EPA-600/2-76-084
March 1976
Environmental Protection Technology Series
U.S.
, B.C.
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U S Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA 600/2-76-084
March 1976
CONOCO DOLOMITE HOT GAS CLEANUP SYSTEM
by
Earl D. Oliver and Konrad T. Semrau
Stanford Research Institute
Menlo Park, California 94025
SRI Project ECU-3570
Contract No.68-01-2940, Task 024
Technical Monitor: S. L. Rakes
Task Officer: Gary J. Foley
Project Officer: Albert Pines
Office of Research and Development
Washington, D.C. 20460
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
OFFICE OF ENERGY, MINERALS AND INDUSTRY
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
v <
-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing]
1, REPORT NO.
EPA-600/2-76-084
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Conoco Dolomite Hot Gas Clean-up System
5. REPORT DATE
March 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Earl D. Oliver
Konrad T. Semrau
8. PERFORMING ORGANIZATION REPORT NO
SRI Project 3570-24
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Stanford Research Institute
333 Ravenswood Avenue
Menlo Park, California 94025
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-01-2940, task 024
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Energy, Minerals, and Industry
U.S. Environmental Protection Agency
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report analyzes a proposal that EPA sponsor a large-scale pilot plant
to develop the Conoco (formerly Consol) Dolomite Hot Gas Clean-up system. The
report includes a history of the prior development program, the technology involved
comparisons with competitive technologies in regard to technical feasibility,
potential efficiency, and environmental benefits. Future funding implications
are explored. The potential benefits in economy and efficiency of energy conversion
are judged to be substantial, but the cost of development is high. The expected
cost of the pilot plant program may be greater than funding ability. The current
laboratory program should either be continued or should be terminated in a way that
will insure that the knowledge already gained is retrievable.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
iulfur removal
'articulates
,ow Btu Gas
Hot Gas Clean-up
Desulfurization of Fuel
Gas
Particulate Removal
B. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
62
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
CONTENTS
LIST OF FIGURES v
LIST OF TABLES vii
I INTRODUCTION 1
II SUMMARY AND CONCLUSIONS 3
III CONOCO DOLOMITE SYSTEM 7
Process Scheme 7
Conceptual Comparison with Cold Gas Cleanup 8
Applications, Prerequisites, and Criteria 11
History of the Program 14
Major Studies 15
IV ADVANTAGES AND DISADVANTAGES OF HOT GAS CLEANUP 19
Potential Efficiency Advantages 19
Environmental Effects 20
V TECHNICAL FEASIBILITY 23
Desulfurization 23
Particulate Removal 25
Liquid-Phase Glaus Reaction 30
Gasification 31
Nitrogen Oxides 31
VI COMPARISONS WITH COMPETING HOT GAS CLEANUP TECHNOLOGY ... 33
VII ECONOMIC EVALUATIONS 37
VIII FUNDING IMPLICATIONS 45
IX FUTURE HOT GAS CLEANUP DEVELOPMENT 49
Appendix--FLOW DIAGRAM FOR PROPOSED PILOT PLANT 53
iii
-------
FIGURES
1 Proposed Conoco Pilot Plant 7
2 Gasification of High-Sulfur Coal 8
3 Hot Gas Cleanup 9
4 Liquid-Phase Glaus Sulfur Recovery 9
5 Cold Cleaning 10
6 Hot Sulfur and Particulate Removal 10
7 Hot Sulfur Removal, Cold Particulate Removal 11
A-l Flow Diagram Portion of Conoco Drawing XF-3542 53
v
-------
TABLES
1 Potential Gasification Process Applications for Conoco
Dolomite System 12
2 History of CO,., Acceptor Process 14
3 History of Hot Gas Cleanup--EPA Sponsored 16
4 Major Evaluations of Conoco Process 17
5 Environmental Comparison with Cold Gas Cleanup 20
6 Environmental Comparison with Flue Gas Desulfurization ... 21
7 Technical Feasibility 24
8 Hot, Dry Systems for Removal of Particulates and Alkali
Compounds 28
9 Competitors to Conoco Process--Hot Gas Only 33
10 Other Possibilities for In-Situ Hot Gas Cleanup 35
11 Conoco Economic Evaluations 38
12 Conoco Economic Evaluation UpdateRegenerative Acceptor
Desulfurization Process . . 41
13 ES&E Economic Evaluations 42
14 Integrated System Cost Summary 43
15 Funding Implications 46
16 Options for EPA 51
A-l Material Balance Portion of Conoco Drawing XF-3542 53
VII
-------
I INTRODUCTION
The EPA has been supporting laboratory and theoretical development
work on the Conoco (formerly Consol) Dolomite Hot Gas Cleanup system.
The results have been encouraging, and Conoco Coal Development Company
has proposed expanding the program to include development on a pilot
plant scale at Rapid City, South Dakota, adjacent to an ERDA-sponsored
gasifier pilot plant. This would entail a substantial increase in fund-
ing at a time when EPA's budget is being reduced. To help decide on the
level of funding and priorities appropriate to hot gas cleanup, EPA con-
tracted with Stanford Research Institute (SRI) to review the program
and prepare briefing reports. A preliminary briefing was given in Wash-
ington, D.C., on October 21, 1975. The present document is the final
briefing report of this project. The main purpose of this report is
to evaluate the Conoco pilot plant proposal.
SRI reviewed available reports and other information on the Conoco
dolomite system and its competitors and made appropriate plant visits
and direct contacts with personnel involved in development programs.
This report includes:
A history of the program.
The technology involved.
Comparisons with competitive technologies in regard to tech-
nical feasibility, potential efficiency, and environmental
benefits.
Future funding implications.
-------
II SUMMARY AND CONCLUSIONS
The main conclusions of this report are:
The Conoco proposal, estimated to cost more than $18 mil-
lion, would probably lead to substantial overruns of cost
and time.
The Conoco prpcess would be applicable to advanced
gasification/combined-cycle (GCC) power systems, expected
to be developed. However, it is not applicable to any
commercialized gasification process.
The Conoco process would reduce coal consumption (estimates
range from 570 to 157») and hence reduce environmental effects
of coal mining. However, it would lead to greater emissions
of SC>2 and probably NOX than conventional cold gas cleanup.
The economic potential of the process is shown by the fol-
lowing estimates of power costs:
- BCR gasifier--Conoco process--16.7 mills per kwh
- BCR gasifier--Selexol process--21.2 mills per kWh
- Conventional coal-fired steam plant--19.8 mills per kWh.
The Conoco process would be operated at higher gas temper-
ature than competing hot gas cleanup systems under develop-
ment. Accordingly, the Conoco process would be more effi-
cient.
SRI concludes that there is little chance of meeting the projected
time and cost schedule with the Conoco process in its present state of
development. This conclusion is based on a comparison of the require-
ments of the proposed pilot plant program with those of the analogous
program to develop the CCL Acceptor Coal Gasification process. The pro-
posed program includes four separate process developments--gasification,
desulfurization, particulate and alkali fume removal, and liquid-phase
Glaus sulfur recoveryrequiring six developmental reactors. The CO^
Acceptor pilot plant includes only two developmental reactors. An addi-
tional basis for the conclusion is the difficulty of solving mechanical
and chemical problems inherent in the Conoco hot gas cleanup process.
-------
However, the Conoco pilot plant proposal offers the advantages of the
use of some existing facilities and a trained operating crew.
EPA support of a particular process development program depends on
whether the successfully developed process would be used on a large scale.
The principal application of hot gas cleanup is expected to be in GCC
power systems. Technological improvements are needed for GCC systems
to be more efficient and economical than conventional coal-fired steam
plants with stack gas scrubbing. SRl's extensive studies of GCC systems
show that they have no advantage if present commercial coal gasifiers
and gas turbines are used. However, anticipated advances in both tech-
nologies will make GCC systems competitive. Other evaluations of the
combined-cycle systems tend to be more optimistic than those of SRI.
This project requires development of a coal gasifier as well as the
hot gas cleanup process because no suitable gasifier is already developed.
Therefore, Conoco has included a development gasifier as part of its
pilot plant program. This gasifier is a dry ash, fluidized-bed type
which is not likely to be competitive with others under development.
However, the hot gas cleanup process would be applicable to gas from
other gasifiers, so the value of the Conoco program would not be negated
by this factor.
The principal environmental advantage of hot gas cleanup over cold
gas cleanup for GCC systems is that of reduced coal usage and a resulting
decrease of environmental effects caused by mining. While potentially
capable of exceeding EPA standards for new large power plants, the Conoco
process nevertheless would lead to higher sulfur emissions than cold gas
cleanup. The Conoco process would produce more solid wastes (spent
dolomite) and probably NO . The comparison of particulate emissions is
not clear.
All available studies indicate the Conoco dolomite process to be
more economical than cold gas cleanup for GCC power plants. However,
the comparisons are between a developmental process and commercial pro-
cesses. Cost estimates usually rise as more is learned about the require-
ments of a process during development.
Competitive hot gas cleanup systems under development either operate
at lower gas temperatures or are judged not to be promising. Iron oxide
processes operate at much lower temperatures and therefore have less of
the efficiency advantage of hot gas cleanup. Air Products and Chemicals,
Inc., has recently recommended termination of its program to develop a
high-temperature, fixed-bed limestone process. The technical feasibility
of other hot gas cleanup systems including molten carbonate and molten
metal types is regarded as doubtful.
-------
Westinghouse proposes to remove sulfur from fuel gas in the gasifier
vessel, which would restrict the use of the cleanup process to a partic-
ular gasifier. However, Westinghouse is also considering the development
of a separate dolomite cleanup system, which would be a direct competitor
to the Conoco system.
Experimental results of the Conoco process to date have been promis-
ing but have been more exploratory than definitive. If the decision is
made to continue support, the question will remain as to the best way to
proceed. The Conoco proposal might make the process available at the
earliest date, but this is not certain because nonoptimal designs in
large-scale equipment might require major modifications.
A more orderly program would start with construction and testing of
a reliable pilot-scale gasifier, presumably funded by ERDA. Concurrently,
further small-scale tests would be made on the desulfurizer-regenerator
system and the liquid-phase Glaus reactor. In addition, process variable
studies, testing of various dolomites, and particulate removal studies
would be carried out on a small scale. In these studies, reliability
would be shown by closing material balances, especially those involving
sulfur and alkali compound fumes.
To save time, design studies could also proceed concurrently with
three objectives:
Evaluate implications on feasibility and economics of new
findings.
Determine whether any reduction in size of the pilot plant
could provide meaningful data at lower cost.
Prepare detailed design of the proposed facilities.
This approach is common in the process industries. This period would
then be followed by construction and operation of the hot gas cleanup
pilot plant.
It is important to stress that should the EPA decide to abandon
efforts to develop the Conoco process on a large scale at the present
time, orderly termination to preserve the value of the work already done
would be desirable. This would include concluding laboratory work, eval-
uation, and careful documentation to make the information retrievable
should the program be reactivated.
-------
Ill CONOCO DOLOMITE SYSTEM
Process Scheme
The Conoco dolomite system has been advanced as a method of desul-
furizing hot producer gas. Particulates and alkali compounds must also
be removed from the hot gas to gain the maximum potential from the system.
The system may be used in combination with various gasifiers. However,
the configuration analyzed here is that proposed to EPA by Conoco for a
pilot plant in Rapid City, South Dakota. (The Appendix shows the con-
figuration and material balance.)
To simplify the discussion, the process is broken down into func-
tional blocks, as shown in Figure 1. Coal is gasified with air to make
a low-Btu gas containing hydrogen sulfide (HoS). The l^S is removed from
the gas by a circulating stream from which sulfur is removed in another
system shown below. The particulates are then removed to produce a hot
high-pressure clean gas. A particular point about Figure 1 is that each
of the four blocks in the Conoco proposal represents a separate develop-
ment project.
Coal
Air
GASIFICATION
PARTICULATE
REMOVAL
Hot,
High-
Pressure,
Clean
Fuel Gas
FIGURE 1 PROPOSED CONOCO PILOT PLANT
-------
The gasification block (Figure 2) represents three reactors: a coal
pretreater, a gasifier, and a carbon burn-up cell. Gasifier development
is an ERDA responsibility. It is not essential to the Conoco hot gas
cleanup development except that a source of hot, dirty fuel gas is de-
sirable for the pilot plant tests, and no suitable gasifier is available
commercially.
Coal
PRETREATER
800° F
Hot, Dirty
Fuel Gas
To Cleanup
GASIFIER
1750°F
Air
1800°F
Char
BURN-UP
CELL
Steam
and Air
FIGURE 2 GASIFICATION OF HIGH-SULFUR COAL
Figure 3 represents the l^S removal and particulate removal blocks.
The H2S is removed by reaction with calcium carbonate in half-calcined
dolomite. Particulates and alkali compound vapors are then removed from
the gas by means not yet well defined. The calcium sulfide formed by
the removal of H2S is regenerated in a separate vessel by reversal of
the reaction with C02 and steam. The offgas from the regenerator goes
to the liquid-phase Glaus system shown on Figure 4 where the H^S reacts
with S02 to form elemental sulfur.
Conceptual Comparison with Cold Gas Cleanup
To show the potential application of this process, it may be compared
with a conventional cold cleaning system for a combined-cycle power plant
(Figure 5). In the cold cleaning system, the gas is cooled by exchange,
wet scrubbed, desulfurized in one of many commercial processes, and re-
heated by exchange. The reheated gas is then burned and expanded through
8
-------
DESULFURIZER
PARTICULATE
REMOVAL
Hot, Dirty
Fuel Gas
Hot, Clean Fuel Gas
To Power Plant
CO2, H2O and H2S
To Liquid Phase Glaus
C02 and H2O
FIGURE 3 HOT GAS CLEANUP
REACTOR
H2O, CO2,
and H2S
Recycle To
Regenerator
Sulfur
310°F
SO.,
FIGURE 4 LIQUID-PHASE GLAUS SULFUR RECOVERY
-------
Particulates
and Alkali
Compounds
Dirty, Hot, A
High-Pressure 1
Gas
HEAT
EXCHANGE
^ WET
SCRUB
H2S
t
^ COLD
PROCESS
1
GAS
TURBINE
STEAM
PLANT
FIGURE 5 COLD CLEANING
a gas turbine. The exhausting flue gas from the gas turbine is passed
through a waste heat boiler to generate steam. In hot gas cleanup (Figure
6), the heat losses and costs represented by heat exchange are eliminated.
Conoco also claims an economic advantage for hot E^S removal even if hot
particulate removal proves impractical and wet scrubbing is required
(Figure 7). This claim is made on the basis that cheaper exchangers can
be used if they do not face the corrosive action of hot H2S and that
other process equipment is also less costly.
Dirty, Hot, j 1
High-Pressure
Gas
CONOCO
PROCESS
HOT
REMOVAL
GAS
TURBINE
STEAM
PLANT
; i
H S Particulates
2 and Alkali
Compounds
FIGURE 6 HOT SULFUR AND PARTICULATE REMOVAL
10
-------
Dirty, Hot,
High-Pressure
Gas
H2S
t
CONOCO
PROCESS
HEAT
EXCHANGE
?
GAS
TURBINE
Particulates
and Alkali
Compounds
t
WET
SCRUB
1
?
STEAM
PLANT
FIGURE 7 HOT SULFUR REMOVAL, COLD PARTICULATE REMOVAL
Applications, Prerequisites, and Criteria
A key consideration in supporting a development program is a judg-
ment of whether it would be used commercially if the development were
successful. Potential applications are shown below:
Power generation, combined cycle.
Power generation, expansion turbine and close-coupled steam
plant.
High-sulfur coals and oils.
Intermediate pressure (^15 atm).
Intermediate gasification temperature (~1775°F or 968°C);
desulfurization (~1650°F or 900°C).
Applications that are excluded are:
High-Btu gas (not suitable before methanation)
Low-sulfur coals and oils
Methanol synthesis
Ammonia synthesis.
Power generation with a combined-cycle system is the most attractive hot
gas cleanup application. Another application would be power generation
in a plant with an expansion turbine and a close-coupled steam plant.
11
-------
Potential applications would utilize intermediate pressures and inter-
mediate temperatures because of the nature of the chemistry of the pro-
cess. Conditions (proposed by Conoco) of gasification at about 15 atmo-
spheres and 1775°F (968°C) are suitable, but considerable variations from
these conditions would be feasible in particular cases. Potential pro-
cess applications are shown in Table 1.
Table 1
POTENTIAL GASIFICATION PROCESS APPLICATIONS
FOR CONOCO DOLOMITE SYSTEM
Process
IGT U-Gas
BCR 3-stage pyrolysis
Foster-Wheeler (air-blown BIGAS)
Synthane (air-blown)*
Winkler (pressurized, Davy Powergas)
Texaco
Pressure
(atm. )
21
£17
£30
20
£14
27
Final Gas Temperature
(°F)
1,500
2,000
1,700
1,400
1,700
2,000-2,500
Note: Processes operating above 1800°F may require quenching the gas
before desulfurization. Pressures greater than 15 atm. will not
permit the degree of desulfurization attainable by the Conoco
process. Increased H20 and C02 concentrations will also reduce
desulfurization.
.j-
Tested in Process Development Unit.
In addition, the Exxon fluid-bed process is probably in the right
range of conditions, the Shell/Koppers-Totzek pressurized gasifier may
be, and the Battelle-Union Carbide process would be if designed for higher
pressure. Processes with low final gas temperatures are unlikely can-
didate's because of tars and oils in the gas; Synthane and U-Gas may be
marginal-on this score.
Applications exclude high-Btu gas because the degree of removal of
sulfur is not adequate before the methanation step. The process is not
12
-------
applicable to low-sulfur coals and oils because these fuels are environ-
mentally acceptable without desulfurization. It is not applicable to
methanol synthesis because nitrogen and part of the sulfur are left in
the synthesis gas, and it is not applicable to ammonia synthesis because
the degree of sulfur removal is not adequate.
Promoters of combined-cycle systems currently disagree as to the
optimum pressure in the combustors. Generally those with industrial
turbine backgrounds favor lower pressures than those with aircraft engine
backgrounds. SRI believes that the pressure of 15 atmospheres chosen by
Conoco will be compatible with electric utility preferences in the fore-
seeable future.
Prerequisites for use of the process include the following:
Process for pressure gasification of high-sulfur coal.
Control of particulates and alkali fumes (according to
application).
Control of NC) .
X
Gasification of high-sulfur coals tends to be more difficult than gasifi-
cation of low-sulfur lignites or subbituminous coals, because the high-
sulfur coals frequently have caking tendencies. Another prerequisite is
the control of particulates and alkali compound fumes according to the
application. Some applications will have more stringent requirements
than others. A third prerequisite is control of nitrogen oxides. The
formation of nitrogen oxides is poorly defined for the potential applica-
tions of the process, as discussed in Section V.
Potential applications of the Conoco process appear to justify its
development, assuming that economical solutions are found for the tech-
nological problems. The following criteria may be applied in deciding
whether a possible process is worthy of development.
At least 90% removal of H2S
The hotter the better, up to gasification temperature
Elemental sulfur product preferred
Minimum process wastes preferred
Economic advantage.
The potential for at least 90% removal of H£S is desirable to
allow for loss of removal capability in large-scale equipment. This
13
-------
potential may be needed to meet environmental regulations equivalent to
over 80% removal. The criterion that a hotter cleanup temperature is
more desirable applies up to the gasification temperature. Obviously,
there is no point in going above the gasification temperature, because
this would require heating. Any decrease of the cleanup temperature
from the gasification temperature means that part of the advantage of
hot gas cleanup is lost. However, some cooling will probably be needed
to remove alkali compound vapors. A process should preferably make ele-
mental sulfur because of market and environmental considerations. Ob-
viously, a process that produces minimum process wastes is preferred,
and there is no incentive to develop a process unless it has an economic
advantage.
History of the Program
Table 2 shows a summary of the history of the CC>2 Acceptor process,
which is a predecessor of the proposed hot gas cleanup process. High-
lights are shown starting with the initial demonstration in 1950. There
was no aggressive development until future shortages of natural gas be-
came more apparent in the late 1960s, when government support accelerated
the work. The costs of the development program are of interest--$9.7
million for the pilot plant itself and $36 million for the total process
development.
Table 2
HISTORY OF C02 ACCEPTOR PROCESS
Date Sponsor Subject
1950 Conoco Laboratory demonstration
1962-4 Texas Eastern/ Laboratory program
Equitable Gas
1964-8 OCR Bench scale work and economics
1966-72 OCR-ERDA Design and construction of pilot plant
1972-present OCR-ERDA Operation of pilot plant
Cost--$9.7 million for pilot plant
$36 million total process
14
-------
Table 3 shows a summary of the history of the hot gas cleanup pro-
cess, as sponsored by the EPA. This was started in 1970 with the in-situ
desulfurization concept of combined gasification and sulfur removal.
Various adaptations of the CC>2 Acceptor process were investigated; the
feasibility of producing boiler fuel was studied; and laboratory work
was performed. The laboratory work included preoxidation kinetics, com-
bined gasification and desulfurization, and acceptor cycle life. It was
concluded that for a low-Btu gas in which the nitrogen from the air ends
up in the product gas, there is no advantage to combined gasification
and sulfur removal. In the period 1972 to 1973, the separate gasifier
concept was developed and a feasibility study was performed. Laboratory
work included tests of the gasifier, carbon burn-up cell, pretreatment,
acceptor cycle, and liquid-phase Glaus reaction. In the 1974 and 1975
period, a preliminary pilot plant proposal was developed, and laboratory
work included deactivation of dolomite, makeup of dolomite to approach
steady state, hardening of dolomite and attrition, reaction of calcium
sulfite and hydrogen sulfide, removal of particulates and alkali fumes,
and the Chance reaction." Proposed work includes the building and opera-
tion of a pilot plant. Supporting laboratory work consists of liquid-
phase Glaus studies, hot fixed-bed filter and cyclone studies, trace
elements analysis, a thermal gravimetric analyzer study, scanning electron
microscope studies, and work on an Illinois dolomite. In addition, an
environmental assessment is proposed. The cost of the program to date
has been $1.2 million.
Major Studies
Table 4 lists major evaluations of the Conoco process reviewed by
SRI. Conoco, Environmental Science and Engineering, United Technologies
Research Center, Foster Wheeler Corp., and Stone & Webster have made or
are making studies. All of these show capital cost advantages for the
Conoco process compared with conventional processes, although some recog-
nize major development problems. Besides these direct evaluations, SRI
considered other major sources, such as the Aerotherm review, Gilbert
Associates report, SRI combined-cycle and gasification multiclient studies,
and SRI private particulate removal studies.
CaC03 + H2S.
15
-------
Table 3
HISTORY OF HOT GAS CLEANUP
EPA SPONSORED
Date
1970-71
Subject
1972-73
1974-75
Proposed
Cost
In-situ desulfurization concept
(combined gasification and sulfur removal)
Three adaptations of C02 Acceptor
Feasibility study
Laboratory work
Revised concept--separate gasifier
Feasibility study
Laboratory work
Preliminary pilot plant proposal
Laboratory work
Build and operate pilot plant
Laboratory work
Environmental assessment
$1.2 million to date
16
-------
Table 4
MAJOR EVALUATIONS OF CONOCO PROCESS
Evaluator
Conoco
ES&E
UTRC/Foster
Wheeler
Stone & Webster
for EPRI
Competitive Processes
Hot potassium carbonate
Hot potassium carbonate
Stretford
Selexol
Flue gas desulfurization
Cold cleanup
Conclusions
Big advantage with hot
removal of particulates.
Less advantage with wet
scrubbing.
Conoco cheapest
Conoco cost competitive
NOX problem
Preliminary: economic
advantage at environmental
penalty
17
-------
IV ADVANTAGES AND DISADVANTAGES OF HOT GAS CLEANUP
Potential Efficiency Advantages
The most promising application for hot gas cleanup processes is
integrated with combined-cycle power generation (Brayton-Rankine cycle)
as previously depicted in Figures 5 through 7. Combined-cycle plants
are now in operation using clean gaseous or liquid fuels, in competition
with steam cycle (Rankine cycle) plants burning the same fuel to raise
steam. There has been only one commercial application of producer gas
in a combined-cycle system (the STEAG plant in Germany). This plant
uses a low-sulfur coal and does not desulfurize the gas. SRl's exten-
sive gasification and combined-cycle studies have shown that combined-
cycle systems using current gasifier and turbine technology give no
economic advantage over conventional coal-fired steam power plants with
flue gas desulfurization. However, expected advances in both gasifier
and turbine technology will make combined cycles relatively more attrac-
tive, especially if the advantages of hot gas cleanup can be realized.
The potential efficiency advantage of hot gas cleanup (both sulfur
and particulates) may not be as great as is sometimes assumed because
some of the available heat can be recovered by exchange prior to cold
gas cleanup. However, hot gas cleanup does offer at least 5% decrease
in the coal required for a given amount of power. Thus, the capacities
of the mine, gasifier, and solid waste disposal can be decreased and
heat exchangers for heat recovery are not required.
If it proves necessary to wet scrub the gas to remove particulates
and alkali compounds, most if not all of the potential efficiency advan-
tage is lost. However, the hot cleanup may possibly have a lower energy
requirement than the cold cleanup, depending on gas composition and
process specifications.
Besides the combined-cycle approach, the hot producer gas may be
expanded through a turbine without prior combustion. Such a system
would be similar to power recovery from catalytic cracking unit regen-
erators in the petroleum industry and might be more tolerant of partic-
ulates in the gas. Hot gas cleanup would have advantages over cold gas
cleanup comparable to those in combined-cycle comparisons. Unfortunately,
19
-------
the thermal efficiency of the expansion turbine approach is less, and
conventional combustion with flue gas desulfurization is expected to
be more economical.
Environmental Effects
Recognizing the potential efficiency and economic advantages of
hot gas cleanup, we should consider environmental effects. Table 5
shows a comparison with cold gas cleanup. The main advantage is that
less coal is required because of the higher efficiency, and this means
less land disturbance by strip mining. The process, although meeting
EPA emissions requirements, is at a disadvantage compared with cold gas
cleanup in the amount of sulfur oxides emitted. Other disadvantages
include more solid wastes and probably more NOX emissions. The higher
NOX emissions result from the residual ammonia in the hot gas. The
comparison of particulate emissions is not cleat:, because so little is
known about the performance of particulate removal equipment at high
temperature and pressure.
Table 5
ENVIRONMENTAL COMPARISON
WITH COLD GAS CLEANUP
Conoco Advantages
Less coal required means less land disturbance
Less steam required if liquid-phase Glaus is successful;
therefore less blowdown
Probably less power required
Possibly less organics in wastewater if applied to pro-
cesses whose product gas contains tars and oils
Conoco Disadvantages
More sulfur emissions
More solid wastes with water pollution from run-off
Probably more NO emissions
Not Clear
Particulate emissions
20
-------
Table 6 compares the Conoco process with a conventional coal-fired
power plant with flue gas desulfurization by lime or limestone scrubbing.
Hot gas cleanup produces less solid waste and probably less sulfur emis-
sions. Comparisons of NOX emissions, land disturbances, and particulate
emissions are not well defined and would depend on specific designs.
Table 6
ENVIRONMENTAL COMPARISON WITH FLUE GAS DESULFURIZATION
Conoco Advantages
Less solid waste
Probably less sulfur emissions
Not Clear
Land disturbance (depends on combined-cycle technology)
NO emissions
X
Particulate emissions
21
-------
V TECHNICAL FEASIBILITY
Table 7 lists items to be considered in regard to technical feasi-
bility that will be discussed in this section. The heart of the hot gas
cleanup idea is desulfurization, which entails fluidized-bed absorption
and fluidized-bed regeneration. Some dolomites tested have shown absorp-
tivity and limited (but perhaps adequate) regenerability without excessive
attrition in small-scale equipment. Operability of large plants may de-
pend on maintaining adequately low attrition rates in the large equipment.
The usefulness of the process depends on having adequate dolomite acceptor
resources with predictable performance characteristics. The failure of
the Air Products and Chemicals' process (discussed in Section VI) and the
great variability of dolomite properties raise some doubt on this point.
Desulfurization
Absorption is by reaction with half-calcined dolomite:
H2S + MgO CaC03 MgO CaS + H20 + C02
Regeneration is by reversal of this reaction. Dolomite is considered
preferable to limestone because removal of the C02 combined with MgO
gives the solid a porous structure. Conoco has determined experimentally
the equilibrium constant for this reaction. Equilibrium constants can
also be calculated from thermodynamic data published by I. Barin and 0.
Knacke entitled Thermochemical Properties of Inorganic Substances
(Springer-Verlag, Berlin, 1973). Constants from the latter source are
about two and one-half times the Conoco values, but the Conoco values
are judged to be more reliable.
Equilibrium limits the potential removal of H2S, but more than 96%
removal is possible in the Conoco scheme. Removal is increased by in-
creasing the temperature but reduced by increasing the pressure. The
temperature may be limited by deactivation of the acceptor. In the pro-
posed pilot plant, which is integrated with a dry ash gasifier, the gasi-
fier temperature is limited by ash fusion. Regeneration is conducted at
a lower temperature than absorption. The raw exit gas from the gasifier
23
-------
Table 7
TECHNICAL FEASIBILITY
Desulfurization
Absorption
Regeneration
Attrition
Plant operability
Acceptor resources
Particulate Removal
Filter
Inertial
Electrostatic
Liquid-Phase Glaus
Reactivity
Contaminant buildup
Gasification
Pretreater
Gasifier
Carbon burn-up
Nitrogen Oxides
Ammonia
Atmospheric fixation
Pressure effects
reheats the regenerated acceptor. Therefore, the absorption temperature
must be lower than the gasification temperature.
24
-------
Regeneration requires reducing the temperature to make the equilib-
rium favorable for regeneration, although this lowers the reaction rate.
There has been limited exploration of the overall effect of temperature.
Raising the pressure would favor regeneration, but operation of absorp-
tion and regeneration at different pressures would raise too many prob-
lems to be recommended.
The concentration of I^S in the regenerator gas is limited by equi-
librium and perhaps by kinetics to relatively low levels, about six times
that in the producer gas. The purpose of hot gas cleanup is to avoid
the heat losses from cooling the producer gas to low temperatures. Un-
fortunately, because of its higher specific heat, the heat capacity of
the regenerator gas in the Conoco scheme is roughly one-fourth that of
the raw producer gas, and the regenerator gas is cooled to a low tempera-
ture in the Conoco scheme.
Attrition of some dolomites has been acceptable in small-scale tests.
Conoco reports that attrition was less in the pilot plant than at the
laboratory scale for the CC>2 Acceptor process. Other processes, such as
the alkalized alumina and the TOSCO II, have experienced greatly increased
attrition of circulating solids at larger scales. The attrition rate
may depend on proper design and operation of the plant. Only large-scale
tests can reliably determine the attrition rate. Therefore, the possi-
bility of excessive attrition remains as a hazard to the success of a
development program. Excessive attrition could affect the operability
as well as the economics of the process; however, this possibility is
not considered serious.
Conoco has found that deactivation of the acceptor through the for-
mation of large crystallites required more make-up than did attrition.
This deactivation, of course, may also be affected by plant size and
design.
Particulate Removal
The collection of particulates present in hot low-Btu gases may be
required for one or more of at least three reasons:
To prevent solids from interfering with the removal of H9S
from the gas.
To prevent emissions to the atmosphere after subsequent use
of the gas as fuel.
To protect the gas turbines in a combined-cycle power gener-
ation system.
25
-------
Evidently, the third purpose in general poses the most severe require-
ments, although future environmental regulations could change this pic-
ture. Protection of the turbines requires not only that the total quan-
tity of particulates be reduced to a very low level, but that the quantity
of particulates larger than 1 to 2 microns should be reduced to a partic-
ularly low level. These measures are necessary to prevent erosion of
the turbines's blades by impact of abrasive particles or buildup of solid
particles on the surfaces of the turbine blades or other components.
A related problem is the presence in the gas of alkali metal com-
pounds that have been volatilized from the coal (or oil) being gasified.
These materials present a special problem because they may be present in
the vapor state at the higher temperatures (perhaps 1200° to 1800°F) en-
countered in gasification systems. Most other inorganic compounds will
be in the solid state in the same temperature range. The alkali-compound
vapors are thus not collectible by the dust collection equipment that
may be applicable to the particulates. However, once the temperature
drops low enough (perhaps about 1200°F), the vapors can condense, char-
acteristically forming a fine fume of solid or liquid particles in the
submicron particle-size range. These fumes can be removed only by high-
efficiency collectors. The Conoco conceptual design shows cooling to
1300°F before particulate removal.
The alkali-compound vapors might be adsorbed on solids in the system.
However, the likelihood that available solids in the systems will have
much affinity for adsorption of the alkali vapors is rather remote. On
the other hand, once the condensation temperature of the vapor is closely
approached or reached, the vapor may tend to condense on the surfaces of
the nonvolatile solids. The most likely surfaces for such condensation
of the alkali compounds are the particulates suspended in the gas stream.
Such nucleation might deposit alkalis on larger particles that can be
more readily collected. There is, perhaps, some evidence for such a phe-
nomenon to be drawn from the distribution of alkalis in the fly ash from
the burning of coal. Nevertheless, it is likely that much of the alkali
material will be self-nucleated and will condense to fume particles di-
rectly. There is ample evidence for such behavior from experience with
various industrial dusts that contain mechanically entrained particles
of nonvolatile compounds as well as fume from the condensation of alkali
metal compounds or other volatile inorganic compounds.
Conoco personnel have cited reports of experience with incineration
of salty sludges in fluidized beds of sand. In these cases, volatiliza-
tion of sodium was suppressed, at least in part, because the NaCI reacted
with water vapor and silica to form the nonvolatile sodium silicate while
releasing the chlorine as HCl. Although the reaction should indeed tend
26
-------
to suppress initial volatilization of the sodium, it does not appear that
the reaction would be of much assistance in collecting the sodium once
it has been volatilized. NaCl can be hydrolyzed in the vapor state by
water vapor, but the resulting compound, sodium oxide, is even more
volatile than the NaCl.
The concern with alkalis arises not so much from the possibility of
abrasion, but from the dangers of corrosion and buildup on the turbine
blades. These dangers are related not merely to the amounts of alkalis
but also to the quantities of other materials, particularly vanadium and
sulfur, that may be present. Sulfates and vanadates of the alkali metals
are known to produce specific types of corrosive attack on ferrous metals
at high temperatures.
As shown by wide variations in the recommendations of manufacturers,
no firm data indicate just what level of particulates or alkalis can be
tolerated in gas turbines. The problem is that the gas turbines that
are expected to make combined-cycle systems competitive have not yet
been developed. The limits proposed by Westinghouse are the most com-
plete and elaborately derived, but their application to turbines of the
future is conjectural, and their relevance to existing turbines is not
clearly established. Future turbines may use transpiration cooling, in
which a cooling fluid moves away from the blades. This fluid could
conceivably protect the blades from deposition. However, such considera-
tions are outside the scope of this study.
For removal of particulates and alkalis (once the latter are con-
densed) from the hot gases, the general methods and devices potentially
available are essentially limited to those presented in Table 8 or to
closely related devices. SRI does not believe that cyclones can effec-
tively remove the smallest particles. Cyclone efficiencies can indeed
be increased by using small-diameter units and high inlet-gas velocities
(the latter at the expense of increased erosion), but all available ex-
perience suggests that the ultimate practical possibilities of cyclones
are distinctly limited. In any case, however, cyclones will probably
be highly useful if not vital as precollectors ahead of high-efficiency
final collectors.
Electrostatic precipitators are in principle well adapted to collec-
tion of fine particles at relatively high temperatures and pressures.
However, their reliability is questionable under circumstances where
even a relatively brief outage might permit severe damage to turbines.
The upper range of temperature for precipitation is set primarily by the
thermal ionization of the gas. As temperature rises, the system pressure
must also be increased to suppress disruptive sparking and preserve a
27
-------
Table 8
HOT, DRY SYSTEMS FOR REMOVAL OF
PARTICUIATES AND ALKALI COMPOUNDS
Device
Cyclone
Electrostatic
precipitator
Granular bed
filter
Woven metal
fabric
Porous metal
sheet (sintered
granules or
fibers)
All types
Principle of
Operation
Inertial deposition
Electrophoresis
Depth filterpri-
marily inertial
deposition
Surface filter--
primarily inertial
deposition
Surface filter--
primarily inertial
deposition
Limitations
Low efficiency in low-micron
and submicron range
Probably limited to not higher
than 1600-1800°F. Reliability
uncertain
Limited efficiency on fine
particles if granules are large
enough to permit easy cleaning
Relatively low face velocities.
Possible blinding and cleaning
problems
(Same)
Removal of alkali and other
vapors not feasible unless
temperature is reduced to pro-
duce condensation.
corona discharge. Currently, it is not known whether it is feasible to
operate a precipitator above about 1600°F. The dimensional stability of
the precipitator structure itself will tend to become critical at higher
temperatures. If alkalis are to be collected, a maximum temperature of
about 1200°F may be permissible to ensure that the alkalis are in the
particulate state.
Filters provide a positive control on passage of particulates, and
probably represent the best available approach. As with the other de-
vices, 1200°F may be a practical upper temperature limit if alkalis must
be efficiently collected.
28
-------
The potential capabilities of different types of high-temperature
filters differ substantially, but the filters can be divided broadly
into two classes surface filters and depth filters. With the surface
filters, the dust collects on the face of the porous filter medium so
that after an initial period the effective filter medium is the layer of
collected dust itself. Such filters (comparable to conventional fabric
filters) are inherently high in efficiency once the initial dust layer
is formed. Filtration velocities are necessarily low if extremely high
gas pressure drops are to be avoided, so that a relatively large filter
area is required.
Cleaning can be effected by shutting off the gas flow and using a
reversed pulse of clean gas. However, it is possible for the filter
medium to become "blinded" by fine particles that may become wedged in
fine pores of the filter medium and thereafter not be dislodged by the
normal flow of clean gas.
With the depth filters (typically composed of deep beds of granules
or fibers), the interstices in the packing are large and particles fil-
tered from the gas are intercepted by the packings and deposited within
the porous structure. Reducing the size of the packings and increasing
the depth of the bed both increase the efficiency of particle collection,
but at the cost of increased gas pressure drop, which in turn may require
use of lower gas velocities and greater filter face areas for a given
gas flow.
The necessity for cleaning the filter at frequent intervals or con-
tinuously imposes restrictions on the form and size of the packings.
Cleaning essentially involves taking the bed apart, removing the collected
dust present in the interstices or adhering to the packings, and then re-
forming the bed. Such a procedure is not practical with fibrous packings,
but can be carried out with granular packings. Also, the granular pack-
ings must be relatively coarse (probably not smaller than 2 to 3 milli-
meters in diameter), since they must be cleanly and easily separated
from the collected dust. The bed must be reformed simply by allowing
the granules to flow back into place.
Both intermittent and continuous cleaning systems have been used
with granular-bed filters. In the first system, the bed is periodically
fluidized by a reverse jet of clean gas, which carries off the collected
dust to a secondary collection system. In the continuous cleaning sys-
tems, the granules are fed into the filter and then withdrawn as they
become loaded with collected dust. The dust is then dislodged and
separated by running the granules over a screen, or by some equivalent
procedure.
29
-------
Because of the relatively large permissible size of the granules,
the collection efficiency for fine dust or fume will tend to be low
unless a deep bed of granules is used. A fairly high gas velocity must
be used to keep the size of the filter within reasonable limits as well
as to increase the filter efficiency.
In general, the basic approaches and devices for particulate collec-
tion at high temperatures and pressures are evidently available, but it
appears that very little has yet been accomplished in developing the
basic design data that will be necessary for selection and application
of appropriate systems to gasification processes.
Liquid-Phase Glaus Reaction
The liquid-phase Glaus (also known as Wackenroder) reaction is:
2H2S + H2S03 3S + 3H20
In the proposed pilot plant, SC>2 from a cylinder and 1^0 are mixed with
the regenerator gas consisting of CC>2 and H^O vapor with t^S at low con-
centration. Conditions are maintained such that little or no l^O con-
denses, since condensation would put a severe thermal penalty on the
process. In a commercial process based on this concept, part of the sul-
fur would be burned and the resulting SC>2 absorbed in water to make
HoSO-j. Such a commercial operation would be a potential source of pol-
lution from the SC>2 content in the absorber tail gas.
The liquid-phase Glaus reaction has been demonstrated in short-term
tests on a small scale. The feasibility of the design may depend on the
reactivity and contaminant build up in large equipment over several
months. The chemistry of aqueous sulfur compound systems is extremely
complex, with disproportionation, other side reactions, and corrosion
being possible. The reaction rate per unit volume may be less in large
equipment because of less efficient contacting of phases.
Conoco believes, and SRI concurs, that no commercial process is very
suitable for the removal of I^S from the mixture of C02 and 1^0 that re-
sults from the regeneration of CaS. A standard Glaus unit is not suitable
because of the low concentration of l^S and the high concentration of l^O.
The high partial pressure CC^ would adversely affect other processes,
such as Stretford. Unfortunately, this means that another process develop-
ment is required for success of the desulfurization process. However,
30
-------
the gas out of the liquid-phase Glaus reactor is not emitted to the atmo-
sphere, so the removal requirement for I^S is lenient, although residual
HoS will repress the regeneration of the acceptor.
From the standpoint of the desulfurization and liquid-phase Glaus
systems, the pressure of gasification should be as high as feasible for
meeting sulfur removal requirements. High pressure promotes regenera-
tion and gives increased concentrations of HoS in the regenerator offgas.
Gasification
The primary purpose of hot gas cleanup is sulfur removal; therefore,
it is applicable only to fuels with sulfur contents too high to meet
emissions standards. Conoco proposes to use Illinois No. 6 coal in the
development program. A dry ash, fluidized-bed gasifier is proposed.
Illinois No. 6 is a moderately caking coal, and successful operation of
this type of gasifier requires that the caking tendencies be destroyed
by low temperature oxidative pretreatment. Both the gas and the pre-
treated coal are transferred to the gasifier. This type of gasifier does
not consume all the carbon; therefore, the solid residue is transferred
to a carbon burn-up cell. The proposed system is similar to the Synthane
pilot plant now under construction.
The proposed gasification system has theoretical disadvantages com-
pared with agglomerating ash and slagging gasifiers now under development.
If any of the other programs are successful, the Conoco gasifier is un-
likely to find wide use. However, the hot gas cleanup system is separate
from the gasifier and could be integrated with other gasifiers.
Illinois No. 6 coal is expected to be much easier to handle than
strongly caking coals like Pittsburgh seam. If the proposed Conoco pro-
gram is successful, a proposal to extend the work to more difficult coals
is likely.
Nitrogen Oxides
Finally, feasibility depends on meeting nitrogen oxides requirements.
The Conoco process will not remove ammonia, which is a capability and
necessity of cold aqueous cleanup systems. Ammonia may lead to more NOX,
depending on the conditions of combustion. NO formation from the com-
X
bustion of ammonia-containing fuel gas does not correlate well with the
equilibrium of the nitrogen-oxygen system.
31
-------
Increased nitrogen oxides formation is not necessarily an unavoid-
able consequence of hot gas cleanup. Some gasification systems form
little ammonia from nitrogen in the fuel. Also, the nitrogen content
of fuels varies widely. Most petroleum fractions have lower ratios of
nitrogen to sulfur than is common in solid fuels.
In the present state of development, neither the formation of NOX
in power systems nor the environmental requirements are known. But: it
is possible that this factor will dictate cold gas cleanup instead of
hot in some applications.
32
-------
VI COMPARISONS WITH COMPETING HOT GAS CLEANUP TECHNOLOGY
Table 9 shows other hot gas cleanup processes that are competitors
to the Conoco process. Only hot gas processes are considered, because
many cold processes are commercially available, and for our purposes
there is no reason to choose among them. There are three iron oxide
processes under development: the Babcock and Wilcox, the Bureau of
Mines, and the Battelle processes. None of these is far along in develop-
ment. All operate at temperatures 300 to 600°F lower than the Conoco
process. Regeneration forms S0?; hence, the most likely by-product is
sulfuric acid, although Foster Wheeler has a process in the development
stage by which elemental sulfur can be made.
Table 9
COMPETITORS TO CONOCO PROCESS--HOT GAS ONLY
Iron Oxide Processes
Babcock and Wilcox
USBM (ERDA) - Sintered with Flyash
Battelle - with Proprietary Additive
300-600°F Lower than Conoco
Make Sulfuric Acid
Fixed Bed Limestone or Dolomite
Air Products - Abandoned
Molten Carbonates
Battelle Northwest
Molten Metal
IGT - Meissner - Proprietary @ 800°F
33
-------
The removal of sulfur is said to be limited by equilibrium in the
reaction:
FeO + HS FeS
However, there is considerable disagreement as to the sulfur removal
attainable. Nevertheless, it is clear that reducing the temperature
favors the desulfurization reaction, an effect opposite to that in the
dolomite scheme, and that the equilibrium removal is independent of the
pressure.
Besides the iron oxide processes, other possibilities include con-
tacting the gas with a fixed bed of limestone or dolomite. The developer
of this process, Air Products and Chemicals, has recommended termina-
tion of the project because of poor results and unfavorable economics.
Although this eliminates one competitor to the Conoco process, it also
raises some question as to the uniformity of results with dolomite.
Unfortunately, the report on the Air Products work is not yet available.
Desulfurization by contacting with molten carbonates is being tested
by Battelle Northwest. SRI believes this process does not merit consid-
eration for combined-cycle systems until it can be proved that vaporiza-
tion and entrainment of highly corrosive molten salts do not damage tur-
bine components.
Finally, contacting with an undisclosed molten metal is said to be
the basis of the IGT-Meissner proprietary process. Because it operates
at about 800°F, it has less potential for efficiency improvement than
the Conoco process.
Other possibilities (Table 10) include in-situ hot gas cleanup by
gasification in molten iron or molten salt, neither of which is in an
advanced stage of development. Two other systems using dolomite or lime-
stone are the EPA-sponsored Esso England CAFB oil gasification system
and one version of the Westinghouse gasifier. Westinghouse is now con-
sidering desulfurization either in the gasifier or in a separate vessel.
The separate vessel process would be a direct competitor to the Conoco
process. The in-situ systems would not be generally applicable to gas
produced by another process.
Not shown on these tables is a suggestion by Kennecott, based on
thermodynamic studies, that copper is the best potential reactant for
H2S in the 1500°F range.
34
-------
Table 10
OTHER POSSIBILITIES FOR IN-SITU HOT GAS CLEANUP
Gasification in Molten Iron
Applied Technology PATGAS
Gasification in Molten Salt
Atomics International
Kellogg
Dolomite or Limestone
Esso England CAFB
Westinghouse gasifier
35
-------
VII ECONOMIC EVALUATIONS
Several economic evaluations of the Conoco dolomite system have
been made, usually giving comparisons with established processes. All of
these are based on one or another conceptual design of the Conoco system,
but the estimates for competitive systems were not all derived in the
same way. The results and comments on their validity are given in this
section.
The most detailed work has been done by Conoco; Table 11 gives the
results of evaluations made in 1973 assuming start-up in 1978. All of
these evaluations are based on comparable conceptual designs for a con-
stant coal rate, which is not constant power output because of varying
energy requirements. The costs include only gas cleanup and sulfur
recovery, not power generation. The figures are considered applicable
to a power plant of about 1,400 MW capacity; based on this capacity and
the higher heating value plus sensible heat above 300°F of the fuel gas,
the heat rate is 8,000 Btu/kWh. In addition, an excess of 160 MW, or
11%, is generated by expansion of the fuel gas to 10 psig. The annual
electrical output is assumed to be 707= of capacity.
The first column in Table 11 applies to the Conoco dolomite process
with hot removal of particulates and alkali compounds, assuming that
cyclone separators are suitable for this job. The second column is for
the same desulfurization system but assumes that the gas must be cooled
by exchange, wet scrubbed to remove particulates and alkali compounds,
and reheated by exchange. This second cleanup method, of course, is
more expensive than the first, but according to these estimates, it is
cheaper than the third method shown in the third column. In the third
method, the gas is cooled by exchange, wet scrubbed, desulfurized by the
hot potassium carbonate process, and reheated by exchange. The advantage
for hot desulfurization even with cold particulate removal depends upon
relative costs of hot and cold desulfurization systems. An additional
cost advantage of hot desulfurization may be cheaper heat exchanger
materials resulting from lessened corrosion potential of the desulfurized
hot gas.
These estimates were made by the Conoco Engineering Department.
The conceptual design of the Conoco dolomite system was based on lab-
oratory data. The conceptual design of the hot potassium carbonate
37
-------
cd
cu
CJ
cx,
O
cd
§
H
4-1
fi
01
o
CJ
CJ
4J
Cd
c
o
rQ
rJ
cd
O
3
H
CO
CO
cd
4J
0
a,
O
W
CJ
en
O i-l
0
O X
txT i-l
CM 1
o-
^ II
CO OO
83,957
OJ 03
X X
CO
O -co-
co
oo
O
f!
u
CJ
S
ion process
4-1
cd
N
r-l
J^
3
14-1
i 1
3
cn
cu
Q
rticulate and alkali remova
cd
a.
14-4
o
*o
o
-C
4-1
o
S
cu
^
H
3
cr
CJ
}-i
r-l
cd
o
o
CJ
r-l
3
4-1
cn
H
^
ed
CJ
r>,
- ..
co
C
O
H
cd
CJ
3
4-1
pq
CJ
3
r-l
cd
60
C
1-1
4-1
cd
o
Si
S-i
o
f"!
60
H
33
producer gas to station
T3
CJ
N
H
r-l
3
14-1
i i
3
en
CJ
Q
(pound mols/hour)
en
O
4J
cd
fe
O
1 1
fe
o
CJ
3
4-J
cd
J-J
cu
a,
e
cu
H
y^
60
en
a
CJ
^
3
en
en
CJ
^j
P-i
cd
CJ
3
4-1
PQ
CU
r-l
cd
60
a
r-l
4J
cd
CJ
si
t-i
CJ
Si
60
r-l
m
3
4-1
2-
4-1
fi
CJ
4-1
a
o
o
4-1
cd
CJ
CJ
i i
rO
H
en
C
CU
en
-f-
£>
pr|
PC
-J;
s-*\
I-l
cd
CJ
en
i
r 1
cd
C
cd
4-1
cn
O
CJ)
plant cost ($ millions)
T)
cu
1 1
i4
cd
4J
CO
a
M
4-1
cn
O
o
CO
i
ON
i 1
OO
r--
i4
0
4-1
C
O
1-1
4J
cd
i i
cd
o
cn
w
during construction
4J
cn
CJ
}-i
CJ
4J
C
1 1
investment
i i
cd
4-1
0
H
en
O
H
i i
r-l
H
6
co-
tI
cd
4-J
r-l
p (
cd
CJ
60
C
r-l
Jat
r-l
o
38
-------
01
T)
O
c
Xi
ca
H
ing costs ($ millions)
4J
ca
0)
ft
o
i i
ce
3
d
0
CO
00
oT
co-
co
o
I 1
CN
CO
in
-co-
co
CO
ca
XJ
00
O
t 1
*\
4J
CO
O
o
60
0
H
4J
cd
fa
0)
ft
o
4-1
O
0)
H
O
1
00
o
00
vD
O
0
0
4J
0
,-1
4J
cd
M
o
4J
ft
Q)
O
CJ
^
co-
co
o
co-
ed
O
o
O
4J
QJ
ft
cn
}-(
ca
i i
i i
o
p
00*
CM
O
CM
OS
i 1
3
4J
pq
0
O
1-1
r 1
i 1
r-l
e
^
QJ
ft
CO
4J
0
QJ
CJ)
product gas to station
14-1
O
to
C5
fa
O)
4-1
0
H
-<=>-
CO
CM
CM
r-l
CM
3
4J
§
H
r-l
H
e
fa
01
ft
tn
4_)
0
QJ
O
s
3
4-J
pq
g
H
1-1
1-1
H
e
fa
QJ
ft
cn
4J
0
QJ
o
CM
CO
CO
CM
r-l
O
CM
4-J
cd
0)
QJ
, |
Xt
H
CO
0
QJ
cn
OS
O
r-l
oo
r-l
generated by expander, MW 160
i-i
01
£5
O
ft
cn
t/3
QJ
0
X
w
fn
o
0
O
CO
M-l
O
0)
}-l
3
4J
ca
r4
O>
ft
e
0)
4-1
4J
QJ
r-<
4J
3
O
fa
QJ
4-J
cd
0)
fa
H
ca
*O
QJ
B
3
tn
tn
ca
0
ca
01
^
o
Xi
ca
4J
0
QJ
4J
0
O
o
4J
ca
0)
X!
0)
XI
H
tn
C
Q)
CO
!<
^
r 1
ca
0
CJ)
cn
3
O
0
H
e
3
4-1
H
pq
e
o
^_i
M-l
cn
ca
O
i i
0)
d
0
cd
01
i i
CJ)
11 1
M 1
O
0
0
1-1
4-1
O
d
o
j_^
PH
~
^
i 1
cfl
4J
0)
0
fa
fa
d
o
PH
O
co
0)
o
3
O
t/3
X*s
CO
1
OS
i I
^4
Q)
fS
g
0)
O
OJ
Q
S-^
o^
^d"
O
1
CO
r--
i
CM
^
O
in
\@
i
^
pLj
w
ietary
fa
ft
o
fa
ft
tn
4J
H
0
O
T3
QJ
tn
ca
X)
o
o
o
0
0
£_p
K*^
^
""O
OJ
JJ
J_(
o
ft
01
fa
0
o
H
4-J
cd
3
1 1
cd
[>
QJ
Q)
4J
id
0
o
XI
^4
cd
o
e
3
H
tn
tn
ca
4-1
'0
ft
4J fa
0 0
PC £
39
-------
system was based on literature data. This estimating group had access
to proprietary cost data but not to data on the selective removal of
HoS in the presence of CC^. With this limitation, the department is
considered competent to perform the cost estimate. However, a disinter-
ested evaluator would be preferred.
Table 12 gives an updated estimate of the Conoco dolomite system
with hot particulate and alkali compound removal. Comparable updates
of the other systems are not available. The later estimate gives an
optimistic by-product credit to sulfur.
The costs of make-up dolomite'in both of these tables is based on
17o loss per cycle, with the dolomite rate 9.2% greater than the theoret-
ical because of its inert content. In 1975, Conoco raised its estimate
of make-up to 270 to account for both attrition and activity loss. This
would increase the operating cost by 4.5%.
Table 13 summarizes economic evaluations of Environmental Science
and Engineering, Inc. (ES&E). Both the Conoco dolomite and hot carbonate
estimates are based on Conoco work for a 1,400 MW facility. The ES&E
estimates were made by adjusting equipment sizes to a 1,000 MW scale,
ES&E states that the coal gasification section is not included because
the costs would be identical for the two systems. This approach fails
to take into account the differing efficiencies of the two desulfuriza-
tion processes. The estimated annual operating cost is higher for the
Conoco system than for hot carbonate because of an increase in the esti-
mated attrition rate (said to be 5% per cycle) for which there is no
sound basis. The cost given for make-up ($1,783,000 per year) does not
appear to be consistent with the stated basis--that is, 5 times attrition
rate and 1.4 times capacity gives $2,316,000 per year. ES&E also esti-
mated the Conoco dolomite total operating cost using a 1% make-up rate
as $8.9 million per year (less than hot carbonate).
The design upon which the Stretford evaluation (third column of
Table 13) is based is characterized by ES&E as "speculative." Therefore,
the accuracy of the estimate is unknown.
Finally, the evaluation by United Technologies Research Center
(UTRC) and Foster Wheeler Energy Corporation is given in Table 14. The
evaluation is based on a Selexol (a commercial process) or Conoco dolomite
cleanup system downstream of a BCR two-stage entrained-flow slagging
gasifier. The evaluations correctly take account of the estimated cleanup
system efficiencies. However, the performances of the two systems are
not the same. The Selexol design includes a water washing step to re-
move ammonia from the fuel gas before it enters the Selexol unit.
40
-------
12
Table 12
CONOCO ECONOMIC EVALUATION UPDATE
REGENERATIVE ACCEPTOR DESULFURIZATION PROCESS
Base PointJuly 1975. Begin Operation--January 1980
Basis: 1,400 MW, 8,000 Btu per kwh, 70% Plant Operating
Factor (6,132 hours per year)
Method of gas desulfurization Hot
Method of particulate and alkali removal Hot
Coal required
Tons per year (6% moisture) 3,422,000
Higher heating value (Btu per year) 81.71 X 10
Desulfurized producer gas to station
Mols per hour 213, 701
Temperature 660° F
Pressure 10 psig
Higher heating value (Btu per year) 65.19 X 101S
HHV + sensible heat content (Btu per year)" 68.68 X 10ls
Cost analysis ($ millions)
Installed plant cost
1975 cost $47.5
Escalation to 1980 12.4
Interest during construction 11.5
Total investment (1980) $71.4
Working capital 2.38
Annual operating cost ($ millions)
Direct operating cost (1980 basis) $10.370
Acceptor at $15 per ton (tentative) 0.973
Interest on working capital at 8.5% 0.201
Capital charges at 18% investment 12.852
Sulfur credit at $25 per metric ton (2.977)
Net annual operating cost $21.419
Desulfurization cost expressed:
In terms of feed coal
Dollars per ton coal $ 6.26
Cents per million Btu(HHV) 26.2^
In terms of product gas to station
Cents per million Btu (HHV) 32.9"
Cents per million Btu (HHV + sensible
heat)" 31.2?
Excess power generated by expander, MW 160
Sensible heat content above an assumed air heater outlet temperature
of 300r'F.
Source: Conoco
41
-------
Table 13
ES&E ECONOMIC EVALUATIONS
Basis: 1,000 MW. Operating Factor Not Stated
Gas Desulfurization and Sulfur Recovery Costs Only
5% Attrition of Acceptor per Pass
Make-Up at $10 per Ton
(Millions of Dollars)
Conoco Dolomite Hot Carbonate Stretford
Total capital investment $31.6 $33.7 $45.7
Annual operating costs
Total direct $ 4.1 $ 3.1 $ 5.6
Total indirect 1.0 1.0 1.3
Costs of capital 5.3 5.7 7.6
Total annual cost $10.4 $ 9.8 $14.5
Source: H. S. Oglesby et al., "Final Report on Engineering Analysis,
Technical Feasibility, and Applicability of the Consol Process
for High Temperature Gas Clean-up," EPA Contract 68-02-1330
(August 20, 1974)
Anhydrous ammonia is recovered from the wash water. Ammonia production
may be economically justified, but the capital cost is somewhat higher
than is necessary for desulfurization and particulate removal. If the
water washing step were eliminated, NH3 would be absorbed along with
H2S in the Selexol system and might cause deactivation of the downstream
Glaus plant catalyst, depending on the NHo concentration. Thus, the
possibility of eliminating the ammonia removal step depends on the ratio
of NH_ to I^S in the gas, which in turn depends on the N and S contents
of the coal and the gasification conditions.
The Conoco dolomite estimate in Table 14 is based on a conceptual
design, and the Selexol is based on cost curves developed from published
data. Because UTRC plans to use higher pressures than Conoco in combined-
cycle power plants, the design conditions are outside the range of the
experimental data. Accuracy is estimated as plus or minus 2570. However,
42
-------
Table 14
INTEGRATED SYSTEM COST SUMMARY
Basis: 8,400 Tons per Day Illinois No. 6 Coal (~1, 000 MW)
70 Percent Annual Load Factor
1974 Costs
BCR Gasifier Conventional
Selexol Conoco Steam Plant
Capital costs ($/kW)
Power system cost 208 190 345
Gasification system cost 117 99
Cleanup system cost 89 35 81
Total plant cost 414 324 426
Owning plus operating costs (mils/kWh)
Owning costs (17% of capital) 11.47 8.97 11.8
Operation and maintenance
Power system 1.19 1.08 1.1
Gasification and cleanup 2.84 1.85 1.1*
Fuel cost at 60^/MM Btu 5.69 4.82 5.8
Total cost of power 21.19 16.72 19.8
*
Stack gas cleanup.
Source: United Technologies Research Center, East Hartford, Connecticut
operating cost of the gasification and cleanup systems are arbitrarily
taken as 8.5% of the capital cost per year.
Table 14 also gives an estimate of a conventional coal-fired steam
plant with stack gas cleanup. The cost is estimated to be higher than
for the advanced gasifier--Conoco dolomite--combined-cycle system.
43
-------
VIII FUNDING IMPLICATIONS
This section considers the funding implications of the Conoco-
proposed development. Table 15 gives a comparison of Hot Gas Cleanup
with the predecessor CC^ Acceptor process. The number of developmental
reactions is two for the C02 Acceptor and six, or three times as many,
for the Hot Gas Cleanup process. The sizes of reactors are shown, in-
cluding the pretreater, gasifier, burn-up cell, desulfurizer, regenerator,
and liquid-phase Glaus reactor. The proposed plant clearly exceeds the
existing plant in both size and complexity. One cost factor shown fav-
oring the proposed plant is that coal handling and preparation facilities
are excluded, since they are already built, besides dolomite handling
facilities, utilities, and general facilities. Also, the trained operat-
ing crew might reduce the time required for successful operation. The
design, construction, and operation time for the CC^ Acceptor plant has
been nine years to date, working with noncaking feedstocks. The corre-
sponding estimated time for the proposed Hot Gas Cleanup process is less
than four years, working with caking coals. This program includes only
moderately caking coals such as Illinois No. 6. If it is successful,
a proposal to extend the program to strongly caking coals such as Pitts-
burgh seam coal is likely. The program cost to date has been $36 mil-
lion for the CC>2 Acceptor, and the estimated cost for the proposed Hot
Gas Cleanup process is $18 million plus a management fee to be negotiated.
Based on this comparison, the extent of experimental work to date and
the inflation that has been experienced, SRI believes that the time and
cost estimates are optimistic.
Another qualitative method of analyzing funding implications is to
consider that there are actually four development processes in series.
These are gasification of caking coal, desulfurization of the gas, re-
moval of particulates and alkali compound vapors, and the liquid-phase
Glaus process. While not every step is essential to operation of the
pilot plant, it is likely that troubles with one step will disrupt
studies of another step. After nine years of work, the COo Acceptor
pilot plant runs are still limited to about three weeks. It is likely
45
-------
ft
i
cu
u
CO
cfl
O
in
X
O
4J
ft
Ol
o
o
Cs
o
o
o)
H
CM
CO
J^
O
4-1
O
cu
^
i i
nj
4-J
G
0)
e
d, en
O ^
.-1 O
CU 4-1
> o
0) nj
13 0)
j_i
t4-|
O 4-1
O
^
0) co
rQ O)
6 N
3 -H
S CO
o m
vO CO
X X
0) 0) 0) 01
G = G G = C
O O O O CO O
la * a ja co 125
co
3
rH
CJ
i-l )-i 01
t-( 0) M co
^i CU N O t8
0) O -H 4-1 J2
4-i !-i !-i nj CL,
cd cu ft 3 ^ '
Ol -H 3m 0) T)
)-l 4-1 | r-| G *H
U -H C 3 CU 3
0) co Jj co 60 tj*
!-i ca 3 CU Ol -H
PJ o PQ o oi hJ
T3
cu
*T3
3
i I
O
C
M
G"
0
H
|J
ctf
J_)
n)
D- co
01 01
M *r-J
a 4J
H
'O ' i
G -t-l
nj o
cB
60 U-l
C
H 1
i 1 (T)
13 H
C 01
nj G
JS CU
60
01
4J 13
H G
g ctf
Q
I 1 -,
O co
T3 01
rH
TJ 4J
C -H
cy i~~i
r-i
i 1 4J
c« 3
O
CJ
O)
4-)
nj
O
4J
J_l
t>^
0>
01
.5
4-1
C
o
H
4-1
IS
^1
cu
ft
o
*"0
3
^
G
0
H
4_)
O
3
4-1
CO
G
O
o
N
C
60
H
CO
0)
P
nj
T>
O
4J
s
^
vO
CO
4-1
cn
0
a
P<
nj
^
60
O
J_j
P-J
46
-------
that gasifier failures in the hot gas cleanup plant would disrupt desul-
furization tests. Delays are costly, especially when revisions to
large-scale equipment must be made.
Unfortunately, no reliable estimate of the actual cost can be made
by this approach. The nature of a development project precludes accurate
predictions of total costs to get a viable process.
47
-------
IX FUTURE HOT GAS CLEANUP DEVELOPMENT
This section discusses the types of information on hot gas cleanup
that would be desirable. Some of the information may be available from
sources outside the scope of this study, so the ideas are given as sug-
gestions rather than recommendations.
The laboratory work by Conoco has included cycling tests of dolomites
but not with make-up at steady state conditions. Projections of steady
state were made by mathematical analyses that appear sound. However,
we suggest that a run with make-up of at least a month would be justified
for a project of this magnitude. Conoco personnel made an offhand esti-
mate that such a run might cost $200,000, which would be no more than 170
of the cost of the pilot plant program.
A steady state run of at least a month is suggested for the liquid-
phase Glaus process because of the complex chemistry of aqueous sulfur
compound systems. Such a run would determine the build-up of contaminants,
corrosion, and possible operating problems. This should be done after
currently scheduled experiments on kinetics and process variables.
Because the Air Products work did not lead to a viable hot gas
cleanup process with limestone or dolomite, it would be appropriate to
make an engineering analysis to determine whether there are any implica-
tions to the Conoco process. The number of dolomites tested so far has
been limited; further work to determine the uniformity of deposits, the
suitability of other deposits, and the adequacy of suitable reserves
would be in order. Some such work is already scheduled and may be
adequate.
Process variables have been more thoroughly studied in the desul-
furization step than in the regeneration step. More regeneration runs
to determine the thermodynamic and kinetic effects of temperature, pres-
sure, and C02/H20 ratio appear desirable to aid in process optimization.
Conoco Coal Development Company is evidently not committed to any
particular particulate collection system for use with its desulfurization
system. However, it has carried on bench-scale studies of a system de-
signed to collect both the particulates and alkali vapors. The basic
process consists of cooling the gas to a temperature (1200 to 1300°F)
49
-------
at which essentially all the alkali vapor is expected to condense, It
is hoped that most of the condensation of alkali will take place on the
dust particles as nuclei. Thereafter, the relatively coarse dust par-
ticles are to be collected by cyclones or by a filter.
In tests made up through the end of August 1975, it appeared that
most of the added alkali was collected in the bench-scale system. How-
ever, it had not been possible to complete material balances to account
for all the alkali and demonstrate where it was actually deposited.
Hence, it is not known whether the system actually worked as conceived.
The experiment is a difficult one to carry out, particularly in such a
small system that has a high internal surface-to-volume ratio.
The use of dust particles to nucleate condensation of the alkali
vapors is probably best studied on a relatively basic level not immedi-
ately related to a gasification process. In such a basic study, the
experimental conditions would have to be very closely controlled and
monitored.
Bench scale work was done by Conoco at rates equivalent to about 6
to 8 pounds per hour of coal feed. The proposed pilot plant capacity is
about 2,400 pounds per hour, a scale-up of about 400. Fluidized-bed
reactors pose scale-up problems, and it is necessary to have fairly
large pilot plants for reliable scale-up to commercial dimensions. The
proposed pilot plant size is considered reasonable, although not neces-
sarily the minimum. If the chemistry and process requirements are ade-
quately defined, construction and operation of a facility intermediate
in size between the laboratory and pilot plant scale does not appear
justified. However, a fluidized-bed gasifier with a capacity of about
50 pounds per hour of coal is currently being built at Research Triangle
Park. This bench-scale pilot plant could perhaps be used to obtain data
to sharpen the pilot plant design. The gas production rate would be
appropriate for the scale we would suggest for studying hot removal of
particulates.
Table 16 summarizes some options open to the EPA in regard to the
Conoco proposal. The first is to proceed as proposed by Conoco if funds
are available. This would make use of a trained operating crew, which
is an advantage that should not be overlooked. It would give earliest
availability of the process unless such major revisions were required
that they overcame the advantage of starting earlier. However, overruns
are probable.
Another option is to modify the program. One of the first steps
could be to build and test a gasifier, presumably under the auspices of
50
-------
Table 16
OPTIONS FOR EPA
I. Proceed as Proposed by Conoco (if funds are available)
Use of trained operating crew
Earliest availability of process
But probable overruns
II. Modify Program
Build and test gasifier (ERDA)
Concurrent lab work:
- Steady state desulfurization
- Steady state liquid-phase Glaus reaction
- Testing of more dolomites
- Particulate removal (^15 acfm)
- Process variables
Concurrent design and evaluation
Desulfurization pilot plant
III. Carry out terminal program
Laboratory work to tie up "loose ends"
Analysis of results
Documentation
Engineering studies for guidance of possible future
work
ERDA. If continuing tests of the CO Acceptor process could be coordi-
nated, this would make good use of the trained operating crew. Concur-
rently, laboratory work could be done. Besides the scheduled work, we
would suggest steady state tests of at least a month to determine any
long-term effects in desulfurization and liquid-phase Glaus sulfur re-
covery. SRI suggests testing other dolomites to determine the suitability
of other deposits and whether there is variability within a deposit.
51
-------
Particulate and alkali vapor removal could be studied profitably at a
scale of not more than 15 actual cubic feet per minute of gas; some such
work is scheduled. Process variables should be further studied to put
designs on a firmer basis. To increase reliability of the data, material
balances should be closed.
Design and evaluation work could profitably proceed concurrently
with the laboratory work to help guide the laboratory work. This work
might determine at an early date whether any new findings would invali-
date the whole concept or whether any reduction in size of the pilot
plant might lead to the same information at lower cost. This work might
expedite the building of the pilot plant when the time comes. This
approach is common in the process industries. The final step in the
modified program would be construction and operation of the desulfuriza-
tion pilot plant and necessary auxiliaries.
A third option is to abandon efforts intended to lead to large scale
development for the present but to provide for orderly termination to
preserve the value of the work already completed. This approach should
entail completion of work for which the added cost would be small com-
pared to the information gained, analysis of the results, process, evalua-
tion and documentation in a final report. The report should give recom-
mendations for further work in the event that the program should be
reactivated.
52
-------
Appendix
FLOW DIAGRAM FOR PROPOSED PILOT PLANT
53
-------
CN
s
O
z
cc
Q
O
o
O
z
O
0
O
I-
cc
O
I
cc
D
O
55
-------
O
O
Lt
Z
O
I-
H
z
LU
Q
CE
I-
(0
LO
2
CO
CN
^~
o
en
CO
r-
CO
»
**
CO
*
o
to
CO
§o
i » S
J eji
8 G (r
t IS 2 §
Is Q
CT ° -
0 i u
0) CO w
b x" o
1 " S *
LL. U O
*- O
8 S.
5? S
CC o
3 °
1 a
5 <
i
1} c ._
S ° S
81 '= O
cc
o
a o S
0) £ Ol
CJ Q^
O ^
Q. £ Qj
U *- LT
C/l "o "*
J 1 S 2
e §
^ D ^
O m
E -1 =
S o "= .5
2 " ra 3
C/5 CJ o
0 J c
<- - 0
< o 3
o
T5 "° TN
S g I
0 u. 5?
c
o
a
u
Q
01 in
r- CN
O O
LO O)
O CM
*- O>
CO -
O O
CN »-
CN CN
0 O
CO *-
CO CN
82
gs
O CO
O *fr
CN CN
o o
LO CN
0) S
s 5 £
2 « -
Q- « i-
. a ^
o § 8
& t- D.
Q.
<
O CN
CO (D *-
CO LO CN
CM
ro
r- CD .
LO CO 1
CD
CN O
O CN LO
O5 Ol CO
CO
CN
CN *fr
CO 00 i
LO '
CD LO
CO LO CN
i- CO CO
O
CO
« ' '
(0 ' I
01 t
o r^ co
SCO CO
CO
i i i
CN
O)
B i i
CO
tf Tj- tO
^! LO (O
CN LO LO
% CN LO
H (p r*-
Oi i |
CD cn
to co r-
CN r- CN
CO
»- CN
^t ^ O
Ol CO O
LO CN *
r^-
o
§ 1 !
e
CD
-
i 1 1
CO o
cc S §
^ 5 *~"
LL "D T3 LL
C C (j
X
in LO o
i i i £ i i ° i i ^
V CO
o 01
111 1 , | 1 | 01
CO CO «t
O) O O
1 1 1 w 1 1 II*
LO LO
CD CO
<- CO
o co
, , CO O O
1 1 | I r- | i CN |
3?
03
O
C
g
O ^ CM COc/5 CMO
O I (MO O Ml CM CNQ CM
cjioozziOc/jn:
ro
O
O
8
0
0
o
o
o
o
o
0
o
o
0
8
o
o
o
o
8
o
8
o
o
o
"
"to
0
1-
CN
S
CO
I
LL
X
I
cc
Q
O
O
o
z
o
o
o
H-
CC
O
LU
O
<
_J
<
DC
uu
CM
<
56
------- |