EPA-650/2-73-049
December 1973
Environmental Protection Technology Series
-------
EPA-650/2-73-049
PRODUCTION OF CLEAN FUEL GAS
FROM BITUMINOUS COAL
by
G. P. Cur ran, J. T. Clancey, B. Pasek,
M. Pell, G. D. Rutledge, and E. Gorin
Consolidation Coal Co. Inc.
Research Division
Library, Pennsylvania 15129
Contract No. EHSD 71-15
ROAPNo. 21ADD-22
Program Element No. 1AB013
EPA Project Officer: D. B. Henschel
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
December 1973
-------
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.
11
-------
ABSTRACT
A process for the production of low-Btu gas from bituminous coals
via fluid bed gasification is described. Coal processing consists
of pretreatment, gasification, and final burnup. Hot fuel gas is
desulfurized with half-calcined dolomite and cleaned of particulates
in high-pressure drop cyclones. The sulfur acceptor is regenerated
with steam and C02. A liquid-phase Claus reactor is used to process
the H2S in the regenerator offgas into elemental sulfur.
Experimental data are presented which demonstrated feasibility of
the major process steps: pretreatment, gasification, carbon burnup,
desulfurization, regeneration, and liquid-phase Claus processing.
In all cases, however, more experimental data are required to
optimize conditions.
An economic evaluation of gas clean-up operations shows that regen-
erative use of acceptor is preferable to once-through, and that
removal of particulates via cyclones, if feasible, is cheaper than
water scrubbing with subsequent reheat. The cost of the gas desulfu-
rization process including sorbent regeneration and sulfur recovery
is of the order of 20^/MM Btu of product gas.
This report .was submitted in partial fulfillment of Contract No.
EHSD 71-15 under the sponsorship of the Demonstration Projects
Branch, Control Systems Laboratory, Office of Research and
Development of the Environmental Protection Agency. The work was
done during the period March, 1972 to June, 1973.
111
-------
IV
-------
TABLE OF CONTENTS
I. SUMMARY 1
II. CONCLUSIONS AND RECOMMENDATIONS , 5
III. FEASIBILITY STUDY 9
A. Introduction 9
B. Overall Process Description 9
Figure 1 - Two Stage Fluidized Bed Partial Combustion
Process 10
Table 1 - Range of Variables Studies in System Analysis 13
Table 2 - Key Stream Flows and Analyses 14
C. Summary - Gas Purification Costs 15
Table 3 - Summarized Economic Evaluation - Regenerative
Acceptor Desulfurization Process 17
D. Process Description - Gas Purification 18
Figure 2 - Block Flow Diagram - Overall System 19
Figure 3 - Alternative Gas Clean-up System for Particulates
Section 6OO 20
Figure 4 - Process Flow Diagram - Low Sulfur Boiler Fuel
Regenerative Acceptor Case, Dwg. AF-3424 22
E. Design Basis 24
F. Material and Heat Balances 27
G. Economic Evaluation 28
Table 4 - Mass and Heat Balance - Gas Desulfurizer 29
Table 5 - Mass and Heat Balance - Acceptor Regenerator 30
Table 6 - Mass and Heat Balance - Acceptor Stripping Column 31
Table 7 - Mass and Heat Balance - Liquid-Phase Claus Reactor 32
Table 8 - Mass and Heat Balance - Sulfur Combustor 33
Table 9 - Mass and Heat Balance - S02 Absorption Tower 34
Table 10 - Overall Mass and Heat Balance 35
Table 11 - Investment Summary 38
Table 12 - Direct Operating Cost Summary Excluding Acceptor
Cost 39
Table 13 - Economic Evaluation - Regenerative Acceptor
Desulfurization Process 40
Table 14 - Acceptor Requirements - Regenerative vs Once-Through 41
Table 15 - Approximate Investment Reduction - Once-Through
Acceptor vs Regenerative 43
IV. EQUIPMENT AND PROCEDURE 45
A. Description of the Continuous Unit 45
Figure 5 - Flow Diagram - Continuous Unit, Acceptor
Reaction Loop, Dwg. CF-33O3 52
Figure 6 - P&I Flow Diagram - Continuous Unit, Acceptor
Reaction Loop, Dwg. AF-3411 53
v.
-------
TABLE OF CONTENTS - Cont'd.
Page
Figure 7 - D-2 Gas Desulfurizer Assembly,, Low Sulfur Boiler
Fuel Unit, Dwg. AV-3306 54
Figure 8 - D-l Regenerator Assembly, Low Sulfur Boiler Fuel
Unit, Dwg. AV-3294 55
Figure 9 - B-l & B-2 Saturator - Continuous Unit, Dwg. BV-3136 56
Figure 10 - C-l, C-4, C-6, C-7 Coolers, Continuous Unit,
Dwg. BV-3426 57
Figure 11 - F-4 & F-7 Condensate Receivers, Continuous Unit,
Dwg. CV-3425 58
Figure 12 - Flow Diagram - H2S Supply System, Dwg. CF-3421 59
B. Operating Procedure for a Typical Two-Vessel Integrated Run 60
C. Equipment and Procedure for a Single-Vessel Operation 61
Figure 13 - Flow Diagram for Coal Pretreatment 62
D. Liquid-Phase Claus Reaction Equipment and Procedure 64
Figure 14 - Schematic Outline of Reactors for Vessel D-l,
Dwg. CV-3420 65
Figure 15 - Configuration of Draft Tubes Used in Draft Tube
and Seeded Coal Tests 66
Figure 16 - Configuration of Draft Tube Used in Seeded
Coal Tests 67
Figure 17 - Flow Diagram - Demonstration Runs, Dwg. CF-3418 68
Figure 18 - Flow Diagram - Char Treatment and Gasification
Kinetics Runs, Dwg. CF-3417 69
Figure 19 - Flow Diagram - Carbon Burn-Up Cell Runs,
Dwg. CF-3416 70
Figure 20 - Carbon Burn-Up Cell Reactor, Dwg. CV-3419 71
Figure 21 - Flow Diagram - Lab Scale Liquid-Phase Claus
Reactor, Dwg. CF-3314 72
Figure 22 - Liquid-Phase Claus System Piped for Batch
Liquid Operation 74
MATERIALS: SOURCE, PREPARATION AND ASSAYS 75
A. Inert Char 75
B. Ireland Mine Coal 75
C. Loveridge Char 75
Figure 23 - Preparation of Inert Char Feeds 76
Table 16 - Properties of Precarbonized Char - 28 x 100
Mesh Fraction 77
Table 17 - Sieve Analysis of -100 Mesh Precarbonized Char 77
Figure 24 - Preparation of Feed Coal 78
Table 18 - Properties of Feed Ireland Mine Coal 79
Table 19 - Sieve Analysis of 100 x 200 Mesh Ireland Mine Coal 79
Table 20 - Char from Loveridge Mine Coal Treatment 8O
vi.
-------
TABLE OF CONTENTS - Cont'd.
Page
Figure 25 - Pretreatment of Loveridge Mine Char 81
Table 21 - Properties of Thermally Treated Loveridge Char 82
D. Acceptors 83
E. Acceptor Assays 83
Table 22 - Results of Raw Stone Assays 84
Table 23 - Compositions of Feed Acceptors 85
Figure 26 - Acceptor Sizing 86
VI. SAMPLE CALCULATIONS 91
A. Acceptor Sulfur Cycle Studies 91
Table 24 - Computation of Number of Cycles - Run A7 93
B. Carbon Burn-up Reactor Calculations - Run CBC-8, Phase 2 98
Figure 27 - Temperature Profile - Run CBC-8, Phase 2 99
C. Liquid-Phase Claus Reactor - Run C6 100
VII. TABULAR CHRONOLOGICAL HISTORY OF RUNS 103
Table 25 - Summary of Seeded Coal Tests 104
-•'' Table 26 - Summary of Preoxidizer Operations 105
Table 27 - Summary of Sulfur Acceptor Runs 106
Table 28 - Summary of Gasifier Demonstration Runs 108
Table 29 - Summary of Gasification Kinetics Runs 109
Table 30 - Summary of Liquid-Phase Claus Reaction Operations 110
Table 31 - Summary of Carbon Burnup Cell Operations 111
VIII. PRETREATMENT STUDIES 113
A. Introduction 113
Figure 28 - Preoxidized Coal Via Draft Tube 115
Figure 29 - Seeded Coal Process 116
B. Draft Tube Studies 117
Figure 30 - Configuration of Draft Tubes Used in Draft
Tube and Seeded Coal Tests 118
Figure 31 - Configuration of Draft Tube Used in Seeded
Coal Tests 119
C. Pretreatment Using the Seeded Coal Process 120
Table 32 - Summary of Conditions and Results for Seeded Coal
Tests - Ireland Mine Coal (lOO x 20O Mesh) 121
vii.
-------
TABLE OF CONTENTS - Cont'd.
Figure 32 - $ Preoxidation vs Temperature for Adiabatic
Constraint 123
D. Preoxidation of Ireland Mine Coal 125
E. Conclusions - Pretreatment of Ireland Mine Coal 126
Table 33 - Preoxidation Conditions and Results 127
Table 34 - Properties of Feed Coal and Products 128
Table 35 - Material Balances for Preoxidation 129
Table 36 - Distribution of Oxygen in Products of
Preoxidation 129
IX. ACCEPTOR SULFUR CYCLE STUDIES 131
A. Introduction 131
B. Cyclic Runs with Tymochtee Dolomite Feed 131
Table 37 - Conditions and Results for Gas Desulfurizer with
Tymochtee Dolomite Feed 132
Table 38 - Conditions and Results for Regenerator with
Tymochtee Dolomite Feed 133
Table 39 - H2S Content of Exit Gases (dry basis) 134
Table 40 - CaS Content of Exit Solids 135
Table 41 - Size Consist of Acceptor Feeds and Products 137
Figure 33 - Runs Al and A2, CaS Content of Exit Solids 139
Figure 34 - Runs A1-A4, H2S Concentration in Reactor Offgas 140
Figure 35 - Run A4 Molar Sulfur Content of Exit Solids 141
C. Dolomite from C02 Acceptor Process 143
D. Runs with Canaan Dolomite Feed 143
Figure 36 - Run A18 Mol % CaS in Exit Solids 144
Table 42 - Conditions and Results for Gas Desulfurizer with
Dolomite Feed from C02 Acceptor Process 145
Table 43 - Conditions and Results for Regenerator with
Dolomite Feed .from C02 Acceptor Process 146
Figure 37 - Run A6 - CaS Content of Acceptor 147
Figure 38 - Run A6 - H2S Concentration in Reactor Offgas 148
Figure 39 - Run A7 - H2S Concentration in Reactor Offgas 150
Figure 40 - Run A7 - CaS Content of Acceptor 151
Table 44 - Conditions and Results for Gas Desulfurizer with
Canaan Dolomite Feed 152
Table 45 - Conditions and Results for Regenerator with
Canaan Dolomite Feed 153
Figure 41 - Run A15 - Mol $ H2S in Regenerator Exit Gas 154
Figure 42 - Run A15 - Mol $ CaS in Exit Solids 155
Figure 43 - Run A16 - Mol $ CaS in Exit Solids 156
Figure 44 - Run A16 - Mol % H S in Exit Solids 157
vin.
-------
TABLE OF CONTENTS - Cont'd.
Page
Table 46 - Density of Sulfided Canaan Dolomite 159
E. Runs with Pennsylvania and Virginia Dolomites 160
F. Runs with Limestone Feeds 160
Table 47 - Conditions and Results for Gas Desulfurizer with
Pennsylvania and Virginia Dolomite Feeds 161
Table 48 - Conditions and Results for Regenerator with
Pennsylvania and Virginia Dolomite Feeds 162
Figure 45 - Run A12 - Mol % CaS in Exit Solids 163
Table 49 - Conditions and Results for Gas Desulfurizer with
Limestone Feeds 164
Table 50 - Conditions and Results for Regenerator with
Limestone Feeds 165
G. Once-Through Operation 166
Figure 46 - Run A13 - CaS Content of Exit Solids 167
Table 51 - Conditions and Results for Gas Desulfurizer -
Once-Through Operation 168
H. "Refractory" CaS - Runs A16 and A17 169
J. Accuracy of H2S Feed Rate - Sulfur Balance 169
K. Corrosion 170
L. Conclusions and Summary of Major Results 170
Table 52 - Change in "Refractory" CaS when Held at
Reaction Conditions 171
Table 53 - Attrition Rate of Acceptors 171
M. Future Work 172
X. CARBON BURN-UP CELL AND GASIFIER DEMONSTRATION RUNS 175
A. Carbon Burn-Up Cell Study 175
Figure 47 - Flow Diagram - Carbon Burn-up Cell Runs,
Dwg. CF-3416 176
Table 54 - Conditions and Results for Carbon Burn-up Cell Runs 177
Table 55 - Compositions of Char Feedstocks 178
B. Gasifier Demonstration Runs 179
Table 56 - Ash Properties 180
Table 57 - Conditions and Results for Demonstration Runs 182
Table 58 - Analyses of Char Feed and Products for
Demonstration Runs 183
IX.
-------
TABLE OF'CONTENTS - Cont'd.
XI. LIQUID-PHASE CLAUS PROCESS
A. Introduction 185
B. Once-Through Runs 185
Table 59 - Results and Conditions for Liquid-Phase Claus
Reaction 186
Table 60 - Sulfur Balances for Liquid-Phase Claus Reaction 187
Table 61 - Yields of Elemental Sulfur 188
Table 62 - Analysis of Claus Product Liquor 190
C. Recycle of Product Liquor 191
D. Batch Liquid Operations 192
Figure 48 - Runs C6-C8 - Sulfur Yields Over Time 193
Figure 49 - Runs C6-C8 - Conversion vs Fraction of Column
Containing Packing 194
E. Reaction Products - Effect on Rate of Sulfur Production 195
Figure 50 - Runs C9-C11 - Conversion vs Inlet H2S Concentration 196
F. Conclusions 197
XII. GASIFICATION KINETICS 199
A. Introduction 199
B. Method 199
C. Feedstock Preparation 199
D. Data Workup 199
E. Results 200
F. Correlation of Rates 200
G. Conclusions 200
Table 63 - Feedstock Compositions for Kinetics Study 201
Table 64 - Properties of Products from Gasification
Kinetics Runs 202
Table 65 - Conditions and Results for 1620°F Runs 203
Table 66 - Conditions and Results for 1700°F Runs 204
Table 67 - Kinetics Summary - System Pressure 15 atm. 205
XIII. BIBLIOGRAPHY 207
x.
-------
TABLE OF CONTENTS - Cont'd.
Page
XIV. APPENDICES
Appendix A - Detailed Investment Costs 209
Table A~l - Cost Estimate: Section 300, Sulfur Removal 210
Table A-2 - Cost Estimate: Section 400,, Sulfur Recovery 211
Table A-3 - Cost Estimates: Section 600, Wet Gas
Scrubbing and Reheat 212
Appendix B - Literature Survey of Liquid-Phase Claus Reaction 213
Appendix C - Thermodynamic Data 217
Table C-l - Effect of Temperature and Sulfur Partial
Pressure on Distribution of Sulfur Species 219
Table C-2 - Equilibrium Constants for Gas Reactions 220
Table C-3 - Numerical Values of Equilibrium Constants
for Table C-2 221
Table C-4 - Heat Capacities at Zero Pressure 222
Table C-5 - Heat Capacities Above 60°F., Gases 223
Table C-6 - Heats of Formation at 25°C, Gases 224
Table C-7 - Equilibrium Constants for Solids Reactions 225
Table C-8 - Mean Heat Capacities Above 60°F, Solids 226
Table C-9 - Heats of Formation at 25°C., Solids 227
Figure C-l - Sulfur Vapor Pressure 228
Figure C-2 - Equilibria for CaS Reactions 229
Figure C-3 - Equilibria for CaS Reactions 230
Appendix D - Modified Sulfur Recovery System 231
Table D-l - Thermodynamic Equilibria in the CaS-CaS04-
CaC03-C02-H20-H2S-S02-S2 System 233
Figure D-l - Modified Sulfur Recovery Section 234
Table D-2 - Equilibria in CaS04 Oxidation of Squires
Product Gas 236
Table D-3 - Composition of Feed and Product 236
Appendix E - Conversion Factors - English to Metric Units 237
xi.
-------
I . SUMMARY
A. Process Description
The process. investigated is illustrated in Figure 1. The basic flow sheet
represents a revision of the original process investigated in the first year
of the EPA Contract EHSD 71-15 as reported in detail in the Annual Report^1)
The revised process separates the sulfur removal from the gasification steps
of the process. Hot sulfur removal is effected in a fluid bed by means of
half-calcined dolomite utilizing the reaction,
H2S + CaC03 = CaS H- H20 + C02
at ca. 1650°F.
The gasification section now no longer utilizes C02 acceptor action .as a
source of heat. Efficient carbon utilization is now effected by use of a two-
stage fluid bed, air-blown gas producer wherein the second stage performs the
function of a carbon burn-up cell.
B. Pretreatment
Two aspects of pretreatment for highly fluid Pittsburgh Seam coals were studied
experimentally. The first was an extension of previous work using preoxidation
to render these coals noncaking in the process. The prior work had demonstrated
that an excessive level of preoxidation compared to the adiabatic level was
required. The use of two stages of preoxidation, with a rising temperature
between stages, and of finer size consist of coal was investigated. Both of
these expedients favor reduction of the level of preoxidation required. The
combination of both did not, however, permit reduction of the preoxidation to
the adiabatic level.
The second phase of the pretreatment work explored the concept of the Seeded
Coal Process wherein the fluidity of the coal is utilized to produce a dense
noncaking gasifier feed. Coal and seed char would be fed into a draft tube in
a fluidized bed at 1000- 1400°F. The coal melts, smears out over the surface
of the char and then solidifies on the completion of pyrolysis. Initial
results were encouraging, but larger equipment will be needed to develop the
process further.
C . Acceptor Sulfur Cycle Studies
A number of acceptors were studied. A pure Nebraska limestone showed little
activity with excellent attrition resistance. An impure limestone, 1691,
gave intermediate activity and good attrition resistance, but tended to form
agglomerates. All four dolomites showed good activity toward H2S. H2S con-
centration exiting the desulfurizer was usually less than 5% of the inlet and
about twice the equilibrium concentration with only an 18-inch bed.
However, of the four dolomites, three showed excessive rates of attrition,
5-10$ of the feed per cycle. The fourth dolomite, Canaan stone, showed good
activity with an attrition rate of less than 1$ per pass.
-------
All stones gave incomplete regeneration, the level of CaS in the acceptor in-
creasing as it was cycled. In the best run, 13$ of the total calcium was being
converted per pass through the regenerator from CaS to CaC03. The longest run
was for 17 cycles at about 8% conversion per cycle, and there was little decline
of activity between the eighth and last cycle.
Tests made with once-through operation showed that the dolomite could achieve a
calcium utilization of about 90% with good removal of H2S in this mode of
operation.
D. Carbon Burn-up Cell and Gasifier Demonstration Runs
Several runs explored the effects of temperature and bed expansion on carbon
burnout and ash slagging in the carbon burn-up cell. Total carbon utilization
was obtained at temperatures as low as 1700°F. At 1900°F, the reactor was run
with no evidence of ash slagging at bed expansions of about 60$ above the
incipiently fluidized bed height.
Operation of the gasifier at process flows and 1725°F was demonstrated with
freedom from ash slagging. However, at 1775°F, some mild slagging of the ash
occurred. Possibly, operation of the gasifier at a higher bed expansion would
lead to successful operation at higher temperatures.
E. Gasification Kinetics
A total of 24 runs were made with char from Pittsburgh Seam coal at 30$ and 6O$
carbon burnoff and at 1620 and 1700°F.
The original process estimates assumed a gasification rate for bituminous coal
char equal to 1/15 the rate predicted from our earlier studies on kinetics of
lignite char gasification. The current study shows that the assumed rates are
too high, mainly because inhibition of the reactions by H2, relative to inhibi-
tion by CO, is greater for the bituminous coal char.
The quantitative effect on the process has not been evaluated. Qualitatively
more steam will be needed in the gasifier' and the thermal efficiency will
decrease somewhat.
F. Liquid-Phase Glaus Process
The liquid-phase Claus reaction, using only water as the liquid, was found to
proceed readily under process conditions. The product liquor contained 0.1$ or
less soluble sulfur. Continued recycle caused no increase in soluble sulfur or
total acidity.
The scope of the effort was insufficient to fully elucidate the behavior of the
system with respect to reactor volume and gas concentration. Preliminary indi-
cations are that the system is between first and second order with respect to
both.
Conversion of over 90$ of the feed H2S-S02 was obtained using a reactor volume
consistent with process estimates.
- 2 -
-------
G. Feasibility Study
The economic feasibility study was limited to consideration of high temperature
(l650°F) desulfurization of pressurized producer gas as fed to a power station
using half-calcined dolomite as the sorbent. Two methods were compared for
cleanup of particulates and alkali, i.e., hot cyclones only and wet scrubbing
with gas reheat. The economic comparison is summarized in Table 3 in Section
III of the report dealing with the feasibility study.
The hot cyclone method of gas cleanup gave an advantage equivalent to about
one dollar per ton of feed coal. The investment cost of the plant facilities
for desulfurization of the producer gas, regeneration of the sorbent and
recovery of elemental sulfur was $32.00 per installed KW. The processing cost
was estimated as 20^/MM Btu of product gas. The basis for the economic evalu-
ation was 1978 operation at 70$ load factor; 7.5$ escalation, interest during
construction, and interest on working capital; and 18$ capital charge.
The "once-through" case wherein the sorbent is not regenerated was also
examined. This alternative was deemed uneconomical because of the high cost
of make-up stone and for treatment and disposal of the spent stone.
- 3 -
-------
II. CONCLUSIONS AND RECOMMENDATIONS
A. Overall Process
The last Annual Report'1/ examined the low Btu gasification process integrated
with advanced power cycles and concluded that the process had economic merit.
This report deals with a revised gasification system (cf. Figure l) using a
separate hot removal of H2S from the producer gas with half-calcined dolomite
(i.e., MgO-CaC03). The economic study given in this report is, however, limited
to consideration of the gas desulfurization and clean-up system. The previous
conclusion,!1) that after hot desulfurization, the hot removal of particu-
lates and alkali is better than wet particulate scrubbing with subsequent gas
reheat within the context of the revised process, was confirmed. In addition,
the regenerative process was shown to be cheaper than once-through processing.
Further economic studies should be carried out, however, to determine whether
the hot gas desulfurization process is also cheaper than conventional wet H2S
removal.
It is recommended that low Btu gas process research and development be continued.
The cleanup of the fuel gas should use a regenerative process for desulfuriza-
tion with a hot system for removal of particulates and alkali. The next phase
of the work should concentrate on development of the hot gas clean-up system
independently of the specific gasification process used. The specific gasifi-
cation system described in this report, however, also has merit and considera-
tion should also be given to its further development. Recommendations follow
with respect to further work needed on the specific process steps of the
gasification and sulfur removal and recovery system.
B. Pretreatment
The pretreatment technique with the most potential is the Seeded Coal Process
which produces a dense, operable gasifier feedstock. Further work should be
carried out in a larger processing unit, so that the particle Reynolds numbers
will be well into the turbulent regime thereby giving good mixing of feed coal
and char. A unit processing about 300 Ib/hr is envisaged.
Earlier work indicated that preoxidation was a viable (within the constraints
of adiabatic operation) pretreatment technique for Illinois No. 6 coal but not
for Pittsburgh Seam coals. Further work remains to be done to determine how
much further reduction in required preoxidation for Pittsburgh Seam coals can
be effected by optimization of the process. An increasing number of stages of
preoxidation, i.e., from two to three, and the use of an optimum temperature
for each stage (rising temperature regime) can significantly reduce the level
of preoxidation. Similarly, the effect of the use of a higher partial pressure
of oxygen approaching the design condition of 3 atmospheres inlet partial
pressure in reducing the level of preoxidation should be investigated.
C. Gasification and Carbon Burnup
Demonstration of operability of these reactors at process conditions was carried
out. Further work on operability characteristics and reaction rates in the
above vessels should be carried out. It would be particularly desirable from
the kinetic point of view to determine operating conditions in the gasifier
- 5 -
-------
under which the presently proven operable temperature could be increased signi-
ficantly above 1725°F. Additional gasification kinetics data and associated
correlation work would be desirable to permit more reliable design of the
gasification system.
D. Acceptor Sulfur Cycle
Additional data on the acceptor sulfur cycle are still needed to more adequately
define acceptor make-up requirements. The acceptor life as a function of condi-
tions in each reaction vessel, gas desulfurizer and regenerator, remain to be
defined. Approach to equilibrium in the gas desulfurizer with acceptor that has
undergone numerous cycles has yet to be demonstrated. In the current work, the
later cycles were conducted with H2S breakthrough in the gas desulfurizer. This
is not necessarily a process restriction but may be attributed to the fact that
the circulation rate was not high enough to supply enough CaC03 to the gas
desulfurizer for efficient H2S absorption.
The prospects for finding widely distributed acceptors with good activity and
attrition resistance remain to be examined. A program for hardening even soft
dolomites should be initiated. Fortunately, as discussed in the feasibility
study, acceptor costs are small for the regenerative case, and fairly high ship-
ping costs can be absorbed.
E. Hot Particulate Removal
Further work remains to be done to demonstrate hot removal of particulates and
alkalis. An "optimum" system would be one wherein both H2S and particulates are
simultaneously removed. Such a study would have to be carried out on larger
scale equipment than used in the present work.
F. Liquid-Phase Claus Reaction
Feasibility of the process step was proven during the current contract period.
Further work is needed to define the optimum system with respect to reactor
volume, gas concentration, temperature and mixing or agitation.
In addition, there are rich possibilities for improvements via the addition of
catalysts and buffers. The area of nonaqueous solvents also remains to be
explored. While the process is workable as is, considerable improvements may
be attainable.
G. Chance Reaction - Disposal of Spent Acceptor
Spent acceptor contains MgO-CaS which cannot be disposed of without further
treatment to change CaS back to CaC03. The Chance reaction was carried out
commercially in the 19th century to dispose of the same nuisance CaS. While
data are expected to accrue from the Chance reactor installed at Consol's Rapid
City gasification plant, research will be needed to define the best mode of
operation for the low Btu gasification process. In particular, the fate of the
magnesium and the reactivity of unregenerable CaS in an aqueous medium should
be studied.
- 6 -
-------
H. Pilot Plant Design
The speed with which this process will be commercialized is a function of the
dollar input to its development. Sufficient data are on hand to design the
front end of a pilot plant comprising preoxidative pretreatment, a gasifier and
a burn-up cell. Presumably, sufficient data will be on hand within a year for
design of the desulfurization and sulfur recovery system.
An alternate scheme would be to adapt the hot desulfurization process described
herein to an existing gasification unit such as one using the Lurgi process.
- 7 -
-------
III. FEASIBILITY STUDY
A. Introduction
The production of low-sulfur producer gas via an adaptation of the C02 Acceptor
Process formed the subject matter that was studied under the first year of the
EPA Contract EHSD 71-15.
A detailed process design and feasibility study of the system as well as an
experimental evaluation was carried out. The results are reported in the
Annual Report . ( 1 )
The experimental evaluation of the system indicated feasibility of all steps in
the process with one exception. It was found that the sulfur recovery from the
acceptor would be incomplete from the regenerator in the Case I scheme of that
report. This necessitated addition of another step in the process in which
sulfur was rejected by the reaction,
CaS + H20 + C02 = CaC03 + H2S,
as shown for Case II in the Annual Report. I1) With this added complication
introduced, further thought was given to refining and simplifying the overall
process .
It became apparent that there is no real advantage in using the C02 acceptor
reaction simultaneously with the sulfur acceptor reaction in the gasifier when
low-Btu fuel gas is the desired product. The gasification system was revised
to provide for disposition of the residual char by use of a carbon burn-up cell
which preheats all the steam and air required for the gasifier. The sensible
heat duty involved in preheating serves as the "heat sink" which is necessary
to prevent ash slagging during combustion of the residual gasifier char. In
the C02 Acceptor Process the heat sink is provided by the endothermic calcining
reaction in the acceptor regenerator.
Since the acceptor no longer needs to be in the gasifier, desulfurization is now
carried out as a separate operation. The sulfur acceptance reaction is conducted
in a dense-phase fluidized bed of dolomite to which the acceptor is fed in the
form of MgO-CaC03. The feasibility study presented here provides the basic
process design parameters for both the gasification and sulfur removal systems.
The economics, however, are limited to the sulfur removal and recovery systems.
B. Overall Process Description
A schematic diagram of the revised process is shown in Figure 1. The hot fuel
gas is desulfurized in the H2S sorption bed by the reaction,
HS + CaCO = CaS + H20 + C02 .
The bed temperature is held at a level below which the acceptor can calcine by
the reaction,
CaC03 = CaO + C02 .
The low-sulfur hot gas is cooled to 1300°F by heat exchange with the water needed
to generate the gasifier steam, and then is cleaned of particulates and alkali
by high pressure drop cyclones.
- 9 -
-------
COAL
o
I
PRE
Figure I
TWO STAGE FLUIDIZED BED
PARTIAL COMBUSTION PROCESS
•REA
1
\
\
ER
v
\
s
f
(5) Product Gas
x —
\
^
With H2S
~^N
Y
CH
AR
1750° F
\
_r
it-
1 (
\/
s
I800°F
Y
|
ASH
BURNUP
CELL ^
GASIFIER
\sChar
I
Steam
D
^
'
•^
1
•x
I652PF
r
i
y
®y
//
/
m
GAS
X
*» 4Mb -* *** "*"
I300°F
V/
DESULFUR-
IZER
D
• CaS-MgO
._ CLEAN
(|l) PRODUCT
1 GAS
/Tj\ f| I3OO°F
-*-* t
REGEr
C(
^2
H20
LJ
";
s. co2
t
5 }jo)
o
?0
x — '
^r L« H20
STEAM
ylERATOR
S02
H20
VENT
AGAS
.Ij LIQUID rS<5n
r~ n^^^j t^ ^^ ^^^.^^^^^^
CLAUS «oounr i iv 'iv
COLUMN COLUMN
H20
H20 k_
x-^ V,U2
__ 1L Steam vA/ .
_, AIR CO
(
D
, t
CHLFUR
i BURNER
SULFUR AIR
-------
The sulfided acceptor is conveyed to the regenerator by continuously recircula-
ting a stream of C02 and steam. In the regenerator the reverse of the H2S
sorption reaction takes place at 1300°F.
CaS + C02 + H20 = CaC03 + H2S.
The regenerated acceptor is returned by gravity to the sorption reactor.
A computer program has been devised to evaluate the heat and material balance
relationships and overall thermal efficiency of the scheme as shown in Figure 1,
and accordingly, includes both the gasification and gas desulfurization sections
of the process.
The program evaluates the interaction among the various components of the system
consistent with the thermodynamic, fluo-solids mechanics and kinetic restraints
on the system.
The entire system was represented by 27 simultaneous linear and non-linear
equations which represent the five basic process steps given below:
1. Carbon burn-up cell.
2. Gasifier.
3. Gas desulfurizer.
4. Steam-product gas exchanger.
5. Regeneration reaction - impact of temperature only.
which also are interrelated by the following quantities:
1. H, C, 0 balance.
2. Heat balances around components 1 through 4, above.
3. Water-gas-shift equilibrium in components 2 and 3,
above.
4. Methane yield correlation.
5. Equilibrium in the reaction,
CaC03 + H2S = CaS + C02 + H20.
The above equations were solved by an iterative procedure for the mols of air
and steam fed to the burn-up cell as a function of the following:
1. Burn-up cell temperature. 3. CaS/total calcium, mol ratio in sorbent.
2. Gasifier temperature. 4. Regenerator temperature.
Once having computed the input and output flows and compositions by the method
outlined above, it was necessary to determine the gasification reactor sizes.
The vessel sizing is determined by the interaction of the fluidization mechanics
of the char particles and the gasification kinetics.
The basis used here was to process 120,800 Ib/hr of coal through each pretreater-
gasifier. The fluidized bed height was fixed at 50 feet. The fluidized bed
density was then calculated using the correlation developed during the work on
the C02 Acceptor Process.!2) It was assumed here that a high density closely
sized char particle would be generated by the "Seeded Coal Process" from Ireland
Mine coal, i.e., the mean particle diameter was taken as 0.04 inch and the initial
particle density as 85 lb/ft3. The reduction in particle density as affected by
carbon burnoff was based on the relationships developed during the work on the
C02 Acceptor Process.!3)
- 11 -
-------
It is now necessary to compute the vessel cross section and fluidizing velocities
which are compatible with the estimated bed inventories by use of reaction kinetic
data.
Extensive differential rate data'4;5;6.) were obtained some time ago on the gasifi-
cation of bituminous coal chars as a function of temperature, pressure and mol
fraction of hydrogen in hydrogen-steam mixtures. Subsequently, extensive kinetic
data were obtained also on the gasification kinetics of lignite char.'3;7) In
that work it was found that the reaction rate was strongly inhibited by the
presence of CO as well as hydrogen. Thus, the prior data on bituminous coal chars
could not be used in practice since the inhibiting effect of CO was not taken into
account. However, under comparable conditions, i.e., in the absence of CO, it
was found that on the average the bituminous coal chars had about l/15th the
reactivity of the lignite chars.
Therefore, in developing the kinetic calculations the equations developed for
lignite char(7) were used with introduction of a correlation factor of l/15th to
account for the lower reactivity of the bituminous coal chars. The differential
kinetics were translated into integral kinetics, i.e., average over the whole bed
by the method given in Appendix D of the Annual Report.i1/
Having calculated the integral rate R-p, the required bed inventory is calculated
from the equation,
Ib Fixed Carbon Gasified -4
T min/lb Fixed Carbon in Bed
The fluidization calculations outlined above are then used to determine the
required vessel cross section and fluidizing velocity.
A total o"f fifteen cases were computed covering gasification temperatures between
1650 and 1750°F. Because of the kinetic and thermodynamic limits, the system is
highly constrained, and the response of the system is quite limited. This is
illustrated by the ranges given in Table 1.
The conditions for the base case, which all things considered, is felt to be close
to a practical "optimum" for the system, are also given in Table 1. A more
complete heat and material balance around the base case is also given in Table 2.
The process concept given here has several potential advantages over the original
process which utilized the C02 Acceptor Process as outlined below;
Operability may be improved in the new process by virtue of;
1. The 02 partial pressure to the burn-up cell is lower than to the previous
regenerator. Steam and air N2 serve as the heat sink. There is less
chance of ash slagging, especially since the burn-up cell can be operated
at much lower temperatures than are needed to calcine CaC03.
2. The 02 partial pressure to the gasifier is lower since the air is diluted
with all of the input steam and all of the products of combustion of the
burn-up cell. Thus, the temperature can be raised to 1750°F as in the
base case to improve kinetics without increasing the danger of ash slag-
ging.
- 12 -
-------
Table 1
Range of Variables Studied in System Analysis
Range of Independent Variables Studied
Gasifier Temperature, °F
Burn-up Cell Temperature, °F
S/S Ca, mol ratio
Regenerator Temperature, °F
Range of Calculated Quantities
Cold Gas Efficiency., %
HHV Dry Product Gas, Btu/ft3
$> Sulfur Removed
Steam Conversion, $
Carbon to Burn-up Cell, Wt fo of C
in Coal Feed
Ib C (gasified) x 1O4
Ib C in Bed, min.
Gasifier Diameter, I.D., ft
Char Particle Density, lb/ft3
Gas Fluidizing Velocity, ft/sec
Gasifier Cross Section Index*
Constant Parameters
System Pressure, atm.
Gas Outlet Temperature, °F
Base Case
Gasification Rate
1650-1750
1750-1950
0.1-0.4
1200-1300
79.1-81.2
143-149
93.1-97.5
49.4-55.7
13.4-16.9
45-88
24.8-25.3
27-36
1.46-1.55
392-415
-« 15
*. ... i son -
1750
1800
0.13
1300
79.8
144
97.0
52.9
13.4
88
24.8
27
1.48
395
* Ft2/109 Btu/hr (HHV of Product Gas)
- 13 -
-------
Table 2
C2)
Stream No.v ' 1
Identification Air
Temperature, °F 398
Mols 11. 7O
Pounds
Composition. Mol $>
CH4
H2
CO
C02
H2O .90
H2S
N2 78.3
O2 20.8
NH3
MgO • CaS
MgO-CaC03
Hydrogen (as H2)
Carbon
(i) Ireland Mine Coal:
Wt.
H
C
N
O
s
Ash
Key Stream FJ'ows and Analyses
Basis; 1OO Ib dry coal(1)
6% moisture as fed
Syste'm Pressure: 15 atm (206 psig)
234 56 7 8 91O
Steam Fuel Burn-up
Char Cell Gas
780 1750 1800
3.264 .814 12.59
21.73
X
X
X
6.19
26.93
X
57.84
9.O4
4.31
95.69
% Dry Basis Mols
4.8 2.381
69.8 5.812
1.2
7.6 .475
4.3 .1341
12.3
Raw Spent "Squires" Regener- Sulfur CO2
Product Acceptor Offgas ated
Gas Acceptor
175O 1652 13OO 130O 31O 4OO
20.50 1.OO1 3.792 1.OO1 .1296 .1296
4.15
1.39 x
16.23 x
19.42 x
7.54 63.95
9.72 32.44
.65 3.61
44 . 8O x
x x
.25
88.4 75.4
11.6 24.6
(2) Stream numbers correspond to those shown in
Figure 1.
11
Clean
Product
Gas
13OO
20. 63
1.38
16.51
18.91
8.51
9.9O
.02
44.52
x
.25
-------
3. More positive contact of dirty gas with the dense-phase bed of acceptor
in the gas desulfurizer is effected.
The cold gas efficiency is definitely improved by virtue of:
1. Lower duty to calcine make-up acceptor. Circulation rate is about l/10th
that of the original concept.
2. 100$ burn up of carbon (vs. 98$).
3. Improved gasification kinetics require less steam. Thus, there is less
latent heat in the product gas.
4. Less air required. Thus, less sensible heat is lost with N2.
Experimental work was carried out during the contract period covered by 'this
report to evaluate at least in a preliminary manner all of the process steps
shown in the schematic diagram of Figure 1. Details of the work are given in
the experimental sections of this report.
The feasibility study itself was limited to an economic evaluation of the treat-
ment of the hot producer gas to remove sulfur by means of the dolomite sorbent;
the regeneration of the spent sorbent and the conversion of H2S in the regenera-
tion offgases to elemental sulfur by means of the liquid-phase Claus process.
C. Summary - Gas Purification Costs
Under this contract an economic evaluation of the process is required. The prime
purpose of this evaluation is to determine the commercial feasibility of the
process, and to delineate those technical areas where further research and
development are best directed. The feasibility study is restricted to the cost
of desulfurizing the producer gas, including sorbent regeneration and sulfur
recovery. The cost of gasification is not included, nor are estimates developed
of the cost of generating electric power from the desulfurized producer gas.
The most attractive end use of clean, pressurized, hot producer gas in the long
term is in combination with a combined cycle power station. However, since only
the gas cleanup end of the plant was evaluated, the commercial evaluation was
based on low-sulfur boiler fuel which is a more immediate practical objective.
The plant was selected to supply a desulfurized fuel gas to a conventional, gas-
burning power station with an approximate rating of 1400 MW. With respect to
quantity of gas, this is the same plant size that was considered in the initial
feasibility study. I1)
The primary evaluation presented in this study is based on these criteria:
1. Sulfur is removed hot with a regenerable dolomite sorbent.
2. Desulfurized producer gas expanded through a power-recovery turbine
before delivery at 1O psig to a power station.
3. Two extremes of further clean up of desulfurized product gas to remove
particulates before expansion;
a. high pressure drop cyclones (or equivalent) only - Case I.
b. conventional water scrubbing with subsequent gas reheat - Case II.
- 15 -
-------
This evaluation, based on an arbitrary start-up date of January, 1978, is
summarized in Table 3.
The cost of the plant facilities to desulfurize producer gas under pressure by
this process amount to $32 to $40 per installed KW (1978 cost basis) depending
on the mode of subsequent gas cleanup, and the annual operating costs correspond
to 1.5 to 1.9 mills/KWH of power generated.
Direct comparison with the economics previously presented in the Annual Report'1)
is not possible since the previous study was concerned with the economics of the
whole system including the gasifier. The cost of the desulfurization section
alone cannot be broken out, since in the former process it was an integral part
of the gasifier. The rough magnitude of costs, however, appear to be consistent
with each other.
It is clear that there is a significant advantage to be gained from the develop-
ment of a "hot" gas clean-up technology for alkalis and solids removal (Case I).
When compared to the alternative - wet scrubbing with gas reheat (Case II) - the
advantage is about $1.00 per ton of feed coal, or 5.3^/MM Btu of product gas.
Over and above the direct cost advantage shown in Table 3 for Case I over Case II,
there is also additional excess power available from the expansion of the product
gas (about 42 MW equivalent). While it is beyond the scope of this study to
evaluate the dollar value of this excess power, it could amount to an additional
credit of 0.3-0.6^/MM Btu of product gas.
A direct comparison of the cost of the "all hot clean-up system" with an "all
cold system" including removal of H2S by wet scrubbing has not been made since it
was outside the scope of this study.
The make-up acceptor cost for this evaluation was assumed at a relatively high
value of $10 per ton of raw dolomite. The requirements for the regenerative
cases are, however, quite modest, and the acceptor cost does not have a signifi-
cant impact on the desulfurization cost. In this study the rate of make-up
acceptor required is estimated as one percent of the circulating acceptor. This
is based on a reasonable extrapolation of the laboratory data obtained to date.
If the make-up rate were doubled, the desulfurization cost would increase only
by 4-5$.
Rather interestingly, the sensible heat content of the desulfurized producer gas
delivered to the power station is not an insignificant factor. In a conventional,
gas-fired power station, the fuel gas is delivered to the burners at ambient
temperature. The combustion air is preheated to 250 to 350°F in an air heater by
indirect sensible heat exchange with flue gas. In this process the fuel gas is
delivered from the expander to the power station at 540-660°F. If the fuel gas
is preheated, then either less air heater surface is required, or more steam can
be generated per unit of fuel burned. An exact determination of the dollar value
of sensible heat in the fuel gas would require an extensive analysis of the design
and cost of the power station proper. This is beyond the scope of this study.
However, a reasonable approximation of the value is to credit the system for the
sensible heat content of the fuel gas above the temperature of the flue gas exit-
ing the air heater. In this study this temperature is assumed as 300°F. In
Table 3, the credit to the process for this excess sensible heat is about 1.0^/MM
Btu or about 5$> of the system cost.
- 16 -
-------
Table 3
Summarized Economic Evaluation
Regenerative Acceptor Desulfurization Process
Plant Operating Factor (6132 hrs/year)
Case
Method of Gas Clean-up
(Particulate and Alkali)
Coal Required:
MM Tons/Year (6/0 Moisture)
Higher Heating Value, Btu/Year
Producer Gas to Station (After Expansion):
Mols/Hour
Temperature
Pressure
Higher Heating Value, Btu/Year
HHV + Sensible Heat Content^1) Btu/year
Desulfurization Plant Investment( 2)
Desulfurization Cost:(2/
$/Ton Feed Coal
MM Btu HHV of Feed Coal
"Hot" Cyclones
only
II
Wet Scrubbing
With Gas Reheat
- 3,422,000 —
81.71 x 1012
213,700
660°F
10 psig
65.19 x 1012
68.68 x 1012
$48.3 MM
4.03
16.9
213,200
540°F
10 psig
64.70 x 1012
66.99 x 1012
$59.8 MM
4.98
2O. 9
j^/MM Btu HHV of Product Gas
jzf/MM Btu (HHV + Sens. Heat) Product Gas
Excess Power Generated by Expander , megawatts
(3)
21.2
20.1
16o(
26.4
25.4
ns(
(i) Sensible heat content above an assumed air heater outlet temperature of 300°F.
(2) January, 1978 operation - includes escalation and interest during construction.
Does not include gasification plant cost or the cost of coal.
(3) After deduction of 157 MW required for compression of process air for
gasification plant.
(4) Assuming 91$> of the isentropic efficiency for the expander, and 89$ of the
polytrbpic efficiency for the air compressors (See Reference 8).
- 17 -
-------
While the basic development of this project to date has been devoted to the
concept of a regenerable sulfur acceptor, there is some interest in evaluating
an alternative process using once-through acceptor. This would result in some
simplification of the sulfur recovery system, at the expense of a much greater
usage of make-up acceptor, projected to be approximately 15 times the rate of
the regenerable acceptor case.
The added make-up acceptor cost for the once-through system (at a make-up
acceptor cost of $10 per ton) amounts to about $9,200,000 per year. The maximum
saving in plant investment would be $10 MM. At a capital charge of 18$, the
annual savings in investment charge is $1,800,000. Make-up acceptor cost
exceeds this saving by a five-fold factor.
The maximum savings in the once-through as compared with the regenerative system
is limited because of the much greater disposal costs in the former case. A
significant cost item here is the necessity to treat the spent stone via the
Chance reaction to convert the CaS to CaC03 before disposal. This step is
expensive since 2 mols of C02 are required per mol of CaS treated since the MgO
is converted to MgC03 in the process.
It is concluded that there is no economic incentive to further consider the non-
regenerative system.
Possibly an alternative disposal method could be developed where spent acceptor
could be disposed of as MgO-CaS04 (e.g., by high-temperature oxidation of
MgO-CaS). However, our limited data show that the spent CaS is difficult to
oxidize and that, during oxidation, some S02 inevitably is released which causes
an air pollution problem.
D. Process Description - Gas Purification
Introduction
The scope of work under this contract covers the experimental development and
economic evaluation of a novel process for desulfurizing pressurized producer gas
as a fuel to a power station. The simple block flow diagram in Figure 2 defines
the various steps in the overall process. Design and evaluation of coal prepara-
tion (crushing and sizing), and of coal gasification (Sections 100 and 200,
respectively), are not covered in this study. Section 300, sulfur removal, and
Section 400, sulfur recovery, have been completely evaluated and form the sub-
stance of this report.
A relatively small quantity of make-up C02 is required in the process (Section
50O). As shown in Figure 2, the C02 could be recovered either from a slip-stream
of the desulfurized producer gas before power generation, or from a portion of the
power station stack gas. Preliminary cost studies indicate approximately equal
costs for the two alternative locations. For the purposes of this study, it is
assumed that the make-up C02 is recovered from the pressurized producer gas before
delivery to the power station. A standard hot potassium carbonate system is
assumed to establish capital cost and utility requirements.
There is continued interest in the added cost of wet scrubbing the particulates
and reheating the product gas before charging the gas to an expanding turbine-
generator set. The equipment required for.this additional step is shown in
Figure 3 and is identified as Section 600.
- 18 -
-------
Figure 2
Block Flow Diagram - Overall
1
r
i
^^i
System
Make-up
C02
i
• Section 500
L
Stack
K
"1
1
1
|_J
1
1
J
IV Ga
Power
Station
s
r
| Make-up
1 C02
i ,, ,
' Section 500
1
1
1
1
J
Desulfurized
Producer Gas
Feed Coal
Coal
Preparation
Section 100
Gasification
Section 200
Producer
Gas
r
• Make-up
C02
Sulfur
Removal
Section 300
Sulfur
Recovery
Section 400
Ash
Sulfur
Note: Alternative locations of Section 500 shown by dotted lines.
-------
C-6O1A
8 Parallel Units
Q = 54.3 MM Btu/hr, Ea.
A = 12,750 ft2 Ea.
Shell 6' 0 12OO°F, 25O psig
Tubes 1820 1-1/4" O.D.-24'Lg.
135O°F 25O psig
FIGURE 3
Alternative Gas Clean-up System for Particulates
Section 6OO
Wet Scrubbing with Gas Reheat
C-6O1B
8 Parallel Units
Q = 54.3 MM Btu/hr, Ea.
A = 12,75O ft2, Ea.
Shell 6' 0 1000°F, 25O psig
Tubes 182O 1-1/4" O.D.-24'Lg.
1100°F 250 psig
C-6O1C
8 Parallel Units
Q = 54.3 MM Btu/hr, Ea.
A = 12,750 ft2, Ea.
Shell 6' 0
Tubes 182O
8OO°F
8OO°F 25O psig
1-1/4 O.D.-24'Lg.
25O psig
Gas from C-501 (Make-up CO2 System)
Gas to Expander
17,970 mph 1220°F
198 psia
Gas from C-302
195,731 mph 1300°F
21O psia
to
o
Make-up Water
213,214 mph 1100°F 198 psia
1090°F
1024°F
C-6O1A
L-601 A-D
Venturi Scrubbers
39,750 ACFM Ea.
500°F 250 psig
699,800 Ib/hr 90°F
Black Water to Pond
J-601 A-C
Water Pumps
70O gprn Ea. 243 ps i
BMP 146 Ea.
F-601 A-B
Surge Drums
10,000 Gal. Ea.
9' 0 x 21'
500°F 250 psig
120°F 722,500 Ib/hr
C-602 A-D
Water Coolers
Q=85.2 MM Btu/hr Ea.
U=90 A=14,350 ft2 Ea.
J-6O2 A-C Black Water Pumps
318O gpm Ea. 58 psi
BPH 165 Ea.
-------
Sulfur Removal
A schematic flow diagram of the proposed commercial embodiment of this gas
desulfurization process is shown in Figure 4 (Consol Drawing No. AF-3424).
Gasifier product gas (Stream No. 4) flows to the bottom of a fluidized gas de-
sulfurizer (D-301). The gas fluidizes a bed of sized dolomite at a temperature
of 1652°F and at 15 atm. absolute pressure. Most (97$) of the H2S in the gasi-
fier gas reacts with the CaC03 component of the dolomite as follows:
H2S + MgO-CaC03 = MgO-CaS + C02 + H20.
The desulfurized gasifier gas then passes through two stages of cyclones (G-301
and G-302, respectively) to remove substantially all of the fines including
attrited acceptor and entrained ash. A portion of the sensible heat content of
the desulfurized gas is used to generate and superheat the steam required for
gasification of the feed coal in heat exchangers C-302 and C-301, respectively.
The desulfurized producer gas at 1300°F and 210 psia (Stream No. 6) is then
delivered to the power station. Subsequent use of this gas is dependent on the
design of the power station, and is discussed in other sections of this report .
Regenerated acceptor (Stream No. 5) at 13OO°F is continuously charged to the top
of the gas desulfurizer; sulfided acceptor is continuously withdrawn via a
standleg (Stream No. 7). The sulfided acceptor is pneumatically conveyed by a
portion of the required recycle gas (Stream No. 9) to a fluidized acceptor
regenerator (D-3O2). Make-up acceptor (Stream No. 8) is introduced into the
system by a lockhopper system comprised of L-301, F-301, F-302, and L-302. This
small stream is also pneumatically conveyed to D-302 by means of recycle gas.
Make-up CO2 required in D-302 (Stream No. 1O) is pressurized by JC-301 and
.'Charged to D-302.
The fluidized acceptor regenerator (D-302) is maintained at 1300°F and 15 atm.
absolute pressure. The sulfided acceptor is recarbonated at these conditions
by the reverse reaction,
MgO-CaS + C02 + H2O = MgO-CaC03 + H2S.
The carbonated magnesium component of the make-up acceptor is also calcined at
these conditions,
MgC03-CaC03 = MgO-CaC03 + C02.
The regenerated acceptor is returned to the gas desulfurizer by gravity flow
(Stream No. 5). Spent acceptor (l$ of the circulating flow) is withdrawn from
D-302 via a lockhopper, F-3O3, and a rotary feeder, L-305 (stream No. 11).
This spent acceptor must be treated before disposal. About 75$ of the calcium
component of this stream is in the form of CaS. If this were disposed of
directly to the station ash pit, H2S gas would slowly evolve as the CaS was
hydrolyzed. To avoid this unacceptable condition, the spent acceptor is
directly contacted with C02 and water in three stages of stirred reactors,
D-303, to convert the CaS to CaC03. At these conditions, the MgO component of
the dolomite would also recarbonate. The overall reaction is,
MgO-CaS + 2 CO2 + H2O = MgCO3-CaC03 + H2S.
- 21 -
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
The spent acceptor is conveyed hydraulically by a dilute recirculating slurry
stream to a hydroclone, G-305. The underflow slurry (controlled at 35 wt. %
solids) flows to the three series converters, D-303A, B and C, respectively.
Each converter is agitated by a turbine-type mixer, L-305, and some reaction
heat is removed in each stage by cooler, C-305. The spent acceptor, stripped
of sulfur and fully carbonated, is pumped to the station ash pit by slurry pump,
J-301. The overflow slurry from hydroclone, G-305, is continuously recirculated
via surge drum, F-304 and circulating pump, J-303. The sensible heat loss from
the spent acceptor is removed by cooler, C-304. Make-up water to this system is
supplied by station pond overflow via J-302 (which also supplies the water to
cool the gasifier ash and convey it to the station ash pond).
Acid gas resulting from the acceptor stripping operation (Stream No. 16) is
compressed to system pressure by JC-302 and flows to the liquid-phase Claus
reactor.
Sulfur Recovery
The process gas exiting D-302 (stream No. 12) is passed through two stages of
cyclones (G-303 and G-304, respectively), to remove entrained acceptor. The
sensible heat content of this stream is then exchanged with recycle gas in
exchanger C-401, and with boiler feed water (required for gasification steam)
in C-402. An electrostatic precipitator, L-401, is provided to remove all
entrained dust. The gas then flows to the bottom of the liquid-phase Claus
reactor, D-401.
The concentration of H2S in the gas from the acceptor regenerator is limited by
the equilibrium restriction at 1300°F and amounts to only about 3.6 volume per-
cent. The liquid-phase Claus reaction developed in this work is uniquely suited
to processing this gas. The liquid-phase Claus reactor, D-401, operates at
310°F and about 210 psia. Product gas from the acceptor regenerator (Stream No.
12), gas from the reject acceptor stripping section (Stream No. 16), and dilute,
aqueous H2S03 (Stream No. 17) are charged to the bottom of the liquid-phase
Claus reactor. Flow is upward through a sparged reactor containing only water,
as was demonstrated during the experimental work described in Section XI.
Liquid sulfur is produced by the reaction,
2 H2S + H2S03 = 3 S + 3 H20.
Liquid sulfur and liquid water flow from the reactor to a decanter-type separa-
tor, F-4O2. Unreacted gas (Stream No. 9), saturated with water vapor at 31O°F,
is compressed by JC-401, reheated in exchanger, C-401 to 975°F and returned to
the acceptor regenerator reactor. The sensible heat content of the liquid water
from F-402 (Stream No. 19), is exchanged with the feed acid (Stream No. 17) in
C-404, further cooled to 90°F in C-405 and charged to the S02 absorption tower,
D-402.
Approximately one-third of the sulfur from F-402 is burned with stoichiometric
air (Stream No. 20) in pressurized combustor, B-4O1, to produce S02. Excess
heat is removed to boiler feed water via cooling tubes in the walls. The exit
gas from the sulfur combustor (Stream No. 21) flows to the base of the S02
absorption tower, D-402. Water (Stream No. 19) flowing down through the packed
tower absorbs the S02 in the gas by,
S02 + H20 = H2S03 aqueous
- 23 -
-------
Most of the exothermic heat of reaction is removed by side stream cooler, C-406.
The vent gas from the absorption tower (stream No. 22) is at 90°F and 205 psia.
It probably would contain some residual S02 (assumed in this case as 0.3 volume
percent). The most practical means of disposing of this gas is to bleed it to
an intermediate stage of the expansion turbine required for the desulfurized
producer gas in the power station. This would decrease the sulfur removal
efficiency of this process from 97.0$ to 96.5$>, but would increase the power
output of the station slightly, and would eliminate an entire stack gas scrub-
bing installation following the S02 absorption step.
Wet Scrubbing - Gas Reheat
For the case where the product gas must be wet scrubbed (for particulate and
alkali removal), then reheated for use in an expanding turbine, the additional
equipment in Section 600 (Figure 3) is required.
That portion of the process gas exiting from the gas desulfurization system that
is not charged to the make-up C02 system (Section 500) is available at 1300°F
and 210 psia. The gas is cooled to 472°F by heat exchange with the scrubbed gas
in three series steps (C-601A, B, and C, successively). The cooled gas then
flows to four parallel Venturi scrubbers (L-601) where it is scrubbed by circu-
lating water at the gas dewpoint of 228°F. The clean gas exiting the scrubbers
is then reheated to 1090°F by the aforementioned exchangers (C-60l). This
cleaned, reheated gas, combined with the product gas diverted to the make-up
C02 system, comprises the total product gas flow to the power station. The
combined gas is at a tempterature of 11OO°F and at a pressure of 198 psia.
The circulating water flow to the Venturi scrubbers flows via surge drums F-601
and black water pumps, J-602. The excess sensible heat in the gas is removed
via coolers, C-602. Make-up water enters the system via water pumps, J-601.
.The make-up water rate is based on an assumed solids loading in the process gas
of 5 grains/cubic foot, and a solids content in the circulating water of 2 wt. %.
Net solids removed from the gas flow as a 2 wt. $ slurry to the station pond.
E. Design Basis
Gasification
Although the detailed design and evaluation of the gasification system is not
covered in this study, it is of interest to at least relate the composition of
the raw gasifier gas to the input feed streams to the gasification section.
1. Feed Coal
This study is based on the use of a high-sulfur content, highly-caking Pittsburgh
Seam coal. The analyses of a typical coal (Ireland Mine) of this seam is shown
below:
- 24 -
-------
Feed Coal Analysis
2.
Moisture, as received
Ultimate Analysis. MF basis
Hydrogen
Carbon
Nitrogen
Oxygen
Sulfur
.' Ash
Higher Heating Value, MF basis
Other Design Constraints
6.0 wt o
4.8 wt $
69.8 wt io
1.2 wt %
7.6 wt io
4.3 wt %
12.3 wt %
12,700 Btu/lb
The total sensible heat content of the gasifier steam is supplied by heat
exchange within the sulfur removal and sulfur recovery sections. Additional
design constraints for the fluidized bed gasifier and carbon burn-up cell are;
System Pressure 15 atm.
Gasifier Temperature 1750°F
Gasification Rate 0.00883 Ib C/min-lb C in bed
Steam Conversion in Gasifier 52.9$
Burn-up Cell Temperature 1800°F
Percent Total Carbon to
Burn-up Cell 13.4$
Gas Desulfurizer
The sulfur in the producer gas (as
reaction with CaC03 as follows;
is removed in the gas desulfurizer by
MgO-CaC03 + H2S = MgO-CaS + C02 + H20
The commercial design basis for this study is adopted from the actual experi-
mental data which are reviewed in a later section of this report. Pertinent
design values are:
Desulfurizer Temperature
Desulfurizer Pressure
Avg. Acceptor Size, Dp arithmetic mean
Acceptor Density, ps
Mol io Ca as CaS - Inlet Acceptor
Mol $ Ca as CaS - Outlet Acceptor
Atols CasV _ /Mols CaS\
\Mols Ca J U VMols Ca /
Avg. Acceptor Residence Time
Outlet Fluidizing Velocity
Approach to Equilibrium
(H20)(C02)/H2S
Removal of H2S from Producer Gas
C02 Driving Force
In
1652°F
15 atm.
0.065 inch
130 Ib/CF
75.4$
88.4$
30 min.
3.4 ft/sec
97$
0.147 atm.
Excluding make-up acceptor.
p - equilibrium p for the reaction, CaC03 = CaO + C02l
- 25 -
-------
Acceptor-Regenerator
The highly-sulf ided acceptor from the gas desulfurizer is regenerated (converted
to CaC03) at a low temperature by the reverse reaction,
MgO-CaS + C02 + H20 = MgO-CaC03 + H2S .
Again, the commercial design basis for this study is adopted directly from the
actual experimental data. Note that the small yields of COS and S2 actually
detected in the experimental program have been ignored in this study.
Presumably these compounds are formed from the reactions:
H2S + C02 = COS + H20
2 H2S = S2 + 2 H2
It is assumed that, for long-term operation, an equilibrium concentration of COS
and H2 would be established in the recycle gas and completely inhibit further
reaction.
The acceptor make-up requirement was arrived at by extrapolation of data from
the continuous unit. The sulfur absorption was assumed to be 13$ of the total
calcium for 13 cycles, and then decline exponentially to zero at infinite
cycles. The rate of decline was assumed to be sufficiently small to permit a
1$ make-up rate .
The required make-up acceptor (as MgC03-CaC03) is charged to the acceptor-
regenerator. The MgC03 component of the dolomite is completely calcined at these
conditions as follows:
MgC03'CaC03 = MgO-CaCO3 + CO2
Other pertinent design values for the commercial acceptor-regenerator are:
Regenerator Temperature 1300°F
Regenerator Pressure 15 atm.
Avg. Acceptor Size, Dp arithmetic mean 0.065 inch
Acceptor Density, ps 132 Ib/CF
Mol $ Ca as CaS - Inlet Acceptor^) 88.4$
Mol $ Ca as CaS - Outlet Acceptor^) 75.4$
(l)
Ca / \ Mols Ca
Avg. Acceptor Residence Time 60 min.
Outlet Fluidizing Velocity 1.55 ft/sec
Approach to Equilibrium, (H20) (C02)/H2S 87$
(i) Excluding make-up acceptor.
Acceptor Converters
In this process the spent acceptor to be rejected from the system must be con-
verted from MgO-CaS to MgC03 .CaC03 before disposal. In the unconverted state,
the acceptor would slowly release H2S to the atmosphere.
The conversion is by the chemical reaction,
MgO'CaS + 2 CO 2 + H20 = MgC03-CaC03 + H2S .
Data for the design of the acceptor converters was obtained from the Bureau of
Mines staff at Salt Lake City. These data are not yet published.
- 26 -
-------
Pertinent design values are:
Conversion Temperature
Conversion Pressure
Wt. "jo Acceptor in Slurry
No. Stages Required
Utilization of C02
Conversion of CaS to CaC03
120°F
Atmospheric
45
3
Approaching 100%
Liquid-Phase Claus
The offgas from the acceptor regenerator is passed cocurrently through a reactor
with a stream of aqueous H2S03 where the following reaction occurs,
2 H2S + H2S03 = 3 S + 3 H20.
The design basis is taken directly from the experimental results from Runs
C-9, C-10, and C-ll as follows:
Reactor Temperature
Reactor Pressure
Volumetric Space Rate
Outlet Gas Velocity
Avg. Vapor Residence Time
Conversion H2S to S
320°F
14 atm.
1340 hr"1
0.25 ft/sec.
27 sec.
90$
S02 Absorption Column
The dilute H2S03 required in the liquid-phase Claus reactor is produced by
absorption'-'of S02 in water in a standard countercurrent flow packed tower. The
design is based on published data for the system.(9>10) Key design parameters
are;
Absorber Temperature
Absorber Pressure
Liquid Rate
Packing
No. Stages
Outlet Vapor Velocity
Percent Flooding Velocity
90-110°F
14 atm.
10,000 lb/hr-ft2
3" porcelain rings
4
0.27 ft/sec
50$
F. Material and Heat Balances
Basic to the process design and economic evaluation of this process on a
commercial scale was the completion of mass and heat balances for each important
processing step. The capacity of the plant in this study (with respect to the
total heating value of the desulfurized producer gas), is the same as in the
initial feasibility study.(1) This plant would supply a modern conventional
gas-fired power station of approximately 1400 MW rating.
The computer program developed by Consol for the integrated system of coal pre-
oxidation, gasification, carbon burn-up cell and producer gas desulfurization
has been described earlier in this section.
- 27 -
-------
With these calculations as a starting point, the following mass and heat balances
were evolved for the commercial process design;
Table 4 - Gas Desulfurizer
Table 5 - Acceptor Regenerator
Table 6 - Spent Acceptor Stripping
Table 7 - Liquid-Phase Claus Reactor
Table 8 - Sulfur Combustor
Table 9 - S02 Absorption Column
Table 10 - Over-all Mass and Heat Balance
These balances define the regenerable acceptor case. As is covered later, it
was not necessary to define in this detail the once-through acceptor system.
G . Economic Evaluation
Introduction and Scope
The process under development in this work produces a desulfurized producer gas,
under pressure, as a desirable fuel gas for power generation. In the first
Annual Report on this project,!1) it was clearly demonstrated that the most
economic application of the process was in combination with a pressurized com-
bined cycle power station. However, these advanced power generation techniques
are only now under development. This study is restricted to delivery of the
desulfurized gas to a conventional gas-burning station at a pressure of 10 psig.
The pressurized producer gas must be expanded from system pressure (about 200
psig) to 10 psig. This is best accomplished in an expanding turbine-generator
set. The power output from this unit exceeds the power requirement to compress
the air for the coal gasification step. The excess power so recovered is a net
credit to the process. However, it is beyond the scope of this contract to
define the actual dollar value of the power credit.
There remains a technical question as to the required quality of gas (with
respect to content and size of particulate matter and alkali) that may be charged
to an expander. It is obvious that the maximum power will be recovered from the
expanding turbine if the pressurized producer gas can be adequately "cleaned" by
high pressure drop cyclones alone. Conversely, the gas may be water scrubbed
(in conventional Venturi scrubbers) then reheated by indirect heat exchange with
hot gas to the scrubbers. This variation assures that the expansion turbine is
protected from particle erosion and/or alkali deposition. There is, of course,
a reduction in the power recovery possible since the gas can only be reheated
(economically) to about 11OO°F, and there is a pressure drop through the scrub-
bing and reheating system (estimated in this study as 12 psi).
A review of the "state of the art" on the use of high-pressure drop cyclones to
remove particulate matter from a gas feed to a turbine is included in a recent
Westinghouse report to EPA.v11)
In this study two cases are presented to cover the range of this technical
uncertainty:
Case I - "Hot clean-up" by cyclones only.
Case II - Wet scrubbing with gas reheat.
- 28 -
-------
Table 4
Mass and Hoot Balonco
Oaf Desulfurizor
Baals:. 1 hour
Datum: 60°P, H2O (l)
miv
Elemental Balance, Lb
Input
(T) Gasifler Gas
CH4
H2
CO
C02
X2
XH3
H2S
HjO (v)
Sub-Total
JlgO CaS
HgO CaCOg
Inert!
Sub-Total
Total!
Output
(^, Producer Gas
CH<
H2
CO
C02
N2
SH3
H2S
H2O (v)
Sub-Total
(7J Acceptor to Ace.
«gO C«S
UgO CiC03
Inerts
Sub-Total
Heat of Reaction
Heat Loss
Totali
Lb
47,950
70,350
1,170,250
713,400
2,700,100
9,150
47,950
376,850
5,135,000
R
890,700
362,850
178,550
1,432,100
6,568,100
47,650
72, 100
1,146,400
810,950
2,700,100
9,150
1,430
386,100
3,174,200
Regon.
1,04>I,300
171,150
178,550
1,393,900
klgO ClC03
CO + HjO <
6,968,100
H.W.
16.04
2.016
28.01
44.01
28.02
17.03
34.08
18,016
112.46
140.41
-
- .
-
16.04
2.016
28.01
44.01
28.02
17.03
34.08
18.016
"
112.46
140. 41
-
-
+ HjS .
Uols
2,990
34,899
41,779
16,210
96,364
939
1,407
20.917
215,106
7,920
2,584
•
10,304
-
2,990
39,750
40,928
18, 427
96,364
539
42
21,431
216,471
9,283
1,219
-
10,304
UgO ClS +
Hoi %
1.39
16.23
19.42
7.34
44.80
0.29
0.69
9.72
100.0
75.4
24.6
-
100.0
-
1.38
16.51
18.91
8.91
44.92
0.29
0.02
9.90
100.0
88.4
11.6
-
100.0
coa + H
• C02 » Hj (891 mola x 1,
-
-
-
H
12,090
70, 330
x
X
X
1,600
2,850
42.200
129,050
X
X
X
X
129, 050
12,050
72,100
x
X
X
1,600
100
43.200
129,030
X
X
X
X
C
35,900
x ••; >
901.790
194,700
x
x
X
X
732,350
x
31,050
-
31,090
763,400
t
39,900
X. :. !
491,990
221,300
x
X
X
• x
748,790
x
14,650
x
14,690
N
X
' X
X
X
2,700,100
7,550
x
X
2,707,050
X
X
X
X
2,707,650
x
X
X
X
3,700,100
7,390
x
X
2,707,630
x
x
X
X
0
x
X
668,600
918,700
x
X
X
334,650
1,621,850
X
124,050
x
124,050
1,645,900
X
x
634,890
989,690
x
X
X
342,900
1,987,400
X
98,900
x
98,900
8
x
x
X
X
X
X
45,100
X
45,100
253,900
x
X
253,900
299,000
x
X
X
X
X
X
1,350
X
1,350
297,690
x
x
297,630
Ash McO Ca
* x
X X
X X
X X
X X
X X
X X
X X
X X
x 630,800
x 207,750
178,550 x
178,550 844,550
178,550 844,650
X X
X X
X X
X X
X X
X X
X X
X X
X X
x 746,550
x 08,000
178,850 x
178,330 844,330
Temp.
Op
1750
1750
1750
1750
1750
1750
1750
1750
1300
1300
1300
1652
1652
1652
1632
1652
1632
1632
1632
1692
1632
1632
Ah
24,125 Btu/ool
11,643 Btu/mol
12,024 Btu/mol
19,929 Btu/mol
12,519 Btu/mol
24,126 Btu/mol
16,812 Btu/mol
34,492 Btu/mol
28,569 Btu/mol
45,746 Btu/mol
300.65 Btu/lb
22,290 Btu/mol
10,969 Btu/nol
12,109 Btu/nol
18,611 Btu/nol
11,707 Btu/nol
22,290 Btu/nol
19,682 Btu/nol
33,476 Btu/nol
37,941 Btu/nol
61,374 Btu/Bol
398.3 Btu/lb
20 (1,365 moll x 30,830 Btu/nol)
430 Btu/nol)
139,090
l •
783, 400
3,707,690
1,649,000
300,000
178,390 844,990
611
MM Btu
72. 1
406.3
940.0
323.0
1,206.0
• 13.0
23.7
721.3
3,305.6
22S.3
118.2
53.7
398.2
3,703.8
66.7
392.1
495.6
342.9
1,128.2
12.0
0.7
717.4
3,159. fl
348. a
74.7
71.1
404.4
42.1
1.3
10.3
3, 703. »
Unit
Btu/lb
23,878
61,100
4,347
x
X
0,680
7,100
x
-
-
-
23,878
61,100
4,347
x
X
9,680
7,100
x
-
-
.
Total
UM Btu
1,149.3
4,298.8
3,087.0
X
X
88.9
340.3
X
10,9130.4
-
-
-
~
1,143.3
4,403.6
4,983.4
x
X
88.9
10.1
x
10,631.3
-
-
.
-
Sole; Circled oumban refor to Strain Number on Coniol Dwf. AT-3434. T«op«ritur»/Pr«iiur« nay dlllar depending on location ot Identifying flag.
-------
Mass and Heat Balance
CJ
9
Acceptor Regenerator
Basis: 1 hour Datum; 60°F
Input
H20 (i)
Lb
Mols
Uol %
Temp.
°F
AH
£11
MM Btu
17 ) Acceptor from Gas Dosulfurlzor
MljO CaS
MgO CaC03
Inerts
Sub-Total
(8) Mako-up Acceptor
V-^ MgC03 CaC03
Inerts
Sub-Total
( 9 J Gao from Liquid- Phase Claus
C02
H2S
H20 M
Sub-Total
QOj Make-up C02 Gas
^ C02
H20 lv)
Sub-Total
Heat of Reaction
UgO CaS + COj + HaO . MgO
Totals
Output
fS) Acceptor to Gas Desulfurlzer
^•^ MgO CaS
UgO CoC03
Inert!
Bub-Total
(lly Acceptor to Stripping Column
MgO CaS
MgO CaC03
Inert*
Sub-Total
Q2) Gas to Liquid-Phase Claus
w coa
.Has
H20 (v)
Sub-Total
Heat of Reaction
MgCOs C*CO3 • UgO CaCOg +
Heat Loss
Total*
1,044,200
171,150
'178,550
1,393,900
19,350
1,800
21,150
1, 121,800
5,150
252,150
1,379,100
49,930
3.550
53,500
9,285
1,219
-
10,504
103
-
105
88.4
11.6
-
100.0
-
—
-
25,489.5 64.31
151.5 0.38
13,906
39,637
1,135
196
.1,331
f
35.31
100.0
85.3
14.7
100.0
1652
1652
1652
60
60
975
975
975
400
400
37,541 Btu/mol
61,274 Btu/mol
398.3 Btu/lb
-
-
9,861 Utu/mol
8,323 Dtu/mol
26,840 Btu/mol
3,314 Btu/mol
21,758 Btu/mol
CaC03 + H2S (1286 mo la x 30, 830, Btu/mol)
2,847,630
890, 700
362,850
178.580
1,432,100
8,900
3,630
1,800
14,350
1,119,760
49,000
232. 800
1,401,250
COj (105 mols x
2,647,700
7,920
2,584
-
10,504
79
26
-
103
96,443
1,437
12.909
39,787
76.4
24.6
-
100.0
75.4
24.6
-
100.0
.6 63.93
.8 J.ei
32.44
100.0
1300
1300
1300
1300
1300
1300
13OO
1300
1300
;
28 ,'569 Btu/mol
46,746 Btu/mol
300.66 Btu/lb
28,569 Btu/mol
45,746 Btu/mol
300.65 Btu/lb
13,955 Btu/uol
11,738 Btu/nol
29,933 Btu/nol
42,850 Btu/nol)
.
348.6
74.7
71.1
494.4
X
X
X
251. 4
1.3
375.6
628.3
3.7
4.3
8.0
39.6
1,170.3
226.3
118.2
63.7
398.2
2.3
1.2
0.5
4.0
353.0
16.9
388.3
758.2
4.5
S.4
1,170.3
Note: Circled numbers refer to Stream Number on Conaol Dug. AF-3424. Temporature/Presaure nay differ
depending on location of Identifying flag.
-------
Table 6
Mass and Heat Balance
Acceptor Stripping Column
Basis: 1 hour Datum: 60°F, H20 (l)
Input
11) Acceptor from Ace. Regen
<^/ MgO CaS
MgO CaC03
Inerts
Sub-Total
13) Make-up C02 Gas
C02
H20 (v)
Sub-Total
(l4) Water (l)
Heat of Reaction
MgO CaS + C02 + H2O =
(79 mols x 30
Lb Mols
8,900 79
3,650 26
1,800
14,350 105
10,150 230
700 40
10,850 270
26,850
MgO CaC03 + H2S
,830 Btu/mol)
Temp. AH
°F AH or CD MM Btu
1300 28,569 Btu/mol 2.26
1300 45,746 Btu/mol 1.19
1300 300.65 Btu/lb 0.54
3.99
140 736 Btu/mol 0. 17
140 19,708 Btu/mol 0.79
0.96
90 29.93 Btu/lb 0.80
2.44
MgO CaC03 + C02 = MgC03 CaC03
(105 mols x 42,850 Btu/mol)
Totals
Output
U5j Reject Acceptor to Pond
MgC03 CaC03
Inerts
Water (l)
Sub-Total
(16} Acid Gas to Liquid Phase
C02
H2S
H20 (v)
Sub-Total
Heat Rejected to Cooling
Totals
52,050
19,350 105
1,800
25,850
47,000
Claus
2,050 46
2,700 79
300 16
5,050 141
Water
52,050
4.50
12.69
120 40.3 Btu/mol-°F 0.25
120 0.21 Btu/lb-°F 0.02
120 59.86 Btu/lb 1.55
1.82
120 542 Btu/mol 0.02
120 491 Btu/mol 0.04
120 1085.6 Btu/lb 0.31
0.37
10.50
12.69
Note: Circled numbers refer to Stream Number on Consol Dwg. AF-3424.
Temperature/Pressure may differ depending on location of
identifying flag.
- 31 -
-------
Table 7
w
to
Basis: 1 hour - Datum; 6O°F, H2O (l)
Inpi
55"
Out
0-
at
Gas from Acceptor Regenerator
C02 1,
H2S
H20 (v)
Sub-Total 1,
Gas from Acceptor Stripping Column
C02
H2S
H20 (v)
Sub-Total
Liquid from S02 Absorber
H2 SO3 •
H20 (1)
Sub-Total
Heat of Reaction
2 H2S + H2S03 = 3S + 3 H2O (1365
Totals 2,
put
Gas to Acceptor Regenerator
C02 1,
H2S
H2O (v)
Sub-Total 1,
Sulfur (l)
Water (l)
Heat Loss
Totals 2,
Mass and Heat Balance
Liquid-Phase Claus Reactor
Temp
Lb Mols Mol % °F
119,750
49,000
232,500
401,250
2,050
2,700
300
5,050
56,000
720,650
776,650
mols H2S
182,950
121,800
5,150
252,150
379,100
65,600
738,200
182,900
25,443.5
1,437.5
12,906
39,787
46
79
16
141
682.5
40,001.5
40,684
x 42,980
25,489.5
151.5
13,996
39,637
2,047.5
63.95
3.61
32.44
100. 0
32.6
56.0
11.4
100.0
1.68
98.32
100.0
Btu/mol)
64.31
0.38
35.31
100.0
380
380
380
305
305
305
220
220
310
310
310
310
310
AH or Cp
3092 Btu/mol
2695 Btu/mol
21,600 Btu/mol
2340 Btu/mol
2049 Btu/mol
20,756 Btu/mol
1.0 Btu/lb
2391 Btu/mol
2092 Btu/mol
20,797 Btu/mol
0.2 Btu/lb °F
251.86 Btu/lb
MM Btu
78. 7
3.9
278. 6
361.2
0.1
0.2
0.3
0.6
122.3
122.3
60.9
0.3
291.1
352.3
3.3
185.9
1.3
Note: Circled numbers refer to Stream Number on Consol Dwg. AF-3424.
depending on location of identifying flag.
542.8
Temperature/Pressure may differ
-------
Table 8
Mass and Heat Balance
Sulfur Combustor
Basis: 1 hour Datum: 60°F H20 (l)
Input
Sulfur(s)
(20) Air
Lb.
22,100
Mols
689.5
Temp. °F
60
02
N2
H20 (v)
Sub-Total
Heat Required
22,050
72,650
500
95,200
to Melt S and
689.
2,593.
29
3,312
Preheat
5
5
to
495
495
495
280°F
Heat Release by Combustion of Sulfur
(689.5 mols x 127,690 Btu/mol)
AH
3.044 Btu/mol
22,565 Btu/mol
AH
MM Btu
Totals
117,300
9.99
0.66
10.65
1.39
88.04
100.08
Output
Gas to Absorption Tower
S02 44,150 689.5 400
N2 72,650 2,593.5 400
H20 (v) 500 29 400
Sub-Total 117,300 3,312
Heat Transfer to B.F.W. Preheater
Heat Loss
Totals
117,300
3487 Btu/mol
2258 Btu/mol
1213.6 Btu/lb
2.40
5.86
0.61
8.87
80.13
11.08
100.08
Note: Circled numbers refer to Stream Number on Consol Dwg. AF-3424.
Temperature/Pressure may differ depending on location of
identifying flag.
- 33 -
-------
Table 9
Mass and Heat Balance
S(>2 Absorption Tower
Basis: 1 hour
Datum: 60°F H20 (l)
Lb Mols
Input
Gas from Sulfur Combustor
Heat of Reaction
S02 + H20 = H2S03 (aq.)
(682.5 mols x 13,720 Btu/mol)
Totals
Out put
(22) Vent Gas
..-. S02
'N2
H2O (v)
Sub-Total
Liquid
849,900
H20 (1)
Sub-Total
450
72,650
150
73,250
56,000
720,650
776,650
7
2,593.5
9
2,609.5
682.5
40,001.5
40,684
Heat Rejected to Cooling Water
Totals 849,900
Temp.
OF
S02
N2
H20 (v)
Sub-Total
Water (l)
44,150
72,650
500
117,300
732,600
689.5
2,593.5
29
3,312
-
400
400
400
90
90
90
90
101
101
AH or Cp
3487 Btu/mol
2258
1213.6 Btu/lb
29.93 Btu/lb
285 Btu/mol
194 "
1072.8 Btu/lb
AH
MM Btu
2.40
5.86
0.61
8.87
21.93
9.36
40.16
x
0.50
0. 16
0.66
1.0 Btu/lb-°F 31.84
7.66
40. 16
Note: Circled numbers refer to Stream Number on Consol Dwg. AF-3424.
Temperature/Pressure may differ depending on location of
identifying flag.
- 34 -
-------
Table 1 O
Ovor-nll Haas and Heat Balance
Basis: 1 hour
Datum: 60T, H20 I/)
CO
Ol
Elemental Balance, Lb
In
"C!
(s
@
Xlt
Gaslfler Gas
Cll,
H2
CO
C02
SH3
H,S
H20 (v)
Sub-Total
y^C03 CaC03
Inert*
Sub-Total
K20 (v)
Sub-Total
Kake-up CO", Cas
C02
K20 (v)
Sub-Total
Totals
t>itpjt
(6J Producer Gas
@
H2
CO
C02
K2
MI3
lf2S
H20
•oste Water <£>
Vent Gas - SO, At
SO,
"2
II20 (v)
Sub-Total
Sensible Heat Tn
Sensible Heat to
Heat of Reaction
Heat Losi
Totals
Lb
47,950
70.350
1, 170,250
713,400
2,700,100
9, 150
47,950
376,850
5, 136,000
19,350
1.800
21,150
22,050
72,650
500
95,200
60,100
4.250
64.350
H2S +
H2S -
5,316.700
47,050
72,100
1,146,400
810.950
2,700,100
9,150
1,450
386,100
5,174,200
19,350
1.800
21.150
43,300
4,600
isorbor
450
72, 650
160
73,250
asforred to Gaa
Cooling Water
CO » HjO • CO2
5,316,700
M.W Hols
16.04 2,990
2.016 34,899
28.01 41,779
44.01 16,210
28.02 96,364
17.03 539
34.08 1,407
18.016 20,917
215, 106
184.42 105
32.00 689.5
28.02 2,593.5
18.016 29.
3,312
44.01 1,365
18.016 236
1,601
1/2 02 » S » H20
3/2 02 = S02 -i- H20
16.04 2,990
2.016 35,750
28.01 40,928
44.01 18,427
28.02 96,364
17.03 539
34.08 42
18.016 21,431
216.471
184. 42 105
32.06 1,338
18.016
64.06 7
28.02 2,393.3
18.016 9
2,609.3
If ler Steam
» HI (851 nols x
Moll
1.39
16.23
19.42
7.54
44.80
0.25
0.65
9.72
100.0
20.8
78.3
0.9
100.0
85.3
14.7
H C N
12,050 35,900 X
70,350 X X
X 501,750 X
X 194,700 X
x X 2,700,100
1.600 X 7,550
2,850 X X
42,200 X X
X 2,500 x
XXX
x 2,500 X
XX X
X X - 72,650
50 X X
50 X 72,650
X 16,400 X
500 X X
100.0 500 16,400 X
(1.358 mols H2S X 113,920 Btu/mol)
(7 mols H2S X 242,080 Btu/mol)
1.38
16.51
18.91
8.51
44.52
0.25
0.02
9.90
100.0
0.27
89.39
0.34
129,600 751.25O 2,780.300
12,050 35,900 x
72,100 x X
X 491,550 X
X 221,300 X
X X 2,700.100
1,600 x 7,530
100 x X
43.200 x . X
129,050 748,750 2,707,650
X 2,500 x
XX X
X 2,500 x
XXX
530 x x
XXX
X X 72,630
XXX
100. Ox X 72, 630
1,430 Btu/Dol)
-
129,600- 731,250 2,780,300
0 8 Ash
X X
X X
668,500 X
518,700 X
X X
X X
X 45,100
334,650 x
8, 400 X X
X X 1.800
8 , 40O X 1 , 800
22,050 X X
X XX
450 X X .
22,500 X X
43,700 X X
3, 750 X X
47, 450 X X
1,600,200 43,100 1,800
X XX
X XX
654,850 X X
589.650 X X
X XX
X . X X
x 1,330 x
342.900 X x
1,587,400 1,350 x
8,400 x x
X X 1,800
6,400 X 1,800
X 43, 500 x
4,050 X X
200 230 x
X XX
130 X X
330 250 x
1,600,200 45,100 1,800
TenD
KcOCa «P
X 1750
X 1750
X 1750
X 1750
X 1750
X 1750
X 1750
X 1750
X
8,450 60
X 60
8,450
X 60
X 60
X 60
X
X 140
X 140
x
8,450
X 1300
X 1300
X 1300
X 1300
X 1300
X 1300
X 1300
X 1300
x
8,430 120
X 120
8,450
X 310
I 90
X 90
X 90
X 90
x
8,450
Higher Heat log
CM Value
<1H or Cp
24,126 Btu/BO
11,643 Btu/mo
12,924 Btu/oo
19,925 Btu/BO
12,515 Btu/BO
24,126 Btu/mo
16,812 Btu/BO
34,492 Btu/BO
19,095 Btu/nol
736 Btu/ool
19,708 Btu/.ol
16,014 Btu/mol
8,533 Btu/mol
9,228 Btu/mol
13.955 Dtu/mol
8.870 Btu/=ol
16,O14 Btu/mol
11,733 Btu/mol
29,932 Btu/mol
40.3 Btu/mol °T
0.21 Btu/lb *T
0.2 Btu/lb °F
29.93 Btu/lb
283 Btu/ool
194 Btu/BOl
19,328 Btu/aol
KM Btu Btu/lb
72.1 23,878
406.3 61,100
540.0 4,347
323.0 X
1,206.0 X
• 13.0 9,680
23.7 7,100
721.5 X
3,305.6
X ~
X
X
x • x
X X
0.5 X
0.5
1.0 X
4.6 X
5.6
31.3
134.7
1.7
3,499.4
305.1 61,100
377.7 4,347
257.1 x
854.8 x
8.6 9,680
0.5 7,100
641.5 x
2,493.2
C.3
0.3
2.2
0.1 X
X X
0.3 s
0.2 X
0.7
834.1
108.4
1.1
.9.1
3,499.4
tai/stu
1,145.1
4,298.8
5,087.0
X
X
88.9
340.3
X
10.960.
X
X
X
X
X
4,403.6
4,983.4
X
X
88.9
10.1
x
10,631.2
X
X
I
X
X
4
Kote: Circled ouben refer to Stren. «u»ber on Conaol D.«. A7-3424. To.peroture/Pr«aure uy differ dopsndlnf on location of Identifying flag.
-------
This defines, in dollar cost, the incentive to develop the hot clean-up tech-
nology. Both these cases are based on a regenerable sulfur acceptor as outlined
in the section on experimental results. The economics of a once-through
acceptor system are briefly explored in Case III.
In any case, the producer gas after expansion to 10 psig is delivered to the
power station as preheated gas (540°F to 660°F). This sensible heat content
serves to improve the efficiency (or the heat rate) of the station proper. In
a conventional gas-burning station, the fuel gas is delivered at ambient tempera-
ture, and sensible heat from the products of combustion is recovered (via com-
bustion air preheat) to an air heater exit temperature of about 300°F. It seems
most realistic to credit the process with the excess sensible heat in the pre-
heated producer gas above 300°F as an approximate means of measuring the
improvement in the power station efficiency.
The sulfur removal facilities are arbitrarily identified as Section 300 (con-
sistent with the numerical identification on Dwg. No. AF-3424), and the sulfur
recovery facilities as Section 400. A conventional "hot pot" system to supply
the required make-up C02 is labeled Section 500 although a detailed system
design is not necessary.
The equipment required for the Venturi scrubbing system and the gas-gas heat
exchangers are shown in Figure 3 and is identified as Section 600.
Plant and Operating Cost Estimates
Based on the experimental data obtained to date, a complete set of heat and
material balances was developed for the entire process. These were discussed
previously and are shown in Tables 4 through 10, inclusive. From these balances,
a process flow sheet was developed as shown in Figures 3 and 4. A preliminary
capital cost estimate was developed by the Consol Engineering Department for
Sections 300, 400 and 600, and is summarized in tabular form in Appendix A as
Tables A-l through A-3, inclusive. Note that Section 500, the make-up C02 system
was assumed to be a conventional "hot pot" system. A detailed estimate was not
required since the costs and utility requirements are adequately defined in the
literature.(12)
The capital cost estimates developed in Appendix A are defined as inside battery
limits (ISBL). To these costs must be added off-site facilities (OSBL) which
include utility costs, electric substations, cooling water towers, distribution
piping, and boiler feed water treating. The accuracy of the capital cost esti-
mates in this study is estimated as ± 20$.
- 36 -
-------
It was required in this study to anticipate the escalation in these costs to a
presumed start-up date of January, 1978. The economic climate at present defies
a logical derivation of these factors. This work is based on these arbitrary
assumptions:
1. July, 1973 costs as a base point.
2. Begin design and construction, January, 1974.
3. Begin operation, January, 1978.
4. Escalation at 7-1/2$ per year average of construction
labor and material.
5. Interest during construction at 7-1/2$ simple interest
on cash flow.
6. Escalation of direct operating labor at 5-1/2% per year.
7. Interest on working capital at 7-1/2$ per year.
The total investment summary for the two cases described above is shown in Table
11. The installed plant cost (present day cost) for the "hot" gas clean-up
scheme (Case l), is $34,800,000. The addition of product gas wet scrubbing with
reheat adds $8,300,000 to the cost for a total of $43,100,000. Escalation of
these costs to January, 1978 (as defined above) increases the total investment
to $48,300,000 and $59,800,000, respectively. Most (approximately 70$) of the
incremental cost of adding wet scrubbing with gas reheat is associated with the
large gas-gas heat exchangers required.
Direct operating costs for these two cases were estimated as shown in Table 12.
Note, that an average plant operating factor of 70$ was assumed reflecting power
station practice. The unit values used to develop operating costs are shown
directly in Table 12.
Evaluation of System Economics
The previous report on this project'1) clearly showed that the most economic
application of this process is in combination with a combined cycle power station.
This current work is more restricted in scope and is limited to a simple exposi-
tion of the costs of desulfurizing the pressurized producer gas. Primarily the
purpose is to focus attention on those areas where additional research and
development work will be most valuable.
Based on the cost estimates presented in the above sections, the system costs
are estimated as shown in Table 13. The net annual operating costs for Case I
(hot cleanup) are $13,810,000 per year; for Case II (wet scrubbing with gas
reheat), the costs are $17,081,000 per year. Approximately two-thirds of this
cost is represented by capital charges. Note, that the low value for sulfur
credit ($8.00/long ton) reflects a conservative view of the sulfur market.
- 37 -
-------
Table 11
CO
00
Investment Summary
Case
300
Sulfur
Plant Section Removal
Utilities Required
Electricity, KW 2,490
Cooling Water, gpm . 1,870
Low-pressure Steam, Ib/hr x
Boiler Feed Water, gpm X
Operating Labor Required
Ken/Shift 4
Investment
Erected Cost ( ISBL) 15,000
Off-sites i Utilities (OSBU
on-sites 1,200
Electrical 200
Cooling Water x
Boiler Feed Water x
V.-.vai OKijL 1,400
2nril,,illC(l i'lnnv. Cost
(July, 1973) 16,400
Escalation to Jan. 1978
Sub-Totnl
Interest during Construction
Total Investment
(Jan., 1978)
400
Sulfur
Recovery
10,030
. 10,340
x
1,234
3
11,400
900
700
•400
300
2,300
13,700
I
500
Make-up
C02
System
344
16,185
78,300
x
1
(in
3,900
200
X
600
X
800
4,700
Total
12,864
28,395
. 78,300
1,234
8
$1000)
30,300
2,300
900
1,000
300
4,500
34,800
6.500
41,300
7.000
48,300
II
600
Product Gas
Wet Scrubbing
with Reheat
512
34,000
x
X
1
6,600
500
x
1,200
X
1,700
8,300
Total
13,376
62,395
78,300
1,234
9
36,900
2,800
900
2,200
300
6,200
43,100
8,100
51,200
8.600
59,800
-------
Table 12
Direct Operating Cost Summary
Excluding Acceptor Cost
CO
ID
Plant Section
Direct Operating Labor
Ken/Shift
Direct Opc-rQtin^ Cost
1. Operating Labor at
&S3, 200/Mn/shif t,'yr
2. Maintenance Labor at
1.6% Installed Plant Cost
3. Direct Supervision
157. of 1 + 2
4. Indirect Overhead
SOT. of 1 + 2 + 3
5. Payroll Overhead
IGTo oT 1 +2 -*- 3 + -1
6. Maintenance Material
2.4% Installed Plant Cost
7. Miscellaneous Supplies
15% Ma int. Material
8. Utilities
Electricity at 9 mllls/KWH
Cooling Water ot 3£/M gal.
BFW at 30?/M Ral.
Low-Pressure Steam
at 75C/M Ib
Chemicals Ct Catalysts
(ex Acceptor)
Sub-Total Utilities
Totals
Escalation to Jan., 1978(1>
Total Direct Operating Cost (Jan.
Basis: 7O% Plant Operating Factor
I II
500 600
300 400 Make-up Product Gas
Sulfur Sulfur CO2 Wet Scrubbing
Removal Recovery System Total with Reheat
4318 1
(in SlOOO/yr)
253 190 63 506 63
262 22O 75 557 133
77 61 21 159 30
296 235 . 80 611 113
133 106 36 275 51
394 328 113 835 199
59 49 17 125 30
138 553 19 710 28
21 114 179 314 375
X 136 X 136 X
X x 360 36O X
XX 13 13 X
159 803 571 1,933 403
1 , 633 1,992 976 4,601 1,022
720
an. 1978) 5,321
Total
9
569
690
189
724
326
1,034
155
738
689
136
360
13
1,936
5,623
877
6,500
(1) Items 2-7 are Investment sensitive and are escalated on the
same basis as plant investment. Items 1 end 8 are at assumed
January 1978 values.
-------
Table 13
Economic Evaluation
Regenerative Acceptor Desulfurization Process
I
O
Basis: 70% Plant Operating Factor (6132 hr/yr)
Case
Method of Producer Gas Clean-up
(Particulate and Alkali)
Coal Required:
Tons/yr (6% Moisture)
Higher Heating Value, MM Btu/yr
Desulfurized Producer Gas to Station
Mols/Hr
Temperature
Pressure
Higher Heating Value, MM Btu/yr
HHV + Sensible Heat Content, MM Btu/yr(1)
Cost Analysis
Installed Plant Cost (1973)
Escalation to 1978
Interest during Construction
Total Investment
Working Capital '
Annual Operating Costs
Direct Operating Cost (1978 Basis)
Acf.opi.oi- at SlO/ion
Interest on v.'orkir.;; Capital at 7.5%
Capital Charges at 18% Investment
Sulfur Credit at SB/metric ton
Not Annual Operating Cost
I
Hot
213,701
660°F
10 psig
65.19 X 1012
68.68 * 1012
$34.8 MM
6.5 MM
7.0 MM
$48.3 MM
$ 1.33 MM
$ 5.321 MM/Year
0. 648 MM/Year
0.100 Mil/Year
8.694 MM/Year
(S 0.953 MM/Year)
513.810 MM/Year
-3,422,000
-81.71 x 106
II
Wet Scrubbing
with Reheat
213,214
54O°F
10 psig
64.70 x 1O12
,12
66.99 x 10
$43.1 MM
8. 1 MM
8.6 MM
$59.8 MM
$ 1.63 MM
$ 6.500 MM/Year
O. G-18 MM/Year
0. 122 Mil/Year
10.764 MM/Year
($ 0.953 MM/Year)
$17.081 MM/Year
Desulf urizat ion Cost Expressed;
In teriiis of feed coal
S/ton coal
c/MM Btu HHV
4.03
16.9
4.98
20.9
In terms of product gas to station:
f/KM Btu HHV
-------
In the cases where the acceptor is regenerated, the make-up acceptor cost has
little impact on the system cost. Since this study does not define a specific
plant site, a definitive cost for delivered acceptor cannot be derived. An
arbitrary cost of $10.00 per ton of raw stone (MgC03CaC03) was assumed. This
somewhat high cost is based on two facts: a) the laboratory studies to date have
shown that some dolomitic stones cannot be used in the process (excessive attri-
tion) thereby limiting the potential supply sources, and b) transportation costs
for raw materials are substantially increasing. However, even with an acceptor
make-up cost of $10.00 per ton, the annual cost is only 4-5$ of the total system
cost. The imprecision of the acceptor cost has only a minor influence on the
evaluation of the total system.
The system costs presented in Table 13 are more meaningful if related to either
the coal required for the system, or to the quantity and quality of the gas
delivered to the power station. The net annual operating costs for Case I are
equivalent to $4.03 per ton of feed coal; for Case II, $4.98 per ton. Expressed
in terms of heating value of the coal, the corresponding costs are 16.9 and
20.9^/MM Btu, respectively.
Considering only the higher heating value of the product gas to the power station,
the corresponding costs are 21.2 and 26.4^/MM Btu for Cases I and II, respectively.
Allowing for the sensible heat credit discussed earlier, the costs are reduced
to 20.1 and 25.4^/MM Btu, respectively. Note, that the sensible heat credit
reduces the process cost by about 4-5$. The addition of wet scrubbing and gas
reheat increases the processing cost by 5.3^/MM Btu or 2671.
As discussed earlier, another credit to the process is the excess power generated
by the expanding turbine-generator set. For Case I this amounts to 160 MW; for
Case II, 118 MW. As noted previously, it is beyond the scope of this contract
.to evaluate the dollar credit associated with this excess power. However, the
impact of the credit upon the processing costs can be at least qualitatively
defined. The real cost of electric power generated by an expanding turbine-
generator set would normally be less than steam-generated power. If the real
cost were one mill/KWH less, then the credit to the process would be for Case I,
1.5^/MM Btu, and for Case II, l.ljzf/MM Btu. These are significant credits, but
could only be evaluated with accuracy through a detailed study.
Non-Regenerable Acceptor
The basic development of this project to date has been devoted to the concept of
a regenerable sulfur acceptor. The feasibility of this concept has been discus-
sed earlier in the review of the experimental data. As pointed out in the
previous section, modest (1-2$) make-up acceptor requirements in the regenerative
case do not significantly influence the process costs. However, this would not
be true for a non-regenerable or once-through system. In a once-through acceptor
system, approximately fifteen times as much raw stone is required. This is
shown in Table 14.
In the once-through acceptor system, the recovery of sulfur would be somewhat
simplified. All of the sulfur removed from the producer gas would be recovered
by stripping the spent acceptor with water and a C02-bearing gas stream. The
acid gas from this system would then be fed to a standard Claus plant for
recovery of elemental sulfur. However, the spent acceptor stripping system would
have to be expanded to handle a fifteen-fold increase in spent acceptor flow, and
the make-up CO2 requirements (to convert MgOCaS to MgC03CaCO3) would also increase
fifteen-fold.
- 41 -
-------
Table 14
Acceptor Requirements
Regenerative vs. Once-Through
Basis: 70$ Plant Operating Factor (6132 hours/year)
Case
Method
Regenerative
Acceptor
Hot Gas Cleanup
III
Once-Through
Acceptor
Hot Gas Cleanup
Coal Required:
Lb/hr (6$ Moisture)
Tons per year (6$ Moisture)
Higher Heating Value, MM Btu/Year
- 1,116,150 -
- 3,422,000 —
81.71 x 1012
Acceptor Required:
Mols Ca Circulating/Hour
Mols Make-up Ca/Hour
Make-up Acceptor (MgC03CaC03):
Lb/Hr
Tons/Year
10,504
105
21,150
64,850
1,598
1,598
321,900
986,900
Annual Cost of Make-up Acceptor
at $10/Ton
$648,500
), 869,000
- 42 -
-------
The maximum savings would accrue if a naturally C02-rich stream could be used
which would obviate the need for a C02 recovery step. Power station stack gas
with a C02 concentration of about 15 volume percent is a speculative possibility
here. The maximum potential savings in investment on this assumption is
approximated as about $10 MM. This is detailed in Table 15 below:
Table 15
Approximate Investment Reduction
Once-Through Acceptor vs. Regenerative
Case I III
Regenerative Acceptor Once-Through Acceptor
Hot Gas Cleanup Hot Gas Cleanup
Installed Cost by Section (Table 11)
Section 300 - Sulfur Removal $ 16.4 MM $ 20.5 MM
Section 400 - Sulfur Recovery $ 13.7 MM $ 4.2 MM(X)
Section 500 - Make-up C02 $ 4.7 MM $ 3.1 MM(2)
Total $ 34.8 MM $ 27.8 MM
Escalation & Interest During
Construction $ 13.5 MM $ 10.8 MM
$ 48.3 MM $ 38.6 MM
(i) Split-flow Claus plant.
(2) Compression cost only - No C02 absorption step.
At an 18fo capital charge, the annual saving from the investment reduction of about
$10 MM would be $1,800,000 per year as opposed to an annual increase in acceptor
make-up cost of about $9,200,000 per year. On this basis we conclude that there
is no incentive to explore further the once-through acceptor case.
This conclusion is based on the requirement that the MgOCaS from the process
must be converted to MgC03CaC03 for disposal. Alternatively, the MgOCaS might
be converted to MgOCaS04 by high-temperature oxidation. However, our limited
data show that the spent CaS is difficult to oxidize and that S02 is formed.
Disposal of CaS04 may create a water pollution problem in some areas.
- 43 -
-------
IV. EQUIPMENT AND PROCEDURE
A. Description of the Continuous Unit
1. Process Piping and Equipment
A simplified flow diagram of the continuous acceptor reaction loop is shown in
Figure 5. A more detailed process and instrumentation diagram is shown in
Figure 6. These figures show the unit as arranged for operation where the
acceptor is recirculated continuously between the gas desulfurizer and regener-
ator reactor. The following description applies to the above system. For
operation of a single vessel, the equipment and procedure, in general, are the
same as described here. The differences which are entailed in operation of a
single vessel are detailed in part C of this section.
2. Process Vessels
The D-2 gas desulfurizer vessel, as shown in Figure 7, is three inches in
diameter. The acceptor bed height was maintained 18 inches by means of a d/P
level controller.
The D-l regenerator vessel, shown in Figure 8, is two inches in diameter, and
the bed height was maintained at 46 inches. The upper portion of the vessel
was enlarged to 4 inches to allow for additional internals above the bed.
Heat was supplied through the walls of the process vessels. The regenerator
heater contained three separate circuits which corresponded to the bottom, middle
and top zones of the reactor. The gas desulfurizer was equipped with four heater
circuits spaced along the length of the reactor. All the regenerator heaters and
the top zone gas desulfurizer heater were controlled by 270 volt Powerstats
supplying 5.1 KW of electrical heat. The three lower heaters on the gas desulfu-
rizer were controlled by 230 volt Powerstats supplying 3.8 KW.
The process vessels and their heaters were contained in pressure shells which
were pressurized with C02, so that the pressure drops across the walls of the
process vessels are zero. Both pressure shells were detailed earlier.!1)
Johns-Manville Sil-0-Cel C~3 insulation was used to fill the annular space between
the heaters and the wall of the gas desulfurizer pressure shell. The gas desulfu-
rizer pressure shell wall temperature was held below 950°F. The annular space
between the heaters and wall of the regenerator pressure shell was filled with
Carborundum "Fiberfrax" insulation. The regenerator pressure shell wall tempera-
ture was held below 7OO°F.
3. Solids Handling
Acceptor was handled in three streams, regenerated acceptor from the regenerator,
regenerated acceptor to the gas desulfurizer, and sulfided acceptor to the
regenerator.
The regenerated stone was fed to the top of the gas desulfurizer bed in an
electrically heated pneumatic transfer line. The sulfided stone was drained from
the bottom of the gas desulfurizer into a purged standleg L-5, which formed the
seal between the gas desulfurizer and regenerator.
- 45 -
-------
The sulfided acceptor entered the top flange of the regenerator, and passed
through a preheater purged with C02 which was introduced into the acceptor return
line above the vessel. The preheater was a stainless steel tube, 7/8" OD x 3/4"
ID x 10" long having nine baffles installed at 60° angles.
Acceptor from the regenerator was withdrawn through a standleg into one of two
lockhoppers, F-6A and F-6B, at a rate dictated by the level of the bed. The
acceptor was removed from the off-stream lockhopper and charged back to one of
two acceptor feed hoppers, F-4A or F-4B. Continuity of acceptor circulation was
maintained by feeding acceptor to the regenerator alternately from the two feed
hoppers. A sight glass was installed below the feed hoppers to show when the on-
stream feed hopper was empty. The feed hoppers and withdrawal hoppers each had
a capacity of six pounds of raw acceptor.
A sampling valve was inserted in the L-5 acceptor return line through which a
sample of sulfided acceptor could be diverted into a sampling cylinder while the
unit was operating.
Sulfided acceptor and acceptor from the regenerator were conveyed by gravity flow
in purged standlegs. These standlegs were purged with inert gas to prevent steam
from either vessel from getting in and condensing in a cold spot. The solids
flow rates were controlled to maintain the standlegs full of solids. To protect
the Teflon seats in the valves used to control the solids flow rates, water-
jacketed coolers were installed in each of the standlegs. These coolers were
detailed earlier.!1^3) Sight glasses (jerguson 12-T-2O) were also installed in
each standleg to allow visual confirmation of the amount and nature of solids
flow.
4-. Gas Flows
The gas desulfurizer inlet gas consisted of recycled product gas with N2, H2S, H2,
C02, and steam flows added to give the desired gas composition.
The metered dry desulfurizer fluidizing gas was passed through the steam generator
which is shown in Figure 9. The water temperature in the steam generator was
controlled (± 0.5°F) to give the desired steam partial pressure. Calibrations
showed that at water temperatures we used, the steam partial pressure was equal to
the equilibrium vapor pressure of water. The steam generator feed water was
passed through a Barnstead demineralizer.
The fluidizing gas passed through an electrically heated line from the steam
generator to the top of the desulfurizer. The wall temperature in the line was
kept above 40O°F to prevent condensation of the steam. The gas inside the gasi-
fier vessel passed through a helically-wound preheat coil and an axial dip tube
which extended into a cone at the bottom of the vessel. The gas passed downward,
reversed direction, and fluidized the bed. The pneumatic lift gas entered the top
of the vessel, disengaged from solids, and exited with the fluidizing gas.
The desulfurizer outlet gas passed through one of two parallel filters where
solids carried over from the gasifier bed were removed using F-porosity Micro-
Metallic filters. Both of the filters and the upstream piping were heated elec-
trically to prevent condensation of steam in the outlet gas. Separate heating
circuits were provided for each of the filters so that they could be cooled
before being removed for cleaning and heated rapidly after being reinstalled.
- 46 -
-------
Steam was condensed in the C-7 cooler and collected in the F-7 condensate
receiver. These are detailed in Figures 10 and 11, respectively. The gas
was further cooled in the C-6 cooler using chilled water at 40°F.
The dry, solids-free desulfurizer product gas was vented to atmospheric pressure
through the back pressure control valve DPCV-1.
The regenerator inlet gas consisted of C02 and steam with air or hydrogen some-
times added. The gas entered the top of the regenerator and passed through a
helically-wound preheat coil and an axial dip tube which extended into the cone
at the bottom of the vessel. The gas passed downward, reversed direction, and
fluidized the bed. In the electrically heated exit piping} solids were removed
in one of two parallel filters which used F-porosity Micro-Metallic filter
elements.
The unreacted steam was condensed in the C-l cooler. The condensed steam was
collected in the condensate receiver F-8 which was mounted below C-l. A liquid
level controller actuated a control valve LCV-5 to drain condensate and maintain
a constant liquid level in the receiver.
The gas was further cooled in the C-4 cooler using water supplied at 40°F from a
water chilling unit. Additional condensate was disengaged in knock-out pot F-9
and collected in the sight glass below F-9.
The dry, solids-free product gas was vented to atmospheric pressure through the
regenerator back pressure control valve PCV-1.
Small metered flows of N2 were used to purge the pressure taps at the tops of
both vessels as well as the L-6 standleg in the acceptor return line while the L-5
stantjleg was purge-d with C02 . All of the purge flows were as small as possible
relative to the fluidizing gas flows.
The'H2, CO2, and N2 used as process gas, pressure shell balance gas, and lock-
hopper pressurizing gas were supplied from cylinders. Regulators downstream of
the cylinder manifolds fixed the supply pressures at 300 psig. Air was supplied
from a compressor. Water and oil were removed by a knock-out drum followed by a
bed of silica gel. C.P. grade hydrogen sulfide was supplied from cylinders. At
15 atmospheres, hydrogen sulfide is close to its liquid point, and therefore can-
not be fed reliably as a gas. Therefore, the cylinder of liquid H2S was turned
upside down so that the H2S could be fed as a liquid. The cylinder was kept in a
compartment thermostated at 90°F to give about 300 psig over the liquid. The H2S
feed system is shown schematically in Figure 12.
Identical single-stage diaphragm compressors (Pressure Products Industries,
Model No. 1073) were used to recycle product gas.
5. Miscellaneous Piping
Wherever possible, connections to the process piping were high-pressure coned
fittings (Autoclave Engineers or Pressure Products) which were used because of
their ruggedness, ease of disassembly, and freedom from leaks.
- 47 -
-------
Special adapter fittings were required to connect the solids outlet tubes
(3/8" x .035" wall) to the standleg piping (9/16" OD x 5/16" ID). The solids
outlet tubes from the vessels slipped inside the adapter fittings and were
sealed using a lava sealant and packing gland. A standard high-pressure
connection was made between the adapter fitting and the standleg.
In order to minimize the pressure drop across the gas outlet piping, 3/8" x .028"
wall tubing was used downstream of C-l and C-6 coolers in the gasifier and
regenerator, respectively.
The steam condensate contained dissolved NH3, S02, H2S, and C02. Types 304 and
316 stainless steel gave excellent service in piping and equipment handling
process gases at temperatures from ambient to 400°F. Previous experience had
shown that condensate and wet process gases are extremely corrosive to copper,
brass and plain carbon steel. These materials were not used, even in the instru-
mentation piping.
Conax packing glands and lava sealants were used where the heater leadwires,
solids outlet tubes, and the vessel skin thermocouples passed through the
pressure shells.
6. Pressure Measurement and Control
The system pressure was controlled at the top of the regenerator vessel using a
Foxboro pressure cell and recorder-controller which operated the back pressure
control valve PCV-1. The pressure at the top of the gas desulfurizer was main-
tained at a fixed differential (usually in the range of zero to five inches H20)
from that at the top of the regenerator using a Foxboro d/P cell, and a recorder-
controller which operated the desulfurizer back pressure control valve DPCV-1.
The discharge pressures of the two recycle compressors were controlled using
Foxboro indicator controllers which operated control valves PCV-2 and PCV-3 and
which vented to the suction side of the compressors.
Identical balance gas systems were used to maintain a zero pressure differential
across the walls of both process vessels. C02 was supplied to the pressure shell
at 300 psig. Barton d/P cells and recorder-controllers operated control valves
DPCV-11A and 11B and DPCV-13A and 13B which either loaded the shell or vented
C02 to the atmosphere to maintain a zero pressure differential between the pres-
sure shell and the process vessel.
The unit was adequately instrumented to monitor pressure drops across the various
transfer lines, standlegs, and fluidized beds. These pressure drops were
measured with respect to the pressures at the top of the gas desulfurizer or
regenerator vessels, and were continuously recorded.
The pressure drops across the solids filters in both the gas desulfurizer and
regenerator outlet gas piping were continuously recorded. The filters were
switched when the pressure drop across the on-stream filter reached 75" of H20.
Panel mounted pressure gauges were used to indicate the pressures at the top of
both vessels, the pressure shell pressures, the compressor discharge pressure and
the pressures immediately upstream of the back pressure control valves. Locally
mounted pressure gauges were used to indicate the pressures in the steam generators
and the various lockhoppers.
- 48 -
-------
7. Temperature Measurement and Control
The bed temperature in both the gas desulfurizer and regenerator were measured
using chromel-alumel thermocouples contained in wells which were immersed in the
fluidized beds. Two thermocouples were used in the regenerator thermowell and
one in the gas desulfurizer.
The bed temperatures were controlled manually by adjusting the Powerstats which
controlled the electrical heat input from the heaters.
In each vessel, the wall temperatures corresponding to each of the heating zones
were measured using thermocouples which were inserted through the side of the
pressure shell and through slots ground in the heater ceramics.
The water temperature in each steam generator was monitored using an Iron-
Constantan thermocouple contained in an axial thermowell. The thermocouple
signal was recorded using a Foxboro temperature recorder-controller. A high or
low temperature activated relay which supplied power to the steam generator
bayonet heater from the appropriate one of two 7.5 KW Powerstats which were set
by trial to supply slightly more and slightly less than the required steady-state
input.
Up to 28 temperatures were monitored of which 26 could be continuously recorded.
The other temperatures included the wall temperatures of the pressure shells,
the transfer lines, the inlet and outlet gas piping, the steam generators, and
solids filters. Thermocouples inserted in the sight glasses in the various
standlegs monitored the temperatures of the solids flowing from the water-
jacketed coolers.
..8. Control of Solids Flow Rates
The solids feed rates were controlled using rotary feeders which consisted of a
tapered Teflon plug rotating inside a stainless steel body. Drawings of the
feeders used in the continuous unit were shown earlier.!1'3) The number, dia-
meter, and depth of the pockets were chosen on the basis of the bulk density of
the solids and the desired range of solids flow rates. Variation of the solids
flow rates over the desired range was accomplished by means of a Graham variable
speed drive. Slots were milled in the feeder body around the inlet and outlet
ports to allow the pockets to begin to fill and discharge before reaching top and
bottom dead center. Since Teflon was used, no lubrication was required between
the feeder body and rotor.
The flows of solids in the L-5 and L~6 standlegs were controlled by air-operated
Contromatics ball valves having a 7/16" bore. The opening and closing of the
valves was determined by the level control system as discussed below.
9. Solids Level Control
Level control in both vessels was regulated by means of differential pressure
measurements. In the gas desulfurizer, a d/P measurement was made between the
top of the vessel and a point two inches below the desired bed height. The
corresponding points in the regenerator were the top of the vessel and a point
four inches below the desired bed height.
- 49 -
-------
At our operating conditions each inch of coverage of the bed pressure probe
produced about 3/4" of water of pressure drop. The pressure drop was changed
into a 3-15 psi signal through a pressure transducer. The latter signal in
turn operated pressure switches connected to solenoid selector valves which
supplied air at one of three preset pressures to a Conoflow cylinder operator.
This adjusted the opening of the ball valve in the appropriate standleg.
The operation of both the L-5 and L-6 valves was monitored continuously by
recording the air pressure supply to the Conoflow cylinder operator. The height
of bed above the pressure tap was monitored by recording the output pressure of
the pressure transducers.
10. Gas Flow Measurements
The inlet gas flows were metered using rotameters which were calibrated with
either a Rockwell dry gas meter or a wet test meter depending on the range of the
rotameter. The gas meters were calibrated against a volume meter using displace-
ment of a mercury sealed piston accurate to 0.2$.
Rotameters using recycle gas were calibrated using two gases having different
molecular weights. The flow rates were correlated with the rotameter float
heights using the following general equation;
o - r u. ci x s ^ C2 x S2
w — \sr
-o (MW) '5 (MW)-5
Where: Q = flow rate, SCFH.
S = float height.
MW = molecular weight of the recycle gas.
..'." C0, C1} C2 = constants for a given rotameter determined via
regression analysis.
The recycle rotameters were calibrated with H2 and N2 at a supply pressure of
300 psig.
The dry product gas flow rates from both vessels were measured using calibrated
Rockwell dry gas meters located downstream of the back pressure control valves.
11. Gas Analysis
The product gas streams from both vessels were analyzed using a gas chromatograph.
During a run, low levels of CO, C02, and S02 were continuously monitored with
infrared analyzers. High levels of C02 were monitored with a thermal conductivity
cell with reference to nitrogen. A paramagnetic susceptibility analyzer was used
to measure the concentration of oxygen. The outputs of the CO, C02, S02, and 02
analyzers and of the thermal conductivity cell all were continuously recorded.
Analyses were made on the gas chromatograph when the above analyzers indicated
that the gas composition was steady.
All of the gas analysis instruments were connected to manifolds so that offgas
from either the regenerator or gas desulfurizer could be fed to each. For inte-
grated operation, the chromatograph analyzed both the regenerator and gas desulfu-
rizer offgases. In addition, gas samples were taken into a stainless steel
cylinder for intensive chromatographic analysis by the analytical laboratory.
- 50 -
-------
Hydrogen sulfide was to be measured with an infrared analyzer, but the perform-
ance of the instrument was not satisfactory. Instead H2S from both vessels was
determined using a standard type of iodine method.(13)
12. Safety Features
The unit was equipped to give an audible alarm whenever a potentially dangerous
upset occurred. A light indicating the source of the upset accompanied an alarm.
In some situations, relays were activated which closed the emergency solenoid
valve to isolate the gas desulfurizer and regenerator vessels and shut off
appropriate solids feeders.
If steam were to condense in the L-5 standleg, the acceptor would become a solid
plug. During normal operations, cross-contamination was prevented by maintaining
the C02 purged standleg above the L-5 valve full of acceptor and by controlling
the pressure differential between the top of the gas desulfurizer and regenerator
vessels such that the pressure drop across the L-5 valve was zero. If an upset
occurred such that the pressure differential between the vessels deviated ± 10"
H20 from the set valve or the pressure drop across the L-5 valve deviated ± 5"
H20 from zero, an alarm sounded and the emergency solenoid valve closed, thereby
isolating the gas desulfurizer and regenerator vessels.
Recycle gas was used to transport acceptor to the gas desulfurizer and was a
major constituent of the fluidizing gases. Pressure switches were installed to
sound an alarm, close the emergency solenoid valve, and shut off the L-4 feeder
if the discharge pressure of the recycle gas compressor fell below a preset
value.
The supply of acceptor was monitored using alarms. High and low AP alarms were
installed to detect plugging or starvation in the acceptor transfer line.
The unit was equipped with a high AP alarm to detect plugging in the gas desulfu-
rizer inlet piping. Another alarm sounded in the event of malfunctioning of the
back pressure valve or of plugging in the gas desulfurizer outlet piping, as
indicated by a high pressure at the top of the vessel. Other alarms warned of
malfunction of the liquid discharge valve below the condensate receiver, as in-
dicated by a high or low condensate level.
Failure of the balance gas system to control pressure differential across the
walls of either process vessel could have resulted in collapse or rupture of the
vessel. An alarm was provided to indicate any deviation in the pressure differ-
ential between either process vessel and its pressure shell greater than 25" of
H20 and another alarm was used to indicate a low pressure in the C02 cylinder
manifold which supplied the balance gas.
Nitrogen was available to supply pressure to the instruments if the supply of
instrument air were lost, as shown by an alarm indicating low instrument air
supply pressure. If cooling water from the cooling tower were cut off, as in-
dicated by a low pressure alarm, water was available from the city water supply.
The H2S feed system was equipped with high and low pressure alarms and relief
valves which vented to an afterburner. During operations, the air was continu-
ally sampled for its H2S concentration with MSA ''sniffer" tubes.
- 51 -
-------
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
CONSOLIDATION COAL CO
MCSCAHCH DIVISION
UBRARY. PENNA.
-------
01
CO
Hq TTE-i
Tb_
V».u/
3.^ 1
*- J
H
A i J/> * /f
'
_= -.
CHILUR.
TJ
K),
Fl —
Tt —
P -
Pj» -
REVISIONS
To
CONSOLIDATION COAL CO.
RESEARCH ft DEVELOPMENT DIVISION
LIBRARY. PENNA.
CF-3411
-------
B. Operating Procedure for a Typical Two-Vessel Integrated Run
1. Prerun Procedure
The acceptor inventory was made up to the desired size consist. About 11,000
grams of raw acceptor was used in a typical run.
After bringing the gas desulfurizer and regenerator to the programmed tempera-
tures, the unit was pressured up with nitrogen to 15 atmospheres and was leak
tested.
2. Startup
With the unit at pressure, the recycle compressor was started and, after setting
the desired compressor discharge pressure, the programmed recycle flows to the
gas desulfurizer were established.
The programmed flows of C02, N2, and H2, were then put into the appropriate
vessels. The steam flow was established by diverting the fluidizing gas through
the steam generator. The Powerstats on the steam generator heater were adjusted
to control the programmed water temperature.
The acceptor feed hoppers were charged and feeding was begun at the programmed
rate. At this time the feeder was fine tuned to give the desired feed rate by
checking the time required to empty a hopper with a known weight of acceptor.
The emergency solenoid valve which isolated the process vessels was kept closed
until the acceptor filled the L-5 standleg and the gas desulfurizer vessel.
The entire acceptor inventory was circulated through the unit in order to
calcine the MgC03 component of the dolomite and to adjust the level controllers.
As the acceptor was circulated, the L-5 valve positions corresponding to normal
and low acceptor levels in the gas desulfurizer, and the L-6 valve setting
corresponding to high, normal and low levels in the regenerator were adjusted
until the optimum settings were obtained.
When an acceptor feed hopper became empty, the parallel hopper was switched on-
stream. Simultaneously, the acceptor withdrawal hoppers were switched. The
off-stream withdrawal hopper was isolated and emptied. After all of the raw
acceptor had been fed to the unit, the acceptor which had been removed from with-
drawal hoppers was combined. Two batches of the combined acceptor weighing about
2100 grams each were made up and charged one batch to each feed hopper. These
two batches constituted the initial external acceptor inventory for the run.
When all of the raw acceptor had been fed, the L-4 feeder was shut off. The gas
desulfurizer bed was presulfided by feeding H2S at the programmed rate long
enough to sulfide 20$ of the useful calcium content of the acceptor in the bed.
Acceptor circulation was then restarted, and this marked the beginning of the
run.
When acceptor circulation was restarted, the Powerstats controlling the electrical
heat input to both vessels were adjusted to give the desired bed temperature.
Axial temperature traverses were taken across both fluidized beds and the elec-
trical heat inputs to the heating zones were further adjusted to give a flat
temperature profile.
- 60 -
-------
3. Routine Run Procedure
Continuous acceptor circulation was maintained by feeding acceptor to the gas
desulfurizer alternately from the two acceptor feed hoppers. The acceptor feed
hoppers and regenerator withdrawal hoppers were switched simultaneously. The
off-stream withdrawal hopper was isolated from the unit and drained. The acceptor
removed from the withdrawal hoppers was charged to the off-stream acceptor
feed hopper. Riffled 20-gram samples were taken at specified times during the
run.
Samples were also periodically removed from the L-5 line during the run.
The concentrations of CO and C02 in the gas desulfurizer offgas were monitored
on infrared analyzers. C02 in the regenerator offgas was followed on the thermal
conductivity cell. During the course of the run, several samples from each
reactor were analyzed on the gas chromatograph. Iodine scrubbers were used to
check the level of H2S in the offgas from both reactors. During some runs, the
exit gases were sampled in a high pressure cylinder and sent to the analytical
laboratory for exhaustive chromatographic analysis.
The flasks which were used to collect the condensate drained from the condensate
receivers were replaced hourly, and the weight of the condensate determined. The
condensates collected in the sight glasses below the C-4 and C-6 coolers were
drained and weighed at least once per shift.
The solids filters in the gas desulfurizer and regenerator outlet gas piping were
changed when the AP across the on-stream filter was 75" H2O. After the filters
were switched, the off-stream filter was removed from the unit, cleaned, and
reinstalled. The contents of each filter were weighed and saved for analysis.
The rotameter settings and product gas meter readings were recorded hourly as
were the important pressures, pressure drops, temperatures and Powerstat settings.
4. Shutdown Procedures
At shutdown, the run was ended by simultaneously shutting off the H2S flow to the
gas desulfurizer, shutting off the acceptor feeder, L-4, and closing the emergency
solenoid valve between the vessels.
The steam and H2 flows were removed from both vessels as soon as possible after
the end of the run. Immediately thereafter, the contents of the regenerator
were withdrawn. The E-2 emergency valve was then reopened and the gas desulfu-
rizer bed was withdrawn through the regenerator. The recycle compressor was
then shut down and the unit was depressurized and allowed to cool.
C. Equipment and Procedure for Single-Vessel Operation
1. Pretreatment Operations
Only the reactor vessel, D-l, was used in the pretreatment studies. A simplified
flow diagram for this system is shown in Figure 13. The system was used for
both the Seeded Coal Process tests and for preoxidation runs. A bed of noncaking
char was fed to the reactor for start-up in both cases.
- 61 -
-------
CT;
to
FIGURE 13
FLOW niAGP.VM FOR COAL PHKTRKAT.VFNT
-P4-
Back Pressure
Control Valve
Tor
Receivers
D-l
Preoxidizcr
Coal
Hoppers
Char Char
i i
A A
Feeders
N3 Air "a
Recycle
Compressor
Product Char
Receivers
-------
Two reactors were used. No. 1 shown in Figure 14 was left from the previous
contract period and was equipped with a boot arrangement. This was changed to
No. 2 when the draft tube Configuration C was installed. The reactor used in
the acceptor sulfur cycle work is shown for comparison.
Coal and char feeds were put in through one outlet line, and product was with-
drawn through the other outlet'line. An overflow weir was used to maintain the
bed level. Tar receivers were connected to the system, and all of the analyzers
were connected to the pretreater vessel. This included the infrared S02 analyzer
and the paramagnetic oxygen analyzer which were not used for the acceptor studies.
Figures 15 and 16 detail the draft tube arrangements which were tied in the
Seeded Coal Process experiments. The reasons for choosing these particular
designs are elaborated upon in the Pretreatment Section of this report (Section
VIII).
Pretreater outlet gas generally contained some heavy tar and pitch which would
rapidly plug the solids filters and the outlet piping. This was alleviated by
installing two parallel tar receivers or knock-out pots to collect the heavy
material as shown in Figure 13. The upstream piping was heated to prevent con-
densation of tar and pitch. The tar receivers were water cooled to ensure
condensation of heavy tars; some water also inevitably was condensed.
To begin preoxidation, the bed was brought to temperature with only nitrogen
being fed. The coal feed and air were both begun at a reduced rate. The nitrogen
flow was diminished in accordance with the increasing air feed. At the same time
the electrical heat input was reduced and the temperature was stabilized at the
programmed level. Since only a fraction of the final air was being fed, severe
temperature excursions were avoided. The final feed rates of coal and air were
approached in several steps.
The run was shut down by first turning off the coal feed. The air flow was kept
at half rate for 15 minutes while the bed was cooling. This was to prevent
possible agglomeration from occurring in the hot bed in the absence of air. Air
was then shut off and the bed was drained under nitrogen.
To begin a Seeded Coal Process test, the bed was first brought to temperature in
nitrogen. Shortly before the coal was to be fed, the programmed air flow was
begun and the Powerstats were cut back to compensate for the oxidative heat
release. Seed char was then fed followed by the fine Ireland Mine coal.
The run was stopped by simultaneously shutting off the coal and air feeds.
2. Demonstration Runs
For the demonstration runs, the vessel, D-l, was used. A simplified flow diagram
for this system is given in Figure 17. The No. 2 reactor of Figure 14 was
used. Char was fed in through the acceptor outlet line, L-6, and removed via
outlet line, L-2. An overflow weir was used to maintain the bed level. All of
the analyzers were connected to the D-l reaction vessel.
A bed of char was fed to the reactor which was then brought to temperature using
programmed flows excepting the substitution of nitrogen for air. To begin the
run, the electrical heat was cut back and air replaced a portion of the nitrogen.
The run was ended by turning off the air and char feeds.
- 63 -
-------
3. Gasification Kinetics Runs
A flow diagram of the system is given in Figure 18. Char was fed into the
reactor through a dip tube which extended into the cone of the reactor, and was
removed through an overflow weir.
The bed of start-up char was brought to temperature using the programmed flows
for all gases. The exit gas analyzers were carefully watched, and the balance
period was begun when gas concentrations in the offgas had reached steady state
and three bed inventories were replaced. The run was shut down by diverting the
gas stream out of saturator and shutting off the char feeder.
The system configuration shown in Figure 18 also was used for pretreatment of
the master batch of char feed to the kinetics runs.
4. Carbon Burnup Cell Tests
For study of the carbon burnup reactor, the D-2 vessel internals were modified
as shown in Figures 19 and 20. The solids and gas inlets were joined inside
the reactor to form a Y so that they entered together. In order to survive the
severe oxidizing atmosphere, all the internals were fabricated of 310 stainless
steel.
The D-2 vessel bottom was plugged so that the only exit was overhead. The exit
line was led to the D-l vessel which was used as a large ash filter and reservoir.
The exit line of the ash reservoir was equipped with two No. S-78-6F Pall Trinity
Micro Corporation filters which were welded into a single long unit. The filters
were 316 stainless steel with a 20 micron mean pore size.
All the gas analyzers were connected to the burnup reactor.
The run was prepared by filling the burnup reactor with an inert bed of sand or
dead-burned dolomite, which was fluidized with the programmed gas flows. The run
was started by reducing the electrical heat input and starting the char feeder.
Ash was collected in the D-l vessel which was emptied after the entire run was
over. The run was ended by turning off the char feeder.
D. Liquid-Phase Glaus Reaction Equipment and Procedure
A schematic diagram of this apparatus is shown in Figure 21. S02 and H2S were
fed into the reactor in two ways. In the beginning of the program, an H2S-C02
mix was used in combination with purchased laboratory grade sulfurous acid. Later,
two separate gas mixes, S02-C02 and H2S-C02 were fed to the column. Steam was
added to the gas mixture in a saturator. Absorbent liquid was fed from pres-
surized feed tanks. Liquid recycle was effected by charging filtered recovered
product back to the feed tanks. All flows were measured with rotameters. The
column was set up for continuous cocurrent downflow operation.
The reactor was a Jerguson sight gauge with an internal cross section of
5/8" x 3/4" and was 54" long. All wetted metal parts were type 316 stainless
steel. Mica covers protected the glass faceplates. The packing consisted of
3 mm glass beads.
Liquid-gas separation took place just below the reactor. A sight gauge then was
used as a settler to separate the absorbent liquid from liquid sulfur.
- 64 -
-------
.4" I
.D.
I
4'
L.T).
REVISIONS
5-1
• • •. LXACM CO. r6H. P
CONSOLIOATION COAL. CO.
RESEARCH A DEVELOPMENT DIVISION
LIBRARY. PENNA.
Op
£.£.1.00
DHAWIMC MO.
CV-34-lQ
-------
FIGURE 15
CONFIGURATION OF DRAFT TU3ES USED IN DRAFT TUBE AMD SEEDED COAL TESTS
O-!
I
RECYCLE
( CONFIGURATIONS A , B and C )
RECYCLE
'lOOx
200 M
COAL
23 x
IOOM
CHAR
CONFIGURATION
Dimension A
Dimension B
8"
24"
Draft Tube .50O"O.D. x 444" I.D.
Cool Feed Line .25O"O.D. x .ISO" I.D.
tip positioned halfway into skirt
Accelerating Gas Line —
RECYCLE
r-
4"
T
r
100 x 28 x 43 x
2COM IOOM IOOM
COAL CHAR CHAR
CONFIGURATION B
8"
24"
750"O.D.x.630"l.D.
.250"C.D.x.l80"l.D.
tip positioned I inch above bottom of tube
IOO x 43 x
200 M !COM
COAL CHAR
CONFIGURATION C
6"
34"
75O"O.D. XJ630"1.D.
.250"O.D. x.ISO" I.D.
tip positioned 5 inches above bottom of tube
.375"O.D. x.305"I.D.
tip positioned I inch above bottom of tube
-------
FIGURE 16
Configuration of Draft Tube Used in Seeded Coal Tests
(Configuration D)
Ronvn 1 f>
Recycle
B
Accelerating
Gas (N2)
Air
N2
Dimension A
Dimension B
Draft Tube
Coal Feed Line
Accelerating
Gas Line
External Baffle
Internal Baffle
CONFIGURATION D
4"
34"
.750" O.D. x .680" I.D.
.250" O.D. x .180" I.D.
Tip positioned 5-inches above bottom of tube
.375" O.D. x .305" I.D.
Tip positioned 1-inch above bottom of tube
3-5/8" x 1-3/4"
Elliptical, 6O° from the horizontal
3/8" D x 60° Cone
Tip positioned 1/2" above coal feed line
- 67 -
-------
CONSOLIDATION COAL CO
RESEARCH ft DEVELOPMENT DIVISION
LIBRARY. PENNA.
-------
~]
01
(O
v/
3>l
K£^C.TQR^
CoOLt-R-
B
JDAfOKA'ToR.
Cm>.g. CH*.R^
"X T
•HOPPE.R&
REVISIONS
CMKO.
•n
7i
CONSOLIDATION COAL CO.
RESEARCH a DEVELOPMENT DIVISION
LIBRARY. PENNA.
•CALK
CF-MH
H CO- MIU Ft
-------
o
i
JSOSSBOL
Courmi
Wt>.L.g E.
CotJl.fc.tJ^kTt
gfcxifc.>\;E.(c
REVISIONS
5-14,
¥^
Nfc-UT
~-r-£-l
0
AriMfsi'
CONSOLIDATION COAL CO.
RESEARCH A DEVELOPMENT DIVISION
LIBRARY. PENNA.
CELL
•CALB
SSI.oo
OIUWIM* no.
OF-MIC,
-------
-------
CONSOLIDATION COAL CO
RESEARCH ft DEVELOPMENT DIVISION
LIBRARY. PENNA.
-------
Very careful temperature control was needed to prevent condensation or evapora-
tion of water at process conditions. The temperature range under study was
just beyond the point of minimum viscosity of liquid sulfur which lies at
300°F. The viscosity of liquid sulfur increases very steeply as the temperature
is increased above 320°F. Thus, temperature aberrations could pose problems
from this source. Close control of temperature was achieved by enclosing the
entire reactor in a 6" diameter by 6' tall glass column and circulating hot air
at the high rate of 150 SCFM. By maintaining a constant temperature outside
the reactor, there was no need for elaborate insulation and accessory heating
devices.
Since condensation of liquid in either the entrance or exit lines could have
resulted in the formation of sulfur which would have caused plugging, numerous
thermocouples were used to monitor temperatures on these heated lines. Exit and
inlet temperatures were printed out on a 24-point recorder. Temperatures inside
the glass column were read out on a digital temperature indicator which had been
calibrated against an ASTM thermometer.
Continuous monitoring of exit gas concentrations was provided by an infrared
S02 analyzer. A wet iodometric volumetric analysis was used to measure and
monitor H2S concentration.
Gas mixes were prepared gravimetrically on a Mastergram Model K-100-100-500 two
pan balance. The balance had a capacity of 100 KG with a sensitivity of 100 mg.
For the last three runs, the unit was repiped as shown in Figure 22. Two
thermocouples were removed. The lower one was replaced with a gas inlet line
terminating in a three-hole sparger, and the upper one was used as a liquid
feed line. Product gas exited through the top and sulfur was withdrawn from
the bottom as before. The packing in this mode was removed and the unit was
operated with a pool of liquid sparged with gas.
The unit was started up by bringing it.up to temperature with the large air
heater, and heating the inlet and exit lines. The reactor was brought to pres-
sure by using C02 at the flow rate of the mix gas. The steam generator was
brought onstream, and then water was fed in. The S02-C02 mix was substituted
for the inert gas and the exit gas composition was followed on the infrared
analyzer. When the exit gas composition was constant, the run was started by
turning on the H2S-C02 mix.
The S02 content of the offgas was continuously monitored with the infrared
analyzer. Hourly collections of sulfur and condensate were taken and the H2S
content of the offgas was also analyzed once per hour.
For once-through operations, the product liquid was collected hourly. For
recycle operation, when a feed hopper was emptied, the alternate hopper was
switched onstream and the product was removed and weighed. The product was
then filtered of sulfur fines and returned to the empty hopper.
The run was ended by simultaneously shutting off the flows of sulfur-bearing
mixes and substituting C02. All the sulfur and solvent liquid were then drained
from the column. C02 was then diverted out of the steam generator and the unit
depressurized. The reactor was allowed to cool slightly before being washed with
toluene followed by methanol.
- 73 -
-------
2, Z
LIQUID
GjAS 10
0
\L
I
QOT
- 74 -
-------
V. MATERIALS; SOURCE. PREPARATION AND ASSAYS
A. Inert Char
The char used in the coal pretreatment studies was produced by carbonization of
Kayford coal (Bethlehem Steel Mine in West Virginia) in Consolidation Coal
Company's Formcoke Pilot Plant. The char was inert and completely noncaking.
The char was sized as shown in Figure 23. Properties of the 28 x 100 mesh char
are presented in Table 16.
To simulate the seed char from commercial equipment (fines from the internal
cyclone in a commercial CBC gasifier), -100 mesh char was used. The sieve
analysis of the char is shown in Table 17.
Coal pretreatment operations were started with inert char beds.
B. Ireland Mine Coal
The coal used in the pretreatment studies was Pittsburgh Seam coal from
Consolidation Coal Company's Ireland Mine. The Ireland Mine coal is a high-
volatile, high-sulfur steam coal which is highly fluid when heated through the
plastic temperature range of 700-900°F.
The raw coal remaining from the previous campaign was dried, ground and sized.
The required sizes were obtained by the procedures shown on Figure 24. The
properties are shown in Tables 18 and 19.
The first attempt to preoxidize "coarse" coal with the draft tube used 24 x 100
mesh Ireland Mine coal remaining from our 1971 campaign. The properties were
published previously^1)
C. Loveridge Char
The raw material for the demonstration and gasification kinetics runs was pre-
carbonized char from Pittsburgh Seam coal from Northern West Virginia (Consol's
Loveridge Mine) produced in Consol's Formcoke pilot plant. Char was screened
at 28 x 65 mesh and was divided into three parts treated separately at 1200,
1600 and 17OO°F in the unit shown in Figure 17. The treatment conditions are
given in Table 20.
The part of char treated at 12OO°F which was not used up in Runs D~ll and D-12
was further treated at 1400°F to eliminate the possibility of coking in Run D-13.
The chars treated at 1600 and 1700°F were used as feedstocks for the gasifica-
tion kinetics runs. The fuel char used in the CBC studies was the product from
the demonstration runs and gasification kinetics runs. The reason for this was
to simulate as closely as possible the feedstock which would be used in a
commercial application.
A block diagram showing the steps for the use of Loveridge char is in Figure 25.
Char compositions and properties after treatment are shown in Table 21.
- 75 -
-------
FIGURE 23
PREPARATION OF INERT CHAR FEEDS
Mesh
PRECARBONIZED CHAR
SCREENING
SCREEN
48 x 100 Mesh USE
C-100 Mesh USEX
(^100 Mesh DISCARD^
- 76 -
-------
Table 16
Properties of Precarbonized Char -
28 x 100 Mesh Fraction
Hydrogen., Wt fo (dry basis) 1.82
Carbon 83.7
Nitrogen 1.56
Oxygen (by difference) 2.05
Sulfur .90
Ash 9.97
Volatile Matter . 5.93
Size. Tyler Mesh. Wt #
24 x 28 .6
28 x 35 17.0
35 x 48 38.9
48 x 65 22.8
65 x 100 17.1
-100 3.6
Mean Diameter, inches^*' .0126
Mean Density, Ib/ft3(2) 82.5
(i) Arithmetic mean.
(2) Measured in mercury at one atmosphere.
Table 17
Sieve Analysis of -100 Mesh Precarbonized Char
Size. Tyler Mesh. Wt %
+65 .0
65 x 1OO 2.0
100 x 150 19.6
150 x 200 30.9
200 x 325 31.5
-325 16.0
- 77 -
-------
FIGURE 24
PREPARATION OF FEED COAL
RAW IRELAND MINE COAL
DRYING
CRUSHING
GRINDING
SCREENING
SCREENING
C
OO x
Mesh
-2OO MESH DISCARD
(i) Nominal, see Table V-3 for detailed size consist.
(2) Nominal, see Table V-4 for detailed size consist.
- 78 -
-------
Table 18
Properties of Feed Ireland Mine Coal
Hydrogen, Wt % (dry basis) 4.87
Carbon 69.88
Nitrogen 1.25
Oxygen (by difference) 6.69
Sulfur 4.65
Ash 12.66
Volatile Matter 41.3
Size. Tyler Mesh. Wt $
+48 .0
48 x 65 16. 1
65 x 100 71.3
100 x 150 11.4
-150 1.2
Mean Diameter, inches(*) .00717
Mean Density, Ib/ft3(2) 82.O
(i) Arithmetic mean.
(2) Reciprocal mean, measured in mercury at one atmosphere.
Table 19
Sieve Analysis of 100 x 200 Mesh Ireland Mine Coal
Size. Tyler Mesh. Wt %
+ 100 1.8
100 x 150 32.8
150 x 200 36.9
200 x 270 20.0
270 x 325 1.0
-325 7.5
- 79 -
-------
Table 20
Char from Loveridge Mine Coal Treatment
Pressure: 15 Atm. (206 psig)
Feedstock
Temperature, °F
Input. SCFH
Fluidizing Gas
N2
Recycle
Lift Gas
Air
N2
Purges
N2
Char Feed Rate, Ib/hr
•* 28 x 65 Mesh Loveridge Char
1200 1400 1600
1700
- 90
160
40
60
20
10
- 80 -
-------
FIGURE 25
PRETREATMENT OF LOVER IDGE MINE CHAR
CHAR FROM LOVERitDGE MINE COAL
MESH DISCARD
RD^)
PRETREATMENT
at 1200°F
TO RUNS
D-ll & D-12
SCREENING
28 x 65 MESH
PRETREATMENT
at 1400° F
TO RUN D-13
PRETREATMENT
at 1600°F
TO RUNS
K-l - K-7
TO RUNS
K-2A - K-7A
TO CBC STUDY
FEE-TREATMENT
at 1700° F
TO RUNS
K-10 - K-16
TO RUNS
K-11A - K-16A
- 81 -
-------
Table 21
Properties of Thermally Treated Loveridge Char
See Table 20 for Conditions of
Treatment Temperature, °F
t
H, Wt %, dry basis
C
N
0 (diff.)
S
Ash
Size. Tyler Mesh, Wt $
+28
28 x 35
35 x 48
48 x 65
65 x 1OO
-100
Density of Size Fraction, Ib/ft3(1)
28 x 35
35 x 48
48 x 65
65 x 100
-100
Mean Density, Ib/ft3(2)
Mean Diameter, inches^3'
1200
1.59
80.58
1.32
2.27
2.09
12.15
.5
19.9
31.4
33.3
13.9
1.0
.0128
Treatment
1400 1600
1.08 .66
82.28 84.32
1.60 1.24
.35 -.77
2.13 2.O3
12.56 12.50
.0
17.7
30.5
30.9
18.9
2.0
89.4
90.5
89.8
91.0
85.2
90.1
.0122
1700
.52
84.72
1.32
-.98
1.85
12.63
(i) Determined in mercury at 1 atmosphere.
(2) Reciprocal mean.
(a) Arithmetic mean.
- 82 -
-------
D. Acceptors
During operations of the continuous acceptor unit, the following natural stones
were used:
1. Tymochtee dolomite, C. E. Duff and Son, Inc.,
Huntsville, Ohio
2. Canaan dolomite, Charles Pfizer, Inc., Adams, Mass.
3. Buchanan dolomite, James River Limestone Company,
Buchanan, Va.
4. Pennsylvania dolomite, G. and W. H. Corson, Inc.,
Plymouth Meeting, Pa.
5. Nebraska limestone, Hopper Brothers Quarries,
Weeping Water, Nebraska.
6. 1691 limestone.
The composition of the stones, as used in the experimental program, are given in
Tables 22 and 23.
The acceptors were ground using a mill which was adjusted to maximize the yield
of the desired size fraction. The oversize fraction was reground and the under-
size material was set aside. The material was then wet screened to remove all
of the fines. This was necessary to make attrition data obtained in the con-
tinuous unit more meaningful. The wet screened acceptor was then dried under
infrared lamps. Acceptor preparation is shown schematically in Figure 26
E. Acceptor Assays
1. Assay of Fresh Acceptors
The main purpose of the assay was to determine the useful CaC03 content of a
potential acceptor. A crushed sample, sized to 28 x 48 mesh, and weighing about
25 grams was placed in a quartz tube reactor which was heated by being immersed
in a fluidized sand bath furnace. After quenching to room temperature and
purging with dry air, the reactor and its contents were weighed on an analytical
balance after each of the following steps:
a. The sample was fluidized in C02 at atmospheric pressure and at
1100°F for 15 minutes. The weight loss represented the moisture
content, hydrate water, the C02 content of unstable carbonates,
one-half of the pyritic sulfur, and pyrolysis of the bitumen
content. This weight loss was less than 0.5$ of the raw stone
for all acceptors used in the operating program.
b. The sample then was fluidized at 1 atm C02 partial pressure at
1550°F for 30 minutes. The weight loss represented the C02
content of MgC03 and the C02 lost during reaction of some of the
CaC03 with impurities as discussed later.
- 83 -
-------
Table 22
Results of Raw Stone Assays
Canaan Dolomite
I
oo
Stone
Gravimetric Factors
F,
F,
ft(l 1
Tymocbtee
Dolomite
1.836
.3894
.2121
Run A7
1.887
.4719
.2501
Runs
A15 & A16
1.874
. 4687
.2501
Pa.
Dolomite
1.843
.4362
.2367
Va.
Dolomite
1.858
.4426
.2382
Nebraska
Limestone
1.745
.7299
.4184
BCR-1691
Limestone
1.595
.4541
.2847
Spentt
Stone
1.228
.2279
.1856
Uols Useful CaO/100 Ib
Raw Stone
.482
.568
.568
.538
.541
Examples;
Let W, - weight on MgCO,'CaCO, basis (CaCO, basis for limestone)
W2 = weight on MgO»CaCO, basis
S, = weight on MgO-CaO basis (CaO basis for limestone)
Wa = weight of sulfided stone
y = fraction of useful Ca converted to CaS
Then, Ft = Wt/W,
F, = (F,)(FS)
W, = (1 + F2) W,/F,
W, = W,/(l + Fa)
Hols useful Ca
in fresh, fully calcined stone =
in fresh half-calcined (MgO-CaCO,) stone °
Ws = W2jl - (F2)(y)(l00.09 - 72.14)/44.Ol/(l +
, )/44.01
.950
. 647
.422
(i) F, o the. weight fraction of COa equivalent to useful CaCO, In the raw, dry stone = (W»-Wi)/Wi.
(•) This stone originally was Tymochtee dolomite which had been exposed to many calcining-recarbonation cycles at
Acceptor Process conditions. About 42% of the original useful CaO had been converted to an Inactive form.
-------
Table 23
Stone
Useful CaO, Wt
MgO
ImpuritiesC 2)
Bound CaO
Compositions of Feed Acceptors
Canaan Dolomite
Tymochtee
Dolomite
49.6
38.6
6.8
4.9
Run A7
60.1
38. O
1.8
.0
Runs
A15 & A16
59.7
37.1
3.2
.0
Pa.
Dolomite
55.6
37.3
7. 1
.0
Va.
Dolomite
56. 4
38. O
5.6
.0
Nebraska
Limestone
93. O
.6
3.3
3. 1
1691
Limestone
57.9
5.5
26. 4(3)
1O.3
Inactive CaO
SpentC1)
Stone
29. 0
38.6
8.1
3.6
20.7
oo
m
i
(i) See Table 22.
(2) In most stones, the impurities largely are silica and alumina. The predominant minerals indentified
by x-ray diffraction, (See Reference 3 ) are;
(3)
Illite
Kaolinite
a Quartz
(Na,K)O'Al2O3-6 SiO2-2 H2O
A12O3.2 SiO2-2 H2O
The raw stone was treated with O.5N HC1} giving an insoluble residue which amounted to
24.7$ of the stone. After ignition in air, the residue composition was;
A1203, Wt %
SiO2
Fe203
Na2O
K2O
Ti02
P205
13.5
75.8
4.4
.2
5.2
.6
.2
-------
Figure 26
Acceptor Sizing
ACCEPTOR AS
RECEIVED
MILL
ROTAP
FINES DISCARD
l>
WET SCREEN
- 86 -
-------
c. The sample then was calcined by fluidizing in N2 at 1600°F for
30 minutes. The weight loss gave the desired useful CaC03
content of the stone.
d. The sample then was recarbonated by fluidizing in C02 at 1500°F
for 30 minutes. This was done to complete the calcining-
recarbonation cycle so that on subsequent handling the acceptor
would not air-slake by reaction with atmospheric moisture.
Characteristically, dolomites showed a weight gain of 95-97$ of
the weight loss on the previous calcining, and limestones showed
66-f
Supplemental tests showed that all reactions were complete in the time intervals
stated above.
Except for dolomites which had very low impurity contents, the weight loss on
calcining the MgC03 was greater than on calcining the CaC03. Since dolomites
with a mol ratio of MgC03/CaC03 greater than the theoretical ratio of unity are
a geological rarity, some of the CaC03 must have reacted with an impurity such
as Si02 according to,
2 CaC03 + Si02 = 2CaO-Si02 + 2 C02.
An experiment showed that indeed this situation had occurred: With a sample of
the Tymochtee dolomite, step b was carried out at 14OO°F using an 8O$ CO2~2O$ N2
mixture as the fluidizing gas. The exit gas composition was monitored by a
thermal conductivity cell using the inlet gas as cell reference. The cell output
showed that the MgC03 component had calcined completely in 15 minutes. The
fluidized bed temperature then was raised to 1550°F over a period of 15 minutes.
Between 1450 and 1550°F the cell output showed that incremental CO2 appeared in
the exit gas and that the reaction producing the C02 was completed at 1550°F.
At these conditions CaC03 could not have calcined since the equilibrium tempera-
ture at 0.8 atm C02 partial pressure is 1617°F. The sand bath temperature was
20°F higher than the fluidized bed temperature of the acceptor. Therefore,
undetected "hot spots" were not the source of the incremental C02.
When a direct determination of the MgC03 content was desired, step b was modified
based on the above finding. The stone was fluidized in C02 at 1350°F for three
hours in order to calcine the MgC03 while avoiding the reactions which bind some
calcium to the impurities. The bound calcium then was determined from the weight
loss which occurred after fluidizing in C02 at 1550°F for 30 minutes.
Results of the assays on the acceptors used in the operating program are given
in Table 22. Two shipments of the Canaan stone were used and the results from
each assay are included in the table. Gravimetric factors, calculated from the
assay results are shown in the table, are useful in the conversion between the
raw, half-calcined, and sulfided weights of the stone. Examples are shown in the
table.
Another way of expressing the compositions of the stones used is shown in Table
23. The components shown were calculated from the various weight losses deter-
mined from assays on the raw stones. The impurity contents were obtained by
difference from 100$. In earlier work,(3) the total impurity content obtained
by this method always was in good agreement with results of chemical analysis for
the noncalcium, magnesium components of the stones.
- 87 -
-------
2. Assay of Sulfided Acceptor
To avoid the long delay in receiving the data on the CaS and CaC03 content of
the recirculating acceptor after submitting the samples to the Analytical
Laboratory, a special assay which could be performed by the operators was
developed. The method is based on measurement of stoichiometric weight changes
involved with straightforward reactions.
a. Only CaS and CaC03 Present
Usually the sample contains no CaO because the partial pressure of C02 is kept
above the equilibrium value at all times during the sulfur cycle. Also, no
CaS04 is present because this material is not stable at the reducing conditions
in the gas desulf urizer . Experience has shown that CaS is not oxidized by steam
or C02 in the regenerator at 1300°F. However, privately sponsored studies show
that the oxidation rates become appreciable at temperatures above 150O°F.
A sample weighing about 20 grams was placed in a quartz tube reactor which was
heated by being immersed in a fluidized sand bath furnace. The sample was
fluidized in prepurified N2 at 1400°F for 30 minutes. After quenching to room
temperature and purging with dry air, the reactor and its contents were weighed
on an analytical balance. The weight loss represents the CaC03 content. The
composition is calculated as follows:
Let X = grams dry sample.
F2 = gravimetric factor from assay of raw, fresh stone
(see Table 22) .
Of = weight loss on calcining in N2.
A = mols CaS present .
B = mols CaC03 present.
C = hypothetical sample weight if acceptor were in form
of MgO-CaO.
Then,
C = X - a - A (72.14 - 56.08) (l)
A + ^ToT = c F 2/44.01 (2)
Solution of equations (l) and (2) gives
F2 (X-Of) - Of
Also,
44.01
Then, the composition of the acceptor, expressed as mol $ of useful calcium
is:
CaC03 = 100 B/(A + B)
CaS = 100 A/(A + B)
- 88 -
-------
b. If CaS04 also is Present
When CaS04 purposefully has been formed during some portion of the acceptor cycle
by oxidation with air, a second step is added to the assay described above. The
sample is fluidized in H2 at 16OO°F for 15 minutes. After quenching to room
temperature and purging with dry air, the reactor and its contents again is
weighed. The weight loss represents the conversion of CaS04 to CaS. The com-
position is calculated as follows:
Let a = weight loss on reduction to CaS.
e = factor for reducible iron content of acceptor.*
D = mols CaS04 present.
Other nomenclature as above.
Then,
D = (a - e C)/16/4 (3)
C = X - a - A (16.06) - D (136.14 - 56.08) (4)
A + TT^7 + D = C F2/44.01 (5)
44 01
Solution of equations (3), (4), and (5) gives:
(x-g) F2 at _ L 80.06 F2~]f a - e (x-q) "1
44.01 " 44.01 L + 44.01 JL 64 - 80.08 ej
16.06 F2 F 8O.O6 F2"|r 16.06 e "I
1 + 44.01 + L + 44.01 J j_64 - 80.06 el
Solution of equations (3) and (4) gives
a - e (x-q) + 16.06 A e
64 - 80.06 e
Also,
a
B =
44.01
Then, the composition of the acceptor, expressed as mol % of useful calcium is,
CaC03 = 100 B/(A + B + D)
CaS = 100 A/(A + B + D)
CaS04 = 100 D/(A + B + D)
Most acceptors contain reducible iron which is present initially in the raw
stone as limonite and pyrite. Using the same general technique as described
for the raw stone assay, the raw stone was calcined by fluidizing in air at
1650°F for 30 minutes. After weighing, the sample then was fluidized in dry
H2 at 1650°F for 15 minutes. The weight loss is loss of oxygen which
represents the reducible iron content. The factor, e, is the ratio:
grams oxygen lost/gram of acceptor in the MgO-CaO form.
- 89 -
-------
c. Refractory CaS
In the following discussion, the assumption is made that CaS04 has not been
formed. The assay technique was developed using samples of freshly prepared
acceptor in the form of MgO-CaC03 in which a known amount of CaC03 was con-
verted to CaS by reaction with H2S. With these samples, the CaS content could
be determined directly by the following method:
After calcining the CaC03 component in N2 at 1400°F as described in section (a)
above, the sample was fluidized in dry air as the temperature was raised from
1300 to 1600°F in 30 minutes in order to oxidize CaS to CaS04. Some generation
of S02 always occurred by the reaction,
1/4 CaS + 3/4 CaS04 = CaO + S02.
The S02 was trapped in an absorption bulb containing Ascarite. After quenching
to room temperature, the reactor and its contents were weighed on an analytical
balance. The composition of the fresh sample was calculated as follows:
Let 3 = weight gain after oxidation in air.
Y = weight gain of Ascarite bulb.
other nomenclature as above.
Then,
* •(»*£$•') /«
B = a/44.01.
Then, the composition of the acceptor, expressed as mol % of useful calcium is,
CaC03 = B/(A + B)
CaS = A/(A + B)
When samples which had been exposed to cyclical sulfiding and regeneration at
process conditions were assayed by the above procedure, it was found that some
of the CaS was not oxidized, even when the final temperature was increased to
1700°F and held for periods up to three hours. This unoxidized CaS has been
termed "refractory CaS."
For these samples, the total CaS was calculated as in section (a) above, i.e.,
F2 (x-a) - a
A =
44.01 + 16.06 F2
and the oxidizable CaS, Al, was calculated as
" - <> *
Then, the composition of the acceptor, expressed as mol % of useful calcium is,
CaC03 = 100 B/(A + B)
CaS = 100 A1/(A + B)
Refractory CaS = 100 (A~Al)/(A + B)
- 90 -
-------
VI. SAMPLE CALCULATIONS
The following operations were conducted during the course of the present con-
tract period:
1. Preoxidation
2. Seeded Coal Pretreatment
3. Acceptor Sulfur Cycle Studies
4. Carbon Burnup Cell Tests
5. Gasifier Operability Demonstrations
6. Gasification Kinetics
7. Liquid-Phase Claus Reaction Runs
Sample calculations involving preoxidation and gasification were detailed in the
last report.(1) Calculations detailing gasification kinetics are discussed in
Section XII. Seeded Coal Process studies were preliminary in nature and
detailed calculations were not carried out. The sample calculations in the next
sections therefore discuss only Items 3, 4, and 7 above.
A. Acceptor Sulfur Cycle Studies
1. Calculation of Number of Process Cycles
Definition of a Cycle
A cycle is completed when the entire inventory of acceptor has passed through
the gas desulfurizer. The inventory consists of the contents of the gas desulfu-
rizer, regenerator and their respective acceptor transfer lines together with the
acceptor in the feed hoppers and product receivers.
ATT
AN = — evaluated from t tot + At.
I •(• X
where; AN = change in number of cycles accumulated.
AH = grams of half-calcined acceptor fed during the period.
I = average internal inventory of reactors and transfer
lines over the period, grams.
X = average external inventory of acceptor over the period,
grams.
t,At = time and time increments, respectively.
The external inventory changes due to attrition and sampling losses. There is a
loss of weight in both the external and internal inventory as CaC03 (MW 100.l)
is replaced by CaS (MW 72.1).
Fractional cycles exist when only part of the inventory has been through the
reactor. Let n + f be the number of cycles, where n is an integer and f is the
fraction of the inventory which has come out of the gas desulfurizer. On the
average, f part of the inventory has seen n + 1 cycles, and 1 - f has seen n
cycles. In our system the mixing occurring in the fluidized beds and in the
product collection pot brings about smooth changes in the sulfur content of the
acceptor. Average cycles are used; no attempt has been made to define a resi-
dence time distribution for the experimental system.
- 91 -
-------
All the data, including regenerator analyses are keyed to the gas desulfurizer.
Generally, about 1/4 to 1/3 of the total inventory is in the regenerator. Thus,
the material exiting the regenerator is about 1/4 of a cycle "younger" than the
material leaving the gas desulfurizer. In the figures which plot CaS content of
acceptor as a function of cycles, the regenerator curve is displaced about 0.25
cycles to the right from the gas desulfurizer curve for purposes of comparing
the change in a given acceptor particle.
Sample Calculation of Process Cycles - Run A7
The number of process cycles was calculated over periods of time corresponding
to about 9,000 grams of feeding. From 384 to 525 minutes into the run, 9,063
grams of acceptor were fed. During this period, the feed hoppers were switched
six times. At the instant that a feed hopper is started the external inventory
consists of the contents of that feed hopper plus the contents of the full
acceptor collection hopper. Thus, the external inventory is the average of two
feed hoppers during the period. Therefore, from 366 to 507 minutes into Run
A7,
orjco
External inventory = X = / = 3,021 grams.
In Run A7, the beds were removed from the reactors just after the half-calcining
period and again at the end of the run. The weights were 3,862 grams and 3,500
grams, respectively. A smooth curve was drawn between the points and the
average internal inventory was estimated from the curve for each calculation.
The change in internal inventory was small compared to the total inventory.
From 366 to 507 minutes into the run,
Internal inventory, I = 3,700 grams,
and I + X = 3,700 + 3,021 = 6,721 grams.
AH 9 063
Change in number of cycles = AN = — = '_ = 1.35
1 + A O, I ZJL
At the start of the period, 2.83 cycles had been accumulated. Therefore, the
number of cycles at the end of the period was;
SN = 2.83+1.35 = 4.18 cycles.
For Run A7, the calculations are summarized in Table 24.
2. Calculations of Conditions and Results for Gas Desulfurizer - Run A7
The numbers used below were taken from Run A7. See Tables 44 and 45 in Section
IX.
Gas Desulfurizer Temperature
Axial temperature traverses showed that the gas desulfurizer bed temperature was
constant to ± 2°F over the entire length of the acceptor bed. The bed tempera-
ture was continuously measured by thermocouples positioned l" and 15" above the
bottom of the bed.
- 92 -
-------
Table 24
to
CO
Computation of Number of Cycles-Run A7
Time,
Start
O
229
384
525
737
919
1O81
126O
1412
Minutes
End
229
384
525
737
919
1081
126O
1412
1531
AH,
Acceptor
Fed,
Grams
11853
10023
9O63
1O626
9177
7551
8386
7191
5698
n,
Number
of Hopper
Changes
6
6
6
8
8
8
10
10
9
x --£&-
X ~ n/2 '
External
Inventory,
Grams
3951
3928^ )
3021
2657
2294
1888
1677
1438
1226
I,
Internal
Inventory,
Grams
3815
3740
37OO
3655
3610
358O
355O
3525
35 1O
I + X,
Total
Inventory,
Grams
7766
7668
6721
6312
5904
5468
5227
4963
4776
AN = ^- ,
I + X '
Change in
Number
of Cycles
1.53
1.30
1.35
1.64
1.55
1.38
1.6O
1.45
1. 19
E AN =
Total
Number
of Cycles
1.53
2.83
4. 18
5.82
7.37
8.75
1O.35
11.80
12.99
(i) Part of hopper fed during restart of unit. X = 11785/3 = 3928.
-------
Acceptor Feed Rate
Over the first 203 minutes of the run, 9,989 grams of half-calcined acceptor
were fed. The feed rate was calculated as,
X 6° = 2,952 gm/hr = 6.50 Ib/hr.
203
Solids Retention Time
The bed height is 1.5 feet and the diameter is 3 inches. The solids occupies
0.35 of the volume and has a density of 135 lb/ft3. The weight of acceptor in
the bed is therefore,
0.785 x 777 x 1.5 x 0.35 x 135 = 3.47 Ibs.
144
Since the feed rate was 6.5 Ib/hr, the nominal residence time in the bed is,
3.47 x 60
=32 minutes.
Exit Gas Composition
The exit gas composition was determined directly from the iodometric H2S analysis
and the GC analysis of the off gas. In the absence of the GC analysis, the exit
composition was calculated from the inlet flows and the condensate collection.
The inlet gas contained excess C02 and H2. Experience has shown that shift
equilibrium is rapidly achieved at the bottom of the reactor. CO and H20 are
produced via the shift reaction. The desulfurization reaction consumes H2S and
produces additional C02 and H20.
Over three hours, 2,568 grams of condensate were collected for an average hourly
rate of 856 gm/hr. Converting to SCFH,
Via dry gas meter readings, the dry exit gas rate converted to standard condi-
tions was 220 SCFH. Iodine scrubber analysis showed the H2S concentration to be
0.049$. Then,
H2S reacted = 3.5 - .00049 x 220 = 3.4 SCFH.
Since each cubic foot of H2S which reacts generates one cubic foot each of C02
and H20, the net water formed via the shift reaction was 37.1 SCFH. The dry
exit gas rate may be estimated as follows:
H2S or Product C02
Inlet C02
Inlet H2
Inlet N2
Purges (5 C02, 15 N2)
Less H20 via Shift
Total Dry Gas
- 94 -
-------
The calculated and observed exit gas rates agreed within 5$. The average of
215 SCFH was taken as the actual dry exit gas rate, and the material balance
value was used for calculational purposes. The contents of the exit and recycle
flows were calculated as shown below:
Exit Gas
via Material
Balance
Inlet
54
73
96
Shift Reaction Puree
-37.1 3.4 5
-37.1
15
37.1
SCFH
(209.4)
.1054
25.3
35.9
111.0
37.1
Mol %
.049
12.1
17.1
53.0
17.7
Actual
Exit
Gas,
SCFH
(215)
.1078
26.0
36.8
114.0
38.1
Recycle,
SCFH
(175)
.08575
21.2
29.9
92.8
31.0
Compound
H2S, .049$
C02
H2
N2
CO
In those cases where the GC analysis was available, the SCFH of each component
in the recycle gas and the exit gas was obtained by multiplying the mole fraction
by the gas flow rate.
' -)
Gaff Flow and Composition Above the Bed
The total flow above the bed = dry exit flow + recycle + steam - purges above the
bed. Thus, the total flow above the bed was,
215 + 175 + 40.5 - 15 = 415.5 SCFH.
On
a wet basis just above the bed we get,
Component
H20
CO 2
H2
CO
H,S
SCFH
40.5
47.2
66.7
191.8
69.1
.1912
Mol #
9.75
11.4
16.1
46.2
16.6
.046
Fluidizing Velocity
The general equation for the fluidizing velocity is as follows:
_ x T
3600
V =
x
144
A
TT
530
or
V = 7.55 x 10
An
where; V = fluidizing velocity, ft/sec.
Q = gas flow rate, SCFH.
A = cross-sectional area of vessel, in2.
T = bed temperature, °R.
TT = system pressure, atm.
The cross-sectional area of the reactor was 7.24 in2. Thus,
7.55 x 10~5 x 2060 x 416
7.24 x 15
= 0.60 ft/sec.
- 95 -
-------
Approach to Equilibrium
The equilibrium ratio for PH2o * PC02/PH2S is 510 at 1600°F. This may be
expressed in mol fractions as,
XH,O x 15 x xco, x 15
510 =
XH s x 15
2
where the subscripted X represents a mol fraction. The equilibrium mol $ H2S
was,
= ^J^HjOJlV^ _ 100 X .0975 , .114 X 15 . ^
510 510
Since the percent of H2S above the bed was .046,
$ H2S in Outlet/Equilibrium % H2S = .046/.0327 = 1.4.
Removal of H2S
The percentage removal of fresh H2S was,
Removal of Feed Sulfur = 100 x (3-5 ~ -108) = 97$.
3.5
The percentage removal of feed plus recycle H2S was,
/ o e _ -^ f^Q \
Removal of Feed + Recycle H2S = 1OO x , i Ao^e = 95$.
O • D T • Uo i D
Conversion of Acceptor
3.5 - .108
Mols of H2S reacted = —OQ, 0 = .00877.
ooo . o
The molar feed rate of acceptor was calculated by first converting the feed rate
to the raw stone basis and then dividing by the pounds raw stone per mol Ca
using the gravimetric factors given in Section V-E.
_* r, * , „ „ J Feed x Fi x Mols Ca/100 Ib Raw Stone 6.50 x 1.887 x .568
Mols of Useful Ca Fed = (i + F2) xlOO = (l + .4719) x 100
= .04735
OO877 -/
Conversion of Acceptor = 100 x 'n>|t7Q<- = 18.52$.
• vJ4 /oD
Rate of Attrition
Including the calcining period, a total of 96,348 grams of acceptor were fed to
the unit. The solids filters yielded 773 grams from the gas desulfurizer and
25 grams from the regenerator. The overall attrition rate was,
Attrition Kate . <"' +„%,' "° . .33*.
3. Calculation of Conditions and Results for Regenerator - Run A7
Regenerator Temperature
Axial temperature traverses showed that the regenerator bed temperature was
constant to within 2°F over the entire length of the acceptor bed. The bed
temperature was continuously measured by thermocouples positioned 16", 21", and
42" above the bottom of the bed.
- 96 -
-------
Solids Retention Time
The bed height was 46" and the diameter was 2". The solids occupied .35 of the
volume and had a density of 135 lb/ft3. The weight of acceptor in the bed was
therefore,
4 46
.785 x — - x — x .35 x 135 = 3.96 Ibs
1 £t
Since the feed rate was 6.5 Ib/hr, the nominal residence time in the bed was,
3.96 x 60
_ „,, =37 minutes
6.50
Exit Gas Composition
The exit gas analysis is determined directly from the iodometric H2S analysis
and the GC analysis of the offgas. In the absence of the GC analysis, the exit
composition was calculated from the inlet flows and the condensate collection
as follows. Over three hours an average of 2324 gm/hr of condensate was
collected. Converting to SCFH,
2324x387
18.02 x 454
Via the dry gas meter, the dry exit gas rate converted to standard conditions
was 133 SCFH. Iodine scrubber analyses showed the H2S concentration to be
0.7$. Then,
H2S released = .007 x 133 = .93 SCFH
Each cubic foot of H2S released consumed one cubic foot each of C02 and H20.
The dry exit gas rate may be estimated as follows:
C02 or Product H2S 110 SCFH
Purges (5 C02, 18 N2) 23
Total Dry Gas 133 SCFH
The calculated and observed exit gas rates were in excellent agreement.
The composition of the exit flow was calculated as shown below;
Compound
C02
N2
H2S
Inlet
110
Reaction
-1
.93
Purge
5
18
SCFH
114
18
.93
Mol %
85.7
13.5
.7
Gas Flow and Composition Above the Bed
The total flow above the bed = dry exit flow + steam - purges above the bed.
Thus, the total flow above the bed was,
133 + 110 - 15 = 228 SCFH.
- 97 -
-------
On a wet basis above the bed we get:
Component SCFH Mol #
C02 109 57.8
H2S .93 .41
H20 110 48.3
N2 8 3.5
Fluidizing Velocity
The area of the reactor is 3.20 in2. Thus, the fluidizing velocity is,
7.55 x 10"5 x 1760 x 228
3.20 x 15
Conversion of Acceptor
= .63 ft/sec.
The exit gas contained .93 SCFH of H2S or .0024 mols. The acceptor feed rate
was .04735 mols/hr. The regeneration of acceptor was,
.0024/.04735 = 5.1#.
B. Carbon Burn-up Reactor Calculations - Run CBC-8. Phase 2
Reactor Temperature and Temperature Gradient
Traverses of the reactor at two times are shown in Figure 27. The target tempera-
ture for the run was 19OO°F, and the run temperature was taken as the maximum
temperature in the bed. The recorded maximum temperature actually varied by 6°F
higher and lower than the target temperature. The gradient was the difference
between the bottom of the bed and the temperature 16" above the bottom. The
nominal bed height was 18", but it was felt that fluctuations in bed height might
leave this point uncovered at times thereby giving an inaccurate value for the
bed temperature. The reported temperature gradient of 22°F was the mean of
21 determinations with a range of 10-30°F.
Char Feed - Run CBC8. Phase 2
Char was fed at the rate of 1.41 Ib/hr. The components of the char and their
fate relative to oxidation are shown in the table below;
Product Gas, 02 Consumed,
Component Ib/hr SCFH SCFH SCFH
Moisture .00505 .11 .11 0
Ash .226
H2 .00688 1.32 1.32 .66
C 1.135 36.55 36.55 36.55
N2 .0105 .14 .14 0
S .0237 .29 .29 .29
Total 1.407 38.4 37.5
- 98 -
-------
to x 10 TO Vi INCH 46 1323
7 X 10 INCHES MtDE III o.».A. .
KEUFFEL A ESSER CO.
-------
Composition of Exit Gas
The composition of the exit gas streams was determined by material balance as
shown below. Analysis of the overhead solids showed that all combustibles in
the char had been burned up.
Compound
H20
N2
CO 2
Total
Oxveen Partial
In,
SCFH
169
88.2
336.8
0
594
Pressure
From Char,
SCFH
1
-37.5
0.1
36.6
Top of Bed
SCFH
170
5
337
37
595
Percent
28.6
8.6
56.6
6.22
Dry + Purges Above Bed
SCFH Percent
0
5
357
37
445
0
11.5
80.2
8.3
The inlet oxygen partial pressure was,
88.2 x 15
594
= 2.23 atm.
The exit oxygen partial pressure above the bed was,
= 1.29 atm.
51 x 15
595
C. Liquid-Phase Claus Reactor - Run C6
Temperature
The temperature in the column was measured with four thermocouples . There was
generally a 2°F difference between the top and the bottom of the column. The
average temperature was used. In Run C6 the temperature was 310°F.
Inlet
Composition
SCFH
The first inlet gas stream was a C02-S02 mix which was humidified in the steam
generator. The temperature was set for 318°F. At the exit of the steam
generator, the second gas mix containing H2S in C02 was added to give the
composition shown below. The dew point of the mix was 310°F.
Compound
H20
CO 2
H2S
S02
Total
The reactor cross-section was 5/8" x 3/4", so that the area was 3.26 x 10~3
sq ft. The throughput rate was,
• 28.83
3.26 x 10"
= 8840 SCFH/ft!
- 100 -
-------
Solvent Liquor Rates
Over a 45-minute period, a hopper containing 442.2 grams of recycle liquor was
fed for an hourly rate of 590 gra/hr. This corresponded to 1.30 Ib/hr or,
- «»
Over the same time period, 453.6 gm or 604.8 gm/hr of product liquor were
collected. This corresponds to 1.33 Ib/hr.
Elemental Sulfur Product Rate
The collection of elemental sulfur tended to be irregular, and the product rate
was taken as the average hourly rate over the last 5.9 hours of the run. In
Run C6 this was 40.9 gm/hr.
Condensate Collection
The condensate was actually Wackenroder' s solution since it contained sulfur
and various soluble sulfur compounds formed from the low temperature reaction of
H2S and S02 in liquid water. The condensate was collected hourly. For Run C6,
the rate was 196.4 gm/hr and the sulfur content of the condensate was 3.87$.
Exit Gas Composition
Since sulfur compounds were removed from the exit gas as the condensate was
formed, the sulfur content of the gas was useful only for material balance
calculations. In Run C6 the exit gas rate was 17.2 SCFH and contained 1.3$ H2S
together with 0.1$ S02.
Sulfur Balance
The sulfur balance for Run C6 is outlined below:
Grams of
Component _ Calculation _ Sulfur /hr Percent
Feed Sulfur ....
— — — — — — 454
Feed H2S 1.086 x -rrr x 32.06 40.8 66.0
oo/
454
Feed S02 .543 x -rr x 32.06 20.4 33.0
38T
Solvent Liquor 590 x .0011 .65 1.0
Total Feed Sulfur 61.85
Exit Sulfur
Product Liquor 604.8 x .0011 .665 1.08
454
Exit Gas 17.2 x (.013 + .001) x •— x 32.06 9.06 14.6
Oo /
Condensate 196.4 x .0387 7.60 12.3
Elemental Product Sulfur 40.9 66.1
Losses and Accumulation (by difference) 3.63 5.9
- 101 -
-------
The losses and accumulation entry merely brings the material balance to 100$.
However, losses do occur in both the sulfur exit piping and in the condensate
receiver as these contain sulfur when they are examined at the end of the run.
It is believed that the sulfur analysis of the condensate may be low because
of the difficulty of removing the solid sulfur when an aliquot sample is
taken.
- 102 -
-------
VII. TABULAR CHRONOLOGICAL HISTORY OF RUNS
The purpose} general run conditions, and major conclusions drawn for
all runs made in the course of this contract period are given in the
tables listed below. All the runs were made at a pressure of 15
atmospheres.
TABLE NO.
25 Summary of Seeded Coal Tests
26 Summary of Proxidizer Operations
27 Summary of Sulfur Acceptor Runs
28 Summary of Gasifier Demonstration Runs
29 Summary of Gasifier Kinetics Runs
30 Liquid-Phase Claus Reaction Operations
31 Carbon Burn-up Cell Operations
- 103 -
-------
Table 25
Summary of Seeded Coal Tests
Run Number
Draft Tube
Conf Igurat ion
Date of Runs
Purpose of Runs
Size of Bed Char
Size of Seed Char
Temperature Range, °F
Results and
Conclus ions
1-5
1, 4/12/72
2-5, 4/13/72
Exploratory
28 x 100
28 x 100
1200-14OO
1.
Poor mixing
suspected.
2. Best result
with 10:1
seed char to
coal ratio.
B
6-8, 4/17/72
9-10, 4/18/72
11-14, 4/20/72
15-19, 4/21/72
Operate with larger
tube having coal
feed line inside
28 x 10O
28 x 100, 48 x 10O
11OO-1500
1. No significant
improvement.
2. Mouth of tube
often choked with
agglomerates.
3. 1SOO°F not operable
20-23
20-22, 4/26/72
23, 4/28/72
Try accelerating
gas to improve
mixing
48 x 100
48 x 100
1100-1500
1. No significant
change.
2. Concluded that
most of fluidlzing
gas was entering
draft tube.
Run Number
Draft Tube
Configuration
Date of Runs
Purpose of Runs
Size of Bed Char
Size of Seed Char
Temperature Range, °F
Results and
Conclusions
24-3O
24, 6/2/72
25-3O, 6/6/72
Operate with baffles
to improve mixing
and fluidization
48 x lOff
None
9OO-15OO
1. Little or no smear-
ing at 90O°F.
2. Caking at 15OO°F.
3. Best smearing at 13OO.
4. Product density
> 80 Ib/ft'.
5. Agglomerates formed
in all cases.
D
6/6/72
Make a long run
to get steady-
• . state product
-1OO
-100
1300
1. Maximum agglomerates
24 mesh.
2. Plug in outlet system
prevented completion
of run.
- 104 -
-------
Table 26
Summary of PreoxifH?•.«•«•• Operations
O
01
Run Number
Unit Revisions
Date of Run
Run Duration, hours
Purpose of Run
Temperature} ° F
Feedstock
Shutdowns
Results and
Conclusions
PRO
Configuration A draft
tube installed
4/6/72
. 17
Does draft tube allow
operation at 85O°F
85O
24 x 10O Ireland Mine Coal
Bed agglomerated
1. Draft tube does not enhance
operability enough to allow
85O°F preoxidizer operation.
2. 8.6% preoxidation.
PR1A
PR1B
No draft
tube
5/15-16/72
14.1
First stage preoxidation
675
48 x ISO Ireland Mine Coal
Voluntary
1. 5.7$ preoxidation.
Configuration D draft
tube installed
5/26/72
2.0
Second stage of
preoxidation
775
PR1A Product
Voluntary
1. Run was operable.
2. 5.7$ preoxidation.
3. Lab test showed product
to be unsuitable gasifier
feed.
-------
Table 27
Summary of Sulfur Accentor Huns
Run Number
Unit Revisions
Date of Run
Run Duration with
H2S Feed, hours
Purpose of Run
Feedstock
Shutdowns
Results and
Conclusions
None
8/3/72
6.1
First exploratory
acceptor run
Changed coll and Inlet
line to 446 S. S.
A4
8/10/72
2.6
Repeat Al. Refine control of
wall temperature in desulfurizer
and reduce gas velocity in re-
generator.
28 x 35 Tymochtee dolomite
None
8/18/72
7.1
Feed air into regenerator to In-
duce CaSO4-CaS Transient liquid
hardening of acceptor.
35 x 48 Tymochtee dolomite
Had high AP in unit and could
not switch solids filters.
1. Good removal of H2S in gas
desulfurizer.
2. Only partial regeneration.
CaS built up in acceptor
3. 2.5$ attrition rate.
Loss of acceptor inventory due to attrition
1. Confirm good HaS removal
in desulfurizer and build
up of CaS.
2. 18$ attrition rate.
1. 5.0$ attrition. Better
than A2, but worse than Al.
Run Number
Unit Revisions
Date of Run
Run Duration with
HaS Feed, hours
Purpose of Run
Feedstock
Shutdowns
Results and
Conclusions
Run Number
Unit Revisions
Date of Run
Run Duration with
HaS Feed, hours
Purpose of Run
Feedstock
Shutdowns
Results and
Conclusions
AS.
None
8/22/72
5.7
Determine attrition resistance
of limestone
28 x 48 Nebraska Limestone
Voluntary
1. Only 0.4% attrition, but
little reaction with H,S.
A8
A6
None
8/25, 31, 9/1/72
24.9
Run with dolomite known to be
hard to see if sulfur cycle
causes weakness
20 x 35 deactivated dolomite
ex OCR studies
8/25 plug in L-5 then dif-
ficulty with HaS feed
9/1 Voluntary •
1. Acceptor stayed hard.
Sulfur cycle per se does
not cause softening.
_A9_
Only desulfurizer in operation
9/11/72 9/12/72
9. 6 10.1
———— Examine once-through operation
28 x 48 Tymochtee Dolomite
Voluntary
1. 86$ solid conversion,
96$ desulfurization.
1. 95$ solid conversion, 89$
desulfurization.
AT
None
9/7, 8/72
25.2
Test attrition and reactivity
of Canaan dolomite
28 x 35 Canaan dolomite
Voluntary
1. Achieved only 0.7$ attrition
with good HaS removal.
2. CaS accumulated in acceptor.
At end, 13 raol $ of acceptor/
pass was being regenerated.
All
Returned to 2 vessel operation
10/3/72
5.2
Test other dolomites for attrition
and activity
28 x 35 Pa. dolomite
Loss of inventory due to attrition
1. O.O$ attrition, good activity.
2. Once-through operation with 85-95$ solid conversion is feasible.
- 106 -
-------
Table 27 (Cont'd.\
Summary of Sulfur Accentor Runs
Run Number
Unit Revisions
Date of Run
Run Duration with
H,S Feed, hours
Purpose of Run
Feedstock
Shutdowns
Results and
Conclusions
A12
A13
None
10/5, 10/72
15.1
Test other dolomites for attri-
tion and activity.
28 x 35 Buchanan dolomite
10/5 Could not switch solids
filters.
10/1O Loss of inventory due
to attrition.
1. 4.1$ attrition, good activity.
None
10/11, 12/72
19.3
Examine attrition and reactivity
of a second limestone,
28 x 35 BCR 1G91 Limestone.
Could not transfer solids from
gas desulfurizer to regenerator
1. Stone formed agglomerates.
2. 0.6jt attrition with fairly
good activity toward H,S.
A15
All Internals made of 310 88
changed to 446 SS. Replaced 31O
SS inner shell
10/16-18/72
35.1
Run Canaan dolomite with 1100*F
regenerator.
28 x 35 Canaan dolomite
Voluntary
1. Only 7% regeneration per pass
after CaS had accumulated.
Run Number
Unit Revisions
Date of Run
Run Duration with
HaS Feed, hours
Purpose of Run
Feedstock
Shutdowns
Results and
Conclusions
A16
None
1O/19, 20, 23/72
42.7
Only gas desulfurizer in opera-
tion
10/25/72
1.5
Aia
Repeat A15 increasing regenerator Examine change in refractory
from 35 to 75 minutes residence CaS as acceptor was hold at
time
28 x 35 Canaan dolomite
Voluntary
1. No effect of increased resi-
dence time in regenerator.
gas desulfurizer conditions
28 x 35 Canaan dolomite
Voluntary
1. No change in refractory
CaS when hold at gas de-
desulfurizer conditions.
Return to 2 vessel operation.
10/26/72
11.6
Repeat A4, but with no air in
regenerator.
Examine effect of H,0/C0,
35 x 48 Tymochtea dolomite
Loss of inventory due to
attrition
1. Hardening due to transient
liquid formation did not occur
in A4.
2. Change in H,0/C0, ratio in
regenerator between .5 and 1.7
had no discernible effect.
Run Number
Unit Revisions
Date of Run
Run Duration with
H3S Feed, hours
Purpose of Run
Feedstock
Shutdowns
Results and
Conclusions
. A19_
None
10/26, 27/72
11.8
Repeat A12 but with gns de-
sulfurizer at 1650°F
28 x 35 Buchanan dolomite
10/27 Plug in HaS feed lino.
10/27 Loss ot Inventory due to
attrition.
1. No change in attrition or
other results due to 1650°F
opuration.
- 107 -
-------
Table 28
Summary of Gasifier Demonstration Runs
o
oo
Run Number
Unit Configuration
Data of Run
Run Duration, hours
Purpose of Run
Temperature, °F
Feedstock
Shutdowns
Results and
Conclusions
Run Number
Unit Configuration
Data of Run
Run Duration, hours
Purpose of Run
Temperature, °F
Feedstock
Shutdowns
Results and
Conclusions
Dll
D12-A
D12-B
Figure 17
11/17/72
12
No change
11/20/72
2
No change
11/20/72
5
Demonstrate operability at simulated conditions at point of entry in gas ifier
1725 1775 1775
••-Char made from Pittsburgh Seam coal pretreated at 12OO°F (Loveridge coal)—*
Voluntary Heating element broke Transfer line plugged.
in steam generator.
1. No ash slagging.
2. Process operable.
D12-C
D12-D
No change
11/28/72
4
Feed line from bottom
11/28-29/72
8
D12-E
No change
11/30/72
4.5
D13
No change
12/18/72
6.2
-Demonstrate operability at simulated conditions at point of entry of gasifier
1775
1775
12OO° F Lover idge Char
1775
Transfer line plugged Transfer line plugged
1. Coke like material
in vessel.
1775
» 12OO°F Loveridge char
pretreated at 14OO°F
High AP and AT in vessel *
*—1. Deposit at bottom of vessel
-------
Table 29
Summary of Gasification Kinetics Runs
Run Number
Unit Configuration
Date of Runs
Run Duration, hours
Purpose of Run
Temperature,°F
Feedstock
Results and Conclusions
K1-K7
Figure 18
12/5-12/72
1O2
Study influence of gaseous components on the
rate of gasification at 3O% char burnoff
162O
Char from Loveridge coal pretreated at 16OO°F
1. Gasification kinetics data on char from
Pittsburgh Seam coal were obtained at
char burnoff.
K2-A-K7-A
No Change
12/12-15/72
71
Study influence of gaseous components on
the rate of gasification at 6O$> char burnof
162O
Combined K1-K7 Product
1. Gasification kinetics data on char fron
Pittsburgh seam coal were obtained at
char burnoff.
Run Number
Unit Configuration
Date of Run
Run Duration, hours
Purpose of Run
Temperature, °F
Feedstock
Results and Conclusions
K1O-K16
Char inlet line through bottom of the vessel
1/2-9/73
1O2.5
Study influence of gaseous components on the
rate of gasification at 3O% char burnoff.
17OO
Char from Loveridge coal pretreated at 16OO°F
1. Gasification kinetics data on char from
Pittsburgh seam coal were obtained at
char burnoff.
Kll-ArK16-A
No change
1/1O-17/73
74.5
Study influence of gaseous components on t
rate of gasification at 6O$ char burnoff.
17 OO
Combined K1O-K16 Product
1. Gasification kinetics data on char were
obtained at 6O% burnoff.
-------
Table 30
Summary Qf T.tnillH-PhAlft f!1aim Rnaotlnn
Run Number
Unit Revisions
Date of Run
Run Duration, hours
Purpose of Run
Liquid Feed
Results and
Conclusions
Cl
12/13/72
2.4
First exploratory run, 320°F
1. Reaction goes readily.
2. Little soluble sulfur In
product liquor.
3. Difficulty in removing product
as Wackenroder's solution in
the condensate.
C2
Enlarged the Condensate
Receiver
12/15/72
2.6
Repeat Cl, get data for sulfur
balance
6$ SO, in HjO
1. About 40$ conversion to
elemental sulfur.
2. Sulfur collection still
difficult.
C3
None
12/21/72
4.1
See If 31O*F operation improves
sulfur collection
1. About 42$ conversion.
2. Sulfur collection O.K., probably
due to large change in sulfur
viscosity, 320-310'F.
Run Number
Unit Revisions
Date of Run
Run Duration, hours
Purpose of Run
Liquid Feed
Results and
Conclusions
C4
Run Number
Unit Revisions
Date of Run
Run Duration, hours
Purpose
Liquid Feed
Results and
Conclusions
Run Number
Unit Revisions
Date of Run
Run Duration, hours
Purpose of Run
Liquid Feed
Results and
Conclusions
None
1/5/73
6.S
Increase liquid flow fourfold
to increase holdup in column.
1.6S$ SO, in H20
1. Very slight increase in
conversion over C3.
C5
C7
Additional gas rotameter brought
onstream
1/15/73
2.S
Examine effect of thlosulfate
addition to feed.
lOf) Na,S203 in H20
1. About 35$ of thiosulfate
decomposed.
2. Conversion of feed gases to
sulfur was 62$.
Packing reduced to 1/2 of column. Packing reduced to 1/4 of column.
1/30/73 2/5/73
6.2 6.7
Examine effect of reducing packed reactor volume
51.7$ conversion.
Conversion decreased.
CIO
Recycle Liquor
43.1$ conversion.
Cll
None
2/23/73 '
5.5
-Study effect of changing gns concentration
H20, one batch, uo circulation
None
3/3/73
6.0
None
1/19/73
7.9
Study recycle of liquid feed.
Recycle liquor
1. 67$ conversion to elemental
sulfur.
2. No buildup of acidity, or
sulfur compounds in product
liquor.
3. Conversion better than with
once-through operation.
C9
Packing renoved. System repiped
to allov gas to bubble through
batch pool of liquid.
2/19/73
5.5
Increase liquid volume by removing .
packing and bubbling gas through
liquid.
H20, one batch, no circulation.
68.5$ conversion in half full
column was better than full
packing column.
1. Conversion decreased i
2. System is betwcon first and second order.—
- 110 -
-------
Table 31
Summary of Carbon BumiD Coll
Run Number
Unit Configurations,
Date of Run
Run Duration, hours
Purpose of Run
Temperature, *F
Inert Bed Material
Feedstock
Shutdowns
Results and Conclusions
CDC-1
Figure 19
1/25/73
2
Explore -the behavior of the car-
bon burnup cell
1850
28 x 35 mesh dead burned dolomite
D-ll Product
Large temperature gradient bet-
ween bottom and top of bed
1. Low bed expansion gives poor
mixing and heat transfer*
2. Ash slagging is not a problem.
3, 10O$ burnoff of carbon In char.
CDC-2
No <±»oge
1/29-11/73
41
Is CBC operable »t higher
tempera tare
1880
35 x 48 mesh dead burned dolomite
Combined K2A-KTA Product
Slag formed at the bottom of
reactor, sudden increase in bed
temperature at the bottom
1. Dead burned dolomite picks
up sulfur, fluidizing char-
acteristics ol bed changed.
2. Ash slagging.
CBC-3
No Change
2/1/73
13
Explore the behavior of CBC
at low bed expansions
1850
28 x 35 nesh dead burned dolomite
Combined K2A-K7A Product
Deposit in steam generator
1. Run operable even wltb
spread of temperature.
2. No ash slagging.
Run Number
Unit Configurations,
Date of Run
Run Duration, hours
Purpose of Run
Temperature, °F
Inert Bed Material
Feedstock
Shutdowns
Results and Conclusions
CDC-4
CBC-5
CBC-6
No change
2/5-7/73
13
No Change
2/8, 9/73
12.5
Operation at higher bed expansion Operabllity of combustion at
different temperature levels
1850
35 x 48 mesh dead burned dolomite
Combined K2A-K7A Product
Voluntary
1. Good operability.
2. Good mixing of bed.
3. Low temperature spread.
4. No ash slagging.
1750-1850
28 x 35 mesh silica sand
D-12 Product
Voluntary
1. Good operability in broad
range of teoperatures.
2. No ash slagging.
No change
2/12-13/73
9
Run at 19OO*F
19CO
28 x 48 mesh silica sand
D-12 Product
Water in supply air (rotameter
ball stuck)
1. Good operability in CBC
combustion.
2. No ash slagging.
Run Number
Unit Configurations,
Date of Run
Run Duration, hours
Purpose of Run
Temperature, °T
Inert Bed Material
Feedstock
Shutdowns
Results and Conclusions'
CBC-7
CDC-8
No change •
2/14/73
6
Same as CBC-6
1900
35 x 65 silica send
D-12 Product
Water in supply air
1. Good operablllty in CBC
combustion.
2. No ash slagging.
No chinge
2/15-2/16/73
26.8
Demonstrate operability in
a long run at high temperature
1900
28 x 65 mesh silica sand
D-12 Product
Voluntary
1. Oporability proved.
2. No ash slagging.
- Ill -
-------
VIII. PRETREATMENT STUDIES
A. Introduction
Eastern steam coals become fluid when heated through the temperature range of
roughly 70O-850°F and, therefore, require pretreatment to prevent caking at
gasification conditions. Prior to the present EPA contract work, considerable
background experience had been developed at Consol Research on pretreatment of
highly-fluid Pittsburgh Seam coals for subsequent fluid bed processing at
atmospheric pressure, i.e., for either low-temperature carbonization or gasifi-
cation.
The work carried out under the first year of Contract No. EHSD Tl-lS^1) led to
the following major conclusions relative to previous work on this subject;
1. Pretreatment under elevated pressure was less effective and the pretreat-
ment operation was subjected to more severe operability restraints than
the corresponding operation at atmospheric pressure.
2. A higher severity of pretreatment is required to establish operability
during gasification at elevated pressure.
3. The severity of pretreatment required to establish operability increases
with increasing particle size. This effect is somewhat unfortunate in
that economic considerations favor use of a relatively coarse coal feed
size consist.
Conditions were established in the limited amount of work carried out wherein
operability was obtained in the pressurized pretreatment-gasification of high-
sulfur Pittsburgh Seam coals. But the amount of preoxidation demonstrated was
in excess of that imposed by the adiabatic restraints on the process.
Illinois No. 6 coals, however, were found to be operable in the pressurized
pretreatment-gasification system within the adiabatic restraints of the process.
LTC chars are likewise satisfactory feedstocks for the process, and these may
be produced from Pittsburgh Seam coals by available processes, such as the
Consol LTC process or the Lurgi flash carbonization process.
In the intervening period between expiration of the first year of the contract
and its extension, additional process review as well as experimental work on
the pretreatment problem were carried out at Consol Research. This work con-
firmed and extended earlier conclusions on the pretreatment problem as outlined
below. These conclusions refer to the main objectives of the pretreatment work,
i.e., to produce a relatively coarse high-density product, within the adiabatic
restraints of the process, which is an operable feed to the gasifier.
1. Pretreatment is most effective at a given level of preoxidation when the
amount of devolatilization accompanying it is maximized. Increased
temperature is most effective in this regard but also increased residence
time is important. Operability limitations in the pretreater itself
determine the maximum level of devolatilization that can be achieved con-
current with preoxidation. Increased pretreater temperature also has an
adverse effect in that it causes a decrease in particle density.
- 113 -
-------
2. Operability in the pretreater is improved when the oxygen partial pressure
to which the coal is exposed is increased. The above holds true at a
constant feed ratio of oxygen to coal. Another favorable effect of in-
creased oxygen partial pressure is an increase in particle density of the
pretreated product.
3. Improved operability is experienced in a high temperature pretreater if
the coal is injected into a fluidized bed of inert solids. This maximizes
preoxidation of the feed coal and minimizes competition for the input oxygen
by the partially devolatilized coal.
4. Operability is improved in a high temperature pretreater by premixing or
"precoating" of the coal feed with fine char or other inert solids.
5. Multistage pretreatment with increasing temperature between stages mini-
mizes the required level of preoxidation to produce an operable gasifier
feedstock and also maximizes the particle density of the pretreated product.
6. Improved operability in a high temperature pretreater is obtained by use of
the draft tube principle. In the limit the pretreater becomes in effect an
air-blown carbonizer.
The last conclusion is derived from a review of experience obtained in the pilot
plant carbonizer operated as part of the CSF Process at Cresap, West Virginia,(14)
which showed that improved operability in the pretreatment step could be achieved
by incorporating the concept of a draft tube. The coal feed enters into the
draft tube where it is mixed with a very high rate of circulating inert bed solids,
i.e., char. The Cresap experience shows that this technique causes a significant
improvement in operability in the handling of highly caking feedstocks. Thus, it
was anticipated that higher temperature operation with more thorough decaking of
the coal with a limited extent of preoxidation would be possible.
Two pretreatment processing techniques incorporating the draft tube concept were
investigated. A schematic diagram of the first process is shown in Figure 28.
Dilution, in the draft tube, of the incoming raw coal by a large amount of
decaked coal from the external bed was to allow the temperature of preoxidation
to be raised sufficiently to decake the coal.
The initial preoxidation reactions occur in the draft tube where the inlet oxygen
partial pressure is 3.15 atm at 15 atm system pressure. Any oxygen not consumed
in the draft tube is reacted in the external bed above the tube. The external
bed supplies additional residence time to allow further devolatilization which
completes the decaking of the coal. The bed is fluidized at a low superficial
velocity with a small fraction of the total air required for adiabatic preoxidation.
A variation on this method is to preoxidize in two stages with injection of pre-
viously oxidized coal into the draft tube of the second stage.
The principle of the preoxidation method of pretreatment is to convert the coal
to a more rigid structure via oxidation such that the fluidity is severely
reduced when the coal undergoes pyrolysis. The Seeded Coal Process shown in
Figure 29 operates on just the reverse principle and actually utilizes the natural
fluidity of the coal. In the process visualized, char would be circulated at a
high rate by means of lift gas through a draft tube immersed in a normal fluidized
- 114 -
-------
Figure 28
PREOXIDIZED COAL VIA DRAFT TUBE
"COARSE"
COAL
AIR
- 115 -
-------
Figure 29
SEEDED COAL PROCESS
11OO°
F -14OO°F
"FINE"
COAL
RECYCLE PRODUCT GAS
or
AIR
AIR
- 116 -
-------
bed. Coal and fine seed char would be fed into the draft tube. The external
fluidizedbed would be maintained at 1000-1400°F, either by injection of air or
hot fluidizing gas from the gasification step. The coal melts, smears out over
the surfaces of the seed char and external bed material and then solidifies on
the completion of pyrolysis.
B. Draft Tube Studies
The pretreater-preoxidizer vessel (Section IV-C) was modified by installation of
s. draft tube to promote internal mixing of the fluidized bed and, therefore,
increase operability. Schematic diagrams of the four draft tube configurations
which were tried are shown in Figures 30 and 31 along with some pertinent
dimensions.
The same modelv2/ used to predict the performance of a vertical pneumatic
transfer line can be applied to a draft tube immersed in a fluidized bed. The
pressure gradient across the fluidized bed external to the draft tube supplies
the driving force to move the bed particles and feed particles through the tube.
Batch beds of precarbonized char were used at 15 atm system pressure. Normally,
all gas flows were N2. Suitable pressure taps were provided to measure the
density of the external bed and the AP across the draft tube. Thermocouples
measured the temperature of the material in the top end of the draft tube and
of the external bed.
With the vessel as shown in Configuration A (Figure 30), 28 x 100 mesh char was
used in the external bed at 9OO°F. The solids flow rate through the tube was
determined by periodically feeding char at room temperature and at known rates
into the draft tube. From the measured decrease in temperature of the material
in the top of the tube, the solids rates were calculated by heat balance.
Results showed flow rates in the range of 500-900 Ib/hr. Using the measured
AP's across the tube, calculations involving the pneumatic transfer line model
showed that, to achieve solids flow rates as high as were indicated by the heat
balances (460,000-840,000 lb/hr-ft2 of draft tube cross section), some of the
fluidizing gas from the external bed must have entered the tube, along with the
solids.
In Configuration B, with a larger diameter draft tube, no quantitative measure-
ments were made. However, qualitative checks by feeding cold char showed that
the solids throughput was in excess of 1000 Ib/hr.
In Configuration C, a different measurement technique was used. The external
bed was 48 x 100 mesh char, operated at 150O°F and 15 atm. system pressure. At
this higher temperature it was shown that all oxygen was consumed when air was
fed to the draft tube. By means of bed AP measurements the bed level was dropped
below the weir to a level corresponding to the top of the draft tube. Then, air
was substituted, mol for mol, for some of the nitrogen entering the solids feed
line. From the measured temperature rise of the material in the top of the tube,
the solids flow rate was calculated as 1500 Ib/hr by heat balance.
Data obtained with Configuration C showed that most of the fluidizing gas fed to
the external bed had passed through the draft tube. Also, particle Reynolds
numbers inside the draft tube were barely above the Stokes Law range, a situation-
not conducive to good mixing. Configuration D (shown in Figure 31) was obtained
by modifying Configuration C as follows: To allow installation of an external
- 117 -
-------
FIGURE 3O
CONFIGURATION OF DRAFT TUBES USED IN DRAFT TUBE AND SEEDED COAL TESTS
oo
I
\
RECYCLE
( CONFIGURATIONS A , B and C )
RECYCLE
100 x 28 x
200 M IOOM
COAL CHAR
\
4"
AIR
100 x
2COW
COAL
28 x
IOOM
CHAR
48 x
ICO.VI
CHAR
RECYCLE
AIR
N2
N2
ICOx 48 x
"£, 200M IOOM
2 COAL Cj±AR
AIR
CONFIGURATION
Dimension A 8"
Dimension B 24"
Drofi Tube .500"O.D. x 444" I.D. >
Cool Feed Line .25O"O.D. x .180" I.D.
tip positioned halfway into skirt
Accelerating Gas Line —
CONFIGURATION B
8"
24"
750"O.D,x.680"l.D.
.250"O.D.x.l8O"l.D.
tip positioned I inch above bottom of tube
CONFIGURATION C
6"
34"
75O"O.D. XJ68O"I.D.
.25O"O.D. x.ISO" I.D.
tip positioned 5 inches above bottom of tube
.375"O.D.x.305"I.a
tip positioned I inch above bottom of tube
-------
Figure 31
Configuration of Draft Tube Used in Seeded Coal Tests
Recycle
(Configuration D)
B
Accelerating
Gas (N2)
48 x
100 M
Char
Air
N2
CONFIGURATION D
Dimension A
Dimension B
Draft Tube
Coal Feed Line
Accelerating
Gas Line
External Baffle
Internal Baffle
34"
.750" O.D. x .680 I.D.
.250" O.D. x .180" I.D.
Tip positioned 5-inches above bottom of tube
.375" O.D. x .305" I.D.
Tip positioned 1-inch above bottom of tube
3-5/8" x 1-3/4"
Elliptical, 60° from the horizontal
3/8" D x 60° Cone
Tip positioned 1/2-inch above coal feed line
- 119 -
-------
baffle, which would maintain fluidization of the external bed, the draft tube
was raised two inches and the inlet lines were lengthened accordingly. An
elliptical baffle 3-5/8" x 1-3/4" x 1/16" thick was welded to the accelerating
gas line below the mouth of the tube at a slope of 60° from the horizontal. To
help promote mixing, a conical baffle was installed inside the tube with the
apex of the cone positioned one-half inch above the end of the coal inlet tube.
Tests with an inert bed of 48 x 100 mesh char at 1500°F and 15 atm. system
pressure showed that the external baffle was effective. The solids circulation
rate upward through the tube was measured as before by substituting a known
amount of air for some of the nitrogen entering the solids feed line. From the
measured temperature rise, the solids flow rate was calculated as 900 Ib/hr by
heat balance. Calculations involving the pneumatic transfer line model showed
that without the external baffle about 270 of the 340 SCFH of nitrogen fed to
the bottom of the external bed had entered the draft tube, whereas with the
baffle, the flow was reduced to about 60 SCFH. This 60 SCFH corresponds closely
to the void volume associated with 900 Ib/hr of solids at the fluidizing condi-
tions which existed in the external bed.
C. Pretreatment Using the Seeded Coal Process
In the Seeded Coal Process, finely sized coal and seed char are fed to a draft
tube. The external bed of decaked material is maintained in the temperature
range of 1000-1400°F by reaction with air. The coal melts, smears out over the
surfaces of the seed char and external bed material, and then solidifies on
completion of pyrolysis.
Thirty-one exploratory tests of the concept of the Seeded Coal Process were
completed. Summarized conditions and results are shown in Table 32. Ireland
Mine coal, sized to 100 x 2OO mesh, was used. Detailed discussion of the results
of individual runs will not be given because, as will be shown below, the general
level of operability was severely hindered. While moderate agglomeration may be
tolerable in a process vessel, the clearances in our draft tube were such that
even small agglomerates were sufficient to jam up the draft tube.
Except in the tests where overt plugs occurred, as noted in Table 32, all the
product material which overflowed the weir contained various ammoUnts of agglo-
merates. Agglomerates were recovered from the product material by screening.
No agglomerates larger than 10 mesh were found. Microscopic examination of the
product at 30 magnification showed that smearing of the melted coal occurred
roughly in inverse proportion to the amount of agglomerates which had formed.
The smeared coal material was readily apparent since it has a higher reflectivity
than does the surface of the precarbonized char used in the external bed, and as
seed material.
Nonuniform smearing showed itself by the presence of agglomerated external bed
particles cemented together by bridges of fresh coal material. In some tests,
the bridges were absent but agglomerates had been formed by film-to-film contact
of the fresh coal material.
The evolution of the reactor configuration was as follows:
The first draft tube was installed in the "boot" reactor used in our 1971 work.
The coal feed line was placed halfway up into the draft tube skirt which was
0.75" O.D. (Configuration A, Figure 30). The results of the preoxidation run
described later, and Tests 1 through 5 (Table 32) showed that a possible cause
- 12O -
-------
Summary of Conditions and Results for Seeded Coal Tests
Ireland Mine Coal (10O x 2OO Mesh)
Test
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
23
26
27
28
29
3O
31
Draft Tube(')
Configuration
"F
120O
1200
13OO
14OO
14OO
1225
1225
1225
13OO
14OO
1100
12OO
13OO
14OO
12OO
125O
125O
1400'
1500
1OOO
1OOO
1OOO
1500
9OO
1OOO
1100
12OO
13OO
14OO
150O
1300
Fluldlzing Velocity: 0. 3O ft/sec at top of external bed
Percent of
Size, Tvler Mesh Feed Rater Ib/hr Adiabatlc Duration of Gas Flow to
Bed Seed . Seed Preoxidation Coal Feeding, Gas Flow to Coal Accelerating
CJiar Char Coal Char Level(a) Minutes Feed Line. SCFH Section. SCFH
28
48
I2.O 0 48 i
2.O O 220
1
28;x 1OO 2.O 5 220 3.3 6O
12.O 5 220
1.5 15 220 <
1OO
(')
2
•28 x 10O
1
(
1OO
)
O
1OO 3
O 3
0 6
3
3
0
3
3
3
2
3
3
100 9
O
1
-1OO -1OO 4
3
*** 1
3
3
i
3
3
5
3
7
3 1
3
3
5
3
3
O
3
O
5O 0
54 96
53 57
37 113
72 78
85 85
3
5
entsilL
Fewer agglomerates than In
Test 2
Fewer agglomerates than in
Test 4
Mouth of tube choked with
agglomerates
Mouth of tube choked with
agglomerates
Mouth of tube choked with
agglomerates. Repeat of
Test 1O
Fewer agglomerates than in
Test 15
Fewest agglomerates of all
tests through 18
Caked in coal feed line
Caked inside draft tube
Little or no smearing
Fewest agglomerates of all tests
Caked in draft tube
24 H maximum agglomerate size
Test
-SSs.
1
2
3
4
5
6
7
a
9
1O
11
12
13
14
IS
16
17
18
19
2O
21
22
23
24
25
26
27
28
29
3O
31
See Figures 3O and 31.
See Figure 32 for adlabatlc preoxidation Itvel at given bed temperature.
») Agglomerates formed in all tests but no plugs occurred except when coal caked In the draft tube.
(•) Seed char waa not fed.
-------
of inoperability was very poor mixing of the feed coal and external bed material
in the skirt of the tube. Test 5, where the feed coal was premixed with ten
parts of inert seed char, produced fewer agglomerates than in any of the pre-
vious runs.
In Configuration B, the diameter of the draft tube was increased to allow the
coal feed line to be positioned inside the tube. No marked improvement in
operability was noted. In some of the tests, as noted in Table 32, the mouth
of the tube became choked with agglomerates which entered the tube from the
external bed. The radial clearance between the coal feed line and the wall of
the draft tube was about 0.22 inch. Very few agglomerates (say 10 x 14 mesh)
would be required to bridge the opening.
For Configuration C, a new vessel was made. Elimination of the reactor boot
allowed a longer draft tube to be installed. To improve internal mixing, a
separate gas stream was fed into the mouth of the tube in order to accelerate
the external bed particles. The coal was fed through a concentric line whose
tip was positioned five inches above the mouth of the draft tube, as shown in
Figure 30. To decrease the sensitivity toward choking by any agglomerates
which may form, the size consist of the external bed of char was decreased to
48 x 100 mesh. Only two temperatures, representing the extremes at which the
Seeded Coal Process was expected to be operable, were used with this vessel
configuration. Agglomerates were recovered by screening at 20 and 48 mesh. No
agglomerates larger than 20 mesh were found.
In most of the tests, air was fed to the draft tube along with the coal. In
principle, the Seeded Coal Process does not require prior preoxidation of the
coal, although this may be desirable in order to alter the fluidity charac-
teristics of the melted coal. Since the retention time of the particles in the
tube was very short (about one-half second), air was used during the experiments
primarily to aid in bringing the feed coal particles to the bed temperatures
quickly.
Figure 32 shows the amount of preoxidation, with adiabatic constraint, required
as a function of temperature, the level of preoxidation being defined by;
percent preoxidation = 100 (Ib 02 reacted/lb dry feed coal). The basis for the
calculations are shown on the plot.
After Test 23 was completed, the test data (Draft Tube Studies) obtained with
the Configuration C vessel were used in calculations involving the pneumatic
transfer line model. Results showed that most of the fluidizing gas fed to the
external bed had passed through the draft tube. Thus, the external bed in Tests
6 through 23 was a moving, fixed bed, an ideal climate for agglomerate formation
from particles which had an incompletely solidified film of fresh coal material.
The calculations also showed that the particle Reynolds numbers inside the draft
tube were barely above the Stokes Law range, a situation not conducive to good
mixing of the feed coal with the external bed material passing through the tube.
Appearance of the agglomerates collected showed that they were coal rich, i.e.,
incomplete mixing of the coal feed with the char bed solids had occurred. In
Configuration D, an external baffle kept most of the fluidizing gas out of the
draft tube, and an internal baffle was used to promote solids mixing in the
draft tube.
- 122 -
-------
-------
Seven tests were made with the Configuration D draft tube, using an external
bed of 48 x 100 mesh char at 15 atm. system pressure. Conditions for the runs
are given in Table 32 . Tests were made at temperatures from 900 to 1500°F, in
100° increments. Temperature limits of operability were established as follows:
(l) at 900°F little or no smearing occurs as was shown by presence of coal-
derived material in the form of hollow spheres in the bed after the run, and
(2) at 1500°F caking occurred in the draft tube.
The products from the runs at 1000 to 1400°F all showed more uniform smearing
than in any of the previous runs without the internal baffle. At the end of each
run, the system was depressured and the bed was drained by removing the coal inlet
line. The hot bed material was quenched rapidly by contact with dry ice in the
catchpot. The entire bed material then was screened at 28 and 48 mesh. A
characteristic of all the run products was that all contained some +48 mesh
agglomerates which were external bed particles cemented together by a thin film
of coal-derived material. No agglomerates larger than 28 mesh were found. The
fewest agglomerates occurred at 1300°F, indicating that this may be the optimum
temperature with respect to uniformity of smearing. The amounts of +48 mesh
agglomerates which formed are listed below:
Temperature, +48 Mesh Agglomerates/Wt . $
_ ^F _ _ of Bed Inventory _
1000 18.0
1100 16.0
1200 15.5
1300 8.0
1400 10.7
The particle density, measured in mercury, for the +48 mesh agglomerates formed
at 1300°F had a high value of 85 lb/ft3.
An attempt was made to run for a prolonged period at 1300 °F and 15 atm. system
pressure to determine the size distribution of the "equilibrium" product. To
simulate the seed char in the commercial embodiment (fines from the internal
cyclones in the gasifier) an initial external bed of -100 mesh precarbonized
char was established. Then, 100 x 2OO mesh Ireland Mine coal and additional
-100 mesh char were fed to the draft tube at rates of 2 and 4 Ib/hr, respec-
tively. The fine char contained a considerable amount of -325 mesh material
which was elutriated from the reactor. The outlet piping system of the present
equipment was not designed to handle large amounts of solids. The run had to
be terminated after 35 minutes of feeding coal because the outlet system began
to plug. Thus, an equilibrium bed was not established. However, analysis of
the bed showed that it contained 50 wt . % of +100 mesh agglomerates, with a top
size of 24 mesh. This material had a high particle density of 82 lb/ft3.
The high particle density achieved is favorable, in that "smearing" of liquid
coal over the seed particles apparently is occurring as desired.
The small size of the existing equipment precludes any further meaningful
studies of the Seeded Coal Process . The radial clearance between the inlet line
and the wall of the draft tube is only 0.15 inch. The mouth of the tube eventu-
ally would become choked by the larger agglomerates which inevitably would be
formed.
- 124 -
-------
D. Preoxidation of Ireland Mine Coal
The first draft tube (Configuration A) was sized to allow the use of 100$ air
(3.15 atm. PQ ) as carrier gas to the draft tube, consistent with the maximum
capacity of the overflow weir (16 Ib/hr of preoxidized coal). At the conclusion
of the draft tube tests, the bed temperature was decreased to 850°F and "coarse"
Ireland Mine coal (24 x 100 M) was fed into the draft tube at conditions shown
below;
Coal Feed Rate 16 Ib/hr
Carrier Gas to Draft Tube, SCFH
Air 62
Fluidizing Gas to Bottom of External Bed. SCFH
Recycle 400
Air 15
N2 110
Total Inlet Flow 587
jo Preoxidation Equivalent to Air Fed
In Draft Tube 6.9
In External Bed 1.7
Total 8.6
The total air flow corresponded to adiabatic preoxidation, 8.6$ at 850°F (Figure
32 ). After ten minutes of coal feeding, the AP across the external bed in-
creased sharply indicating that massive caking had occurred. On disassembly of
the vessel, the bed contained large chunks of agglomerated material. The coal
particles had cemented together the char particles present in the start-up bed.
A series of runs were then made in which a finer size consist, nominally
65 x 100 mesh, of the raw coal feed was used in conjunction with two stages of
preoxidation. From prior work both expedients are in the direction of reducing
the excessive preoxidation required in single stage treatment of relatively
coarse coal.
a. First Stage - Run PR1 - The draft tube was removed from the reactor for this
run. At the low temperature level of 675°F chosen for the run, the draft tube
probably would give no benefit since the plastic range of the coal lies at higher
temperatures and caking/agglomeration should not be a problem.
Programmed' conditions called for the adiabatic level of 6.3$ preoxidation. The
actual coal feed rate was somewhat higher than the programmed value, with the
result that 5.7$ preoxidation was achieved. The run was completely operable
and scheduled shutdown was made after making 143 pounds of steady-state product.
b. Second Stage at 775°F - Run PR1B - After steady-state conditions at 750°F
had been reached in Run PR1A in which complete operability was demonstrated, the
bed temperature was raised to 775°F for Run PR1B. The preoxidation level was
increased purposefully to give a conservative total level for both stages of
11.3$ preoxidation, which is 50$ greater than the adiabatic level at 775°F.
This run also was completely operable with no caking or .agglomeration occurring.
- 125 -
-------
The Run PR1 product and all the air were fed into the Configuration D draft tube.
In a preoxidation run, the draft tube probably makes no intrinsic contribution
to the process. In the tube, the incoming first-stage coal product and air are
immediately diluted with a large amount of external bed material. The draft tube
was used to assure a smooth transition from the start-up conditions involving an
initial bed of inert char and to prevent localized hot spots caused by the highly
exothermic preoxidation reactions. Axial traverses across the entire bed showed
that the temperature was constant to within 4°F.
c. Second Stage Preoxidation Attempt at 800°F - In an effort to obtain a more
highly devolatilized and, therefore, a more thoroughly decaked product, the bed
temperature was raised to 800°F via the electrical heaters at the end of the
material balance period of Run PR1B. After about one-half hour, the pressure
drop across the bed had increased and a large temperature gradient had developed.
Inspection of the reactor after shut down showed that it was full of large chunks
of agglomerates.
Detailed results of Runs PR1 and PR1B are given in Tables 33 through 36. Inspec-
tions of the preoxidized coal product from Run PR1B show that the material was
considerably swollen and had a particle density of 49 lb/ft3, compared with the
raw coal value of 82 lb/ft3. This result was not anticipated because our 1971
work had indicated that the smaller size fractions of the preoxidized coal pro-
ducts had suffered relatively little swelling. Some comparisons are shown below
with data from two runs made during the earlier work:
Run Number
Raw Coal Size Consist
Temperature, °F
Percent Preoxidation
Particle Density, lb/ft3
Mean of Entire Product
65 x 100 mesh Fraction
IP
2P2
PR1B
*—24 x 100 Mesh —
700 750
19.5 18.6
61.6
76.8
53.1
73.1
65 x 150 Mesh
775
11.3
48.6
54.4
The lower density, 65 x 1OO mesh, product produced in PR1B as compared with pre-
vious work may be due to the lower level of preoxidation and/or the higher
temperature used.
The laboratory shock-heating test showed that the Run PR1B product would have
been inoperable at gasifier conditions. The amount of agglomerates formed in
the test was greater than for the Run 2P2 product which actually was fed to the
gasifier at process conditions and which was shown to be inoperable.
E. Conclusions - Pretreatment of Ireland Mine Coal
The results of the exploratory studies strongly indicate that future studies of
the Seeded Coal Process should be made with a larger reactor unit in which the
draft tube diameter would be at least 2 inches (vs .680 inch at present). The
potential advantages of the process are that it will supply a feedstock which
assuredly is operable with respect to caking/agglomeration at gasifier condi-
tions, and that it can produce a dense, closely sized feedstock substantially
free of fines which will allow a higher gasifier throughput than was envisioned
previously.
- 126 -
-------
Table 33
Preoxidation Conditions and Results
System Pressure; 15 atm (206 psig)
Feed Coal Size Consist 48 x 150 Tyler Mesh
Ireland Mine Coal
Run Number PR1 PRIB
1st Stage 2nd Stage
Temperature, °F 675 775
Inlet 02 Partial Pressure, atm .18 .40^*'
Fluidizing Velocity (top of bed), ft/sec 0.30 0.30
Moisture in Coal,, Wt as Fed 1.83 0.71
Input
Coal Feed Rate, Ib/hr (dry basis) 1O.44 5.64
Lift Gas. SCFH
Air 35 18.4
N2 170 126
Fluidizing Gas. SCFH
Recycle 417 417
N2 00
Purges (N2), SCFH to bed 5 5
Purges (N2), SCFH above bed 7 11
Output
Exit Gas Rate, SCFH (dry basis) 212 159
Exit Gas Composition. Mol % (dry basis)
02 .14 <.01
C02 .76 .89
CO . 30 44
CH4 .09 .44
C2H6 .03 .10
C2H4 <.01 .01
C3H8 .02 .03
C3H6 <.01 .01
COS .02 .03
S02 <-01 .01
N2 (by difference) 98.64 98.04
Flow Rate, SCFH at top of bed 642 585
Watert Ib/hr (corrected for coal moisture)
Condensate .677 .258
Moisture in Exit Gas .008 .006
Tar, Ib/hr .106 .092
Preoxidized Coal, Ib/hr 9.89 5.31
$ Preoxidation(2) 5.71 5.67
Duration of Steady-State Period, hr 14.1 2.0
Total Product, ex. overhead fines, Ib 143 10.4
(1) In draft tube.
(2) Lb 02 consumed/100 Ib dry feed to stage.
- 127 -
-------
Table 34
Properties of Feed Coal and Products
Ireland Mine
Run Number Feed Coal
Temperature, °F
Hydrogen, Wt $ (dry basis) 4.87
Carbon 69.88
Nitrogen 1.25
Oxygen (by diff.) 6.69
Sulfur 4.65
Ash 12.66
Volatile Matter 41.3
PR1
PR1A
1st Stage
675
4.64
70.37
1.08
6.29
4.60
13.02
35.5
PR1B
2nd Stage
750
775
3.52
70.44
1,24
4.23
4.55
16.02
3.52
70.60
1.35
3.92
4.52
16.10
24.2
23.8
Size, Tvler Mesh
+28
28 x 35
35 x 48
x 65
Wt Jo Density Wt "& Densitv Wt 1o Density Wt % Density
48
65 x
100 x
100
150
-150
16. 1
71.3
11.4
1.2
Mean Diameter, inches(2)
Mean Particle Density,
Ib/ft3(3)
.00717
82
--
--
0.7
27.4
67.3
4.4
0.2
—
—
—
62. O
69.6
76.7
—
0.5
1.8
10.2
42.9
40.0
4.0
0.6
—
—
49.7
46.7
51.5
53.2
—
0.6
6.6
14.8
34.5
38.1
4.5
0.9
—
44.3
39.0
46.4
54.4
56.9
58.8
.00774
67.6
.00914
5O.5
.00923
48.6
(i) Measured in mercury at 1 atm.
(2) Arithmetic mean.
(3) Reciprocal mean.
- 128 -
-------
Run Number
Table 35
Material Balances for Preoxidation
Basis: 100 Ib Dry Feedstock
PR1
Temperature, °F
Preoxidized Coal
Overhead Fines
Tar
Carbon in CO + C02
Hydrogen in H20
Hydrocarbons (Cj-Cg)
Sulfur in S02 and COS
Carbon and Sulfur in Condensate
Coal Oxygen to Products
Total
PR1B
675
94.05
.72
1.01
.67
.73
.17
.03
.10
1.70
775
92.23 ~
1.95
1.63
1.17
.52
.88
.09
.10
1.39
Output, Ib.
99.18
99.96
Run Number
Temperature, °F
Output
To Water
Condensate
Moisture in Gas
To CO2
To CO
To S02
To COS
In Tar
Unreacted 02
Total
Input
From Coal (by diff .)
Lb/hr 02 from Air
Total
Table 36
Distribution of Oxygen in
Products of Preoxidation
PR1
675
Lb/hr Percent
.602
.008
.133
.026
0
.002
.002
.012
.785
.177
.068
76.7
1.0
16.9
3.4
0
.2
.2
1.6
1OO.O
PR1B
.785
Lb/hr
.229
.006
.118
.029
.001
.002
.013
0
.398
.078
.320
775
Percent
57.5
1.5
29.5
7.3
.3
.5
3.4
0
100.0
.398
- 129 -
-------
Within the framework of an adiabatic preoxidation process, use of a fine size
consist of the feed coal coupled with two stages of preoxidation cannot produce
a feedstock which is operable in both the preoxidizer and gasifier. Such a
process variation should reduce the total preoxidation required, but from the
present work the magnitude of the reduction has not been demonstrated. It is
clear, however, that the required total preoxidation level exceeds 11.3$.
Conceivably, a three-stage adiabatic preoxidation with the third stage at 800°F+
in which inlet oxygen partial pressures higher than we are able to provide with
the existing equipment would produce an operable feedstock with a lower level of
preoxidation. However, the evidence obtained so far indicates that even in this
process the product would have a small particle size and a low particle density
which, in turn, would lead to a low gasifier throughput.
All of our work to date has shown that preoxidation at elevated pressure of the
highly fluid coals such as Ireland Mine coal will always require a preoxidation
level in excess of the adiabatic level.
- 130 -
-------
IX. ACCEPTOR SULFUR CYCLE STUDIES
A. Introduction
The reaction,
CaC0
H2S = CaS
H20
C0
2,
is used for both gas desulfurization and acceptor regeneration. Conditions in
the gas desulfurizer and regenerator are such that the equilibrium H2S content
of the cleaned fuel gas is low, and the equilibrium H2S content of the
regenerator- off gas is relatively high.
Run conditions in this study generally simulated those used in the process
design calculations, as given below;
Gas Desulfurizing
Reactor
Regenerator
Temperature, °F
Inlet Gas Composition. Mol jo
H2
CO
CO 2
N2
H2S
H20
Acceptor Conversion,
Mol % of Ca Converted
ca.
1600
17
18
8
46
1.0
9
20
1300
50
50
20
B. Cyclic Runs with Tymochtee Dolomite Feed
Runs A1-A4 and Run A18 were made using Tymochtee dolomite. Reactor conditions
and results for the gas desulfurizer and regenerator are given in Tables 37 and
38^ respectively. For Runs Al and A2, producer gas was simulated by feeding
H2, C02, and H20, and making CO in the reactor by way of the water gas shift
reaction,
H,
+ C02 = CO + H20.
For later runs, operations were simplified by also making all necessary steam
via the shift reaction. Analysis of the exit gases and condensate showed that
shift equilibrium was always attained.
Table 39 gives the analyses of the dry exit gases from both vessels. The CaS
content of exit solids as a function of cycles is given in Table 40. As discus-
sed earlier (Section V-E), a small fraction of the calcium in the stone reacts
with impurities the first time it is heated. The numbers in Table 40 represent
mols of CaS as a percentage of the calcium remaining after reaction with impuri-
ties in the stone is complete. Table 41 lists the size consist of the feed and
of the last sample collected from the regenerator.
- 131 -
-------
Table 37
Conditions and Results for Gas Desulfurizer With Tvmochtee Dolomite Feed
to
I
System Pressure:
System Temperature:
IS Btra.
16OO°F
(206 psig)
Run Number
Acceptor Size Consist, Tyler Mesh
Feed Rate, Ib/hr (half calcined basis)
Solids Residence Time, minutes
Inout. SCFH
Recycle to Bed
H2S
C02
H2
H20
N,
Purges (C02) to Bed
Purges (Na ) above Bed
Recycle Acceptor Lift Gas, above Bed
Output in Cycle
Exit Gas Rate, SCFH (dry basis)
Composition, Mol &
II,
CO
C02
N2
H2S
Outlet Gas, Top of Bed
Composition, Mol 1>
H 2O
H2
CO
C02
N2
H2S
Flow Rate, SCFH, top of Bed
Fluidizing Velocity, ft/sec
Attrition, "f> of Feed Rate
Duration of Circulation with H2S Feed, hr
Removal of Feed Sulfur, %
Removal of Feed plus Recycle Sulfur, £
£ H,S in Outlet/Equilibrium % H2S
Conversion of Acceptor/Pass, mol %
350
3.5
4O
53
15
67
5
15
124 (once-through Na)
1-3
264
9.8
10.2
7.7
72.3
.038
9.3
8.9
9.2
6.9
65.3
.034
525
.78
2.5
6.1
97
93
1.8
21
26O
3.5
51
53
15
68
S
15
158
1-3
154
13.2
20.4
14.4
51.9
.075
9.9
11.8
18.4
13.0
46.8
.068
443
.66
18
2.6
97
92
1.8
28
130
3.5
35
48
65
5
15
92
1-3
148
18.3
17.3
12.1
52.1
.09
1O.9
16.3
15.4
1O.8
. 46.5
.OS
275
.40
5.6
7.1
97
94
2.3
23
10
14
18
11
47
.O835
292
.42
4.9
11.6
96
93
2.6
19
(>) Estimate. Gas chromatograph recorder not functioning.
-------
CO
CO
Run Number
Solids Residence Time, min.
Input. SCFH
H2O
C02
Air
Purges (N2) to Bed
Purges (N2) above Bed
Purges (CO3) above Bed
Output in Cycle
Exit Gas Rate, SCFH (dry basis)
Composition. Mol ^
C02
N2
H2S
S2
Outlet Gas. TOD of Ded
Compos 1 tlon. Mol ^a
II 20
C02
N2
H2S
Flow Rate, SCFH, Top of Bed
Fluidizing Velocity, ft/sec
Regeneration of Acceptor, Mol *ja
System
System
Al
25
187
188
1O
i n
iu
5
1-3
214
9O.2
9.3
.49
— O
46.6
48.1
5.O
.26
~ O
•:',;
388
1.11
6.2
Pressure:
Temperature:
A2
42
1O7
108
1O
1 C\
i\j
5
1-3
133
84.2
15.0
.83
~ O
47.6
47.6
4.4
.49
~ O
279
.80
8.3
15 atm.
130O°F
A4
4O
89
82
7.5
1O
i n
i\J
5
1-3
no
76.1
23.0
.8
.05
43.5
47.3
8.6
.5
.03
186
.53
6.7
(206 psig)
Phase 1
82
82
—
^
2
105
82
17
.885
~ 0
47
47
5
.54
~ 0
171
.48
6.4
8
A18
Off
57
1O7
3
ISO
85
14
.73
~ O
33 .
62
5
.55
~ O
172
.48
11. 0
Pha^UL.
91
53
--
5
76
74
24
2.2
~ O
59
34
. 5
1.10
~ 0
152
.43
12. 0
-------
Table 39
H,S Content of Exit Oases (dry baaig)
Run
Ho.
Al
A2
A4
AS
A6
A7
AS
A9
All
A12
A13
A15
A16
Reeenerator
Gas Desulfurlzer
Cycle
No.
.6
1.7
.8
.5
1.8
2.6
.9
2.0
6.6
9.1
1O.1
11.8
.3
1.4
5.5
7.5
8.6
12.6
—
—
—
3.0
7.O
0(s
.9
5.3
6.8
8.8
11.0
11.9
15.5
.6
1.2
2.4
3.7
5.2
6.6
6.8
7.7
8.5
9.9
1.6
2.8
3.3
4.1
2.5
4.9
7.0
Mol %
HgS
.037
.039
.075
.115
.090
.084
>1.4
.105
.144
.476
.511
.684
.046
.049
.048
.058
.064
.672
.059
.16
—
.122
.094
) .123
.114
.093
.207
.817
.99
.93
.645
.046
.071
.048
.048
.039
.047
.06
.132
.421
.645
.095
.088
.088
.081
.054
.065
.074
$ H,S
Removal
97
97
97
95
96
96.5
<8
87
82
41
37
15
97
97
97
96
96
59
96
89
—
93
94
93
93
94
87
50
40
43
61
95
92
94
94
95
94
93
84
50
23
96
96
96
97
97
96
95
Cycle
No.
1.1
2.1
1.1
.7
2.1
2.5
1.2
2.3
4.5
7.0
9.6
9.8
12.0
.5
1.5
5.7
7.8
9.2
12.5
—
—
1.5
1.7
2.4
7.1
2.2
5.4
4.1
7.0
8.8
11.0
11.9
15.5
.9
1.2
2.5
3.8
5.3
6.6
6.8
7.7
8.5
10.1
1.6
2.8
3.2
4.1
5.6
2.4
4.8
7.1
Mol %
HjS_
.384 '
.492
.831
.80
.82
1.34
.057
.262
.380
.411
.404
.354
.367
.671
.927
1.39
1.69
1.40
1.90
—
—
1.41
2.01
1.68
1.89
2.14
1.44
1.45
1.41
• 1.41
1.35
1.20
..-.1.18
.283
.456
.456
.508
.464
.705
.58
.790
.445
.629
.885
.756
.703
2.27
2.14
1.09
1.32
1.38
$ Recovery of
HaS Fed to
Desulfurirer
23. S
30
32
25
26(0
42(0
2
29
41
45
44
39
40
25
33
fto
53 0
72(0
—
—
54
76
64
72
83
56
64
62
62
59 ,
S3
52
24
39
39
43
39
60
49
67
38
S3
29
31
29
54
51
39
48
SO
A18
A19
(>) The condensate contained ca. 3# of the feed sulfur as elemental sulfur.
(>) Taken during the presulfiding (tap.
- 134 -
-------
Table 4O
Run No.
Al
A2
A4
A5
A6
CO
tn
A7
A3
A9
All
A12
CaS Content
of Exit Solids
Gas Desulfurizer
Cvcle No.
.6
1.6
2.3
.5
.8
1.55 •
.8
1.25
1.8
2.3
2.9
ca. .9
.5
.8
1.2
1.6
2.1
3.6
4.6
12.6
12.6 + 150 min.
2.5
6.0
8.0
11.3
12.9
12.9 + 120 min.
—
—
2.0
2.7
1.2
2.1
3.0
4.0
6.1
7.3
Mol * CaS
32.9
49. 0
60.6
32.1
41.8
56.1
40.4
52.6
58.9'
7O.2
81.1
1.1
15.3
17.2
16.7
20.1
21.2
24.1
27.9
. 42.0
49.9
33.5
66.7
78.8
96.8
98.4
98.4
85.6
94.6
32.6
39.6
26.4
33.9
37.4
42.4
55.3
61.0
Regenerator
Cycle No.
.5
1.5
2.15
.2
.5
.7
1.3
.4
1.4
2.4
3.0
ca. 1.2
.2
.7
1.3
2.O
2.8
3.3
6.2
9.0
10.4
12.4
12.6 + 150 min.
.8
3.4
9.0
12.9
12.9 -I- 120 min.
1.5
1.9
2.3
2.7
.7
4.2
4.8
7.2
Mol $ CaS
16.7
34.7
49.6
5.2
11.8
15.9
28.9
25.8 ^
42.1 I
52.9 1
65. 7J
O.9
7.3
14.5
16.4
17.2
15.6
23.5
24.4
25.5
26.0
30.5
27.8
12.2
32. 0
71.6
85.3
75.9
15.7
20. 2
23.8
25.3
10.2
31.5
34.5
44.9
Includes ca.
4# CaSO4
-------
Table 40 (Cont'd.)
CaS Content of Exit Solids
Gas Desulfurizer
Regenerator
A15
I
H1
CO
I
A16
A18
A19
Cycle Ko.
2.9
4.1
5.3
6.4
7.6
8.8
10.0
4.6
8.6
11.4
13.0
14.2
17.2
.7
3.2
4.1
6.8
7.4
8.5
1O.5
2.O
3.3
5.6
3.5
5.5
7.2
Mol CaS
70.9
72.7
95.7
88.5
95.4
97.5
98.8
77.2
96.6
95.6
96.3
95.6
96.4
23.3
59
65
94
95
96
95.8
35.7
52.5
70.5
43.2
58.3
72.8
Cycle No.
1.6
3.9
6.0
8.3
10.4
.4
1.9
3.0
4.1
5.2
8.3
10.2
12.1
13. 0
14.8
16. 0
17.0
.5
3.0
4.O
6.8
7.3
8.5
9.5
10.5
1.7
3.2
5.6
3.6
6.5
7.2
Mol $ CaS
44.6
64.0
83.7
88.6
92.5
35.8
37.5
48.2
62.2
71.9
88.6
88.5
88.6
89.5
89.3
90.0
89.9
15. 0
49. 0
58.7
81.9
85.0
87.7
88. 0
88.3
26.3
4O.9
58.5
39.0
51.7
57.6
-------
.' Table 41
Consist of Acceptor Feeds and ProductsC*/
Data given on a weight % basis
Smallest size includes fines
CO
-a
Run Number
Si7f Tyler
A2
AS
A6_
A7
A12
A13
A16
20
28
35
43
65
x 28
x 35
x 48
x 65
x 100
-100
Size, Tvler Mesh
20
28
35
48
65
28
35
48
65
1OO
4.7
84.1
9.7
1.4.
100*
6.1
70.5
15.1 61. S.
8.3 23. 0
6.4
8.8
100*
100*
33. 0
62.5
4.3
0.1
0.1
15. S
84.2
61.4
29.2
7.5
1.9
61. 0
29.6
6.3
1.5
1.4
30. 0
• -38.6
31. 4
42.1
41.2
9.9
6.8
3.2
48.4
40. 1
8.2
3.1
48.3
46.6
2.1
4.9 l.O
48.
41."
5.
3.
48.
46.
2.
2 39.9
8 20. 3
1 38. S
1
3
6
1
6.7
35.6
25.7
17.2
14.8
17.3
70.7
7.9
4.1
13.
76.
8.
1.
15.
80.
4.
0.
6
4 37.8
9 36.8
1 25. 4
0
1 91. 0
4 9.0
4
33.4 2.O 23.9
32.1 70.8 19.1
31.5 14.2 22.2
13. 0 34.8
91. O 1OO*
9.0 1°°*
Sizef Tvlcr Mesh
20 x 28
28 x 35
35 x 48
48 x 65
65 x 10O
-1OO
Tvle
Mesh
2O x 28
28 x 35
35 x 48
48 x
65 z
65
100
For once-through Runs AS and A9 analyses are of stone exiting the gas desulfurlzer.
All other product analyses are of the last sample taken from the regenerator.
Nominal size consist.
-------
Run Al
Approximately 97$ of the H2S fed was removed in the gas desulfurizer. Gas exit-
ing the bed contained about twice the calculated equilibrium H2S concentration.
However, only partial regeneration was achieved, and the CaS concentration
continued to build up in the acceptor. Figure 33 presents the concentration of
sulfur in solids exiting the gas desulfurizer and the regenerator as a function
of time. Figure 34 gives the H2S concentration in the dry gas exiting each
vessel.
The hourly attrition rate was 2.5% of acceptor being fed, far higher than the
same acceptor had given in the C02 cycle for the C02 Acceptor Process.(3) The
run was shut down due to difficulties in switching the solids filters.
Run A 2
Run A2 was a refined repetition of Al. The fluidizing velocity in the regenerator
was lowered, and the wall temperature of the sulfur reactor was closely controlled.
In Run Al the wall temperature exceeded 1750°F and calcination of CaC03 may have
occurred. For all later runs, the wall temperature did not exceed the bed tempera-
ture by more than 60°F.
As in Al, about 97$ removal of H2S was obtained in the sulfur reactor; this corre-
sponds to twice the equilibrium H2S content in the exit gas. Regeneration was
incomplete, and the sulfur content of the acceptor went up with time. This is
shown graphically in Figure 33.
The attrition rate was disastrously high, i.e., 18$ of the acceptor fed per pass.
Loss of inventory due to attrition forced an early shutdown.
The acceptor was soft when it came out of the regenerator. Yet all of the
attrited material was collected in the sulfur reactor filters. This led to the
conclusion that attrition occurred in the feeder, the pick-up chamber, and the >
lift line to the gas desulfurizer. The feeder was disconnected from the lift line,
and then, at atmospheric pressure, regenerator bed material from Run A2 was run
through the feeder. About 2$ attrition, as measured by the -150 mesh fraction,
occurred in the feeder alone. Details are given below:
Feed: Run A2 Regenerator Bed Material
Into Feeder,
Wt. $
Out of Feeder,
Wt. 1o
Run A3
+28
28 x 35
35 x 48
48 x 100
100 x 150
150 x 200
-200
0.1
70.5
17.6
6.9
1.3
0.7
2.9
2.0
67.0
17.9
7.0
1.5
2.5
3.2
Run A3 was to be made with 20 x 28 mesh acceptor which required a higher fluidi-
zing velocity, 0.96 ft/sec. It was not possible to establish an even temperature
profile in the sulfur reactor with such a high gas rate, and H2S feeding was never
initiated. As in previous runs, the attrition rate was. high.
- 138 -
-------
(0 =
<0
I K
5 s
s s
- u
0 i u
x-
ox
70
60
0
-------
rn*
N
n
8
I "
5 s
i s
_ LJ J
Ox u
Ho t
O? 3
-r, U
S «
Ki-
IE
•J1QJU
o
-------
m«
CM «
O
0 I u
Ho t
O? 3
-o g
V. ^ *
Sll
- 141 -
-------
Run A4
On the premise that during the OCR C02 Acceptor development work dolomite had
been hardened due to the formation of transient liquid from CaS04-CaS, air was
added to the regenerator to convert about 4$ of the calcium to CaS04. There
was no build-up of sulfate with cycling, since it was reduced in the gas
desulfurizer. The attrition rate was lower than in the comparable Run A2 but
was still imacceptably high. Either 1600°F is too low a temperature for tran-
sient liquid formation or the sulfate was reduced before liquid formation could
occur.
An interesting result of the run was that about 10$ of the total sulfur in the
regenerator offgas was elemental sulfur. Previously, only trace quantities
had been collected.
The sulfur contents of product gases and solids are given in Figures 34 and 35,
respectively. Again, H2S removal was good in the gas desulfurizer and regenera-
tion was incomplete. The concentration of H2S in the regenerator offgas went up
as the sulfur content of the acceptor increased. The run was terminated because
of attrition of the acceptor inventory.
The attrition from Run Al was lower than that from Runs A2, A3, and A4. Condi-
tions in Run Al were substantially different from the others only in that the
wall temperature in the sulfur reactor was higher in Run Al. Hardening may have
involved calcination or sintering because of the wall temperature exceeding
175O°F.
The initial level of CaO conversion to CaS for Runs Al, A2, and A4 was high.
For Run A4 this was expected since the gas desulfurizer bed was presulfided
before circulation of solids began. In Runs Al and A2, it is believed that
extra H2S was fed while the rotameter was being adjusted. This adjustment is
difficult and large slugs of liquid H2S can get through while it is being made.
Run A18
Run A18 was made at conditions duplicating Run A4, except that air was not added.
The attrition rate of Tymochtee dolomite in Run A4 was lower than in the roughly
comparable Run A2, but still too high at 5.6$. Run A2 was made with 28 x 35
Tyler mesh stone and a lift gas rate of 158 SCFH; Run A4 was made with 35 x 48
mesh stone and a lift gas rate of 92 SCFH. While the results of Run A4 were
encouraging with respect to hardening, the lesser lift gas flow may also have
reduced attrition, thereby confounding the experiment.
Run A18 was made with 35 x 48 mesh Tymochtee stone and a low lift gas rate to
more closely duplicate Run A4. No air was used. The attrition rate was 4.9$
compared to 5.6$ in Run A4. It is concluded that hardening due to transient
liquid formation did not occur in Run A4. The results imply that attrition occurs
mostly in the solids pick-up chamber and lift line, the flow rate of lift gas and
the particle size being an important factor.
Although hardening due to transient liquid formation did not occur in Run A4,
such hardening would be expected if the two-stage process pictured in Figure D-l
of Appendix D were used. The two-stage modified sulfur recovery system is
described in detail in Appendix D. CaS is oxidized to CaS04 in a riser
reactor, and the temperatures should be high enough to induce transient liquid
formation with the attendant hardening of the stone. The CaS04 then acts as an
oxidizing agent toward Squires reactor product gas to produce a Claus feed.
- 142 -
-------
Having established the objective of checking the attrition rate, the steam/C02
ratio in the regenerator was also altered from 1/1 to 0.5/1 and then to 1.7/1.
The goal was to find quickly whether this was a key kinetics variable. Results
are detailed in Table 38 and Figure 36, and are summarized below:
Run Number A18-1 A18-2
Mol $ CaS in Feed to Regenerator 35.7 52.5
Mol $ of Entering CaS Converted to CaC03 18 21
H20/C02 1.00 0.53
Cycle Number 2.0 3.3
There is no marked effect of variation of the steam/C02 ratio. However, the
results are obscured by the inevitable increase in acceptor CaS content with
increasing number of cycles.
The experiment should be repeated, using three separate runs.
C. Dolomite from C02 Acceptor Process
In Run A6, the feed was a composite of acceptors which had been used in bench-
scale development of the C02 Acceptor Process under the Office of Coal Research
(OCR). About 40$ of the calcium was in the form of inactive CaO. This material
had shown a low rate of attrition in the previous program, and when cycled in
the presence of H2S, the attrition rate remained low, about 0.2$.
Run conditions are detailed in Tables 42 and 43. Based on the original Ca
present in the stone, 42$ of the Ca was in the form of inactive CaO. Conversion
was set for 10$ Ca conversion of the remaining 58$ of the Ca. The initial
absorption of H2S was good and then rapidly declined after seven cycles. The
regeneration step was incomplete, but instead of improving, as the CaS content
of the acceptor increased, exit H2S content remained the same. This is pre-
sented in Tables 39 and 40, and graphically depicted in Figures 37 and 38.
At the end of the run, the solids feed was stopped and both vessels were run as
batch beds for 150 minutes. This was done to see whether further regeneration
would occur. The H2S content of regenerator offgas fell from 0.37$ to 0.06$,
i.e., conversion dropped to a negligible level. Based on analysis of regenerator
solids, the CaS content fell from 30$ to 28$. There also was an increase in the
CaS content in the gas desulfurizer from 42$ to 50$.
Run A6 showed that there is nothing about our run conditions which causes weaken-
ing of the stone upon cycling. By implication, a strong fresh dolomite should
perform well with respect to both attrition and H2S removal.
About 50$ of the CaC03 was unreactive toward H2S. Similarly, only a portion of
the CaS which was formed was easily regenerated in the "Squires" reactor. Since
this stone had an unusual history, there is no way to predict whether such a
problem might become important to a freshly cycled dolomite.
D.
Runs with Canaan Dolomite Feed
Run A7
Run A7 was made at simulated process conditions to test the activity and attrition
resistance of Canaan dolomite from Connecticut. The attrition rate was low, about
- 143 -
-------
co«
OS
X *
5 8
5 a
_ u J
Oi u
l-o J-
ui
i
80
70
60
50
40
30
20
10
ti
- 144 -
-------
Table 42
Conditions and Results for Gas Desulfurizer With
Dolomite Feed From C02 Acceptor Process
System Pressure: 15 atra. (206 psig)
System Temperature: 1600°F
Run Number
A6
Acceptor Size Consist, Tyler Mesh
Feed Rate, Ib/hr (half calcined basis)
Solids Residence Time, min.
20 x 35
11.3
24
Input. SCFH
Recycle to Bed
H2S
C02
H2
H20
N2
Purges (C02) to Bed
Purges (N2 ) above Bed
Recycle Acceptor Lift Gas, above Bed
Output in Cycle:
Exit Gas Rate, SCFH (dry basis)
Composition. Mol jo
H2
CO
C02
N2
H2S
Outlet Gas. Top of Bed
Composition. Mol $
H20
H2
CO
C02
N2
350
1.7
66
74
29
87
5
15
90
1-2
211
6.2
20.2
13.7
59.8
.105
12.2
5.6
18.2
12.4
51.5
.0947
Flow Rate, SCFH, Top of Bed
Fluidizing Velocity, ft/sec
Attrition, % of Feed Rate
Duration of Circulation with H2S Feed,
Removal of Feed Sulfur, $
Removal of Feed plus Recycle Sulfur, %
is H2S in Outlet/Equilibrium $> H2S
Conversion of Acceptor/Pass, mol %
hr
622
.90
.15
24.9
87
72
2.1
8
- 145 -
-------
Conditions and Results for Regenerator with
Dolomite from C02 Acceptor Process
System Pressure; 15 atm. (206 psig)
System Temperature; 1300°F
Run Number A6
Solids Residence Time, minutes 27
Input. SCFH
H20 131
CO2 162
Purges (N2) to Bed 10
Purges (N2) above Bed 10
Purges (C02) above Bed 5
Output in Cycles 1-2
Exit Gas Rate, SCFH (dry basis) 185
Composition. Mol %
C02 89
N2 11
H2S .26
Outlet Gas, Top of Bed
Compos ition} Mol ^
H20 43.5
C02 53.1
N2 3.2
H2S .16
Flow Rate, SCFH, Top of Bed 301
Fluidizing Velocity, ft/sec .85
Regeneration of Acceptor, Mol % 2.6
- 146 -
-------
u
X f,
t.
I
1
50
n
-------
n<
(O
X
u
_
0 i u
hu J-
±
.6
.0
.8
.4
• 12
6 8
- 148 -
10
12
-------
0.8$>, for the entire run. Good H2S removal in the gas desulfurizer was main-
tained for about eleven cycles. Breakthrough occurred in the gas desulfurizer
because of the build up of CaS which resulted in insufficient CaC03 to react
with the input of H2S.
Details of the run conditions are given in Tables 44 and 45. The sulfur-bearing
contents of exit gases and solids are given in Tables 39 and 40, respectively.
The data are presented graphically in Figures 39 and 40.
When H2S breakthrough occurred in the gas desulfurizer, the solids feed was
stopped and both vessels were run as batch beds for two additional hours. This
was done to see whether additional regeneration would occur or not. There was
a possibility that the CaS was "dead" and might not react any further. After
two hours, the CaC03 content of acceptor in the regenerator rose from 15 to 24
mol %, thereby showing that additional regeneration would occur if residence
time were increased in the regenerator.
Several encouraging conclusions may be drawn from Run A7;
1. The CaC03 in dolomite is effective in removing H?S up to high con-
versions in the solid. At the e
solid had been converted to CaS.
versions in the solid. At the end of Run A7, 98% of the CaC03 in the
2. The attrition problem encountered with the Tymochtee stone is not
common to all dolomites. The average attrition rate for Run A7 was
O.8$ of the acceptor feed rate.
3. The regeneration reaction works to the extent that at least moderate
regeneration of CaC03 from CaS occurred with a forty minute residence
time.
At the end of Run A7, about 70$ of the inlet H2S to the desulfurizer was being
recovered in the regenerator with a solids residence time of forty minutes.
This corresponded to conversion of about 13 mol j> of the calcium in the stone.
If conversion were to stay constant in the regenerator and the H2S input were
reduced, it might be possible to circulate acceptor for a large number of cycles
at 13$ conversion per pass between ca. 93 and 8O$ CaS in the solid while main-
taining close to equilibrium H2S removal in the sulfur reactor.
During Run A7, small amounts of sulfur in the regenerator condensate caused
drainage problems. In later runs, about 5$ of hydrogen was added to the
regenerator feed to inhibit dissociation of H2S.
Runs A15 and A16
Two runs were made with Canaan dolomite at 1100°F in the regenerator instead of
130O°F as used in Run A7. The run conditions are given in Tables 44 and 45, and
results are plotted in Figures 41 through 44.
Run A15 was carried out at the same conditions as Run A7, except for the lower
regenerator temperature. In A15 breakthrough occurred in the gas desulfurizer
at about 6 cycles as compared with after about 10 cycles in A7. Gas analyses
had not been taken prior to this, but analyses of the solids indicate that good
pickup was being achieved.
- 149 -
-------
Hit
2.0
1.8
1.6
1.4
1.2
1.0
.8
.6
to-
rn*
I "•
5 s
5 a
i
Oi
Ho
x
o
,07
,06
1*1 .05
,04
.03
,02
.01
6 8
- 150 -
10
12
14
-------
FIGURE!. 40:
• \'
•HttT
100
"i
iBtofa
CafrlQQNT ENTrrOF TftCCEP FOR
m&
90
^EaS^tBes
Ltorzf
S5
80
70
•HeU
60
50
rn
40
30
-r/-
7
20
10
10
~t —- —_j
12
14
- 151 -
-------
Table 44
Conditions and Results for Gas Desulfuriaer with Canaan Dolomite Feed.
cn
to
System Pressure:
System Temperature*
Run Number
Acceptor Size Consist, Tyler Mesh
Feed Rate, Ib/hr (half -calcined basis)
Solids Residence Time, minutes
Incut r SCFH
Recycle to Bed
HaS
CO,
Ha
H20
Na
Purges (COa) to Bed
Purees (Na) above Bed
Recycle, Acceptor Lift Gas, above Bed
Output. SCFH in Cycle
Exit Gas Rate, SCFH (dry basis)
Composition. Mol t
H,
CO
CO,
N2
H,S
Outlet Gasr TOD of Bed
Comnosition. Mol %
11 aO
Ha
CO
CO.,
Na
H2S
Flow Rate, SCFH, Top of Bed
Fluldizlng Velocity, ft/sec •
Attrition, ^ of Feed Rate
Duration of Circulation with HaS Feed, brs
Removal of Feed Sulfur, $
Removal of Feed + Recycle Sulfur, %
$ H,S in Outlet /Equilibrium % H,S
Conversion of Acceptor /Pass, mol %
15 atm (2O6 psig)
1600° F
A7
28 x 35
6.5
32
175
3.5
• 54
73
-
96
5
15
71
1-2
215
17 \
18 /
12 S(»)
53
.osj
9.8
16
17
11
46
.046
416
.60
O.7
25.5
97
95
1.4
19
A15
28 x 35
6.7
32
178
3.5
54
73
-
96
5
15
71
6-7
213
17.5
17.9
9.93
54.4
.207
1O.O
16.4
16.8
9.28
47.3
.194
418
.60
.77
35.1
87
79
6.8
16
A16
28 x 35
3.3
65
178
1.8
54
73
-
96
5
15
71
1-3
215
18.2
18. O
S.74
54. O
. O48
9.47
17.1
16.9
9.18
47.3
.045
418
.60
.89
42.7
94
9O
1.7
18
(i) Estimate.
-------
Table 45
Conditions and Results for Regenerator
with Canaan Dolomite Feed
System Pressure; 15 atm. (206 psig)
Run Number
A7
A15
A16
Solids Residence Time, minutes
Temperature , °F
Input. SCFH
H20
C02
H2
Purges (N2) to Bed
Purges (N2) above Bed
Purges (CO2) above Bed
Output in Cycle
Exit Gas Rate, SCFH (dry basis)
Composition. Mol $
C02
N2
H2S
H2
CO
COS
Outlet Gas, TOD of Bed
Composition Mol %
H20
C0
H2S
H2
CO
COS
37
1300
110
110
0
8
10
5
1-2
133
85.7
13.5
0.7
ND
ND
ND
48.3
47.8
3.5
0.4
Flow Rate, SCFH, Top of Bed
Fluidizing Velocity, ft/sec
Regeneration of Acceptor, Mol
228
.63
5.1
36
1100
100
118
12
8
10
5
6-7
154
78.4
11. 4
1.41
8.0
.76
.02
41.6
48.5
3.2
.91
5.15
.49
.01
238
.60
11.5
74
110O
109
118
12
8
10
5
1-3
152
77.4
12.8
.456
8.3
1.01
Trace
44.1
46.0
3.9
.282
5.14
.63
Trace
245
.61
7.5
ND - Not Determined.
- 153 -
-------
n<
CM«
(Oi
X tt
u £
_
Ol
1.6
1.4
1.2
i.o
o.g
0.6
I
±1:
- 154 -
-------
10 X 10 TO Vi INCH 46 1323
7 X 10 INCHES HADE IK U.S.*.
KEUFPEL A ESSER CO.
Eiiii
1OO
80
tn
en
60
40
2O
10
12
14
16
18
-------
n
N
n
v
to
o
u
I g
u S
5 a
- u
0 i u
hg t
0=3
"0 g
x - "
x
100
10
- 156 -
-------
0 =
"• £
IDS
u
Z a
o S
5 IS
SIS
-o S
- *
!i
I)
!i
- 157 -
-------
The progress of the run was defined by the CaS content of solids, which is
given in Figure 42. Where the desulfurizer curve reaches its plateau, the
reduction in CaS content in going through the regenerator is 8 mol % absolute.
For Run A7, it was 13$. Going from 8$ to 13$ conversion for a 200°F change in
temperature is equivalent to an activation energy of the order of 7-11 Kcal/gm
mol depending on the mechanism assumed. The system is thus relatively insensi-
tive to change in temperature as long as equilibrium is not limiting.
A very important result of Run A15 is that the observed decline in activity was
quite low as cycling was continued. From cycle 8 to cycle 17, the amount
regenerated dropped from 8 mol $ to 6 mol $. This is confirmed by the gas data
presented in Figure 41 which shows the exit H2S concentration from the regenerator
slowly dropping from 1.4$ to 1.2$.
More data are required for confirmation, but it appears that good activity could
be sustained for a large number of cycles under Run A15 conditions if the circu-
lation rate were increased such that only 6-8$ of Ca were sulfided per pass
through the sulfur reactor.
The same tentative conclusion applies to Run A7 conditions, but here the amount
of Ca sulfided per pass could be increased to 13$.
Run A16 was comparable to A15 with the feed rates of H2S and acceptor halved.
This maintained the same conversion of acceptor per cycle, and doubled the resi-
dence time of acceptor in both vessels. It was hoped that increased residence
time in the regenerator would increase the degree of regeneration per pass.
Figure 43 presents the solids data from Run A16 together with data from Run A15.
There was no effect of increasing the residence time. The data points from both
runs fall on the same curves.
The exit gas H2S content is given in Table 39 and in Figure 44. Since only half
the usual H2S was fed, the H2S content of the regenerator offgas was correspond-
ingly lower.
At the end of Run A16, acceptor circulation was halted and the vessels were run
as batch beds. Samples were withdrawn from the regenerator and analyzed. At
the end of Run A16, the initial bed feed solids contained 3.6 mol $ CaC03 and
the final bed product 11.7 mol $ of CaC03 after a residence time of 75 minutes.
An additional holding time of ten hours increased the CaC03 content to only
17.5$.
The conclusion to be drawn from the comparison of Run A15 with A16, and from the
experiment at the end of A16 is that most of the CaS is in a form which is un-
reactive, and increasing the residence time will provide only a marginal
improvement in conversion in the regenerator. It seems that the conversion of
CaS per cycle is limited to about 6-8$ per pass at 11OO°F regardless of resi-
dence t ime.
Density of Sulfided Product
As dolomite is sulfided, its density tends to decrease as CaC03 is converted to
CaS. Table 46 lists the density of several sulfided samples. Although density
decreased with degree of sulfidation, the change was not as great as would have
- 158 -
-------
Table 46
Density of Sulfided Canaan Dolomite
Cycle Mol % Densityt1)
Number CaS Lb/ft3
A7 .8 9.6 142.6
5.7 48.O 139.O
12.3 85.0 131.6
A15 0 0 176.o(2)
.4 35.8 134.9
8.3 88.6 130.5
12.1 88.6 130.8
17.1 89.9 130.4
A16 10.5 88.3 126.3
(i) Measured in mercury at 1 atmosphere.
(2) Raw uncalcined feed.
- 159 -
-------
been predicted from molecular weight calculations. The reasons for this are not
clear at this time. In Run A15, there was no change in density once the sample
had been thoroughly sulfided.
E. Runs with Pennsylvania and Virginia Dolomites
Three runs were made to examine the attrition resistance of two other dolomites.
Results are summarized in Tables 47 and 48.
The feed for Run All was supplied by G. & W. H. Corson, Inc., Plymouth Meeting,
Pennsylvania. This stone gave an attrition rate of 9$ of the feed rate. The run
was shut down after three cycles due to depletion of the feed. Based on the
analysis of solids and a single gas analysis, it was concluded that pick up of
H2S and subsequent regeneration were about the same as with other dolomites.
The James River Limestone Company supplied the Buchanan, Virginia dolomite which
was the feed for Run A12. The run was shut down due to depletion of the feed as
a result of a relatively high attrition rate of 4$.
The activity of the Buchanan dolomite was good and its performance on regenera-
tion was at least equivalent to that of the Canaan dolomite as noted by the
relatively high recovery of about 70% of the H2S fed to the desulfurizer in
regeneration of the acceptor (cf. Table 39).
Another possible criterion to judge the maintenance of activity is the diver-
gence between curves representing the CaS content of acceptor leaving the sulfur
reactor and regenerator, respectively, versus the number of cycles. As long as
this divergence is maintained then regeneration is being achieved. The Buchanan
dolomite also performed well based on this criterion as the curves in Figure 45
show.
The principal disadvantage of the Buchanan dolomite then is its relatively high
attrition rate.
A sample of regenerator gas from Run.A12 was found to contain COS, a component
for which we did not usually check. Equilibrium considerations call for COS to
be present at a level of about 3$ of the H2S.
For Run A19, the gas desulfurizer was run.at 1650°F with Buchanan dolomite feed.
The purpose was to see whether sufficient sintering would occur at the higher
temperature to reduce the attrition rate. The run was similar to Run A12,
except for the higher temperature and the need to increase the C02 content of the
gas to prevent calcining at the hot walls. The attrition rate was 5.0$, compared
to 4.1$ in Run A12. It was concluded that hardening due to sintering does not
occur in going from 1600 to 1650°F in the gas desulfurizer.
F. Runs with Limestone Feeds
Two varieties of limestone were tested to see whether the attrition problem
experienced with dolomites would occur. Results are summarized in Tables 49 and
50 . Ruth, et al.(15) have found limestone and calcite to be inactive in the
reaction of CaC03 with H2S. On the other hand, Esso EnglandC16 ) used 1691
limestone from Seneca Falls, New York in the reaction of CaO with H2S and
obtained good results.
- 160 -
-------
Conditions and Results for Gas Desulfurizer with
Pennsylvania and Virginia Dolomite Feeds
System Pres
Run Number
Acceptor and Size Consist, Tyler Mesh
Temperature , °F
Feed Rate, Ib/hr (half-calcined basis)
Solids Residence Time, minutes
Inmit, SCFH
' Recycle to Bed
HSS
C02
H2
H20
N2
Purges (C02) to Bed
Purges (Na) above Bed
Recycle, Acceptor Lilt Gas, above Bed
Cutout . SCFH in Cycle
Exit Gas Rate, SCFH (dry basis)
Composition. Mol 4
Hz
CO
C0a
N2
H2S
Outlet Gas, TOD of Bed
Comuos^tion, Mol %
11 aO
Ha
CO
CO,
Na
HjS
Flow Rate, SCFH, Top of Bed
Fluidlzing Velocity, ft/sec
Attrition, i> of Feed Rate
Duration of Circulation with H9S Feed, hr
Removal of Feed Sulfur, £
Removal of Feed + Recycle Sulfur, £
£ H2S In Outlet/Equilibrium % H,S
Conversion of Acceptor/Pass, mol jt
sure; IS atm (206
All
28 x 35
Plymouth
Meeting, Pa.
Dolomite
1600
7.0
30
178
3.5
54
73
-
96
5
IS
71
1-2
215
18")
18 r (0
10 I ^ '
54 J
~
10
17
17
9
47
~
418
.60
e.o
5.2
Good
Good
-
ca. 20
psig)
A12
20 x 35
Buchanan, Va.
Dolomite
16OO
7.4
29
178
3.5
54
73
-
96
5
15
71
2-3
214
20.7
17.7
9.2
52.4
.11
9.8
19.4
16.6
8.66
45.6
.105
418
.60
4.1
1S.1
93
88
4.1
16
A19
28 x 35
Buchanan, Va.
Dolomite
1650
7.2
30
6.6
13
20
IS
46
.0635
408
.60
S.O
14.8
96
93
2.6
18
(i) Estimated value. Gas chromatograph recorder not functioning.
-------
Table 48
Conditions and Results for Regenerator with
Pennsylvania and Virginia Dolomite Feeds
System Pressure:
System Temperature;
15 atm. (206 psig)
1300° F
Run Number
Solids Residence Time, minutes
Input, SCFH
H20
CO 2
H2
Purges (N2) to Bed
Purges (N2) above Bed
Purges (C02) above Bed
Output in Cycle
Exit Gas Rate, SCFH (dry basis)
Composition, Mol Jo
CO 2
N2
H2S
H2
CO
COS
105
105
11
8
10
5
1-2
135
771 ( 1
13J ^ * '
1.41
61
2 (0
OTV
97
105
11
8
10
5
2-3
133
72.5
16.5
1.68
6.4
3.0
.03
102
102
11
8
10
5
4-5
130
76'
14^
1.32^
7.3"
1.1
ND,
(0
Outlet Gas, Top of Bed
Composition. Mol %
H20
C02
N2
H2S
H2
CO
COS
Flow Rate, SCFH, Top of Bed
Fluidizing Velocity, ft/sec
Regeneration of Acceptor, mol
46
45
3
.87
4
1
ND
220
.62
10
45.6
42.1
5.5
1.03
3.9
1.8
.02
217
.61
11
45.7
44
4
1.32
4.5
.7
ND
210
.59
9
(i) Estimate. Gas chromatograph recorder not functioning.
ND - Not Determined.
- 162 -
-------
n =
*• z
o s
a S
Z 01
- u
z««
3 I u
- 163 -
-------
-Table 49
Conditions and Results for Gas Desulfurlzer with Limestone Feeds
System Pressure: 15 atm (2O6 psig)
System Temperature; 16OO°F
I
M
CT)
I
Run Number
Acceptor and Size Consist, Tyler Mesh
Feed Rate, Ib/hr .(half-calcined basis)
Solids Residence Time, minutes
Input. SCFH
Recycle to Bed
H2S
H20
N2
Purges (CO2) to Bed
Purges (N3) above Bed
Recycle, Acceptor Lift Gas, above Bed
Output. SCFH in Cycle
Exit Gas Rate, SCFH (dry basis)
Composition. Mol *t
Hz
CO
CO,
Na
HaS
Outlet Gas. Top of Bed
Composition,. Mol %
HaO
Ha
CO
CO,
Na
HaS
Flow Rate, SCFH, Top of Bed
Fluldizing Velocity, ft/sec
Attrition, $ of Feed Rate
Duration of Circulation with HaS Feed, hr
Removal of Feed Sulfur, %
Removal of Feed + Recycle Sulfur, $
% H2S in Outlet/Equilibrium % HaS
Conversion of Acceptor /Pass, mol %
A5
A13
28
x 48 Nebraska
Limestone
4.6
'6O
(1)
(O
< 1.1
28 x 35
BCR-1691 Limestone
6.7
34
2OO
3.75
59
8O
1O6
5
15
71
O-l
223
15
21
9.7
54
.11
11
14
19
8.9
47
.10
46O
.66
O,6
19.3
94
83
3.2
19
(i) Estimated value.
-------
Table 50
Conditions and Results for Regenerator
with Limestone Feeds
System Pressure; 15 atm (206 psig)
System Temperature; 1300°F
Run Number A5 A13
Solids Residence Time, minutes 68 39
Input. SCFH
H20 126 109
CO2 118 112
Air 7.7
H2 12
Purges (N2) to Bed 10 8
Purges (N2) above Bed 10 10
Purges (C02) above Bed 5 5
Output. SCFH in Cycle 1-3 5-6
Exit Gas Rate, SCFH (dry basis) 146 146
Composition. Mol *&
C02 84.1 77
N2 15.8 13
H2S 0.06 1.44
H2 6
CO 2
Outlet GaSj. Top of Bed
Composition,. Mol ^
H20 52.7 46
C02 42.6 45
N2 4.7 3
H2S .03 .88
H2 4
CO - 1
Flow Rate, SCFH, Top of Bed 277 238
Fluidizing Velocity, ft/sec .78 .67
Regeneration of Acceptor, mol % 0.5 11.6
- 165 -
-------
Run A5
The feed to Run A5 was Nebraska limestone. This limestone showed very little
activity toward H2Sj only about 1% conversion to CaS occurred. However, the
rate of attrition was 0.4$.
Runs A10 and A13
Run A10 was to be made with 1691 limestone to determine whether our results
showing Nebraska limestone to be a poor acceptor were peculiar to that stone.
A preliminary result was that the attrition rate was 0.9$ of the feed rate, but
the run was plagued with numerous operating difficulties and could not be
brought to a successful conclusion.
Run A13 was made with 1691 limestone as a repeat of aborted Run A10. The
attrition rate was acceptably low at 0.6$ of the feed rate and H2S removal was
good. The regeneration of the stone did not appear to proceed as readily as
with Canaan dolomite, i.e., compare Figures 46 and 40. With the Canaan dolo-
mite (Figure 40), it took 13 cycles to sulfide the acceptor to 99$ CaS, and
upon regeneration, 13$ of the CaS was removed. The corresponding figures for
the 1691 stone (Figure 46) were 10 cycles and only 6$ regeneration of the CaS.
Run termination was caused by inability to remove acceptor from the gas desulfu-
rizer. Upon disassembly, the contents of the gas desulfurizer were found to be
agglomerated in a single mass. In addition, difficulty in transferring acceptor
between the vessels and the presence of +28 mesh agglomerates had been experi-
enced during the entire run.
The agglomerates from the sulfur reactor were raspberries with a hard core of
yellow material. The agglomerates were 50$ soluble in cold, 0.5 N HC1.
The sample of 1691 limestone which we received had an unusually high level
of impurities. The stone contained only 64 wt. $ CaC03 and showed a 9.5$ weight
loss when heated at half-calcining conditions (1550°F, 1 atm. C02).
The good activity toward H2S was probably due to the porosity resulting from
the weight loss which occurred at half-calcining conditions. The 1691 .behaved
like a cross between a dolomite and a pure limestone. The agglomerates were
probably caused by a low-melting liquid formed from the impurities, possibly in
conjunction with CaS.
The 1691 was a rather unusual stone. It is doubtful that the results which
were obtained would apply to ordinary limestones or dolomites.
G. Once-Through Operation
Runs A8 and A9
Two runs were made using once-through operation to provide a basis for cost
comparisons with process variations using regeneration. Run conditions are
listed in Table 51. Dolomite was fed in the raw condition; calcination of MgC03
was assumed to occur as soon as the solids hit the hot bed. The C02 of calcina-
tion is therefore not included in the calculated gas composition just above the
bed. It is also excluded from the equilibrium calculation. Samples were taken
after some four or more bed turnovers had occurred.
- 166 -
-------
S g
1OO
90
80
n<
CM.
(O
70
H
•m
o 3
-o g
X- *
ox
*• r*
60
£
50
40
30
6 8
- 167 -
1O
12
-------
Table 51
Conditions and Results for Gag Desulfurlaer - Once-Through Operation
System Pressure:
System Temperature;
15 atn (206 palg)
1600* F
Run Number
Acceptor and Size Consist, Tyler Mesh
Feed Rate, Ib/hr, Raw Stone
Solids Residence Time, minutes
A8
Input. SCFH
Recycle to Bed
H2S
C02
H2
HaO
N,
Purges (CO2) to Bed
Purges (N2) above Bed
Recycle, Acceptor Lift Gas, above Bed
Output , SCFH
Exit Gas Rate, SCFH (dry basis)
Composition. Mol t
28 x 48
•— Tymochtee Dolomite —
2.01 1.68
95 116
160
2.8
46
64
82
5
15
71
160
2.8
46
64
82
5
15
71
189
01
00
CO
C02
N2
II2S
(0
(O
Outlet Gasf Ton of Bed
Compos ttion. Mol 5t
H,0
CO
CO,
112S
.Flow Rate, SCFH, at Top of Bed
Fluidizing Velocity, ft/sec
Attrition, % of Feed Rate
Duration of Circulation with H,S Feed, hrs
Removal of Feed Sulfur
Removal of Feed + Recycle Sulfur, $
% H2S In Outlet/Equilibrium % HaS
Conversion of Acceptor/Pass, mol $
9.4
16.7
16
45.4
.057
362
.52
1.3
9.6
96
93
9.7
16.1
16.7
45.7
.153
365
.53
.8
10.1
89
81
(i) Estimate.
!i) CO, of MgCO, calcination not Included.
• ) Based on gas analysis. Assays on solids show
and 95^ for A8 and A9 respectively.
-------
Based on assays of the solids, the conversion was 85$ in Run A8 and 95$ in Run
A9. However, H2S removal was only 89$ in Run A9 as compared to 96$ in Run A8.
The data show that once-through operation in the range of 85-90$ solids con-
version is feasible. The attrition rate using the raw stone feed was about 1$
of the feed rate. The high conversion obtained was consistent with the results
of Run A7.
H. "Refractory" CaS - Runs A16 and A17
Part of our solids assay procedure involved air oxidation of CaS by burning in
an air-fluidized bed which is slowly heated from 1300-1600°F, and then held at
temperature for 30 minutes. As originally set up, all of the CaS would be
reacted to CaS04 or CaO. It was found that generally a portion of the CaS in
reactor samples was resistant to oxidation, and this material was called
refractory CaS. Although the reactions and reaction conditions in the regenera-
tor and assay differ, the possibility that formation of refractory CaS would
correlate with inactivity of CaS in the regeneration reaction was considered.
At the end of Run A16, the level of refractory CaS was studied as a function of
time at regenerator conditions. Run A17 was designed to examine the change in
refractory CaS with time in the gas desulfurizer by sulfiding a batch of acceptor
at simulated process conditions, turning off the H2S feed, and then sampling from
the vessel several times.
The results are given in Table 52. The data show that refractory CaS is con-
verted slowly to ordinary CaS under regenerator conditions. There was no change
in the refractory CaS under gas desulfurizer conditions.
It was found that regenerator performance does not correlate with refractory CaS
as determined from our assay. At the end of Run A7, the refractory CaS from the
regenerator was 48$ and the CaS conversion per pass was 13$. In Run A16, the
refractory CaS was 26$ and the conversion of CaS was only 8$.
It seems that refractory CaS does not form from ordinary CaS, but rather is made
when the CaS is first formed. One would speculate that inactive CaS differs
morphologically from the active material giving rise either to diffusion limita-
tions on a microcrystalline level or to a different adsorptive surface for
reacting species. Build up of inactive CaS also may be responsible for poor
kinetic results reported when batch cycle regeneration of CaS is carried out.(17)
A number of samples were sent to our analytical laboratory for sulfur analysis
using a variation of the Eschka method as outlined in ASTM Method D-271. The
results were in error on the low side. Oxidation of CaS to CaS04 is involved
and not all of the sulfur was picked up.
J. Accuracy of H2S Feed Rate - Sulfur Balance
We have not been able to properly close sulfur balances in this experimental
system. In Run A9, for example, the sulfur balance based on the feed rate of
solids and the H2S content of inlet and outlet gas shows 80$ molar conversion of
CaC03 to CaS. The assay of the product shows 95$ conversion of CaC03 to CaS.
For cyclic operation, calculations become more uncertain because of the unsteady
state nature of the inlet and outlet CaS concentrations, accumulation in the
reactor and the exit gas composition. However, such balances also close ± 2O$.
- 169 -
-------
Our most accurate numbers .are the measured exit gas concentrations and the
solids assays. The most inaccurate number is the feed rate of liquid H2S.
Although it is not nearly as sensitive to temperature as is gas density,
temperature variations can cause a 10$ change in liquid density. Finally, it
is known that H2S can react with the reaction vessel.
The accuracy of solids assays was improved as our understanding of the behavior
of inactive species improved. A revised H2S feed system is under consideration,
but could not be installed during this contract period.
K. Corrosion
At the start of the sulfur cycle studies, the gas desulfurizer vessel and in-
ternals were made of type 310 stainless steel. At the end of Run Al, the gas
desulfurizer was found to have accumulated a blob of molten, yellow, metallic
sulfides at the gas inlet line. Part of the inlet line was severely corroded
and embrittled. A new coil and gas inlet line for the sulfur reactor were fabri-
cated from type 446 stainless steel which is more resistant to a reducing atmos-
phere than type 310. The remainder of the internals were replaced with type 446
stainless steel during a shut down of Run A10. At that time, the reactor body
showed scaling, but was replaced with type 310 since the type 446 stainless for
the body was not on hand.
No corrosion problems were encountered in the regenerator which had both a body
and internals made of 31O stainless steel.
L. Conclusions and Summary of Major Results
Attrition
The attrition rates for all stones tested are presented in Table 53. An attri-
tion rate of 1$ or less is desired, and only the Canaan dolomite was satisfactory,
Limestones had good attrition rates, but the Nebraska stone was not active and
the 1691 limestone agglomerated in the gas desulfurizer. ,
It is probable that many dolomites can be found which will have the requisite
attrition resistance. There is some evidence that limestones are generically
inactive toward H2S.(15)
By using a strong, deactivated acceptor from previous OCR experiments, it was
shown that the sulfur cycle under consideration does not intrinsically cause
softening and attrition to occur.
Activity in the Gas Desulfurizer
All the natural dolomites showed good reactivity toward H2S. Using only an 18-
inch deep bed with gas residence times of 2 to 4 seconds, the exit H2S con-
centration was only 2 to 3 times the calculated equilibrium value. Typically
this activity was maintained with CaS loadings of 80$ in the solid. Better
results should be obtained in deeper, commercial sized units.
Tests with once-through operation showed that CaS loadings of 85-90 mol $ are
feasible.
- 170 -
-------
Table 52
A16
A17
A7
A15
Change in ' Refractory CaS When
Held at Reaction Conditions
Material and Conditions
Regenerator bed.
10.5 cycles,
held at regenerator
conditions .
Newly sulfided bed.
Held at gas desulfurizer
conditions .
Regenerator bed,
12.9 cycles, held at
regenerator conditions.
Holding
Time,
Hours
0
.5
1
2
10
0
.5
1.0
4.0
0
2.0
CaC03
11.7
13.1
14.0
14.6
17.5
56.0
58.7
53.3
58.3
14.7
24.1
Mol $
Total
CaS
88.3
86.9
86.0
85.6
82.5
44.0
41.3
39.4
41.7
85.3
75.9
Refractory
CaS
25.5
23.3
21.0
18.6
14.6
6.3
5.1
7.4
5.3
47.7
35.7
Regenerator Bed, 17.1 cycles
10.1
89.9
39.4
Type of Stone
Dolomite
Deactivated )
Dolomite )
Limestone
Table 53
Attrition Rate of Acceptors
Origin
Plymouth Meeting, Pa.
Tymochtee formation
Buchanan, Va.
Canaan, Conn.
Cycling runs for OCR
Nebraska
BCR-1691
Attrition Rate,
Wt # of Feed Rate
9
5
5
0.8
0.2
0.4
0.2
- 171 -
-------
Regeneration, CaS -» CaC03
The regeneration step has not gone to completion under the conditions employed.
The CaS content of the acceptor builds up until it is 95 mol $ or more. The
CaS content then cycles between this upper limit and the CaS content of the
regenerated stone. The best conversion observed was 13 absolute mol $ in Run
A7. The longest run was Run A15 which ran for 17 cycles. From cycle 8 to 17,
the regeneration activity declined from 8 mol $ per pass to 6 mol $.
It appears that as the acceptor is cycled, the CaS changes into an inactive
form. The reaction times in the regenerator produce only marginal additional
conversion to CaC03. Changing the temperature from 1300 to 1100°F did not
improve the regeneration. Varying the steam/C02 ratio also yielded no improve-
ment .
It is concluded that in a commercial process, the CaS would cycle between two
high levels such as 90 and 75$. Sulfidation and regeneration would be 10 to 15$
per cycle and a large number of cycles will be needed.
Toward the above end, one run has given 13$ conversion in the regenerator per
cycle, and another has demonstrated 17 cycles without serious loss of activity
over the range of 8 to 17 cycles.
M. Future Work
Demonstration of Activity for a Large Number of Cycles
It has been concluded that the acceptor could be cycled at 13$ conversion per
pass without losing its ability to remove H2S in the gas desulfurizer. A
demonstration of this conclusion is needed. It has not been actually shown
that high removal efficiency can be obtained with a cycled stone which is cycled
from about 77$ to 90$ conversion. In our runs so far, H2S breakthrough in the
gas desulfurizer occurred because of 20$ per pass of H2S being fed. Also, at
the 13$ conversion per pass, only about 12 cycles were actually demonstrated.
Squires Reaction - Approach to Equilibrium
The work done to date simulated only the bottom of the regenerator since, the
maximum H2S concentration was about 1$ (wet basis) compared to about 4.5$ at
equilibrium. The process design assumes about 3.6$ H2S or 87$ approach, -and
this has yet to be demonstrated.
The kinetics of the reaction remain to be further investigated. Specifically,
the effect of the H20/C02 ratio and the effect of temperature as they affect
rate of reaction, approach to equilibrium and the regenerability of the stone
require more work.
Attrition
Of the four dolomite acceptors studied, three had excessive rates of attrition.
It has to be established that satisfactory acceptors can be obtained in areas
close to projected gasification plant sites. Alternately, conditions or treat-
ments must be determined to render other acceptors more resistant to attrition.
- 172 -
-------
Effect of Particle Diameter
The studies to date were restricted to 28 x 48 Tyler mesh particles. It remains
to be determined how the larger particles, which would be used in the scaled up
process, would behave.
Blocked Out Operation
In order to unequivically define the relationship of CaS content to number of
cycles, it will be necessary to operate the gas desulfurizer-regenerator combina-
tion so that all of the acceptor is run through before the next cycle is started.
This will necessitate running with a large starting batch of feed and discarding
the start-up and shut-down beds as the run proceeds.
- 173 -
-------
X. CARBON BURN-UP CELL AND GASIFIER DEMONSTRATION RUNS
A. Carbon Burn-Up Cell Study
1. Introduction
The CBC concept provides for preheat of the steam-air mixture for the gasifier
by combustion of the gasifier residue char.
Combustion of char at high oxygen partial pressures poses the threat of localized
hot spots and consequent ash slagging. The incoming process steam and air to the
CBC are relatively cold, and thus provide a heat sink which should prevent ash
slagging.
The experimental work presented here was designed to demonstrate the limiting
operating conditions in the CBC under which efficient utilization of the feed
char and lack of ash slagging are simultaneously obtained.
Combustion was carried out in inert beds of dead burned dolomite or silica sand
in the 3-inch I.D. reactor used as the gas desulfurizer in the sulfur cycle study.
The bed height was 18 inches. A flow diagram of the system is shown in Figure 47.
Run conditions were chosen within the framework established by the process heat
and material balance calculations.
Eight runs were made using pregasified Pittsburgh Seam coal char as fuel at
inlet oxygen pressures of 2.2 atmospheres. Conditions and results are shown in
Table 54. Table 31 gives the descriptions of the runs not specifically mentioned
in Table 54. Compositions of the char feedstocks are shown in Table 55.
These feedstocks were partially gasified Pittsburgh Seam coal char, used to
simulate the actual gasifier product which is the feed to the carbon burn-up
cell.
2. Runs CBC-1 through CBC-4
These runs were made with dead burned dolomite as the inert bed material. They
explored the effect of bed expansion on bed temperature gradient and on ash
slagging.
Run CBC-2, over a period of about 20 hours at 1880°F and 55$ bed expansion, gave
no evidence of ash slagging but did show a moderately high temperature gradient.
The inlet gas flow rate then was decreased to give a bed expansion of 41$.
After about 20 hours of additional operation at this latter condition, the
temperature at the bottom of the bed increased suddenly to 1950°F and the run
was ended. Inspection of the reactor showed that slag, amounting to about 1$ of
the total ash fed, had formed at the bottom.
Run CBC-3, made at the slightly lower temperature of 1850°F showed that, at least
for 13 hours, no slagging occurs at the low bed expansion of 30$. However, a
large temperature gradient existed.
- 175 -
-------
-J
o
DRAWN
5-IC,
Tj)
CONSOLIDATION COAL CO.
RESEARCH A DEVELOPMENT DIVISION
LIBRARY. PENNA.
•CALI
£51.00
i m. LUCH eo, r
DMAWIM* MO.
CF-.MiG
-------
Conditions and Results for Carbon Burn-un Cell Runs
-a
I
Run
Bed Material
Size, Tyler llesh
Ter.periture
Fuel Char
Input. SCFH
Steam
Air
Lift Gas
Fluidizing Gas
Purges (N2)
To Bed
Above Bed
Fuel Char Feed Rate, Ib/hr
Output. Top of Bed. Mol ^
Steam
02
Na
C02
Flow Rate, SCFH at Top of Bed
Ash, Ib/hr
Exit Cas Composition. Mol %. dry basis(')
C02
Fluidizing Velocity, Top ol Bed, ft/sec
Og Partial Pressure, atm
Inlet
Outlet, Top ol Bed
Bed Temperature Gradient, *f\*i
Run Duration, hours
% Bed Expanslon(')
CBC-2
Vessel
System
Dead Burr.ed Dolomite
35 x 48
1880
K2A-K7A Product
132
7O
279
5
20
1.14
27.3
9.O
57.7
6.O
487
.217
11.8
80.4
7.8
.79
2.26
1.36
57
21
55
110
70
220
5
20
.997
27.3
8.9
57.6
6.2
406
.190
11.4
8O.7
7.9
.66
2.26
1.33
117
20
41
I.D. 3" - Bed Height 18'
Pressure: 15 atm. (206
CBC-3
Dead Burned Dolomite
28 x 35
1850
K2A-K7A Product
158 '
70
350
5
20
1.46
27.2
8.7
57.8
6.3
584
.278
11.5
8O.2
8.3
.94
2.27
1.31
124
13
30
psig)
CBC-4
Dead Burned Dolomite
35 x 48
1850
K2A-K7A Product
167
70
350
5
2O
1.39
28.3
8.9
56.9
5.9
593
.265
11.9
8O. 2
7.9
.95
2.23
1.34
20
31
75
Silica Sand
28 x 35
17OO 175O 18OO
D12 Product
.89 .91 .93
17 5 7
48 52 52
185O
.94
38
54
CBC-8
Silica Sand
28 x 65
1850-1900
D12 Product
169
7O
350
5
2O
1.41
28.6
8.6
56.6
6.2
595
.226
11.5
8O. 2
8.3
.98
.23
.29
22
25
67
(i) Includes all Na purges.
(a) Difference between bottom of bed and top of be4.
(a) Percent bed expansion above Incipiently fluidized bed height, 1OO$ expansion ia twice the Incipient height.
-------
Table 55
Compositions of Char Feedstocks
Used in Run CBC-1 -» CBC-4 CBC-5 -» CBC-8
Produced in Run K2A -» K7A D12
Hydrogen, Wt % (dry basis) .50 .46
Carbon 78.84 81.11
Nitrogen .54 .75
Oxygen (by difference) .37 -.18
Sulfur .47 1.70
Ash 14.28 16.16
Sulfide Sulfur .85
- 178 -
-------
In Run CBC-4, the bed expansion was increased to 75$ at 1850°F by substitution
of 35 x 48 mesh dead burned dolomite for the 28 x 35 mesh stone used in CBC-3.
The temperature gradient decreased to only 70°F. No slag had formed after
31 hours.
The bed of the dead burned dolomite proved to be reactive toward S02 formed
during combustion of the fuel char. Nearly quantitative pickup occurred with
the result that the particle size and density of the bed continually changed
during the runs. To prevent changes in fluidizing characteristics of the bed,
silica sand was used as bed material in place of dolomite in Runs CBC-5 through
CBC-8.
3. Runs CBC-5 through CBC-8
The behavior of silica sand as inert bed material was checked in Run CBC-5, and
combustion characteristics at the low end of the envisioned temperature range
were determined. Operation was smooth at all conditions, with no ash slagging
and low or moderate temperature gradients. Inspection of the product ash
showed that the char carbon was completely consumed, even at 1700°F.
In Run CBC-8 the bed temperature was increased to 19CO°F at a bed expansion of
67$. After 25 hours, no slag had formed and the temperature gradient was
moderate throughout. Typical temperature profiles are shown in Figure 27.,
4. Ash Properties
The product ash from all the runs had no detectable carbon content. Some
properties of the Loveridge char ash are shown in Table 56. Also shown are the
size consist and ash particle density for Run CBC-5. The product ash from each
of the other runs was nearly identical. The fuel char feedstock to the carbon
burn-up cell for all runs was substantially 100$ +100 mesh. As Table 56 shows,
the product ash is nearly all -100 mesh.
5. Conclusions
The experimental work shows that the carbon burn-up cell will be tr.ouble-free.
Complete carbon consumption occurs at temperatures as low as 17CO°F within a bed
height of 20 inches. Freedom from ash slagging and a moderate temperature
gradient has been demonstrated at 1900°F. What is required is sufficient mixing
of the inert bed particles. Our data show that 60-70$ bed expansion is suffi-
cient to provide the necessary mixing.
B. Gasifier Demonstration Runs
1. Introduction
At the bottom of the gasifier where the preheated air and steam from the carbon
burn-up cell first contact the incoming pretreated coal feedstock, the possi-
bility exists that ash slagging will occur or that ash-derived deposits will
form.
For the experimental program, the char feedstock was pretreated at 1200°F to
avoid caking problems. Thus, it simulated the product from the Seeded Coal
Process.
- 179 -
-------
Table 56
Ash Properties
Ash Composition. Sulfur-Free Basis
A1203, Wt
Si02
Fe203
CaO
MgO
Na20
K20
Ti02
ASTM Ash Fusion, Oxidizing
Initial
Softening
Hemispheric
Fluid
2300
2380
2420
2600
24.3
44.1
17.6
8.7
1.8
.9
1.4
.8
.4
Product Ash. Run CBC-5
Size. Tyler Mesh. Wt
+65
65 x 100
1OO x 200
-20O
.0
4.7
20.1
75.2
Particle Density, Ib/ft3* 100 (100 x 200 Mesh Fraction)
* Measured in mercury at one atmosphere.
- 180 -
-------
The runs reported here were carried out in the vessel used for regeneration of
the sulfur-laden acceptor. The bed diameter was 4-inches and bed height 48
inches.
2. Results
Three runs were made at process conditions where the steam, C02, and 02 partial
pressures were the same as in the effluent from the carbon burn-up cell in order
to simulate conditions at the bottom of the process gasifier. The purpose of
the runs was to determine whether ash slagging or wall deposits occur. The
feedstock was char prepared from Loveridge Mine coal (Pittsburgh Seam) .
Run Dll was made at 1725°F and there was no evidence of ash slagging or of wall
deposits. Run conditions and results are shown in Table 57. Feed and product
char compositions are given in Table 58.
To help define the temperature limit for the onset of ash slagging, Run D12 was
attempted at 1775°F with the 1200°F pretreated feedstock. A large coke-like
deposit formed quickly. Although the formation of coke from a material which had
been exposed previously to 1200°F was considered unlikely, another batch of feed-
stock was prepared at 1400°F to eliminate any possibility of coke formation.
Run D13 was made at 1775°F with the 1400°F pretreated feedstock. After about
six hours of char feeding, the temperature at the bottom of the fluidized bed
increased rapidly to 2000°F, and the run was ended. Inspection of the vessel
showed a massive deposit similar in appearance to that of the previous run. A
small amount of slagged ash was present at the bottom of the vessel. Slagging
clearly was caused by continued encroachment of the deposit into the zone where
the incoming oxygen and char met. The normal mixing pattern of the fluidized bed
thereby was altered, allowing a hot spot to develop.
Conditions and results for Run D13 are shown in Tables 57 and 58. The material
balance period for the run was made before the hot spot developed.
The superficial appearance of the deposits from both 1775°F runs was misleading.
Neither deposit was coke, as shown by their carbon content of one weight percent.
Chemical analysis of the deposits showed that their composition was nearly
identical with that of the char ash shown in Table 56. Thus, the deposits were
built up by mild slagging of the ash.
3. Conclusions
Complete operability has been demonstrated at 1725°F. In contrast with the carbon
burn-up cell runs, in which the same char ash was present, sintering or incipient
slagging caused a buildup of ash at the bottom of the reactor at 1775°F. The
carbon burn-up cell runs were made, in general, at considerably higher tempera-
tures and at strongly oxidizing conditions.
Ash fusion temperatures can be altered drastically under reducing conditions
such as exist in the gasifier. Possibly, operation at 1775°F would be successful
if better mixing of the char were achieved though the use of higher bed expansions
and fluidizing velocities. Existing equipment limitations preclude higher fluidi-
zing velocities than were used in Runs D12 and D13.
- 181 -
-------
Table 57
Conditions and Results for Demonstration Runs
System Pressure: 15 atm (206 psig)
Feedstock: Pretreated Loveridge Char (Pgh. Seam)
Vessel ID - 4" - Bed Height 48"
Run Number Dll D13
Feedstock Pretreatment Temperature, °F 1200 1400
Run Temperature, °F 1725 1775
Input
Steam, SCFH 98 96
C02 23 21
Air 134 158
Lift Gas, N2 91 76
Purges (N2) to bed 10 1O
Purges (N2) above bed 15 15
Char Feed Rate, Ib/hr 6.06 7.53
Output
Exit Gas Rate, SCFH (dry basis) 377 406
Composition. Mol $
H2 13.94 14.47
CH4 .56 .64
CO 14.97 19.12
C02 11.55 10.06
H2S .24 .22
N2 58.98 55.49
Outlet Gas, top of bed
Composition, mol %
H20 14.38 10.93
H2 12.44 13.39
CH4 .50 .59
CO 13.36 17.69
C02 10.31 9.32
H2S .21 .20
N2 •"• 48.80 47.89
Flow Rate, SCFH at top of bed 422 439
Fluidizing Velocity, ft/sec 0.38 0.40
Product Char, Ib/hr 3.32 4.06
Duration of Char Feeding, hr 12.0 6.2
Bed Weight, Ib 7.95 7.93
Carbon Gasified, Ib/hr 2.45 3.O9
% Carbon Burnoff(*) 48.4 49.9
% Steam Conversion 38.0 50.0
(i) 1OO (Ib/hr net carbon in exit gas)
(Ib/hr carbon in with char)
- 182 -
-------
Table 58
Analyses of Char Feed and Products
for Demonstration Runs
Run Number
Hydrogen, Wt %, (dry basis)
Carbon
Nitrogen
Oxygen (by difference)
Sulfur
Ash
Dll
D13
Feed
1.59
8O. 58
1.32
2.27
2.09
12.15
Product
.48
77.45
.73
-.76
1.62
20.48
Feed
1.08
82. 3O
1.60
.33
2.13
12.56
Product
.41
8O. 37
.76
.46
1.62
16.38*
Run Dll Product
Size. Tyler Mesh
28 x 35
35 x 48
48 x 65
65 x 100
-100
Wt %
12.4
32.9
29.5
20.4
4.8
Particle Density, lb/ft3
75.5
70.9
71.1
70.1
* Ash content is low because of deposit which remained in the gasifier.
- 183 -
-------
XI. LIQUID-PHASE GLAUS PROCESS
A. Introduction
The composition of the gas leaving the regenerator as projected in the
feasibility study in the Annual Report!1) contained about 4% H2S, 48$ H20, and
48$ C02. Elemental sulfur is recovered by contacting this gas with sulfurous
acid at the saturation temperature, ca. 325°F. The reaction taking place is:
2 H2S + H2S03 = 3 S + 3 H20.
Elemental sulfur is recovered as a liquid.
Since nothing was known of the kinetics of the liquid-phase Claus reaction,
conditions were chosen to match the gas concentrations and vapor liquid ratio
given in the Annual Report. Flow rates were chosen to give the same vapor
retention time, and the feed rate per unit volume of reactor space was the same
as projected in the Annual Report. A total of eleven runs were made. The data
are summarized in four tables as given below:
Table 59 - Results and Conditions for Liquid-Phase Claus Reaction
Table 60 - Sulfur Balances for Liquid-Phase Claus Reaction
Table 61 - Yields of Elemental Sulfur
Table 62 - Analysis of Claus Product Liquor
B. Once-Through Runs
Runs Cl and C2
Run Cl was made at 320°F. The outlet S02 concentration was near zero for the
entire run; the H2S concentration was not determined. Sulfur could be seen in
the column, but none could be collected at the bottom. The condensate was
bright yellow, the color of Wackenroder's solution of sulfur together with
soluble sulfur byproducts. Analysis of the product solvent showed only 0.06$
sulfur. The run ended when the exit line plugged.
The run indicated that sulfur can be readily produced with little by-product
formation under process conditions. There were insufficient data for a sulfur
balance.
Run C2 was a repeat of Cl. The same observations with respect to sulfur collec-
tion and sulfur content of the condensate and product solvent were made.
Analysis of the exit gas gave 3.4$ H2S and 0.2$ S02. H2S and S02 which had not
reacted in the column were absorbed in the condensed steam to form Wackenroder's
solution. The ratio of H2S to S02 in the dry exit gas indicates that S02 was
preferentially absorbed by the cooled condensate. Sufficient data were on hand
to make a crude sulfur balance as shown in Table 60. The balance showed 40$
conversion of the feed to elemental sulfur. However, after 2-1/2 hours of
operation, only 25$ of one hour's feed was collected as elemental sulfur.
Liquid sulfur was visible in the column and a wash with hot toluene yielded some
additional sulfur.
- 185 -
-------
Table 59
Results and Conditions for Liquid-Phase Claus Reaction
I
M
00
I
Run Number CI
Temperature, °F 32O
Absorbent Liquid •
Packing Volume, cc 415
Liquid Feed Rate, Ib/hr 1.31
Liquid Feed Rate, Ib/hrrft2 4O2
Gas Feed Rate, SCFH 24.1
Gas Feed Rate, SCF/hr-fta 74OO
Inlet Gas Composition. Mol %
II2O 38.8
CO, 58.7
H,S 4.5
SO, O
Outlet
Liquid Product, Ib/hr 1.2
Weight % S .06
"Condensate", Ib/hr .57
Weight % 3 ca. .4
Product Sulfur, Ib/hr N.D..
Exit Gas Race, 8CFII, (dry basis) 14.1^
H,S, Mol £ K.D.
SO,, Mol O.O
$ Conversion of Feed Sulfur to
Elemental Sulfur(') N.D.
C2.
32O
System Pressure:
Packing:
31O
- &f> S0a in H2O »
415 415
1.38
424
24.1
7400
38.8
56.7
4.5
O
1.2
.11
.46
3.3
1 -°S
1 14. l(
3.4
0.2
1.45
445
26.2
8O50
34. 0
61.9
4.1
O
1.2
.07
.46
2.8
O •057(,,
1.9
.1
15 a tin (2O6
3 mm glass
C4
31O
1.68% SO,
in H,O
415
5.38
1650
28.5
87 4O
35.2
61. O
3.8
O
5.08
.06
.58
2.7
.062
17.6la)
1.5
.06
PSig)
beads
C5
310
1O$ Na,S9O,
in H2O
415
1.57
48O
29.1
8930
35.1
59.3
3.74
1.88
1.2
2.99
.41
1.74
.084
17.4(O
1.4
.03
C8
310 31O 31O
Recycle Solvent Liquor
4l(0
42
46.0
62.2
415
1.30
4OO
28.8
88 4O
35.4
59.0
3.77
1.88
1.33
.11
.43
3.9
. .O9O
17.2'
1.2
O.I
66.7
208
1.42
430
29.5
9O6O
34.5
59.9
3.68
1.84
1.57
.02
.49
4.0
.O7O
18.2
2.56
.06
51i7
101
1.59
49O
29.5
9O6O
34.5
59.9
3.68
1.84
1. 6O
.04
.53
7.4
.058
18.O
1.11
.40
43.1
- C9..
310
29.1
8940
32.6
61.8
3.73
1.86
.06
.44
5.6
.092
18.2
1.30
.62
68.5
CIO
310
CM.
31O
29.7
011O
34.3
64.0
1.151
.03
.48
.95
.023
19.2
.85
.22
54.8
28.6
8770
34.6
64.6
.540
.270
.01
.45
.SO
.0083
IB. 5
.47
.12
43.3
(i) Approximate value, determined by difference.
'(i) Maximum value, assumes no CO, loss due to solution in product liquids.
(s) In column only, excludes Wackenroder sulfur In condenser, sulfur in
recycle liquor, or thlosulfate in liquid feed.
(4) Runs C9, CIO and Cll were made with a single batch of water tilling
one half of the column. No packing was used.
H.D. - Hot determined.
-------
Table 6O
Sulfur Balances for Liquid Phase Glaus Reaction
I
(->
oo
Run Number
Feed Sulfur
As H2S
As SO 2
As S2O3=
In Recycle Liquor
Exit Sulfur
In Liquid Product
In Condensate
In Exit Gas
Elemental Sulfur Product
Losses and Accumulation
Run Number
Feed Sulfur
As H2S
As SO2
In Recycle Liquor
Exit Sulfur
In Liquid Product
In Condensate
In Exit Gas
Elemental Sulfur Product
Losses and Accumulation^3'
All
C2
Grams
4O. 8
18.8
—
—
O. 6
6.8
19. 1
24. 2^
8.9(2
C7
Grams
4O. 8
2O. 4
. 13
.14
8.9
17.9
31.7
2.6
Weights
Percent
68.5
31.5
—
—
1
11
32
) 41
) 15
Percent
66.5
33.3
O. 2
.23
14.5
29.2
51.7
4.3
on an Hourly Basis
C3
Grams
40.8
19.7
—
—
0.4
5.9
12.4
25.7
16.3
C8
Grams
4O. 8
20.4
0.3
O.3
17.8
1O. 2
26. 4
6.8
C4
Percent
67.3
32.5
—
—
0.7
9.7
21
42
27
Percent
66.3
33.2
O. 5
O. 5
28.9
16.6
42.9
11. O
Grams
4O.8
2O. 5
—
— —
1.4
7.2
1O. 6
28.1
14. O
C9
Grams
40.8
2O. 4
—
—
11. 1
9.5
41.9
(-1.4)
Percent
66.6
33.4
—
—
2.3
11.7
17.3
45.9
22.9
Percent
66.7
33.3
— —
—
18. 1
15.6
68.5
(-2.2)
C5
Grams
4O.8
20.6
18.3
—
15.7
3.3
9.2
38.3
13.3
CIO
Grams
12.86
6.43
—
—
2.08
7.73
1O. 58
(-1.1)
Percent
51.2
25.8
23. O
—
19.7
4.1
11.5
48.0
16.7
Percent
66.7
33.3
— —
—
1O. 8
4O. 1
54.8
(-5.7)
C6
Grams
4O. 8
2O. 4
—
O. 65
0.67
8. 4
7.6
4O.9
4.3
Cll
Grams
5.81
2.90
—
—
1.O2
4.13
3.78
(-.21)
Percent
66. O
23. O
—
1.0
1. 1
13.6
12.3
66. 1
6.9
Percent
66.7
33.3
—
—
11.7
47. 4
43. 4
(-2.5)
(i) By difference.
(2) Estimated as 15$ of feed.
(3) A negative number indicates the material balance exceeds
-------
Table 61
Yields of Elemental Sulfur
Average Sulfur
Yield of Yield After
Run Time, Elemental Sulfur, 2 Hours Lining
Run No. hours % of S0g + H9,S Feed Out Period, %
C3 1 0
2 12'7 42.4
3 35.6
4.O8 49.2
C4 1.5 0
2.5 43.2
3.5 39.3
A r- r- -i i- ^*^ • **
4.5 51.5
5.5 44.0
6.5 48.9
C5 .5 42.0
1.5 62.0 62.3
2.5 62.3
C6 1 0.5
2 31.8
3 62.5
4 63'° 66.7
5 63.9
6 71.0
7 76.9
7.9 62.8
C7 1 10.0
2 39.0
3 54.2
4 58.7 51.7
5 47.9
6 47.2
6.2 50.3
- 188 -
-------
Table 61 (Cont'd.)
Yields of Elemental Sulfur
Run No.
C8
C9
CIO
Cll
Run Time,
hours
1
2
3
4
5
6
6.7
1
2
3
4
5
5.5
2
3
4
5
5.5
1
2
3
4
5
6
Yield of
Elemental Sulfur,
& of S02 + H-S Feed
7.7.
17.6
39.4
41.8
48.4
37.7
48.0
43.6
60.3
71.2
67.5
69.9
65. 4
50.8
53.4
54.4
60.6
50.8
0.0
8.0
52.8
34.4
41.3
44.8
Average Sulfur
Yield After
2 Hours Lining
Out Period. jt>
43.1
68.5
54.8
43.3
- 189 -
-------
Table 62
Run
No.
Cl
C2
C3
C4
C5
C6
C7
C8
C9
CIO
Cll
Analysis of
Approxi-
mate Cycle
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
3
4
5
6
1
2
3
4
5
1
2
3
4
5
6
(3)
(3)
(3)
_fiB_
2.3
2.5
2.2
2.4
2.0
1.9
2.0
2.1
2.1
2.1
3.7
3.8
3.8
2.7
3.O
2.9
2.8
2.8
2.8
2.5
2.1
2.2
2.5
2.4
2.5
2.3
2.2
2.3
2.1
2.1
2.1
2.5
2.8
Glaus Product Liquor
Total S,
Weieht %
.06
.11
.11
.04
.10
.08
.07
.05
.02
.06
2.85
2.80
2.99
.03
.13
.09
.10
.11
.09
.02
.04
.04
.02
.01
.02
.04
.06
.03
.04
.04
.06
.03
.01
S as S04,
Weight %
.04
.04
-
-
-
-
.06
.04
.04
.04
.81
.75
.68
.01
.03
-
.03
-
.04
.01
-
.02
-
.01
.01
-
.02
-
-
.04
.03
.02
.01
S via Iodine( 2)
Titration. Wt •&
.03
.03
.03
.02
0.0
(i) Runs C1-C5 were not cyclic. Several analyses
from once-through operation are given.
(2) Calculated as if S203~ were being detected.
(3) Runs C9, CIO and Cll did not involve cyclical
operations. The analysis given is for liquid
withdrawn at the end of the run.
- 190 -
-------
Run C3
At 320°F the viscosity of sulfur is 23 cp as opposed to 7 cp at 310°F. Run C3
was made at 310°F to see whether this would improve sulfur removal from the
column. During the third hour of operation, elemental sulfur equivalent to 42$
of the feed sulfur was obtained. Results are detailed in Tables 59 through 62.
Run C4
Run C4 was made at about four times the liquid throughput rate of Run C3; the
gas rate was constant. It was anticipated that the higher liquid rate would
increase holdup in the column. The increased liquid volume available for
reaction should then have increased the conversion.
Reference to Table 61 shows that the average sulfur yield increased only 3.5$
absolute. Assuming the dynamic liquid holdup to be proportional to the one-
third power of the liquid flow rate,(18) a 3.7-fold increase in flow would
produce a 55$ increase in dynamic holdup. However, if the static liquid holdup
were a large fraction of the total, this would explain the small response to
the change in flow rate.
Run C5
Run C5 was made with 10 wt. $ sodium thiosulfate (anhydrous) added to water.
Sulfur dioxide was fed mixed with C02 as was the H2S. This method is equiva-
lent to feeding an S02-H20 solution, since essentially all the SO2 flashes from
the S02-H20 solution as the process temperature is approached.
According to previous work at ambient temperature,'19) thiosulfate is supposed
to enhance the reaction. Conversion of sulfur based on H2S and S02 feed was
62$. Although this represents an improvement over the previous runs, the
improvement was not as dramatic as had been hoped for. In addition, about one-
fifth of the feed thiosulfate ended up as sulfate at run conditions, and the
sulfur balance showed another 14$ of the feed thiosulfate apparently ended up
as sulfur which added 4$ to the conversion of H2S and S02. The analyses of
several product samples are given in Table 62.
C. Recycle of Product Liquor
Runs C6 to C8
These runs were made with S02 and H2S entering the gas feed. Liquid product
was collected, filtered of sulfur fines and returned to the feed pots. Samples
were taken periodically; a cycle was completed in about 1.4 hours. Run C7 was
carried out with half of the column packing removed, and Run C8 was carried out
with three-fourths of the column packing removed.
In Run C6 an average conversion1 of 66.7$ was obtained. This was the best
result yet obtained in the column. The most comparable run, C3, showed only
42.4$ average conversion. The major difference was the use of recycle liquor
in Run C6. These observations indicate the build up of a component in the
recycle liquid which reaches a steady-state concentration or level of activity
after a few cycles.
- 191 -
-------
Reference to Table 61 and Figure 48, shows that there was no trend in sulfur
yield with cycling after the initial two hours of operation. Runs C6 to C8
ran for about six cycles. The lab setup was not equipped to run for many more
cycles than this. Table 62 shows no trend in pH or sulfur content of the
product liquor.
All the runs made without additives showed a pH of 2 to 3 in the product liquor.
Apparently, this acidity was quickly established and remained constant. The
product of Run C4 because of high throughput, saw the least reaction per unit
liquid, and the pH was 2.1. The product of Run C6 went through the unit six
times and the pH was 2.8.
All liquor products showed residual sulfur contents between 0.03 and 0.11 wt. %
sulfur. A significant fraction of this was analyzed to be the sulfate form as
can be seen from Table 62. The product liquor was also titrated with iodine,
and the results of these titrations are given in Table 62. Since a number of
sulfurous compounds can be detected via this method, it is not certain which
species is actually present. The analysis merely shows the presence of a
sulfur-bearing compound other than sulfate. It is also noted that the low con-
centrations involved preclude a high degree of accuracy at this time.
For runs C6 to C8, the conversion to sulfur is plotted versus the fraction of the
column containing packing in Figure 49. Allowing for scatter, the data could
fit either a first or second order system. The first order dependency can exist
in a diffusion controlled or reaction rate controlled regime. Second order
dependence is consistent with reaction control mechanisms. If the system were
reaction rate controlled, increasing the liquid volume would increase conversion.
D. Batch Liquid Operations
Runs C9 to Cll
Runs C9 through Cll were made with no packing and the column half filled with
liquid, thereby increasing the volume of liquid per unit volume of reactor. The
system was repiped so that gas entered through a sparger at the bottom and was
withdrawn overhead. There was no liquid circulation; sulfur was withdrawn from
the bottom of the column.
The average conversion in Run C9 was 68.5$. This was the highest conversion
yet observed and it was obtained with only one half the column volume. The
result may be compared with that of Run C7 when the column was 50$ packed and
the conversion was 51.7$.
Runs CIO and Cll were made at decreasing inlet gas concentrations. The inlet
of CIO was the calculated outlet gas composition of C9, and the inlet of Cll
was the outlet of CIO. For example, the H2S concentration leaving the column
in Run C9 was (wet basis): 3.73 x (l - 0.685) =1.175$. The inlet gas for
Run CIO was 1.15$. Thus, a single long column was simulated by taking three
reaction sections. Combination of the results gives 86$ conversion for 100$
of column volume (68.5 + .3 5 x 54.8) and 92$ conversion for 150$ of column
volume. Since the gas throughput per unit volume of column and the gas con-
centration both match the process flow sheet, it is concluded that the column
size listed in the Annual ReportC1/and in this report is in reasonable
concurrence with experimental results.
- 192 -
-------
(O
s g
5 a
_
0 i "
i-y t
- 193 -
-------
(M
CO
to
I •
o S
_
Ol |j"
-O g
X- "
- 194 -
-------
Figure 50 shows conversion as a function of inlet gas concentration for Runs
C9 through Cll. In a first order system, the fractional conversion would be
independent of concentration. For second order kinetics, conversion would be
more dependent upon concentration than our results show. Our system appears
to be somewhere in between first and second order.
Table 62 gives the analysis of the aqueous product liquid remaining at the
end of the run. The pH, sulfur and sulfate figures are consistent with earlier
recycle data. Since the liquid from the later runs was in the column for the
entire run, it represents more residence time than the longest recycle run.
The absence of change in the liquor composition corroborates the conclusion
that no problems will occur due to recycling.
E. Reaction Products - Effect on Rate of Sulfur Production
At first it was thought that conversion was low early in a run due to the time
required to build up an inventory of slowly draining sulfur in the column.
Washes of the column after Runs C7 and C8, gave 7 to 9 grams of sulfur,
respectively. This was not sufficient to account for the sulfur that should
have been made during the first two hours if the rate were the same as the
average of the last four hours. Thus, the buildup of some component in the
liquid which improves conversion is indicated.
Figure 48 shows the rate of sulfur production as a function of time for the
recycle runs in which the height of packing was progressively reduced. If the
delay in reaching final level of conversion were due to slow drainage of sulfur,
one would expect the delay to decrease as the height of packing was reduced.
Such was not the case, and it appears that there is simply an initiation period
in which compounds that are beneficial to the reaction buildup.
The above conclusion fits the earlier observation that the conversion was higher
in recycle Run C6 than in the once-through Run C3.
Table 61 shows that Runs C9 and CIO did not experience a significant initiation
period. Presumably this was due to the liquid not being circulated so that
beneficial products accumulated rapidly. Run Cll did show an initiation period
but this may have been due to sampling difficulties as well as the low concen-
tration of sulfur in the feed gas. Run C5 also did not display an initiation
time, and this may have been due to the presence of thiosulfate and its decompo-
sition products in the liquid feed.
Using a batch aqueous Claus system at ambient conditions, Mathieu(2°) found it
required about 20 minutes for desulfurization to reach 90$ of its maximum value,
and about 10O minutes to attain the maximum value.
Corrosion
Our reaction medium is potentially very corrosive. The stainless steel valve
through which sulfur was drained showed pitting. The column was not examined
under magnification, but there was no obvious corrosion or scaling.
- 195 -
-------
N
CO
to
o
S g
u J
Ol U
»-- U
.,°^
X
ox
: i :
|
±
; ;
\;>
EE
- 196 -
-------
F. Conclusions
The liquid-phase Claus reaction proceeds readily under simulated process condi-
tions. There is no build up of acidity, sulfur or sulfate as the solvent liquor
is exposed to the reaction. Conversions corresponding to over 90$ conversion of
the feed sulfur have been demonstrated.
The reactor volume needed to reach 90^ recovery of sulfur via the liquid-phase
Claus is in the range given in the previous process estimate.V1)
The system has a gas concentration dependence between first and second order.
- 197 -
-------
XII. GASIFICATION KINETICS
A. Introduction
Four series of gasification kinetics runs were made to supply information needed
for gasifier design. Two temperatures (1620 and 1700°F) and two carbon burnoff
levels (nominally 30$ and 60$) were used. All runs were made with chars produced
from a Pittsburgh Seam coal from the Loveridge Mine.
To avoid large changes in the partial pressures of the reactants and products
over the length of the gasifier bed, run conditions were chosen to give low
gasification rates.
B. Method
The reactor was the 4 inch I.D. vessel used in the coal pretreatment studies.
A flow diagram of the system is given in Figure 13.
The inlet gases were various mixtures of steam, recycle gas, and added N2, H2;
and C02 to give the desired inlet partial pressures. By varying the ratio of
added H2 and C02, the desired inlet partial pressure of CO was obtained after
the entire inlet gas mixture had reached water gas shift equilibrium. Experience
has shown that this equilibrium is established rapidly in the bottom of the
reactor.
Char was fed continuously and the unreacted product was removed continuously via
an overflow weir which controlled the bed height at 41 inches. The weight of
material in the bed was measured by the pressure drop across the entire bed.
The runs in each of the four series were made continuously and sequentially,
except for weekend shutdowns. Temperature and inlet gas flows and compositions
were held constant for at least three nominal bed inventory changes and then at
least two, 2-hour balance periods were made. The inlet gas condition then was
changed to that for the next run and the above procedure was repeated. In all,
26 runs were made.
C. Feedstock Preparation
Preparation of the feedstocks for both 30$ burnoff series is described in Section
V. The feedstocks for the two 60$ burnoff series were the composited products,
including the lineout material, from the corresponding 30$ burnoff series.
D. Data Workup
The data workup was straightforward, being based on measured values for the
input and output streams, the dry exit gas rates, condensate rates (unreacted
steam), and the dry product gas analyses. The gas chromatograph used for the
product gas analyses was not designed to give complete resolution of CO and N2.
The resolution is such that the CO content can be measured to within 10$, e.g.,
10 ± 1$, which is sufficient for most purposes. In this study, the CO content
of the dry gas was calculated to correspond with water gas shift equilibrium in
the total gas leaving the top of the bed.
- 199 -
-------
E. Results
The compositions of the four feedstocks which were used are given in Table 63.
Properties of the char products are shown in Table 64. Run conditions and
results for the 1620°F and 1700°F runs are in Tables 65 and 66, respectively.
Some of the entries in the tables are given with more significant figures than
the accuracy of the data warrants. This was done only to avoid errors due to
rounding.
Results from two runs, K3A and K16A, are not included in the tables. Steady
state never was reached in these runs because of char feeder aberrations.
F. Correlation of Rates
For correlative purposes the assumption was made that an "effective" gas com-
position can be assigned to each run. For this work the effective gas composi-
tions were taken, somewhat arbitrarily, as 90$ of the outlet gas partial
pressure plus 10$ of the inlet gas partial pressures listed in Tables 65 and
66.
Rate correlations were attempted using the reaction models developed in an
earlier study with lignite and subbituminous coal charsV3>7>21) but without
success.
Table 67 gives the reaction rates, effective partial pressures, and the mean
burnoff level for each run.
G. Conclusions
In the absence of any kinetics data for bituminous coal chars which involve the
combined inhibiting effects of both H2 and CO, our process estimates assumed a
total gasification rate of 1/15 the rate predicted by our kinetics correlation
for lignite char gasification.^,7)
Comparison of the calculated rates (multiplied by 1/15) for the effective
partial pressures given in Table 67 with the rates measured at high levels of
inhibition shows that the calculated rates are considerably higher and that the
deviation becomes greater as the H2/CO ratio increases. Although we have no
new data at low levels of inhibition at process conditions, the end result
undoubtedly will be that the total gasification rate will be lower than previously
predicted.
The process implication is that more steam will be needed in the gasifier, which
in turn will decrease the thermal efficiency. Design calculations cannot be
made until the new data are correlated successfully. Additional experimental
data are desirable in order to achieve a more reliable correlation.
- 200 -
-------
Table 63
Feedstock Compositions for Kinetics Study
Feedstock Used in Run
Hydrogen, Wt . % (dry basis)
Carbon
Nitrogen
Oxygen (by difference)
Sulfur
Ash
Size. Tyler Mesh. Wt . %
28 x 35
35 x 48
48 x 65
65 x 100
-100
Particle Density. lb/ft3
28 x 35
35 x 48
48 x 65
65 x 100
-100
Kl -» K7
.66
84.32
1.34
-.77
2.03
12.50
17.7
30.5
30.9
18.9
2.0
89.4
90.5
89.8
91.0
85.2
K1A -» K7A K10 -» K16 K10A -» Kli
.53 .52 .41
82.93 84.72 82. 2O
.78 1.32 .71
-.58 -.98 -1.00
1.12 1.85 1.15
15.22 12.63 16.53
- 2O1 -
-------
Table 64
Properties of Products from
Gasification Kinetics Runs
CO
Product from Run
Hydrogen, Wt . $ (dry basis)
Carbon
Nitrogen
Oxygen (by difference)
Sulfur
Ash
Size. Tyler Mesh. Wt. %
+28
28 x 35
" .35 x 48
48 x 65
65 x 1OO
-1OO
Density of Fraction r lb/ft3
28 x 35
35 x 48
48 x 65
65 x 10O
Mean Density, lb/ft3'1'
Mean Diameter, inch(a)
Product from Run
Hydrogen, Wt . $ (dry basis)
Carbon
Nitrogen
Oxygen (by difference)
Sulfur
Ash
Size. Tyler Heshr Wt . I
+28
28 x 35
35 x 48
48 x 65
65 x 1OO
-1OO
Density of Fraction,, lb/ft3
28 x 35
35 x 48
48 x 65
65 x 1OO
Mean Density, Ib/ft3(1)
Mean Diameter, lnch(3)
(i) Reciprocal mean. •
(a) Arithmetic mean.
Kl
.'54
82.56 '
.72
-.44
1.27
15.35
.O
12.6
33.9
32.0
19.6
1.9
78.2
78. 0
72.9
73.1
76.8
.0119
K10
.36
81.81
.66
-1.01
1.23
16.95
.O
14.8
33.5
30.6
18.5
2.6
77.2
76.3
76.6
76.6
76.6
.O12O
K2
.57
82.96
.73-
-.56
1.16
15.14
.0
14.0
34.2
31.4
18.4
2.O
77.5
81.4
80.1
79.3
80.0
.O12O
Kll
.40
81.99
.65
-.53
1.12
16.37
.O
15.8
31.8
31.4
18.3 .
2.7
79.2
75.6
. 75.3
77.4
76.4
.0121
K3
.59
81.84
.75
-.97
1.38
16.41
.O
12.1
33.2
32.6
20. 5
1.6
78.2
76.4
7O.9
75.9
74.6
.0118
K12
.39
80.77
.67
-1.31
1.31
18.17
.O
12.6
31.2
32.4
20. 6
3.2
76.6
72.9
73.4
72.5
73.4
.0116
K4
.58
82.81
.73
-.61
1.O4
15.45
.O
17.5
32.9
30.9
17.3
1.4
82. 0
79.8
76.9
80.6
79.4
.01 24
K13
.40
81. 07
.68
-.53
.89
17.49
.O
15.9
34 .O
31.4
16.9
1.8
76.8
75.1
75.5
73.9
75.3
.O123
K5
. .62
82.17
.70
-.16
l.OO
15.67
.O
14.3
32.5
31.9
19.2
2.1
79. 0
8O.7
79.9
76.8
79.3
.0120
K14
.43
81.21
.67
-.66
1.02
17.33
.O
17.9
32.2
30.5
17.1
2.3
77.8
73.9
72.8
76.8
74.8
.O124
K6
.59
83.27
.77
-.62
1.10
14.89
.O
14.3
32.8
32.3
18.7
1.9
86.2
82.2
81.8
81.2
82.4
.O120
K15
.41
82.73
.67
-1.01
l.OO
16.20
.O
18.5
33.5
29.0
17.0
2.0
80.7
79.2
77.6
79.5
79.1
.01 25
K7
.53
82.86
.89
-.34
1.40
14.66
.0
17.5
31.1
29.9
18.9
2.6
8O.8
84.2
83. 0
83.4
83. 0
.0122
K16
.42
83. OO
.71
-.89
1.27
15.49
.O
11.9
32.0
36.2
18.4
1.5
80.7
SO. 6
81.1
81 .0
80.9
.0118
K2A
.53
8O.O5
.61
+ .18
.44
18.19 .
.O
10.9
33.6
33.1
20. 3
2.1
73.4
71.8
71.9
71.6
71.9
.O117
K11A
.41
78.89
.52
-1.27
.49
20.96
.O
17.2
34.0
30.1
16.3
2.4
67.9
66.5
67.5
67.7
67.2
.01 24
K4A
.54
78.64
.64
-.25
.35
2O. O8
.O
12.1
31.7
31.7
21.6
2.9
74.5
71.8
67.2
7O.O
70.1
.0116
K12A
.32
75. 09
.53
-.11
.51
23.66
.O
11.1
32.6
35.4
19.6
1.3
67.2
62.2
58.1
62.6
61.3
.O117
K5A
.51
78.25
.61
+ .05
.35
20.23
.O'
15.7
33.8
30.6
17.6
2.3
70.4
66.8
71.2
67.4
68.8
.0122
K13A
.32
76.18
.52
-.33
.31
23. OO
.O
12.5
32.3
31.6
2O. O
3.6
69.0
61.6
61.0
69.4
63.9
.O117
K6A
.55
79.47
.62
+ .O8
.38
18. 9O
.O
13.9
31.3
30.7
21.3
2.8
74.5
68.6
7O.8
72.1
7O.9
.0118
K14A
.34
75.52
.SO
-.42
.34
23.72
.0
11.4
32.2
32.9
20. 6
2.9
65.7
64.4
68.2
61.8
65.1
.O116
K7A
.53
79.71
.60
-.17
.62
18.71
.O
15.5
35.1
29.2
17.5
2.7
73.7
73.7
73.3
69.9
72.8
.0122
K15A
.42
77.89
.49
-.90
.30
21. SO
.O
14.2
33.6
31.2
18.9
2.1
69.5
64.1
62.8
62.8
64.1
.O12O
-------
Table 65
8
co
Run No.
Feedstock
Feedrate, Ib/hr MF
Recycle Flu id iz ing Gas, SCFH
Recycle Lift Gas, SCFH
Inlet Gas
Steam, SCFH
Ha
C02
N2
Total Inlet Gas, SCFH
Purges (N2) above Bed, SCFH
Inlet Gas Partial Pressure, atm
H2O
H2
CH,
CO
C02
Outlet Gas
Product Gas, SCFH
N2 Purges
Condensate
Total Outlet Gas, SCFH
Fluidizing Velocity, at top of Bed, ft/sec
Partial Pressures, at top of Bed.. atm
CH4
CO
CO,
Ha
Product Gas Composition, mol
Ha
CH.
co
co2
% Steam Conversion
Char Bed Density, lb/ft'
Id
Conditions and Results for 162O°F Runs
System Pressure; 15 atm (2O6 psig)
Reactor ID - 4" Bed Height - 41"
K2
K3
K4
KS
K6
K7
K2A
K4A
K5A
K6A
K7A
f Loveridge Char Treated at 16<
1.69
97. O
7O.5
65. S
18.0
18.0
61. 0
330. 0
3.O25
1.8O3
.O7O
.985
1.309
7.8O9
146.7
56.8
356. 0
.306
2.390
2.176
.126
1.467
1.278
7.563
16.49
.95
11. 09
9.66
61.85
18.4
31
2.26
72.9
69.9
93.5
34. 0
14.O
39.0
323.3
4.348
3.137
.117
.879
.965
5.554
137.4
82.8
348.0
.299
3.564
3.49O
.223
1.288
1.O42
5.392
28.93
1.85
10.67
8.64
49.90
15.5
33
2.89
78.2
71.5
V47.O
.O
4O.O
.0
336.7
6.256
1.387
.087
1.181
4.22O
1.87O
111.9
121.4
368.0
.316
4.947
2.111
.143
2.113
3.925
1.761
19.81
1.34
19.83
36.83
22.19
16.0
31
3. 2O
19.8
65.6
153.5
46. 0
32. 0
16. O
332.9
6.978
2.899
.099
.891
1.698
2.435
154.6
133.9
358.9
. 3O8
5.598
3.475
.270
1.449
1.847
2.361
34.64
2.69
14.44
18.41
29. B3
17.3
32
V*>1P
3.94
.O
69.2
187.2
28. 0
53. O
.O
337.4
8.181
1.758
. O66
.780
2.877
1.337
167. 0
161.3
382.5
.328
6.326
2.718
.210
1.583
2.92O
1.242
29.35
2.27
17. 09
31.53
19.76
16.8
32
3.96
.O
65.5
136.8
8O.O
.O
49. 0
331.3
5.913
4.812
.098
.113
.110
3.953
189.2
15 O
116.7
356.4
.306
4.9O6
5.O4O
.372
.476
.367
3.839
47. 04
3.47
4.44
3.42
41.63
14.6
31
4.62
43.1
68.7
138.5
.O
.O
65. 0
315.3
6.1O5
1.315
.082
.174
.638
6.685
158.3
97. 0
352.1
.302
4.133
2.567
.185
. 8O9
1.032
6.274
22.30
1.61
7.02
8.97
60.10
27.9
34
1.66
68.7
67.3
95.6
34. O
21. 0
51. O
337.6
4. 36O
2.509
.077
.882
1.215
5.958
149.6
88.5
359.1
.308
3.70O
2.8OO
.158
1.215
1.27O
5.858
23.44
1.32
10. 17
10.63
54.44
13.7
29
2. 1O
21.3
65.7
150.9
47. 0
32. 0
8.Q
324.9
7.062
3.04O
.101
.955
1.757
2.O86
137.1
137.2
346.3
.297
5.946
3.465
.254
1.393
1.892
2.O48
35.67
2.62
14.35
19.48
27.88
14.3
28
Ite KL - K? Product -
2.50
.O
68.8
176.9
32. 0
51.0
.O
328.7
8.159
1.934
.057
.863
2.883
1.1O4
141.5
16O.8
356.1
.306
6.771
2.619
.168
1.439
2.948
LOSS
29.58
1.89
16.25
33.29
18.99
13.2
28
2.60
.O
65.6
136.3
78. 0
;O
47. 0
327.2
5.970
4.788
.086
.094
.093
3.969
178.4
118.3
347.3
.298
5.113
4.972
.317
.378
.308
3.912
47.12
3.OO
3.58
2.92
43.38
13.1
29
x.
3.05
48.6
68.5
159.5
.O
.O
49. O
325.6
X
y
6.697
1.238
.065
.143
.612
6.245
144.2
131. 0
377.3
.324
5.205
2.300
.134
. 56O
1.004
5.797
22.17
1.29
5.40
9.68
61.46
16.3
29
-------
Table 66
Conditions and Results for 17OO°F Runs
Feedstock
Feedrate, Ib/hr MF
Recycle Fluidizing Gas, SCFH
Recycle Lift Gas, SCFH
Inlet Gas
8
Steam, SCFH
H2
C02
N2
Total Inlet Gas, SCFH
Purges (N2) above Bed, SCFH
Inlet Gas Partial Pressure, atm
H2O
H2
CH4
CO
C02
N2
Outlet Gas
Product Gas, SCFH
Nj Purges
Condensate
Total Outlet Gas, SCFH
Fluidizing Velocity, at top of Bed, ft/sec
Partial Pressures, at top of Bed, atn
H,0
H,
CH4
CO
C02
N2
Product Gas Composition, mo l
H,
CH4
CO
C02
$ Stean Conversion
Char Bed Density, Ib/ft*
System Pressure:
Reactor ID - 4"
no
f
1.78
117.8
70.4
52.3
18.0
23. 0
64. 0
345.5
^
2.411
1.751
.056
1.497
1.438
7.847
161.0
44.7
378.9
.338
1.767
2.181
.099
2.23O
1.261
7.462
15.78
.72
16.14
9.13
58.23
22.9
31
m
KL2
KL3
LoVGrid^** PVia f T»«»n t*»H ft *•
2.24
95.2
68.8
68.7
34. 0
19.0
46. O
331.7
3.268
2.997
.097
1.4O1
1.065
6.172
152.9
56.8
358.7
.320
2.375
3.404
.181
2.065
1.006
5.968
25.68
1.37
IS. 58
7.58
49. SO
24.8
30
2.80 3.18
119.7
71.1
118.3
.O
46. O
.O
355.1
4.739
1.433
.101
2.O27
4.676
2.024
120.2
92.2
388.2
.346
3.563
2.O05
.155
3.296
4.O79
1.9O3
16.70
1.29
27.45
33.97
2O.59
20.0
30
60.2
67.9
122.1
50.0
29.0
11.0
340.2
5.569
3.747
.130
1.496
1.550
2.508
159.6
106.9
379.6
.338
4.23O
4.313
.274
2.277
1.556
2.350
37.87
2.41
2O.OO
13.66
26. 06
19.O
32
15 atm (2O6
Bed Height -
K14
3.86
47.8
70.1
139.2
32. O
53. 0
.O
342.1
6.296
2.374
.093
1.692
3.129
1.416
161.2
118.3
382.4
.341
4.641
3.161
.205
2.807
2.871
1.316
28.87
1.87
25.63
26.22
17.41
2O. 7
29
PSig)
• 41"
K15
3.91
35.3
66.8
1O8.6
83.0
.O
45. O
338.7
4.574
5. 688
.123
.274
.154
4.188
197.8
83.5
368.4 '
.328
3.401
6.022
.348
.866
.341
4.021
49.31
2.85
7.O9
2.79
37.95
22.6
30
K16
4.60
87.7
70.1
115.8
.O
.O
55.0
328.6
4.752
2.O56
.077
.515
.829
6.771
157.3
73.8
373.9
.333
2.959
3.216
.142
1.5O1
.964
6.218
25.43
1.12
11.88
7.63
53.92
32.2
34
K11A
i
1. 49
95.3
68.9
69.5
36. O
19. O
44. 0
332.7
3.334
3.061
.095
1.363
1.035
6.113
146.6
62.4
358.2
.319
2.613
3.386
.174
1.883
1.O14
5.930
26.03
1.33
14.47
7.80
SO. 36
19.2
29
K12A
1.89
118.1
70. 1
111.1
.0
45.0
.O
344.3
4. 592 '
1.274
.091
1.933
4. 86O
2.249
111.8
89.9
. 374.9
.335
3.588
1.773
.137
3.O6O
4.315
2.127
14.81
1.15
25.55
36. 03
22.46
17.4
3O
K13A
2.06
59.2
66.8
118.5
51. O
32. 0
11.0
338.5
5.531
3.584
.122
1.538
1. 654
2.571
149.1
1O9.8
369.9
.330
4.451
4.011
.255
2.158
1.668
2.457
35.96
2.29
19.35
14.95
27.45
15.8
27
KL4A
2. 44
47.5
69.6
141.7
32.0 '
52.0
.0
342.8
6. 461
2.350
.085
1.656
3.174
1.273
148. 0
124.9
375.0
.334
4.999
3.039
.182
2.599
2.983
1.198
28.63
1.72
24.49
28.11
17. 04
17.81
28
K15A
V
2.60
35.3
66.9
1O2.2
85.0
.0
47. O
336.4
4. 426
5.749
.116
.211
.113
4.384
189.3
84. O
36O. 5
.321
3.495
3.964
.326
.646
.264
4. 305
49.19
2.69
5.33
2.18
40.61
19.6
25
-------
Table 67
to
O
01
Kinetics Summary
System Pressure 15 atm
Run No.
Temperature, °F
Feedstock
Mean Burnoff, %
15 - PNJ
Effective Partial Pressure, atm
H20
H,
CO
CH4
Gasification Rate*
Total Gasification Rate, measured
Methane Formation Rate, "
Total Gasification Rate, "calculated"
Run No.
Temperature, °F
Mean Burnoff, %
15 - PN2
Effective Partial Pressure, atm
H,O
H2
CO
CH«
Gasification Rate*
Total Gasification Rate, measured
Methane Formation Rate, "
Kl
<.
30
6.13
2.454
2.139
1.419
.120
9.6O
.97
28.0
K1O
* — —
39
6.22
1.831
2.138
2.157
.095
13.46
.82
K2
25
8.56
3.642
3.455
1.247
.212
9.74
1.64
41.5
KL1
I
30
8.0O
2.464
3.363
1.999
.173
13.34
1.5O
K3 K4
Char Treated at
32 26
9.27 1O. SO
5.O78 5.736
2.O39 3.417
2.O20 1.393
.137 .253
17.68 15.39
1.O6 2.78
35.6 55.4
K12 K13
38 33
8.95 11.08
3.681 4.364
1.948 4.256
3.169 2.199
.ISO . 2SO
21,39 19.88
1.13 2.67
K5
1 gfyy f . ,
3O
10.83
6.512
2.622
1.503
.196
21.48
2.54
60.4
K14
17OQP F —
32
10.78
4. 8O7
3.O82
2.696
.194
25.24
2.27
K6
162O - -
20
1O. 81
5.007
5.017
.440
.345
14.88
4.54
100.8
K15
24
1O. 64
3.518
5.989
.807
.326
18.29
4.O9
K7
N
22
7.69
4.330
2.442
.746
.175
17.72
1.62
86.2
K16 -
26
7.78
3.138
3.10O
1.4O2
.136
2O. 67
1.13
K2A
44
7.87
3.766
2.771
1.182
.ISO
9.21
1.5O
48.3
K11A
s
\
57
8.O4
2. 685
3.354
1.831
.166
12.30
1.54
K4A
48
11. 07
6.O58
3.423
1.349
.239
14. SO
2.89
60.0
K12A
65
8.49
3.688
1.723
2.947
.132
20.11
1.O2
K5A
Id - K7
49
11.00
6.910
2.551
1.-381
.157
17.46
2.14
7O.O
K13A
USA
42
10. SO
5.199
4.954
.350
.294.
12.86
4.06
115.8
K14A
59 6O
1O.86
4.559
3.968
2.O96
.242
19. SO
2.94
1O.79
5.145
2.970
2. SOS
.172
23.68
2.11
K7A
\
f
46
8.16
5.354
2.194
.518
.127
18.14
1.38
136.3
KL5A
•v
49
10.44
3.588
S.943
. 6O3
.305
17.42
4.59
Total Gasification Rate, "calculated"
33.4
46.1
34.7
63.6
S3.6
148.2
91.O
56.2
39.3
7O. 8
63.0
187.8
Units are; Ib carbon gas if ied/ninute/lb carbon in bed xlO*
-------
XIII. BIBLIOGRAPHY
1. Curran, G.P.; Clancey, J.T.; Fink, C.E.; Pasek, B.; Pell, M.; and
Gorin, E., "Development of the C02 Acceptor Process Directed
Towards Low-Sulfur Boiler Fuel/' Annual Report to Control Systems
Division, Environmental Protection Agency, Under GAP Contract
EHSD 71-15, Period: Sept. 1, 1970 to Nov. 1, 1971.
NTIS Accession No. PB 210-840.
2. Consolidation Coal Co., Research & Development Report No. 16 to the
Office of Coal Research, U.S. Dept. of the Interior,
Under Contract No. 14-01-OOO1-415 (1970).
Book 1, "Studies on Mechanics of Fluo-Solids Systems."
Gov't. Printing Office Catalog No. 163.10:16/INT3/Book 1.
3. Ibid.
Book 3, "Operation of the Bench-Scale Continuous Gasification Unit."
Gov't. Printing Office Catalog No. 163.10:16/INT3/Book 3.
4. Goring, G.E.; Curran, G.P.; Tarbox, R.P.; and Gorin, E.,
Ind. Eng. Chem., _4_4, 1051 (1952).
Ibid. .44, 1057 (1952).
5. Goring, G.E.; Curran, G.P.; Zielke, C.W.; and Gorin, E.,
Ind. Eng. Chem., 45. 2586 (1953).
6. Zielke, C.W.; and Gorin, Everett, Ind. Eng. Chem., 47. 820 (1955).
Ibid., 49, 396 (1957).
7. Consolidation Coal Co., Research & Development Report No. 16 to the
Office of Coal Research, U.S. Dept. of the Interior,
Under Contract No. 14-01-0001-415 (1970).
Book 2, "Laboratory Physico-Chemical Studies."
Gov't. Printing Office Catalog No. 163.10:16/INT/Book 2.
8. Robson, F.L.; Giramonti, A.J.j Lewis, G.P.; Gruber, G.; "Technology and
Economic Feasibility of Advanced Power Cycles and Methods of
Producing Nonpolluting Fuels for Utility Power Stations."
United Aircraft Research Laboratories Report No. J-970855-13,
Dec., 1970, Under NAPCA Contract CPA 22-69-114.
9. Sherwood, T.K., Ind. Eng. Chem., 17. 745 (1925).
10. Adams, F.W., Trans. Am. Inst. Chem. Engrs.. 25. 424 (1933).
11. Westinghouse Research Laboratories, Report to the Office of Air Programs,
Environmental Protection Agency, Under Contract No. CPA 70-9 (1971),
Volume III, Appendix M.
12. Katell, S.; and Faber, J.H.j Petr. Ref.; 39. No. 3, 187-190 (i960).
13. Kulik, M.D., Consol Internal Report, RM-11968 (1968).
- 207 -
-------
14. Consolidation Coal Co., Pilot Scale Development of the CSF Process.
Research & Development Report No. 39, Vol. IV, Book 3. Prepared
for the Office of Coal Research (August, 1971).
U.S. Gov't. Printing Office Stock No. 2414-0033.
15. Ruth, L.; Squires, A.M.; and Graff, R.A.; Environmental Science and Tech..
6., 12, 1009-1014, Nov., 1972.
16. Craig, J.W.T.; Johnes, G.L.; Moss, G.; Taylor, J.H.; and Tisdall, D.E.;
"Study of Chemically Active Fluid Bed Gasifier for Reduction of
Sulfur Oxide Emissions," Final Report to GAP on Contract CPA 70-46
from Esso Research, England, June, 1972.
17. Jasulaitis, W.A., Consol Internal Reports, RM-11141 and RM-11168 (1966).
18. Satterfield, C.N.; Pelossof, A.A.; and Sherwood, T.K.;
AIChE J.. 15., 2, 226-234 (1969).
19. Feld, W.; U.S. Patent No. 1,079,291 (1913).
20. Mathieu, P., Annales du Genie Chimique. .3, 107-115 (1967).
21. Curran, G.P.; Fink, C.E.; and Gorin, E.; IEC Proc. Design and Development.
8., 559 (1969) .
22. JANAF Tables, Dow Chemical Company, Clearinghouse for Federal Scientific
and Technical Information, U.S. Department of Commerce,
No. PB 168-370 (1965) and No. PB 168-370-1 (1966).
23. Preuner, G.; and Schupp, W.j Z. Physik. Chem.. 68. 129 (1909).
24. West, J.R.; Ind. Eng. Chem.. 42. 713 (1950).
25. Curran, G.P.; Fink, C.E.; and Gorin, E.; "C02 Acceptor Gasification Process -
Studies of Acceptor Properties," in "Fuel Gasification," Advances in
Chemistry Series No. 69, ACS, Washington, D.C. (1967).
26. Zawadski, J.; Z. Anorg. Chem.. 205. 180 (1932).
27. Hill, K.J.; and Winter, E.R.S.; "Thermal Dissociation Pressure of Calcium
Carbonate," J. Phys. Chem., 60. 1361-1362 (1956).
28. Uno, T.j Tetsu to Hagana. 37. 14-17 (1951) .
29. Rosenqvist, T.; "A Thermodynamic Study of the Reaction CaS + H20 = CaO + H2S
and the Desulfurization of Liquid Metals with Lime," J. Metals Trans..
AIME. 3, 535-540 (1951).
30. Batchelor, J.D.; Yavorsky, P.M.; and Gorin, E., J. Chem. & Eng. Data. 4.
(3), 241 (1959).
31. Terres, E.; and Schaller, A.; Gas und Wasserfach. 65. 761 (1922)
32. Batchelor, J.D.; Consol unpublished data.
- 208 -
-------
Appendix A
Detailed Investment Costs
- 209 -
-------
TuUlo A-l
Coat EntiiKto; Suction 3UO; Su
I
eo
r->
o
I
l>4Ul Hctmt
Steaii Sup.tr Hatfter»
Vaate Ho»t boiler
Quench Mater Cooler
Trlai Cooler
Quench Water Cooler
Sullur Reactor
Begen. Reactor
P II
Acceptor Lock Hopper
Speot Acceptor Lock lluppwr
Duet Collector*
Duet Collector!
Du«i Collector*
Quit Collectors
Hydroclonee
Slurry f\iep
Black Vater PUep
Queach Vater Ctrc. Pu*p
Uake-up COj Coap'r
Acid Gaa Coaip'r
COa Blower
Acceptor Screw Feodor
Acceptor Elevator
H Taap. Rotary Feeder
U leap. Rotary Feeder
U Teaip. Rotary Feeder
Cooverter Agitator
Piling
fouodat lone
Structural
Electrical
laaulatloa
BuildlKc*
PtpUf
InatrtMMntation
Nunber
C-301
C-302
C~303
C-304
C-303
C-304
D-301
D-302
D~3Q3
F-301
F-302
F-303 •
F~304
0-301
C-302
0-303
0-304
0-303
J— 301
J-302
J-303
JC-301
JC-302
JC-303
L-300-
L-301
L— 302
L-303
L-304
L-303
I- lOt
DuHurlpt ton
a b. 1333$ 1-1 f*"- Float Ha., i65o°F,
230», 316 8.8. Tub.
8 b. 212001-1 Paaa. Plo.t Ha., 1600°F.
230«, 2-1/4 Cr-lUo Tub.
3 Km. 630 if U-Tub. 2a''0 Fig. In D-3U3
7S», JSO'f, 304 B. 8. Tub*
3400 4. 1-1 M». fl«. Hd. . 193*
304 B. 8. Tub.
173 0), U-Tub., C.I. , S0"0 6h.ll «
10'L, 3/«"9 Tub* i 14 B.W.C.
3400^, 1-1 rtm. Fix. Hd. , 1!3>
304 3.8. Tub.
6 E>.. 21'-6"O.D. X 47'0.8.S., 23Q«,
1700*r, B.fr. Lln.d, C-3I 4,
2 C>. 21'-3"0.0. > IS'O.S.S., 250«,
1300*r, R.tr. Lln.d, C-81 1,
3 b 3' 1 D X 7'-6"0 8 8 13f
120"r, Con. Sot., FLD, 3U4 s'. S.
• Cone. b«ie
1/4" C.8. I
4 t«. 6' 1.0.t 1730 am
1 b. 1124 CPU, 36 pal, Au.t. CBL-3"B
It 1180 BFH •
2 b. 239 CHI, 106 p»l, Auit. Iron Cent r 11.
• t 1730 BMI
7620 ACFII, 13 — 240 p.l>, 2 8tg. W/pr.cool.r
and Ibt.rcool
B80 ACFU, 13**215 ptila, 2 Stg. W/lntarcool.
Baclp.
1430 ACFII, 3 PHI, Kot.ry w/W.C.
B.arlnga, 3600 BHI
2 b. 300 O'H, 25' Ig XB" Dlu. lit 3U° Uloijo
X-H C.S.
2 b. 300 CFH, B3'C-C, Webat.r «C-123B
12 0*0 Cailng
4 b, 130 CFH 6"0 Hot Vane 100°F 230# C S
6 la. 3230 Cril, 18"B Bot. Van., 30"B Fig.
Batr. Lin. 230», 1700°F
6 Sa. 33 CFU,6"0 Bot. Van., 1B"0 Fig.,
Ba(r. Lin. 250,, 1700°F
2 la. 203 CFU,B"B Bot. Van., 20"0 Fig.,
B.(r. Lin. 23U«, 1330*F
3 la. 14"H I.p.llar, 13«, 1>U*F, 304 8.8.
Totala Major Equlpa.nt
H + L at 3UU
H » 60, L - 00
H - .21, L • 122
H • 60, L - 30 •
II - .10, L - 1.70
II • .(0, L - .60 at 43.000 cr
»v- .30 Haj. «. L . .70 Uv
H|» .08 IUJ, N, L • .40 H|
Tbtala Minor Kqulpaant
Total H t L
v.lght
•
160,800
2UO.OOO
13,000
42,000
3.000
22,000
703,1100
BO2.000
U.UOO
5, bOO
b,7QO
40, :i2G
B^,2UU
4,U60
51)4,000
379,600
121,200
112,000
4,000
2,100
3,000
3,600
BO, 000
r,
24,000
a.auo
3,100
13.U20
9 60O
42 , 000
27,000
10,200
3,300
H*rfcup. loci. •««., Supervision. Purch. , Field, UOM OCMc*,. Profit and Cunt Infancy
Total InvMitnunt
Uat.rliil
18B.600
220,000
30.73U
.43,000
2, 40U
43, OOU
1,OU4, 40U
4U7. 4OO
11,100
1,620
3,!>t>U
2b, 000
23, ^UO
5,?t>0
2UU.UUU
273,400
37,200
34,OOO
6, 1OO
3,300
6,020
11,100
193,000
39,300
5, l>oo
3, 420
M.3UU
22 000
7S,OOO
43,200
16,200
B.7IIU
3,173,330
147.01KJ
66,OUU
403,000
206,000
182,000
21.OOO
•SI, 000
190, OOP
2,182,000
B. 730,000
6.243,imO
13,000,000
_J_ Q1 H.I'. f C. ¥ 60 1 E.,
7,2011 3,7bU 73,9OO
12,OOO ti,240 67,90O 3U 20
710
1 , 200 1 , 000 3
150 400 1
1 , 200 1 , 000 3
1,020, tiOO 2b,tiuo 2J6 Uw
394,200 11,0110 1211 4M
tiOO 1,3011
1,-lSO 'J7
44O
J,b0i> l.ntiii
6,I10O O.OOII
440 J
7U,200 U,3-tO 3(3,000
72,000 9,340 36,0011
16,000 1,900 12,000
14, BOO 1,800 11,3(K)
500 1 , 000
200 10 2
300 120 7
400 BO 3
10,000 1,300 33 6
3,000 300 12
200 40 j
7OO 6 4uo 2
3.100 10 11,74(1 3
1 000 B 2 40O
5,700 300 30 7,200
4,200 IbO 12 4,800
1,6OO 70 6 1,800
430 IS 1,2UO
1,1,02.680 73,130 3,137 293,340 496 Ia2
4UO
(jti,(IOO |( IUO
317,000 1,440,0110
172,000 3,440
4OV,OOO 240, OUU
27,000
666,000
76. OOP
1,732,000
• Nut itiuvn OB Figure 4.
-------
Table A-2
Cost Estimate: Section 40O; Sulfur Recovery
Equipment
Major Sulfur Combustor
Claus Gas Reheater
Claus Gas Reheater
Heat Exchanger
Tempering Gas Cooler
Acid Cooler
Recycle Cooler
Sidestream Coolers
Liquid Phase Claus Rx.
SO2 Absorber Columns
Dust Hoppers
Separators
1
Number
B-401
C-401A
C-4O1B
C-402
C-403
C-4O4
C-405
C-406
D-401
D-402
F-401
F-402
Liquid Sulfur Storage Tank F-403
to
M
l-i Recycle Gas Comp'r.
Centrifugal Air Comp'r.
Sulfur Pumps
Acid Circ. Pumps
Acid Sidestream Pumps
B.F. Water Pumps
Absorber R. C. Pumps
Elec. -Static Precipitator
Minor Piling
Foundations
Structural
Electrical
Insulation
Buildings
Piping
Instrumentation
Mark-up. Including Eng'r. ,
JC-401
JC-402
J-401
J-402
J-403
J-404*
J-405
L-401
Purchasing,
Description
14"0 x 34' Lg. , 250 psig, Refr. Lined,
Water Wall, 88 MM Btu/hr
4 Ea. 5400 1J3, l"0 x 16 BWG., 316 S.S.
Tube, 22' Lg. , 250#, 1300°F, Refr. Lin.
4 Ea. 5400$, l"0 x 16 BWG., ICr-^ Mo Tube,
22' Lg. , 2500, 950°F, C-Mo. Shell
4 Ea. 5000 ij, l"0 x 14 BWG., C.S., 20' Lg. ,
250#, 750°F, C.S. Shell
1 Ea. 630 rf[, 3/4"0 x 14 BWG., C.S. ,10' Lg. ,
"u" Bend, 250#, 450°F, C.S. Shell
4 Ea. 280orfr, 316 S.S. Tubes, 3/4"0x 16 BWG.
"U", 28~S C.S. Shell, 250#
4 Ea. 7835 I{J, C.S. Tubes, 3/4"0 x 14 BWG.
Fix. Hd. , 1-2 Pass, 250#
2 Ea. 32751)3, 316 S.S. Tubes, 3/4"0 x 16 BWG
"U", 24' Lg. 1-4 Pass, 250#
2 Ea. 9'0 x 90'O.S.S w/skirt, 316 S.S. Clad,
250#, 450°F, Ellip. Hd.
2 Ea. 7'3"0 x 57'O.S.S. w/skirt, 316 S.S. Clad,
250#, 450°F, 44' Pack.
2 Ea. 5'0 x 10'O.S.S., 250#, 400°F, C.S. £
w/valves
2 Ea. 14'6"0 x 16'6" O.S.S., Horiz, 316 S.S.
Clad, 250#, 450°F
5'0 x 15' O. S.S., Horiz, w/Stm. Coil, 140,
310°F, C.S. E
3 Ea. 13,200 ACFM, 310°F, 200 — >-242 psia,
Centrif. , 8000 RPM
20,900 CFM, 14.7— »219.7 psia, 2 Stg.
w/Intercool, 8000 RPM
2 Ea. 25 GPM, 215 ps i Rot. Gear, 316 S.S.,
W.C. Bearings
3 Ea. 750 GPM, 210— »262 psia, Aust. I.
Centrif. 1750 RPM
3 Ea. 750 GPM, 210— >-216 psia, Aust. I,
Centrif, 1150 RPM
2 Ea. 1232 GPM, 0 — *250 psig, C.S., 2 Stg.
Centrif, 1750 RPM
3 Ea. 750 GPM, 200 — >310 psia, C.S. , Centrif.
1750 RPM
2 Ea. 14,400 ACFM, 380 °F, 210 psia, 1 Stg.
86-8"0 Tubes
Totals Major Equipment
M + L = 300
M = 60, L = 60
M = .28, L = .22
M = 60, L = 45
M = .80, L = 1.70
M = .60, L = . 60 at 100,000 C.F.
Mp= . 25 Ma j . M. , L = . 60 Mp
Mj= .06 Maj. M. , L = . 40 tal
Totals Minor Equipment
Total M + L =
Superv. , Field, Home Office, Profit t Conting.
Total Investment
Weight Material
If $
353,000 83,800
200,000 368,800
160,000 192,400
148,000 140,000
5.0OO 4,200
72,000 121,200
240,000 138,000
40,000 70,800
269,000 268,000
308,780 207,740
13,600 10,500
170,600 120,200
5,320 2,190
276,000 630,000
180,000 411,000
800 2,540
4,200 12,630
3,000 9,060
8,600 13,400
8,400 14,070
200,000 236,000
3,056,530
63,000
42,000
115,000
1,033,000
24,000
60,000
764,000
92,000
2,193,000
7,086,280
4,313, 720
$11,400,000
Labor Insulation
S $
61,800 500
12,000 3,270
8,000 2,940
8,000 1,360
300 160
4,000
12,000
2,000
22,200 5,800
18,440
900
102,400 2,740
560 320
30,000
20,000
200 40
450
450
600
450
20,000
324,750 17,130
_
42,000
90, 000
774,000
51,000 30,000
60,000
458,000
37,000
1,512,000
Electrical Structural
H. P. It
7,800
32,560
34,240
2,000
300
4,080
7,900
600
5,300
6,000
9,000
7,000
15
120
30
500
300
80 27,940
17,045 128,720
410,000
17,220
Pound's Piling
C. Y 50 T Ea
20 12
17 14
15
2
8
13 8
8
75 30
43 12
20 8
3
106 24
56 12
2
6
6
14
10
17 -
441 120
210
700
*Not shown on Figure 4.
-------
Table A-3
I
to
(->
to
Equipment
Major Heat Exchanger
Heat Exchanger
Heat Exchanger
Circ. Water Coolers
Make-up Water Pumps
Black Water Circ. Pumps J-402
Surge Drums
Venturi Scrubbers
Minor Piling
Foundations
Structural
Electrical
Insulation
Buildings
Piping
1ns t rument a t ion
Mark-up,including Eng'r., Superv
Cost Estimates: Section 600; Wet Gas Scrubbing and Reheat
Number
C-601A
C-601B
C-601C
C-602
J-401
J-402
F-601
L-601
Superv. ,
8
8
8
4
3
3
2
4
M
M
M
M
M
M
M
M
Weight
Description ff
Ea. 12750 '2 , 250#, 1300°F, 316 S.S. Tube
l-l/4"0, Cr-Mo Shell 800,000
Ea. 12750;;, 250#,1100°F, Cr-Mo Tube,
1-1/4"0,C.S. Shell 920,000
Ea. 12750 J, 250#,800°F, C-Si Tube,
1-1/4"0,C.S. Shell 920,000
Ea. 14350 ~, 250#,220°F, C. S. Tube &
Shell, 3/4"0 Tube, Fix. Hd. 480,000
Ea. 700 GPM, 243 psi, 2 Stg. Centrif.,
3500 RPM, C.S. 13,200
Ea. 3180 GPM, 58 psi, 1 stg. Centrif,
1750 RPM, Cr-Stl. 20,100
Ea. 10,000 Gal. 9'0 x 21' O.S.S., Vert.
C.S., 250#, Ellip. Hd. 104,120
Ea. 3975O CFM at 50O°F, 2500, C.S. Drum
w/Cr-Stl. Ajd. Throat 188,000
Total Major Equipment
+ L = 300
= 60, L = 60
= .28, L = .22
= 60, L = 55
= .80, L = 1. 70
= .60, L = . 6O x 10,000 C.F.
p= . 15 Ma j . M, L = . 60 Mp
j= . 02 Maj. M, L = .40 Mj
Total Minor Equipment
Total M + L =
Purchasing, Field, Home Office, Profit &. Cont'g.
Material
S
1,408,000
796,000
568,000
255,600
16, 740
23, 190
43,800
142,400
3,253,730
21,000
22,800
50,400
69,000
72,000
6,000
488,000
65,000
794,200
4,832,430
1,767,570
Labor Insulation Electrical Structural Foundations, Piling
S ij H. P # C. Y. 50 T Ea
40,000 13,720 17,650
48,000 9,160 23,200
48,000 6,860 26,450 138 64
24,000 26
900 450 18
1,800 600 45
5,200 3,300 48
13,200 4,180 47,230
181,100 33,920 1,050 117,830 275 64
70
22,800 380
39,600 180,000
63,000 1,150
153,000 90,000
6,000
293,000
26, OOP
,400
Total Investment
$6,600,000
-------
APPENDIX B
Literature Survey of Liquid-Phase Glaus Reaction
Introduction
The product of the reaction of H2S with S02 in water is called Wackenroder's
solution. Therefore, "Wackenroder Reactor" was chosen as a convenient title
for characterizing the process sulfur recovery scheme. However, the reaction,
2 H2S + S02 v y » 2 H20 + 3 S,
also is equivalent to the Claus reaction which is more widely recognized. In
this study, the above reaction will be called the liquid-phase Claus reaction
in order to include non-aqueous solvents
Literature references in this section refer to the list on pages 215 and 216.
Aqueous Solvents
When H2S and S02 react in an aqueous medium, byproducts such as thiosulfate and
polythionic acids (H2SX06) may occur. Schroeterl1) reviews the relevant
reactions in some detail. Side reactions may be minimized by maintaining the
ratio H2S/S02 at 2.0.
Dupasquier(2) reported that production of polythionates increases with tempera-
ture in the range of 0-25°C. On the other hand, Mathieul3) reports that poly-
thionic acid formation is minimal at about 50°C, and sulfur recovery is at a
maximum. No literature data for high temperatures (l50°C, 300°F) have been
found.
In 1933, Lincoln!4) patented a process in which H2S was contacted with sulfurous
acid to produce sulfur. He claimed that a slight excess of S02 was desirable to
insure complete removal of H2S.
Rosensteim5) proposed use of only a catalytic amount of liquid water on a
packing of insoluble aluminum compounds. The reaction was to be carried out
above the melting point of sulfur. «:
Continental Oil Company!6) patented the use of up to 5$ of an aliphatic alcohol
in water as the reaction medium.
LeahyC7) suggested the use of at least two electrolytes with unlike ions as
additives. Townend and Kelly(8) proposed the addition of 0.5 to 5% aluminum
sulfate plus 1 to 5% sulfuric acid.
In a patent granted to SNPA,(3) the use of an aqueous solution containing 10 to
20 grams sodium chloride and 0.6 to 85 grams magnesium sulfate per liter is
disclosed. Sea water is included as a. solvent. It is claimed that sulfur
removal remains high as the liquid is cycled.
Modell, et al.(9) investigated use of the liquid-phase Claus reaction for
possible use in stack gas clean-up operations. They used water at 30°C and S02
concentrations of 2OO-7OOO ppm. At this writing, results are incomplete.
- 213 -
-------
The Bureau of Mines(1Q) has disclosed work on a process, which makes use of the
liquid-phase Claus reaction, to clean up S02-bearing stack gases. Sulfur
dioxide is first absorbed in a citrate buffer solution. It is then reacted
with a stoichiometric volume of H2S in a stirred vessel at 50-70°C. It is
claimed that the reaction can be brought to completion in one minute.
Stauffer's Aquaclaus Process!11) is similar to the USBM process, but the buffer
material in the Stauffer process has not been disclosed.
Water Plus Organic Solvents
Townsendv 12>13/ teaches the use of an organic solvent containing dissolved
water. The solvent is to contain at least two non-CH atoms, at least one lower
chalcogen such as oxygen or sulfur, and no more than two vicinyl hydroxy
radicals. The patent lists over 40 specific solvents, but glycols are pre-
ferred. Water is claimed to be catalytic for the reaction, but harmful in
large concentrations.
Stolfa, et al.'14/ claim the use of water-saturated hydrocarbons together with
a catalytic amount of free water. Various aromatic compounds, straight chain
paraffins, cyclic paraffins, and olefinic hydrocarbons are listed. A similar
patent(15) discloses the use of a detergent additive to improve the recovera-
bility of the sulfur product from the above solvents.
Shell Oil(16) patented a process which conducts the reaction in a medium consist-
ing of 7O-98$ organic sulfone, 2-20$ water, 0.1-10$ ferrous salt, and a chelating
agent .
Urban and Massey! 17 > 18 j19^ teach the use of from 0.5 to 40$ by volume of water
in any alcohol, any ester, or any carboxamide having the general formula,
s
R — c —NX
R2
Water-Free Solvents
In 1909 Feld and Jahl!20' patented the use of tar oils as a solvent for conducting
the Claus reaction. They indicated that basic organic nitrogen compounds in the
tar were beneficial to the reaction.
A German patent'21/ in 1941 referred to the use of an alkylated aromatic base
such as dimethylaniline .
Renault'22) patented the use of liquid esters of phosphoric acid. A variation
of the process was to add an alkaline agent such as soda, potash, alkali metal
sulfides, an amine, or an ammonia type compound.
Eickmeyer' 23) suggested use of liquid sulfur as an absorption and reaction medium
for the Claus reaction.
Freeport Sulfur Company! 24) patented a process using liquid sulfur as the solvent
with a base nitrogen compound as a catalyst. A major advantage of this process
is that the reaction product is the same as the reaction medium, and therefore no
separation step is required.
- 214 -
-------
I.F.P. (institut Francais du Petrole)I25J teaches the use of a solvent contain-
ing an alkali or alkaline earth salt of an organic monocarboxylic acid, poly-
carboxylic acid or partial esters of them. Polyethylene glycol is the preferred
solvent, but numerous others are listed. The process has been commercialized
in Japam26/ and Canada.(27)
Summary and Conclusions
The literature shows that a large variety of solvents may be used for the liquid-
phase Claus reaction. From the point of view of cost and simplicity, water is
by far the best choice.
Any non-aqueous solvent would have to have a low vapor pressure at 320°F in
order to limit solvent losses. Ideally the solvent should have a high solu-
bility for H2S and S02 and low solubility for H20 and S.
Clearly, experimental work is needed to set a design basis for the sulfur
recovery section. A priori, water is the best solvent, but non-aqueous systems
should not be ignored.
Most of the available literature is in the form of patents. There appears to
be little or no reliable data on process yields or kinetics under our proposed
operating conditions.
References
1. Schroeter, Louis C., Sulfur Dioxide. 91-94, Pergamon Press, 1966.
2. Dupasquier, J., Ann. Fac. Sci., Marseille J31, 155-214 (1951).
3. Mathieu, P.A., U.S. Patent No. 3,595,966 (1971).
4. Lincoln, B.H., U.S. Patent No. 1,901,249 (1933).
5. Rosenstein, L., U.S. Patent No. 1,941,623 (1934).
6. Every, R.L., and Grimsley, R.L., U.S. Patent.No. 3,318,666 (1967).
7. Leahy, M.J., U.S. Patent No. 1,995, 545 (1933).
8. Townend, R.V. and Kelly, D.H., U.S. Patent No. 2,563,437 (1949).
9. Modell, M., Margolis, G., and Meissner, H.P., "The Reduction of S02 by H2S
in Aqueous Solution," report prepared for EPA, 1972.
1O. Rosenbaum, J.B., George, D.R., Crocker, L., Nissen, W.I., May, S.L., and
Beard, H.R., "The Citrate Process for Removing S02 and Recovering Sulfur
from Waste Gases," paper presented at AIME Environmental Quality
Conference, Washington, B.C., June, 1971.
11. Kayford, J.S., Vanbrocklin, L.P., and Kuck, M.A., "Stauffer AQUACLAUS
Process," paper presented at 74th National AIChE Meeting, New Orleans,
March, 1973.
- 215 -
-------
12. Townsend, F.M., U.S. Patent No. 2,881,047 (1959).
13. Anon, Oil and Gas J., 56. 41, 120-124 (1958).
14. Stolfa, F., Gleira, W.K.T., and Urban, P., U.S. Patent No. 2,994,589 (1961).
15. Urban, P., and Gleim, W.K.T., U.S. Patent No. 2,998,304 (1961).
16. Deal, C.H., and Papadopoulas, M.N., U.S. Patent No. 3,363,989 (1968).
17. Urban, P., and Massey, L.G., U.S. Patent No. 3,023,088 (1962).
18. Urban, P., and Massey, L.G., U.S. Patent No. 3,050,370 (1962).
19. Urban, P., and Massey, L.G., U.S. Patent No. 3,099,536 (1963).
20. Feld, W., and Jahl, A., U.S. Patent No. 927,342 (1909).
21. Wohlwill, M., German Patent No. 707,132 (1941) Through C.A. 36:2099-2.
22. Renault, R., U.S. Patent No. 3,441,379 (1969).
23. Eickmeyer, A.G., U.S. Patent No. 2,994,588 (1961).
24. Wiewiorowski, T.K., U.S. Patent No. 3,447,903 (1969).
25. Deschamps, A., and Renault, R., U.S. Patent No. 3,598,529 (1971).
26. Hirai, M., Odello, R., and Shimamura, H., Chem. Eng., 79. 8, 77 and 79 (1972)
27. Bonnifay, P., "The IFF Gas Purification Plant at the Nevis Operators'
Committee Gas Plant," paper presented at the 74th National AIChE Meeting,
New Orleans, March, 1973.
- 216 -
-------
APPENDIX C
Thermodynamic Data
Introduction
Presented here is a body of self-consistent thermodynamic data for use in equili-
brium and heat balance calculations. Some of the data given here were published
previously in our 1971 Annual Report to EPA.v1/ These data are included here
for convenience to the reader.
Previous operations of the continuous unit have shown, that equilibria in many
of the possible reactions are approached closely. Apparently, the reasons for
this are the high temperatures involved and catalysis by the acceptor and char ash.
Thus, because of close approach, equilibrium calculations provide not only a good
check for operation of the continuous unit, but also a reliable guide to predict
performance of an actual plant.
Sources of Data
A. Gaseous Phase
Nearly all of the standard free energy of formation values, heat capacities of
zero pressure, and the standard heats of formation were taken from the JANAF
Thermochemical Tables.(22) The equilibrium constants were calculated from the
free energy values, so the set of equilibrium constants is linked and meshed
together to give a high degree of consistency. Exceptions in the sources of
data were:
1. The heat capacity and heat of formation for S6/K\ were calculated
from the JANAF tables for S8/ \ and the experimental data in the
equilibrium constants measurea experimentally by Preuner and
Schupp.(23)
2. The vapor pressure and heat of vaporization of liquid sulfur
were taken from West.'24)
3. The heat content of steam above liquid water were values given
in the Keenan and Keyes steam tables.
B. Solid Phase
Sources of data involving calcium compounds are primarily experimental equili-
brium measurements which were supported privately by Consol during the 1960's.
Consoll25) published equilibrium constant data in the reactions:
3/4 CaS04 + 1/4 CaS = CaO + S02 (l)
CaO + H2S = CaS + H20 (2)
and CaC03 = CaO + C02 (3)
- 217 -
-------
The results were in agreement and extended the prior experimental measurements
of the above reactions.(26j27,28>29)
The reaction,
CaC03 + H2S = CaS + C02 + H20 (4)
by which gas desulfurizing and acceptor regeneration occurs is the sum of
reactions (2) and (3) above.
For reactions involving solids, we prefer to use experimentally determined values
of the equilibrium constants in order to avoid the errors involved in: l) the use
of the sometimes uncertain or discordant free energy values in the literature,
and 2) the extrapolation of free energy data to temperatures of interest to the
process.
Equilibria in many reactions for which we have no experimental data can be calcu-
lated accurately by summing one or more of the reactions (l), (2), and (3) above,
with gaseous reactions for which the free energies are well-established. For
example, consider the reaction,
CaO + 3/4 S2 = CaS + 1/2 S02
This reaction can be expressed as the sum of the following reactions;
Equilibrium Constant
CaO + H2S = CaS + H20 K12
H20 + 1/4 S2 = H2 + 1/2 S02 (1K3)V2
H2 + 1/2 S2 = H2S K2
CaO + 3/4 S2=CaS + 1/2 S02 K13
Thus, (K12)(K2)
K13 - (K3)1/2
The above equilibrium constants are those identified in Tables C-2 and C-7.
Presentation of Data
Distribution of sulfur vapor species are presented in Table C-l.
Equilibrium constants for gas reactions are shown in Table C-2 and numerical
values of the constants as a function of temperature are in Table C-3. Heat
capacities are in Table C-4 and the mean heat capacities above 60°F, derived
from the Table C-4 data, are in Table C-5. The heat content of steam above
liquid water at 60°F also is given in Table C-5.
Heats of formation and heats of reaction at 25°F are given in Table C-6. Equa-
tions for the vapor pressure of liquid sulfur are shown in Figure C-l, which is
a plot of the equations. An equation for the heat of vaporization of liquid
sulfur also is given in Table C-6.
Equilibria for the pertinent solids reactions involving calcium compounds are
given in Table C-7. Mean heat capacities above 60°F are shown in Table C-8.
Heats of formation and heats of reaction are shown in Table C-9.
Plots of the equilibrium constants versus temperature for reactions involving
CaS are shown in Figures C-2 and C-3.
- 218 -
-------
TAI1I.E C-I
Effect of Temperature and Sulfur Partial Pressure
on Distribution of Sulfur Species
Mol Fraction
S,
.8575
.8472
.8O8O
.7567
.7022
.8249
.eooo
.7473
.6916
.6230
.7566
.7158
.6SO1
.5841
.5072
.6903
.5903
.5204
.441O
.3682
.6079
.5651
.4937
.4133
.34O1
.2623
.1958
.4614
.4146
.34O4
.2612
.1934
.1279
.O801
.3287
.2822
.2117
.1423
.0910
.0499
.0264
.217O
.1751
.1164
.0670
.O369
.0176
.O084
.O016
.0977
.0572
.0288
.01 43
.0063
.0029
.O005
.0055
.0025
.0010
.0004
.OOO1
.OOOO
s.
. 1424
. 1527
.1917
.2426
.2963
.1749
. 1997
.2520
.3068
.3735
.2428
.2831
.3474
.4108
.4819
.3O8O
.4045
.4693
.5374
.5906
.3873
.4274
.4913
.5558
.6O19
.6273
.6163
.5159
.5512
.5955
.6175
.6018
.5380
.4431
.5949
.6O75
.5997
.5448
.4582
.3398
.2368
.597O
.5726
.5012
.3899
.2822
.1817
.1149
.O398
.4625
.3561
.2423
• . 1588
.0946
.O571
.0189
.O87O
.0523
.0294
.0172
.OO55
.OOO7
S,
.OOO1
.0001
.O003
.O007
.0015
.OO02
.0003
.O007
.0016
.0035
.OO06
.OO11
.OO25
.O051
.O109
.0017
.0052
.0103
.0216
.0412
.O048
.0075
.O15O
.O309
.0580
.1104
.1879
.0227
.O342
.0641
.1213
.2O48
.3341
.4768
.0764
.1103
.1886
.3129
.4508
.6103
.7368
.1860
.2523
.3824
.5431
.6809
.8007
.8767
.9586
.4398
.5867
.7289
.8269
.8991
.9400
.9806
.9O75
.9452
.9696
.9624
.9944
.9993
Partial Pressure of
Total Sulfur, ntn Atoi
250° F
4.85 x 10-'
3.54 x 10- J
1.24 x 10-5
.396 x 10-»
. 142 x 10-a
300° F
2.33 x ID"*
1.25 x 10-4
.401 x 10-4
.144 x lO'4
.0482 X 10- 4
4OO°F
3.11 x 10'J
1.40 x 10-'
.461 x 10-'
.171 x 10-'
.0592 x ID"'
5OO°F
2 . 42 X 1O" "
.508 x 10-'
.192 x 10-'
.O680 x 10- *
.0272 x 10- *
600° F
9.81 x 10-J
5.31 x 10-'
2. O2 x 1O-1
.726 x 10-2
.294 x 10-*
.114 x 10-'
.O511 x 10-'
700° F
. 13OO
.O724
.0294
.O115
.O0517
.00234
.O0124
800° F
.1825
.1063
.0472
.0211
.0110
.00601
.O0378
9OO°F
.2764
. 1714
.0859
.04 48
.0271
.O171
.0118
.O0609
10OO° F
.3O71
.1749
. 1O42
.O698
.0475
.0345
.O186
12OO* F
.546
.398
.*87
.216
.120
. O328
mf Sulfur/Mot Total Sulfur
7.714
7.694
7.615
7.511
7.398
7.649
7.599
7.492
7.377
7.232
7.311
7.427
7.29O
7.148
6.971
7.374
7.16O
6.999
6.795
6.572
7.196
7.100
6.927
6.703
6.448
6.O83
5.640
6.832
6.692
6.424
6.037
3.568
4.919
4.253
6.352
6.123
5.669
5.O33
4.378
3.658
3.1O6
3.690
3.341
4.7O3
3.962
3.350
2.832
2.510
2.169
4.436
3.768
3.142
2.721
2.416
2.246
2.078
2.381
2.224
2.124
2.072
2.O22
2.003
- 219 -
-------
TABLE C-2
to
CO
o
Kl CO + 1/2 S2 = COS
K2 H2 + 1/2 S2 = H2S
K3 2H2 + SO2 = 1/2 S2 + 2H20
K4 2CO + SO2 = 1/2 S2 + 2CO2
K6 3S8 = 4S6
K8 S8 = 4S2
K5 2H2S + S02 = 3/2 S2 + 2H2O
K7 H2O + CO = H2 + CO2
K9 CO + 3H2 = CH4 + H2O
Equilibrium Constants
for Gas Reactions
Temperature in ° Kelvin
InK = A +
A
-15.5992
-10.35O3
-3.3749
-4.9853
-4. 8OO8
-1.4871
-15.7562
-14.3343
21.9713
-33.2782
1.99482
-6.21477
-33. 415
BT + CT2 +
B
.O122924
. OOO6332
-.OO27184
-.OO07315
. OO3597O
-.OO2161O
.OO52624
. OO159O7
-.OO3582
. 132324
. OO895O2
. O01689O
.OO30594
D/T
C x 1O6
-8.O1O42
-. 1OO12
1.O1362
. 18766
-2.26137
. 43958
-2.65427
-.35856
3.5055
-63.97O7
-4.6424
-.344O1
-1.03191
D
12,017.2
11, 327. O
1O,O39.6
10,480.5
15,353.9
14,786.0
25, 277. O
25,314.0
-15,135.4
-29, 596. O
-437O.84
5,187.25
28 , 1O4
Temperature
Range . ° K
3OO-90O
9OO-14OO
30O-9OO
9OO-14OO
30O-90O
9OO-14OO
3OO-9OO
90O-14OO
3OO-8OO
3OO-8OO
3OO-9OO
9OO-14OO
9OO-13OO
-------
TABLE. C-3
Numerical Values of Equilibrium Constants for Table C-2
Temperature in °F
oF
2OO
3OO
4OO
5OO
600
to
M 7OO
8OO
9OO
1OOO
12OO
140O
16OO
18OO
2OOO
Kl
1.O5 x 10s
1.66 x 1O7
8.O4 x 1O5
73,9OO
1O,6OO
2,09O
5O5
158
58.1
11.4
3. 18
1. 15
.501
.251
K2
1.55 x
2.76 x
1.57 x
1.6O x
2.49 x
5.27 x
1. 42 x
109
108
107
10 6
105
104
104
4,62O
1,750
353
99
35
15
7.
.6
.9
.5
64
1.8O
1.45
2.26
8.36
5.72
6.15
9.29
1.82
4.39
4.O1
6.35
1.38
3
1
K3
x
X
X
X
X
X
X
X
X
X
.X
X
1015
1014
1012
1010
109
108
107
107
106
1O5
104
104
,900
5.27
8.45
9.35
4.35
5.62
1.55
7.55
5.73
6. 16
1.55
9. 15
9
1
,34O
K4
x 1021
x 1019
x 1016
x 1014
x 1012
x 1011
x 109
x 108
x 107
x 106
x 104
,28O
,420
296
K6
3. 17 x 10~8
3.88 x 1O~7
2.44 x 1O~5
6.62 x 1O~4
9.81 x 1O~3
.0937
.644
3. 43
15. 1
188
--
—
—
—
K8
2.O3 x 1O
2.61 x 1O
5.75 x 1O
-29
-26
-21
1.56 x KT16
8.48 x 1O~13
1.26 x 1O~9
6.3O x KT7
1.25 x 10
.01O8
9.54
~4
K5
,OOO749
.OO19O
.OO914
.O325
.O925
.221
. 461
.856
1. 44
3.22
6.39
1O. 7
16.3
22.9
K7
1.965
1.2O
.819
.603
.471
K9
6.97 x 1O~2
1.569 x 1O~2
1.224 x KT3
1.486 x 10~4
-------
TABLE C-4
Heat Capacities at Zero Pressure
Temperature in "Kelvin
Cp = A + BT + CT2 + DT3 + E/T2
B
C x 106 D x 109
Ss
s6
S2
COS
H2S
SO 2
N2
CO
CO 2
H2
HO
2 (g)
CH4
02
Air
28.51
23.13
8.633
9.678
7.344
5.855
7.098
5.551
5.101
7.219
7.757
1. 3609
6.055
5.8254
.03771
.01852
. 000272
. 005978
.OO 1851
.01534
-.001431
.003185
.015568
-.000674
. OOOOO03
.021486
.0036021
. 002572
-25.25
-3. 137
-.1903
3.364
-11.002
3.490
-.8609
-10.238
.6840
3.219
-5.80317
-1.6265
-.52567
-1.623
2.842
-1.348
2.552
-1.125
.2825
-87,826
-123,658
48,814
-16,003
112,176
36,576
Temp. Range,
°K
300-800
300-1000
300-1400
300-1000
- 222 -
-------
COS
H2S
SO 2
N2
CO
CO 2
H2
H20
CH.
(g)
Air
TABLE C-5
Mean Heat Capacities Above 60° F. Gases
Temperatures in °F
Cp = A + BT + CT2 + DT3
Btu/Lb Mol/°F
36.903
27.940
7.680
9.718
8.065
9.274
6.941
6.925
8.632
6.910
7.974
8.1915
6.9211
6.953
B
C x 106
.006734
. 004678
.001251
. 003033
.000916
. OO2798
. 000037
. 000105
.002958
. 000135
. 000410
.0033201
.O0077623
-.OOOOO3471
-2.598
-.321
-.640
-1.096
.214
-.898
.252
.266
-.844
-.044
.239
. 4754
-.1442
.3839
D x 10s
.126
.177
-.070
.121
-.058
-.070
.109
.022
-.048
-.2312
.012092
-.1108
Temp. Range}
°F
60-1000
60-1300
60-200O
60-1340
Heat content of steam (l atm) above liquid water at 60°F.
Btu/Lb = 1021.08 + .4854OT - 35.83 x 1O~6 T2 + 34.46 x 10~9 T3 220-1000
Btu/Lb = 1039.25 + .41916T + 46.6O x 10~6 T2 1000-1600*
Can be extrapolated accurately to 2000°F.
- 223 -
-------
TABLE C-6
Heats of Formation at 25°C. Gases
cal/g mol
AHf
S8/ . +24,200
(SI •'
+25,580
S2, , +30,840
COS -33,080
H2S -4,880
SO2 -70,960
CO -26,416
C02 -94,052
H2°(g) -57,798
H20, . -68,430
CH4 -17,889
Heats of Reaction at 25°C. Gas Reactions
cal/g mol
Reaction
1 CO + 1/2 S2 = COS -22,080
2 H2 + 1/2 S2 = H2S -20,3OO
3 2H2 + S02 = 1/2 S2 + 2H20, . . -29,220
(Si
4 2CO + S02 = 1/2 S2 + 2CO2 -48,890
5 2H2S + S02 = 3/2 S2 + 2H20, , +11,380
6 3S8 + 4S6 +29,720
8 S8 = 4S2 +99,160
6a 1/6 S6 = 1/2 S2 +11,160
8a 1/8 S8 = 1/2 S2 +12,400
5a 2H2S + SO2 = 1/2 S6 + 2H20, . -22,090
5b 2H2S + S02 = 3/8 S8 + 2H20, < -25,8OO
Heat of Vaporization of Sulfur
Temperature in °F
Btu/Lb = 211.1 - .3302T + 434.6 x 10~6 T2 - 214.4 x 10~9
- 224 -
-------
TABLE C-7
I
to
en
K1O
Kll
K12
KL3
K14
K15
K16
1/4 CaSO4 + CO = 1/4 CaS + CO
CaO + H2S = CaS + H2O
CaO + 3/4 S2 = CaS + 1/2 SO
CaSO
S = CaS + 2 SO
CaCO3 = CaO + CO2
CaCO3 + H2S = CaS + CO
Equilibrium
Constants for
Solids Reactions
Temperature in °Farenheit
In K = A + BT + CT2 + D/T
CaO + SO 2
'aS + CO 2
/2 SO 2
102
>2 + H20
A
71.5076
3.OO4O
17.5646
-7.22273
-6.55O15
-.84O165
16.670O
B
-.03125O
-.OOO751
-.OO92926
. OO2533O
.0118713
.OO74O15
-.OO18535
C x 1O6
5.79655
.O1932
1.7716
-.3688O
-1.86265
-1.28243
. 48O912
D
-67,1O4
4, 232
-1,O7O
15,20O
-12,576
-12,932
-13,977.2
Temperature
Range, °F
15OO-2OOO
15OO-2OOO
13OO-2OOO
15OO-2OOO
15OO-2OOO
all
11OO-2OOO
-------
TABLE C-8
Mean Heat Capacities Above 60°F? Solids
Temperature in "Farenheit
Cp = A + BT + CT2
Btu/lb Mol/T
CaS04
CaS
CaO
MgO
MgO-CaO
MgO- CaC03
Char Carbon
MgO- CaS
A
23.57
10.14
10.98
9.77
20.75
30.64
1.447
19.91
B
.00645
. 00230
.OO071
. 00081
.00152
. 00506
.O029 1
.00311
Btu/lb/°F
C x 106
-.0383
-.502
-.0772
-.0387
-.116
-.193
-.643
-.541
Ash
Coal
.207
.OOO029
-.0034
Heat Content Above 60°F
Btu/lb
-12.2
.1879
.O0025
Temperature
Range. °F
60-2OOO
(30,31)
60-200o(31)
60-800
Also used for impurity content of acceptor.
Including products of pyrolysis.
- 226 -
-------
TABLE C-9
Heats of Formation at 25°C, Solids
cal/g mol
AHf
CaS04 -344,090
CaS -113,550
CaC03 -288,280
CaO -151,900
Heats of Reaction at 25°C, Solids Reactions
cal/g mol
3/4 CaS04 +1/4 CaS = CaO + S02 +63,600
1/4 CaS04 + CO = 1/4 CaS + CO2 -10,002
CaO + H2S = CaS + H20, . -14,570
CaO + 3/4 S2 = CaS + 1/2 S02 -20,260
CaS04 + S2 = CaS + 2 S02 +57,780
CaC03 = CaO + C02 +42,330
CaC03 + H2S = CaS + C02 + H20, . +27,760
CaS + 3/2 02 = CaO + S02 -109,310
CaS + 2 02 = CaS04 -230,540
Heat of Combustion
Btu/lb mol
Char Carbon 178,540-4T* (32)
* T is maximum temperature to which the char has been exposed,
°Farenheit.
-------
FIGURE C-l
m<
o"
-------
ooo
000
1000
1100
12OO
1300 1400
- 229 -
1500
16OO
100
1700
-------
^
I
0
J! U
(JO S
(2 S
> u
o <
Sd
5x£
-,S3
|u
I
LOOOO
1000
100
1500
16OO
17OO
18OO / 1900
- 230 -
-------
APPENDIX D
Modified Sulfur Recovery System
Introduction and Summary
It would be desirable to produce directly in the regeneration of sulfided lime
acceptor a feed suitable in composition to conduct a liquid-phase Claus reaction,
i.e., one containing a ratio of H2S/S02 of 2. Perhaps the most straightforward
method of doing this would be to inject oxygen into the offgases from the Squires
reactor. Air cannot be used since it introduces nitrogen as a diluent which
would accumulate in the recycle system. The system investigated here utilizes
CaS04 as an oxidizing agent or oxygen carrier.
Two systems were evaluated here, namely, the single and two-stage processes.
The first appears to be impractical while the second seems to have merit.
The single-stage operation at 15-20 atmospheres, requires an operating tempera-
ture above 1400°F to achieve the desired result of H2S/S02 ratio = 2. Under
these conditions, however, the total "atomic" percent of sulfur in the product
gas is limited to a maximum value of about 1 to 1.25$. The "optimum" tempera-
ture at this pressure level is about 1600°F. The optimum temperature as well as
the sulfur concentration in the offgas increases with increasing pressure.
A more practical system involves use of a two-stage process where the CaS-CaC03
couple, i.e., the "Squires" reaction is carried out at a lower temperature and
the product gas is oxidized to produce Claus feed via the CaS-CaSO4 couple.
In this case, the temperature and pressure at which the Squires reaction is con-
ducted determine the sulfur concentration in the product gas. The second or
oxidation stage must now be carried out at a higher temperature to achieve the
desired H2S/S02 ratio. The minimum temperature for the second stage increases
with increasing steam pressure and is 1600°F for a steam pressure of 6.4 atmos-
pheres .
Another alternative is to condense steam out of a fraction of the Squires product
gas (at least 1/3.,of the gas) and oxidize this fraction by means of the CaS-CaS04
couple to produce a gas containing a low H2S/S02 ratio. This gas would then be
blended with the untreated gas to produce a Claus feed.
Discussion
Single-stage Operation
The thermodynamic performance is determined by establishment of equilibrium
simultaneously in the reactions below;
C02 + 3/4 CaS04 + 1/4 CaS = CaC03 + S02 Kj. (l)
C02 + 1/4 CaS04 h 3/4 CaS = CaC03 + 1/2 S2 K2 (2)
4/3 H2S + 2/3 S02 = S2 + 4/3 H20 K3 (3)
Establishment of equilibrium in the above reactions automatically insures that
equilibrium will be maintained also in the "Squires" reaction, i.e.,
CaS + H20 + C02 = CaC03 + H2S (4)
- 231 -
-------
The net reaction neglecting the relatively small amount of elemental sulfur
produced is:
3/2 CaS + 1/2 CaS04 + 4/3 H20 + 2 C02 = 2 CaC03 + 4/3 H2S + 2/3 S02 (4a)
It is assumed that part of the CaS in the acceptor will be oxidized to CaS04
by air in a transfer line reactor in which the acceptor is passed from the
sulfur removal step to the acceptor regeneration step. It also is assumed
that a sufficiently high partial pressure of C02 will be maintained to prevent
calcination of the acceptor.
A restraint that is put on the system is that the product be a suitable feed
to a Claus unit, i.e., the ratio PH2S/PSO ~ 2- With this restraint it is easy
to show that the partial pressure of steam is given by the relationship:
3/2
i.e., the steam partial pressure at equilibrium is independent of the applied
C02 pressure. The H2S, S02 and S2 partial pressures are given by:
S02 = KI PC02
P
S2
(PC02)
n — QD — O V "D
H <5 ~
H2b bU2 L.U2
The results of the equilibrium calculations at two total pressure levels of
15 and 20 atmospheres, respectively, are given in Table D-l. It is noted that
the maximum total "atomic" percent sulfur is very low, i.e., 1.1% at 1600°F
and 15 atmospheres and 1.28$ at 1625°F and 20 atmospheres. The equilibrium
sulfur concentration goes up slowly with pressure and still amounts to only
2.05$ at 1710°F and 50 atmospheres. It is concluded that the single-stage
regeneration procedure is impractical.
Two-Stage Process
The flow sheet of the two-stage process is given in Figure D-l. The H2S-C02~H20
mixture leaving the "Squires" reactor at about 1300°F is oxidized at a higher
temperature by CaS04 to generate a Claus feed via the reactions:
1/2 CaS04 + 2 H2S = 1/2 CaS + S2 + 2 H20 K4 (4)
S2 + 4/3 H20 = 4/3 H2S + 2/3 S02 l/Kg (3)
Net Reaction,
1/2 CaS04 + 2/3 H2S = 1/2 CaS + 2/3 S02 + 2/3 H20 K5 = K4/K3 (5)
Conditions must be chosen such that the net product from the CaS04 reactor has
the proper H2S/S02 ratio = 2 for the subsequent liquid-phase Claus reaction.
- 232 -
-------
Temperature, °F
K,
(*L\
\ K2 /
Pressures. Atm.
H2°
S2
Total
Atom, % S
PHoO
SO,
TT
Table D-l
Thermodynamic Equilibria in the
CaS-CaS04-CaC03-C02-H20-H2S-S02-S2
System
Atom, % S
1390
.0017
.0088
3.75
0.14O
0.46
14.4
0.0245
O.O49
0.016
14.93
0.71
0.46
19.4
0.033
0.066
0.029
20.00
0.785
1540
.0039
.0073
5.10
1.45
2.64
12.2
0.0475
O.O95
0.008
15.00
1.06
2.64
17.1
0.067
0.134
0.016
19.95
1.17
1600
.0053
.0068
5.63
3.40
5.02
9.80
0.052
0.104
0.0045
15.00
1.10
5.02
14.7
0.078
0.156
0.010
19.96
1.27
1625
.0059
0.0665
5.85
4.62
6.28
8.55
0.0505
0.1010
0.0035
14.98
1.06
6.28
13.5
0.080
0.160
0.008
20.03
1.28
1710 1710
.0009 y
. 7o T
13.98 f
11 15 f -»
0.50 35.0
0.0045 0.312
0.0089 0.624
0.0000 0.047
14.96 50.3
0.09 2.05
14.45
5.45
0.049
0.097
0.002
20.05
0.75
- 233 -
-------
FIGURE D-l
Modified Sulfur Recovery Section
Flue Gas Out
P.
o
o
o
•H
•P
a
•a
•H
x
O
Air
CaS in from
CaS04
in
Oxidation
Reactor
~ 1600°F
CaS
Spent
Acceptor to
Oxidation
Gas Desulfurization
CaC03 out to
Gas Desulfurization
Fresh
Acceptor
In
Squires
Reactor
~ 1300°F
Liquid-
Phase
Glaus
~ 310°F
Sulfur
Out
Spent
Acceptor
Discard
C02 in
- 234 -
-------
Now, if equilibrium is established in the CaS04 reactor, we have:
2/3
p l = K5 = K4/K;
• £
But, since PH g/pSO = 2> the above relationship can be used to define the
temperature at whicn the reaction must be carried out, consistent with the
partial pressure of steam.
A typical calculation is given in Table D~3 for operation of the "Squires"
reaction at 1300°F, with 80$ approach to equilibrium, and a total pressure of
15 atmospheres.
It is noted that a suitable operating temperature for the oxidation reactor is
about 1600°F. The heat required to raise the temperature to 1600°F is "auto-
matically" supplied by the heat of oxidation, i.e., by reaction (6) below;
1/2 CaS + 02 = 1/2 CaS04 (6)
which combined with reaction (5) gives the overall reaction where 1/3 of the
H2S is oxidized to S02 via reaction (5),
2/3 H2S + 02 = 2/3 S02 + 2/3 H20 AH ^ 161,000 Btu/mol (?)
Another alternative is to condense the steam out of 1/3 or more of the gas and
oxidize it more completely to SO2 while the remainder of the gas is by-passed
around the oxidation reactor.
The use of other oxygen carriers than the CaS-CaS04 couple may be considered
to avoid the cost of an oxygen plant. However, the above couple appears to be
the mo:;t "natural" to use in the present context.
Perhaps a more "natural" application of the system would be to the regeneration
of CaS04 produced in a fluid bed boiler. The cycle in this case would comprise:
l) partial reduction with coal or a reducing gas to produce a CaS-CaS04 mixture,
and 2) the product from step (l) would then be passed in series through the
oxidation and "Squires" reactors of Figure D-l. The CaC03 product leaving the
"Squires" reactor would be returned to the fluid bed boiler.
- 235 -
-------
Table D-2
Equilibria in CaS04 Oxidation
of Squires Product Gas
Temperature. °F 1400 1500 1600
K4 2.16 5.43 12.28
1/K3 0.262 0.212 0.178
K5 = K4/K3 0.565 1.16 2.18
Equilibrium ?„ Q atm
0.85 2.5 6.45
If Pu o2 _ 2.0
Table D-3
Composition of Feed and Product
Squires Reaction Product at 1300°F After CaS04 Oxidation at 16OO°F
Partial Pressure, Partial Pressure,
Atm. Mol jo Atm. Mol #
H20 6.35 42.4 6.453 43.23
C02 8.10 54.2 8.000 53.62
H2S 0.51 3.4 0.297 1.98
S02 ~ 0.149 0.99
S2 — 0.027 0.18
Total 14.96 100.0 14.926 100.00
- 236 -
-------
APPENDIX E
Conversion Factors ~ English to Metric Units
Length
Area
Volume
Mass
Pressure
Temperature
Energy
English System
inch
foot
square foot
gallon
cubic foot
pound
pound per square inch
atmosphere
0 Fahrenheit
Btu
horsepower hour
Metric Equivalent
2.54 centimeter
0.305 meter
0.093 square meter
3.785 liter
28.32 liters
453.6 grams
*
51.70 millimeters Hg
760 millimeters Hg
1.8 ("Celsius) + 32
1055 joules
2.69 x 106 joules
- 237 -
-------
TECHNICAL REPORT DATA
(1'lease read lauruclivns un the reverse before completing!
1. REPORT NO.
EPA-650/2-73-049
4. TITLE ANDSUBTITLE
Production of Clean Fuel Gas from Bituminous Coal
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
December 1973
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
G.Curran, J.Clancey, B. Pasek, M.Pell, et al.
8. PERFORMING ORGANIZATION REPORT NO
^PERFORMING ORGANIZATION NAME AND ADDRESS
Research Division
Consolidation Coal Co., Inc.
Library, PA. 15129
10. PROGRAM ELEMENT NO.
ROAP 21ADD-22
11. CONTRACT/GRANT NO.
EHSD 71-15
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, N.C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
FINAL
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The report describes the initial experimental study of a process for
producing clean low-Btu fuel gas from caking bituminous coals. The process consists
of a sequence of coal-processing steps—coal pretreatment, gasification of the
pretreated material, and combustion of the residual char to provide process heat—
followed by desulfurization of the resulting gas at high temperature and pressure.
A fluidized bed of half-calcined dolomite is used for desulfurization. Partially
sulfided sorbent is regenerated using steam and CO2, and elemental sulfur is
recovered from the regenerator off-gases using a liquid-phase Claus process. Most
process steps have been experimentally investigated, and an economic evaluation
has been conducted. The gas could be utilized to generate electricity with low
environmental impact, employing an advanced power cycle (or a conventional power
boiler).
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Held/Group
ECONOMIC ANALYSIS
07A, 13B
Air Pollution
Coals
Desulfurization
Coal Preparation
Coal Gasification
Coal Gas
Fluid-Bed Processing
Dolomite (Rock)
Regeneration
(Engineering)
Air Pollution Control
Clean Fuel Gas
Claus Process
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report}
UNCLASSIFIED
21. NO. OF PAGES
239
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
- 239 -
------- |