
1.56
Sulfur Balance, Pounds per Hour
In Feed - Pyritic
Sulfate
Organic
Total Input
6,416
80
1,340
7,836
Coal Loss
F62(S°4)3
Product Sulfur
In Product Coal to Boiler
In Net Product Coal - Pyritic
Sulfate
2
1,828
1'78°
2,438
125
298
53
Organic 1,242
Elemental " 70
Total Output
7,836
Thermal Efficiency
HHV of dry coal feed at 12,140 Btu/lb
HHV of net dry coal product
at 12,747 Btu/lb s
Thermal efficiency = Heat out/Heat in
2,428 billion Btu/hr
2,235 billion Btu/hr
92.1$
604
-------
quality improvements attainable is variable, being constrained by
processing objectives, cost, processing technology, and coal
characteristics.
Plants are designed to produce a product or products of
definitive characteristics for one specific customer. The
preparation plant is designed to remove the non-combustibles from
the coal at the minimum practical operating cost and at the
optimum practical yield. However, the ROM coal is prepared only
to the extent that is necessary to make the product salable.
The technical limitations of the preparation process relate
primarily to the very small component particles existing in coal.
Many of these particles are residual structures of vegetation and
minerals, generally irregular in shape. The pyrite particles in
many coals are less than 1 micron (0.00004 inch) in their longest
dimension. Particles smaller than 50 microns cannot be
practically separated from each other, and separating them is
usually inefficient. Larger particles, or those less homogeneous
in composition, respond more readily to separation.
To be separable, impurity-containing particles must have masses
greater than the pure coal particles. The difficulty in
separating particles of less than 50 microns results from their
slower response to the acceleration of gravity than larger
particles: they literally float within the coal. Moreover,
since most of the separation is done in water systems, there is
the further complication that removal of the water from the small
particles is significantly more difficult and more costly than
removing water from the larger particles due to the smaller
porosity of the smaller particles or of the combination of
particles. Because of the technical difficulty in separating
small particles, the separation costs increase as the particle
size decreases. The processes which will remove more pyrite from
605
-------
the coal necessarily utilize smaller particle sizes and are
considerably more costly. Accordingly, coal cleaned primarily
for ash removal is cleaned with as large a particle size as is
practical. It is for this reason that coal processing plants
which were not designed for sulfur removal often do not function
well as pyrite removers.
The economic limitations of coal preparation are varied and
numerous. Cleaning of coarse coal is relatively simple and less
costly than cleaning of the finer sizes. The fine coal portion
in the raw coal feed has materially increased as mechanization of
mining process has increased, thus adding considerably to clean-
ing plant costs. Wet cleaning units for fine coal are not
expensive; it is the equipment necessary to dewater and dry the
product that adds significantly to the cost. Clarifying the
process water and thermal drying substantially increase plant
capital investment. Yet many modern cleaning plants must contain
this equipment in order to obtain the desired ash, sulfur, and
moisture in the product and still recover the greatest amount of
salable coal.
In 197^, 258 million one of a total of 603 million tons of U.S.
coal produced were cleaned by wet methods (9,p.3D. The
following table (11,p.1-3) shows total U.S. coal production and
the amount cleaned for the years 1964-1972.
• 606
-------
PRODUCTION AND CLEANING OF U.S. COALS
Millions of Tons
196M 1965 1966 1967 1968 1969 1970 1971 1972
Total Pro-
duction M87 512 53M 553 5^5 561 603 552 595
Quantity
Cleaned 310 332 3*U 3M9 31*! 335 323 271 293
Percentage
Cleaned 63.7 6M.9 63.8 63.2 62.5 59.7 53-6 H9.1 U9.2
TABLE 9-7 shows the types of cleaning methods employed for U.S.
coals for the years 1967-1972.
However, the degree of sulfur reduction attainable via physical
cleaning is naturally limited, since only pyritic and sulfate
sulfur are removed.
Pyritic sulfur is the mineral pyrite which occurs in coal as dis-
crete particles, although often of microscopic size. It is a
heavy mineral which has a specific gravity of about 5.0, compared
to coal which has a maximum specific gravity of only 1.7. The
pyrite content of most coals can be reduced significantly by
utilizing coal preparation methods of size reduction and gravi-
metric separation.
The U.S. Bureau of Mines performed a comprehensive study (4) of
the washability of U. S. coals, involving M55 coal samples from
the 6 principal coal producing areas. The standard against which
the coal cleaning processes and the washability of the coals was
judged was the present federal standard of performance for solid
fossil fuel fired steam boilers: Maximum SO emission rate of
1.2 Ib/MM Btu heat iraput. On this basis the U.S.B.M. concluded:
607
-------
TABLE 9-7.
METHODS OF CLEANING U.S. BITUMINOUS
COALS AND LIGNITE*
Jigs
Tables
Class!
Launde
Flotat
Pneuma
Dense
Magn
Sand
CaCl
Total
Millions of Tons Cleaned
1967
160
s 50
ifiers 4
ers 5
tion 8
atic 21
Media:
netic 65
d 33
1 3
3^9
1968
159
47
5
4
9
17
71
27
2
341
1969
155
45
3
5
10
19
72
24
2
335
1970
140
44
4
5
10
18
77
23
2
323
1971
155
36
2
5
9
15
69
18
2
281
1972
128
40
3
5
13
12
75
15
2
293
•From Item 11, pp. 1-3, in reference list
608
-------
o "If all the coals were upgraded at a specific gravity of
1.60, the analyses of the clean coal products of the various
regions would range on the average from 5.1 to 8.3 percent
ash, 0.10 to 1.80 percent pyritic sulfur, 0.56 to 3.59
percent total sulfur, 12,799 to 14,264 Btu per pound and
would produce 0.95 to 5.5 pounds of S02/MM Btu at Btu
recoveries ranging from 91.7 to 97.6 percent. The
corresponding SOg removal efficiencies required to comply
with the current EPA emission regulations of 1.2 pounds
S02/MM Btu would range from none to 78 percent." The
evaluation data are summarized in TABLE 9-8.
o The 455 U.S. coal samples evaluated contained on the average
1.91 percent pyritic sulfur and 3.02 percent total sulfur.
Only 14 percent of the raw coal samples could meet the cur-
rent EPA S02 emission standard of 1.2 pounds SO_/MM Btu with
no preparation. If a 50 percent Btu recovery was accepta-
ble, then 32 percent of the samples could be upgraded to
meet the standard when crushed to 14-mesh top size.
When these results are contrasted with the most stringent state
standard (Wyoming) of 0.2 Ibs SC^/MM Btu, it is obvious that
physical coal cleaning may be considered only as a partial solu-
tion for the S02 emissions problem; however, as a first step, it
will lighten the load on the downstream sulfur removal/recovery
systems.
Desulfurization of Coal During Combustion in a Fluidized Bed—
Although fluidized bed combustion (FBC) techniques were developed
primarily for the electric power industry so that steam generator
systems could utilize high sulfur coals, the principles of FBC
may be applied in coal conversion plants to incinerators that
incidentally produce steam or to utility boilers whose prime
purpose is to generate process steam.
609
-------
TABLE 9-8.
SUMMARY OF COMPOSITE PRODUCT ANALYSES BY REGION FOR
CRUSHED AND CLEANED COALS
(Coals crushed to 3/8 inch top size
and cleaned at 1.60 specific gravity.)
Cumulative analyses of float 1.60 product
o\
M
O
Northern Appalachian
Southern Appalachian >
Alabama
Eastern Midwest
Western Midwest
Western
Total United States
Percent
Btu
recovery
92.5
96.1
96. U
9H.9
91.7
97.6
93.8
Ash
8.0
5i1
5.8
7.5
8.3
6.3
7.5
Pyritic
Sulfur
.
0.85
.19
.M9
1.03
1.80
.10
.85
Total
Sulfur
1.86
.91
1.16
2.71
3.59
.56
2.00
Pounds
soy MM
Btu(1)
2.7
1.3
1.7
H,2
5.5
.9
3.0
Calorific
content,
Btu per
pound(2)
13,766
1*4,197
1U.261
13,138
13,209
12,779
13,530
SO, removal
efficiency
required
percent(3)
56
8
29
71
78
None
60
(1) Based upon the moisture-free Btu value for the float coal and assuming all the sulfur is
converted to S02- Actual emissions will vary depending upon the as-fired coal moisture content
and the amount of sulfur that actually goes out the stack as S02.
(2) The calorific content (moisture-free basis) was used to calculate the S02 removal efficiency
required.
(3) S02 removal that must be accomplished in treatment of flue gases to meet the Federal new source
performance standard. Values may require adjustment to account for the as-fired coal moisture
content and the amount of sulfur that actually goes out the stack as S02.
-------
In general, an FBC system will include the following major
components:
o The fluidized bed is a mixture of inert bed material (i.e.,
sand, alumina, coal ash, etc.), coal particles and and SO
absorbent, typically limestone or dolomite. The particle
size in the bed can vary from -1/4 inch to 100 mesh. The
static bed depth can vary between 0.5 and 3 feet with an ex-
panded depth of 1 to 6 feet. The bed is fluidized with a
superficial gas velocity between 1 and 15 ft/sec and oper-
ates in a temperature range of 1,400-2,000°F.
o Heat transfer surfaces are located both within the fluidized
bed and external to the bed. Internal bed surfaces include
rows of tubes placed vertically, horizontally or at some in-
termediate pitch. Membrane type water walls can also be
used for internal bed surface. External surfaces include
tubes placed directly above the bed and downstream of the
combustion zone, such as economizer surface. The tempera-
ture profiles in the tubes depend upon the location of the
surface and the conditions of the internal fluid.
o Solids handling systems include the transfer of feed mate-
rials to the fluidized bed, the removal and/or recycling of
bed material, and the return of elutriated solids to the
bed. The feed materials injected into the fluidized bed are
at ambient temperature and can include coal, limestone, or
dolomite, inert bed material and additives. All other sys-
tems require handling hot ( ^ bed temperature) solids.
o Elutriated material recovery. Solids elutriated from the
bed include unburned carbon and other bed material. Unburned
carbon can be collected and combusted in a carbon burnup
cell or simply returned to the primary combustion bed.
Other material is usually collected in cyclone type separa-
tors and either discarded or returned to the bed with or
without further processing. In some cases, baffles are
611
-------
placed above the fluidized bed to eliminate entrainment of
particles in the combustion gases.
The coal-burning fluidized bed combustion boiler is formed by an
enclosure usually consisting of waterwalls (abutting boiler tubes
in which water and steam flow), or a lined uncooled shell. The
pressure in this enclosure is approximately 1 atmosphere. A
plate distributes the air flow uniformly over the base of the
enclosure. The air then passes at velocities in the range of 1
to 15 ft/sec through a bed of particles at temperatures from 760
to 1,095°C. These particles, whose maximum size is generally 1/4
inch, are comprised of ash or other inert material, both reacted
and unreacted sorbent (lime, dolomite) and small quantities (less
than 3 percent) of unburned coal or carbon. Coal, sorbent, inert
bed material and other additives are fed to the bed by various
techniques through the air distributor or the containment shell.
From 50 to 60 percent of the heat released in burning the fuel
with air is transferred to the water/steam in the tubes surround-
ing and submerged in the bed.
The air and combustion gases passing through the bed cause consi-
derable agitation. Particles are thrown from the bed into the
empty volume above, called the freeboard. Larger, heavier parti-
cles fall back into the bed. Smaller, lighter particles are car-
ried out of the enclosure by the gases. Much of the ash formed
during the combustion of washed coals is eventually carried from
the bed in the combustion gases. Larger ash particles may accu-
mulate in the bed and require continuous removal.
Convective heat transfer surfaces can be located in the path of
the combustion products to generate more steam from the sensible
heat of these gases. A reasonably high heat transfer coefficient
from the combustion gases to the tube surface requires high gas
velocities. These velocities can be obtained by narrowing or
612
-------
restricting the gas passage in the convection section or by pack-
ing tubes into a broad gas passage with small clearances between
tubes.
Ash, fragments of the limestone/dolomite sorbent, and unburned
coal char or carbon carried from the fluidized bed enclosure by
the combustion gases can be captured by particulate collectors.
To obtain high combustion efficiencies, these captured particles
must be recycled to the boiler bed or to a separate fluidized bed
combustor (a carbon burnup cell) where burning of the char is
completed. A secondary collector can collect particles not
collected in the primary collector for disposal.
Additional heat can be extracted from the combustion gases in the
heat recovery section, which may be an economizer (that preheats
boiler feedwater) and/or an air preheater. Final cleanup of the
combustion gases is carried out in a secondary particulate
removal system.
FBC can be operated in several ways to effect the desired
solid-gas reactions:
o Downflow, where limestone and fuel are fed into a precalcin-
ed bed of lime. Gas velocities are usually low so that
little lime, but most of the coal ash, is elutriated. Sol-
ids, consisting of CaSOj , CaO,and some ash, are removed
through an overflow pipe, thus maintaining a constant bed
level. The cooling surface can be immersed directly in the
bed to maintain the bed temperature at the desired level.
o Upflow, where the bed consists of coarse inert material.
Gas velocities are higher and feed materials are finer so
that both spent lime and ash are carried overhead. Because
the solids retention time is short, this mode of operation
is suitable if the gas-solids reactions are rapid.
613
-------
o Multi-Solids Combustor, under development by Battelle
Memorial Institute, uses a high specific gravity material
for the base bed with a recirculating entrained bed of fine
particles above it. Coal and limestone are fed into the
dense bed, SO 2 is absorbed there and the released heat
maintains the temperature of the entrained bed in which the
steam tubes are located. Space velocities up to 3^ ft/sec
have been reported. Satisfactory SCfc removal with Ca:S
ratios as low as 1.4 is claimed.
0 Cyclonic, patented by Babcock 4 Wilcox, is designed with a
high velocity center tube through which all combustion gases
pass and an annular fluidized bed of lime. The
configuration allows feed to enter at the top of the annular
bed then the char produced enters the bottom of the central
tube to supply heat to the solids while the combustion air
maintains a high velocity vortex in and at the top of the
tube. High unit capacity is claimed.
Operating temperatures of 760 to 870°C are predicted according to
limestone calcination theory and have been proved in practice:
temperatures above 760°C are required to promote rapid combustion
and for reasonable S02 reaction rates, while above 870°C sorbent
activity appears to decrease and corrosive alkali metal salts
begin to volatilize.
Best operation is obtained with excess air ranging from 5 to 15
percent. Above 15 percent, NOX formation becomes a problem and
excessive heat is lost in the stack gases. Below 5 percent leads
to emissions of combustibles and freeboard burning.
When coal is burned with excess air in the pressure of calcined
limestone or dolomite the sulfur in the coal is oxidized to S02
and reacts with limestone to form calcium sulfate or with
614
-------
dolomite to form a mixture of calcium and magnesium sulfates.
Although thermodynamic studies indicate that sorption of S02 or
I^S by limestone is possible and should proceed nearly to
completion, experimental results indicate that calculated
equilibrium values are not approached and that the reaction
kinetics appear to be controlling. Careful design based on both
theoretical and actual considerations is required for
satisfactory performance.
Reference 7 discusses the design factors, their application and
results of their application on pp.3-1-3-9 and 6-1-6-2.
Flue Gas Desulfurization—
If sulfur in fuels fed to combustion equipment is not removed
before or during combustion by methods discussed previously, then
flue gas desulfurization ( FGD) must be employed to meet
stringent environmental standards.
The average U.S. coal contains 3.02 percent total sulfur and has
a higher heating value of 12,574 Btu/lb (4,p.2). If this fuel is
burned without controls, an S02 emission of 4.8 Ib/MM Btu
results. To meet the most stringent standard of 0.2 Ib/MM Btu,
an FGD efficiency of about 96 percent would be required.
Obviously, if the coal contains less sulfur a lower SO, removal
efficiency is required, and conversely.
The tar produced from a gasification facility processing average
U.S. coal is likely to contain about 1.81 percent sulfur (60
percent of 3.02 percent). The uncontrolled SO- emission would be
about 2.2 Ib/MM Btu. Therefore, to meet a standard of 0.34 Ib/MM
Btu, an FGD efficiency of about 85 percent is required.
More than 100 FGD processes in various stages of development are
reported in the literature (9,p.67). A Kellogg study addressed
615
-------
such processes and recommended 4 for further study (8). These
were:
o USBM Citrate
o Wellman-Lord
o Chiyoda Thoroughbred 101
o Pullman Kellogg Weir
The Federal Power Commission, reporting on 7 commercial FGD
processes, listed the process charactistics (9,p.40) as in TABLE
9-9.
In a later section of the report they describe 15 advanced FGD
processes which are estimated to be commercially available from
1976 to 1988. These are shown in TABLE 9-10.
There were about 45 FGD units installed at 11,000 MW of
generating capacity as of the end of 1977. Projections indicate
that about 70 FGD units, installed on 26,000 MW of generating
capacity, will be in operation by 1981 (9,p.50). In 1975, the
weighted average capital cost of the operating FGD systems was
$90/KW and the operating cost was 3.1 mills/KwH (9,p.24). The
costs were somewhat higher for retrofitted systems as opposed to
new systems and for systems treating high sulfur coal as opposed
to those treating low sulfur coal.
FGD processes are normally subdivided into two types:
o Regenerable, where the product sulfur removed is in
marketable form such as elemental sulfur (S), sulfuric acid
(H^OJ, or concentrated sulfur dioxide (SO-).
o Non-regenerable, where the product sulfur removed is in
waste form (normally a sludge containing CaSO , CaSO ,
unreacted CaCO , fly ash, and water).
616
-------
TABLE 9-9. CHARACTERISTICS OF COMMERCIAL FGD PROCESSES
Process
Wet Lime/Limestone
Alkaline Fly Ash
Sodium Carbonate
Double Alkali
Dilute Sulfuric
Acid/Gypsum
Magnesium Oxide
Wellman-Lord
Primary Removal Agent
Lime or Limestone
Alkaline Fly Ash
Sodium Carbonate
Sodium Hydroxide
Sulfuric Acid
Magnesium Oxide
Sodium Sulfite
Form of Principal Operational
Regenerable? Sulfur Product Mode
No
No
No
Yes
Yes
Yes
Yes
CaSO (Throwaway)
CaS03 (Throwaway)
Na SO (Throwaway)
CaS04 (Throwaway)
CaSO (Throwaway)
so2
SO,
Wet
Wet
Wet
Wet
Wet
Wet
Wet
-------
TABLE 9-10. CHARACTERISTICS OF ADVANCED FGD PROCESSES
Process
GO
Agglomerating Cone
Allied/Wellman-Lord
Ammonia Scrubbing
Basic Aluminum Sulfate
- Gypsum
Catalytic Oxidation
Citrate
Copper Oxide (Shell)
Dry Adsorption
Electrolytic Regeneration
(Stone 4 Webster/Ionics)
Manganese Oxide
Aqueous Carbonate
Nahcolite Injection
Organic Absorbent
Potassium Thiosulfate
Phosphate
Primary Removal Agent
Phosphate Rock Slurry
Sodium Sulfite
Ammonium Solution
Aluminum Sulfate
Vanadium Pentoxide
Citric Acid
Copper Oxide
Activated Carbon
Caustic Soda
Manganese Oxide
Liquid Carbonate
Nahcolite Ore
Glyoxylic Acid
Sulfates, Sulfites
Phosphate Buffer
Form of
Regenerable? Primary Sulfur Product
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Fertilizer
Sulfur
Sulfur
Gypsum
H2S04
Sulfur
SO 2
Sulfur
H2S04
SO,
H2S
Na 2SO 4
SO 2
Sulfur
Sulfur
Operational
Mode
Wet
Dry
Wet
Wet
Dry
Wet
Dry
Dry
Wet
Dry
Wet
Dry
Wet
Wet
Wet
-------
This study will be confined to the description of several of the
more promising of the candidate systems.
Citrate Process—Research sponsored by the U.S. Bureau of Mines
has developed a process where the flue gas is first cooled and
scrubbed with an ash slurry to remove solid particles and sulfur
trioxide. The gas then enters an absorber where over 95 percent
of the SO contained in the treated gas is removed by contact
with a citric acid/sodium citrate/sodium thiosulfate solution.
The clean gas is reheated and exhausted to the atmosphere.
The SO rich solution flows by gravity to a stirred reactor
vessel and is reacted with H S, which is available from the acid
gas removal system, to precipitate sulfur. The resultant slurry
is separated by flotation as a 10 to 15 percent solids product,
leaving clear regenerated citrate for recycle to the absorption
tower.
The sulfur flotation product is filtered and the solids are
heated to melt the sulfur at about 138°C. The hot liquid passes
into a settler tank from which molten sulfur is tapped. The
citrate solution is withdrawn from the top of the settler for
reuse (8).
Advantages of the citrate process include:
o Removal efficiencies on the order of 95 to 98 percent have
been obtained on pilot scale operations.
o Precipitation of sulfur compound takes place outside of the
boiler; hence, there is no plugging or scaling.
o The system has a high capacity for short-term SO overloads.
619
-------
o The reactions in the reactor tank occur rapidly, resulting
in a fast-settling sludge.
o The low liquid-to-gas ratio leads to a low pressure drop
across the absorber. Energy requirements are thus
minimized.
o The product is elemental sulfur.
The process has some disadvantages and possible problem areas:
o The process of separation of sulfur from solution in a
decanter may be difficult.
o Regeneration by production of ^ S may prove to be costly due
to expenses associated with the gas feed requirements.
However, when f S is available, this is no longer a factor.
o Lack of commercial experience may be the main negative
factor (9, p. 72).
Although no commercial plants are presently in operation, one is
being installed at the St. Joe Minerals Company's G.F. Weaton
Power Station at Monaca, Pennsylvania. This unit will treat
156,000 SCFM (234,000 ACFM) of flue gas from a 60 MW coal fired
boiler. Completion of construction is scheduled for the fall of
1978. A one year demonstration program will follow.
Capital and operating costs for an FGD unit handling flue gas
from generation of 1000 MW fed with 2.5 percent sulfur coal are
estimated to be $73/KW and 1.97 mills/KWH or $5.U9/ton of coal
(1977). These figures include the H2S generation unit. If H2S
generation is not needed, the capital cost drops to about $67/KW
and the operating cost is also reduced (10, pp. 1-4,25-27).
620
-------
Allied/Wellman-Lord Process--In t.his process the flue gas is
first scrubbed by a recirculating ash slurry to remove fly ash.
The concentration of the circulating stream is controlled by a
purge stream which is directed to a disposal pond.
The cool gas then enters an absorber where it is contacted with a
sodium sulfite solution which absorbs S02 to form sodium
bisulfite. A side stream of this solution is purged for sulfate
removal by crystallization. The clean flue gas is discharged to
a stack after reheating.
The absorber liquor is heated in an evaporator-crystallizer,
where SO and water vapor are released and sodium sulfite
crystals are removed as a slurry. The water in the overhead
vapors is condensed and used to redissolve the crystals for
recycle to the absorber. Prior to redissolving, the slurry is
thickened and the clear solution is returned to the evaporator.
A small purge is withdrawn to control the thiosulfate
concentration. The SCL gas is compressed and can be sent to a
Claus plant where it reacts with IL S available from the complex
to recover sulfur.
Advantages of the process are:
o Elemental sulfur is recovered in the process, allowing easy
handling and yielding a potentially marketable product.
o The operation can quickly adjust to flue gases of any S02
concentration.
o The process has good commercial experience with 10 units
recovering S02 in operation. Another 18 units are in
various design stages.
621
-------
o S02 removal efficiencies of greater than 95 percent are
possible.
Possible disadvantages and problems are:
o The use of methane as a reducing gas can lead to
significantly high materials costs. It is not required when
H 2$ is available.
o Capital cost for this process appears somewhat higher than
for others. A Pullman Kellogg study indicated that the
capital investment is about 33 percent higher than the
citrate process (8) (9,pp.67,70).
Ammonia Scrubbing Proces3--In this process ammonia compounds
e.g., ammonium bisulfite, possessing very high SO absorption
rates, are used as the absorbent. The sulfites and sulfates thus
produced can be regenerated by chemical reduction. Elemental
sulfur is recovered by the Glaus process by conversion of
recovered SO with hydrogen sulfide recovered from the complex
(9,p.70).
Process advantages are:
o The very high SO absorption rate for ammonia permits a low
liquid-to-gas ratio.
o Extensive regeneration of the spent scrubber solution is
possible due to the volatile nature of ammonia.
622
-------
Possible disadvantages and operational problems are:
o An environmentally hazardous plume tends to form if process
control is not rigidly maintained. Ammonium sulfate has
been found to be the primary constituent of the exhaust gas.
o In practice it has sometimes been difficult to separate the
fly ash and the ammonium sulfite.
Shell FGD Process — In this process, the flue gas is first passed
through precipitators, of a type determined by the particulate
size distribution, for solids removal.
Flue gas then passes over a static, packed bed of copper oxide
acceptor plates at about 400°C. SC^ and oxygen in the flue gas
react to form copper sulfate. When one bed is spent, it is
regenerated with hydrogen, also at 400°C. The reactions are
summarized as follows:
CuO + 1/2 02 + S02 —
CuSOi| + 2H2 - fr-Cu + S02 * 2H20
Cu + 1/2 02 — *• CuO
The regeneration off-gas passes through a quench-absorber
stripper section which reduces the temperature of the gases by
waste heat recovery and delivers a concentrated S&> stream for
further processing.
If recovery of elemental sulfur is required, then HL S from the
facilities is used for the Glaus reaction (8) (9, p. 73).
623
-------
Advantages of the process are:
o Since this is a dry process, there is no handling of wet
materials and water input requirements are minimal.
o Acceptance and regeneration occur at approximately the same
temperature, obviating any heating or cooling of the absorp-
tion beds.
o The flue gas passes over the surface of the acceptor materi-
al rather than through it. This arrangement prevents
plugging of the acceptor bed.
o By alternating the acceptor and regeneration units, contin-
uous processing can be maintained.
o Operating costs are minimal due to absence of reheating and
to low water requirements.
o The process has the potential for expansion to SO and NO
x x
removal by ammonia injection into the acceptor bed. The CuO
will act as a catalyst for NO absorption.
Jv
o There is no disposal of spent sludge.
Process disadvantages and possible operating problems include:
o Equipment and installation costs are high.
o A hydrogen source is needed for regeneration. Operating
costs may be high due to the large hydrogen requirements.
o Recovery is in the form of gaseous S^ , requiring further
processing for marketing.
624
-------
o Damper operation becomes important when switching
operational modes at high temperature.
o The stripper requires larger quantities of steam resulting
in high energy input requirements. The acidic wastewater
must be neutralized prior to discharge.
o The hydrogen feed may not be adjustable to fluctuations in
S02 input concentrations.
Pullman Kellogg Weir Scrubber--Pullman Kellogg offers a
multistage, countercurrent, limestone scrubbing process with a
low pressure drop, high SO2 conversion and high limestone
utilization. In the first step, hot gases are quenched to
saturation temperature and passed along a series of horizontal
scrubbers where a limestone slurry is sprayed vertically into the
contact chamber. The gases are reheated to discharge temperature
and enter the stack for discharge to the atmosphere.
The limestone slurry, containing catalysts, is collected in a
slurry tank and is recirculated to the scrubber. Net make of
sulfite-sulfate slurry is oxidized by air or oxygen and decanted.
Thickened slurry is filtered on a rotary vacuum filter and the
cake is either directed to disposal or further processed for
marketing as commercial gypsum. Clear liquid is sent to
limestone slurry makeup. Catalyst, usually magnesium salts or
oxide, must be added to compensate for losses due to retention of
some liquid in the washed filter cake (8) (9,p.43).
The process has the following advantages:
o Preliminary economics indicate that this process has one of
the lowest capital outlays of all the processes being
considered.
625
-------
o The basic process is fairly simple. Very few process steps
are included.
o Reserves of absorbent materials are fairly abundant in the
United States.
o SO2 removal efficiencies are generally high, on the order of
95 percent.
o The two-stage treatment of flue gases allows for the
simultaneous removal of SO2 and particulates.
o The process is the method most commonly used by utilities
for S02 control, exclusive of low sulfur fuel. Commercial
installations have been operating for more than four years.
Therefore, the lime/limestone process is the most fully
characterized of the existing FGD processes. Operational
experience has led to a greater understanding of basic
principles.
o Successful performance, particularly in terms of SO
removal, on coal-fired systems has been demonstrated.
o The process is not adversely affected by fly ash in the
system.
Disadvantages and possible process problems are:
o Large quantities of waste require disposal in an
environmentally acceptable manner.
o If not designed carefully or operated attentively, lime/
limestone systems have a tendency towards chemical scaling
plugging, and erosion. These problems can frequently halt
operation of the system.
626
-------
o There appears to be a deficiency in the understanding of the
factors that cause or prevent the oxidation of sulfite to
sulfate, thus enhancing or inhibiting serious scaling.
Excess air, pH, fly ash, residence time in the reaction
tank, and the presence of N02 in the flue gas are suspected
to be contributing factors.
o Very high liquid-to-gas ratios are required in the scrubber.
o In the wet lime process, the sludge has poor settling
properties due to the high sulfite content.
Chiyoda Process—The Chiyoda Thoroughbred 101 Process requires a
front-end water scrubber to eliminate solids carryover in the
flue gas. The SO in the gas is absorbed in 2 to 5 percent
sulfuric acid where it is oxidized to S03 and H2S04 by air in the
presence of a ferric sulfate catalyst. The gas is reheated for
release through the stack.
The net make of acid is neutralized with lime or limestone,
producing gypsum which is separated by filtration. The mother
liquid is returned to the scrubbing step. The gypsum is of high
grade, suitable for use in wallboards or cement (8) (9,p.46).
Advantages of the process are:
o The process flow and plant structure are uncomplicated.
Capital cost is reasonably low.
o If gypsum can be recovered as a marketable product, the need
to dispose of waste products is obviated. If it cannot be
recovered, the form of the waste product is relatively easy
to handle.
627
-------
o High SO removal efficiencies, on the order of 97 percent,
have been reported.
o Scaling problems have proved to be minimal, relative to
those encountered with the lime/limestone process.
o No plugging problems have been encountered, since the
circulating fluid is highly acidic.
o The operational reliability of this process has been
relatively high.
o The gypsum produced is of good quality.
o The process has good commercial experience.
o The process is not adversely affected by fly ash in the
system.
Disadvantages and possible operational problems are:
o Since the process involves handling sulfuric acid solutions
special corrosion-resistant materials are required.
o There is a possibility of problems resulting from poisoning
of the ferric sulfate catalyst.
o The marketability of gypsum in the United States is highly
questionable.
o The process design requires large pumps and fairly large
scrubbers and oxidizers.
o A high liquid-to-gas ratio is required in the scrubber.
628
-------
o The recovered gypsum is dewa'tered in a centrifuge which
requires high maintenance.
o Handling difficulties arise from the large volumes of
sulfuric acid passing through the system.
References—
1. Hofferl, F. D. , Soung, W.Y., and Stover, S.E., "Summary of Gas
Stream Control Technology for Major Pollutant in Raw
Industrial Fuel Gas." Hydrocarbon Research, Inc. (Draft of
EPA Report), Oct. 1977. 899*
2. Glaser, F., "Emissions for Processes Producing Clean Fuels."
Booz-Allen & Hamilton, Inc. EPA 450/3-75-028, March 1974
315*
3. Magee, E.M., "Evaluation of Pollution Control in Fossil Fuel
Conversion Processes. Coal Treatment: Section 1. Meyers
Process." EPA 650/2-74-009K, September, 1975. 297*
4. Cavallaro, J.A., Johnston, M.T., and Deurbrouck, A.W., "Sulfur
Reduction Potential of U.S. Coals." EPA 600/2-76-091,
April, 1976. 596«
5. Deurbrouck, A. W., and Hudy, J., Jr., "Performance Character-
istics of Coal Washing Equipment: Dense-Medium Cyclones."
Bureau of Mines PB212656, 1972. 326*
6. Fluor Utah, Inc., "Economic System Analysis of Coal Precon-
version Technology, Volume 4: Large Scale Coal Processing
for Coal Conversion." July, 1975. 421*
•Pullman Kellogg Reference File Number
629
-------
7. Dowdy, T., Lapple, W. , Kitto, J., Stanoch, T., and Boll, R. ,
"Summary Evaluation of Atmospheric Pressure Fluidized Bed
Combustion Applied to Electric Utility Large Steam
Generators." EPRI FP-308, October, 1976. 773*
8. Pullman Kellogg Technical Report for Getty Oil (Eastern)
Company. Sulfur Emissions Study, June 1976 (Confidential)
9. Federal Power Commission, "The Status of Flue Gas Desul-
furization Applications in the United States: A
Technological Assessment." July, 1977. 6l8»
10. Madenburg, R.S., and Kurey, R.A., "Citrate Process Demonstration
Plant. A Progress Report." U.S. EPA Flue Gas Desulfuriza-
tion Symposium, November, 1977. 89U«
11. Hall, E. , Peterson, D. , Foster, J., and Kiang, K. , "Fuels
Technology, A State of the Art Review." EPA 650/2-75-03^,
April, 1975. 201*
.630
-------
Processes for Control of Hydrogen Sulfide
The Glaus Process—
The Claus Process is a commercial process for production of
sulfur from gas streams that contain H S. The system is very
flexible with regard to the H2S content of the feed gas: H2S
concentrations as low as 7 percent (1, p.121) and as high as
essentially 100 percent can be sucessfully treated. Conversion
of H_ S to sulfur can be as high as 95 to 97.5 percent for
three-stage plants treating rich feed streams (6, p.137), while
the conversion will fall off to about 90 to 91* percent for lean
feed streams.
The process essentially consists of an oxidation step, where
one-third of the H S is oxidized with air to SO , followed by
several catalytic reaction steps where the rest of the H S reacts
to form sulfur. Overall, the reaction can be represented by the
equation:
3H2S + 1.5 02 *(3/x)Sx + 3H20
The presence of hydrocarbons in the feed gas can cause a
deterioration of the plant operation in that:
o Conversion of H S to sulfur is reduced
o Combustion air requirements are increased
o Gas volume flowing through the plant can increase sub-
stantially
For example, a feed containing 5 percent methane will
increase air requirements by 35 percent and will increase
gas flow by 27 percent (1, p.120)
o COS and CS2 may be formed from the hydrocarbons (1, p.121)
Normally it is recommended that the hydrocarbon content of the
feed be reduced to a low level prior to the Claus operation.
631
-------
The presence of ammonia in the feed gas presents a potential
problem. For a conventional Claus plant, 2 to 3 percent ammonia
in the feed gas is a practical limit. However, when the feed
gas contains 30 percent or more of CO , the ammonia content
should be maintained below 0.05 to 0.10 percent (1, p.121).
Ammonia can be removed from the feed gas by scrubbing with water.
For these various reasons, a Claus plant alone cannot be expected
to reduce I^S in process offgases to a level low enough to meet
environmental standards. Developments in technology for control
of H2S have been directed toward treatment of Claus offgas and
have been shown to be successful. Combination of offgas
treatment processes with the basic Claus process, therefore
results in systems that will, in most cases, yield final offgas
streams that can be vented to atmosphere.
Combined Claus and Beavon Processes—
This design combines a Claus unit for bulk sulfur conversion and
recovery with a Beavon unit for conversion of the remaining
sulfur compounds present in the Claus tail gas to elemental
sulfur.
The Beavon process is a tail gas cleanup process based on the
catalytic conversion of sulfur species to H2 S by hydrogenation
and hydrolysis followed by conversion of H^ to elemental sulfur
by the Stretford process. The process is commercial and is
licensed by The Ralph M. Parsons Company. There are about thirty
Beavon installations in operation (4,p.4). It is highly
successful in reducing sulfur emissions to the atmosphere.
Effluent from the absorber is guaranteed to contain less than 100
ppm (v) total sulfur of which less than 10 ppm is HJS.
632
-------
Combined Glaus and SCOT Processes—
This combination links a Claus plant to the SCOT (Shell Claus
Off-Gas Treating) process for tail gas treatment. The SCOT
process treats the gas by a catalytic conversion step which
converts various sulfur species (COS, CS2 , SO 2, S, etc.) to H2S
via hydrogenation and hydrolysis reactions. Fuel gas (light
hydrocarbon) and air are needed to produce the reducing gas for
these reactions, as in the Beavon process. The ^ S is then
absorbed into an alkanolamine solution in a scrubbing system by
countercurrent contact with the solvent. The overhead gas stream
is released to atmosphere.
Rich solvent is pumped through a heat exchanger to the stripper
where H2S is released overhead in a fairly concentrated stream
that is recycled to the Claus plant. Steam supplies heat to the
stripper reboiler and cooling water is normally used in the
overhead condenser.
The process is commercial with a number of installations in
operation. It is licensed by Shell Development Company. A high
sulfur cleanup efficiency is supposedly achievable by the
process. It is claimed that the absorber overhead gas contains
less than 100 ppm (v) total sulfur species with less than 10 ppm
(v) H2S. Reported data, however, indicate that the total sulfur
content of this stream may be as high as 300 to 350 ppm (v) (7).
Combined Claus and ARCO Processes—
This design combines a Claus plant with an ARCO (Atlantic
Richfield Company) process for tail gas treatment. No
performance data are available on the process, but information
from Parsons (2) indicates that the concept is essentially the
same as the Claus/SCOT combination.
633
-------
Combined Glaus and Incineration Processes —
This system simply combines a Claus unit with an incineration
step to oxidize the tail gas sulfur species to SCL . A flue gas
desulfurization (FGD) process is required following incineration.
This concept has the advantage of removal of hydrocarbons (by
oxidation) which may otherwise be present in the tail gases. it
is, of course, commercial.
Combined Claus and Lucas Processes —
This system combines a Claus plant with a Lucas plant to treat
the Claus tail gas. In the Lucas process, all sulfur species
present in the Claus tail gas are incinerated to SCL . The SO- is
recovered in concentrated form and returned to the Claus plant.
The Lucas process consists of three basic stages. In the first
(incineration), tail gas is burned with air and fuel to convert
most of the sulfur compounds to S02 . The second stage consists
of a hot coke treatment of the incinerator off gas where the
following reactions take place:
02 + C— *C02
2S03 + C — *2S02 + C02
H2S + 1.502 — » S02 + H20
2NO + C— >N2 + C02
The third stage is the removal and concentration of SCU by an
absorption/regeneration process using aqueous alkali phosphate
solution. Sulfur dioxide is absorbed as in the following
reaction:
+ H20 + S02 - *NaH2P04 + NaHS03
In the regenerator the reaction is reversed and SO is released
for recycle to the Claus unit (8, p. 110).
634
-------
The process has been tested on a semi-commercial scale in a
German refinery. To our knowledge it has not yet been applied on
a fully commercial scale. The absorber off-gas supposedly
contains practically no H2S, less than 200 ppm SC^ and less than
150 ppm COS and CS2 • From the standpoint of pollution control,
this process does not appear to compare favorably with those
previously discussed.
The Stretford Process for Primary Sulfur Recovery—
As an alternate to primary sulfur production by Glaus processing,
the Stretford process may be employed in one of the following
manners:
o Pressure Stretford process for removal of ^S from the main
process gas stream and subsequent sulfur production.
o Low pressure Stretford process for sulfur production from an
H2S-rich stream from an acid gas removal (gas purification)
system such as Selexol, Rectisol, MEA, hot carbonate, etc.
The Stretford process is described in detail in "Integrated
Schemes for Emission Control," later in this section of the
report. When the Stretford process is employed, some means must
be employed to convert organic sulfur compounds (COS, CS2»
mercaptans,and thiophenes) to H2S upstream of the Stretford unit,
otherwise they would pass through the Stretford absorber to the
atmosphere. A number of catalytic conversion processes are
available for this purpose. Some of these are:
o Holmes-Maxted (commercial)
o Carpenter-Evans (commercial but not widely accepted)
o British Gas Council (proposed)
635
-------
In these processes the gas is heated and passed through a
catalyst bed where organic sulfur compounds are catalytically
converted to H2S by hydrogenation and hydrolysis. The heated
gas is then suitable for feed to the Stretford process. A
simplified flow scheme and the reactions taking place are shown
in Figure 9-3*1.
When a hot potassium carbonate process is used for H S removal
from process gas, a considerable degree of hydrolysis of COS
occurs:
COS + H20—*H2S + C02
In this case the H S-rich gas may be suitable for use in the
Stretford process without a catalytic conversion process.
The Stretford process itself is commercial with 31 plants in
operation (5). A later article (1978) reports 36 plants in
operation. Conversion efficiencies for H S are reported to be
over 99$ and H-S concentrations in the absorber off-gas are
reported to be between 5 and 8 ppm (v).
References—
1. Chute, A. E., "Tailor Sulfur Plants to Unusual Conditions."
Hydrocarbon Processing, April, 1977.
2. Personal communications with M. H. Griebe, The Ralph M. Par-
sons Company on analysis of the Claus/Beavon system, March
through May, 1978.
3. Pullman Kellogg, "Engineering Evaluation of a Process to
Produce 250 Billion Btu/Day Pipeline Quality Gas." Prepared
for Panhandle Eastern and Peabody Coal Companies
(confidential).
636
-------
TYPICAL REACTIONS:
CS2 + 2II2
COS + II2
RCH2 SH
C + 2
•CO +
PREI1EATERS
CONVERTER
SPRAY
:OOLER
-»— TO STRATFORD,
PRPCESS
"2°
Figure 9-34. Typical schematic and reactions for catalytic conversion processes.*
*From Item 9, p. 111-31 in reference list
-------
4. Beavon, D. K., "Four Years' Experience with the Beavon Sul-
fur Removal Process." APCA 70th Annual Meeting, Toronto
June, 1977. 905*
5. Vasan, S.,and Moyes, A. J., "Holmes-Stretford H2S Removal Pro-
cess Proved in Use." Oil and Gas Journal, September 2
1974. 889*
6. Goar, B. G., "Tighter Control of Glaus Plants." Oil and Gas
Journal, August 22, 1977.
7. Naber, J. E., Wesselingh, J. A., and Groenendaal, w., "New Sh 11
Process Treats Glaus Off-Gas." Chemical Engineering
Progress, Vol. 69, No. 12, December 1973. 567*
8. Doerges, A., Bratzler, K., and Schlauer, J., "Lucas Proces
Cleans Lean H.S Streams." Hydrocarbon Processing, October
1976.
9. Booz-Allen & Hamilton Inc., "Evaluation of Techniques to
Remove and Recover Sulfur Present in Fuel Gases Produced in
Heavy Fossil Fuel Conversion Plants." Report No. 9075-ois
January, 1975. 606*
638
-------
Techniques for Control of Particulates
Particulates are chemical elements, compounds or mixtures of them
in either solid or condensed liquid droplet form. They are
usually described in terms of the physical characteristics which
affect the mechanisms for their separation. The most important
of these are physical size and density. Figure 9-35 is a summary
of particulate sources, their size ranges and the equipment
usually used to remove them from gas streams.
The size, density, and nature of particulate matter directly af-
fect equipment selection and design, as shown in Figure 9-36.
The types of particulate collectors may be arranged in order of
increasing particulate collection efficiency, complexity and cost
as:
o Cyclone collectors
o Wet scrubbers
o Fabric collectors
o Electrostatic precipitators
Cyclones--
The most widely used type of particulate control device is the
cyclone collector. An example is shown in Figure 9-37. Particu-
late-laden gas enters tangentially at the top of the cyclone body
and its liniar velocity is translated into angular acceleration.
Because the density of the particulates is greater than the den-
sity of the gas, centrifigal force moves them rapidly to the
cyclone wall where they collect. The force of gravity encourages
movement of the collected particulates downward to the discharge
at the cyclone bottom.
Of the many parameters in the design of a cyclone for a specific
application, pressure drop and particle collection efficiency are
639
-------
PARTICLE DIAMETER. MICRONS, fj. IMM
00001 0001 001 Oil O CO I.QOO O,000
EQUIVALENT
SIZES
TECHNICAL
DEFINITIONS
COMMON
ATMOSPHERIC
OISPERSOIOS
TYPICAL
ARTICLES
AND CAS
OISPERSOIOS
TYPES OF
CAS
CLEANING
EQUIPMENT
GAS
&***«
SOlOS
SOLID
LIQUID
_M£T/
ousi
ZINC
0«lO£
FUME
COLLOID!
SlLI
sot
HOC
-* VIF
» _ _ —
— SMO(
K-S«
U.LURGIC
rs AND F
•-C4X9C
BL*C
It'**'
:*
-ifMOSC
«!--
t ,
uses*
HIGH E
AlR FlL
ELECTS
PRECiPl
*
i —^— -*-
OIL
«OKCS«
AL
'•flri^i'Mili
US SCREEN MCSH
•«|>«!» '«|||j
-tj CLOUOS -.-
~*1*~UtO f06~"?
I '
J FtRTIL'
•) LIMIS
FLY ASH—
»-COAL DUST—*
UMES
COHW U_C
•KV*^
P»IHT .
*~ »ioMt»*rs •**
• Ih.'
— *l
»-%a
MtWIC 01
COMBUST
HuO.lt
• K-E
LOTM CO
FFiClENC
TERS
ICAL .
TATORS
1
GROUND
^
*-%
JST
ION
ACTERIA
0 SCRUB
LLECTOF
T-~H
(••
E^T-*
"ULVERli
COAL
mxcTici
OUSTS
TAUCI*-
• — 1
SR°— <
•1
r«-
—rt
*...._
8ERS-*
S — ft
MECHiN
SEPARA
— ••(
C"iZS.£ '
IK'IH RAlN-^
•1ST
[*.GK9bN
TON£
•H
ED-*4
1C
POLLEN
HYORAL
NOZZL
— DROP
PNEUVtT
NOZ2I.C C
—SETT^
^HAMB
COL
TORS*
0
;
LIC
;— »,
C
mo»»
G
ERS
00001 0001 OOt 01 ' (0 tOO 1000 10000
PARTICLE DIAMETER. MICRONS. M IWM
Figure 9-35. Characteristics of particles and
particle dispersoids. (Adapted
from Stanford Research Institute*
*From Item 1, p. 5-3 in reference list
640
-------
.001
1000
CTi
PARTICLE SIZE. MICRONS
U.S. SCREEN MESH
HIGH
ENERGY
SCRUBBER
LOW
ENERGY
SCRUBBER
DRY
CYCLONE
COLLECTOR
ELECTRO-
STATIC
PRECIPITATOR
CLOTH
COLLECTOR
CARBON BLACK
H SO MIST
2 4
PAINT PIGMENT
STOKER FLYASH
PULVERIZED FLYASH
METALLURGICAL FUMES
I
ZINC OXIDE FUMES
METALLURGICAL DUST
PHOSPHATE ROCK DUST
TYPICAL
PARTICLES
MAGNESIUM OXIDE
ALKALI FUMES
CEMENT DUST
i
MILLED FLOUR
GROUND LIMESTONE
NH CL FUMES
4
.001
0.1 1.0
PARTICLE SIZE. MICRONS
100
1000
Figure 9-36.- Types of collectors for various constituents.
-------
Figure 9-37.
too
80
5 60
a.
u
y
u.
20
Operating principle of a dry inertial cyclone
riol 1 *>rtor. *
collector
10 15 20
ARTICLE DIAMETER. MICRONS
25
JO
Figure 9- 30.
Typical fractional performance curves for a
multitube mechanical collector.*
*From Item 1, pp. 5-7, 5-9 in reference list
642
-------
the most important, pressure drop because of its effect on duct
sizing, fan sizing and fan horsepower, and efficiency because of
its relationship to pressure drop and cyclone configuration.
Both of these parameters become particularly important when large
volumes of gas must be cleaned and have been the incentive for
development of cyclone types that depart considerably from the
classic configuration of Figure 9-37. One such design sub-
divides the gas stream among several small cyclones to maintain
separation efficiency without excessive pressure drop. The
performance of such an array is shown in Figure 9-38.
Collection efficiency as depicted in Figure 9-38 is typical for
cyclone collectors in that efficiencies drop rapidly for parti-
cles below about 10 microns in diameter. Cyclones are most com-
monly used in these applications:
o When particulates are mainly in the coarser size ranges
o When particulate concentrations are fairly high, e.g.,
above 3 grains per scf
o When high collection efficiency is not critical
o When they can serve as pre-collectors in conjunction with
other types of collectors that are more efficient in
removing fine particulates
Cyclones have the lowest capital cost of the four general types
of particulate collectors. Costs of $0.08 to 0.10/ACFM* for
units in the capacity range of 100,000 ACFM are reported. Special
custom designed units may cost as much as $0.35/ACFM. Installa-
tion will usually add about 25 percent to the cost. These
figures are based on 1971 data (1, p.5-10).
•ACFM s Actual cubic feet per minute
643
-------
Wet Scrubbers—
In particulate collection, wet scrubbers are next in terms of
relative complexity and initial cost. There is a great variety
of individual types and configurations. The scrubbers operate on
the basic principle of confronting the particulates with
impaction targets which can be either wetted surfaces or, most
frequently, individual droplets. The efficiency of a wet
scrubber is a function of several variables. It will be higher
when particle diameter, particle density, and relative velocity
between the particle and the target droplet are high and when the
gas viscosity and target droplet size are low. Thus, to obtain
efficient particulate removal from a given stream, it is
advantageous to employ a high liquid-to-gas ratio and to produce
a high degree of atomization of the scrubbing liquid.
The efficiency of collection of particulates of a given particle
diameter and density is related to the energy consumption of the
scrubber system, as measured by the pressure drop across the
system, and is also related to the manner in which the energy
consumption is utilized (1, p. 11-11):
Collection Efficiency at Pressure Drop,
5 Microns 2 Microns cm Water
Spray Tower 9H% 81* 8
Orifice
Impingement 97 91* 8
Self-Induced
Spray 93 78 15
Venturi 99.6 97 60 to 75
The orifice impingement scrubber maintains its high collection
efficiency with low energy consumption as particle size decreases
and is roughly comparable in efficiency to the venturi in effi-
ciency at particle sizes below 1 micron, but with only 10 to 12
percent of the venturi's energy consumption. The efficiency of
644
-------
the spray tower is roughly comparable to that of the self-induced
spray but with half the energy consumption. Choice of collector
therefore must take into account the particle size distribution
of the particulates, the required collection efficiency and the
overall energy consumption.
High energy wet scrubbers (venturi scrubbers) are normally used
where:
o Fine particles must be removed at high efficiency
o Cooling is desired and moisture addition is not objection-
able
o Gaseous contaminants as well as particulates are involved
o Volumes are not extremely high (because of the relatively
higher operating cost per ACFM)
o Relatively high pressure drop is tolerable
o Contamination of the scrubbing liquid with materials removed
from the gas poses no problem
Initial cost (1971) of wet scrubbers sized for about 100,000 ACFM
ranges from $0.25 to 0.35/ACFM in carbon steel and about
$0.65/ACFM in alloy steel with cost of erection adding about 25
percent (1, p.5-13)•
Fabric Filters—
Various configurations of fabric filters are used in which porous
fabrics remove particulate from gas streams by allowing clean gas
to pass though while preventing the passage of the particles.
The buildup of particulate matter on the filter medium aids in
the particulate removal process, as shown in Figure 9-39.
Fabric filters are usually arranged as a number of cylindrical
tubes, or bags, suspended in an enclosure. Design variations
include arrangements for gas passage from the outside or from
645
-------
Figure 9-39. Operating principles of a surface type fabric
filter. Initial condition at left; filtering
effect of "cake" indicated at right.*
From Item I/ p. 5-13 in reference list
646
-------
the inside of the bags and arrangements for removal of collected
dust from the surfaces of the bags. These two principal
variations are frequently related: bags cleaned by mechanical
shaking and by high pressure air jets inside the bags usually
collect dust on the outside of the bags, while bags cleaned by
pressurizing a section of the bag enclosure or by high pressure
air jets outside the bags collect dust on the inside of the bags.
Overall, there is little advantage of one general type over the
other, with choice finally being influenced by such
considerations as rate of bag wear, ease of replacement,
auxiliary power required, etc.
Very high efficiencies can be attained with fabric filters (99 +
percent). The principal design considerations are:
o Superficial velocity (air to cloth ratio)
o Fabric resistance coefficient (permeability)
o Cake resistance coefficient
o Weight of cake per unit area
o Gas viscosity
o Means of filter cleaning (mechanical shaking, collapsing,
reverse flow, shock wave, pressure pulse, etc.)
Fabric filters are normally employed where:
o High efficiencies are desired
o Operation is above the gas dew point
o Temperatures are moderate
o Valuable material is to be collected dry
o Water availability and disposal is a problem
Initial cost of these units (1971) ranges from $0.50 to 1.20 per
ACFM, depending on the filter medium. Erected cost is reported
to be about 30 percent of the equipment cost.
647
-------
Electrostatic Precipitators—
In the electrostatic precipitator (ESP) particles are charged in
an electric field, collected on the passive collecting electrode
and removed by mechanical means to hoppers located below the
collectors. The efficiency of collection is dependent primarily
on the total surface area of collecting electrode per unit volume
of gas treated and the precipitation rate parameter. The latter
is directly proportional to field strength and particle diameter
and inversely proportional to gas viscosity. The factors which
affect field strength and ultimately the collection efficiency
are the gas density and the particle bulk resistivity. High gas
density and low particle resistivity result in high efficiency,
all other factors being equal. The particle resistivity can be
controlled by selection of the proper operating temperature or by
use of conditioning agents. The resistivity of fly ash as a
function of temperature and coal sulfur content is shown in
Figure 9-^0.
ESP's are most commonly used when:
o Very high efficiencies are required for fine materials
o Volumes of gas are very large
o Water availability and disposal are problems
o Valuable dry material is to be recovered
The purchase price (1971) of an ESP in the 100,000 ACFM capacity
range is about $0.80/ACFM while that of one ten times as large is
about $0.40/ACFM. Erection cost adds about 70 percent (1, p.5-
20).
Sources and Control of Particulate Emissions—
The major particulate emissions from a coal conversion facility
are expected to come from the following sources:
o Coal storage and reclaiming facilities, estimated to be
648
-------
E
u
I
I
o
>.
">
jK
in
cr
200 400 6OO
Temperoture,°F
Figure 9-40- Resistivity of fly ash.*
*From Item 2, p. 188 in reference list
649
-------
0.025 to 0.04 Ib/ton of coal.«
o Coal crushing, screening, and conveying operations, estimated
to be 0.05 Ib/ton of coal.*
Dust emissions from coal storage areas are not expected to have a
very strong impact on the environment because of the relatively
large particle sizes encountered. Only particles having a
diameter less than 30 microns have long range drift potential
(greater than 1,000 feet). A very small percentage of the coal
in storage is expected to be in this size range (3, p. IV-16) and
therefore, dust from this source is expected to have only a
localized impact, for the most part.
Control methods for particulates originating from handling and
preparation activities are the following (4, p.22):
o Enclosing and ventilating screening and coal fines cleaning
operations and employing wet scrubbers and bag houses to
collect particulates from the ventilation air stream
o Using covered conveyors
o Reducing the stacker conveyor height to minimize the free
fall of coal onto the storage pile
o Using water sprays containing a wetting agent at all trans-
fer points, truck dump hoppers, crushers,and screens
o Employing completely enclosed coal storage facilities
The amount of fly ash in the exhaust gases from the power boiler
is dependent on the boiler type, as shown by the following data
(3, p.IV-9) for uncontrolled bituminous coal boilers:
•By Wyoming Coal Gas Co. as reported in reference 4, p.22,
.650
-------
COAL ASH APPEARING IN BOILER EXHAUST GASES AS PARTICULATES
Boiler Type Average as Particulates
Pulverized - Dry Bottom 85$
- Wet Bottom 65
Cyclone 20
The particle size distribution of the fly ash from the uncontrol-
led bituminous coal boilers is reported (3, p.IV-9) as follows:
SIZE DISTRIBUTION OF PARTICULATES FROM UNCONTROLLED BOILERS
Particle
Size, Microns
5
10
20
40
Pulverized Type
Range*
12-32
28-56
42-79
61-93
Average*
25
42
65
81
Cyclone Type
Range*
20-72
40-96
70-99
81-100
Average*
40
65
81
92
•Weight percent less than size
Control of the fly ash in the boiler exhaust is accomplished by
cyclones followed by ESP's.
Dust from ash handling is a problem only when the ash is dry, as
in operation of some types of power boilers, since gasifier ash
is water-quenched. Handling all ash as moist solids or as a
slurry appears to be the most practical and economical means of
control of particulates.
References—
1. Lund, Herbert F., "Industrial Pollution Control Handbook."
McGraw-Hill, Inc, 1971. 900*
.Pullman Kellogg Reference File number
651
-------
2 Coughlin, R., Sarofim, A., and Weinstein, N. . "Air Pollution And
Its Control." AIChE Symposium Series, No. 126, Volume 68,
1972. 902«
3. Coal Conversion Program, Energy Supply and Environmental Co-
ordination Act (as amended). Section 2, Volume I. Federal
Energy Administration. FEA G-77/1^5, May 1977.
4. Sinor, J. , "Evaluation of Background Data Relating to New
Source Performance Standards for Lurgi Gasification." EPA
600/7-77-057, June 1977. 552»
652
-------
Control of Cooling Tower Drift
Cooling tower drift is defined as mechanically entrained water
droplets which are carried along with the air flowing through the
tower and exhausted to the atmosphere. These water droplets have
essentially the same chemical composition as the circulating
water in the cooling tower.
Drift resulting in deposition of water and its impurities on
objects in the vicinity of the tower is potentially objectionable
because corrosion problems may result, electrical equipment may
be damaged, a public nuisance may be caused and vegetation may be
damaged, especially with salt or brackish water cooling towers.
To determine the environmental significance of drift, it is first
necessary to establish the total drift emission rate as well as
the drift particle size and mass distribution. A drift
measurement system, including a cyclone separator to collect
entrained water droplets and an isokinetic sampling device, has
been developed by Ecodyne Cooling Products Company.
To minimize the impact of drift on the environment, it is
desirable to reduce its quantity. Several designs are now
available to accomplish this result, of which the system
deveJoped by Ecodyne (5, pp.10,14) is chosen as an example.
Ecodyne incorporates:
o A new drift eliminator design, the "Hi-V"
o Provision for positive sealing at structural members
o A positive drainage system for collected water droplets.
Figure 9-41 is a sketch of the Ecodyne Hi-V drift eliminator
system.
653
-------
en
I. "HI-V'PVC DRIFT ELIMINATO
2 PLYWOOD 0 t AIR SEAL.
3. STRUCTURAL TIE.
4 CAB AIR SEAL
5 STRUCTURAL COLUMN.
6 TRANSVERSE PARTITION.
7 0 E BLADE
8. DRAIN SLOTS.
9 Z t 3 0 E SUPPORT
10 RIVULET OF WATER OROPLE
EXTRACTED FROM AIRSTRE*
FOLLOWS z«30E SUPPORT
COLO WATER BASIN
AIR - FLOW
TYPICAL DE. CONSTRUCTION
Figure 9-41. Sketch of Ecodyne "Hi-V" drift eliminator system*
*From Item 5, p.10 in reference list
-------
Drift tests have been conducted recently by Ecodyne on industrial
cooling towers. Both standard towers, equipped with the two-pass
drift eliminator configurations that are typical of the industry
for the past twenty years, and towers equipped with the Ecodyne
Hi-V system were tested. The results of 22 drift tests are as
follows:
Standard Hi-V
Drift, % of Circulation
Range 0.02-0.12 0.001-0.008
Typical Value 0.05 0.004
Particle Size Range, microns 22-2,400 22-2,400
These tests indicate that drift can be reduced by over 90 percent
through use of the Ecodyne system, or one having similar effects.
The dynamic behavior of drift is a function of the original
droplet size, condensation or evaporation rates, aerodynamic and
gravitational forces, and meteorological conditions. Ecodyne
developed a computer program to evaluate the trajectory path of
drift droplets. The program takes into account the change in
drop size and position with respect to time on an incremental
basis, together with such factors as original drop size, tower
operating parameters, atmospheric conditions, fall velocity and
evaporation rates. TABLE 9-11 shows the size and mass
distribution of drift particles. Figure 9-42 shows the fall
velocity of water drops as a function of size.
An eight cell crossflow tower designed to cool 134,000 GPM of
salt water with the same chemical composition and salinity level
as the sea was chosen for an example. The plant location is
assumed to be two miles from the ocean on an estuary or bay. The
drift rate is 0.004 percent of the circulating water rate and
655
-------
TABLE 9-H.
SIZE AND MASS DISTRIBUTION OF DRIFT PARTICLES
(HI-V DRIFT ELIMINATORS)
DROPLET DIAMETER
(MICRON)
22
29
44
58
65
87
108
120
132
144
174
300
450
600
750
900
1050
1200
1350
2250
2400
% OF SAMPLE
BY NUMBER
24.0
36.0
26.0
6.3
4.0
1.4
0.67
0.43
0.28
0.26
0.65
O.I I
0.027
0.01 I
0.0055
0.0033
0.0024
0.0019
0.0016
0.00095
0.0010
% MASS BY
DROPLET SIZE
0.43
1.49
3.76
2.09
.86
.56
.43
.26
.09
.32
5.81
5.04
4.17 .
4.01
4.00
4.03
4.57
5.46
6.80
17.99
21.83
656
-------
1000
too
z
2
o
o
Ul
>
_l
10
10
100
1000
DROPLET SIZE, MICRON
Figure 9-42. Fall velocity of water drops as a function
of size.*
*From Item 5 in reference list
657
-------
the drift mass size distribution of TABLE 9-11 is used in this
example. The conditions were established as 17°C dry bulb
temperature and 50 percent relative humidity with stable
atmospheric conditions and an onshore wind of 20 miles per hour
at the plant location.
Figure 9-43 illustrates the drift dynamic behavior for this
example. The results show that 68.7 percent of the drift mass
hits the ground in the first 400 feet. Note that roughly half of
this total mass falls out in the first 150 feet. Drift
originally smaller than 100 microns in diameter represents 12
percent of the total mass and, for the atmospheric conditions
chosen, droplets originally smaller than 450 microns evaporate to
a diameter of 100 microns or less before hitting the ground.
Thus 31.3 percent of the total drift mass remains in suspension
and its subsequent behavior is evaluated using an atmospheric
dispersion model.
The dashed lines of Figure 9-43 show the trajectory history when
the atmospheric conditions are changed to -1°C dry bulb tempera-
ture and 80 percent relative humidity. Note that for this typi-
cal winter condition the droplets fall slightly closer to the
tower since the evaporation rate is diminished. Further, there is
a slight change in the total airborne drift quantity since the
450 micron droplets now hit the ground.
A parametric analysis, where relative humidity, salinity levels
and atmospheric and exit air temperatures were varied, revealed
that the dynamic behavior of drift droplets in the 400 to 700
micron size range is the most sensitive to changes in these
variables. This example shows, however, that those drift drop-
lets that hit the ground will do so in the first 500 feet with
20 mph wind.
.658
-------
tn
vo
20 MPH WIND
o
31.3 % OF DRIFT MASS GOVERNED
BY ATMOSPHERIC DISPERSION
130 174 186 203 228 270
DISTANCE TRAVELED, FEET
386
17°C,
T • *
~ -L V* f
11 PI n n n n
50% RH
80% RH
% OF DRIFT MASS
Figure 9-43. Dynamic behavior of cooling tower drift.*
*From Item 5 in reference list
-------
Evaluation of other conditions including weather extremes show
that even under the most adverse conditions all drift droplets
that will reach the ground will do so in the first 1,000 feet.
The relative significance of the airborne drift in terms of salt
concentrations must now be determined. Natural airborne salt
nuclei are generated by the bursting of air bubbles on the
surface of the sea. This is caused by wind, waves, and surf
action. Meteorologists have been actively interested in these
sea salt nuclei and their role in droplet formation in clouds and
precipitation for many years. It has been shown that wind speed,
direction, and the distance from the shore line will determine
the natural sea salt nuclei concentration levels. Typical values
are shown in Figure 9-1*1*. The air salt concentration is shown to
be directly proportional to inblowing wind speed and inversely
proportional to the distance from the shoreline. Note that the
airborne sea salt concentrations two miles from the shoreline
vary anywhere from 9 g/m3 to over 150yg/m3 for normal wind speed
variations. Obviously local plant life in the area must be
capable of withstanding these natural airborne salt levels if
they are to survive. A recent comprehensive environmental report
(7) established an important correlation between airborne salt
concentration levels and injury to vegetation. Based on field
observations it was shown that exposure of local vegetation to
airborne salt concentrations above lOC^g/m3 for several hours
would result in some foliar injury. There was no visible damage
for concentration levels below 60 g/m3 . This information
suggests that a conservative plant damage threshold level can be
established at 60y g/m ., For the example chosen, the background
natural airborne sea salt concentration is 4?u g/m3 (Figure 9-4i|)
and application of a dispersion model shows that the total of the
airborne drift plus background concentration will reach 60U g/m3
at a position approximately 2,200 feet downwind from the tower.
Therefore no plant damage is anticipated beyond this distance.
660
-------
10
60O—
500—
400—
300-
200-
o |0°-
^ 90-
5 80-
tt 70-
2 60-
Ul ___.
O 50—
1 4CH
b
to
UJ
-I
cc
I
30-
20-
10-
Q_
8-
7-
6-
5-
4-
I
4
I
12
20
28
I
36
i
44
Figure 9-44. Natural sea salt concentration in air.*
*From Item 5 in reference list
661
-------
Note also that the drift-related increase is 13 g/m3 at this
position. Figure 9-44 shows that a 13 v g/m 3 change in airborne
salt level is approximately equivalent to a 3 mph change in
average wind speed. Thus the fractional increase in airborne
salt concentrations due to drift is insignificant when compared
to normal variations caused by changes in atmospheric wind
conditions.
In summary, it can be generally concluded that cooling tower
drift effects on the environment are localized and that beyond
some reasonable distance, that is usually within the plant site
boundary, drift does not significantly affect the environment.
All field experience during the last 20 years where salt or
brackish water has been used in cooling towers supports this
general conclusion. Further, it can be concluded that the use of
modern drift eliminator systems can reduce drift losses by about
90 percent when compared to older systems.
References—
1. Wistrom, G.K. , and Ovard, J.C., "Cooling Tower Drift, Its
Measurement, Control, and Environmental Effects." Cooling
Tower Institute Annual Meeting, Houston, Jan. 1973. 893*
2. Chilton, H. , "Elimination of Carryover from Packed Towers
with Special Reference to Natural Draught Water Cooling
Towers." Trans. Institution of Chemical Engineers, Vol. 30
1952. '
3. "Assessment of Environmental Effects." General PubliG
Utilities Unpublished Report, January 1972.
•Pullman Kellogg Reference File Number
.662
-------
Miscellaneous Control Techniques
Control of Lock Hopper Vent Gases—
In the Lurgi process, as well as in certain others, coal is fed
to the gasifiers in a cyclic operation using a pressurized lock
hopper. The recommended sequence of operations is as follows:
o Fill the lock hopper to about 90 percent capacity with coal
from coal bunker
o Pressurize the lock hopper with cooled crude gas
o Feed coal from the lock hopper to the gasifier
o As the pressure tends to drop in the lock hopper, add
cooled crude gas by use of a recycle compressor or, as an
alternate, gas from the top of the gasifier can back flow
through the entering coal stream for pressure equalization
o When the lock hopper has been emptied of coal, bleed the
gas into the fuel system or return it to the process by
means of a compressor
o Displace the residual gas in the lock hopper with C02 or
nitrogen and direct this stream to the fuel system or to
incineration
o Open the lock hopper and recharge it with coal
The use of the scheme recommended prevents the discharge of
pollutants (CO, H^S, COS, hydrocarbons) to the atmosphere while
recovering the fuel value of the lock hopper pressurizing gas.
Control of Ash Quench Vent—
Ash produced in the Lurgi gasifier is discharged through the
bottom of the gasifier into a pressurized ash lock. After the
ash lock is filled with ash, the top ash lock cone valve is
closed, isolating the ash lock chamber. High pressure gases in
the ash lock at this point are mainly steam. The chamber is
vented to a close coupled direct contact condenser, where the
steam is condensed with a water spray. The bottom ash lock
663
-------
valve is then opened and the ash falls out. After the ash is
dumped, both cone valves are closed and the ash lock chamber is
repressurized with steam. The top ash lock valve is opened and
ash flow from the producer is re-established. The look dumps
approximately every 20 minutes into an ash quench system where a
mixture of water streams from the plant are added. During the
quenching process a large amount of steam containing fine ash
dust and clinkers is produced.
The recommended control method for disposal of this stream is to
send the mixture first to a wet cyclone to remove the clinkers
and then to a direct contact condenser to condense the steam and
remove fine ash particles. Along with the steam, some amount of
non-condensable gases (hydrocarbons) may be formed from organic
materials in the quench water and unreacted carbon in the ash.
The quantity and composition of this gas stream is not known
precisely; however, data from SASOL reveals that the gas is
mostly steam. Nevertheless, final treatment by incineration is
recommended.
Control of Miscellaneous Leaks—
In any processing plant, certain leaks will arise from valve
stems, pump packing glands and mechanical seals, flanges, relief
valves, instrument and piping connections,and compressor seals.
In most cases, emissions from these sources can be reduced or
eliminated by proper plant design and thorough maintenance
programs.
Some specific methods of control are the following:
o Provide enclosures for valve stems, pump seals, compressor
seals, flanges, and instrument and piping connections.
Escaping gases can be collected in a central system, or
systems, and routed to the incinerator
664
-------
o Provide a closed relief valve 'system which vents to the
incinerator
o Provide a positive pressure sealing system for compressors
in which an inert gas, such as nitrogen, can provide posi-
tive pressure in the outer seal chamber. Process gas which
passes the inner seal mixes with the inert gas and is bled
to the incinerator through an intermediate seal chamber
o Provide covers and collection systems for escaping hydro-
carbon vapors from open systems such as API separators
Control of Emissions from Storage Vessels—
In a coal conversion facility a number of products, byproducts
and chemicals must be stored at the site. Some of these are tar,
tar oils, phenols, ammonia, naphtha, sulfur, sodium hydroxide,
sulfuric acid, methanol, isopropyl ether, Stretford solution
components, Selexol solvent, and water treating chemicals.
Emissions from storage will consist of tank breathing, leaks,
spills, venting of tanks while filling and vaporization (boil-
off) of volatile materials.
Liquid spills are handled best by providing containing dikes
around the storage vessels to prevent spreading of spilled liquid
while recovery or disposal operations are carried out.
The means of control for emissions depends on the vapor source:
o Vapor collection systems for atmospheric storage tanks
should be so designed that then filling these tanks the
displaced vapors will be diverted to the fuel system,
incinerator or recovery systems. Secondary wiper seals to
reduce hydrocarbon losses from floating roof tanks should
be provided
o Refrigeration systems will control boil-off from cryogenic
665
-------
or volatile liquid storage
o Scrubbing systems which use low volatility solvents to
absorb escaping vapors may be provided
o Adsorption systems using activated carbon or other suitable
materials may be provided as first stage or backup methods
to prevent the escape of storage vapors.
Cost of Emission Control—
It is difficult, if not impossible, to quantify the cost of emis-
sion control for the miscellaneous sources previously mentioned.
If a choice existed, control measures would be applied first to
those sources which contributed the most emissions. It ia
interesting to note that a DuPont study (1) concluded that:
o Almost 75 percent of inventoried organic emissions came
from sources which emit more that 500 Ib/hr and these could
be reduced by 85 percent for about 10 percent of the in-
vestment cost that would be needed for controlling all
sources which emit more than 3 Ib/hr. See Figure 9-M5.
o About 95 percent of emissions come from sources which emit
more than 23 Ib/hr. Figure 9-45 indicates that these could
be reduced by 85 percent for about 50 percent of the in.
vestment cost that would be needed for controlling all
sources which emit more than 3 Ib/hr.
o Figure 9-M6 demonstrates that the operating cost of emis-
sion control on a unit basis begins to increase rapidly for
those sources which emit less than 100 Ib/hr. It increases
very rapidly for sources emitting less than 20 to 30
Ib/hr.
One may conclude from the above information that:
o First priority for control should be given to the source
with the greatest emission rates. This conclusion i
reached from consideration of both environmental impact and
economics.
666
-------
BASIS: 100% INVESTMENT - $350MM (1976 DOLLARS)
Sources > 3 Ib/hr or 15 Ib/day
100 _
80 .
60 -
40 .
20 .
100
500 100 3 Ibs/hr
ALL SOURCES GREATER THAN INDICATED EMISSION RATE CONTROLLED
TO 85%
Figure 9-45. Investment vs. emission reduction.*
*From Item 1 in reference list
667
-------
2.0
1.5 —
oo
S 1.0
0.5 —
40
I
_L
80 120 160 200
SOURCE EMISSION RATE (LBS VOC/HR)
240
200
320
Figure 9-46. Organic abatement operating cost.*
*From Item 1 in reference list.
-------
Means to reduce the unit cost of control for small emis-
sions should be developed, such as central collection sys-
tems which tie in to a number of emission sources.
Some consideration might be given to relaxing regulations
as far as small emissions are concerned. An emission
source of 200 Ib/hr which is reduced by 85 percent results
in a final emission of 30 Ib/hr. This is equal to 10 small
sources, each emitting 3 Ib/hr. Requiring emission control
systems for all 10 small sources may not be entirely
practical because of the limited effectiveness and high
cost of those systems.
Reference—
1. Kittleman, T. , and Akell, R., "Cost of Alternative Organic
Emission Control Regulations." AIChE Meeting, New York, Nov
1977. 860»
•Pullman Kellogg Reference File number
669
-------
INTEGRATED SCHEMES FOR EMISSIONS CONTROL
The Lurgi Dry Ash process was selected as the base gasification
case for study of integrated schemes for emissions control. As
previously mentioned, the block flow diagram and material balance
of Figure 9-1 was assembled from the conceptual designs of C. F.
Braun, Cameron Engineers and Pullman Kellogg for operation of the
Lurgi process on western, low sulfur coal. From this flow
diagram and material balance the base case flow diagrams for the
sections of the plant were developed as shown in Figures 9-2
through 9-9, then the overall feed and product weight balance of
Figure 9-10 and the gaseous emission streams shown in Figure 9-11
were calculated. These flow diagrams, calculations and material
balances were primarily directed toward establishment of the
operating characteristics of the sulfur recovery unit, the
composition of the offgas stream from sulfur recovery and the
required operating characteristics of the unit for control of the
sulfur emissions from the coal conversion plant. The overall
sulfur balance for the Lurgi base case is shown in Figure 9-M7.
The changes in the flows and the material balance when the Lurgi
process is fed with high sulfur coal are shown in Figure 9-U8.
The Lurgi Dry Ash process produces phenols, oils and tars (p/o/t)
that are separated from the gas stream and either processed
further for sale or are sent to an incinerator/boiler. Because
of the quantity of materials that must be disposed of by
incineration, the incinerator/boiler is an important part of the
coal conversion plant, its operation is closely integrated with
the sulfur recovery unit and treatment of its offgases provides a
second source of product sulfur.
On the other hand, the high temperature gasification processes
exemplified by the Bi-Gas process, produce little or no p/o/t
and the importance of the incinerator/boiler as a means of waste
670
-------
en
Figure 9-47.- Sulfur balance : Lurgi gasification base case with
low sulfur coal. (Sulfur flows in tons per day.)
-------
STE«M
ro
1 —
MIS
IT
d1
f
iEAVON
UNIT
3PFGAS
0.02
VENT
HATER ^
.16
SULFUR
CONTAMINATED LIQUOR
Figure 9-48.- Sulfur balance : Lurgi gasification with high sulfur coal,
(Sulfur flows in tons per day.)
-------
disposal is reduced. Process steam .is raised more from coal and
less from waste, the overall sulfur balance changes and the de-
mands on the sulfur recovery and emissions control units change.
The overall sulfur balance for Bi-Gas operation was based on the
C. F. Braun conceptual design for operation with low sulfur coal
and is shown in Figure 9-49. The changes in the sulfur balance
when high sulfur coal is fed to the process are illustrated in
Figure 9-50.
For liquefaction, the Ralph M. Parsons conceptual design for the
SRC II process was selected as representative. As previously
discussed, the liquefaction flow diagram and material balance are
shown in Figure 9-12 and those for sections of the plant are
shown in Figures 9-13 through 9-20. From these diagrams the
overall feed and product weight balance shown in Figure 9-21 was
calculated and the gaseous emission streams shown in Figure 9-22
were established. As in the gasification processes, calculation
of the material balances was primarily directed toward deter-
mining the operating characteristics of the sulfur recovery unit,
the offgas stream composition and the demands on the unit for
control of the sulfur emissions. The overall sulfur balance for
liquefaction is shown in Figure 9-51.
The properties of the low and high sulfur coals that are con-
sidered as the feeds to the illustrated processes are shown in
TABLE 9-12.
Incineration and Steam Generation
Processes Producing Phenols, Oils and Tars: The Base Case—
In the Lurgi process, chosen as being typical of those conversion
processes producing phenols, oils and tars (p/o/t) in the gasi-
fier, steam is required for the process and for generation of the
electric power (about 350,000 horsepower for production of
673
-------
9727 (116.7)
Figure 9-49.- Sulfur balance: Bi-Gas process with low sulfur coal.
-------
en
~j
01
Figure 9-50.- Sulfur balance: Bi-Gas gasification with high sulfur coal.
-------
ESP
370.947
SULFUR
REMOVAL
1 .WASTE GAS
' ,
»
SULFUR
370.914
FUEL GAS
0.007
SOUR HATER TO TREAT
Figure 9-51.- Sulfur balance: SRC-II liquefaction.
-------
TABLE 9-12. COAL PROPERTIES
To Gasification
Proximate Analysis,
As Received, Wt$
Moisture
Volatile Matter
Fixed Carbon
Ash
Ultimate Analysis (dry), WtJ
Carbon
Hydrogen
Nitrogen
Oxygen
Sulfur
Ash
Western
Coal
22.0
29.4
42.6
6.0
100.0
67.70
4.61
0.85
18.46
0.66
7.72
100.00
Eastern
Coal
6.0
31.9
51.5
10.6
100.0
71.50
5.02
1.23
6.53
4.42
11.30
100.00
To Liquefaction,
Midwest Coal
69.06
4.73
1.34
8.94
3.80
12.13
100.00
Heating Value of Dry Coal
Btu/lb (HHV)
Heating Value of Coal as
Received, Btu/lb (HHV)
11,290 13,190
8,800 12,400
12,460
12,125
677
-------
250 billion Btu/day of SNG) required for the conversion plant
complex. The steam generator is considered as a process auxili-
ary in most process designs and is usually included in the off-
site section of the plant complex. In this illustrative Lurgi
case, however, with operation on western coal, the heating value
of the liquid byproduct and waste streams, and of the waste gas
streams, amounts to about 58 percent of the total energy required
for steam and power generation. Therefore, the steam generator
may be considered as an incinerator and as a part of the conver-
sion process within battery limits. The remaining 42 percent of
the energy requirement is supplied by part of the total plant
coal feed in the Lurgi design, but alternately may be supplied by
other fuels.
In addition to providing a useful and economical means of dispos-
al for combustible liquid and gaseous byproducts and wastes, the
incinerator/boiler system aids in reducing or eliminating emis-
sions of:
o Particulates from coal firing
o CO, hydrocarbons (CH4, C2H4, C2H6) and ammonia from waste
gas streams
o Sulfur from gas, liquid and solid (coal) feeds. Sulfur and
sulfur compounds are oxidized to S02 (and S03) and then are
removed in the flue gas desulfurization (FGD) unit. The
FGD unit will be discussed later
o Nitrogen oxides (NOX)
To illustrate the integration of the incinerator into the proceaa
scheme,., the design conditions of the incinerator unit were taken
as follows:
Heat Input 4678.5 MM Btu/Hr (HHV)
Steam Flow 2,844,000 Lb/Hr
Steam Pressure 1,500 psig at superheater outlet
.678
-------
Steam Temperature 910°F psig at superheater outlet
F. W. Temperature 456°F
Stack Temperature 300-350°F
Air Temperature 95°F
The design plan contemplates burning a combination of fuels in
the boiler(s). Miscellaneous gas streams are to be incinerated
to destroy Hj, CO, hydrocarbons,and ammonia and to oxidize sulfur
compounds. Liquid byproducts are to be burned, together with
coal fines to supply the remaining heat.
The gaseous fuels are described in TABLE 9-13. The liquid fuels
are described in TABLE 9-14. Additional fuel required is com-
prised of coal fines with the ultimate analysis (dry basis) shown
in TABLE 9-12 and containing 22 percent moisture.
The net coal required by the boilers is determined by deduction
to be about 175,190 Ib/hr (dry) or 224,603 Ib/hr (wet) which
furnished 1,977. 9 MM Btu/hr (HHV).
The total input to the boilers is summarized as follows:
Gas Liquid Coal Total
Lb/hr 1,640,947 164,282 224,603 2,029,832
MM BtuXhr(HHV) 144.6 2,556.0 1,977.9 4,678.5
Heat input, % 3.091 54.633 42.276 100.000
Combustion air required at 15 percent excess is 160,160 moles per
hour dry plus 3,994 moles of water. Total weight of air is
4,711,627 Ibs/hr. The calculated flue gas composition is shown
in TABLE 9-15.
Information received from Combustion Engineering indicates that
three boilers are needed to satisfy the demand. Each of the units
679
-------
00
O
TABLE 9~13. GASEOUS FUELS TO INCINERATOR
C02 Stripper
H2
CO
co2
CH4
C2«6
NH3
H2S
COS
cs2
Dry
Total
Tot*l Ib/hr
MM Btu/hr (HHV)
Exp. Gas
MPH
-
2.50
1356.51
3.12
9.24
—
1371.37
112.68
1484.05
62,165
3.73
Acid Gas
MPH
-
-
194.73
8.22
—
202.95
492.37
695.32
17,720
1.98
Flash C02
MPH
24.8
0.8
33.9
28,331.0
197.9
84.2
112.1
3.1
—
28,787.8
44.1
28,831.9
1,257,785
134.7
Overhead
MPH
0.1
1830.0
0.3
5630.8
0.5
1.3
1.7
0.8
0.1
-
7465.6
24.8
7490.4
299,649
2.30
Excess HjS*
MPH
0.02
0.03
62.69
0.18
0.09
0.12
0.10
6.85
0.03
0.03
70.14
6.91
77.05
3628
1.90
Total
MPH
24.9
1830.8
36.73
35,575.73
201.70
85.59
113.92
0.9
24.41
3-13
0.03
37,897.8
680.9
37,578.7
1,640,947
144.6
•Return from the citrate process flue gas desulfurization system.
-------
TABLE 9-14. LIQUID FUELS TO INCINERATOR
Compound Tar Tar Oil Phenols Naphtha Total
Naphtha (CgHg"1") - 15,622 15,622
Phenols - - 11,260 - 11,260
Tar Oils (^C-^H^) __ 48,600 - - 48,600
Tars (/V'C24H50) 88,800 - - - 88,800
Total, Lbs/hr 88,800 48V600 11,260 15.622 164,282
MM Btu/hr (HHV) - 1,376.4 753.3 156.3 270.0 2,556.0
Nitrogen, Wt% 0.85 0.85 - - 0.71
Total Nitrogen
Lbs/hr 754.8 413.1 - - 1,167.9
Sulfur, Wt% 0.46 0.26 - - 0.33
Total Sulfur,
Lbs/hr 411.8 125.2 - - 537.0
681
-------
TABLE 9-15. INCINERATOR FLUE GAS COMPOSITION
MPH
N2
°2
A
CO,
so2
so3
MOX.
Dry
H2°
Total MPH
Total Lbs/hr
SCFM § 60°F
ACFM € 300°F
13 psla
126,781.8
4,277.6
1,601.6
57,726.3
78.6
1.6
54.75
190,522.25
23.581.75
214,104.00
6,724,608
1,3^4,350
2,237,675
MOL % (Wet)
100.000
NO
N02
Total
% (v)
95
5
MPH
52.00
2.75
54.75
Mol %
0.024
0.001
0.025
Lbs/hr
1,560
126
1,686
682
-------
is equipped with the normal complement of components including
firebox, burners, superheater, air heater, economizer, cyclones,
and electrostatic precipitators. Estimated efficiency of the
units is 86 percent. Combustion Engineering states that special
design can accommodate firing of the liquid fuels and coal such
that these fuels will pose no combustion problems. They further
state that the waste gas streams should be introduced into the
upper furnace to avoid interfering with the combustion of liquid
fuels and coal.
Particulate Emission—Ash content of the western coal that is
used in the Braun study amounts to 7.72 percent of 175,190 Ibs/hr
of coal feed, or 13,525 Ibs/hr. If 80 percent (2, p.IV-9) is
assumed to be flyash, with the remaining 20 percent being bottom
ash, then 10,820 Ibs/hr leave the boiler and enter the cyclone
separators. The expected particulate removal efficiency of 75
percent (3, p. 202-205; 4,p.23-20) results in 2705 Ibs/hr
entering the electrostatic precipitators (ESP's).
The most stringent standards for particulates from combustion are
those of New Mexico:
Coal 0.05 Ib/MM Btu
Oil 0.005 Ib/MM Btu
Gas 0.03 Ib/MM Btu
Based on the relative heat inputs of coal, oil, and gas, the
allowable emission becomes:
E = 0.423(0.05) + 0.546(0.005) + 0.031(0.03)
= 0.0248 Ib/MM Btu
or 0.0248 (4678.5) = 116 Ibs/hr
The ESP's need to remove 2,589 Ibs/hr. This amounts to a 95.7
percent removal efficiency. These units commonly remove
683
-------
99 percent or more (3,p.l88; 4,p.23-21) of the entering parti-
culate matter. Therefore, no problem is foreseen in meeting or
exceeding the most stringent of the present pollution standards.
If 99 percent removal is attained, the emission of particulates
would be only 27.1 Ibs/hr or 0.00578 Ib/MM Btu, which is less
than 25 percent of the most stringent present standard.
It should be mentioned that additional particulate removal will
be accomplished in the spray coolers preceding the SO absorbers
in the FGD unit. Therefore, the ultimate particulate emission
will be essentially nil for all practical purposes. Conversely,
consideration may be given to elimination of the ESP's when FGD
is included in the flowsheet.
Flyash from pulverized coal boilers normally ranges in size from
1 to 200 microns, with most of the particles in the size range of
5 to 40 microns (2,p.IV-9, 4, p.23-10). Electrostatic precipita-
tors are effective in removing particles as small as 0.01 micron
and as large as 40 to 50 microns (4,p.23-10).
For the low sulfur coal cases, the flyash resistivity will be, at
300 to 350°F, about 3 x 1011 ohm-cm. Therefore, the ESP's must
be located upstream of the air heaters where the temperature is
around 600°F so that the resistivity becomes sufficienty low (1 x
109 ohm-cm) to sustain efficient particle removal (3,p.188).
CO Emission—The most stringent standard for CO emission from
fuel burning equipment is 200 ppm (Illinois) . The expected un-
controlled CO emissions from coal, oil and gas fired units are
given in the following table (2,p.IV-5):
684
-------
0.042
0.021
0.017
1,977.9
2,556.0
144.6
83.1
53.7
2.4
Lb/MM Btu MM Btu/Hr Lb/Hr
Coal
Oil
Gas
139.2
Thus, the total expected CO emission is 139.2 Ib/hr or 4.97 MPH.
Expressed as ppm by volume this becomes 23.2 ppm (v), or about 10
percent of the most stringent standard.
Hydrocarbon Emission—Illinois has the most stringent standard
for hydrocarbon emission from fuel burning equipment. This
standard states that incineration must reduce hydrocarbon
emission to 10 ppm CH equivalent or less or must convert 85
percent of the hydrocarbons to CO and water. The expected
uncontrolled emissions from coal, oil and gas fired units are
given in the following table (2,p.IV-5):
Lb/MM Btu MM Btu/Hr Lb/Hr
Coal 0.013 1,977.9 25.7
Oil 0.014 2,556.0 35.8
Gas 0.001 144.6 0.1
61.6
Thus the expected hydrocarbon emission is 61.6 Ibs/hr or 3.84 MPH
of CH^ equivalent, or 17.9 ppm (v). While this figure exceeds
the previously mentioned maximum of 10 ppm, the standard is still
met by converting the necessary percentage of hydrocarbons to CC^
and H20:
HC converted = (401.2-3.84)(100)/401.2 = 99.0$
This exceeds the required 85 percent conversion of the Illinois
standards.
685
-------
Ammonia Emission — New Mexico's standard for ammonia emission from
gasification plants is 25 ppm maximum. Even if none of the
ammonia entering the incinerater/boiler is burned, this standard
will be met easily. The maximum ammonia concentration in the
stack, assuming no combustion, will be 4 . 2 ppm (v) . Since part
or all of the ammonia will be burned, the actual ammonia
concentration in the stack gases will be considerably less than
4.2 ppm.
Sulfur Emission — Sulfur compounds enter the incinerator/boiler as
H2S, COS, and CS2 in the gas streams and as organic sulfur com-
pounds in the tar and tar oil. The coal contains pyritic, organ-
ic and sulfate sulfur. The sulfur compounds oxidize during in-
cineration and, as determined in Pullman Kellogg1 s experience, 98
percent are converted to S02 and 2 percent to S03.
If there were no FGD, the incinerator offgases would contain 71.8
moles of S02 and 1.5 moles of SO 3< The emission, therefore
would be 0.98 Ib/MM Btu. This emission far exceeds the most
stringent standard of 0.2 Ib/MM Btu (Okla.) for gas fired equip-
ment, of 0.25 Ib/MM Btu (Ky.) for combination fired equipment or
of 0.2 Ib/MM Btu (Wy.) for coal fired equipment. The FGD step
will be described later.
NO Emission — The most stringent standard for NO emissions is
based on New Mexico's regulations which state that maximum
emissions of NOV shall be:
J^
NO, Ib/MM Btu
Coal 0.45
Oil 0.30
Gas 0.20
.686
-------
Nitrogen oxides are formed by oxidation of organic nitrogen in
the fuel (fuel NO ) and by oxidation of nitrogen in the
Ji
combustion air (thermal NO ).
J\
Typically, about 30 to 60 percent of the fuel nitrogen will be
converted to NO (3,P«55;6,p.96;7,p.9ff)• Therefore, it is clear
J^
that reducing the nitrogen content of the fuel will be advan-
tageous. Generally speaking, NO formation is a function of
X
temperature (being greater at higher temperatures), oxygen avail-
ability and, as mentioned above, the nitrogen content of the
fuel. Boiler modifications which lower the flame temperature and
reduce the oxygen availability result in lower NOV formation.
J\
These modifications can be:
o Two-stage combustion
o Low excess air firing
o Flue gas recirculation
Expected NO emissions for both uncontrolled and controlled
H
boiler conditions are shown below (2,p.IH-55; 5,p.21; 6,p.95;
7,P.5):
NOyin ppm
Uncontrolled With Boiler Controls
Coal 500 370
Oil 280 150-210
Gas 200 85-110
The allowable NOx emission, based on New Mexico standards, for
the Lurgi low sulfur case becomes:
Weighted Average NOX = 0.423(0.45) + 0.546(0.3) * 0.031 (0.2)
= 0.3604 Ib/MM Btu
687
-------
Based on a heat input of4,678.5 MM Btu/hr, the allowable NCL
emission isl,686.1 ib/hr or 54.75 MPH. Expressed in concentra-
tion by volume this becomes 256 ppm(v).
The expected N^ emission with no boiler controls is calculated
as follows:
NOV = .423 (500) + .546(280) + .031(200)
= 371 ppm(v) (which is above the New Mexico standards)
x
Using similar calculation techniques, the use of boiler controls
can be shown to reduce this figure to 241 to 275 ppm.
In the case under study, the nitrogen content of the liquid fuels
isl,168lb/hr and that of the coal fuel isl, 489 Ib/hr for a total
of 2,657 Ib/hr. With no controls, the expected NOX emission is
projected to be 371 ppm or 79.3 MPH or 2,443 Ib/hr of NO . in
J\
"Processes and Techniques for Control of Nitrogen Oxides" it was
shown that about 30 percent of the fuel nitrogen is converted to
or 797 Ib/hr of N, equivalent to 1,752 Ib/hr of NO . Thermal
= 2,443-1,752 = 691 Ib/hr.
If 80 percent of the nitrogen in the liquid fuel is removed by
hydrodenitrogenation (8 , p. 14-21 , 140) , then a substantial reduc-
tion in NOX emissions can be expected. The nitrogen content of
the fuel oil becomes 234 Ib/hr, the coal nitrogen remains un-
changed and the total fuel nitrogen becomesl,723 Ib/hr. Conver-
sion of 30 percent of the fuel nitrogen, or 517 Ib/hr, to NO
yields NOx from fuel of 1,136 Ib/hr. The thermal NO^ remain's
constant at 691 Ib/hr and the total NOX emission becomes 1,827
Ib/hr, or 59.3 MPH or 277 ppm(v), a reduction of 25.3 percent
from the uncontrolled emissions level.
Therefore, the reduction of the nitrogen content of the liquid
688
-------
fuels is shown to be about as effe-ctive in reducing total NO
A,
emission as are boiler modifications. If both techniques are
employed and it is assumed that the percentage reduction of NO
J^
by boiler modifications is constant at 26 to 35 percent, then the
NOX levels may be reduced to the following:
NO Emission, ppm(v)
Maximum allowed 256
With no control 371
With boiler modifications 241-275
Remove 80$ of liquid fuel nitrogen 277
Combined boiler modifications and
N removal 180-205
!»'
It may be concluded that the most stringent present standards may
be met by using boiler modifications and/or liquid fuel denitro-
genation techniques. However, if standards are lowered in the
future, more complete NO removal will be required.
J\
TABLE 9-16, which appeared in "Chemical Engineering" for February
14, 1978, indicates the possible goals for NO emissions in 1980
Ji
and 1985. The projected 1980 goals may be met by the above
mentioned combination of techniques. This is not so for the 1985
goals. The maximum allowable NO emissions, based on the
relative heat inputs, is projected to be:
NOX, Max. 1985=0.423(100) + 0.546(90) * .03K50)=93 ppm(v)
This figure is about one half of that attainable with both boiler
controls and liquid fuel denitrogenation. Therefore, other meth-
ods for NOV control will be needed if these stringent standards
J^
are imposed. There are 48 flue gas denitrogenation processes
listed in the literature with 42 being described in detail in a
report (7,p.XIV, XV). These processes are generally described
689
-------
TABLE 9-16. POSSIBLE FUTURE GOALS FOR NOV EMISSIONS
ppm NOy at 3% excess 0
Source Current technology 1980 goal 1985 goal
Utility boilers
Gas 150 (a) 100 50
Oil 225 (a) 150 90
Coal 550 (a) 200 (c) 100
Industrial boilers
Gas 150 80 50
Residual Oil 325 125 90
Coal 450 150 (c) 100
Reciprocating engines
Spark ignition-gas 3,000 1,200 (d) 400
Compression ignition-oil 2,500 1,200 (d) 800
Gas turbines
Gas 400L150 (b)] 75 ^
Oil 600[225 (b)] 125 (d) 25
(a) Current NSPS.
(b) Estimated achievable with wet control technology.
(c) Developed and field-applied technology.
(d) Developed technology.
Source: EPA Combustion Research Branch (Research Triangle Park
N.C.).
690
-------
as being dry or wet. There are 5 commercial processes (all dry)
which have operated in 50 MW or larger boilers. Of these, the
following processes have been chosen for further evaluation:
o UOP/Shell Copper Oxide
o Hitachi Zosen
Both processes use ammonia to reduce nitric oxide to molecular
nitrogen as shown below:
6 N0(g) + 4 NH3(g) * 5N2(g) + 6 H20(g)
Each of the processes is capable of NO removal in excess of 90
percent. Therefore, the NO emission in the base case without
A
controls could be reduced to about 35 to MO ppm using these
processes.
Processes Producing Phenols, Oils,and Tars: The Alternate Case—
Incineration and steam generation in Lurgi gasification operating
with high sulfur coal feed are similar in operation to the base
case operation with low sulfur coal but different in the demands
on the emission control techniques.
The analysis of the high sulfur coal is given in TABLE 9-12. It
is assumed for this illustration of techniques that the required
heat input from coal is the same as in the base case. It is also
assumed that with the exception of sulfur content the gas and
liquid streams would not change significantly in composition and
that the flow rates and heating values would be the same as those
in the base case. These simplifying assumptions result in the
following total heat input to the boilers:
691
-------
_ Gas __ Liquid Coal Total
Lbs/hr 1,640,947 164,282 159,396 1,964, 62!
MM Btu/hr (HHV) 144.6 2,556.0 1,977.9 4/678.5
Heat Input, % 3.091 54.633 42.276 100.000
Particulate Emission — The ash content of the eastern coal is 11.3
percent of the 149,832 Ib/hr of dry coal feed, or 16,931 Ib/hr.
Flyash at 80 percent of the total ash amounts to 13,545 Ib/hr
leaving the boiler. Cyclone removal efficiency of 75 percent
results in 3, 386 Ib/hr entering the ESP's.
Since the relative heat inputs from gas, oil, and coal are the
same as in the base case, the allowable particulate emission is
the same at 116 Ib/hr. The ESP's need to remove 3,270ib/hr for a
removal efficiency of 96.6 percent. As in the base case, no
problem is foreseen in meeting or exceeding the most stringent of
the present pollution standards. Further, if 99 percent
particulate removal is attained in the ESP's, the particulate
emission would be 33.9 Ib/hr, or 0.00725 Ib/MM Btu, which is less
than 30 percent of the most stringent present standard.
previously mentioned, the spray coolers that precede the FGD
unit will remove virtually all remaining particulates from the
gas stream and therefore the final particulate emission will be
essentially zero. And conversely, elimination of the ESP's may
be considered when FGD is to be used.
The resistivity of the flyash from high sulfur coal combustion is
about 1 x 109 ohm-cm at 350°F. This is low enough to allow
installation of the ESP's after the air heaters (3, p. 188).
CO, Hydrocarbon, and Ammonia Emi33ions--A3 in the base case, no
problem is seen in meeting present standards.
692
-------
MPH
206.71
162.74
96.00
Lbs/hr
6,628
5,218
3,078
MPH
206.71
117.27
96.00
Lbs/hr
6,628
3,760
3.078
Sulfur Emission--Sulfur compounds in the incinerator feed are
estimated to be, expressed as sulfur:
With FGD Without FGD
MPt
In coal
In gas
In liquid
Total sulfur feed 465.45 14,924 419.98 13,466
If there were no FGD and 90 percent of the sulfur is converted to
SO , the resulting SO content of the flue gas if 5.64 Ibs/MM Btu,
far exceeding the most stringent standard. Flue gas desulfuriza-
tion (FGD) is required. This process will be described later.
MO Emissions--With respect to NOV emissions, the incinerator
—rr *
flue gas is not expected to be significantly different from that
of the low sulfur coal, therefore the same arguments are assumed
to apply in both cases, leading to the same solutions.
Processes Producing No Phenols, Oils,or Tars—
The Bi-Gas process was selected as being typical of those conver-
sion processes producing no phenols, oils, or tars in the gasi-
fier. In this process the liquid byproduct and waste streams are
eliminated and the waste gas streams are reduced drastically. As
a result, the steam generator operates as a coal-fed boiler with
CO flash gas and C00 stripper overhead from the acid gas removal
2
section as the only significant gas streams to be considered.
The required heat input to the boiler is 3,588 MMBtu/hr, supplied
by 317,803 Ib/hr of dry western coal or 272,025 Ib/hr of dry
eastern coal.
693
-------
Particulate Emission—The most stringent standard for particu-
lates is 0.05 Ib/MM Btu (N. Mexico). The allowable emission
therefore, is 179.4 Ib/hr from the coal-fired boiler. Combustion
of the coals is expected to yield 19,627 Ib/hr of flyash from
western coal and 24,591 Ib/hr from eastern coal. Cyclones should
reduce the flyash loading of the flue gas to 4,907 Ib/hr and 6,149
Ib/hr, respectively, for the two coals. ESP's following the
cyclones would be required to operate at collection efficiencies
of 96.3 percent and 97.1 percent for the two coals, both of which
are well within the demonstrated capabilities of ESP's.
As remarked in the discussion of Lurgi operation, the FGD scrub-
ber following the ESP's should reduce particulates to essentially
nil, or the FGD scrubber may eliminate the need for ESP's.
CO Emission--The expected uncontrolled CO emission from coal
fired boilers is 0.042 Ib/MM Btu or 150.7 Ib/hr or 5.38 MPH for
eastern coal. The CO content of the flue gas is therefore 46. q
ppm(v) and 48.2 ppm(v), respectively, for the two coals, or less
than 25 percent of the current most stringent standard (Illinois)
of 200 ppm(v).
Hydrocarbon Emission—Standards for hydrocarbon emissions do not
apply to fuel burning equipment (boilers). It is noteworthy
however, that the expected hydrocarbon emission from coal fired
boilers is 0.013 Ib/MM Btu and thus the expected uncontrolled
hydrocarbon from western coal combustion is 46.6 Ibs/hr or 2.Qp
MPH of CH equivalent or 25.4 ppm(v). For eastern coal th
expected hydrocarbon is 26.1 ppra(v) of CIL equivalent.
Sulfur Emission--Sulfur compounds in the boiler feed an
estimated to be, expressed as sulfur:
694
-------
Western Coal
In coal
In gas
Total
With
MPH
65.40
10.04
75.44
With
MPH
374.99
63.01
438.00
FGD
Lbs/hr
2,097
322
2,419
Eastern
FGD
Lbs/hr
12,023.
2,020.
14,043.
Without
FGD
MPH Lbs/hr
65.40 2,
3.62
69.02 2,
Coal
Without
MPH
5 374.99 12
4 20.33
9 395.32 12
097
116
213
FGD
Lbs/hr
.023.
652
,675.
5
5
In coal
In gas
Total
Without FGD and assuming that 98 percent of the sulfur is con-
verted to SO , the resulting SO content of the flue gases is
1.21 Ib/MM Btu for western coal and 6.92 Ib/MM Btu for eastern
coal. Both of these values exceed the most stringent standard of
0.2 Ibs/MM Btu (Wy.) for coal fired equipment, and FGD is needed.
Liquefaction—
In the SRC II liquefaction conceptual design of Ralph M. Parsons,
the fuel for steam and power generation is gas produced in a
slagging Bi-Gas gasifier together with tail gas from the sulfur
removal system in the process gasifier section.
The boiler is designed with staged combustion and operates with
low excess air to minimize NO formation. Estimated emission is
A
100 Ibs/hr or 50 ppm(v) of NO , which is less than the most
stringent standard.
Total S0? emissions from the power boilers are estimated at 0.001
lb/ MM Btu, far less than the most stringent standard.
695
-------
Since the boilers are fired with clean gas, no particulates are
expected.
CO emissions are estimated to be 1U7 ppm with application of the
50 percent excess air correction, less than the 200 ppm
(corrected) most stringent standard (Illinois).
References—
1. Detman, R., "Factored Estimates for Western Coal Commercial
Concepts." FE-2240-5, October 1976.295*
2. Coal Conversion Program. Energy Supply and Environmental
Coordination Act (as amended), Section 2., Volume 1. 847«
3. "Air Pollution and Its Control." AIChE Symposium. Series
126, Volume 68, 1972. 902»
4. Lund, Herbert F., "Industrial Pollution Control Handbook."
1971. 90o«
5. Do, N. Loan, and Hunter, W.D. (Pullman Kellogg), "NO Control
J\.
Technology." Report No. RD-77-1342, September 1977 (Confi-
dential)
6. Siddiqi, Aziz, Tenini, J.W., and Killion, L.D., "Control NO
Emissions from Fixed Fireboxes." Hydrocarbon Processing,
October 1976. 578*
7. Faucett, H. L., Maxwell, J. D., Burnett, T. A., "Technical
Assessment of NO Removal Processes for Utility Application."
November, 1977.
•Pullman Kellogg Reference File number
696
-------
8. Satchell, D. P., "Development of. a Process for Producing an
Ashless, Low-Sulfur Fuel From Coal." Volume IV - Product
Studies - Part 6 - "Hydrodenitrogenation of a Coal Derived
Liquid". 232«
Flue Gas Desulfurization (FGD)
In the preceding discussion of incinerator/boilers in coal
conversion plants the necessity for desulfurization of flue
gases from gasification process boilers was demonstrated.
Since the SRC II process apparently, according to Parsons,
produces very little S02 as flue gas and no FGD is needed,
the discussion that follows will be confined to FGD in
gasification.
Consideration of regenerable vs. non-regenerable FGD pro-
cesses led to the conclusion that production of sulfur for
sale was more attractive than scrubbing the flue gases with
limestone and disposing of the sludge. Consequently, regen-
erable processes were studied and the U.S. Bureau of Mines'
citrate process was selected for integration into the gasi-
fication flowsheets.
With reference to the sulfur balances in Figures 9-47, 9-48,
9-49 and 9-50 the performance of the citrate process FGD
scrubber is shown in TABLE 9-17.
For Lurgi gasification, the coal contains about 49 percent of
the total sulfur in the feedstock to the incinerator/boiler,
whereas for Bi-Gas gasification the coal contains nearly 95
per cent of the total sulfur in the feedstocks to the boiler.
Since, in the Lurgi case, the liquid feedstocks have a higher
heating value than the coal, the net result is that the S02
feed to the scrubber, per million Btu, is about 81.5 percent
697
-------
TABLE 9-17. FLUE GAS DESaLFURIZATION WITH THE CITRATE PROCESS
Lurgi
Bi-Gas
To scrubber*
as S, Ibs/hr
as S02, Ibs/hr
as S02 Ibs/MM Btu
From scrubber
as S, Ibs/hr
as S02, Ibs/hr
as S02, Ibs/ MM Btu
S02 removal In
scrubber, %
Low
Sulfur
2,571.6
5,138.1
1.10
345.4
690.1
0.15
High
Sulfur
14,924
29,818
6.37
345.4
690.1
0.15
Low
Sulfur
2,419
4,833
1.35
359.8
718.9
0.20
High
Sulfur
14,043.9
28,059.8
7.82
359.8
718.9
0.20
86.57
97.69
85.13
97.44
•includes the 5 percent excess F^S that is fed to the scrubber
and sent to the incinerator/boiler. See text.
.698
-------
of the SO- feed to the Bi-Gas scrubber, and the Lurgi flue gas
volume is lower. These effects combine so that the Lurgi FGD
system is able to reach a final S02 emission level of 0.15 Ibs/MM
Btu while, at substantially the same sulfur removal efficiency,
the Bi-Gas scrubber can reduce the S02 emission only to 0.20
Ibs/MM Btu. Although both operations meet or exceed the present
most stringent emission standard of 0.20 Ibs S02/MM Btu
(Oklahoma), the differences in performance emphasize the
variations that may be expected in practice.
In pilot plant operation of the citrate process, S02 removal
efficiencies of 95 to 98 percent have been obtained. However,
for the Bi-Gas high sulfur case to reach the SG^ emission level
of 0.15 Ibs/MM Btu, a sulfur removal efficiency of 98.1 percent
would be required.
Reference—
Moyes, A. J., Mills, B., and Reeve, R. N., "The Citrex Process for
Desulphurisation of Gas Streams." International Conference of
European Federation of Chemical Engineers, Salford, England, 6
April 1976. 959*
Glaus Sulfur Recovery
In the Lurgi base case, the acid gas stream from the Selexol ^ S
removal unit is divided, with about two-thirds being routed to
the flue gas desulfurization unit and the remaining one-third
being scrubbed with water for ammonia removal before entering the
Claus unit. Total gas entering the Claus unit for the Lurgi base
case is 780 MPH. This stream contains 9.2U mol percent H2S (wet
basis) and trace amounts of COS and CS .
•Pullman Kellogg Reference File number
699
-------
In the Glaus unit, one-third of the entering H2S is oxidized with
air to S02 with liberation of heat:
H2S + 1.502 * S02 + H20
Then the remaining H2S reacts with SO formed in the oxidation
step to form elemental sulfur:
2H2S + S02 > (3/x)Sx •«• 2H20
This exothermic reaction is carried out over bauxite catalyst.
Heat is recovered as steam in the oxidation step as well as after
each of the three reaction steps. Overall, the reaction is
3H2S + 1.502 * (3/x)Sx + 3H20
Molten elemental sulfur formed in the process flows from the
sulfur condensers to a sulfur pit. In the Lurgi base case, 26.5
STPD of sulfur are produced in the Glaus unit corresponding to a
9^.3 percent recovery. In the Lurgi high sulfur coal case, 111.0
STPD are produced corresponding to a 96.4 percent recovery (2)«
It is noteworthy that the Glaus plant would approximately triple
in size for both cases if a different FGD system (not requiring
H2S) had been chosen.
Tail gas from the final sulfur condenser contains 4,477 ppm(v)
sulfur (as S1) for the low sulfur coal case and 7,891 ppm /v\
sulfur for the high sulfur coal case. These concentrations are
far too high for discharge to the atmosphere. Therefore, tail
gas treatment to reduce sulfur emissions is required. For this
•Item in reference list following "Beavon Tail Gas Treating Unit"
700
-------
treatment the Beavon process was chosen. The Ralph M. Parsons
Company, licensors for the Claus/Beavon process, provided design
and cost data for the system.
In the SRC II liquefaction case about 65 percent of the sulfur
entering with the feed coal is recovered as salable sulfur in the
process gasifier system via the Claus/Beavon process route.
About 30 percent of the entering sulfur is recovered as salable
sulfur in the fuel gas gasifier system via the Stretford process.
(About 1 percent leaves in the ash and U percent in the fuel oil
product.) In the Claus unit, fed with offgas from the Rectisol
acid gas removal system, from the acid gas removal system in the
liquefaction section and from sour water stripping, 818 STPD of
sulfur are produced, corresponding to a 95 percent recovery.
Tail gas from the final sulfur condenser contains 11,972 ppm(v)
of sulfur as ^ S and S02 • As in the Lurgi cases, tail gas
treatment is required and for this the Beavon process was
selected.
A typical flow sheet for a three-stage Claus process with partial
bypass of the acid gas feed is shown in Figure 9-52.
Beavon Tail Gas Treating Unit
The Beavon process, licensed by The Ralph M. Parsons Company, was
chosen for treatment of the tail gas from the Claus units in the
Lurgi cases and in the SRC II case.
There are about thirty Beavon processes in commercial operation.
The Claus/Beavon combination was selected because it appears to
be superior to Claus/SCOT or Claus/ARCO processes, at least for
these applications, for the following reasons (2):
701
-------
O
to
Hf STUM
MtlOWER
JUmffl PUMP
Figure 9-52. Glaus process flowsheet*
*From Item 1, p. 123, in reference list
-------
o About 15 percent lower capital cost
o Superior operation when processing acid gases containing
high C0? concentrations
o Lower fuel gas requirements
A simplified flow diagram of the Beavon process is shown in
Figure 9-53. The process is based on the catalytic conversion of
sulfur species to H2S by hydrogenation and hydrolysis followed by
reaction of F^ S to elemental sulfur in a Stretford unit. Reducing
gas is produced via incomplete combustion of hydrocarbon gas
(SNG, for example) in air. Tail gas from the Glaus unit is mixed
with reducing gas and the stream is passed via a catalyst where
the following reactions take place at about 700°F (4, p. 11):
SQ + 8H2 — * 8H2S
S02 + 3H2 — * H2S + 2H20
COS + H20_+, H2S + C02
CS2 + 2H20 -> 2H2S + C02
CO + H20 — > C02 + H2 (water-gas shift)
Essentially complete conversion to H2 S is achieved. Heat is
recovered in a boiler downstream of the reactor. Final cooling
to about 100°F is required before the gas stream enters the
Stretford absorber. Here the gas is contacted with an aqueous
solution of sodium carbonate activated with sodium raetavanadate
(SMV) and anthraquinone disulfonic acid (ADA). In the absorber
the H £ reacts as follows:
H2S + Na2C03 - * NaHS + NaHC03
In the holding tank following the absorber, SMV reacts with HS~
to form solid elemental sulfur:
HS-
703
-------
COOLER
r~^i
El
J
T
COMCNT
• LOnoOWH
CONOEHSAIt TO
SOUK WATCH
Figure 9-53. Beavon tail gas treating process - typical flow diagram*
*Frora Item 4 in reference list
-------
This reaction may be shown more completely as:
2NaHS + 4 NaV03 + H20—* 2S+ Na^Og + UNaOH
The ADA present in reduced while oxidizing the SMV back to V*5:
Na V 0 + 2NaOH + H 0 + 2ADA—* 4NaVO, + 2ADA (reduced)
249 2 3
The solution next flows to the oxidizer where oxygen from air is
used to oxidize the reduced ADA:
2ADA (reduced) + 02 » 2ADA + 2H20
Excess air from the oxidizer, free of pollutants, is vented to
the atmosphere. The overall reaction for the Stretford process
can be written as follows:
H2S + 1/202 » S + H20
The absorber overhead gas is relatively free of sulfur compounds,
containing only 1 ppm H2S and 62 ppm COS (2).
The sulfur formed appears as a froth containing about 6 to 8
percent sulfur on top of the oxidizer liquid. Filtration and
melting with low pressure steam yield a liquid sulfur product of
about 99.8 percent purity (M,p.7).
Several side reactions are possible in the process. If the
sodium hydrosulfide contacts absorbed oxygen in either the
absorber or the oxidizing tank, sodium thiosulfate will form as
follows:
2NaHS + 202 —» Na2S203 + H20
705
-------
A liquid purge stream is normally needed to rid the system of
byproduct salts. This stream is expected to have the following
composition:
Wt %
ADA 0.16
NaVQj 0.69
Na2C03 1.18
7.12
15.85
H20 75.00
100.00
The quantity to be purged is expected to be 210 and 360 gallon
per day, respectively, for Lurgi gasification of low and high
sulfur coal (2). By analogy, a purge of about 80 gallons per
is expected for the liquefaction case.
A system has been developed by Parsons for treatment of the pura
stream to recover its contained ADA and NaV03 and reject th
byproduct salts as sodium sulfate to eliminate any possible wat
pollution problem. Alternatively, Union Oil Research ha
developed a variation of the process which eliminates th
formation of byproduct salts and consequently the need for
purge (4,pp.7-8).
Stretford Sulfur Recovery
The Stretford process was selected by Parsons, in their conce
tual design for SRC II liquefaction, for recovery of sulfur f
the fuel gas gasifier system, apparently because the react
gases are presumed to contain sulfur only as H2S. The H s
absorbed in a redox solution which is then regenerated in an
blown oxidlzer. The sulfur is skimmed from the oxidizer
froth, melted,and transferred to storage.
.706
-------
Where sulfur is present i*h the feed gases only as & S the
Stretford process removal efficiency is nearly 100 percent and
yields fuel gas containing about 1 ppm(v) of H2S. Sulfur
production is 370.9 STPD.
References- -
1. Chute, A. E. . "Tailor Sulfur Plants to Unusual Conditions."
Hydrocarbon Processing, April 1977.
2. Griebe, M. H., The Ralph M. Parsons Co., Analysis of the
Claus/Beavon System Applied to Two Cases for Sulfur Recovery.
Private communications, March through May 1978.
3. Pullman Kellogg, Engineering Evaluation of a Process to Pro-
duce 250 Billion Btu/Day Pipeline Quality Gas, June, 1972.
4. Beavon, D. K., "Four Years' Experience with the Beavon Sulfur
Removal Process ." APCA 70th Annual Meeting, Toronto, June
1977. 905»
5t Moyes, A. J., and Vasan, S., "Holmes-Stretford H2S Removal Pro-
cess Proved in Use." Oil and Gas Journal, September 1974.
889*
•Pullman Kellogg Reference File number
707
-------
COSTS FOR CONTROL OF GASEOUS EMISSIONS
Development of costs for processes for control of gaseous
emissions is based primarily on published data. Updating of the
published costs to the end of 1977 was accomplished through use
of the Plant Cost Index, compiled and published by "Chemical
Engineering" in the May 8, 1978 issue.
The processes selected for investigation are considered to be
representative of the best available technology. The develop-
ments are simplified hypothetical cases, intended to demonstrate
the types of studies that would be required for a more rigorous
treatment, and cannot be interpreted or used as definitive
estimates. In-depth studies will be needed in order to make
specific process recommendations.
Some of the simplifying assumptions that were made for these
studies are:
o Power plant size of 500 MW with a heat rate of 9,000 Btu pe
hour per kilowatt or 4,500 MM Btu per hour
o Costs are calculated to the fourth quarter of 1977 (CE Cost
Index = 210, based on 1956 to 1959 = 100) by prorating from
the year in which costs are quoted
o Profit and return on investment are not included in the
developed costs
More rigorous and extensive studies are needed to consider th
many variables that affect the economics of the control
technology and, ultimately, the economics of the coal conversio
process. Such studies could include the effects on capital
operating costs of:
o Control process size (capacity, throughput)
o Changing analyses of coal feeds
708
-------
o Conversion plant location
o Conversion plant capacity
o Varying control process efficiency or, stated another way,
the effect of meeting varying (increasingly stringent)
environmental standards
o Alternate financing schemes for the conversion plant
o Time required for installation of the control process as it
affects conversion process downtime for retrofitting or
overall construction time for new installations
Particulate Control
The control of particulates in coal conversion facilities will be
governed largely by established practices. Particle size, range,
density, resistivity, concentration, composition, the degree of
removal required, the allowable pressure drop, and other factors
will all influence the selection of the particulate control
method.
The four most common types of particulate collectors may be
arranged in order of increasing efficiency, complexity and cost:
o Cyclone collectors
o Wet scrubbers
o Fabric filters
o Electrostatic precipitators
Cyclones are most commonly used in these applications:
o When particulates are mainly in the coarser size ranges
o When particulate concentrations are fairly high, e.g.,
above 3 grains per standard cubic foot (SCF)
o When high collection efficiency is not critical
o When they can serve as pre-collectors in conjunction with
709
-------
other types of collectors that are more efficient in
removing fine particulates
Cyclones have the lowest capital cost of the four general types
of particulate collectors. Costs of $0.08 to $0.10/ACFM (ACFM =
actual cubic feet per minute) for units in the capacity range of
100,000 ACFM have been reported, with special custom designed
units costing as much as $0.35/ACFM. Installation will usually
add about 25 percent to the cost. These figures are based on
1971 data (1, p.5-10). Escalating these figures to 1977 based on
the CE cost index (1971 = 130, 1977 = 210) and adding 15 percent
for contractor overhead and profit results in the following
installed costs for 100,000 ACFM units:
Costs of Cyclone Collectors, per ACFM
Equipment only, Installed cost
1971 1977
$0.08 $0.186
0.10 0.232
0.25 0.581
0.35 0.813
High energy wet scrubbers (venturi scrubbers) are normally used
where:
o Fine particles must be removed at high efficiency
o Cooling is desired and moisture addition is not objection-
able
o Gaseous contaminants as well as particulates are involved
o Volumes are not extremely high (because of the relatively
JT
higher operating cost per ACFM)
o Relatively high pressure drop is tolerable
o Contamination of the scrubbing liquid with materials re-
moved from the gas poses no problem
710
-------
Initial cost (1971) of wet scrubbers sized for about 100,000 ACFM
ranged from $0.25 to $0.35/ACFM in carbon steel and about
$0.65/ACFM in alloy steel with cost of erection adding about 25
percent (1, p.5-13). Escalating these costs to 1977 and adding
contractor overhead and profit yields the following installed
costs for 100,000 ACFM units:
Costs of Venturi Scrubbers, per ACFM
Equipment Only, Installed Cost,
1971 1977
$0.25 $0.580
0.35 0.813
0.50 1.161
0.65 1.509
Fabric filters are normally employed where:
o High efficiencies are desired
o Operation is above the gas dew point
o Temperatures are moderate
o Valuable material is to be collected dry
o Water availability and disposal is a problem
Initial cost of these units (1971) ranged from $0.50 to $1.20 per
ACFM for 100,000 ACFM units, depending on the filter medium used.
Erection cost is reported to be about 30 percent of the equipment
cost (1, p.5-16). Adjusting these costs as before yields the
following installed costs for 100,000 ACFM filters:
711
-------
Costs of Fabric Filters, per ACFM
Equipment Only,
1971
$0.50
0.75
1 .00
1 .20
Installed
1977
$1.207
1.811
2.414
2.897
Cost
Electrostatic precipitators (ESP's) are most commonly used where:
o Very high efficiencies are required for fine materials
o Volumes of gas are very large
o Water availability and disposal are problems
o Valuable dry material is to be recovered
The purchase price (1971) of an ESP in the 100,000 ACFM capacity
range was about $0.80/ACFM while that of one ten times as large
was about $0.40/ACFM. Erection cost adds about 70 percent
(1,p.5-20). The adjusted installed cost (1977) for 100,000 ACFM
becomes $2.527 per ACFM.
The ranges of installed capital costs for the four particulate
control devices are shown in Figure 9-54. Variation of capital
cost with capacity was obtained by multiplying the base capital
cost by the ratio of new to base capacity raised to the 0.7
power.
Operating costs are highly variable, using functions of particu-
late size, properties, loading, degree of removal required, pres-
sure drop,and other factors. Therefore, no general operating
costs have been developed.
712
-------
I— i—•—-i—j-H-~ H -) - -T-
0.01
Figure 9-54. Installed capital costs of particulate control
devices.
713
-------
Sulfur Dioxide Control
Desulfurization of Coal by Coal Cleaning and by the Meyers
Process—
The range of coal cleaning processes now being practiced in the
United States may be generalized into four individual levels of
preparation. These levels may be defined as follows:
o Level 1—no preparation, direct utilization of the run-of-
mine product.
Level 2—removal of gross non-combustible impurities, plus
control of particle size and promotion of uniformity (typi-
cally 95 percent material yield and 99 percent thermal re-
covery) . Little change in sulfur content.
o Level 3—single-stage cleaning allowing little component
liberation. Particle sizes less than 3/8 inch usually are
not prepared. 80 percent material yield and 95 percent
thermal recovery. Limited ash and sulfur content.
o Level 4—multi-stage cleaning with controlled pyrite lib-
eration. Usually incorporated dewatering and thermal
drying. 70 percent material yield and 90 percent thermal
yield. Maximum ash-sulfur rejection and calorific content
of product.
Preparation practice for most coals used by electric utilitie
lies between levels 2 and 3. The preparation practices fOr
metallurgical coals are typically level 4. The relative costs of
these different levels are indicated in TABLE 9-18. The exte
to which a specific coal can be cleaned is dependent upon th
. 714
-------
TABLE 9-18. PREPARATION PLANT CAPITAL AND OPERATING COSTS (1)(2)
Eastern Bituminous Coal
Design Capacity
Clean Coal
Tons/Yr
3,000,000
2,000,000
1 ,000,000
Western Subbituminous
Utility Coal (4)
10,000,000
5,000,000
3,000,000
2,000,000
1 ,000,000
Cleaning Cost $/Ton (7)
Level 4
$25,200,000(3)
17,500,000(3)
9,000,000(3)
Level 3
$11,200,000
8,100,000
4,350,000
0.80(5)
1.74(6)
0.45(5)
0.87(6)
Level 2
$3,200,000
2,500,000
1,500,000
6,720,000
3,360,000
2,040,000
1 ,580,000
1,200,000
0.05(5)
0.17(6)
0.05(8)
0.12(8)
(1) From Item 9, pp. 53-56 in reference list
(2) Mid-1974 dollars
Level 4 - Detailed, elaborate facility (15% recovery).
Level 3 - Removal of liberated mineral matter (15% recovery).
Level 2 - Removal of only gross mineral matter (95?
recovery).
(3) The capital costs utilized for cleaning eastern bituminous
coals at Level 4 ranged between $23,000 and $25,000 per ton
of raw feed capacity per hour. Utilizing the "Best
Practice" would increase this value to about $30,000 per ton
of raw feed capacity per hour. The value would increase to
an estimated $40,000 per ton hour if the "Best Cleaning
Technology Available" were developed.
(4) Only Level 1 or 2 is applicable. Lignite - Level 1 only
considered necessary.
(5) Includes labor, power, maintenance - no amortization or
return on investment. Thermal drying adds about 25% to
capital costs and 30% to operating costs.
(6) Includes straight line financing at &% interest, 20 years
life and 5% ROI.
(7) Eastern Bituminous coal cleaning - three million tons per
year.
(8) Western Subbituminous coal cleaning - ten million tons per
year at Level 2.
715
-------
characteristics of the coal and the sophistication of the prep-
aration process. The limitations are often both economic and
technical.
Examples of the operating characteristics for three types of coal
cleaning plants are shown in TABLE 9-19. The characteristics of
the raw coal feed are different for each plant and represent
progressively more difficult separation efforts (2, pp.12, 52).
Capital costs for the three plant types were taken from a report
by Fluor Utah, Inc. and were updated to 1977. These capital
costs are shown in Figure 9-55 as functions of plant feed
capacity.
Capital costs were developed for a standard plant size of 75,000
TPD run-of-mine coal delivered to the incoming stockpile for 250
days per year. Included are raw coal receiving and breaking
blending and stockpiling, prepared coal loadout and support
facilities.
The yield factor is the TPD output of prepared coal divided by
the TPD input of run-of-mine coal. The rated production of a
non-standard plant is calculated by dividing the required Btu per
day output by the HHV of the prepared coal and by the yield. The
cost of a non-standard plant is calculated by multiplying the
cost of the standard plant by the ratio of the rated production
of the non-standard plant and the rated production of the
standard plant raised to an exponent (the cost scaling factor) .
Values of these parameters are as follows:
716
-------
TABLE 9-19. COAL PREPARATION- PLANT CHARACTERISTICS
Raw
Coal
Finished
Coal
Tons
Removed
Hypothetical Crushing and Screening Plant
Tonnage per day:
Ash
Sulfur
Other
Total
Heat Value:
Btu/lb
Billion Btu/day
Btu yield
13,950(18.6*) 13,360(18.40*)
1,130( 1.510 1,110(1.50*)
59.920
75,000
59,780
74,250(99.0*)
10,990
1,648
11,030
1
638
99.4%
Hypothetical Baum Jig Plant
Tonnage per day:
Ash
Sulfur
Other
Total
Heat value:
Btu/lb
Billion Btu/day
Btu yield
10,350(13.8*) 4,890(7.90*)
1,065( 1.42%) 375(0.61*)
63.585 56,610
75,000 61,875(82.5*)
12,760
1,914
13,236
1,638
85.6
Hypothetical Heavy Media Plant
Tonnage per day:
Ash
Sulfur
Other
Total
Heat value:
Btu/lb
Billion Btu/day
Btu yield
10,875(14.5*) 5,015(8.50$)
1,120( 1.49%) 420(0.71*)
63.005 53.590
75,000 59,025(78.7*)
13,084
1,963
13,870
1,638
83.4%
590 (4.2*)
20 (1.8*)
140 (0.2*)
750
5,460(52.8*)
690(64.8*)
_6,975(11.0*)
13,125
5,860(53.8*)
700(62.5*)
9.415(14.9*)
15,975
717
-------
125
100
(0
O
U
(8
CJ
25 50 75 ~ """ TOO
Plant Feed Capacity, TPD 1000's
125
Figure 9-55. Coal Preparation Capital Cost
718
-------
Crush & Baum Heavy
Screen Jig Media
Capital cost (1975), $MM 9.5 52.1 70.2
Standard plant capacity,
TPD 75,000 75,000 75,000
Yield factor 0.99 0.825 0.787
Cost scaling factor 0.80 0.75 0.70
The 1975 operating cost for coal preparation in a heavy media
plant is stated (9, p.56)(16) to be, per ton of feed coal:
Coal cleaning $1.90
Coal storage and refuse disposal 0.36
Total operating cost $2.26
From TABLE 9-19 it is seen that in heavy media cleaning 78.7
percent of the weight of the feed coal is recovered as cleaned
product and that the higher heating value (HHV) of the cleaned
coal is 6 percent higher than the HHV of the feed coal.
The Meyers process for pyritic sulfur removal was studied by TRW
(U, pp.217-230). In this economic study the HHV of the feed coal
was taken at 12,291 Btu/lb. The assumption is made for this pres-
ent comparison that weight recovery and increase in HHV for
heavy media cleaning will be the same for feed coal at 12,291
Btu/lb as at 13,084 Btu/lb, thus allowing the two processing
methods to be compared on an equal basis.
The operating cost of heavy media cleaning is calculated as
follows:
($2.26 x 1,000,000)/(78.7*)(12,291 x1.06)(2000) = $0.110/MM Btu
of cleaned coal
719
-------
The cost of the cleaned coal is calculated from the cost of the
feed coal:
Cost of feed coal = $C/ton
= (C x 1,000,000)7(12,291 x 2.000)
= $0.0407(C)/MM Btu
Cost of cleaned coal = (0.0407)/(0.787 x 1.06)
= $0.0488(C)/MM Btu
Incremental cost of cleaned coal = $0.0081(C)/MM Btu
The total incremental cost of coal cleaning is the sum of the
operating cost and the incremental cost of cleaned coal per MM
Btu = $0.110 + $0.0081(C).
If the incremental cost is to be escalated to 1977, the operating
cost portion of the total incremental cost is increased by 17
percent and the total incremental cost per MM Btu becomes $0.129
+ $0.008l(C).
The Meyers process is an alternate to physical coal cleaning for
removal of pyritic sulfur. Pyrite, FeS2> in the coal is removed
by chemical methods to produce ferric sulfate, ferrous sulfate
and elemental sulfur. An economic study of the process was per-
formed by TRW, who considered the 4 cases that are outlined in
TABLE 9-20 (4, pp.217-230).
The run-of-mine coal fed to each of the four process configura-
tion cases contains 20 percent ash and 3 to 4 percent pyritio
sulfur. Approximately 90 to 95 percent of the sulfur is removed
during processing. The capital costs include the battery limits
and offsite facilities required for processing, interest for
construction, startup costs and working capital.
720
-------
TABLE 9-20. COAL CLEANING WITH THE MEYERS PROCESS*
Case 1: Cleaned fine coal (14 mesh top size), physical
cleaning, Meyers fine coal configuration.
Case 2: Run-of-mine coarse coal (1/4 inch top size), Meyers
coarse configuration.
Case 3 & 4: Deep cleaned fine and coarse coal with 50% Meyers
bypass.
Feed: Rate, TPH
Ash, %
Pyrite S, %
Trillion Btu/yr
Physical Cleaning Ash:
Rate, TPH
Ash, %
Pyritic S, %
Trillion Btu/yr
Product:
Rate, TPH
Ash, %
Pyritic S, %
Trillion Btu/yr
Case 1
Case 2
Cases 3 & 4
120
20
3-4
23.6
20
75
10-14
1.4
93
6
0.1
21.3
100
20
3-4
19.7
85
15
0.2
17.5
240
20
3-4
47
40
75
10-14
2.
185-190
6
0.
Fine 43.
.2
8
2
4
Coarse 42.2
•From Item 4, pp. 217-230 in reference list
721
-------
The required sales prices for processed coal are calculated by
the following equation:
P = (aX + bY + cZ)/d
where
P = Required sales price for processed coal, $/MM
Btu
X = Working capital for raw materials and supplies,$
Y = Sum of the total plant investment and start-up
cost, $
Z = Annual total operating cost, $/year
a, b, c = Constants given below
d = Annual energy output, MM Btu/yr
Investor Utility
Constant Financing Financing
a 0.391 0.1MO
b 0.505 0.121
c 1.016 1.006
The ugrading costs are determined by deducting the cost of the
dirty energy ($0.4l/MM Btu at $10/ton; $0.81 at $20/ton and $1.22
at $30/ton).
For Cases 1, 3 and 4 processing costs include ash reduction as
well as sulfur reduction. For Case 2, physical cleaning was
assumed to be coupled with pyrite removal which results in a
major reduction in ash from about 20 percent to about 6 percent.
Prorating cost for plant capacity used the ratio of new plant
capacity to base capacity to the 0.7 power.
722
-------
Capital investment costs for the four cases are shown as a
function of capacity in Figure 9-58.
The incremental processing cost for the Meyers process for each
of the 4 cases is shown as a function of coal cost, both investor
and utility financed, in Figure 9-59. Also shown is the
operating cost for heavy media processing.
Inspection of the figures leads to the conclusion that the Meyers
process requires more capital than the heavy media process and
that its operating costs for the best case are somewhat higher:
25.000 TPD Feed (1975)
Meyers, Heavy
Case 4 Media
Capital investment, $MM 101 33
Incremental operating cost*, $/MM Btu 0.330 0.272
•Coal at $20/ton. Utility financed
However, a pyritic sulfur reduction of 95 percent is claimed for
the Meyers process while a reduction of only 75 to 85 -percent is
obtained with heavy media washing. Therefore, in terras of sulfur
removal effectiveness, the most favorable Meyers case (Case 4)
may be about equivalent to heavy media washing.
Desulfurization of Coal During Combustion in a Fluidized Bed—
Fluidized bed combustion (FBC) of coal to generate steam with
capture of S02 by limestone was compared to a conventional boiler
equipped with flue gas desulfurization (FGD) (5). The comparison
considered two coals, at 3.6 percent and 0.4 percent sulfur, two
capacities, of 100,000 and 400,000 pounds per hour of steam, and
three installations, one at a coal fired plant, one at an oil
fired plant and one at a grass roots plant. Capital investments
723
-------
200
10
Plant,Feed Capacity, TPD 1000's
Figure 9-58. Meyers process capital cost.*
*From Item 4 in reference list
724
-------
E-i
10
O
U
O
H
H
CO
CO
u
u
§
1.20
1.00
0.80
0.60
014
0.2
1975 Costs (CE Index =j!80) : ~T
~Por 1977 Costs~tTse~CE Index .= 2TO"
Meyers Proces
|.:--:
Case -l^i
Meyers Process*
-Heavy Media i
f-Process-**
I .. I •- t • ) , T I *^*.^^WVifc»0 7 •
-4-4-^-teb;^--ha:^
10 20 30
Raw Coal Cost, $ per Ton
Figure 9-59. Meyers process incremental processing
cost.
*From Item 4 in reference list
**From Item 9 in reference list
725
-------
and operating costs for the cases are shown in TABLE 9-21 for a
single boiler added to a coal fired plant, in TABLE 9-22 for a
single boiler added to an oil fired plant and in TABLE 9-23 for
grass roots boiler plants with backup.
Figure 9-60 compares the capital cost of FBC to conventional
installations with FGD in the several configurations and at
various steam rates when operating on Illinois No. 6 coal at 3.6
percent sulfur.
Figure 9-61 compares the cost of steam production in the two
boiler types in the same configurations and steam rates as in
Figure 9-60, when fed with the high sulfur coal.
A review of the report as well as the tables and figures leads to
the following conclusions:
o For low sulfur compliance coals, which require no SQ.
controls, a conventional boiler equipped with only an ESP
appears to be superior to the FBC.
o For high sulfur coals, the FBC appears to be the better
choice. A lower capital investment and steam production
cost is obtained for all cases. At a capacity of 400,000
Ibs/hr, the FBC shows the following steam cost advan-
tages:
- added to coal fired plant $0.28/M Ib
- added to oil fired plant lO.S^/M Ib
- grass roots plant $0.?2/M Ib
o One of the advantages of the FBC is its higher heat
release rate. It is reported to be 100,000 Btu/hr/oubi
foot of expanded bed volume (50,000 to 60,000 Btu/hr/CP
of firebox) as opposed to 20,000 Btu/hr/CF of firebox fo
a conventional boiler (5, p.11).
. 726
-------
TABLE 9-21. COMPARISON OF INVESTMENTS AND COST OF STEAM
(EX FUEL) FOR SINGLE BOILER ADDED TO COAL FIRED PLANT"
Fuel (1)
Boiler Type
Steam Rate, KPPH
Investment, M$
Fuel Handling
Additions
Boiler and Stack
Envtl. and Waste
Disp.
Total, M$
High Sulfur Coal
Low Sulfur Coal (2)
Fluidized
Bed
Combustion
100
0.6
3.1
ill
1.8
100
0.9
7.6
2.9
U.I
Conventional
With
Scrubber
100
0.2
2.9
2.3
5.1
100
0.3
8.6
6.1
15.0
Fluidized
Bed
Combustion
100
0.6
3.1
0.9
1.6
100
0.9
7.6
2.1
10.9
Conventional
With
ESP
100
0.2
2.9
0.7
3.8
100
0.3
8.6
1.8
10.7
Unit Cost of Steam (ex Fuel).
-------
CO
TABLE 9-22. COMPARISON OF INVESTMENTS AND COST OF STEAM
(EX FUEL) FOR SINGLE BOILER ADDED TO OIL FIRED PLANT
Fuel
High Sulfur Coal
Fluidized
Bed
Boiler Type
Steam Rate, KPPH
Investment, M$
Fuel Handling
Allowance (2)
Boiler and Stack
Envtl. and Waste
Disp.
Total, M$
Unit Cost of Steam
Direct Op. Costs
(ex Fuel and BFW)
Boiler Feed Water
Capital Charges
Combustion
100
1.8
3.1
1.3
6.2
(ex
112
60
157
MP
2.7
7.6
3 .4
*13.7
Fuel), */k
102
60
87
Conventional
With
Scrubber
10_0
1.8
2.9
2.6
7.3
Ib.
150
60
185
400
2.7
8.6
6.9
18.2
108
60
115
Low Sulfur Coal
Fluidized
Bed
Combustion
100
1.9
3.1
1.1
6.1
100
60
j.55
400
2.9
7.6
2.9
13.1
60
60
85
(1)
Low Sulfur
Fuel Oil(l)
Conventional
With
ESP
AQO
1.9
2.9
1.0
5.8
83
60
147
400
2.9
8.6
2.6
14.1
52
60
_89
Package
100
0.1
1.5
__
1.6
46
60
41
400
0.2
3-7
__
3.9
31
60
25
Total, */k Ib.
(ex fuel) 359 2U9 395 283
315
205
290
201
147
116
(1)' Low sulfur coal and fuel oil by definition are sufficiently low in sulfur that no
needed to meet whatever environmental limits are applicable.
controls are
(2) In some cases where coal is reliably available by truck delivery, the capital costs for fuel receipt
and storage could be significantly* reduced. In such cases, however, the delivered price of coal would
rise more or less correspondingly so that the overall cost of steam would not be changed markedly.
-------
•;ARLE 9-23. COMPARISON OF INVESTMENTS AND COST OF STEAM
VEX FUEL) FOB GRASS ROOTS BOILER PLANTS WITH BACKUP
-J
ho
vo
Fuel
Boiler Type
Steam Rate, KPPH
Investment, H$
Fuel Handling
Allowance (2)
Boiler and Stack
Envtl. and Waste
Disp.
Total, M$
Unit Cost of Steam
Direct Op. Costs
(ex Fuel and BFW)
Boiler Feed Water
Capital Charges
High Sulfur Coal
Fluidized
Bed
Combustion
100
1.8
5.8
1.9
9.5
(ex
180
60
241
400
2.7
14.3
5.0
^22.0
Fuel),
-------
50 70 100
Figure 9-60. Capital cost comparison:
boiler with FGD*
*From Item 5 in reference-list
730
FBC vs. conventional
-------
U)
OQ
o
o
o
1-1
w
D
O
Q
D
w
O
U
CO
100
200
500
600
300 400
STEAM RATE, M:SB/HR
Figure 9-61. Steam cost comparison: FBC vs. conventional boiler w/FGd.*
*From Item 5 in reference list
-------
o Another advantage of the FBC is the formation of CaSOj.
which is disposed of in reduced quantities as a dry solid
mixed with ash and unreacted limestone. Wet limestone
scrubbing systems form a sludge consisting of
CaS03» 1/2H20, CaS04»2H20, unreacted limestone, flyash and
water (normally about 50$). Assuming a Ca/S mol ratio of
3 for the FBC and 1.2 for the wet limestone scrubbing
system results in a 20 to 30 percent reduction in solid
wastes for the FBC (5,pp.126-130).
o Thermal NO emissions should be reduced for the FBC
yt
system due to its lower operating temperature of 1,400° to
1,600°F as opposed to 2,500°F or higher for conventional
boilers (5,p.9). Ex'xon estimates NO at 0.5 Ib/MM Btu
X
for FBC as compared to 0.7 for a conventional boiler
(5,p.128).
o Further development of the FBC appears warranted.
Flue Gas Desulfurization—
Desulfurization of stack gases by either dry or wet gas cleanup
systems has been studied for several years. There are large
scale test programs presently underway to determine system
reliability and cost for several sulfur dioxide removal
processes. Some of the EPA-sponsored tests are summarized in
TABLE 9-24. Cost data for these processes have been published.
(12)(13) in great detail and may be referred to for support of
the summarizations that follow.
The design of sulfur dioxide removal facilities is dependent upon
the actual quantities of gas to be handled and the sulfur dioxide
emission rates. TABLE 9-25 shows these parameters for several
variations.
732
-------
TABLE 9-2*<. EPA-SPONSORED STACK GAS DESULFURIZATION DEMONSTRATION SYSTEMS*
EPA-Sponsored Process
(byproduct)
Limestone slurry scrub-
bing (sludge)
Lime slurry scrubbing
(sludge)
Magnesia slurry scrub-
fa ing- regeneration
(98J sulfuric acid)
Catalytic oxidation
( reheat)
(80J sulfuric acid)
-j Sodium scrubbing-
OJ regeneration
**• (sulfur}
Cooperating
Utility
TVA
TVA
Boston Edison
Illinois Power
Northern Indiana
Public Service
Co .
Process
Developer
Bechtel and
others
Chemico, Bechtel
and others
Cheraico-Basic
Monsanto
Davy Powergas
Allied Chemical
Location
Shawnee Unit 10
Paducah, Ky.
Shawnee Unit 10
Paducah, Ky.
Mystic Station 6
Boston, Mass.
Wood River Station 4
East Alton, 111.
D. H. Mitchell
Station 11
Garv. Ind.
Unit Size
and Type
10 MW
coal
10 MW
coal
155 MW
oil
110 MW
coal
115 MW
coal
•From Item 12 in reference list
-------
TABLE 9-25. FLUE GAS AND SULFUR DIOXIDE
EMISSION RATES FOR NEW COAL FIRED POWER PLANTS
Plant
Size, MW
200
500
500
500
1,000
Fuel Sulfur
Content^ %
3.5
2.0
3.5
5.0
3.5
Gas Flow to FGD,
M ACFM (310°F)
630
1,540
1.540
1,540
2,980
S02 to FGD
Ibs/hr
9,310
13,010
22,760
32,510
44,000
TABLE 9-26. REQUIRED REMOVAL EFFICIENCIES IN FGD UNITS
Sulfur Content
of Fuel. %
2.0
3.5
5.0
Particulate
Removal, %
98.7
98.7
98.7
SO-
Removalt <
58.9
76.3
83.4
734
-------
The FGD systems are designed to meet the existing federal
allowable emissions for coal fired boilers of 0.1 Ib/MM Btu heat
input for participates and 1.2 Ibs/MM Btu heat input for sulfur
dioxide. TABLE 9-26 shows the required removal efficiencies to
meet the standards.
In TABLE 9-27 are shown the estimated capital investments for FGD
units using the limestone slurry and lime slurry processes, both
of which produce sludge that requires disposal, and the magnesia
slurry and catalytic oxidation processes, both of which produce
sulfuric acid. The comparison is based on the following:
o New coal fired units
o Midwest location
o Costs as of 1974
o 90 percent sulfur dioxide removal
o 30 years life
o Stack gas reheat to 175°F
o On site disposal pond 1 mile from power plant
o No flyash disposal facilities
TABLE 9-28 shows the operating costs for the FGD units under
consideration, based on 7,000 hours of operation per year and 1975
operating costs.
Figure 9-62 shows the effect of power unit size on the capital
investment for the limestone slurry and lime slurry FGD
processes. The same presentation is made in Figure 9-63 for the
magnesia slurry and catalytic oxidation processes.
Figure 9-64 illustrates the effect of the sulfur content of the
coal feed on the operating costs for the 500 MW unit. Figure
9-65 shows the effect of plant size on the operating costs with
fuel containing 3.5 percent sulfur.
735
-------
TABLE 9-27. CAPITAL INVESTMENT FOR FGD UNITS
Plant Capacity, MW 200 500 1000
Sulfur in Fuel, % 3.5 2.0 3.5 5.0 3.5
Limestone Slurry, $MM 13-03 22.60 25.16 27.34 37.73
$/KW 65.16 45.20 50.32 54.69 37.73
Lime Slurry, $MM 11.75 20.23 22.42 24.27 32.77
$/KW 58.75 40.46 44.84 48.54 32.77
Magnesia Slurry, $MM 14.14 22.96 26.41 29.36 38.87
$/KW 70.60 45.92 52.82 58.71 38.87
Catalytic Oxidation,
$MM 19.54 42.52 42.74 42.93 69.89
$/KW 97.69 85.04 85.47 85.86 69.89
TABLE 9-28. OPERATING COST FOR FGD UNITS
Plant Capacity, MW 200 500 10QQ
Sulfur in Fuel, % 3.5 2.0 3.5 5.0 3.5
Limestone Slurry
Mills/KWH 2.80 1.94 2.20 2.43 1.70
-------
I
or
>
c:
•rl
04
«J
u
«•
60
40
20
bOU oUU ±UUU
600 800 10OU
Power Plant Size, MW
Figure 9-62. Capital investment for limestone and lime slurry FGD processes,
-------
60
30
200
400 600 800 1000 200
Power Plant Size, MW
400 600 800 1000
Figure 9-63.
Capital investment for magnesia slurry and catalytic oxidation
FGD processes.
-------
3.5
3.0
CO
el
•J
H
S
CO
O
O
H
I
W
2.5
2.0
1.5
LIMESTONfi
—-(-2-\-LIME-SLURRY
-(3) MAGNESIA1 SLURRY"~500 MW :POWER iPLANT
CATALYTIC OXIDATION ;-_
I-
1
35
D
2
H
w
CQ
30
CO
oc O
2i 0
O
20
1..0
2.0 3.0 4.0
SULFUR IN COAL, %
5.0
Figure 9-64. Effect of sulfur content of coal
feed on FGD operating cost.
739
-------
3.5
w
3.0
o
u
O 2.5
S3
2.0
1.5
rrrtir
!':..(2j_LLimeiljSlurry-I!.F::!:
"J~L:(3)i;Magnes.i!a •
JCatalyeicT*
r.-r_:z
-.—I----K—h
rj^Lt;.
[vt::jr
'•;—'.i;_;:.
"sui'fur"rin: Feed
200
400 ' 600 800
POWER PLANT SIZE, MW
1000
Figure 9-65. Effect of plant size on FGD operating
costs.
740
-------
Comparison of these data show that capital investment for the
throwaway processes is lower than for the product recovery
process and that the lime slurry process has the lowest capital
requirement. Operating costs for the limestone slurry process
are, however, lower than those for lime slurry.
It is of interest to note in Figure 9-64 that variation in the
sulfur content of the feed coal has very little effect on the
operating costs of the catalytic oxidation process.
Capital investment and operating costs of the limestone slurry
process and the citrate process for sulfur recovery are compared
in TABLE 9-29 for a 500 MW power plant in a midwest location fed
with coal containing 3.5 percent sulfur and for removal of 90
percent of the sulfur dioxide from the flue gas. The citrate
process costs were calculated from published data and information
(13) to a basis comparable to that of the limestone slurry
process in order to develop the comparison. The hydrogen sulfide
that is needed for the citrate process was assumed to be supplied
from the feed to the sulfur recovery section of the coal
conversion process.
The capital and operating costs are shown graphically in Figures
9-66 and 9-67, respectively, for the processes by years. It is
of interest to note the apparent convergence in both figures to
an intersection around 1984,
Comparison of the costs for the citrate process and the limestone
slurry process indicates that the citrate process with credit for
sulfur sales is a close competitor of the limestone slurry
process. Further, the citrate process has the advantage of
sulfur production as opposed to sludge production. For these
reasons the citrate process was selected for inclusion in further
economic .studies.
741
-------
TABLE 9-29. COST COMPARISION OF LIMESTONE SLURRY AND CITRATE FGD
Capital Investment, $MM
1974
1977
1979
Operating Cost,
1975
1977 (2)
1977 (3)
1980 (2)
1980 (3)
Operating Cost,
1975
1977 (2)
1977 (3)
1980 (2)
1980 (3)
Operating Cost,
1975
1977 (2)
1977 (3)
1980 (2)
1980 (3)
$MM/yr
Mills/KWH
Btu (4)
Limestone
Slurry
25.16
43.32
7.70
13.56
2.20
3.87
24.45
43.03
Citrate (1)
47.63
51.35
12.28
13.67
14.40
15.35
3.51
3.91
4.11
4.39
39.00
43.41
45.70
48.72
(1) Hydrogen sulfide supplied 'by coal conversion process (2) With
sulfur sales at $40 per ton (3) without sulfur sales (4) HHV hea?
input l
742
-------
I
H
CU
<
u
LIMESTONE SLURRY
500 MW POWER PLANT
20 -
10
1974
1976
1978
19SO
Figure 9-66.
Capital Cost Comparison: limestone
slurry vs. citrate process without
IS generation.
743
-------
B
I
BTU
•o-
OPERATING COST
I-1
•
0
1974
1976
1978
1980
Figure 9-67. Operating Cost Comparison: limestone
slurry vs. citrate process without
H-S generation.
744
-------
Sulfur Dioxide Control for Coal Fired Boilers—
Sulfur dioxide may be controlled by FGD alone or by a combination
of pyritic sulfur removal and FGD. To develop the operating
costs for these systems the following basis was chosen:
Plant capacity 500 MW
Plant Heat rate 9.000 Btu/hr/KW
Heat input 4,500 MM Btu
Plant operation 90$ Onstream (7,890 hrs/yr)
Cost basis 1977
Coal cost $20/ton
Coal heat content 12,57M Btu/lb
Coal feed rate 357,882 Ibs/hr
Sulfur in coal 3.02* (10,808 Ibs/hr)
If the utility boiler at a coal conversion facility is fired with
coal only, the following cases may be considered:
o Case 1: Sulfur dioxide control by flue gas desul-
furization (FGD) alone.
o Cases 2 and 3: Remove 80 percent of the pyritic sulfur
from the feed coal by heavy media washing
and then apply FGD as necessary to meet
sulfur dioxide emission standards.
Two coals were considered for this study. Their sulfur contents
are shown as follows:
Case 2 Case 3
U.S. Average Coal Appalachian Coal
Pyritic Sulfur 1.9U 2.M2$
Organic Sulfur 1.11 0.60
Total Sulfur 3.02$ 3.02$
745
-------
The Northern Appalachian coals are found in Maryland, Ohio
Pennsylvania and West Virginia.
The coal sulfur analysis used is typical for the middle and lower
Kittanning coal bed. The organic sulfur content of these coals
is lower than any other in the United States, with the exception
of western coals which average about 0.45/6 organic sulfur.
When two coals have the same total sulfur content but the type of
sulfur present varies, the coal with the higher pyritic sulfur
content will be more amenable to a cleaning process such as heavy
media washing.
The most stringent sulfur "dioxide emission standard for coal
fired boilers with a heat input greater than 250 MM Btu/hr is
that of Wyoming at 0.2 Ib/MM Btu. The next most stringent is
that of New Mexico at 0.34 Ib/MM Btu. The federal standard, and
that of most states, is 1.2 Ibs/MM Btu.
Figure 9-68 shows the operating costs for the three cases in i/MM
Btu and $MM/yr vs. the S02 emission standard. The processing
costs used are shown below (based on 1977):
o Heavy media washing: 29.1
-------
25 ^
ffi
V)-
CO
O
W
s
80% OF PYRITJCIs 'REMOVED FBOM
E-"
D
s
W
S3
,. 0.2
(7T3 076 0.8 1.0 1.2
S02 EMISSIONS, LBS/MM BTU HEAT INPUT
1.4
1.6
1.8
2.0
CO
O
U
O
2
W
Cu
O
Figure 9-68. Sulfur dioxide control costs for coal fired boilers.
-------
circumstance that the emission standards are relaxed, the com-
bined cases, particularly Case 3 for Northern Appalachian coal
show less of a disadvantage.
Without FGD the sulfur dioxide emission from Case 2 would be 2.37
Ibs/MM Btu and from Case 3, 1.72 Ibs/MM Btu. The latter point is
the reason for the break in the Case 3 line in Figure 9-68
Heating values of both coals are upgraded 6 percent by cleaning
from 12, 574 to 13,331 Btu/lb.
The following table shows the components of the treating costs
and the percentages of each treatment employed, to meet varying
S02 emission standards.
Operating Cost.
-------
with hydrodesulfurization of the oil. The basis for the study
was:
Plant Capacity 50 MW
Plant heat rate 9,000 Btu/hr/KW
Heat input 4,500 MM Btu/hr
Plant operation 90% Onstream (7,890 hrs/yr)
Cost basis 1977
Oil type Residual
Oil heat content 6.24 MM Btu/barrel
Oil feed rate 17,310 Barrels per day
Sulfur in oil feed 1.81% (60% of sulfur in coal feed)
For Case 1, hydrodesulfurization costs for 92 percent sulfur
removal, with an allowance of 25 percent of the cost as credit
for salable sulfur, were calculated to be $1.82 per barrel of
product oil (equivalent to $1.73 per barrel of feed oil) or
$0.292/MM Btu of heat input to the boiler. Sulfur in the resi-
dual oil feed was 44^1 Ibs/hr. If all the oil were desulfurized,
the the desulfurized oil to the boiler would contain 355 Ib/hr of
sulfur which, when burned would yield a flue gas containing 0.16
Ib/MM Btu of sulfur dioxide. For the various flue gas composi-
tions studied, varying amounts of residual oil were desulfurized.
The resulting costs are shown in Figure 9-69.
For rase 2, costs for the citrate FGD process were developed in
the same manner as for the preceding study on sulfur dioxide con-
trol for coal fired boilers. Corresponding costs are shown in
Figure 9-69.
Inspection of Figure 9-69 leads to the conclusion that for oil
fired boilers hydrodesulfurization of the feedstock is a more
economical method for control of sulfur dioxide emissions than is
FGD and, ,as the maximum allowable sulfur dioxide content of the
flue gas is reduced, the advantage increases.
749
-------
-J
tn
o
0.4
0.6 •
0.8 1.
S02 EMISSION,
0 1.2
LBS/MM BTU
1.4
1.6
1.8
2.0
O
13
W
IS
t-3
H
2!
O
o
o
en
t-3
w
^3
d
Figure 9-69. Sulfur dioxide control costs for oil fired boilers.
-------
Nitrogen Oxides Control
NOV Control by Boiler Modifications—
Jv
It is generally recognized that the first step taken in the
direction of lowering NO emission from fossil fuel fired boilers
Ji
will be boiler modifications. Design changes which lower the
flame temperature and reduce oxygen availability result in lower
NO formation. These changes can be one or more of the
,Jt
following:
o Two-stage combustion
o Low excess air firing
o Flue gas recirculation
The approximate percentage reduction of Np^ emissions by the
different techniques is shO"*> below.
Percent NO Reduction
Coal Oil Gas
Staged Combustion 35 35 45
Low Excess Air 20 25 25
Combined Staged Combustion
and Low Excess Air 40 40 50
Flue Gas Recirculation 25 25 ' 45
Thus, it is concluded that a 40 to 50 percent reduction in NOX
emissions can be achieved by boiler modifications.
The most stringent standards for NO emissions from fuel
x
combustion at present are those of New Mexico. These standards
along with projected future standards for NO emissions are:
x
751
-------
New Mexico NOX Projected Future
Standard N0y Standards
1980 1985
ppm(v) ppm(v)
200 100
150 90
150 50
The table which follows gives expected NC^ emissions for both
uncontrolled and controlled boiler conditions:
Expected NO^ Emissions
Coal Oil Gas
Coal
Oil
Gas
Ib/MM Btu
0.45
0.30
0.20
ppm(v)
338
225
150
Uncontrolled, ppm 500 280 200
Controlled, ppm 370 150-21") 85-110
NO Reduction with
A.
Controls, % 26 25-46 U5-57
It may be concluded that boiler modifications are sufficient to
meet the most stringent present standards for NOX, perhaps with
coal being borderline. These modifications also appear to meet
the projected 1980 goals for oil and gas, but not for coal.
However, none of the projected 1985 goals can be met using only
boiler modifications.
»
Boiler modifications are believed to be the lowest cost NO
control measure. However, neither investment costs nor operating
costs for these changes were available. Therefore, they are not
incorporated in the following sections which address economics of
the control of NO . For example, in these sections, a coal fired
Ji,
boiler is assumed to have 500 ppm(v) NOX in the flue gas and an
oil fired boiler is assumed to have 280 ppm(v).
752
-------
NOV Control by Flue Gas Denitrification--
A
About 40 to 50 flue gas denitrification and denitrification/-
desulfurization processes are reported in the literature. They
are both the wet type, which normally removes both NO and S07 ,
X ^-
and the dry type. A report by TVA studied these processes and
recommended eight for further study (11, p.370). Included in
these are the following processes for which some preliminary
economic data was available:
o UOP - Shell (S02/NOX)
o UOP - Shell (NOX only)
o Asahi Chemical (S02/NOX)
Figure 9-70 shows estimated capital investment costs of these
processes as a function of capacity in MW. An exponent of 0.7
was assumed to prepare the figure. It should be noted that the
flue gas flow is assumed to be 330 moles per hour (MPH) per MW.
Therefore, a 500 MW power plant will produce about 165,000 MPH of
flue gas.
Capital investment costs for the three processes (1977) for
operation in a 500 MW power plant are reported to be (11):
$/KW
UOP-Shell (S02/N0x) 131
Aashi Chemical (S02/N0x) 127
UOP Shell (NOX) 31
Operating costs for the processes are shown below:
Operating Cost»
Mills/KWH $/MM Btu
UOP-Shell (SO /NO ) 5.0 0.5555
fc X
Asahi Chemical (S02/N0x) 7.4 0.8222
UOP-Shell (NO) 1.4 • 0.1555
J^
•Heat rate assumed to be9,000 Btu/hr/KW
753
-------
(1) UOP-SHELL (SO2-NOX)
" " "ASAHI"" ~
200
100
_ 60
- 40 M
a
5
20
100
200 ' 400
POWER PLANT SIZE, MW
600
1000
10
Figure 9-70. Capital investment
for NO /SO- control.
4b
754
-------
These costs are based on treating a flue gas containing 2,580
ppm(v) SO- and 634 ppm(v) NO for the UOP-Shell processes and
2200 ppm(v) S02 and 600 ppm (v) NO for the Asahi process.
Since the UOP-Shell process (SO ./NO ) has a considerably lower
^ X
reported operating cost than the Asahi process, it was selected
for further study. For NO removal only, the UOP-Shell process
was also selected.
NOV Control for Oil Fired Boilers—
A
If a residual fuel such as tar, tar oil, and oil skimmings
produced at a coal conversion plant and containing, typically
about O.U percent N, is burned in a utility boiler, the
uncontrolled NO emissions are expected to be:
ppm(v)
Fuel NOX 207 (45* conversion)
Thermal N0x 73
Total 280
These emissions exceed even the present most stringent standards,
not to mention projected future standards. Therefore, some type
of control will be required. Three alternate strategies were
examined:
o Case 1: Flue gas denitrification alone
o Case 2: Fuel hydrotreating alone
o Case 3: A combination of 100 percent fuel hydrotreat-
ing plus the necessary degree of flue gas de-
nitrification to meet a given standard
Costs of the treatment methods are:
o FG Denitrification = $0.1555/MM Btu at 90 percent NOX
removal (based on UOP-Shell process)
o Fuel Hydrotreating = $0.2918/MM Btu at 90 percent
nitrogen removal efficiency
755
-------
The operating costs for each of these cases are plotted as a
function of the NO emission for each of the cases in Figure
9-71. The basis for these costs was:
Cost basis End 1977
Plant capacity 500 MW
Plant heat rate 4,500 MM Btu/hr
On-stream factor go%
UOP-Shell NO removal: efficiency 90$
J"L
cost $0.1555/MM Btu
Hydrodenitrification: efficiency <)Q%
cost $0.2918/MM Btu
After examination of the information available the following
conclusions may be drawn:
o Case 1, flue gas denitrification alone, is far
superior to the other c^ses. The proposed 1980
standard of 150 ppm NO^ can be met by treating 51.6
percent of the flue gas at a cost of 8.02
-------
I I
0.45
0.40
i . 1 ; • .: .! ..:
.LIMIT - 13.8 PPM ——]-_-
" ' ' ~~'"""'
CASE 3: FUEL HYDRODENITROGEHATION
FLUE GAS DENITRIFICATION •"•
0.35
0.30
Si 0.2
t.
u 0.2
0.1
0.1
0.0
CASE2:FUEL i •"... _
'; \HYDROCENITROGENATION
i : ; \HYDROCENITROGENATION
ISO
200
250
Figure 9-71. Operating costs for NO,
control for oil fired
boilers.
757
-------
fuels with a lower nitrogen content exhibit a higher
conversion to NO .
x
Case 3, the combination of 100 percent fuel hydro-
treating and partial to complete flue gas denitrifi-
cation, is the most costly of the methods examined. A
lower ultimate NO level can be achieved, however.
This limit is expected to be about 14 ppm at a cost of
Btu.
Combined Sulfur Dioxide and Nitrogen Oxide Control
Control of S02/NOX by Fuel Oil Hydrotreating--
Hydrotreating of gas oils (boiling range 650 to 1,100°F) and
residual oils (boiling range >650°F) is a widely practiced scheme
for desulfurization of these fuels. Sulfur present is converted
to hydrogen sulfide which is then fed to a Claus unit for
elemental sulfur production. Sulfur removal efficiencies of 97
percent for gas oils and 92 percent for residual oils are re-
ported (6, p. 5). Hydrodenitrogenation occurs simultaneously as
nitrogen in the feed is converted to ammonia. A nitrogen removal
efficiency of about 80 percent, perhaps as high as 90 percent
can be attained (7, pp. 15, 17).
In Figure 9-72 are shown the estimated capital investments (1977)
and in Figure 9-73 » the operating costs for gas oil and residual
»
oil hydrotreating as a function of feed capacity. The basis for
development of the operating costs is as follows:
Cost Basis End 1977
Plant capacity 500 MW
Plant heat rate 4,500 MM Btu/hr
On-stream factor 90$
Product/feed 0.9H7 .
758
-------
VD
' S_
s •'•
RESIDUAL
OIL
OIL i
(b);
ESTljMAT:
ESTIJHATl
I
! I
ill
T
KELLOGG _
RESEARCH
"NSTiTt TE
68 20 40 60
CAPACITY, BARRELS PER DAY, 1000'S
80 100
200
400
Figure 9-72. Capital investment for fuel oil hydrotreating,
-------
-4
CTl
O
•d
a)
-------
No credit was allowed in development, of the costs for byproduct
naphtha and fuel gas. If the credit were allowed, operating
costs would decrease about 25 percent.
For each capacity a range of costs is shown in Figure 9-72, of
which the upper line is derived from Pullman Kellogg estimates
and the lower from Stanford Research Institute estimates. Sulfur
in feed and product to hydrotreating are:
Feed Product Removal
Residual oil 3«90$ 0.3* 92*
Gas oil 3.25 0.1 97
For calculation of capital costs at various capacities an ex-
ponent of 0.6 was used for residual oil and 0.7 for gas oil.
A large coal conversion facility of 250 billion Btu/day SNG
requires a utility boiler which is about equivalent to a 500 MW
power plant. If the plant is assumed to have a heat rate of 9,000
Btu/hr/KW, then 4,500 MMBtu/hr of fuel are needed. Tars and tar
oils produced in low temperature gasification processes have a
heating value of about 16,500 Btu/lb. The specific gravity of
these materials is about 1.08 resulting in a heating value of
about 148,600 Btu/gallon or 6.24 MM Btu/barrel (42.gallons/-
barrel). To supply 4,500 MM Btu hr, 721.1 barrels/hr or 17,310
barrels/day of fuel are needed. The capital investment cost
(1977) of a residual oil hydrotreater of this capacity is esti-
mated to be in the range of $29 to $35 MM. The operating cost is
expected to be about $2.30/barrel feed without credit for by-
products. After taking a 25 percent credit for byproducts and
adjusting for the slightly reduced output of 94.7 percent of the
feed, the operating cost becomes $1.82/barrel of product or
$0.2918/MM Btu.
Control of S02 and NOX for Oil Fired Boilers—
In previous sections, S0? and NO control for oil fired boilers
fc A.
761
-------
were addressed separately. In this section, the combined control
of these pollutants will be discussed. Three cases were studied:
o Case 1: Hydrotreat only. With full hydrotreatment
the ultimate SO emission is 0.176 Ib/MM Btu and the
ultimate NO emission is 138 ppm.
x
o Case 2(a): Hydrotreat 91.9 percent of the fuel oil
sufficient to meet an SO emission standard of 0.3U
Ib/MM Btu. Remove NO to varying levels with
Jv
UOP-Shell process.
o Case 2(b): Hydrotreat 97.8 percent of.the fuel oil
sufficient to meet.an SO, emission standard of 0.22
Remove NO to varying levels with the UOP-
• Jx
Ib/MM Btu.
Shell process.;
Case 3: No hydrotreating
by UOP-Shell .process.
Combined SO-/NO removal
Figure 9-74 is a plot of operating cost vs
S02 and NOX emissions
for the cases.
cases.
TABLE 9-30 .summarizes emissions and costs for the
Costs were developed from the following basis:
Cost basis
Plant capacity
Plant heat rate
On-streara factor
Fuel hydrotreating:. sulfur.,removal ;
nitrogen removal
cost
UOP-Shell NO removal
cost
UOP-Shell SO /NO removal:
£ A
SOV NO removal
2 x
cost
End. 1977 .
500,,MW, ,
> - ' ', - •
45,00 MM. Btu/hr
90* r
925 *
901,
$0... 2 9 4,8 /MM,,
90*
90$
$0.5555/MM Btu
-762
-------
TABLE 9-30. CONTROL OF NOX/S02 FOR OIL FIRED BOILERS
Full Hydrotreat (100»
SO,
Ib/MM Btu
NO ,ppm
Colt, $/MM Btu
Partial Hydrotreat (97-8J)
S0_, Ib/MM Btu
NO , ppm
Co§t, $/MM Btu
Partial Hydrotreat (91.9*)
SO,
Ib/MM Btu
NO , ppm
Colt, $/MM Btu
Full Flue Gas Treat (100*)
SO-
Ib/MM Btu
Case 1
Hydrotreat
Only
0.176
138
0.292
0.220
111
0.285
0.310
118
0.268
Case 2(
Hydrotreat 91
UOP-Shell NO
0.310
90
0.336
0.310
13
0.391
a)
.9% plus
Process
0.310
28
0.109
Case
Hydrotreat
UOP-Shell
0.220 0
90
0.318 0
2(b)
97. 8% plus
NO Process
.220
13
.101
0.220
28
0.124
Case 3
UOP-Shell
SO-/NO Process
NO' ppm
CoSt, $/MM Btu
Partial Flue Gas Treat (93.9*)
SO-
Ib/MM Btu
NO i ppm
Colt, $/MM Btu
0.220
28
0.555
0.310
13
0.522
-------
NO EMISSION, PPM(V)
X
0.5
1.0
S02 EMISSION, LB/MM BTU
1.5
2.0
Figure 9-74, Operating costs~for NO /S09
; X ^
control for oil fired boilers.
-------
The following conclusions may be drawn:
o To meet an SCU emission standard of 0.34 Ib/MM Btu and
the proposed 1980 NOV goal of 150 ppm, Case 1 with 91.9
X
percent hydrotreating is clearly superior, with a cost
of $0.268/MM Btu
o To meet the same S02 standard and the proposed 1985 NOX
goal of 90 ppra, Case 2(a) is preferred with 91.9
percent hydrotreating and UOP-Shell NOX and a cost of
$0.336/MM Btu. It is possible that a combination of
hydrotreating and boiler modifications (which reduce
NOV emissions by 30 to 40 percent) may suffice to meet
X
the 1985 goal. If so, this case would cost less than
Case 2(a). The NOV emission can be lowered to 43 ppm
A
by treating more flue gas, but with a cost of $0.391/MM
Btu.
o Case 3 does not appear to be competive with Case 2(a).
To meet S02 standards of 0.34 Ib/MM Btu and 43 ppm NOX
costs $0.522/MM Btu.
Control of S02/NOX for Coal Fired Boilers—
Two cases were analyzed for control of SO? and NOX from coal
fired boilers:
o Case 1: Removal of NOV by the UOP-Shell process and
J^
removal of S(>2 by the citrate process
o Case 2: Removal of both S02 and NOX by the UOP-Shell
process
Costs were developed for the two cases on the following basis:
Cost Basis End 1977
Plant capacity 500 MW
Plant heat rate 4500 MM Btu/hr
On-stream factor 90/t
S02/NOX removal efficiency 90$
765
-------
Operating costs: UOP-Shell N0x $0.1555/MM Btu
Citrate S02 $0.3889/MM Btu
UOP-Shell S02/N0x $0.5555/MM Btu
If all of the flue gas is treated for both cases with 90 percent
removal efficiency for both S09 and NO , the costs and emissions
^ A
for the two cases are:
Case 1 Case 2
Operating cost, $/MM Btu 0.5444 0.5555
SO , Ib/MM Btu 0.48 0.48
N0x, ppm(v) 50 50
As shown in Figure 9-75, no clean-cut choice can be made based on
operating costs. The SO,, emissions obtained are adequate for
most states, but they exceed the most stringent standard of 0.2
Ib/MM Btu set by Wyoming. The NO emissions are lower than the
projected 1985 goal of 100 ppm(v).
If 88.9 percent of the flue gas is treated, so that the NO
emissions are 100 ppm for both cases, the operating costs and
emissions are:
Case J^ Case 2
Operating cost, $MM/Btu 0.484 0.494
S02, Ib/MM Btu . 0.96 0.96
NO , ppm(v) 100 100
Again, the operating costs are too close to choose between the
cases.
The SO emissions meet the present standards set by most states
but obviously not the most stringent standards.
766
-------
S02 EMISSION, LB/MM BTU
•vj
0>
-J
100
200
300
400
500
NO EMISSION, PPH(V)
X
Figure 9-75, Operating costs for SO /NO
^ H
control for coal fired boilers.
-------
In order to meet the most stringent standards of 0.2 Ib S02/MM
Btu and 100 ppm NO , Case 1 appears to be required because of its
a
higher SO removal capability. The costs and emissions to meet
these standards are obtained by partial treatment of the flue
gas, before the air heater, by the UOP-Shell process and full
treatment of the flue gas after the ESP by the citrate process
operated to obtain 95.8 percent removal efficiency:
Case 1
Operating cost, $/MM Btu
UOP-Shell N0x 0.1382
Citrate .40
Total 0.5382
S02: Ib/MM Btu 0.2
NOX: ppm(v) 100
If Case 2 can be operated to achieve 95.8 percent S02 removal, it
will probably serve as well.
Production of Elemental Sulfur from H2S and Control of
Organic Sulfur Emissions
The Glaus process and the Stretford process are mo'st nearly
universally applicable in converting H2S-rich gas from a solvent
system in a coal conversion plant to elemental sulfur. Neither of
these processes produces a tail gas that is environmentally
acceptable, a circumstance that has led to combination processes
that minimize pollution from H2S, COS, CS2,and S02:
Processes Combined Processes Combined
with Claus ._ with Stretford
Beavon Hot Carbonate
SCOT Holmes-Maxted
ARCO Carpenter-Eyans
Incineration British Gas Council
Lucas Iron Oxide
768
-------
Selection of the optimum processing technique can 'depend on a
number of factors, including:
o The relative amounts of C02 and H2S in the acid gas
(C02/H2S ratio)
o The presence of organic sulfur compounds (COS, CS2) in
the feed gas
o The presence of ammonia and hydrocarbons in the feed
o The volumetric flow of acid gas
o The concentration of H2S
o The emission standards in effect for various sulfur
species
o Reducing gas cost and availability
o Relative costs of raw materials, utilities and man-
power
o Capital related costs
It was beyond the scope of this task to evaluate all of these
variables. Some economic data for the Claus/Beavon combination
were supplied by the Ralph M. Parsons Co. Capital costs for
these plants are plotted as a function of capacity in Figure
9-76, using the exponent 0.6.
The H2S concentration in the feed is shown to have a considerable
effect on the investment required. Decreasing the H2 S content
from 40 to 10 mol percent results in a 70 percent increase in
capital investment for the same sulfur production. Therefore, it
is clear that a concentrated H2S feed is highly desirable.
Although no specific economic data were supplied for the other
processing schemes, Parsons stated that a Claus/ARCO unit
typically costs about 15 percent more than Glaus-Beavon.
769
-------
20
40
60 80 100 200
SULFUR PRODUCTION, STPD
400 600 800 1000
2000
Figure 9-76. Capital investment for Claus-Beavon
sulfur recovery.
-------
Estimated operating costs for the Claus-Beavon combination are
shown in Figure 9-77, developed with the following criteria:
Cost basis 1977
On-stream factor 90$
Capital charges 25%
Steam credit $3/1000 Ib
Fuel cost $3/MM Btu
Power cost $0.025/KWH
Labor cost 2 men/shift ($M80/day)
References
1. Lund, H. F., "Industrial Pollution Control Handbook."
McGraw-Hill, 1971. 900*
2. Fluor Utah, Inc., "Economic System Analysis of Coal Pre-
conversion Technology, Phase I, Volume 4: Large Scale Coal
Processing for Coal Conversion." July 1975. U21*
3. The M. W, Kellogg Co., (Pullman Kellogg). "High Sulfur
Combustion Assessment, Task No. 30 Final Report." EPA
Contract No. 68-02-1308, February 1975. 895*
M. TRW Systems Group, "Meyers Process Development for Chemical
Desulfurization of Coal, V.olume I." EPA-600/2-76-lH3(a), May
1976. 276(a)»
5. Exxon Research and Engineering Co., "Application of Fluidized
Bed Technology to Industrial Boilers." January 1977. 531*
•Pullman Kellogg Reference File number
771
-------
to
300
200
100
80
60
40
J+i
I'll
-L-l-o.-:
it i
20
U.L.
I i
uu
ffll
^
IT
TTT
il
i
ti
l\
FEEI
1C
4(
' I
IOL! *•
; ;
CONTENT OF
3ASIS
i
HOL<%
I i
I'M
,, i
It!
111
10
20
40. 60 80 100 200
SULFUR PRODUCTION, STPD
400 600 800 1000
.Figure 9-77. Operating costs for Claus-Beavon
sulfur recovery.
-------
6. Stanford Research Institute, "Petroleum Desulfurization."
Supplement A, Report No. 47A, July 1975.
7. Satchell, D. P., "Development of a Process for Producing an
Ashless, Low Sulfur Fuel From Coal, Volume IV - Product
Studies, Part 6 - Hydrodenitrogenation of a Coal Derived
Liquid." May 1974. »232
8. Ricci, L. J., "EPA Sets Its Sights on Nixing CPI's NOV
Ji
Emissions." Chemical Engineering, February 14, 1977.
9. J. J. Davis Associates, "Coal Preparation Environmental
Engineering Manual." May 1976. »300
10. Do, N. Loan, and Hunter, W. D., "NO Control Technology." Pullman
X
Kellogg Report No. RD-77-1342, September 1977.
(Confidential).
11. Faucett, H. L., Maxwell, J. C. , and Burnett, T. A., "Technical
Assessment of N(^ Removal Processes for Utility Applica-
tions." November 1977.
12. Tennessee Valley Authority, "Detailed Cost Estimates for Ad-
vanced Effluent Desu1furization Processes."
EPA-600/2-75-006,January 1975. »279
»
13. Torstrick, R. L., Benson, L. J.,and Tomlinson, S. V., "Economic
Evaluation Techniques, Results and Computer Modeling for Flue
Gas Desulfurization." U.S. EPA Flue Gas Desulfurization
Symposium, November 1977.
14. Kohn, P. M., "CE Cost Indices Maintain 13-.Year Ascent."
Chemical Engineering, May 8, 1978.
773
-------
15. Griebe, M. H., The Ralph M. Parsons Co. Communications
with Pullman Kellogg from March through May 1978.
16. The M. W. Kellogg Co. (Pullman Kellogg), "Evaluation of the
Controllability of Sulfur Dioxide Emissions for Iowa Power
Boilers". EPA-650/2-74-127, December 1974.
17. Madenburg, R. S., and Kurey, R. A., "Citrate Process
Demonstration Plant. A Progress Report." U. S. EPA Flue Gas
Desulfurization Symposium, November 1977.
NEED FOR ADDITIONAL DATA, INFORMATION AND DEVELOPMENT
Coal Pretreatment
Sulfur must be removed from coal before conversion or be re-
covered in the conversion processes. Physical pyritic sulfur
removal by, for example, heavy media washing, is incomplete but
relatively inexpensive, and the sulfur values are usually thrown
away. Chemical pyritic sulfur removal by, for example, the Meyers
process, is effective in reducing sulfur and yields salable
sulfur and iron sulfate that would probably be thrown away.
Neither process is effective in removing organic sulfur. Use of
either type would reduce the requirement for sulfur recovery and
sulfur dioxide control in the coal conversion processes.
Development of economic data on coal cleaning methods and on the
facilities within the coal conversion process is required for the
making of meaningful decisions to:
o Reduce sulfur by physical coal cleaning and recover the
rest in the conversion plant
•774
-------
o Reduce sulfur by physical coal cleaning, recover part in
the conversion plant and throw away the rest as FGD sludge
o Reduce sulfur by chemical coal cleaning, recover or thow
away the sulfur and recover or throw away the remaining
coal sulfur in the conversion plant
More information is needed on operability of the Meyers process.
Particulates
Further study and evaluation are needed on the quantities and
compositions, including particulates, of gases released from coal
feeding devices such as lock"hoppers.
More data are needed on paniculate collection efficiencies and
costs as applied to such coal dust control devices as cyclones,
baghouses and electrostatic precipitators.
Data on quantity and particle size distribution of dusts evolved
in coal crushing, sizing, pulverizing and drying steps would be
helpful in developing control methods and meaningful economics.
Ash Quench
While much data and information are available on gases evolved
from quenching in Lurgi Dry Ash gasification, little real data
have been collected on the gases evolved from other processes.
If, as in many conceptual designs, ash is quenched with
contaminated water, the contaminants themselves may volatilize,
together with products of reactions of the contaminants with the
ash, or the ash with water.
775
-------
Data are needed on quantity and composition of the ash quench
streams from the various conversion processes and the variations
for any one process as the quench water composition varies.
Particulate Removal from Gasifier Offgas
Particulates are carried in the gasifier offgas as ash, char, and
unreacted coal. When the gas stream is water quenched for cool-
ing and tar condensation purposs, as in the Lurgi process, the
particulates are removed from the gas stream by the quench water.
Since the Lurgi process has been in commercial operation for
many years, the problems of particulate separation from the
liquid streams have reached satisfactory solutions.
In the high temperature slagging gasif-'.ers little or no phenols
oils and tars are produced and direct quenching of the raw gas
stream for cooling and condensation is not practiced. In the
Parsons conceptual design for a liquefaction plant, for example
there are two slagging gasifiers, one oxygen blown to produce
process gas and the other air blown to produce fuel gas for pro-
cess heating and steam generation. In both of these, the raw gas
carries char and the char is mechanically separated from the gas
streams by combinations of cyclones, dust filters and elec-
trostatic precipitators. While it is permissible to stipulate
particulate removal efficiency in a conceptual design for process
purposes, final engineering design of a full scale conversion
plant requires as exact a set of data as can be assembled.
Indications from the literature, concerning investigations on
high temperature, high efficiency, high volume cyclone designs
(which parameters are usually considered as being mutually exclu-
sive) , are that much more data are needed on performance at ex-
pected operating conditions. The same impressions are gained
concerning hot gas filters. Electrostatic precipitator
776
-------
performance on hot fly ash is well documented and, given the dust
loading, resistivity and operating conditions, design should
offer no unusual difficulty.
.Acid Gas Removal
While removal of hydrogen sulfide and carbon dioxide from the
process gas stream is not usually considered to be an emissions
problem, the composition of the hydrogen sulfide stream affects
the performance of the sulfur recovery system, and thus affects
the treatment that must be applied so that the final vent gas
stream will meet environmental standards. In a like manner, the
composition of the carbon dioxide stream affects the performance
of downstream process steps and thus affects the vent gas
treatment step.
Suggested avenues of investigation in several of the acid gas
removal processes are:
o Selexol: Study effect of feed acid gas composition on
removal efficiency at various operating
temperatures and pressures
o Rectisol: Determine the solvent retention of heavy
hydrocarbons and the composition and quantity
of the fugitive carryover from the process at
high pressures
o Monoethanolamine (MEA): Study the effects of operating
conditions on the formation of non-regenerable
compounds, on excessive solvent losses, on
corrosion and on foaming
777
-------
o Diisopropanolamine (DIPA): Study effect of operating
pressure on hydrogen sulfide removal efficiency
o Diglycolamine (DGA): Study the effects of operating
conditions on the formation of non-regenerable
compounds and the effect of feed acid gas
composition on removal efficiency at various
operating temperatures and pressures
o Diethanolamine (DBA): Study means of removal of the fine
particles that cause foaming, as removal
efficiency vs. operability vs. cost. Data are
needed on utilities requirements vs. operating
temperature and pressure
o Fluor Solvent: Determine utility requirements and study
effect of feed gas composition on removal
efficiency at various operating temperatures
and pressures
o Sulfinol: Determine solubility of hydrocarbons in the
sulfinol solvent and study process economics
vs. operating parameters
o Estasolvan: Study the effect of operating pressure on
acid gas removal efficiency. Study methods of
treatment for the blowdown stream
o Benfield: Study the process when it is operated selec-
tively for the hydrogen sulfide content in the
carbon dioxide stream. Determine the extent of
COS hydrolysis vs. the requirements for
Stretford process feed
778
-------
o Araisol: Determine utility requirements and study effect
of feed gas composition on removal efficiency
at various operating temperatures and pressures
Sulfur Recovery and Tail Gas Cleanup
Further study of the sulfur recovery and tail gas cleanup process
should include characterization of inlet and outlet gas streams,
vent streams, byproducts, sulfur removal efficiency vs. operating
parameters, and reactant degradation. Suggested areas of
investigation in examples of processes include:
o Glaus process: Determine extent of removal of HCN and
ammonia from the feed gas stream and fate of CO
and hydrocarbons in the feed gas. Study the
effect on sulfur conversion of the presence of
oxidizable compounds in the feed gas. Determine
the economics of the process for operation on
various feed gas compositions, with particular
emphasis on the effects of variations in
hydrogen sulfide concentration
o Stretford and Beavon processes: Determine conversion of
organic sulfur compounds in the presence of
high concentrations of carbon dioxide.
Characterize oxidizer vent gas stream and
solvent blowdown stream. Determine degree of
removal of mercaptans and ammonia
779
-------
Sulfur Dioxide Control Processes
For the citrate process, data are needed on the effects on sulfur
dioxide removal and on the process economics of changes in feed
gas composition.• These same data are needed for the Kellogg-
Weir scrubber. For the Wellman-Lord process, data are needed on
the efficiency of removal of HCN and ammonia.
For all processes, data and information are needed to supplement
those already available on composition and means of disposal of
the products from each process, such as sulfur, sulfuric acid,
concentrated sulfur dioxide,and sludges.
These examples are cited to illustrate the types of data that are
needed for process and economic evaluation of sulfur dioxide
control processes.
Hydrocarbon Control Processes
The hydrocarbon content of waste gas streams can be reduced or
eliminated by incineration, by absorption,or by adsorption.
*R. S. Madenburg of the Morrispn-Knudson Company, Boise, Idaho
presented a paper, "Industrial Application of Citrate FGD
Technology," at the June, 1978 meeting of the Air Pollution
Control Association, indicating the areas of use of the process
its efficiency and costs. More data are needed, however, on
process variations such as feeding dilute hydrogen sulfide at
elevated pressure in order to maintain a partial pressure of
hydrogen sulfide at 10 to 14 psi.
780
-------
Incineration by oxidation at high temperature is attractive as a
means of hydrocarbon destruction because the waste gas stream can
be introduced directly into the utility boiler. Boiler design is
affected by the quantity and composition of the waste gas streams
and data are needed on the economic effects of these design
changes.
Absorption into the solvent of low vapor pressure followed by
stripping and solvent regeneration has been proposed. More data
are need on process types, efficiencies and economics as these
are affected by stream composition and quantity.
Carbon adsorption is effective for removal of some hydrocarbons
and, like absorption, yields' a concentrated hydrocarbon stream
upon regeneration. Data are needed on efficiencies and economics
vs. gas stream composition and quantity.
Nitrogen Oxides Control Processes
Nitrogen oxides emissions can be controlled by combustion
modification in boilers and furnaces, by using fluidized bed
combustion techniques, by hydrodenitrogenation of li-quid fuels,
and by flue gas cleaning. The available data on these methods
are drawn from power plant boiler experience and petroleum
refinery experience. Data are needed on application of the
techniques to the combustion problems peculiar to coal conversion
plants, particularly where the utility boiler may also act as an
incinerator for waste process streams. These data should include
the effects on economics of the installation of the control
methods at various levels of control efficiency. Investigations
might be started in pilot plants and later be extended to full
scale plants.
781
-------
As an example of the types of data needed, suggested areas of
investigation into fluidized bed combustion (FBC) are as follows:
o Determine the sulfur retention in the ash from FBC and
spreader-stoker boilers for representative coals and mixed
fuels
o Determine nitrogen oxides emissions for the same boilers
and fuels
o Determine the fate of trace elements, including flue gas
desulfurization systems
o Develop economics as a spur to development of the process
For UOP-Shell, Hitachi-Zosen and other processes more
investigation is needed into the effect on the processes of
particulates from coal fired boilers.
More data are needed on nitrogen oxides formation in boilers with
and without combustion modification in order to determine the
benefits of the modifications. Development of cost data is
needed.
For liquid fuel denitrogenation, data are needed on the nitrogen
content and type in coal derived liquids. Study of denitro-
genation of these liquids is needed, together with development of
economic data.
Exxon Research and Engineering Company is developing and applying
a system described in "Reducing Nitrogen Oxides Emissions by
Ammonia Injection," a paper given by R. K. Lyon at the June, 1978
meeting of the Air Pollution Control Association. The described
system achieves reductions of around 70 percent by injecting
782
-------
ammonia into the boiler flue gas. There have been 8
demonstrations in commercial boilers and furnaces, to date.
Further development of applications, efficiencies, and costs for
this simple process should be encouraged.
Cooling Tower Drift
More technical and economic studies are required to develop more
data on means of reducing cooling tower drift. Economic
projections would be useful in contrasting high velocity cooling
tower designs with low velocity and hyperbolic designs.
Lock Hopper Vent Gas
Study is needed to determine the optimum type or types of lock
hopper operation and methods of disposal of excess lock hopper
vent gas in the various coal conversion processes.
Developent of extrusion type coal feeders for high pressure
reactors is progressing. Use of these would reduce or eliminate
vent gases. Inclusion of such process wastes as tars, oils and
biological oxidation sludges with the coal feed might aid the
operation of the feeders and incidentally solve a disposal
problem.
783
-------
SECTION 10
ENVIRONMENTAL DATA ACQUISITION: CONTROL OF SOLID WASTES
The many types of solids with widely varying chemical and
physical characteristics that are involved in the operations of
coal conversion plants pose a variety of problems concerned with:
Fugitive dusts from:
Bulk handling of:
Coal storage and reclaiming operations
Coal storage piles
Ash and scrubber sludge piles
Limestone storage piles
Sulfur storage, handling,and shipment
Coal
Coal fines and dust
Conversion process ash slurry
Incinerator/Boiler ash
Scrubber sludge
Spent catalysts
Evaporated solids
Limestone
Sulfur
784
-------
Disposal of: Coal fines and dust
Ash
Sludges
Spent catalysts
Evaporated solids
These many problems may be broadly grouped into the general
categories of dust control and solid waste management. Efforts
in both categories are directed toward minimizing or eliminating
deterioration of the environment. In the subsections that
follow, the overall problems are examined and controls are
proposed to meet existing and proposed environmental standards.
Developing control technology is examined in view of future
environmental goals. In all cases the viewpoint of the
conversion plant operator is taken: solutions for environmental
problems should be realistic, operable,and economical.
LITERATURE SURVEY AND DATA GATHERING
In the survey of available literature on coal conversion and
related processes data and information were gathered on the
evolution of fugitive dusts, methods of control that are in
general use, suggested means of more efficient dust control, the
extent to which solids disposal problems are recognized, proposed
means of solving solids disposal problems and possible future
development of more efficient or lower cost solids management
methods. Close collaboration was maintained with the effluent
and emission study groups in this project for recognition and
solution of problems that required consideration of more than one
state of matter.
785
-------
Personal Contacts, Trips.and Meetings
In addition to the list of contacts and meetings that was given
in Section 5 of this report, there were others that yielded
information on solids handling:
o A visit to the Public Service Company in San Antonio,
Texas, to observe coal receiving and stockpiling
operations including dust suppression, reclaiming,and
handling of dust and fines
o A meeting with Dravo Corp. engineers for a general
discussion of coal and dust handling and a particular
discussion of the Bi-Gas pilot plant at Homer City, Pa.
Visits to the pilot plant, a coal fired power generating
station and U. S. Steel's coal preparation plant in
Kirby, Pa.
o Attendance at the Coal Waste Technology Seminar in
Houston.
Telephone contacts (TC) and correspondence (C) were held with the
following:
o Dravo Corp. - F. A. Zuhl (TC). Information on "Calcilox"
waste solidification system
o
Cherafix Inc. - Alan Cohen, Richard Patent (TC).
Information on I.U. Conversion Systems solidification
process
o McDowell-Wellman Engineering Co. - J. Wellman (TC). Dust
control at rail car dumping stations
'786
-------
o Allis Chalmers Co. - B. K. Smay (TC). Dust control
o EPA Office of Solid Waste Management Programs - Allan
Geswein and Robert Landreth (TC, C). Liners for land
disposal sites
o Austin (TX) Water Control Board - J. Snow, J. Carmichael,
G. Maxon (TC, C). Ash evaporation ponds and liners
o DuPont Inc. - Phillip Rizzo (TC), C. S. Glover (C).
Information on pond liners
o American Colloid Co. - William Hahn (TC, C) . Bentonite
in waste disposal areas
o Dowell Division of . ow Chemical Co. - Dan Hunt (TC, C) .
Soil sealant
o B. F. Goodrich General Products Co. - H. F. Cumraings (C).
Hypalon liners
o National Ash Association, Washington, D. C. (TC).
Disposal pond operation
o Johnson-March Co. - C. L. Burchsted (TC, C) . Dust
suppression
TARGET POLLUTANT RESIDUALS
Guidelines and standards of federal, state, regional,and
international jurisdictions for solid waste disposal requirements
were reviewed and synopsized in the same manner as were those for
air emissions and liquid effluents.
787
-------
Solid Waste Disposal and Management
The majority of the solid waste disposal requirements are much
less definitive, with regard to establishing design requirements,
than those criteria established within the air and water regu-
latory areas. The regulations tend to establish requirements
directed more toward the operation of a disposal facility, such
as adequate rodent control and proper compaction and cover for
solid waste, than to the design.
It should be expected that the regulatory activity in this area
and especially with respect to hazardous wastes will continue to
increase as a result of the Solid Waste Disposal Act as amended
by the Resource Conservation and Recovery Act of 1976, Title II
Solid Waste Disposal (42 USC 6901 et s«q.).
One provision generally common to the states reviewed allows for
solid waste disposal on one's own property without a permit so
long as no nuisance conditions are created.
Texas, one of the states surveyed, has issued Technical Guide-
lines for solid waste disposal and indicates that by following
these guidelines all solid waste disposal requirements will be
satisfied. These Technical Guidelines are available from the
Texas Water Quality Board, which has responsibility in this
area.
In "Guidelines for the Land Disposal of Solid Wastes" (
is the EPA requirement that location, design, construction, and
operation of land disposal sites shall conform to the most
stringent of applicable water quality standards. States require
leachate collection and treatment and may require water
monitoring (N. Dak.) or leaching and drainage are to be prevented
(Ky., Mont. , Wy.).
788
-------
The Dominion of Canada sets criteria for impermeable soils for
disposal sites:
o Permeability to be not more than 10~6 cm/sec
o Not less than 30 percent passing U.S. standard 200 mesh
sieve
o Interceptor drains or collectors may be required
Other most stringent landfill standards and guidelines applicable
to ash and inorganic sludges are as follows:
o Dust control measures are required (N. Dak.)
o Cover the fill site at least every 20 days (Br. Col.)
o Slope the cover 2 to 4 percent (Can. Fed.)
o Neutralize and dewater sludges prior to storage (Br.
Col.)
o Compaction is required (Alaska, Pa.)
o Closure with 24 inches of compacted earth (N. Dak.)
o Surface drainage to be consistent with surrounding area,
not to cause interference with adjacent drainage nor
allow runoff to become concentrated (N. Dak.)
o Area to be seeded after closure (N. Dak.)
o No solid waste is to be discharged into ground or surface
waters (Tex.)
789
-------
Pennsylvania has regulations applying to coal mining refuse piles
which presumably could be used as solid waste disposal
guidelines:
o Maximum height of the pile to be 100 feet above average
elevation of the immediate area
o Refuse is to be deposited in layers no more than 2 feet
thick and is to be compacted
o Surface water is to be managed (in a manner similar to
that described previously).
Fugitive Dusts
Fugitive dust standards and guidelines are similar to those for
solid waste disposal in that there are few numerical values and
much subjective direction:
o "No material shall be handled, transported, stored, or
disposed of... .without taking reasonable precautions to
prevent particulates from becoming airborne." Use of
water, asphalt, or oil emulsions to control dust is
suggested (N. Hex., Ala. and others)
o "Coal preparation plants. All crushers, conveyors
screens, cleaners, hoppers,and chutes, which are designed
for continuous transportation or preparation of coal
shall be equipped with hoods, shields,or sprays where
reasonably necessary to prevent airborne particulate
matter" (N.Mex.)
o "....without taking reasonable precuations to prevent
particulate matter from becoming airborne in amounts
which cause a public nuisance" (N. Dak., Okla. Va.
Wash.)
• 790
-------
o "....without taking reasonable precautions to prevent
particulate matter from becoming airborne" (0., Tenn.)
o "Any source must have control equipment to minimize
emissions" (W. Va.). Applies to ash or fuel piles,
transport and handling
o Precautions are to be taken to control fugitive dusts
through use of water, chemicals or covering (Texas)
There are a few numerical values stated:
o Fugitive dusts shall not exceed 20 percent opacity except
when the wind exceeds 30 MPH (Colo.)
o No particles larger than 40 microns mean diameter shall
be airborne beyond the property line except when the wind
exceeds 25 MPH (111.)
o Maximum allowable .round level concentration at the
property line is 2 milligrams per cubic meter above such
ground concentrations for periods of not more than 10
minutes in any hour (Kan.)
o Cannot exceed ambient air standards (Wy.)
o Visibly passing property lines is prohibited (Colo.,
Okla.,and others)
Odors
With the exception of Missouri, West Virginia,and Wyoming, which
describe apparatus or evaluation methods, the statements
concerning odors are all subjective, as "malodorous air shall not
be emitted such that odors are detected beyond the property
line," "no odor that would be objectionable to a person of
ordinary sensibility shall be emitted," "best available control
technology shall be used,""best practical control technology
shall be used," "prohibited to allow emission of odorous gases,
liquids or solids in such quantities as to cause air pollution."
791
-------
Consideration of these regulations leads to certain conclusions
regarding odor control in solids handling and waste disposal in
caol conversion plants:
o Spontaneous combustion in coal storage piles should be
avoided to avoid emissions of hydrogen sulfide, sulfur
dioxide,and other odorous materials
o Organic materials should not be allowed in ash and
ash/sludge disposal areas
o pH control of the liquid portion of the ash slurry to
disposal may be necessary to prevent odor-causing reactions
within the disposal area
DUST CONTROL
Suspension of finely divided solids in ambient air as fugitive
dusts is governed by the size, shape, and density of the solids
and by the air velocity and direction. In air suspension or by
settling in areas surrounding coal conversion plants the dusts
can become a fire hazard (in the case of coal and sulfur), a
physical hazard, a contributor to deterioration of the environ-
ment by discouraging plant growth or a toxic factor in the envi-
ronment (through inhalation by man and animals or by the leaching
out of soluble compounds into adjacent land and water) . Control
of these dusts is important from the environmental standpoint and
from the economic considerations of loss of usable materials or
of loss of useful land.
Fugitive dust clouds are generated when a stream of dry solids
falls freely in air as at a conveyor transfer or discharge point
when the stream of solids is agitated as at rail car dumping or
reclaiming from pile storage and when wind blows across piles of
solids. There is little quantitative data on fugitive dust
792
-------
generation. Available information is principally concerned with
gross material losses and not with measurement of cause versus
effect. Process operators, however, are making efforts to reduce
dust evolution from materials handling operations and, aided by
equipment manufacturers, have developed systems for dust control
that may be characterized as dust suppression and dust
collection.
Definition of the Problem
Run-of-tnine coal is crushed and screened during the preparation
operations for removal of inerts and pyrite-bearing particles.
The size distribution of the crushed particles is affected by:
o Coal source and type
o Crusher type
o Crusher design
o Crusher discharge opening
o Crusher speed
o Coal grindability
The effect of these variables on the size distribution of the
crushed particles is illustrated in TABLE 10-1 for crushers
designed by two manufacturers operating on several types of
coal.
TABLE 10-2 contrasts the operation of these two crushers and a
third model all operating with 1.25 inch crusher opening on
various coals. The size data are shown in graphical form in
Figure 10-1 .
Other data on size consists is reported for Indiana and Illinois
coals (2) and for "western" and "eastern" coals (3) and is shown
in TABLE 10-3. Unfortunately, the crusher size opening was not
specified and comparison with the data in TABLES 10-1 and 10-2
793
-------
TABLE 10-1. SIZE DISTRIBUTION OF PRODUCTS FROM COAL CRUSHING
CRUSHER A» ON ILLINOIS NO. 6 COAL, % RETAINED ON SCREENS
Screen Opening
1.50 in.
1.25
1.00
0.75
0.50
0.375
0.25
0.125
10M»»
28M
Crusher Opening, In. (S=Slow Speed, F=Fast Speed = 1.65 to 1.95S)
2. DOS
15
30
19
68
75
81
91
95
98
1.25S
1.5
12.9
28.9
19-7
61.1
72.8
82.9
87.0
93.0
1 .003
1.0
5.3
15.1
38.0
50.1
65.1
80.6
81.3
92.1
0.75S
2.0
1.5
32.2
15.7
61.9
77.2
82.7
90.9
0.50S
2.9
16.0
26.5
57.1
71.2
82.9
0.25S
1.2
30.5
17.8
71.5
1.25 F
3.1
9.1
25.1
15.6
57.8
68.2
78.7
83.3
89.8
1.00F
2.8
11.3
31.1
15.9
61.5
76.1
80.2
87.8
0.75F
0.2
6.2
23.3
11.3
57.1
72.8
77.6
85.1
0.50F
1.2
1.1
10.2
31-7
52.2
65.5
78.3
0.25F
2.1
26.3
13.6
70.3
to
CRUSHER B» ON VARIOUS COALS. % RETAINED ON SCREENS
Screen
Opening.
2.00 in.
1.75
1.50
1.25
1.00
0.75
0.50
0.375
0.25
0.125
10M««
28M
Crusher. W. Ky.
Opening, In. 1.50
2.7
12.6
31.9
51.3
71.1
85.3
W. Ky.
1.00
6.8
31.8
58.1
77.9
88.2
95.2
Ind.
1.00
5.1
31.3
61.3
71.5
81.6
91.9
111.
1.25
5.5
21.0
52.3
73.1
88.5
91.9
111.
0.75
11.0
50.5
80.9
91.9
Ohio
2.0
~076~
7.1
51. Oq
65.5
77.2
88.2
Ohio
1.50
3.0
17.3
11.5
67.8
83.2
Ohio
1.25
0.5
11.2
35.1
61.2
79-3
88.6
W. Va.
1.50
3.6
28.1
17.9
68.5
80.2
91.6
W. Va.
1.25
8.8
26.9
55.9
73.1
88.3
W. Va.
1 .00
2.1
20.6
51.6
78.1
Wy.
2.0
~5TS
10.7
27.5
37.3
18.9
62.1
69-3
77.6
85.7
* Crusher manufacturers' names confidential, by request
•• Tyler standard mesh size designation
-------
TABLE 10-2. COAL CRUSHING WITH 1.25-IN. OPENING
Screen
Size
1.25 in.
1.00
0.75
0.50
0.25
0.125
10M»*
28M*»
i
(Percent Retained on Screens)
Illinois
A*
4.5
12.9
28.9
49.7
72.8
82.9
87.0
93.0
B*
5.5
24.0
52.3
73-1
88.5
94.9
W.
C»
7.2
22.5
46.8
66.1
82.0
Ky. Ohio
B C
3.9 23.0
17.9 35.7
44.5 49.5
65.4 64.9
81.9 79.7
87.0
B
0.5
11.2
35.4
61.2
79.3
88.6
W. Va.
B
8.8
26.9
55.9
73.4
88.3
* Crusher manufacturers' names confidential, by request
** Tyler standard mesh size designation
795
-------
Q
U
z
Of
W
2
3
O
i.o
fc.O
)6-o
1.0
1 0
Jo
4.e
So
09.0
• 1
.3
• 4
.5
.C,
.ft
99-9
SU
KO
S
\
5
)\
\s
\
VL
^
V
CUR\ E CI
\
\\
\ '
\
\
\
USHEF
\
A
\
COAL
111.
rhin
JLl.
. Va
\>
\
\
\
\
LI CN
CO f
O
INCHES
O
O
o o
o o
-------
TABLE 10-3. SIZE CONSIST OF AS-RECEIVED COALS
(Percent Retained on Screens)
A* B_ C D E F
2.0
1.0
12.0 8.5 3.3
27.9 12.9 14.0
44.5 47.0 24.0 34.5
43.6 41.0
57.5 71.4 43.0 63.4
87.3 74.0
74.0 80.9
88.0 90.0
97.7
200M 98.7
1.500 in.
1.250
1.000
0.750
0.500
0.375
0.250
0.125
10M»*
28M
100M
3.0
12.0
21.0
45.0
61.5
83.5
3.0
15.5
44.5
57.5
74.0
87.0
» A = Old Ben No. 1, Indiana Strip Mine Washed (2)
B = Old Ben No. 24, Illinois Raw Coal (2)
C = Consolidation Norris, Illinois Strip Mine (2)
D = Old Ben No. 26 , Illinois Deep Mine (2)
E = Western Coal (3)
F = Eastern Coal (3)
»» Tyler Standard Screen Mesh
797
-------
can be made only by inference.
The impact of fugitive dust from coal storage and handling was
assessed (1, p.IV-l6ff.) by comparing coal piles to piles of
crushed stone and to road and soil dust. The conclusions reached
in the study may be applied to coal conversion plants, since the
operations are very similar.
The study concluded that because coal in a storage pile would not
purposely be pulverized until just before injection into a coal
conversion plant, it is reasonable to assume that only a small
portion of the particles is capable of wind transport over long
ranges; in general, only particles of diameter less than 30
micrometers have long-range drift potential, i.e., greater than
1,000 feet from origin.
By comparison, dust normally encountered from unpaved roadways
construction sites, and agricultural soil operations contains
from 65 to 85 percent of particles that have diameters smaller
than 30 micrometers. Of these, over half are smaller than 2
micrometers. Particles in this latter category have drift
potential of hundreds of miles. Therefore it may be reasoned
that dust generated from coal storage piles and coal handling
generally has localized (1,000 feet) impact, while dust generated
from such sources as construction sites, farming activity, and
unpaved roadway travel has potential for long-range air quality
impacts.
Although considerable uncertainty is introduced when the curves
of Figure 10-1 are extrapolated to 400 Tyler mesh (37 micro-
meters) , the inference may be drawn that the content of particles
in crushed coal less than 400 mesh appears to be in the range of
0.2 to 2 percent and may average about 1 percent. From the
foregoing study, this amount of coal dust represents the maximum
amount of dust that may drift beyond property lines and be
considered in excess of environmental standards.
•798
-------
The conclusions of the assessment study do not address the
problems of fugitive dust within the plant boundaries, however,
and these, because of the potential fire and toxic hazards, must
be considered in assessing the overall problem.
The relationship between particle size and distance that a
particle will be carried may be demonstrated by use of the
following equation, assuming Stokes' law is applicable:
D = [(18 uHV)X(gLd)] °-5
D = minimum particle diameter collected
u = air viscosity
H = vertical distance (pile height)
V = air velocity
g = local acceleratio- of gravity
L = horizontal travel distance
d = (density of particle) - (density of air)
Thus, for a 30 micrometer particle to reach the ground 1,000 feet
from the peak of a coal pile 45 feet high, the crosswind velocity
can be no higher than about 2 miles per hour.
Distance from the coal pile and wind velocity have a profound
effect on the diameter of the smallest particle that will reach
the ground from the peak of a 45 foot pile:
Crosswind Velocity
Distance from peak, ft.
100
500
1,000
2 raph
95<170)»
42(325)
30(575)
5 mph
150(100)
66(200)
47(300)
10 mph
212(65)
94(170)
67(200)
20 mph
300(48)
133(115)
95(170)
•Particle diameter in micrometers. Approximate equivalent stand-
ard Tyler screen mesh shown in parentheses.
799
-------
From Figure 10-1, on the order of 0.5 to 4 percent of the coal
could reach the ground 500 feet from the pile with a 20 MPH
crosswind. In a gasification plant of 250 billion Btu/day
capacity, this potential dust accumulation could be 75 to 700
tons per day, depending on the coal source and the conversion
process.
Another investigation determined the minimum size of particles
that could safely be exposed to wind action (7), with the
following results:
Particle Diameter Tyler Screen
Micrometers Opening Wind Susceptibility
0 to 420 < 35 Highly erodible
420 to 840 35 to 20 Difficultly erodible
840 and over > 20 Usually non-erodible
Dust Suppression by Water Sprays
The usual operation at the coal unloading station in power plants
involves dumping cars by inversion into a receiving hopper and
conveying the coal to a stacker for distribution in the storage
area. All of these operations are potential producers of fugi-
tive dust.
Dust at the car dumping station is usually controlled by enclos-
ing the area and directing water sprays into the car and into the
receiving hopper to wet the coal particles at the moment of dis-
charge so that the small particles agglomerate and do not become
airborne. Surfactants may be added to the spray water to in-
crease the wetting action, most frequently at a dilution of about
one part surfactant to 1,000 parts water. Solution application
rate averages about 2 gallons per ton of coal, increasing for dry
coal and decreasing for wet coal.
800
-------
When wetting agents are used there is a carryover effect to the
storage pile that aids in reducing dust generation at the dis-
charge of the stacker to outside storage piles. Additional
sprays may be used during reclaiming operations to ensure that
all coal is wetted.
When plain water is used in the spraying system, on the order of
5 to 8 percent of moisture, or about 12 to 20 gallons of water
per ton of coal, must be applied, in contrast to the (approxi-
mately) 0.8 percent moisture addition at 2 gallons per ton when
surfactants are used.
Design of the wetting system is critical, since the coal is dis-
charged from the car by a rotary dumper in about 30 seconds. The
principal objectives are (5):
o Confinement of the dust within the dust producing area by a
curtain of moisture
o Wetting the dust by direct contact between the particles and
water
o Formation of agglomerates too heavy to remain airborne, or
too heavy to become airborne, by encouraging adhesion of
particles to each other or to larger coal particles with
water as the adhesive
Careful attention to formulation of the surfactant yields a
wetting solution whose surface tension may be about 35 to MO
percent of that of water. The effects of the surface tension
decrease are (5) :
o Dust can pentrate into the water droplets, rather than only
coating the surfaces. The entire volume of the droplet is
available for dust collection
o Fine particles, because they are more readily wetted, are
cemented to each other and are effectively encapsulated by
the surface active droplet
801
-------
o The aggregates, because of increased weight, either drop
from the air stream or cannot become airborne
o Sprays of untreated water yield large droplets with low
surface area and large, wasted volumes. Treated water,
because of the reduction in surface tension, is more readily
atomized, more droplets are produced per unit volume, total
surface area of the droplets is greatly increased and the
contact potential with dust particles is vastly improved.
The surfactant selected for use should exhibit such
characteristics as the following (5):
o Excellent surface tension reduction for optimum wetting
spreading,and penetration
o Heat stability, freeze-thaw stability
o Miscible with water in all proportions
o No flash or fire point
o Non-toxic
o Non-corrosive
o Contain biocidal additives to inhibit growths of slimes that
may clog sprays, strainers,or parts of the proportioning
system
Completely engineered systems for dust suppression by spraying Of
water solutions of surfactants are available from specialists in
this field. Observation of systems in operation when receiving
coals that are already wet or that contain normal (6 to 10
percent) moisture led to the following conclusions:
o Dust at the car dumping station is greatly reduced by the
sprays
o Dust is evolved, despite the sprays, in quantities
sufficient to constitute a nuisance to personnel in the
enclosed or semi-enclosed dumping areas and to constitute
possible potential explosion or fire hazard from dry dual-
802
-------
settling on surfaces of the dumping station
o Periodic washdowns of the dumping area are needed to control
buildup of settled dust
o Dust suppression at the storage pile stacker is effective
for most coals
o The sprayed-on material evaporates rapidly enough in live
storage to lose much of its effectiveness with the result
that wind-generated fugitive dust from the pile remains a
problem
In general, the water-plus-surfactant spraying systems have
greatly reduced the fugitive dust problem, but have not elimi-
nated it. This is especially apparent when dry coals are
received and the deficiencies of the control system are more
readily visible. Much of the difficulty seems to lie in the
necessarily short exposure tn sprays of the coal stream dumping
from the cars: while most of the surfaces of the larger coal
particles are wetted, much of the dust that becomes airborne
during dumping escapes contact with the water droplets and is not
collected. This situation might be improved by installation of
more sprays of the misting type at the dumping station. A
likelier solution may well be the use of a more dilute solution
of surfactant, for example 1:UOOO, in flooding type sprays
directed downward into the cars before they are dumped to more
nearly saturate the coal and cause dust particles to adhere to
the larger particles or to agglomerate. Excess water draining
from the cars could be recycled through strainers to the flooding
sprays.
Costs have been published (mid 197*1) for a spray system installed
in a 1,000 TPH rock crushing plant to spray 1.5 gallons of
wetting agent solution per ton of rock processed (6) . Installed
equipment cost was estimated at $61,676. Operating costs were
estimated on the basis of 1,920 hours per year plant operation
803
-------
(one shift at 8 hours per day, 5 days per week, 48 weeks per
year) with spray systems operation assumed to be required only 40
percent of the plant operating time due to the moisture content
of the material or the prevailing weather conditions. Total
operating costs on this basis were reported as $17,050 per year,
or $0.009 per ton of stone produced.
These costs will be treated later in this report for application
to coal conversion plants.
Dust Suppression by Total Wetting
As an alternate to spraying coal with surfactant solutions, con-
sideration might be given to totally wetting the coal prior to
unloading. In this case no surfactant would be used, the whole
carload of coal would be wet and maximum agglomeration would be
achieved.
Bituminous coal as received at the conversion plant will contain
on the order of 10 percent moisture, subbituminous on the order
of 15 to 25 percent and lignite on the order of 25 to 35 percent.
This moisture is the water remaining after processing in the coal
preparation plant, handling, storage,and transport to the conver-
sion plant site. The amount of water that can be held by the
coal is partly a function of the overall particle size distribu-
tion and partly a function of such particle surface characteris-
tics as porosity and the tendency to be hydrophilic or hydro-
phobic. There are no actual data on the water-holding charac-
teristics of various coals, probably because of the many vari-
ables involved. Consideration of the following general state-
ments from Davis Associates' "Coal Preparation Environmental
Engineering Manual" (4) may aid in estimating the total water
content:
804
-------
1. Natural drainage will reduce surface moisture to 5 to. 6
percent for particle sizes larger than 0.5 inch.
2. Vibrating screens reduce moisture to about 10 percent on
particles coarser than about 0.75 inch.
3. Finer fractions can be expected to contain on the order of
20 to 40 percent moisture after screening, centrifuging or
filtration.
U. The natural moisture content for coarse bituminous coal
refuse (median about 10mm) is about 11 percent and for
fine refuse (median about 0.1mm) is about 31 percent.
5. An example is given (p.330) for a typical coal cleaning
plant producing:
Coarse Coal
Intermediate coal
Fine coal
270 tons 6* moisture (approx.)
363 1056
61 30/6
697 tons Ave. 10$ moisture
Coarse refuse
Fine refuse
190 tons
113
303 tons
Ave.
11/6 moisture (approx.)
31*
18.5/6 moisture
The average coal as shipped contains 0.11 pounds of water per
pound of dry solids while the average refuse as piledsin disposal
areas naturally contains 0.23 pounds of water per pound, of dry
solids. Recognizing that the refuse particle sizes are generally
smaller than the coal sizes .and that the water holding capacity
of the refuse is therefore greater than that of the coal, the
natural (saturation) water content of the coal may be estimated
to be an average of about 0.17 pounds of water per pound of dry
solids, or an increase of about 50 percent over the shipped
805
-------
moisture content.
Applying this reasoning to three coals received at a conversion
plant producing 250 billion Btu/day of synthetic natural gas
yields the following water requirements for saturating all coal
received:
Bituminous Subbituminous Lignite
Dry coal feed 17,000 TPD 17,000 TPD 17,000 TPD
Moisture as
received - wt* 10 20 30
- Ib/lb
dry solids " 0.11 0.25 o.43
Saturation,
Ib/lb solids 0.17 0.37 o.64
Moisture gain,
Ib/lb solids 0.06 0.12 o.21
Water required- <£
gal/day 245,000 490,000 858,000
As received coal 18,900 20,400 22,100
Cars per day,
100T capacity 189 204 221
Water per car, gal. 1,300 2,400 3,900
The operating concept is: flood the car just before dumping
over a collecting basin; drain the basin, the dump hopper, the
conveyor feeders, and the conveyor tunnel to a sump; recycle the
water to the flooding nozzles. Since the nozzles are large, only
a strainer would be needed to remove large particles from the
water.
Dust Suppression by Chemical Binders
The stockpile of feed coal for conversion plants is usually
806
-------
considered to be about 1M days' feed, or about 300,000 tons.
Site specific conditions, such as the logistics of coal movement,
may require considerably more stockpiling. Reclaiming from
stockpiles is usually accomplished by conveyors in a tunnel or
tunnels beneath the stockpile, fed to a series of openings.
The volume of coal directly over the openings flows to the con-
veyors by gravity to form conical depressions with walls sloping
roughly MO to M5 degrees from the horizontal. This center, or
"live", storage volume moves through the storage area in a re-
latively short time and, if a surfactant solution is well dis-
tributed on the coal during unloading and stacking, will con-
tribute little to the fugitive dust problem because of the re-
sidual effect of the solution. Depending on the stacking method
and the number of openings under the total storage pile, the live
storage volume may vary ..rom 20 percent of the total volume (one
pile, one opening) to 55 percent of the total (multipile
openings) (M, p.280).
The remaining 45 to 80 percent of the total volume of coal stored
is moved only when the live storage volume is exhausted. This
"dead" storage volume may therefore be immobile for considerable
periods of time, any carryover effect of surfactants from the un-
loading sprays is soon lost and the pile may become a source of
fugitive dust.
Chemical binders of several types are available for application
to the outside of dead storage piles to alleviate the dusting
problem (7). These chemicals coat the topmost particles with a
thin film, promoting adherence of particles to each other, and
forming a tough, durable crust that is resistant to wind and rain
erosion but through which moisture can penetrate and run in the
void spaces between particles. While the crust is intact the
pile is protected from windage loss and erosion from rainfall and
807
-------
the dust problem is eliminated. Once dried, the film is
insoluble and tough enough to be resistant to the strains of
freezing and thawing. Examples of chemical binders are:
o Compound SP (Johnson-March Corp.) is a water base synthetic
organic polymer that will coat all types of materials and
sizes of materials and dries to an inert, non-toxic film
that does not affect coal burning qualities. Various
grades are made for various conditions. Application rate
is 1 gallon per 100 square feet of pile surface.
o Coherex (Golden Bear Oil Co.) is a mixture of semi-liquid
petroleum resins and a wetting liquid containing sequester-
ing agents. The resins form films and bind dust while the
wetting liquid carries the res-ns, wets the particle
surfaces and renders the compound miscible with water in
all proportions. The mixture is non-flammable. Usual
application rate is about 1.5 gallons of Coherex per 100
square feet of pile surface. The Coherex is diluted with 4
parts of water prior to application. Information for
Coherex (7) shows a cost of $0.225 per gallon for the
material.
o Dowell Binder (Dowell Division of the Dow Chemical Co.) is
a specially formulated synthetic liquid adhesive that is
diluted to about 1 part in 24 parts water for spraying. On
drying a crust about 1.5 to 2 inches thick is formed that
eliminates wind and rain erosion from coal in transit as
well as coal in stockpiles.
o Aerospray Binders (American Cyanaraid Co.) are polymeric
(Aerospray 70) and alkyd resin (Aerospray 52) water-disper-
sable materials, the former for short term use, the latter
for long term. Both Aerosprays form surface films when
808
-------
particles are small and dilution is small and penetrate
when particles are larger and dilution is greater. Usual
dilution for Aerospray 70 is about 1:*l with application of
about 5 to 6 gallons of solution per 100 square feet. For
Aerospray 52 the dilution is about 1:10 with application of
about 3 to 4 gallons per 100 square feet.
Dust Suppression by Physical Binders
Stockpiles may be coated with asphalt or road tar to provide an
airtight seal. The procedure is quoted (7) from the "BCI Fuel
Engineering Data Book", Section D-3, prepared by the Bituminous
Coal Institute:
"In this method the top and sides of the stockpile
are covered with an air-tight seal of asphalt or road
tar. This seal may be applied with spraying equipment
similar to that used by highway departments. The top
of a large storage pile can be capped by the use of a
hose and hand nozzle. Small piles can be capped by
hand spraying only.
A covering or cap, 1/8 in. thick and requiring
about 1 gal. of asphalt per 10 sq ft of area, has been
found satisfactory. The sides exposed to prevailing
winds should be treated somewhat more heavily for
greater assurance against access of air.
A preference has been shown for the AE-U grade of
asphalt emulsion because of its quick water-separation
and superior coating properties."
Dust Suppression by Compaction
In another quotation from the "BCI Fuel Engineering Data Book"
(7) are given recommendations for compaction sealing stockpiles
809
-------
against weather. The sealing method has the additional advantage
that wind erosion is minimized:
"Plants having reserve piles larger than 500 tons
should always build a compacted coal pile. Compaction
seals out air and minimizes spontaneous heating. it
also reduces heating-value loss to one percent or less.
Dumping coal on the pile segregates it, creates
flues and promotes heating and fires. The pile should
be built in successive layers 1 to 2 ft thick. Each
layer should be thoroughly compacted by repeatedly
running a bulldozer or weighted roller over the coal.
Dress the pile by sloping the tops of successive
(coal) layers and compacting the pile toward the sides.
The sides of the pile must be compacted and dressed.
This prevents rain, melting snow and ice from penetrat-
ing the pile. Pile slopes should not be too steep; and
angle of 30 deg to the horizontal is a good slope to
maintain. Steep side slopes can cause segregation and
erosion from wind and rain.
Crown the top of the pile slightly or allow for
slope in one direction. Saucer-shaped tops create a
problem of standing water, wetting the coal so that it
clogs chutes and conveyors.
It is recommended that, insofar as possible, only
one type of coal be placed in the stockpile. For large
compacted reserve storage piles, the coal should be a
slack or nut and slack size not exceeding 2 in. top
size. Size breakdown should be such as to fill all the
voids in the coal as it is compacted. Large size or
double-screened coal should never be compacted.
A 6-in. layer of fine coal (1/4-in. x 0) [should]
be placed over the top and slopes of the shaped stock-
pile. As an anchor, place over this a U-in. layer of
larger size material (at least 2 in. x 0). The com-
bination will materially prevent wind and rain ero-
sion."
810
-------
Dust Reduction by Agglomeration—A Concept
As an alternate to dust suppression, if dust were removed from
coal at the mine, agglomerated and then recombined with the coal
before it is loaded into cars, most of the problems caused by
evolution of dust from the car loading station, from the cars
during transportation and from the car unloading, coal stacking,
storage, reclaiming and conveying operations at the conversion
plant could be alleviated or eliminated.
Tests have shown that losses in transit can be substantial.
Youngstown Sheet and Tube Company, for example, found that about
2,700 to 3,000 pounds of coal were lost as dust from 70 ton cars
during the 200 mile transit from mine to point of use (8). This
represents a weight loss of about 2 percent from the top layer of
coal in the car. For a coal conversion plant receiving 18,000 to
20,000 TPD of coal in these cars, potential windage losses could
be on the order of 120,000 to 135,000 tons year distributed along
the right of way.
When the settling rate formula that was discussed earlier' is
applied to the railroad car problem, with H taken as 2 feet, V
taken as HO MPH and L of the car varying from 4 feet to MO feet,
the diameter of the largest particle that could be lost from the
back of the car is about 470 micrometers, or about 32 Tyler mesh,
and from the front about 148 micrometers, or about 100 Tyler
mesh.
As a first approach to the problem, then, the statement may be
made that if all particles in the loaded car were 32 mesh or
larger, windage losses during transit would be reduced or
eliminated. In handling and storage, with reference to the
previous discussion on fugitive dust loss from storage piles, in
a 20 MPH wind 32 mesh particles could be expected to reach the
811
-------
ground no further than about 45 feet from the peak of the pile.
Reduction of the dust content of the coal before it is loaded
into cars has many advantages, among which may be listed:
o Fugitive dust losses at car loading are reduced
o Dust losses during transport are reduced
o Fugitive dust problems during unloading, storage, reclaim
and handling are reduced
o Need for dust suppression systems at car loading and at
the conversion plant is reduced or eliminated
o Need for chemical binders on loaded cars in transit is
reduced or eliminated
With reference to Figure 10-1, the crushed coal at the mine may
be estimated to contain on the order of 1.4 to 6.5 percent of
particles passing 32 mesh. Separation of these particles may be
accomplished by wet or dry screening or by wet or dry elutria-
tion. Agglomeration may be by compaction or granulation, with
the object of agglomerating the fines into masses having a
nominal diameter of 32 mesh or larger or a range of about 20 to
32 mesh (833 to 470 micrometers). The estimated load on the
agglomeration step would be 14 to 65 TPH for a 1,000 TPH coal
preparation facility.
For compacting, the particles are squeezed between two revolving
cylinders, or rolls. Roll separating force, maintained hydrau-
lically or by springs, may range up to about 400 tons. Capacity
is a function of peripheral speed of the rolls, the spacing
between rolls and the roll face width. A compactor with 30-inch
812
-------
face rolls of 6-foot diameter running at 120 RPM with an 800
micrometer spacing will compact on the order of about 18 TPH of
fines, using about 30 horsepower. The compacted material is
usually broken in a saw tooth crusher and screened to remove
fines. Fines are recycled to the compactor. Assuming 15 percent
of the compactor discharge will be fines, the maximum total feed
to the rolls would be .75 TPH and a battery of 5 mills would be
needed to handle the load.
Briquettes may be produced in equipment that is similar to the
smooth roll compactor except that the rolls have depressions in
their faces into which the dust is squeezed and formed into
pillow or almond shapes. Commercial machines usually produce
briquettes of one inch diameter or larger. Because the bri-
quettes are large, capacity of the machines is greater, with
respect to roll diameter and length, than the smooth roll
compactors. For example, a briquetting machine manufactured by
K. R. Komarek Inc. has rolls 36 inches in diameter and 12
inches wide, a roll separating force of 450 tons and a capacity
of about 40 TPH, using about 75 horsepower. On this basis, two
briquetters would suffice to process the fines from production of
1,000 TPH of coal.
Depending on such coal characteristics as the size consist,
moisture content and susceptibility to plastic flow under
pressure, the requirements for addition of moisture and a binder,
and the amount of each, will differ for different coals in the
compaction and briquetting processes. Addition of moisture leads
in most cases to a requirement for a drying step which, in
addition to removing moisture, also hardens the agglomerates.
Binders may be needed, such as bentonite clay, lignosulfates that
are byproducts of wood pulp manufacture, pregelatinized starches,
causticized humic acids, fuel oil, waxes, asphalt,and pitch, and
Oay be used alone or in combinations.
813
-------
Coal fines may be granulated either on an inclined rotating disc
or in a horizontal drum. Fine, free-flowing materials are fed to
the disc or drum at a uniform rate and are sprayed with a binder
that is usually a water solution, but can be a melt of wax or
asphalt or similar material. Small pellets form as a result of
the continuous rolling, tumbling action in the granulator and
serve as nuclei for further pellet growth by layering of the
fines and binder on the nuclei.
The inclined disc has an inherent classifying action that
separates fines feed, nuclei, and product pellets into distinct
zones. Therefore, only the desired size range of pellets is
discharged and production of off-size pellets is minimized.
Pellet size is controlled by such variables as pan inclination,
pan speed and the location of the feed point and the binder
sprays. Product pellet sizes usually range from 0.125 to l.o
inch in diameter. For coal dust granulation, however, micro
pellets in rather irregular shapes less than 0.125 inch (131
micrometers or 6 Tyler mesh) can be produced with proper choice
of operating variables. Where small sizes are desired and the
tolerable size range is wide, capacity of a disc increases. For
example, it is estimated that a 20-foot diameter disc would
produce only about MO to 45 TPH of 0.5-inch pellets, requiring
125 horsepower, but would produce over 65 TPH of 6- to 20-mesh
pellets with the same drive power. Therefore, one 20-foot disc
appears to be adequate to granulate the dust from a coal
preparation plant producing 1,000 TPH.
The drum granulator builds pellets in the same manner as the
disc. Because the material flow through the nearly-horizontal
cylinder is much less controllable than the flow across a disc
the size range of the pellet discharge is much wider with the
drum. Firmness of the pellets may vary considerably, inversely
with diameter, because of exposure of the pellet nuclei to
814
-------
varying amounts of binder and fines feed. There is almost always
unpelletized dust in the drum discharge. This requires that the
product be screened with subsequent recycle of fines. If a
recycle ratio of 2:1 is assumed, the total throughput for a drum
granulator would be 195 TPH in order to produce 65 TPH of
granulated fines. To process this quantity of material, 2 drums
would be needed, each 11 feet in diameter and 30 feet long and
requiring about 75 horsepower to run at about 8 to 9 RPM. Bed
depth in each drum would be about 3 to 3»5 feet. Retention time
would be about 5 minutes.
Comparison of roll compaction, briquetting, disc granulation,and
drum granulation leads to ^neral conclusions as follows:
o Power requirements for the four types of agglomeration
machines are approximately equal at 150 horsepower for
production of 65 TPH of agglomerates
o Auxiliary equipment is required in compaction and drum
granulation for separation and recycle of unagglomerated
or created fines
o Presence of unagglomerated fines in the products of disc
pelletizing or of briquetting is probably insignificant
and separation probably would not be required
o Although meaningful capital and operating costs for the
four types of processes were not available, it appears
that the least costly, per ton of agglomerated fines
produced, would be disc pelletizing, followed in order of
increasing cost by drum granulation, compaction ,and
briquetting, with possibly little real difference between
the last two.
815
-------
o Disc pelletizing appears to be superior to other agglom-
eration methods from the standpoints of simplicity, ease
of control, suitability of product, and costs
Coal fines may be pelletized on a disc by addition of water
alone. Product size in this case is a direct function of product
moisture content. For the micropellets desired for dust
elimination, the optimum moisture content is on the order of 20
percent, with variation depending on the size consist and the
coal type. Because of the low water solubility of the
constituents of the coal, there is little residual tendency for
the individual particles to adhere to each other when the
moisture is removed by drying, with the result that the pellets
tend to be fragile and to disintegrate readily.
Use of a binder improves the strength of pellets made by water
addition (9):
o Bentonite is the most effective, but is expensive and
increases the ash content of the coal
o Liquid lignosulfates, byproducts of wood pulp manufac-
ture, are lower in cost than bentonite and do not
increase the ash content of the coal. They are water
soluble, however, and the upper layers of pellets in
outside storage or transport will deteriorate. They are
usually applied at a rate of 1.0 percent by weight of dry
solids
o Pregelatinized corn flour starch is highly effective as a
binder and, when dried, is not affected by rainfall on
the storage pile or in transit, but is more expensive
than the lignosulfates
816
-------
o A mixture of equal parts of the starch and lignosulfates
applied at a rate of 0.5 to 1.0 percent by weight of
solids has been shown to be effective
o A mixture of equal parts by weight of dry solids of
lignosulfates and bentonite, applied at a rate of 0.5 to
1.0 percent by weight of dry solids, resulted in
excellent pelletization and little increase in the coal
ash content
o A binder made by reacting humic acids with calcium
hydroxide has been shown to be as effective as bentonite
in the pelletization process and in final pellet strength
and costs much less than bentonite
o In general, increasing the proportion of binder yields
stronger, but larger, pellets
Use of water and binders is effective in pellet production but
requires a drying step to reduce the pellet moisture content to
about 2 to 3 percent and to increase pellet strength. High
drying rates have been shown (9) to decrease final pellet
strength. Although pellets of 0.5 inch diameter and larger may
be dried in conveyor type thermal dryers, supported on wire mesh
for through circulation of hot air, the micropellets that are
under consideration here would require, for example, fluid bed
drying for least attrition.
Drying may be eliminated if the coal fines contain little water
as fed to the pelletizer and if a nonaqueous binder is used.
Tars, pitches,and waxes are examples of such binders that may be
applied to the pelletizer through sprays as melts, singly or in
combination, or in mixtures with fuel oil. The product pellets
would need only cooling to become durable enough for transport,
817
-------
and would be waterproof. Elimination of the drying step should
compensate, at least in part, for the increased cost of the
binding agents.
Attention has been given so far only to the fines that are
created during the coal crushing operations. Where coal fines
are recovered as a slurry from flotation or from settling ponds
and are filtered for water separation, the moisture content of
the damp mass of solids will be on the order of 25 to 30 percent
(10). With admixture of a binder, and of dry fines to adjust the
liquid content to the proper level for the desired degree of
pelletization, these solids may be pelletized and dried to
increase the overall yield of the mining operation.
Where both dry and wet fines are to be pelletized, an economic
study would be required to determine the best process route
given the following choices:
o Two trains, one handling dry fines with a melt type
binder and no dryer, the other handling wet fines with a
solution type binder and a drying step
o Mix the wet and dry fines and feed to discs for water and
binder pelletizing followed by drying
o Partial drying of the wet fines and mixing with the dry
fines so that the total mixture has a moisture content of
about 3 percent, then pelletizing with a melt type binder
and no dryer.
Consideration may also be given to use of one or more of the
chemical binders described previously under the heading "Dust
Control*" Experimentation would be needed to optimize the
scheme, with cost as the criterion.
818
-------
Collection and Disposal of Coal Dust
Coal is reclaimed from storage by several means, all of which are
potential dust generators:
o Through openings under the piles leading to feeders that
discharge to conveyors
o By bucket wheel reclaimer, in which toothed buckets are
attached to the rim of a large wheel that rotates on a
horizontal axis. The wheel assembly is mobile so that it
may be advanced into the pile or moved parallel to it.
The buckets scoop up the coal and dump it onto a conveyor
o By trencher, similar in action to the bucket wheel except
the toothed buckets are mounted on a traveling chain
Dust is usually controlled at the reclaiming station by water
sprays that may contain a surfactant, as described in preceding
paragraphs. Coal is transported by conveyors to the conversion
process feed preparation area. Covering of the conveyors will
normally prevent fugitive dust loss and, if the coal is wetted
during reclaiming, dust generation at transfer points will be
minimized.
At the primary crushing station the coal passes across a scalping
screen, or grizzly, where the small particles of coal are
separated from the larger lumps. The primary crusher reduces the
large lumps of coal to a size consist that, depending on the
conversion process, may be fed to the reactor without further
processing or may be fed to a pulverizer. The discharge of the
crusher is combined with the finer fraction that passed through
the grizzly and the whole stream is (usually) transported to live
storage silos. In the primary crushing area the discharge of the
819
-------
feed conveyor, the grizzly, the crusher and the conveyor feed
point for the intermediate size coal product are all enclosed and
vented to a dust collection system. Pickup points for evolved
dust are strategically located in the crusher area. Air velocity
at the openings of the pickup nozzles varies from a minimum of
about 200 up to about 400 feet per minute, depending on, for
example, the coal type, range of moisture content and the
particle size distribution at the points in the system. Air
velocity in the ductwork connecting the pickup points to the
baghouse is normally 2,500 to 3,000 feet,per minute to prevent
settling of the dust particles.
A similar system is used to control dust in subsequent transport
and grinding/pulverizing operations.
Baghouse collectors are used almost universally to separate coal
dust from the transporting air, since coal dust particle sizes
range upward from about 2 micrometers and for this duty the
collection efficiency is rarely less than 99 percent. Although
the capital cost of bag collectors is high and they require the
most space for installation of any of the various types of dust
collection equipment, they require much less energy than the
other types to achieve their high-efficiency recovery and require
no water for operation. In the dust collection system the bag
collectors may be preceded by cyclone collectors which will
remove most of the dust particles of 10 or more micrometers
diameter and will thus relieve the bag collector of part of the
load.
Storage silos are either vented separately through small,
individual bag collectors or air filters or are vented together
through a common bag collector. The individual collectors
discharge their collected dust into their respective silos. The
common dust collector may discharge to the orginal silos or the
collected dust may be combined with the dust from the transport
820
-------
and milling operations. In the latter system, all collected dust
may be conveyed, usually pneumatically, to a point in the
conversion plant coal preparation system where it may be
conveniently combined with the main stream of the coal feed. As
an example, the pneumatic conveyor may discharge into the top of
a surge bin, from which the dust may be fed as needed, and in
which a fabric filter separates dust from the transporting air
before the air is discharged to atmosphere.
Alternately, the dust may be slurried with water or oil and fed
to the conversion process reactor, as in the Texaco gasification
or the SRC liquefaction processes, respectively.
In the Lurgi Dry Ash process the coal feed can contain no fines.
In this case the process feed is screened to remove fines and
fines and collected dust are fed as fuel to the
incinerator/boiler that is part of the process scheme.
Control of Other Process Dusts
Ash and slag from the coal conversion processes are quenched as
they are discharged from the reactors. Much of the free water is
removed during passage of the ash slurry through hydrocyclones
and sieve bends until the combined ash stream from conversion and
from steam generation contains on the order of about 20 to 25
percent water. In this form the solids stream is easily trans-
ported by conveyor belt or in bulk carriers and is not dusty
while the surface layer is wet. Dust control while the material
is within the process battery limits may therefore be limited to
wetting devices such as sprays or high pressure hose streams on
any piles of slag or ash awaiting transfer to permanent disposal
areas.
Limestone for flue gas desulfurization may be received as lumps
for eventual in-plant grinding or as powder for direct feeding to
821
-------
the scrubbers. Dust control methods in handling of lump
limestone from car unloading through storage, reclaim and primary
crushing may be similar to those in coal handling: wetting the
particles during unloading with water containing sufficient
surface active agent to provide a carryover effect and
supplementing the wetting at unloading with wetting at transfer
points and during reclaim. Following reclaim, control of dust
evolved during crushing, screening, fine grinding, silo storage,
reclaim,and feeding to the flue gas scrubber system can be
accomplished dry, by bag collector, or wet, by wet cyclone, or
other wet collectors. If wet collectors are used, liquor from
the flue gas scrubber may be the collecting medium, for example,
with the collected dust being discharged to the scrubber system
as part of the feed.
Where ground limestone is received in carload lots, the cars may
be unloaded pneumatically or by dumping into under-track hoppers.
With the pneumatic system, cars of special design ("Airslide,"
for example) discharge into closed systems that do not permit
dust to escape. The ground limestone is finally discharged into
closed silo storage, vented through fabric filters. Dust control
is a greater problem when cars dump into under-track hoppers and
the limestone is to be transported, stored, and fed as a powder:
since wetting is not permitted, dust must be collected by air
streams and then separated from the air in a bag collector. The
volume of air to be handled can be minimized by reducing to a
practicable minimum the open area between the car and the hopper
and sealing the hopper bottom to the powder transport mechanism
which can be a screw conveyor or a pneumatic conveyor.
As an alternate in powdered limestone handling, the cars could
dump into receiving tanks to become slurries containing on the
order of 30 to 50 percent solids. The slurrying medium could be
either makeup water for the scrubbing system or scrubbing system
. 822
-------
liquor. In either case, sprays of the liquid could be used to
control dust evolution.
Spent catalysts from conversion processes are changed
periodically because of deactivation through poisoning. Quanti-
ties are small in comparison with other solids streams, but dusts
evolved in catalyst handling are hazardous and/or toxic and
require control. Examples of the catalysts are:
o Cobalt-molybdenum, catalyst for naphtha hydrogenation
o Methanation catalyst for SNG production
o Shift conversion catalyst for hydrogen production
o Claus catalyst fc, sulfur production
o Copper-based catalyst for flue gas denitrification
o Desulfurization and denitrogenation catalysts
Catalysts are unloaded from their containing vessels in the
conversion process sequence by dumping or by hand raking and are
received in covered containers. Dust control is accomplished by
small, individual mobile units consisting of fans, bag filters
(or other fabric medium),and appropriate flexible ductwork.
Collected dust is placed in the covered catalyst containers. The
containers are trucked either to the solid waste disposal area,
to be mixed with the ash/slag stream at a convenient point, or to
the pug mixer in which quenched ash/slag, ash,and evaporator
bottoms are combined for transport to disposal. In either case,
the catalysts are dispersed in the damp solid wastes and dusting
beyond the dispersal point is not a problem.
823
-------
COSTS OF DUST CONTROL
Dust Suppression by Water Sprays
Comparison of water spraying systems at the coal handling
facilities that were visited during this project, and con-
versations with suppliers of spraying systems, lead to the
conclusions that there is little uniformity in design and
installation of the the spraying systems, that despite the
"custom design" of any one installation there is little or no
attempt at scoping or defining the individual dust problems by
either the plant operator or the spraying systems designer and
that, finally, dust suppression sprays are installed by rule of
thumb with sufficient spare capacity so that if suppression is
not achieved at the design sprayin& rate, then the rate may be
doubled or trebled to achieve suppression. Capital and operating
costs for spraying systems are difficult to evaluate because of
these many variables and empirical factors in system design and
operation.
A study leading to the determination of spraying systems costs
was carried out in 1974 by the Bureau of Mines for three sizes of
hypothetical hard rock crushing plants (6). While the conditions
in rock crushing operations may not exactly parallel those in
coal handling in conversion plants, there is sufficient similar-
ity to warrent cautious comparison. For example, the assumption
in the Bureau of Mines study that 1.5 gallons of surfactant solu-
tion, with a surfactant to water ratio of 1:1,000, would be re-
quired per ton of rock processed is in the same range of solution
usage (1.5 to 2.0 gallons per ton) that is stated as "normal" for
coal handling.
In the Bureau of Mines study the rock crushing plant operated
1,920 hours per year atl,000 tons per hour. The surfactant
824
-------
solution was sprayed at 1.5 gallons per ton. The spraying system
therefore had an operating output ofl,500 gallons per hour of
solution. In contrast, a coal conversion plant producing 250
billion Btu per day of gaseous or liquid fuels will require on
the order of 20,000 tons per day of coal. If it is assumed that
the coal feed will be unloaded during 2 shifts per day for 6 days
per week during 48 weeks per year, then the coal unloading rate
becomes 1,435 tons per hour. If it is further assumed that the
coal will require 2 gallons of surfactant solution per ton of
coal, then the operating output of the spraying system will be
2,870gallons per hour, or 1.9 times the capacity of the system in
the Bureau of Mines study.
The installed cost of the Bureau of Mines spraying system,
including winterization, is shown as $61,676. When the 0.6
exponent is used to scale up the capital cost for the increase in
the capacity and 17 percent is added for escalation of costs from
mid 1974 to the end of 1977, the capital cost for the coal
spraying system becomes $106,100. In both cases it is assumed
that these are turnkey installations and that electric and water
utilities are in place to supply the systems.
Operating costs are calculated on the basis of operating 2 shifts
per day, 6 days per week, 48 weeks per year, or 4,608 hours per
year. Because the system is controlled automatically in
conjunction with operation of the coal reception station,
operating labor is minimal and is estimaced at 8 hours per week
or 384 hours per year. Maintenance labor in the Bureau of Mines
study was estimated at 8 hours per week for 40 hours per week
system running time. For the coal system, with 96 hours per week
operating time, the maintenace labor is estimated to be 16 hours
per week or 768 hours per year. Operating costs for the coal
spraying system calculated from these bases are shown in TABLE
10-4.
825
-------
TABLE 10-4. CAPITAL AND OPERATING COSTS
FOR WET DUST SUPPRESSION
Basis: Coal rate 1,435 TPH
Spray rate 2 Gal/T
Operation: 2 Shifts/Day, 6 Days/Wk, 48 Wks/Yr
Depreciation: 20 Yrs., Straight Line
Fixed Capital $106,100
Working Capital (60 Days Operating Cost) 19 t4op
Total Investment $125,500
Operating Costs per Year
Power 10 hp x 4608 hrs. x 1.8
-------
Dust Control by Chemical Binders
Application of chemical binders to the outside surfaces of dead
storage piles has been shown to reduce the problem of fugitive
dust (7). The chemicals in water solution are sprayed on the
piles and form stable, inert, non-toxic and non-flammable films.
Spraying is accomplished with conventional equipment, such as
that used for spraying orchards, or with custom built spray
trucks with large capacity tanks, high pressure pumps and special
nozzles.
A stockpile containing about 15 days supply of coal for a 250
billion Btu per day conversion plant, or 300,000 tons, may
consist of U piles, each U5 feet high and about 1600 feet long.
Total pile area to be coated in this configuration is about
825,000 square feet.
Compound SP, made by Johnson-March Corporation for coating piles,
is a synthetic organic polymer in a water base. It is used as
received, without dilution. Usual application rate is one gallon
per 100 square feet of surface area. Effective protection is
provided for about one year. Various grades of the material have
been developed to cope with such variables as differences in
types of materials to be coated, particle size, surface
wettability,and climatic conditions. Cost of Compound SP is
about $2.50 per gallon, and therefore the cost of the 8,250
gallons of the compound to coat the total pile surface of the
stockpile is about $20,600.
Coherex, manufactured by Golden Bear Oil Company, is an emulsion
of petroleum resins, wetting agents,and sequestering agents in a
water base. The usual dilution for use is one part Coherex to U
parts water. The rate of application varies with the character-
istics of the material being sprayed, but an average rate of
827
-------
2 gallons of diluted Coherex per square yard, or 22 gallons per
100 square feet, may be considered. On sandy materials the
penetration is about 2 inches at this rate of application. Coat
of the Coherex concentrate is about $0.26 per gallon. About 4.5
gallons of concentrate are required to treat 100 square feet;
therefore, the cost of 37,125 gallons of Coherex for one applica-
tion to the total surface of the stockpile is about $9,700. It is
probable that coating would be needed twice a year, bringing the
annual cost of material to about $19,400.
A special tank truck carrying high pressure pumps and spray
nozzles may cost about $30,000, fully rigged. Amortized over 6
years yields an annual cost of $5,000.
Spray rate for the undiluted Compound 3P is estimated at 10 gal-
lons per minute, at high pressure to atomize the spray, requiring
about 14 hours to coat the pile. Spray rate for diluted Coherex
is estimated at 40 gallons per minute at a pressure of 40 to 60
psig, requiring about 75 to 80 hours to coat the pile. If it is
assumed that spraying will be done during 6 of the 8 hours of a
shift, then 2.5 days once a year are needed for Compound SP and
12.5 to 13.5 days twice a year are needed for Coherex. Operating
costs for the two coating systems may be compared:
Compound SP Coherex
Amortization of spray truck $ 5,000 $ 5,000
Labor: 3 operators at $10/hr 2.5 days/yr 600
26 days/yr 6,2,40
Maintenance labor 1,500 1,500
Fuel and other supplies 400 2,000
Coating compound 20,600 19,400
Taxes and insurance 1,000 1,500
Overhead 35% of labor 735 2,709
Indirects 40? of maintenance and supplies 9.000 9.160
TOTAL ANNUAL OPERATING COST $38,835 $47,509
OPERATING COST PER SQUARE FOOT COATED $0.047 $0.058
828
-------
References
1. Office of Coal Utilization, "Coal Conversion Program.
Energy Supply and Environmental Coordination Act (as
Amended). Section 2. Volume 1." FES-77-3. May 1977. 847*
2. Institute of Gas Technology, "Preparation of a Coal Conver-
sion Systems Technical Data Book." Quarterly Report, May 1-
July 31, 1976. FE-2286-4. 587»
Detman, R., "Factored Estimates for Western Coal Commercial
Concepts." FE-2240-5. October 1976. 294»
4. Davis, J. J. Associates, "Coal Preparation Environmental
Engineering Manual." EPA 600/2-76-138. May 1976. 300»
5. Guimond, J. A., "Dust Suppression from Mine Face to Car
Loading." Proceedings, Rocky Mountain Coal Mining
Institute, Estes Park, Colo., July 1969.
6. Evans, R. J., "Methods and Costs of Dust Control in Stone
Crushing Operations." Bureau of Mines Information Circular
1C 8669. 1975.
7. Matthews, C. W. , "Chemical Binders: One Solution to Dust
Suppression." Part 6 of a series "Stockpiling of
Materials," Rock Products, Jan. 1966.
8. "Crusting Agent Minimizes Loss of Coal in Transit." Reprint
from Railway Age, Sept. 1974, for Dowell Division of Dow
Chemical Co., Tulsa, Okla.
•Pullman Kellogg Reference File number
829
-------
9. Luckie, P. T.,and Spicer, T. S., "The Application of the
Palletizing Process to the U. S. Coal Industry."
Proceedings, Ninth Biennial Briquetting Conference, The
Institute for Briquetting and Agglomeration, 1965.
10. Montgomery, C. T., and Beafore, F. J., "Enhancement of Coal
Preparation Plant Production through Use of Polymers,"
Proceedings, Third Symposium on Coal Preparation, National
Coal Association and Bituminous Coal Research Inc.,
Washington, D. C., 1977.
SOLID WASTE DISPOSAL AND MANAGEMENT
Problem Definition
Coal conversion processes generate solid wastes with widely
divergent characteristics:
o Ash from gasifier reactors that operate with a maximum
temperature below the ash fusion point is composed of
loose agglomerates of small particles that break readily
into discrete particles. Average bulk density of the dry
solids is MO to 45 pounds per cubic foot. Dry particles
easily become airborne. Gasifier ash is quenched with
water in all conversion processes. The resulting slurry,
if it is allowed to settle undisturbed in ponds, will
eventually compact to about 25 pounds of dry solids per
cubic foot (l,p.IV-l8) with an equivalent water content
of about 65 percent. When subjected to classification
thickening and filtration, however, the water content can
be reduced to 15 to 20 percent
830
-------
Ash from gasifier reactors that operate with a maximum
temperature above the ash fusion temperature is
discharged from the reactor as a melt. During subsequent
water quenching the melt shatters into discrete particles
resembling sand. Average bulk density of the dry
quenched solids is about 100 pounds per cubic foot. Dry
particles become airborne only with difficulty. Solids
in the slurry from the quenching step may be separated
from the liquid by classification, thickening,and
filtration to yield a final stream containing only about
10 to 15 percent water
Ash discharged from utility boilers or incinerator/
boilers may be a combination of bottom ash and flyash or
may be bottom ash alone, depending on the boiler con-
figuration and method f operation. In these cases the
ash characteristics are similar to those from low
temperature gasifiers. With the increasing attention
being given to fluidized bed combustion, in which
limestone is mixed with the coal fuel to act as an
absorbent for sulfur dioxide, the ash may be a mixture of
"normal" ash, calcined limestone, calcium sulfite and
calcium sulfate. Quantity will depend on the utility
demand of the conversion process plant and the proportion
of the total heat requirement that is furnished by
incineration of organic residues from the conversion
process. Physical characteristics will, in general, be
similar to "normal" boiler ash. The total ash discharge
may be sluiced to the ash handling system or it may be
quenched only enough to reduce the temperature to the
range than can be tolerated by conventional transport
equipment
831
-------
Flue gas desulfurization (FGD) sludge varies greatly in
composition depending on the coal composition, the degree
of fly ash removal prior to scrubbing, scrubber operating
characteristics, and the type and amount of additives.
Sludge from a lime/limestone wet scrubber is mainly
calcium sulfite, calcium sulfate, calcium carbonate and
ash. The sludge may be dewatered to an average solids
content of about 50 percent and an average bulk density
of around 70 pounds per cubic foot. In settling ponds
there is less dewatering and the fluid settled sludge may
contain considerably less than 50 percent solids. Both
dewatered and settled sludge have low bearing and
compressive strength and are thixotropic
Organic sludges from biological oxidation processes are
susceptible to bacterial degradation when they are
exposed to air, or possibly to anaerobic fermentation if
they are covered. Although the quantity of organic
sludge that is generated in coal conversion plants is
small in comparison to the quantity of ash and FGD
sludge, odors originating in the organic sludge could
become a nuisance and, possibly, a legal problem. In the
treatment schemes that have been examined in this study,
the preferred method of disposal of organic sludges is
incineration. This has been shown in Section 8 in the
integrated schemes for Lurgi gasification and SRC
liquefaction. Bi-Gas and similar high temperature
gasification processes do not require biological
oxidation in the water treatment sequences and therefore
do not produce organic sludges.
Inorganic salts accumulate in boiler and cooling tower
blowdowns and originate in water treatment processes. As
proposed in the integrated schemes for water treatment in
832
-------
Section 8 of this study, the inorganic salts may be
concentrated by evaporation. The constituents of the
concentrate will vary widely, depending on raw water
analysis and methods of treatment of the raw and recycl-
ing water. The concentration of the salts in the evapor-
ator bottoms will vary depending on the limitation of
concentration due to scaling of heat exchange surfaces
and the type of evaporator. For example, the salts may
be evaporated to near dryness in an oil suspension, then
may be separated from the oil by centrifuging. The quan-
tity of inorganic salts, on a dry basis, is small in com-
parison to the quantity of ash. The possibility exists
that the inorganic salts may react with components of the
ash or FGD sludge to yield soluble compounds. In any
case, the evaporator bottoms will probably contain both
soluble and insoluble compounds.
o Spent catalysts are discharged periodically in quantities
that are small in comparison to the ash stream. They con-
tain metals or metallic compounds that may be harmful to
the environment, such as cobalt, molybdenum, copper, and
zinc and may during their active life pick up other
potentially harmful metallic compounds. The spent
catalysts are discharged dry from their containing
vessels in the conversion process sequence.
The Pullman Kellogg study of control of gaseous emissions
(Section 9 of this report) did not consider flue gas desulfuriza-
tion by lime/limestone scrubbing, preferring instead to take
advantage of the rich hydrogen sulfide stream that is fed to
Glaus sulfur recovery to supply the reducing gas to the citrate
process sulfur recovery unit on the incinerater/boiler. By so
combining the two sulfur recovery systems the total amount of
sulfur for sale is increased, the capital cost for emission
control is reduced and overall operating costs for control of
sulfur compound emission is reduced.
833
-------
If, however, FGD by lime/limestone scrubbing were the only means
of controlling sulfur compound emissions from coal conversion
plants, production of FGD sludge may be estimated from the
following information, which is based on power plant operation
(1 , p.IV-32ff.):
o 95 percent of the sulfur in coal fed to boilers becomes
sulfur dioxide. The rest of the sulfur remains with the
ash
o Density of dewatered sludge is 57 to 85 pounds per cubic
foot and averages about 71 pounds per cubic foot
Since about 93 to 98 percent of the sulfur compounds in coal fed
to conversion processes are converted to hydrogen sulfide which,
if not recovered, must be oxidized to sulfur dioxide and removed
by FGD scrubbing, for the purposes of determining the amount of
FGD sludge and evaluating its effect on the overall solid waste
management scheme the conversion processes can be considered to
be roughly equivalent to boiler operation. Accordingly, the
variation in ash and FGD sludge production with changes in coal
composition may be evaluated as in TABLE 10-5.
If sulfur is recovered via the Glaus process from streams rich in
hydrogen sulfide, all other sulfur-bearing streams, including
Glaus tail gases, are oxidized to sulfur dioxide, and FGD scrubs
the gases, the dry solids in FGD sludge become:
FGD Solids Sludge/Ash Ratio
Gasification, high sulfur coal 1,119 TPD 0.60
Gasification, low sulfur coal 244 0.16
Liquefaction 1,288 0.62
Maximum sulfur recovery in the conversion processes, including
834
-------
TABLE 10-5. ASH AND FGD SLUDGE PRODUCTION WITHOUT
SULFUR RECOVERY
Basis: Production of 250 billion Btu/day of gaseous or liquid
fuels
Feed sulfur to gasification FGD scrubbing:
High sulfur coal 98.6*
Low sulfur coal 95.0*
Feed sulfur to liquefaction FGD scrubbing:
High sulfur coal 93.4$
Gasification
High Sulfur Low Sulfur
Coal Coal Liquefaction
Coal feed: Ash, % 11.30 7.72 11.80
Sulfur, % 4.42 0.66 3.70
Btu/lb 13,190 11,290 12,125
TPD 14,778 17,265 17,680
Ash production, TPD 1,851 1,553 2,070
Sulfur to FGD, TPD 644 114 611
FGD Sludge, TPD
dry solids* 4,025 713 3,819
Ratio, Sludge/Ash 2.17 0.46 1.84
» Typical composition: CaSO^.O.SHgO 58*
CaC03 33*
Sulfur Content = 16*
835
-------
sulfur recovery from stack gas scrubbing, will yield no FGD
sludge.
These comparisons illustrate the extreme variations that may be
encountered in coal conversion plants, depending on the economics
of sulfur recovery vs. sulfur disposal.
With reference to Figure 8-54, the water balance for Lurgi
gasification of low sulfur coal, the solids-bearing streams are
shown to be:
Total Solids Water % Solids
Evaporator bottoms, TPD 153.6 18.0 135.6 11.6
Boiler ash, TPD 264.0 264.0 0.0 100.0
Superheater ash, TPD 18.0 "lo.O 0.0 100.0
Dewatered gasification
ash, TPD 1,495.3 1,271.0 224.3 85.0
Total to disposal, TPD* 1,930.9 1,571.0 359.9 81.4
•Biological oxidation sludge not included
If there is no sulfur recovery, the FGD sludge at 50 percent
solids added to the above total yields a total mixture of 3,414.6
TPD, containing 2,284 TPD of solids, or about 68 percent.
With high sulfur coal in Lurgi gasification, assuming no change
in evaporator bottoms, boiler ash and superheater ash, the total
solids to disposal becomes 2,351.4 TPD containing 1,869.0 TPD of
solids, or 80.2 percent. Addition of FGD sludge yields a total
mixture of 10,401.4 TPD containing 5,912 TPD of solids or 56.8
percent.
With high sulfur coal feed to liquefaction, maximum sulfur
recovery and no FGD, reference to Figure 8-58 shows that the
836
-------
total to disposal is 2831.7 TPD, containing 2,123.8 TPD solids, or
75 percent. If maximum FGD sludge is added, the total stream to
disposal is 10,469.7 TPD, containing 5,942.8 TPD of solids, or
56.8 percent.
As previously noted, the dry bulk density of the ash discharged
from gasifiers operating below the fusion temperature of the ash
is about 40 to 45 pounds per cubic foot. The density may be in-
creased to about 60 to 65 pounds per cubic foot by compacting the
disposal pile. In contrast, the ash from gasifiers operating
above the fusion temperature of the ash, and the ash from boilers
and incinerators, will have a dry uncompacted bulk density of
about 100 pounds per cubic foot, and a compacted bulk density of
about 110 pounds per cubic foot. A bulk density of 130 pounds
per cubic foot may be assumed for the solids in the evaporator
bottoms. The total volume of uncorapacted solids discharged from
coal conversion processes without FGD may then be estimated as
shown in TABLE 10-6. Uncompacted densities were used so that
structural considerations of the disposal areas could be
simplified.
From TABLE 10-6, the volume of the solid wastes may range from
31,400 to 75,600 cubic feet per day, depending on the coal type
and the conversion process operating characteristics.
Considering operation at 330 days per year yields an estimated
solids volume of 10.4 to 25 million cubic feet per year.
Solids Transport
The solids mixture as it is produced in the conversion plant,
without FGD sludge, is a damp mass that may be transported by
conveyor belt or by truck to the disposal area. Choice of
transport method would be based on economics, considering the
distance between plant and disposal area, the terrain and
837
-------
TABLE 10-6. VOLUME OF SOLIDS FROM COAL CONVERSION
WITHOUT FGD
Ash from gasifier:
TPD
Density, PCF*
Volume, MCFD*
Ash from boilers:
TPD
Density, PCF
Volume, MCFD
Evaporator bottoms:
TPD
Density, PCF
Volume, MCFD
Total dry solids,
TPD
Low Temp.
1,271-1,569
45
56.5-69.7
282
100
5.6
18
130
0.3
High Temp. Liquefaction
1,218-1,452 590
100 100
24.4-29.0 H,8
335-399 1,480
100 100
6.7-8.0 29.6
18 54
130 130
0.3 0.8
1,571-1,869 1,571-1,869 2,124
Total solids volume,
MCFD
Moisture in total
solids, %
Bulk density of
wet solids, PCF
62.4-75.6
19-20
62
31.4-37.3
30
143
42.2
25
134
•PCF = pounds per cubic foot of compacted solids
MCFD = thousands of cubic feet per day
838
-------
climatological conditions. Trucking stabilized FGD sludge at 60
percent solids is evaluated in a TVA study (2). In the study
capital costs were developed for mid-1979 for trucks and earth-
moving equipment to haul about 2,400 TPD of sludge 1 mile to dis-
posal for the IU Conversion Systems (IUCS) Poz-0-Tec process:
Hauling Distance, mi Capital Cost for Transport
1 $581,000
3 617,000
5 641,000
10 700,000
Because of the tendency of wet, stabilized FGD sludge to dewater
during truck transport and cause spills on the roads to the dis-
posal site, hauling distance must be limited to 2 to 3 miles,
according to an evaluation x-y Columbus and Southern Ohio Electric
(3). In C&SOE's operating system, freshly-mixed stabilized FGD
sludge is carried by conveyor belt to a radial stacker which
forms a surge, or curing, pile from which earthmoving equipment
pushes the partly cured Poz-0-Tec stabilized material to final
disposal, where the sludge is laid down in layers about 2 feet
thick and then compacted to a dry solids density of about 65
pounds per cubic foot. Projected final height of the pile is 100
feet.
These findings on transport of stabilized FGD sludge may be ap-
plied to transport of ash and the other solids from the conver-
sion plant to the disposal site. The problems in ash trans-
portation appear to be less severe than with stabilized FGD
sludge:
o Solids content of the mixed solids, as shown in TABLE 10-6,
varies from 70 to 81 percent
839
-------
o The ash mixture is not thixotropic, as untreated, freshly-
mixed or partly cured FGD sludge is, and therefore leakage
from trucks during road hauling should be greatly reduced.
Special construction, such as gasketed tail gates on dump
bodies, could eliminate leakage.
o Transport by conveyor belt and stacker has been demonstrat-
ed and is in operation for freshly-mixed stabilized FGD
sludge. The transport problem is greatly reduced for the
ash mixture because of its higher solids content and lack
of thixotropic tendencies. Further, transport by conveyor
of wet ores of various particle sizes is common commercial
practice.
Ash and mixed solids may be transported by aerial tramway (5).
Tramcar capacities vary from 10 to 90 cubic yards and are able to
travel at rates up to 1750 feet per minute and seldom underl.000
feet per minute (4). Since spans of 1500 feet or more between
towers are being operated, the system may be considered as a via-
ble means of transport and practical from the standpoint of being
able to span the disposal area at any desired height. Towers 350
feet tall are in operation. Considering the solids from low
temperature gasification, as shown in TABLE 10-6, loaded into 90
cubic yard tramcars, each car would carry 75 tons of wet solids
or 60 tons of dry solids and 31 carloads per day would satisfy
the plant. If the number of cars is conservatively estimated at
35, if operating time for the tramway is 7 hours per day and if
the traverse rate is 1,000 feet per minute, then the disposal area
could be located over 2.25 miles from the conversion plant.
Variations on this scheme would require study to determine the
most economical combination of the following:
o Operate 24 hours per day with smaller tramcars to reduce
cable loading and capital cost. Also reduces stockpiling
of mixed solids awaiting transport to a minimum surge pile
840
-------
o Possible advantages of increasing tramcar speed, or the
number of cars in transit in order to increase the prac-
tical transport distance
o Need for transfer points, and the required equipment, for
long transport distance
o Hazards and nuisances involved in overhead transport of wet
solids
Economic evaluation of belt conveyors as a means of transport
must be carefully done, considering the following:
o Cost per ton mile tends to remain constant, no matter how
far the conveyor is extended
o Distribution at the discharge end of the belt may be by
truck, scraper or stacker units. Bins must be provided if
trucks or scrapers are used, but are not needed with
stacker units
o Belt conveyors may be combined with tramways for negotia-
tion of difficult terrain
o In the belt conveyor layout provision should be incorporat-
ed for reverse return where the belt, after passing the
discharge point, is twisted 180 degrees so that the wet
side of the belt does not come in contact with the return
idler rollers. Wear on the return idlers is reduced,
buildup of sticky material on the return idlers, with
consequent problems with maintaining belt alineraent, is
avoided and droppings of waste on the beltway are reduced
Wheeled vehicles may be used to transport and disperse waste.
Rear dump trucks, side dump trucks/and scraper loaders have been
841
-------
used for coal refuse disposal (M) and would be suitable for con-
version plant waste transport and disposal. These vehicles can
spread the waste in a thin layer in the disposal area and can
compact it by driving the vehicles over the area while the loads
are being discharged. Vehicles have great flexibility in opera-
tion, the means to adapt the discharge pattern to changing con-
tours in the disposal area and to develop area stability, togeth-
er with their intrinsic flexibility in capacity that is charac-
teristic of multiple vehicular transport. In addition, the cost
per ton mile for wheeled vehicles tends to decrease as the haul-
age distance increases. Economic evaluation of vehicular trans-
portation and distribution of conversion plant wastes must
consider:
o Special vehicle construction that may be required by the
characteristics of the material being handled
o Special construction that may be required so the vehicles
can traverse the disposal area
o Environmental regulations that may limit the effectiveness
of vehicles, such as restrictions on roads travelled, re-
strictions on spillage or droppings on roads, and restric-
tions on creation of hazards or nuisances from dust and
noise
Management of the Solid Waste Disposal Area
In the discussion that follows the wastes to be considered are
ash/slag and evaporator bottoms. As previously pointed out, the
quantities of such miscellaneous solids as spent catalysts are
small, or they are discharged only intermittently, and therefore
would have little effect on the total waste quantity or on the
physical or chemical characteristics of the waste.
842
-------
Flue gas desulfurization (FGD) sludge is considered only inci-
dentally in this discussion because the decision to include an
FGD system in a coal conversion plant is site-specific and econom-
ically oriented. Information and data on means of disposal of
FGD sludge have been gathered, however, and will be applied to
solid waste management wherever such inclusion may be helpful in
defining, or advancing solutions for, waste disposal problems.
Elimination of organic wastes from the solid waste stream avoids
nuisance problems of odors caused by decomposition. It is
assumed in this study that the organic wastes will be incinerated
and that any residue will be inorganic and combined with the ash
stream.
The following is quoted fr^m reference 1, p. V-6:
"The effects of sludge and ash production will include in-
creased demand for discharge sites, disruption of natural
wildlife and vegetation and possible contamination of local
water resources. When solid wastes are to be transported for
disposal to an offsite landfill by trucks, secondary trans-
portation impacts may occur, including increased use of fuel
and equipment, increased air pollutant emissions and increased
traffic, dust, and noise in the area traversed. In developed
areas, land will be converted from use in agriculture, housing,
and industrial development to uses serving coal production and
utilization. Land use will be permanently altered or dis-
rupted when permanent facilities for coal transport and use
are developed. Other areas used in mining and solid waste
disposal may be reclaimed, eventually, but the time required
and the degree of success of reclamation efforts are not known
at this time. It is unlikely that natural topographical
features and an exact replica of ecological systems will
result from any reclamation program."
843
-------
The quotation serves as a guideline for the discussion of manage-
ment of solid waste disposal areas, together with the applicable
environmental standards, both present and proposed.
A basic premise of the present study is that the control tech-
nology applied to coal conversion operations should be so chosen
that gaseous emissions may be reduced to a level at or below the
most stringent environmental standards that are now in force and
that liquid effluents from the conversion processes should be
treated for recycling within the process and should not be
released to receiving waters. This premise has been presented,
and the means to achieve it have been described in detail, in
Section 8 of this report for liquids and Section 9 for gases.
Increasing the Solids Content of the S'lids Waste Stream—
In the schemes proposed for treatment of liquids in coal conver-
sion, commercial control technology was applied in various
combinations to permit recycling the water. A residual amount of
water remained, bearing inorganic salts, after evaporation to a
concentration limited by the operating characteristics of the
evaporator. As discussed previously, this evaporator bottoms
stream was combined with quenched ash/slag from the gasifier
reactor and the dry ash from the incinerator/boiler and the
process steam superheater to yield a mixture contaioning 70 to 80
percent solids.
Although the damp ash mixture can be handled by conventional
means, the solids content may be increased by use of an evaporat-
ing system marketed by Dehydro-Tech Corp., East Hanover, N. J.t
in which the water and solids are suspended in a circulating bath
of oil (5). In this process, oil of the correct volatility,
viscosity and surface tension is added to the aqueous waste and
the mixture is fed to a multiple-effect falling-film evaporator.
The mixture passes through the evaporator, leaving the solids
844
-------
suspended in the dry oil as a fluid slurry. The oil is then
centrifuged off to be used again, and the dry solids are left.
Water is recovered and recycled to the coal process, the oil is
recycled, and the water-soluble solids emerge in a dry powder
form.
The oil maintains fluidity throughout the system, improves heat
transfer in the evaporator, and prevents scaling and fouling as
the salts become completely dry. This patented technique is
efficient for the removal of water without problems of
thickening, scaling, and fouling.
It is reported that oil-fluidization plants have turned out to
cost approximately one-third less than equivalent spray-drying or
incineration plants, due mainly to economics of multiple-effect
heat utilization. Evaporation is energy-intensive but the use of
a strong brine as feed, along with multiple-effect heat cost
savings, bring it within the limits of reasonable economy, given
that clean-up of the aqueous waste is required.
Makeup oil for the process may be available as one of the
products or byproducts of the conversion processes. Choice is
governed by such oil properties as gravity, viscosity, and vapor
pressure (boiling range). There is no theoretical reason against
using the phenols fraction as the evaporator oil (5). This could
be a small diversion from saleable products to be used as make-up
oil, or it could be once-through on the way to incineration.
The oil, of whatever type, withheld from sales would be only
make-up. Most of the oil would be recycled into the evaporation
cycle, and the oil purge, if viscous impurities collect, would be
recycled into the coal conversion process. The vapor pressure of
the oil is such that about one pound of oil is vaporized with
845
-------
every pound of water*. This can be diminished or augmented as
desired by selecting oil fractions of appropriate boiling range.
The cooled oil can then solvent-extract organics out of the
water, which is then treated further or recycled to the coal
process as needed.
For evaporation equipment, cost has been explored (1976) as
affected by feed concentration and feed rate. The economics of
scaleup in feed rate are apparent from the figures summarized in
TABLE 10-7. It is also apparent that pronounced economies result
from designing for the higher feed concentrations. These effects
should be analyzed in an economic study to demonstrate the inter-
action of oil-fluidized evaporation with the overall water treat-
ment system and the subsequent solids handling operations.
The advantage of oil-fluidized evaporation is that, with such an
operation to remove water, the solids content of the mixture from
low temperature gasification may be increased to about 87.5
percent, that from high temperature gasification to about 84.3
percent and that from liquefaction to about 89 percent. This
reduction in water content further reduces any solids handling
problems.
The decision to install oil-fluidized evaporation instead of
conventional evaporation, or for there to be any evaporation step
at all, depends on the following:
o Whether or not FGD sludge will be part of the solid waste
stream
o Economics of transporting very wet (50 to 60 percent
solids), damp (20 to 30 percent solids) or semi-dry (10
to 15 percent solids) wastes to disposal
846
-------
TABLE 10-7.
CAPITAL COSTS OF OIL-FLUIDIZED
EVAPORATION (5)
Feed Concentration,
Wt Salts
Feed Rate,
Pounds/Hour
Number of Effects
100 PSIG Steam,
Pounds/Hour
Cooling Water, GPM
Labor, 24 Hr Day,
Men
Demonstration Unit
20
1.32
Commercial Unit
20
Capital Cost
$374,500
$2,386,500
$864,000
1.32
8,000
3
3,200
285
127,868
6
31,500
5,600
40,000
4
12,000
1,422
631,000
6
157,500
28,000
1.5
$6,270,000
847
-------
o Means, and effectiveness of the means, of avoiding damage
to the environment and the ecosphere as affected by the
physical and chemical characteristics of the total waste
stream
References
1. Office of Coal Utilization, "Coal Conversion Program. Energy
Supply and Environmental Coordination Act (as Amended).
Section 2. Volume 1." FES-77-3. May 1977. 847*
2. Barrier, J. W., Faucett, H. L. , and Henson, L. J., "Economics of
FGD Waste Disposal." EPA Flue Gas Desulfurization Symposium,
Hollywood, Fla. Nov. 1977. 751*
3. Boston, D. L., and Martin, J. E., "Full-Scale FGD Waste Disposal
at the Columbus and Southern Ohio Electric1 s Conesville
Station." 1977. 750*
4. Davis, J. J. Associates, "Coal Preparation Environmental
Engineering Manual." EPA-600/2-76-138. May 1976. 300»
5. Brown, J. A., "The Carver-Greenfield Process." Technical
Memorandum of Dehydro-Tech Corporation, 1976.
•Pullman Kellogg Reference File number
848
-------
The Solids Disposal Problem--
Production from coal of liquid or gaseous fuels containing 250
billion Btu per day will produce, as previously noted, on the
order of 10.4 million to 25 million cubic feet per year of mixed
solids consisting of ash with small amounts of inorganic salts,
sludges, and catalysts. The volume may be reduced to some degree
by compaction as the mass accumulates, but bulk densities cannot
be expected above about 110 pounds per cubic foot for slag and
about 60 pounds per cubic foot for non-slag ash.
Since the compacted slag and ash densities are in the general
range of packed earth, there will be little difference in overall
appearance between mounds made by depositing ash/slag directly on
level ground and covering it with a layer of earth and mounds
made by digging pits or trenches in level ground, filling them
and then piling earth on top. For example, if the ash or slag
for one year's production is deposited in a square area, 1000
feet on a side, and covered with 2 to 4 feet of earth, the mounds
will range in height from 20 to 26 feet above grade. In 20 years
the total area covered in a square configuration would be about
4500 feet on a side, amounting to about 465 acres. Obviously,
piling higher would reduce the acreage, but against this must be
considered the effect on the overall environment and the resident
escophere. Final decisions on height and square versus
rectangular or other plot areas can best be made on a
site-specific basis that includes an environmental and an
economic study.
Not the least of the problems involved in solid waste disposal is
preventing leachates from contaminating the soil areas
surrounding the disposal pile and the surface and ground waters.
Interposing an impervious membrane between the wastes and the
environment offers an attractive means of avoiding environmental
damage. Such membranes will be discussed later.
849
-------
Disposal of solid wastes in abandoned areas of strip mines
appears to be an attractive alternative to piling the wastes at
grade, particularly when it is considered that the volume of ash
may range from 5 to 15 percent of the volume of coal fed to the
conversion plant. While this method of disposal appears to be
convenient, as when the mine is adjacent to the conversion plant,
there are other considerations, such as:
o Means of impounding the solid wastes to prevent water
erosion
o Means of access to the disposal area
o Means of preventing intrusion of ground water
o Means of preventing leachates from entering the
environment
Construction of dams of earth or overburden is practical for
impounding the solid wastes, provided that the other criteria can
be met. Because the abandoned mine is a stripped-out area, the
contours and the mine surfaces may be difficult to seal so that
no intrusion of water or leakage of leachate will occur.
Problems that maybe encountered are discussed in connection with
FGD sludge disposal for Columbus and Southern Ohio Electric (1)
and include the necessity for extensive regrading to obtain a
sealed subsurface, the probability that leaching could not be
prevented because of the presence of random piles of soil and
rock overburden and the high cost of transportation of the
stabilized sludge to the area. These same criteria must be
applied to a feasibility study for solid waste disposal in strip
mines.
850
-------
Control of Leaching by Membranes—
Leachate from the mixed solids in the disposal site may be a
portion of the free liquid phase of the solids as transported to
the disposal site together with seepages of ground water through
the site and drainage from precipitation. If the solids are in
direct contact with the soil the possibility of percolation of
the leachate into the soil under the site and movement both
horizontally and vertically could eventually lead the leachate
into subsurface waters. If the soil at the disposal site is
relatively impervious, or the quantity of leachate is large,
either continuously or intermittently, runoff may lead to
contamination of surface waters.
It appears to be possible *o isolate the solids from the soil by
interposing an impervious membrane, or liner. With attention to
grading, leachate can be conducted to a gathering sump from which
it may be pumped back to the conversion plant for recycling as
ash quench water or to treatment and preparation for use as
process water. These conclusions are deduced from the results of
investigations into use of liners in sanitary landfills, and the
differences between sanitary landfill operation and conversion
plant solid waste landfill operation:
o As previously noted, solid wastes contain no organic
constituents. Sanitary wastes contain, among others,
hydrocarbons, fats, animal and vegetable oils and, as
bacterial action progresses, degradation products from
these. The organic compounds have been shown in
experiments to affect some types of disposal site liners
(2)(4), resulting in failure, loss of strength or
increases in permeability.
o Sanitary leachates have a pH around 5 (2)(4). Liner ma-
terials in both sanitary and inorganic waste service have
851
-------
been shown experimentally to be resistant to acidic waters
at this pH (2)(3)CO. Trials of liners with nitric acid
solutions showed (3) that cracking, hardening and surface
blistering can be expected at low (1.5) pH , effects that
may be partly caused by the oxidizing action of the nitric
acid. A bentonite-sand admixture failed in the pH5 acid
exposure while a soil-cement admixture apparently was
resistant (3) .
o Conversion plant ash usually has a pH around 8, varying
with the coal composition. In short term tests, liner
materials exposed to spent caustic at a pH of 11.3 showed
no visible change, the bentonite-sand admixture failed and
the soil-cement admixture appeared to be resistant to
attack (3)•
o Lined pits may be established in clay beds or by use of
remolded clay (6). Consideration of the sorption and/or
ion exchange characteristics of clays and the permeability
of the prepared natural clay beds is essential. Permeabi-
lity may be very low, on the order of 0.1 x 10~^ to 1 x
10*7 centimeters per second for distilled water, for clay
landfill liners (6).
Asphalt Liners—In sandy soils a firm base for installation of
the liner must be prepared by stabilizing the sand to a depth of
4 to 6 inches with asphalt emulsion. Specifications and
instructions for application are contained in "Asphalt for Waste
Water Retention in Fine-Sand Areas," published by The Asphalt
Institute, and are included in Appendix D of Reference 5.
Preparation of the lagoon area is of utmost importance, in that
all debris, vegetation, and organic materials must be removed,
areas to be paved must be graded and free of excess material and
weak areas must be repaired. The sand layer and the asphalt are
852
-------
mixed by a travel plant, rotary or mechanical mixing, or by motor
graders, then the mixture is spread evenly and compacted by
conventional road machinery. Normal application of asphalt is
0.5 to 0.7 gallons per square yard per inch of compacted depth.
Where the soil is stable when dry, the asphalt stabilization may
not be needed. The decisions are site-specific and are made by
soil engineers. The precautions for site preparation are the
same, in either case.
When the water depth (hydraulic pressure) is 8 feet or less an
asphalt membrane seal is applied to the base over a tack coat.
There are three allowable alternatives:
1. Hot-sprayed asphalt placed in 3 applications to a total
of at least 1.5 gallons per square yard
2. Asphalt emulsion and 0.375 inch aggregate placed in 2
applications to a total of at least 0.5 to 0.7 gallons of
asphalt per square yard plus 24 to 36 pounds of aggregate
per square yard.
3. Hydraulic asphalt concrete placed in one course to a
compacted thickness of 2 inches. The hot-mixed concrete,
containing 6.5 to 9.5 percent by weight of asphalt with
the balance 0.375 inch aggregate, is spread and compacted
with conventional road building equipment.
When the water depth or hydraulic pressure is above 8 feet but
not above the maximum allowable depth of 12 feet, hydraulic
asphalt concrete is applied over a tack coat in two courses with
overlapping joints to a total compacted depth of 3 inches.
853
-------
The seals and linings must be extended beyond the crests of
slopes as a means of anchoring the lining in the embankment
roadway or top in order to prevent erosion damange.
It should be noted that use of hydraulic asphalt concrete as a
liner allows operation of mechanical equipment in the area for
spreading or compaction of the deposited solids when the lagoon
impounds landfill solids.
Asphalt Sealed Fabric Liners—An example (5, Appendix E) of this
type of membrane is nonwoven polypropylene fabric, fused one
side, coated to a final thickness of 100 mils (0.1 inch) with two
coats, totalling l.M gallons per square yard, of a mixture of:
Anionic asphalt emulsion SS-lh 100 gal
Asbestos fiber 7M-02 60 Ib
Water 44 gal
Wetting agent (Phillips or equiv.) 2 Ib
Joints in the fabric must be machine sewn. Edges are anchored in
a perimeter trench and backfilled after sealing.
These specifications were formulated by the Soil Conservation
Service (Engineering Standard 521-E-l), who also specified, among
other characteristics, that the completed liner should have a
Mullen hydraulic burst strength of 200 pounds per square inch and
should exhibit no water loss when a hydraulic head of 35 feet of
100° F water is applied for 7 days.
Catalytically Blown Asphalt—This material is used as a seal for
asphalt canal liners and to seal off layers of expansive soils
under pavement (2). Prepared by air blowing hot asphalt in the
presence of a catalyst (ferric chloride or phosphorus pentoxide)
it is applied at 200 to 220° C in two applications totalling
-854
-------
about 1 to 1.5 gallons per square yard and forms a film about 0.2
to 0.3 inches thick. The cooled membrane is flexible, tough,
impervious to water and remains flexible at low temperatures.
The membrane is usually covered with a layer of soil to protect
it from traffic and from damage by light.
Flexible Membranes—Extreme care is required in site preparation
so that the foundation area is smooth and free of projections
that might damage the lining: stumps and roots must be removed
and rocks and hard clods must be removed, rolled into a sub-base
or covered with a fine soil cushion (Soil Conservation Service
Standard S-521-A-1) . Lining is spread smoothly and field
spliced. The joints are required to develop a minimum of 60 to
80 percent of the film shear strength, depending on the film. A
cover at least 6 inches thick of earth or earth and gravel (9
inches where the liner is cposed to livestock), with the bottom
3 inches no coarser than silty sand, is applied to protect the
membrane.
Butyl rubber, chlorinated polyethylene, chlorosulfonated poly-
ethylene, ethylene propylene rubber, polyethylene,and polyvinyl
chloride are examples of materials that have been used to con-
tain fluids, as in ponds. With the plastic films, a minimum
thickness of 8 mils is required for application over sands and 12
mils for application over gravels. When rubber sheeting is rein-
forced with nylon the respective minimum thickness are 20 mils
and 30 mils. Unreinforced rubber requires a minimum thickness
of 30 mils for both soil types.
Soil-Cement Liners—Although mixtures of soil and cement have not
been used as landfill liners, they have been in use for several
years as paving materials and have shown excellent resistance to
moisture penetration. Suggested specifications for the soil-
cement base course have been assembled by the Portland Cement
855
-------
Association (5, Appendix I). The soil must pass a 1-inch sieve
and not more than 20 percent may be retained on a No. 4 sieve.
The mixture may be prepared in place or in a central plant.
For in-place mixing, the cement, at 3 to 20 percent by weight of
the soil, is spread uniformly on top of the soil and is mixed to
a depth of about 6 inches with a disc harrow. Water is applied
and mixed in and finally the whole mass is compacted and graded.
Moisture retention for curing is accomplished by application of
"bituminous material" applied at a rate of about 0.7 gallons per
square yard. Curing is completed in about 7 days. According to
the Portland Cement Association, the soil-cement course will
withstand light local traffic immediately and all traffic in 7
days, "provided the soil-cement has hardened sufficiently to
prevent marring or distortion of the s-rface."
Soil-Clay Membranes—The sodium variety of bentonite clay swells
to up to 10 times its volume in water and, when mixed with a
permeable soil, will reduce the permeability in direct proportion
to the amount and type of bentonite added. The bentonite is
distributed on the prepared soil base at 18 to 36 pounds per
square yard and then mixed with the soil to a depth of 4 to 6
inches with a disc harrow. The soil mixture, after compaction,
is ready for use.
Because bentonite, like other clays, exhibits ion exchange
properties, the composition of the leachate in contact with it
may affect its sealing properties. For example, calcium
bentonite does not swell as does sodium bentonite. Sodium
bentonite will readily exchange its ions for the calcium ions in
solution and become less effective as a sealant (7).
A proprietary mixture of sodium bentonite and polymers that swell
in water is more resistant to the effects of ion exchange than
856
-------
bentonite alone (8). The material is applied in the same manner
as bentonite alone, but normally at a rate of 9 to 18 pounds per
square yard. Applications noted by the manufacturer include
sanitary landfills, liners under stockpiles and mine refuse piles
and sewage and process water lagoons.
Evaluation of Liners—
In an ongoing EPA study (4) a variety of liner materials were
exposed to leachate from simulated sanitary landfills. The study
was undertaken with these objectives:
o To determine the effects of exposure to leachate from
compacted municipal refuse on the physical properties of
lining materials (excluding soils and clays) that are
believed to be potentially useful for the lining of
sanitary landfills
o To estimate the effective life of liner materials when
exposed to prolonged contact with leachate under
conditions comparable to those encountered in a sanitary
landfill
o To determine the effects of exposure for 12 to 24 months
to sanitary landfill leachate on the physical properties
of the 12 liner materials mounted in the bases of the
simulated sanitary landfills and on the 42 smaller
specimens buried in the sand placed above the mounted
liners
o To analyze the costs of these materials for lining
sanitary landfills. This analysis will include liner
costs, installation costs, and the benefits from longer
durability
857
-------
The lining materials tested were:
Hydraulic asphalt concrete (3 percent voids)
Paving asphalt concrete (6 percent voids)
Soil asphalt
Soil cement
Blown asphalt (canal lining asphalt)
Emulsified asphalt on fabric
Butyl rubber
Chlorinated polyethylene (CPE)
Chlorosulfonated polyethylene (Hypalon)
Ethylene propylene rubber (EPDM)
Polyethylene (PE)
Polyvinyl chloride (PVC)
After a year of exposure to the leachate from the simulated
landfill there were these observations:
o The admix liners containing asphalt, although losing
drastically in their compressive strength, maintain their
impermeability to leachate. The asphalt itself became
softer, indicating possible adsorption of organic
components from the leachate
o During the year's monitoring of the cells, in only three
of the cells did the leachate enter the base below the
liners. Two of these liners, soil asphalt and paving
asphalt concrete, leaked. The leakage in the third was
caused by a failure of the epoxy sealing compound around
the periphery of the specimen
o The soil cement lost some of its compressive strength'
however, it hardened considerably during the exposure
period and cored like a Portland cement concrete. Its
permeability decreased somewhat
•858
-------
3 Inhomogeneities in the admix materials, which probably
caused the leakage in the paving asphalt and soil asphalt
liners, indicate the need for considerably thicker
materials in practice (2 to 4 inch thick liners were
selected for this experiment to give an accelerated test
and were designed with an appropriately sized aggregate)
o The asphaltic membranes withstood the leachate for 1 year,
although they did swell slightly. There was no indication
of disintegration or dissolving of the asphalt
o All of the polymeric liner materials withstood a 1-year
exposure to the leachate, although several, e.g.
chlorinated polyethylene and Hypalon, swelled appreciably.
Swollen liners softened but did not lose tensile, tear, or
puncture resistan '. Preliminary tests of the exposed
liners indicated some increase in permeability, probably
because of swelling.
o Variation occurred among polymeric membrane liners based
upon a given polymer, which may reflect variations in
polymer source, compound composition, and possibly methods
of manufacture.
o The seams of the polyvinyl chloride, Hypalon and
chlorinated polyethylene liners deteriorated in strength.
The polyethylene retained its strength best.
o The quality of the leachate in all 24 of the cells was
similar, indicating that the initial composition of the
refuse was controlled and that the comparison among the
liner materials would be valid.
859
-------
The average pH of the leachate was 5.1. The COD averaged 4'6
grams per liter and the volatile acids (acetic, propionic,
butyric, isobutyric) averaged 22 grams per liter.
The overall conclusion may be drawn that the deterioration in
physical properties that was noted in the asphaltic liners and in
some of the polymers may have been due, at least in part, to the
presence of organic materials in the leachate. It would be
logical to suppose that these same liners, if exposed to the
conditions in the solid waste disposal area of a coal conversion
plant, where there are no organic materials, would exhibit less
of a deterioration of physical properties. As a consequence,
selection of lining materials for solid waste areas may be made
primarily on the basis of such parameters as initial cost, cost
of installation, ease of installation (seaming methods, special
construction at joints), resistance to puncture or tearing,and
resistance to oxidation and sunlight. In all cases the base, or
subgrade, must be carefully prepared.
Choosing a Liner—
Following are some of 'the criteria that must be considered in
choosing liner materials:
o Clay soils, including bentonites, can lose their
impermeability when impounding strong acids, strong bases,
or brines. Testing and careful evaluation may be required
for selection of a clay that will remain impermeable
o Exposure of liner to sunlight. Some polymers, such as
polyvinyl chloride, may become brittle and crack, either
by loss of plasticizer by evaporation or by degradation by
ultraviolet light. Special plasticizers and/or inhibitors
may be required in polymer formulations. Butyl rubber is
susceptible to cracking from ozone attack.
860
-------
Exposure of liner to weather.. Soil liners may be damaged
by drying or by freeze-thaw of exposed areas.
Variability in the strength and chemical resistance of
field splices of the polymeric materials. Polyethylene
may be heat sealed or (usually) solvent sealed, as are
Hypalon, chlorinated polyethylene and polyvinyl chloride,
Te trahydrof uran showed promise in the splicing of
polyvinyl chloride( 4 ) . There are variations in the
cements or solvent systems for the various polymers. In
general, splices with solvent washes are difficult to make
in the field
Liner Protection —
None of the proposed liner materials should be used directly as a
pavement. While some of the materials can easily support
rubber-tired construction equipment, no manufacturer recommends
allowing distribution vehicles to use the liner as a pavement
because of the high wheel loadings. Equipment with crawler
treads should not be allowed to operate directly on the liner.
Manufacturers recommend protecting the liner with an earth cover
one to two feet thick. This material should not contain jagged
rocks or other sharp objects that could damage the liner.
Similarly, the first lift of solid waste placed in the fill site
should be placed carefully and spread so that the depth of the
cover protecting the liner is effectively increased.
Control of Leaching by Chemical Stabilization —
Processes for chemical stabilization of waste materials have been
receiving increasing attention due to the requirement for substi-
tution of coal for natural gas and oil in power generation and
the requirement for low sulfur dioxide emissions to the atmo-
sphere. The public utilities appear to adhere to the principle
that the principal effort should be generation of power and that
861
-------
there should be minimum effort required in handling emissions.
The result of this attitude seems to be lack of interest in
recovering sulfur from stack gases for sale and considerable
interest in scrubbing the sulfur dioxide from the stack gases and
disposing of the resulting sludge.
The volume of the flue gas desulfurization (FGD) sludge is
expected to be large and the physical and chemical
characteristics are poor (9). Processes have been developed for
chemical stabilization of the sludge so that the soluble matter
is immobilized, the permeability is reduced to a level comparable
to that of the more impervious soils and the physical bearing
strength is increased from an extremely low level to one
resembling well-compacted earth. Although disposal of FGD sludge
is not a real part of the present study, the results of the work
done by various organizations in stabilizing the sludge may be
applied toward solution of the general problem of solid waste
disposal. Further, as pointed out earlier, the decision to
recover sulfur or dispose of it will be site specific and will be
a management choice based on the economics of the situation.
Consequently, the solid waste from coal conversion plants may
include varying amounts of FGD sludge, ranging from zero upwards.
Since these sludges in general are thixotropic and rarely contain
more than 50 percent solids, mixing them with the other solid
wastes may, depending on the proportion of sludge, reduce the
bearing strength of the total mixture to such a low level that
the solids waste area cannot be reclaimed and may be a hazard.
Chemical stabilization has been applied to FGD sludge to improve
its physical characteristics. The final result of the stabili-
zation process may resemble a cement-like monolith of very high
density. Permeability, of the solid mass is usually less than
10~6 cm/sec. Chemical stabilization appears to offer excellent
possibilities for economical disposal of ash/slag alone
862
-------
ash/slag plus FGD sludge or FGD sludge alone. Where both
ash/slag and FGD sludge must be disposed of, an economic study
may demonstrate advantages in two separate handling and disposal
systems, one to dispose of the ash without stabilization and one
to stabilize and dispose of the FGD sludge.
There are three chemical stabilization systems in commercial use:
the Dravo Corporation Synearth process, the IU Conversion Systems
(IUCS) Poz-0-Tec process and the Cherafix (division of the
Carbonundum Company) treatment process. In the Dravo and IUCS
processes there are cementitious reactions involving calcium,
aluminum, and silicon compounds in the wastes and in the
additives. In the Chemfix process, sodium silicate and Portland
cement are added to the wastes and react with each other and with
the wastes.
The Dravo Process—In the sequence of operations, the sludge is
first dewatered to about 35 percent solids. "Calcilox" (a
furnace slag) is added at about 7 percent of the dry sludge
solids together with lime at about 2 percent of the dry sludge
solids. Actual amounts may vary, depending on sludge charac-
teristics. The slurry is pumped to a disposal pond where the
solids settle, reaction ensues and the solids harden, in about 30
to 45 days, to a mass having an unconfined compression strength
of around 4,000 pounds per square foot in 30 days to 6.000 or more
pounds per square foot in 45 days (10). According to Dravo
reports (11), the sludge mixture will dewater to about 45 to 50
percent solids.
The Dravo process is also being used to stabilize fine coal
refuse (12). The final stabilized mixture at 42 percent solids
showed compression strengths of about 1,200 to 1,700 pounds per
square foot after 30 days, considerably less than the stabilized
FGD sludge. In appearance, the stabilized material resembled a
863
-------
fine silt. The undisturbed material had a permeability ranging
•7
from 2 to 9 x 10 centimeters per second, depending on the
proportions of Calcilox and lime mixed into the refuse.
The concept of operation of the process for solids disposal
involves preparation of a pumpable slurry of the solid waste and
the reactants and pumping to a dammed or diked area for settling,
solidification and partial dewatering. The supernatant liquid is
returned to the FGD scrubber or to the coal cleaning plant for
reuse. When the disposal area is full of solids it may either be
pumped dry and covered with earth for area reclamation or left as
a lake. Because of the high pH of the stabilizing process it
would appear that the advisability and the acceptability of the
latter alternative is highly questionable.
The IUC5 Process—Stabilization of fly ash, FGD sludge and mine
refuse, alone and in combinations, have been successfully
demonstrated on a large scale (13). Laboratory and pilot plant
work has been correlated with the full scale results. Controlled
mixing of the wastes and such additives as lime and, for FGD
sludge, flyash is prescribed from laboratory investigations on
<
the specific waste material.
With FGD sludge, lime, and flyash are added and mixed in a pugmill
at about 60 percent solids (1). The damp mass is discharged to a
surge pile and allowed to set for 3 to 6 days and is then trans-
ported, at a consistency similar to clay, to the landfill area,
where it is distributed and compacted. The mixture develops a
compressive strength that can range from around 10,000 pounds per
square foot (1) upward to 20,000 pounds per square foot and
higher, depending on the end use of the disposal area (13).
Permeability is on the order of 1 x 10~" centimeters per second
or less (13). The stabilized material has been used to line a
pond by applying successive layers each about 6 inches thick, and
864
-------
has shown a permeability in this service of about 1x10
. i L y xn uiij.s »ervj.ue ui ciuuuu i x lu
centimeters per second
Coal waste has been successfully stabilized by the IOCS system
micrometers. In the several mixes evaluated in the laboratory,
solids content of the stabilized material ranged from 77 to 81
percent, compression strength after 28 days ranged from 3,000 to
15,000 pounds per square foot and permeability ranged from 1 to 3
x 10~ centimeters per second. The ranges in properties resulted
from varying the proportions of additives to the mix.
Chemfix—For FGD sludge stabilization the dewatered sludge is
mixed with sodium silicate and Portland cement. The silicate
forms insoluble compounds with polyvalent metal ions, then reacts
with the cement, while the cement hydrates and also reacts with
sludge components (15). Proportions of the additive are varied,
depending on the final characteristics desired in the processed
material. For FGD sludge the sodium silicate usage is about U
percent and the cement usage is about 7 percent of the weight of
dry sludge solids (16).
No data are available from Chemfix on test results on stabilized
FGD sludge or on mine refuse. Compressive strengths of various
stabilized wastes from other sources ranged from 1,000 to 10,000
pounds per square foot. No permeability data were available.
In an economic study of FGD solids disposal, conducted by TVA
(16), the Chemfix process was compared to the IUCS and Dravo
processes. The system used in the study mixed the sodium sili-
cate and the cement with FGD slurry that had been dewatered to 60
percent solids. The mixture is then trucked to the disposal site
and compacted.
865
-------
References
1. Boston, D. L. , and Martin, J. E., "Full-Scale FGD Waste Disposal
at the Columbia and Southern Ohio Electric's Conesville
Station." Presented at EPA Symposium on Flue Gas Desulifuri-
zation, Florida, November 1977. 746*
2. Haxo, H. E., Jr., "Assessing Synthetic and Admixed Materials
for Lining Landfills." Presented at Rutgers University
Research Symposium, March 1975. EPA 600/9-76-004, March
1976. 712*
3. Haxo, H. E., Jr., "Evaluation of Selected Liners When Exposed
to Hazardous Wastes." Presented at University of Arizona
Research Symposium, February 1976. EPA-600/9-76-015, July
1976. 708*
4. Haxo, H. E., Jr., and White, R. M., "Evaluation of Liner Materi-
als Exposed to Leachate" Second Interim Report. EPA-600/2-76-
255, September 1976. 494*
5. Geswein, A. J., "Liners for Land Disposal Sites--An Assess-
ment." EPA/530/SW-137, March 1975. 519*
6. Sanks, R. L. LaPlante, J. M. , and Gloyna, E. F., "A Survey of
Suitability of Clay Beds for Storage of Industrial Solid
Wastes." Center for Research in Water Resources, University
of Texas, Austin, Texas, June 1975. 711*
7. Hughes, J. , "A Method for the Evaluation of Bentonites as
Soil Sealants for the Control of Highly Contaminated Indus-
trial Wastes." Presented at the annual Purdue University
Industrial Pollution Conference, May 1977.
866
-------
8. Dowell Division of the Dow Chemical Company, Tulsa, Oklahoma.
Technical Bulletin, "Dowell Soil Sealant Service," June
1976.
9. Office of Coal Utilization, "Coal Conversion Program. Energy
Supply and Environmental Coordination Act (as Amended).
Section 2, Volume 1." FES-77-3, May 1977.
10. Lobdell, L. W. , and Rothfuss, E. H., Jr. "Eighteen Months of
Operation (of the) Waste Disposal System (at the) Bruce
Mansfield Power Plant, Pennsylvania Power Company." Presented
at Flue Gas Desulfurization Symposium, Florida, November
1977.
11. Freas, R. C., "The Stabilization and Disposal of Scrubber
Sludges - The Dravo Pr. ess." Presented at American Petro-
»
leum Institute Committee on Refinery Environmental Control,
Utah, September 1975.
12. Snyder, G. A., Zuhl, F. A. , and Burch, E. F., "Solidification of
Fine Coal Refuse." Presented at AMC Coal Convention,
Pennsylvania, May 1977.
13. Minnick, L. J., "Stabilization of Waste Material Including
Pulverized Coal Flyash." Presented at Second National
Conference on Complete WateReuse., Chicago, May 1975.
1H. Taub, S. I., "The Poz-0-Tec Process for Coal Waste Stabiliza-
tion." Presented at AIChE annual meeting, New York, November
1977.
15. Conner, J. R., "Ultimate Disposal of Liquid Wastes by Chemi-
cal Fixation." Presented at 29th Annual Purdue Industrail
Waste Conference. Reprint by Chemfix, Pittsburgh, Pa.
867
-------
16. Barrier, J. J., Faucett, H. L. , and Henson, L. J. , "Economics of
FGD Waste Disposal." Presented at Flue Gas Desulfurization
Symposium, Florida, November 1977. 751*
COST OF SOLIDS DISPOSAL
Preparation of a Lined Disposal Site
The annual volume of solids produced in coal conversion processes
was estimated previously to be 10.M to 25 million cubic feet. To
contain this volume of solids, and to provide means of collection
of any runoff waters, the mass should be confined. One method of
preparing a relatively flat area for a disposal site is shown in
Figure 10-2.
In Figure 10-2 the configuration of the disposal area was devel-
oped from the following:
1. The earth removed below grade shall be used to construct
the walls
2. The wall slope shall be 3 horizontal to 1 vertical
3. The top of the walls shall be a roadway 20 feet wide
4. The disposal area shall be square in order to provide min-
imum perimeter with maximum interior surface area
5. The total volume shall be 110 percent of the solid wastes
volume
868
-------
GRADE
00
-------
The volume below grade = W2D - U(3WD2 -
where W = Width at grade = Length at grade
D = Depth below grade
The total impounded volume is:
W2H + 6 WH2 + 12 H3 + W2D - 12 WD2 + 36 D3
where H = Height above grade
The volume of the wall above grade =
4(3 WH2 + 20 WH + 18 H3 + 180 H2 + 400 H)
where H = Height above grade
From these relationships the volume of earth to be moved and the
area of liner required may be calculated for various lengths
(widths). Results of the calculations are shown in TABLE 10-8.
In TABLE 10-9 are shown installed costs of various lining
materials for disposal areas. The two opinions of Haxo and Clark
on the installed costs have been updated to the 1977 basis for
comparison. There is considerable difference between the two,
with Haxo generally being higher. Private communications with
several manufacturers and installers of the pond liners indicate
that the Clark film prices are probably representative, while for
the asphalts, bentonite,and the soil-cement liners the Haxo
prices are probably representative.
There are several possibilities for collecting the drainage and
runoff from solid waste impoundment areas. One method would
involve placing a layer of washed sand on the liner and topping
the sand with gravel. Drainage pipes through the walls of the
impoundment area would be manifolded or would discharge into a
peripheral trench which would conduct the drainage to a central
sump. Waters collected in the sump would be pumped back to the
870
-------
TABLE 10-8. ESTIMATED DISPOSAL SITE CONSTRUCTION COSTS
A. Site
L
Length
1250 ft.
1000
750
B. Site
00 L
~-> Length
r-1
1000 ft
750
600
for Disposal of
V
Width
1250 ft.
1000
750
for Disposal of
W
Width
1000 ft
750
600
25 Million Cubic Feet of Solids
H
Height
Above Grade
14 ft
19
26
D
Depth
Below Grade
3 ft
7
29
Earth
to be
Moved
168,600 CY(1)
237,900
356,40
Earth
Moving
Cost (2)
$202,300
285,500
427,700
Internal
Surface
Area (3)
199,300 SY (1)
139,800
94,200
Plan
Area ,
Acres
43
30
21
10.4 Million Cubic Feet of Solids
H
Height
Above Grade
9 ft
14
18
D
Depth
Below Grade
2 ft
5
14
Earth
to be
Moved
33,700 CY
62,100
108,600
Earth
Moving
Cost
* 40,000
74,500
130,300
Internal
Surface
Area (3)
124,300 SY
78,400
57,100
Plan
Area ,
Acres
27
17
9
(1) CY = Cubic yards, SY = Square yards
(2) Coat s $1.05 per cubic yard for earth moving plus $0.15 per cubic yard for compaction and final grading (mid
1978)
(3) Area to be covered or treated for reducing or eliminating permeability
-------
TABLE 10-9. INSTALLED COSTS OF DISPOSAL AREA LINERS
Installed Cost per square yard(l)
Haxo (2)
Polyethylene, 10-30 mil
Polyvinyl chloride, 10 mil
20 mil
30 mil
10-30 mil
Chlorinated
polyethylene, 20 mil
30 mil
20-30 mil
Hypalon, 20 mil
30 mil
20-45 mil
Butyl rubber, 31.3 mil
46.9 mil
62.5 mil
31.3-62.5 mil
Ethylene-propylene
diene monomer, 31.3 mil
46.9 mil
62.5 mil
31.3-62.5 mil
Paving asphalt + sealer,'2 in,
4 in,
Hot Sprayed asphalt,
1 gal/sq. yd.
Asphalt membrane, 100 mil
Soil-bentonite, 9 Ib/sq. yd.
18 Ib/sq. yd.
Soil-cement + sealer, 6 in.
$1.27 - 2.04
1.66 - 3.06
3.44 - 4.59
4.08 - 4.33
4.60 - 5.66
3.44 . 4.84
1.70 - 2.41
3.33 - 4.60
1.74 - 2.45
1.78 - 2.65
1.02
1.66
1.77
Clark (3)
$1.07
1.63
1.95
2.89
3.78
2.89
3.78
3.33
4.00
4.60
3.22
3.89
4.55
2.22
1.55
1.55
(1) Includes material and labor. Does not include cost of
subgrade preparation or earth cover
(2) From Reference 5, p.17, updated from 1973 to 1977 by Chemical
Engineering Cost Index
(3) From Reference 5, p.18, updated from 1974 to 1977 by Chemical
Engineering Cost Index
872
-------
conversion plant for reuse as ash quench water or, following
treatment, as process water. Another method of collecting
drainage would be to grade the bottom of the impoundment to a
central sump from which a pipe or channel would carry the
drainage to a pumping station. Of the two methods, the former
would probably be preferred because all piping would be
accessible for inspection and maintenance. Costs for the
drainage system, not including sump, pumps or pipelines to the
conversion plant, are estimated to be on the order of $0.75 to
$0.90 per square yard of impoundment surface area.
Comparison of the estimated capital costs of impoundment areas of
various dimensions may be made with the following criteria:
o Land cost at$3,500per acre
o Earth moving cost as in TABLE 10-8
o Drainage system at a median cost of $0.83 per square yard
o Liner costs from TABLE 10-9
These costs are summarized in TABLE 10-10 for the two quantities
of solids that are under consideration. It is seen that in both
Case A and Case B the advantage lies with the smaller, deeper
impoundment areas. The advantage is particularly apparent when
the total disposal area for the life of the conversion plant,
which may be 20 to 30 years, is considered.
Disposal Sites with Soil and Clay Lining
It is generally stated by soil engineers that soils with a perme-
ability rate of 10~" centimeters per second or less may be consid-
ered as being impermeable since, at this rate, migration of
liquid through such a soil would be on the order of 12.5 inches
per year. The most stringent of environmental regulations
873
-------
TABLE 10-10. ESTIMATED CAPITAL COSTS OF LINED DISPOSAL AREAS
oo
A. Disposal of 25 Million Cubic
Feet of Solids
Impoundment Dimensions,
Land Cost
Earth moving cost
Drainage system cost
Liner cost: high (1)
medium (2)
low (3)
Finish cover installation (1)
Total, high cost liner (1)
medium cost liner (2)
low cost liner (3)
B. Disposal of 10. 1 Million Cubic
Land cost
Earth moving cost
Drainage system cost
Liner cost: high (1)
medium (2)
low (3)
Final cover installation (1)
Total, high cost liner (1)
medium cost liner (2)
low cost liner (3)
1250x1250
$150,500
202,300
111,100
797,200
566,000
352,800
138,900
$1,133,000
$1 ,201 ,800
$ 988,600
Feet of Solids
1000x1000
$ 91,500
10,100
92,200
197,200
353,000
220,000
88,900
$813,200
$669,000
$536,000
1000x1000
$105,000
285,500
92,200
559,200
397,000
217,500
88,900
$1,130,800
$ 968,600
$ 819,100
Impoundment Dimension,
750x705
$ 59,500
71,500
51,900
313,600
222,700
138,800
50,000
$519,500
$158,600
$371,700
Feet
750x750
$ 73,500
117,700
51,900
376,800
267,500
166,700
50,000
$979,900
$870,600
$769,800
Feet
600x600
$ 31,500
130,300
23,100
228,100
162,200
101,100
38,000
$151,300
$385,100
$321,000
(1) Paving asphalt + sealer, 1 inch thickness, $3.33 to $1.60 per square yard, median for
calculation = $1.00
(2) Average of costs in TABLE 10-9 at $2.81 per square yard
(3) Soil cement * sealer, 6 inches thick, $1.77 per square yard
(1) Earth cover, 2 feet thick, installed and compacted at $0.80 per square yard
-------
(Dominion of Canada) requires this same maximum permeability
rate. It is possible to achieve this permeability by compaction
of suitable soils, admixture of native clay and remolding of
native clay. Aside from the difficulty of creating a homogeneous
barrier layer, clays have been shown to be susceptible to ion
exchange, with consequent changes in their physical characteris-
tics, in contact with leachate from sanitary landfills. Much
work has been done in evaluating barriers for handling FGD
sludges. Ponds have been in commercial operation for many years
for storing flyash slurry from power plants and fine coal and
refuse from mining operations. None of the storage or impound-
ment conditions, however, duplicate those to be found in impound-
ment of gasifier ash. Judgement as to the suitability of native
soils and clays, whether virgin, remolded or compacted must,
therefore, be by inference.
i
Cost of preparing a clay lined site may be estimated from the
costs of preparing the base for membrane liner and then adding
the following as appropriate:
o Clay for the lining assumed delivered to the site at $1.00
per cubic yard
o Spreading and compacting assumed at $1.20 per cubic yard
o Liner to be 3 feet thick
o If native soil is suitable, assume $0.60 per cubic yard for
spreading, compacting,and final grading.
Estimated costs are shown in TABLE 10-11.
From the foregoing discussion it is obvious that despite the
apparent economics to be realized through use of available clays
875
-------
TABLE 10-11. ESTIMATED CAPITAL COSTS OF SOIL AND CLAY
LINED DISPOSAL AREAS
A. Disposal of 25 Million Cubic Feet of Solids
Impoundment Dimensions, Feet
1250x1250 1000x1000 750x750
Cost of Land, earthmoving,
drainage system, and
final cover $551,300 $463,700 $444,800
Liner cost: clay 442,400 314,200 215,400
native soil 120,700 85,700 58.700
Total: clay liner $993,^00 $777,900 $660,200
native soil liner $672,000 $459,400 $503,500
B. Disposal of 10.4 Million Cubic Feet of Solids
4
Impoundment Dimensions. Feet
1000x1000 750x750 500x500
Cost of land, earthmoving,
drainage system and
final cover v $316,000 $235,900 $210,600
Liner cost: clay 273,200 174,700 104,500
native soil 74.500 47.600 28.500
Total: clay liner $589,200 $410,600 $315,100
native soil liner $390,500 $283,500 $239,100
876
-------
and compacted native soils, experimentation to determine
suitability for the service is a necessity.
Disposal Sites with Chemically Stabilized Solids
Published information on chemical stabilization does not include
stabilization of ash alone. Much has been published on stabili-
zation of FGD sludge and mine refuse, and the following
discussion and cost development is based on this information by
inference.
To determine the characteristics of stabilized ash/slag solids
with regard to permeability and strength in compression and
shear, to reach decisions on the suitability of chemical
stabilization for this service, and to determine the economics of
the systems, experimentation on actual materials is needed.
Leo, Fling and Rosoff (1) develop summarized costs for
stabilization and landfilling of FGD sludge containing 40 percent
flyash from a 500 megawatt power generation plant over a 30 year
life. The land used was 320 acres, or about 11 acres per year,
with sludge piled 30 feet high. Land price was $5000/per acre.
Dry solids in the sludge generated totalled 250,000 tons per
year. A previous estimate of total disposal costs considered the
Dravc, IUCS and Chemfix stabilization systems. No distinction
between processes is made in the statement that, in 1977, the
total disposal costs were 1.53 mills per kilowatt hour or $12.27
per ton of dry solids. (It should be noted that the sludge
disposal cost calculated from the plant capacity and the stated
cost per kilowatt hour is:
500,000 x 4,380 x $0.00153/250,000 = $13.40 per ton of dry
sludge solids.)
This cost includes capital and operating costs for all
877
-------
facilities, chemicals, utilities,and manpower required for the
preparation and disposal of the FGD sludge.
In a TVA study (2) the Dravo, IUCS and Chemfix processes were
compared in systems for disposal of about 250,000 tons per year
of FGD sludge containing flyash from a 500 megawatt power station
operating at conditions similar to the previous case. In this
study a midwestern plant location was selected. Land costs were
assumed to be $3,500 per acre. Economic assumptions were as
follows:
o All capital cost estimates based on Chemical Engineering
cost indexes (labor index—237.9, material index--26M .9) .
Capital costs projected to mid-1979. Project assumed to
start in mid-1977 and be completed in mid-1980.
o Direct capital costs covered process equipment, piping and
insulation, transport lines, foundations and structural,
excavation and site preparation, roads and railroads, elec-
trical, instrumentation, buildings, pond construction/
and earthmoving equipment. Material and labor (fabrication
and installation) costs for each of these items were
estimated.
o Indirect capital costs included engineering design and
supervision, architect and engineering contractor expenses,
construction expenses, contractor fees, contingency, allow-
ance for startup and modifications, and interest during
construction. Working capital and land were included in
the total capital requirements. Estimates were based on
current industry practice and authoritative literature
sources.
For the Dravo system, the thickened sludge at- 35 percent solids
was treated with Dravo additives: Calcilox at 7 percent of dry
878
-------
sludge, and Thiosorbic lime at 2 percent of dry solids. A
gravity thickener was used for dewatering the sludge. Treated
sludge was pumped to a clay-lined pond located about 1 mile from
the scrubber facilities. The ponded sludge was assumed to settle
in the pond to 50 percent solids, with excess water recycling to
the scrubber system. The treated settled sludge fixed as a
soillike material in the pond.
For the IUCS system, the effluent from the scrubber system was
dewatered to 60 percent solids using a thickener and rotary drum
filter. The dewatered sludge was fixed by mixing with lime at 4
percent of dry sludge solids. Trucks transported the treated
sludge to the landfill disposal site located 1 mile from the
scrubber facilities. The treated material was assumed to have
claylike properties that allowed placement and compaction in the
landfill with earthmoving -
For the Cherafix system the effluent from the scrubber system was
dewatered to 60 percent solids using a thickener and rotary drum
filter. Dewatered sludge was fixed by mixing with Chemfix
additives: Portland cement, at 7 percent of dry solids, and
sodium silicate, at 4 percent of dry solids. The thickened
sludge, at 35 percent solids, was pumped 1 mile to the disposal
site where it was filtered to remove additional water and mixed
with additives. The treated material was then hauled to the
landfill, placed, and compacted with typical earthmoving
equipment.
The capital investment requirements" for the three processes were
estimated to be:
879
-------
Capital Investment, $1000
IUC5 Chemfix Dravo
Direct costs(1) $ 4,301 $ 5,775 $ 4,943
Other direct costs(2) 581 442 7,410
Indirect costs(3) 2,955 3,700 6,381
Startup and interest(4) 1,666 2,138 3,381
Land (5) 676 693 1,450
Working capital 538 783 550
Total capital investment $ 10,717 $ 13,531 $ 24,115
(1) Does not include trucks, earthmoving equipment*or pond
(2) Trucks and earthmoving equipment for IUCS and Chemfix, pond
for Dravo
(3) Includes 20 percent contingency on engineering, procure-
ment, construction and contractor's fee
(4) Allowance for startup and modification and for interest
during construction
(5) 193 acres for IUCS, 198 acres for Chemfix, 41*4 acres for
Dravo
i
The TVA study concludes that the unit revenue requirements for
sludge disposal are:
Per Ton of Dry Solids Mills/KWH
IUCS $12.55 1.51
Chemfix $16.51 2.00
Dravo $15.32 1.91
Thus, IUCS has an apparent advantage over the others in both
capital and operating costs, while Chemfix has a lower capital
cost than Dravo but a higher operating cost.
In a report (3) on the operation of the Bruce Mansfield power
plant, the capital cost (1975) is presented as $90 million for a
880
-------
complete Dravo FGD disposal system for 3 power units generating a
total of2,475 megawatts. Escalating to 1977 and factoring
downward to 500 megawatt capacity yields an estimated capital
investment of $36 million, which may be compared to the TVA
estimate of $24 million with these modifiers:
o At Bruce Mansfield the slurry is transported 7 miles, as
against TVA's assumption of 1 mile
o At Bruce Mansfield the impoundment area is 1,400 acres and
includes a dam that will eventually reach a height of 420
feet, 2,200 feet long at the crest and 1,550 feet thick at the
base. Although the proportion of the total capital repre-
sented by the impoundment area is not stated, it would
appear to be substantial and account for much of the fac-
tored difference in capital cost
IUCS has applied their Poz-0-Tec process to stabilization of a
mixture of coal mine refuse and FGD sludge (4). In the labora-
tory work the mixture stabilized satisfactorily and exhibited
permeabilities of 1x10~^ to 3x10 "^ centimeters per second. No
costs for this operation were developed. IUCS states that for
FGD sludge the typical cost may range from $3 to $7 per ton of
dry solids. The basis for these costs was not stated, and they
are apparently not on the same basis as those developed by the
TVA study.
In another report of operation of the IUCS system for disposal of
FGD sludge (5) from 1,625 megawatts of generating capacity costs
are developed for sludge disposal onsite in a system engineered,
built and operated by IUCS. About 304,000 tons per year of
solids (dry basis) are impounded in an area of 50 acres. Because
of the demonstrated load bearing characteristics of the stabli-
lized sludge, at over 10,000 pounds per square foot, construction
881
-------
of a 100 foot mound over a period of 20 years appears to be
feasible. The electric company pays an annual fee to IUCS and
supplies operating and maintenance labor, heavy equipment
maintenance, auxiliary power and lime when the demand is over 3.5
percent of the dry solids processed. On this basis, the total
operating cost to the elctric company is $10.78 per ton of dry
solids when the plant load factor is 51 percent. At a 70 percent
load factor the conversion system processes about 416,000 tons a
year and, because the IUCS fee and operating labor are constant,
the total operating cost drops to $8.53 per ton of dry solids.
When these cost figures are prorated to the processing of 250,000
tons per year of dry solids, the total operating cost becomes
$12.60 per ton of dry solids, showing very close agreement with
the TVA study.
The conclusions that may be drawn from this information are as
follows:
o Chemical stabilization appears to offer great promise as a
means of economically disposing of solid wastes without
endangering the environment
o The IUCS stabilization process has the lowest capital and
operating costs of the three processes studied by TVA
o The results of the TVA study of process operating costs for
the IUCS process are borne out by commercial operation
o The available useful information on stabilization of solid
wastes concerns FGD sludge disposal
o Operation of the stabilization processes with the solid
wastes from coal conversion plants may differ from operation
with FGD sludge and may require different process operating
parameters and changes in the types or quantities of addi-
tives, which in turn may affect capital and operating costs
882
-------
o Laboratory and large scale testing of stabilization pro-
cesses is needed to determine the permeability and compres-
sive strengths of stabilized solid wastes from conversion
processes.
o Results of the permeability tests will aid in determining
whether or not the disposal area must have a lining to pre-
vent leachates from reaching ground and surface waters
o Economic studies of the three stabilization processes that
have been discussed, and others that show promise, are
needed to establish the similarities and differences between
processes and between operation with FGD sludge and conver-
sion plant solids
If the assumption is mad'* that data from FGD sludge stabilization
with the IUCS system may be applied to'coal conversion solid
wastes, then estimates may be made of stabilization costs for the
solids from gasification of low and high sulfur coal and from
liquefaction of high sulfur coal. These annual operating costs
and the cost per dry ton of solids processed are shown in TABLE
10-12. The advantage of size of operation, in comparison to the
previously-developed costs for FGD sludge disposal, is immedi-
ately apparent.
References
1. Leo, P. P., Fling, R. B. , and Rosoff, J., "Flue Gas Desulfuriza
tion Waste Disposal Field Study at the Shawnee Power Sta-
tion." Presented at Flue Gas Desulfurization Symposium.
Florida, November 1977. 746*
•Pullman Kellogg Reference File number
883
-------
TABLE 10-12. ESTIMATED ANNUAL OPERATING COSTS FOR
STABILIZATION OF COAL CONVERSION PLANT SOLID WASTES
WITH THE IUCS PROCESS
Gasification
Dry solids, TPY
IUCS annual fee
Operating labor
Maintenance labor
Heavy equipment
maintenance
Auxiliary power
Lime usage
Total annual cost
Cost per ton of
dry solids
Low Sulfur High Sulfur
518,400
616,800
$2,384,000 $2,384,000
140,000 140,000
87,000 96,000
174,000 207,000
159,000 176,000
800,000 949,000
$3,744,000 $3,952,000
$7.22
$6.41
Liquefaction
High Sulfur
700,900
$2,384,000
140,000
104,000
236,000
190,000
1 ,079,000
$4,133,000
$5.90
Note: Costs prorated from data in Reference 5
884
-------
2. Barrier, J. W., Faucett, H. L. , 'and Henson, L. J., "Economics of
FGD Disposal." Presented at Flue Gas Desulfurization
Symposium. Florida, November 1977. 751*
3. Lobdell, L. W., Rothfuss, E. H., Jr., and Workman, K. H.,
"Eighteen Months of Operation (of the) Waste Disposal System
(at the) Bruce Mansfield Power Plant, Pennsylvania Power
Company." Presented at Flue Gas Desulfurization Symposium.
Florida, November 1977. 7^8*
M. Taub, S. I., "The Poz-0-Tec Process for Coal Waste Stabili-
zation." Presented at AIChE annual meeting, New York,
November 1977.6^9*
5. Boston, D. L. , and Martin, J. E., "Full-Scale Waste Disposal at
the Columbus and Southern Ohio Electric1s Conesville Sta-
tion." Presented at Flue Gas Desulfurization Symposium.
Florida, November 1977. 750*
NEED FOR FURTHER STUDY
Coal Dust Control
Conversion of power generation plants to coal firing, construc-
tion of new coal fired generation facilities and the establish-
ment of coal conversion processes all increase the movement of
coal from mines to points of use. Since it is unlikely that coal
users will always be located within conveyor distance of coal
producers, rail transport of coal may increase by an order of
magnitude by 1985 to 1990. Dust problems arising from coal
transportation may be expected to increase in proportion. There
885
-------
appears to be a need for studies directed toward dust control or
dust elimination, with emphasis on the latter.
Coal fines have been treated as a necessary evil in coal clean-
ing. Creation of fines begins at the seam face and continues
through the crushing and screening operations. Reduction in the
fines content of the coal broken from the seam face may be possi-
ble by modification of the tooth profiles of continuous miners,
the angle of attack of the teeth against the face or the tooth
velocity at impact. There is probably little that can be done to
minimize fines creation during strip mining operations, since the
coal must be broken from horizontal seams, instead of vertical
faces as in underground mining.
As pointed out in delineation of the coal dust problem, crusher
design and operation have a significant effect on fines creation.
An investigation into the effect on dust creation of such oper-
ating parameters as peripheral speed, angle of attack of the
crusher mechanism against the coal, tooth or jaw design and feed-
ing method might lead to methods of operation or machine design
that would reduce fines, decrease power requirements, and decrease
machine costs.
Separation of the fines and dust from the main coal stream may be
accomplished by dry or wet screening or by elutriation of dry
fines by an air stream. In most coal preparation plant flow-
sheets the initial breaking of lumps to separate coal from refuse
is a dry operation and dust created at this point may be
collected in baghouses for later combination with wet fines
streams. The total dust stream, however, may contain significant
quantities of undesirable materials, such as clay and silt. In
this case, separation of the non-combustible substances may be
accomplished by air elutriation by taking advantage of
differences in density of coal and non-coal particles.
886
-------
Following the breaker, the coal stream is usually wet screened
and separated into various size fractions from which refuse is to
be separated. Attention to the means of applying water to the
coarse and intermediate size coal streams during their residence
on the screens could ensure that all fines are washed from the
coal and sent to the fine coal cleaning section. Normal
procedure for treating coal fines after separation from refuse
involves combining the fines stream with the intermediate size
stream and drying the mixture. The coal fines are thus
recombined with the main coal stream.
A study of the mechanics and economics of compacting or agglomer-
ating the fine coal stream to larger, non-dusting sizes, could
yield attractive processes for eliminating dust and, possibly,
for increasing the value of the coal: it must be rememberd that
eliminating the dust at th~ source eliminates the need for dust
control methods during transport, storage and handling at the
mine and at the conversion plant site, and that these savings
should defray part or all of the cost of dust elimination.
Solids Waste Disposal
As has been pointed out, the investigations on the applicability
of impermeable liners for disposal areas has been confined to
sanitary landfill application and, to a much lesser extent, to
FGD sludge applications. Because the characteristics of ash and
slag, and mixtures of these with other solids, from coal conver-
sion processes are quite different from those of sanitary land-
fills, FGD sludges or power plant ash, conclusions reached in the
fines testing programs that are now in operation cannot be
applied with confidence to coal conversion solid wastes. A
program for evaluation of liners in coal conversion waste
disposal service would be a most valuable aid in establishing
applicability of liners and economics of their use. This
887
-------
information may become of prime inportance in the future if
environmental standards that concern permeation of the soil by
leachates become more stringent.
With regard to chemical stabilization of solids, for the three
systems (IOCS, Dravo, and Chemfix) that have been discussed and
others that are as yet untried on a large scale, the principal
application has been in stabilizing FGD sludge. Successful as
this application has been, the parameters of chemical stabili-
zation of FGD sludge cannot be applied immediately to solid
wastes from coal conversion processes. For example, the ash/slag
may be either easier or more difficult to stabilize, entailing
more or less additive usage, special handling procedures or
special waste are a management procedures, any of which may
affect the economic attractiveness of the methods. Further, the
permeability of stabilized ash/slag may differ widely from that
of, say, FGD sludge: it may be greater, and thus possibly
unacceptable by present standards, or it may be significantly
less and thus offer an excellent possibility for solid waste
disposal that will meet present standards and possible future
environmental goals. A testing program to determine the physi-
cal, chemical,and economic aspects of chemical stabilization
appears to be necessary and highly desirable.
888
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA-SOO/7-79-228b
4. TITLE AND SUBTITLE
Coal Conversion Control Technology
Volume II. Gaseous Emissions; Solid Wastes
7. AUTHOR(S)
L.E. Bostwick, M.R. Smith, D.O. Moore, and O.K. Webber
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Pullman Kellogg
16200 Park Row, Industrial Park Ten
Houston, XX 77084
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
. RECIPIENT'S ACCESSION NO.
. REPORT DATE
October 1979
. PERFORMING ORGANIZATION CODE
. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
EHE 623A
11. CONTRACT/GRANT NO.
68-02-2198
13. TYPE OF REPORT AND PERIOD COVERED
Final; 4/77 - 11/78
14. SPONSORING AGENCY CODE
EPA/600/13
,6. SUPPLEMENTARY NOTES IERL_RTp project officer is Robert A. McAllister, Mail Drop 61,
919/541-2160.
This volume is the product of an information-gathering effort relating
to coal conversion process streams. Available and developing control technology
has been evaluated in view of the requirements of present and proposed federal,
state, regional, and international environmental standards. The study indicates
that it appears possible to evolve technology to reduce each component of each
process stream to an environmentally acceptable level. It also indicates that
such an approach would be costly and difficult to execute. Because all coal
conversion processes are net users of water, liquid effluents need be treated
only for recycling within the process, thus achieving essentially zero discharge.
With available technology, gaseous emissions can be controlled to meet present
environmental standards, particulates can be controlled or eliminated, and
disposal of solid wastes can be managed to avoid deleterious environmental effects.
This volume (II) deals with the control technology of gaseous emissions and
solid wastes.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COS AT i Field/Group
Pollution
Coal Gasification
Coal Preparation
Wastes
Pollution Control
Stationary Sources
Coal Conversion
Solid Wastes
13B
13H
131
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
381
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (»-73)
888a
------- |