EPA-650/2-74-012
February 1974
Environmental Protection Technology Series
m£: ;>:;:;:;:g;
'
m*
-------
EPA-650/2-74-012
EVALUATION OF THE REGENERATIVE
PRESSURIZED FLUIDIZED-BED
COMBUSTION PROCESS
by
T. E. Johnson, A. G. Sligcr,
P. A. Lcfrancois, and D. 0. Moore
M. W. Kellogg Company
1300 Three Grcenway Plaza East
Houston, Texas 77046
Contract No. CPA 70-68 (Task 9)
ROAP No. 21ADE-10
Program Element No. 1AB013
Task Officer: P.P.Turner
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
February 1974
-------
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
-------
MWKLG-RED-74-1284
EVALUATION OF THE REGENERATIVE PRESSURIZED
FLUIDIZED BED COMBUSTION PROCESS
TASK NO. 9 FINAL REPORT
CONTRACT NO. CPA 70-68
by
THE M.W. KELLOGG COMPANY
RESEARCH & ENGINEERING DEVELOPMENT
HOUSTON, TEXAS
Submitted to
CONTROL SYSTEMS DIVISION
NATIONAL ENVIRONMENTAL RESEARCH CENTER
RESEARCH TRIANGLE PARK, N.C. 27711
Prepared for
OFFICE OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
FEBRUARY, 1974
-------
A en ai
RESEARCH AND ENGINEERING DEVELOPMENT
I KELLOGG |
EVALUATION OF THE REGENERATIVE PRESSURIZED
FLUIDIZED BED COMBUSTION PROCESS
TASK NO. 9 FINAL REPORT
Submitted to
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND MONITORING
CONTROL SYSTEMS DIVISION
CONTRACT NO. CPA 70-68
Approved:
Project Directo:
Manager
Chemical Engineering Development
Director t/
Research and Development
-------
THEM..W. KELLOGG COMPANY
A DIVISION OF PULLMAN INCORPORATED
RltKAMCM ft ENCINCKRINC DKVCLOPMBNT
[UU060I
PAGE HO.
REPORT *Q- RED-74-1284
EVALUATION OF THE REGENERATIVE PRESSURIZED
FLUIDIZED BED COMBUSTION PROCESS
TASK NO. 9 FINAL REPORT
EPA-ORM-CSD CONTRACT NO. CPA 70-68
FEBRUARY, 1974
Staff:
Period Covered:
RDO No.:
Distribution:
T.E. Johnson, P.A. Lefrancois, A.G. Sliger,
D.O. Moore, Jr.
August, 1971 to February, 1974
4092-9
Copy No,
Office of Research & Monitoring, EPA 1-51
L.C. Axelrod 52
C.W. Crady 53
W. Cronkright 54
J.B. Dwyer 55
J.A. Finneran 56
S.E. Handman 57
T.E. Johnson 58
P.A. Lefrancois 59
D.O. Moore, Jr. 60
W.C. Schreiner 61
A.G. Sliger 62
M.J. Wall 63
R.I.D. (4) 64-67
Authors: / ,
-------
ABSTRACT
ESSO Research and Engineering has conducted extensive studies
toward developing a regenerative pressurized fluidized limestone
bed coal combustion process using either limestone or dolomite
as the S0_ acceptor. This process requires a regeneration step
to convert the sulfated acceptor back to lime for recycle to
the combustor.
Results of an in-depth literature search into the regeneration
chemistry of CaSO. and CaSO.-MgO and an evaluation of this
chemistry are presented along with recommendations for potential
processing schemes. Included are process flow sheets, material
balances, process descriptions, and solids handling techniques
for both one-step and two-step regeneration. Areas where additional
experimental data are needed to confirm assumptions are delineated.
-------
TABLE OF CONTENTS
PAGE NO,
I. Introduction 1
II. Summary and Conclusions 2
III. Basis of Evaluation 4
IV. Regeneration Chemistry 6
A. 1-step Process
B. 2-step process
V. Process Descriptions 7
A. One-step regeneration process 7
B. Two-step regeneration process 12
C. Special Equipment for Sulfur Condensation 16
VI. Discussion of Solids Handling 17
A. One-step regeneration - low pressure case 17
B. One-step regeneration - high pressure case 20
C. Two-step regeneration 21
VII. Process Design Notes 23
A. Combustor Design Basis 23
B. Regenerator Calculations 25
C. Fluidized Vessel Calculations 26
1. Minimum Fluidization Velocity
2. Slugging Height
3. Transport Disengaging Height
4. Solids Entrainment
D. H2S Plant 29
E. Sulfur Plant 30
-------
TABLE OF CONTENTS (cont'd.)
PAGE NO.
VIII. Process Appraisals 32
A. Comparison of the Processes 32
1. Utilities
2. Solids Handling and Operability
3. Processing Steps
B. Areas Requiring Further Investigation 34
1. Fly Ash
2. Lime Attrition
3. Kinetics of Regeneration
4. Elemental Sulfur in Regenerator Effluent
5. Trace Elements
IX. References 39
X. Appendices 40
A. Process Flow Sheet - One Step Process (Figure 1) 42
B. Process Flow Sheet - One Step Process - Sulfur Plant
& H2S Plant (Figure 2) 44
C. Stream Summaries - One Step Process (Table 1) 45
D. Process Flow Sheet - Two Step Process (Figure 3) 52
E. Process Flow Sheet - Two Step Process - Sulfur Plant
(Figure 4) 54
F. Stream Summaries - Two Step Process (Table 2) 55
G. Solids Handling Diagrams (Figures 5, 6, 7) 60
H. Combustor Heat Balance (Table 3) 63
I. Equilibrium Yields - One Step Regeneration (Table 4) 64
J. Equilibrium Yields - Two Step Regeneration - 1st Stage 65
(Table 5)
K. Equilibrium Yields - Two Step Regeneration - 2nd Stage 66
(Table 6)
L. Combustor & Regenerator Dimensions & Design Parameters
(Table 7) 67
M. Reduction of CaSO.-Summary of Literature Search
68
-------
LIST OF TABLES
TABLE NO. DESCRIPTION PAGE NO.
1
2
3
4
5
6
7
Stream Summaries - One Step Process
Stream Summaries - Two Step Process
Combustor Heat Balance
Equilibrium Yields for One
ration
Equilibrium Yields for Two
ration - 1st Stage
Equilibrium Yields for Two
ration - 2nd Stage
Combustor and Regenerator
Design Parameters
Step Regene-
Step Regene-
Step Regene-
Dimensions and
45
55
63
64
65
66
67
-------
LIST OF FIGURES
FIGURE NO. DESCRIPTION PAGE NO.
1
2
3
4
5
6
7
Process Flow Sheet - One Step Process
Process Flow Sheet - One Step Process -
Sulfur Plant and H2S Plant
Process Flow Sheet - Two Step Process
Process Flow Sheet - Two Step Process -
Sulfur Plant
Solids Handling - One Step Process - Low
Pressure Regeneration
Solids Handling - One Step Process - High
Pressure Regeneration
Solids Handling - Two Step Process
42
44
52
54
60
61
62
-------
I. Introduction
This report covers work performed under Contract No. CPA
70-68, Environmental Protection Agency, Office of Research and
Monitoring. Task specifications were delineated by ORM in
Task #9 contract specification request and supplemented later by
letter following discussions between ORM and MWK based on results
obtained in the first phase of the work.
The objective of this task was to provide data to supplement
Esso Research and Engineering's work directed toward developing a
continuously regenerable pressurized fluidized limestone bed com-
bustor using either limestone or dolomite as the S0_ acceptor.
The task, in short, specifies that an in-depth literature survey
be conducted into the high pressure ("10 atmospheres) regeneration
chemistry of CaSO. and CaSO.-MgO followed by an evaluation of
this chemistry. In addition, recommendations on potential
processing schemes were requested, as were suggestions for tasks
necessary for further evaluation of these schemes.
The report is based primarily on the Interim Report (January
1, 1971-June 1, 1971) of Esso Research and Engineering Company(3)*.
The Final Report from Esso issued in May, 1972, was not made
available to M.W. Kellogg during the time the work was done.
*References included as Section IX of this report.
-------
II. Summary and Conclusions
An extensive literature search was conducted for data on
the reduction of CaSCK or CaS04«MgO under pressure. Very little
pertinent data were found for this reaction under pressure so
analysis of the system was based on the literature and experience
at atmospheric pressure. Feasibility of several key reactions
have been established at 220 psia but kinetic data apparently
have not been obtained; however, the pressure effect should be
advantageous in most of the major reactions in question. (The
results of the literature search are discussed in detail in the
Appendix, Part M of this report).
Preliminary process designs consisting of flow sheets and heat
and material balances have been prepared for the desulfurization
of coal in a 15 MW fluidized bed boiler using the regenerable
limestone concept. Data necessary for the designs have been
taken from Esso Research and Engineering Company Interim Report
January 1, 1971 - June 1, 1971, along with pertinent information
taken from the literature. The emphasis of this study has been
on investigating the viability of the lime regeneration schemes
by demonstrating the sulfur recovery procedures and the solids
handling techniques which could be used. Thus, it was assumed
th-at the kinetics of the various reactions were feasible. No
optimization of process design was made nor was any detailed
equipment design done.
Two processes embodying different regeneration concepts have
been developed:
• A 1-step regeneration process which operates at high
temperatures and produces a concentrated S02 stream for
further processing to elemental sulfur.
• A 2-step regeneration process which operates at relatively
low temperatures and produces an H2S stream for conversion
to elemental sulfur.
-------
With the exception of the regeneration reactors and the
combustor, it has been found that the processes need to contain
only commercially proven concepts and technology to be workable.
From this standpoint then, the regenerable limestone/fluidized
bed boiler concept provides the potential nucleus for a viable
coal desulfurization process.
-------
III. Basis of Evaluation
The following general overall design bases and parameters
used in the process evaluation were mutually derived by ORM and
MWK:
• The effort is limited to a technical feasibility study only;
no cost estimates are included.
• A complete system is considered, all equipment and processing
steps necessary for the production of elemental sulfur, with
recycle of sulfur plant tail gas, is included.
• The plant size studied will be in the 10-20 MW range.
• Only pressurized operation of the fluid bed combustor is con-
sidered. The pressure used will coincide with Esso's pilot
plant design, viz., 150 psig (10 atmospheres).
• A dolomitic limestone is used as the desulfurizing agent and
about 10% fresh stone is added as makeup, basis stoichiometric
requirements.
• Regeneration is evaluated for both pressurized (150 psig) and
atmospheric operation.
• It is assumed that natural gas is available for the regeneration
thereby eliminating the need for reducing gas production.
• Data needed for the evaluation are taken from an Esso report(3)
supplemented by literature data as required.
-------
Regeneration for the 1-step process at both 40 psia and 152 psia
was considered with the result that the only important difference
found between these two was in the ease of solids handling in the
high pressure case. Only high pressure design was considered for
the 2-step process since chemical equilibrium limitations existed
at atmospheric pressure. Feasible solids handling circuits for
both regeneration processes were developed.
A list of topics requiring further laboratory study for purposes
of eventual process design has been prepared. Most important
among the items requiring further development are:
o Details on the behavior of coal fly ash in the combustor
o Data on attrition of lime in the fluidized beds
o Data on the kinetics of the three regeneration reactions
-------
IV. Regeneration Chemistry
The basic chemical reactions involved in the two regeneration
processes which were considered are based on reduction of CaSO.
4
with CO and H~ at elevated temperatures:
A. 1-step process
CaSO. +
4(s)
CO
H2
(g) *
co;
H2°
(g)
S0
2(g)
Ca°
(s)
(1)
B. 2-step process
CaSO.
4(s)
CaS(g) +
CaCO,
+ 4
CO
H2.
(g)
H,0, . + CO-
2 (g) 2(g)
t CaO, , + C0_.
CaS
(s)
+ 4
co
(g)
CaCO-
(s)
(g)
2(g)
(2)
(3)
(4)
Reaction (1) has been observed to proceed rapidly to equilibrium
at temperatures of 2000°F and 1 atm pressure (1). Below 1800°F
equilibrium favors reduction to CaS via reaction (2) although there
are very little experimental data to back up the feasibility of
this reaction. At a temperature of 1340°F and a pressure of 10
atmospheres, equilibrium predicts only CaS via reaction (2).
There is reason to believe that addition of a catalyst will enhance
the rate of reaction (2) at temperatures below 1500°F(2).
A separate discussion based on the results of a literature sur-
vey of reaction (3) is enclosed in Appendix X, part M, for ready
reference.
-------
V. Process Descriptions
A. One Step Regeneration Process
The one step regeneration flow sheet contains four sections:
• Combustor
• Lime Regenerator
• Sulfur Plant
• H2S Plant
In the combustor, the S0_ formed from combustion of sulfur-
bearing coal is absorbed in a fluidized lime bed via the following
reaction:
Ca°(s) +S°2(g)
°2(g) ?CaS°4(s)
Sulfated lime (i.e., CaSO.) from the combustor is transferred to
the regenerator where it is reduced to lime via the reaction:
CO,
CaS0
4(s)
CO
(g)
Ca°(s) + S02(g) +
(g)
Off gas from the regenerator containing SO^ is cooled, mixed
with a gas rich in H S, and sent to the si
Glaus reaction produces elemental sulfur:
with a gas rich in H S, and sent to the sulfur plant where the
2H2S(g)
S0
2(g)
2H2°(g)
3S
(g)
Some of the sulfur produced in the sulfur plant is then sent
to the H«S plant where the following two reactions are
carried out:
CH4(g) * 4S(g, I CS2(g) + 2H2S(g)
CS2(g) + 2 H2°(g) I C02(g) + 2 H2S(g)
-------
H^S from this step is then mixed with the regenerator off
gas and fed to the sulfur plant as described above.
Figures 1 and 2 are the process flow diagrams illustrating
the one step regeneration process which utilizes either a high
pressure or a low pressure regeneration scheme. The chemistry
and the process flows for the two cases are virtually identical,
and for this reason only one material balance and one flow
sheet are provided. The primary difference in the two cases
is in the handling of solids between the combustor and the
regenerator; this will be discussed in a separate section.
Stream summaries for this process are presented in Table 1.
Pulverized coal dispersed in air is transported to the
bottom of the combustor vessel D-101, above a grid plate
provided to insure even distribution of gas over the vessel cross
section. (Note that coal grinding equipment is not shown.)
Combustion air from compressor J-101 as well as a recycle
gas stream from the sulfur plant tail gas compressor J-201
are introduced below the grid. A more detailed drawing
of the combustor-regenerator circuit is presented later (Figures
5 & 6). Coal combustion occurs in a fluidized bed of lime with
15% excess air at 1700°F and 150 psia. Immersed in the bed and
surrounding the bed are heat exchanger tubes which extract the
heat of combustion by generation of high pressure steam.
The sulfur in the coal is initially oxidized to SO in the
bottom of the bed and then absorbed by the lime to produce
CaSO4. A make-up stream of fresh limestone (CaCO ) is
introduced directly into the combustor bed as is a recycle
stream of regenerated lime. The fresh limestone is required
to maintain the activity of the lime bed. It undergoes
calcining directly upon entering the combustor bed:
CaCO- -»• CaO. . + CO-
3(s) ts)
-------
Combustion gases from the bed exit through a cyclone
where entrained lime particles and large fly ash particles are
separated before the gases reach the turbine and economizer.
Small fly ash particles pass through with the flue gas.
Rock, containing CaO and CaSO. (str. 1)*, is transported
from the combustor bed to the regenerator, D-102, where the
CaS04 is reduced to CaO in a fluidized bed at 2000°F and 40
psia (152 psia in the high pressure case). Reducing gas at
2680°F (str. 6), formed by the partial oxidation of natural
gas with air in combustion chamber B-101, is introduced below
the regenerator grid. Immediately above the grid more air is
added such that the total air added is 80% of that necessary
for the complete combustion of the natural gas. It is necessary
to withhold some of the air from B-101 in order to limit the
temperature rise of the gas below the grid to a level which
is acceptable from a materials of construction standpoint.
Regenerated rock (str. 2) is returned to the combustor after
a small slipstream of rock (str. 3) is removed.
Gas from the reduction reaction (str. 7) containing 7.5%
S02 and 28.6% HO, passes through a cyclone to separate
entrained lime, and then through a cooling train (C-100, 101,
101A, 102) where the gas is cooled from 2000°F to 100°F and
water is condensed. The heat removed from stream 7 is used to
reheat steam from the first stage of the power plant steam tur-
bine in C-100 and generate 1000 psig steam in C-101 and C-101A.
Water is condensed in C-102 and is separated from the gas in
drum F-101. Water is condensed at this point in an attempt
to improve the yield of elemental sulfur in the sulfur plant.
It was felt that this processing scheme would improve the Claus
Plant efficiency. However, no attempt was made to optimize the
* Refers to stream numbers in Table 1
-------
system and it is possible that the Shell method (Shell Report,
April, 1972, Figure 2, which was not available to M.W. Kellogg)
which does not include condensing water before feeding the
Glaus reactors is more economical. After H20 condensation,
a gas from the H S section (str. 30) containing approximately
80% H_S is blended with the regenerator off gas and the
mixture is heated to 169°F in C-103 by exchange with hot gas
(str. 28) from hydrolysis reactor D-302, and then to 425°F
in C-105 by exchange with hot gas from incinerator B-201.
At 425°F the gas is ready for entry into the Glaus reactors
for sulfur production.
In the sulfur plant, the blended gas from the reheat
train passes through two reactors, D-201 & D-202 where
sulfur is formed by the Glaus reaction. Referring to Figure
2, gas from the reheater, C-105, passes through reactor D-201
where heat from the exothermic reaction raises the temperature
to 743°F. The gas from the first reactor is cooled to 375°F
by exchange in reheat exchanger C-201 with cool gas coming
from F-201 and in C-203 via 50 psig steam generation. Sul-
fur is condensed in C-203. Gas from drum F-201 (str. 14)
is reheated to 425°F in C-201 and reacted to produce more
sulfur in D-202. Hot gas from D-202 is cooled to 300°F
in C-203 where sulfur condenses and the resulting liquid-
gas mixture then flows into F-202 where liquid sulfur is
separated and recovered.
Gas from F-202, containing some sulfur vapor as wel] as
unreacted H_S and SO. is sent to a natural gas fired combustion
chamber, B-201. In B-201, H.S and sulfur vapor are oxidized
at 1450°F to S02. This is necessary because the incinerated
gas is eventually fed to combustor D-101 for SO- removal.
10
-------
Hot gas from B-201 (str. 18) is cooled in C-105 by exchange
with sulfur plant feed and in C-204, C-204A and C-205 by
generation of 100 psig steam and exchange with cooling water.
Water is condensed in C-205 to save compressor horsepower and
separated in F-203. From here cooled gas goes to J-201
where it is compressed to 153 psia. The compressed gas (str.
20) is fed back into the combustor D-101 where residual S0_
is absorbed and eventually returned to the sulfur plant.
The net sulfur product from F-201 and F-202 is pumped to
a sulfur storage pit. The rest of the sulfur production
(str. 21) however, is sent to the H S plant. Molten sulfur
(str. 21) is pumped to heater F-301, where the temperature
is raised from 300°F to 500°F in a non-flow, stirred tank
heater. This type heater is necessary in order to avoid
pumping sulfur through a shell and tube exchanger at a
temperature range in which sulfur has peculiar viscosity
characteristics. Liquid sulfur from F-301 is pumped through
B-301 where sulfur is vaporized and heated to 1165°F.
Furnace B-301 also preheats a CH feed stream (str. 23) to
760°F. The preheated CH. and sulfur vapor are mixed and fed
to CS_ reactor D-301 where the following reaction occurs:
CH4(g) + 4 S(g) J CS2(g) + 2 H2S(g)
Heat released by the reaction raises the temperature of the
outlet gas (str. 24) to 1229°F. Outlet gas from D-301 is
cooled to 248°F by exchange in sulfur heater F-301 and C-301
11
-------
and C-302. Condensed sulfur is separated from reactor effluent
gas in F-302. From here unreacted sulfur is recycled to the
reactor (D-301) via F-301 where it is reheated. Separator
overhead gas (str. 25) is reheated in C-301 to 600°F and mixed
with steam (str. 27) prior to entering the hydrolyzer reactor,
D-302. The following reaction occurs in the hydrolyzer:
CS2(g) + 2 H2°(g) I C02(g) + 2 H2S(g)
Since the above reaction is strongly exothermic, it is necessary
to limit the temperature rise of the exit gas (str. 28) to
600°F by generating 150 psig steam in D-302. Product gas from
D-302, containing approximately 11% H-O, is cooled in C-103 by
exchange with sulfur plant feed gas. Water is condensed in
C-104 raising the H2S concentration in the gas from F-102 to
about 80%. From this point the gas (str. 30) is mixed with
regenerator off gas and fed to the sulfur plant.
M.W. Kellogg chose technology currently available for
conversion of sulfur to H_S via first producing CS_. No
attempt was made to find the lowest cost method of producing
the H_S required for the Glaus reaction. It is possible that
H-S may be produced directly from S0_ eliminating the need
to recycle sulfur but this was not investigated.
B. Two Step Regeneration Process
The second process to be described is the two step
regeneration process. This process contains the following
basic sections:
• Combustor
• First stage and second stage lime regenerators
• Amine treating unit
• Sulfur plant
12
-------
The principle of operation of the combustor remains the
same as in the previously discussed one step process and, for
this reason, will not be discussed again. The basic difference
between the two processes is in the regeneration of CaSO.;
two regeneration steps are used in this instance and sulfur
is released as H2S rather than SCL. Only pressurized regenerator
operation was considered because it was found that the chemistry
of the second regeneration step required high pressure. In
the first stage of regeneration sulfated lime from the combustor
is reduced to CaS via the reaction:
CaSO
4(s)
+ 4
CO
H,,
(g)
CaS(s) + 4
co2
H2°
(g)
C02 produced in the above step is recovered in an amine absorber
and is subsequently used as a reactant in the second stage
regenerator where CaS is reacted to CaCO_ by the reaction:
CaS(s) +C02(g) +H2°(g)
+H2S(g)
H2S produced in the second stage is cooled and sent directly
to a conventional sulfur plant where 1/3 of it is oxidized
to S02:
H2S(g)
S0
2(g)
H2°(g)
The SO- produced from the above then takes part in the Claus
reaction to produce elemental sulfur:
2H2S(g)
+ SO
2(g)
2H2°(g)
+ 3S
(g)
Stream summaries for the 2-step process are presented in
Table 2. Referring to Figure 3, sulfated rock from the com-
bustor (str. 1) is transported to the first stage regenerator
where it is contacted with hot reducing gas (str. 7) resulting
from the partial oxidation of CH with 60% of theoretical
air in combustion chamber, B-101. Gas from B-101 must be
13
-------
cooled in order to maintain the first stage regenerator
(D-102) in heat balance at the relatively low temperature
of 1340°F. This is accomplished by generating steam and
preheating boiler feed water in C-101 and C-101A followed
by exchanging with gas (str. 9) from the CO,, compressor,
J-102. Rock from D-102 (str. 2) containing CaS is transported
from the first stage to the second stage regenerator and
exiting gas (str. 8) containing CO- and HO goes to the
amine absorber, D-301.
From the overhead of the first stage regenerator, str.
8 is cooled from 1340°F to 150°F by rebelling amine solution
in C-301 and by exchanging with cooling water in C-303.
Water is condensed in C-303 and is separated in F-301. Dry
gas from F-301 (str. 11) goes to the amine absorber, D-301,
where some of the CO- and essentially all of the trace of H S
contained in the gas is removed by contact with a circulating
amine solution. Rich amine solution at 148°F from the bottom
of the absorber exchanges heat with lean solution from the
bottom of the stripper, D-302, and goes to the top of the
stripper where CO- and the small amount of H_S present are
flashed overhead. Lean solution from the bottom of the stripper
is cooled from 245°F to 110°F by exchange with rich solution
in C-304 and cooling water in C-306. Lean gas leaves the
top of the absorber at about 115°F and goes to the stack.
C02 and traces of H_S from drum F-302 go to CO- compressor
J-102 where the gas is boosted from 20 psia to 154 psia.
Steam is added to the compressed CO- to bring the H_O com-
position up to 50% (str. 9). From J-102 the gas goes to
C-102 where it is heated to 471°F for entry into the second
stage regenerator, D-103. In the second stage regenerator,
the rock which overflows from the first stage at 1340°F is
contacted with the gas from C-102 containing equimolar
amounts of CO- and H20 along with some traces of H-S
which were produced in the first stage. Gas (str. 9) enters
below a grid plate which is provided to give good distribution
over the vessel cross section. Gas (str. 10) exits from the
second stage at 980°F and contains about 16% H S and 42% H_0.
14
-------
From the second stage, rock containing CaO and CaCO_ is
returned to the combustor after a small slip stream of rock
is purged. Tn the combustor, the CaCO3 made in the second
stage undergoes rapid calcining to produce CaO and CO.
'2'
From the second stage, overhead gas goes to a cooling
train (Fig. 4) which condenses H2O prior to entry to the
sulfur plant. Gas is cooled from 980°F to 345°F by generating
steam in C-103 and preheating boiler feed water in C-103A.
Gas is cooled from 345°F to 100°F in C-104 by exchange with
cooling water. Condensed water is separated in F-201 and
the dry gas (str. 14) is reheated to 450°F in C-105 by exchange
with gas from incinerator B-201 (str. 15).
Gas from C-105 at 450°F is fed to a conventional sulfur
plant which is not shown in detail in Figure 4. This
sulfur plant would be similar to the one in Figure 2 with the
exception that a furnace would be included to oxidize 1/3
of the entering H2S to S0_. The reactor and condenser sequence
would be the same as in Figure 2. Molten sulfur from the
condenser flows to a storage pit. Gas from the sulfur plant
final condenser, containing unreacted H_S and SO , goes to
the tail gas incinerator, B-201. In the incinerator which
operates at 1400°F, sulfur present in the tail gas is oxidized
to S02. Hot gas from the incinerator (str. 15) is cooled
from 1400°F to 120°F before going to recycle compressor
J-201 where the gas is boosted to 153 psia and sent to the
combustor for removal of residual SO.. Hot gas (str. 15) is
cooled by exchanging with sulfur plant feed in C-105 and
then generating 100 psig steam and preheating boiler feed
water in C-201 and C-201A. Water is condensed from the gas
by exchange with cooling water in C-202.
As mentioned in the description of the 1-step process,
water is condensed from the Claus plant feed gas in an
attempt to improve efficiency of the sulfur recovery system.
However, again no effort was made to optimize the process
15
-------
and it is possible that it is not necessary to condense
water before feeding the Claus plant.
C. Special Equipment for Sulfur Condensation
A special sequence of processing steps is required to
avoid exchanger plugging in the event elemental sulfur is
present in the off gas from the 2nd stage regenerator in
the case of the 2-step process, or the regenerator of the 1-step
process (see Figures 1, 4). As the processes were originally
described, any traces of sulfur present in the off gas from
the regenerators would be condensed and solidified at 100°F
thereby causing plugging in exchanger tubes. No process
calculations have been made for this section of equipment;
rather a method of handling the problem has been indicated.
All temperatures and flow rates shown on the regenerator
effluent cooling trains of Figs. 1 and 4 are based on the
assumption that no elemental sulfur is present at this point.
The following description applies to both regeneration pro-
cesses .
Gas, at about 400°F containing elemental sulfur, enters the
bottom of a baffled tower. Here gas is contacted with a cold
slurry of solid sulfur and water, thereby cooling the gas
and causing water to condense and sulfur to solidify. Slurry
is pumped from the bottom of the baffle tower, through a
cooler, and back to the top of the tower. A slip stream
of slurry is withdrawn upstream of the cooler and is sent
to a centrifuge where the slurry is concentrated. From the
centrifuge, concentrated slurry goes to a stirred heating
tank maintained at a temperature high enough to cause the
sulfur to liquefy and the remaining water to flash overhead.
Water flashed overhead, containing some sulfur vapor, is
sent to the bottom of the baffle tower. Molten sulfur from
the heating tank is pumped to a pit.
16
-------
VI. Discussion of Solids Handling
In order to provide a realistic appraisal of how a regenerable
limestone process might work in practice, preliminary diagrams
of the combustor/regenerators and their associated solids handling
circuits have been prepared. It is felt that although many
details have not been considered (materials of construction,
gas distributor design, etc.) and considerable experimental work
is still to be performed, the fundamentals of the schemes presented
are sound.
Three regeneration schemes are illustrated: (A) one step
regeneration with pressurized combustor and low pressure regenerator,
(B) one step regeneration with pressurized combustor and high
pressure regenerator, and (C) two step regeneration with pressurized
combustor and pressurized regenerators.
A. One Step Regeneration - low pressure case
Figure 5 is a diagram of how the combustor and regenerator
are envisioned to operate in this case. Pulverized coal is
conveyed to one of two feed hoppers, H-l or H-1A. At all
times one feed hopper is depressurized and is receiving coal from
the conveyor, and one hopper is feeding coal to the combustor
at a controlled rate while under air pressure via a line to the
discharge of the air compressor, J-101. Such a system of alter-
nating pressurized feed hoppers is referred to as a lock hopper
system. Coal is fed to the combustor at a controlled rate through
rotary feeders located in the bottom of H-l and H-1A. Pressurized
air from J-101 picks up the coal from the rotary feeders and trans-
ports it in dilute phase to the upper side of the combustor grid.
Combustion air from J-101 and sulfur plant tail gas from J-201 are
introduced below the grid and pass up through the grid into the
fluidized lime bed (150 psia) where combustion occurs. In all
three cases the combustor vessel was assumed to have a cylindrical
geometry for convenience sake although the eventual shape of
a commercial size unit may be different.
17
-------
Heat exchanger tubes immersed in the fluidized lime bed
extract combustion heat through generation of high pressure
steam. Preheated boiler water is fed into the steam drum which
is a "doughnut" shaped bustle pipe surrounding the combustor.
Water from the steam drum circulates through the tubes by
thermosyphon action and is heated to its boiling point.
Steam from the drum is superheated in a tube section which is
located above the dense bed in the dilute phase. Gases from
coal combustion pass up from the dense bed a distance equal to
the transport disengaging height for entrained solids. At this
point, gases with a residual amount of entrained solids enter
an enlarged section of the combustor which houses a cyclone.
In the cyclone, large rock and coal particles are separated
from the gas. Air is introduced on the upstream side of
a slide valve in the standpipe, thereby keeping the downmoving
solids fluidized at a density of about 40 pounds per cubic
foot (#/cf) and returned to the bed, while small fly ash particles
exit the cyclone with the flue gas. Combustion gases pass from
the cyclone to a turbine (not shown) for power extraction.
Sulfated lime from the fluidized combustor bed is with-
drawn via an aerated standpipe which extends through the grid
and out of the vessel. From the standpipe, solids drop into
the sulfated lime hoppers, H-3 and H-3A. These are lock hoppers
which, in essence, reduce the pressure between the combustor
and the regenerator. Lock hoppers are required for this
service rather than a slide valve alone because the high
pressure drop would result in severe valve erosion. The
method of operation of these hoppers is similar to that of
the coal feed hoppers. When receiving sulfated lime, the
hoppers are vented to the combustor to reduce the pressure
drop across the slide valve. When feeding lime to the
regenerator, the hoppers are vented to the regenerator to
reduce the pressure drop across the rotary feeder. Lime
being fed to the regenerator goes through the rotary feeder
to be picked up by steam and transported in dilute phase to
the regenerator fluidized bed. The density in the transport
18
-------
line is l#/cf.
Reducing gas is introduced into the regenerator (D-102)
below a grid which is provided to distribute the gas over
the vessel cross section. In the regenerator bed, CaSO. is
reduced to CaO at 2000°F. A small amount of rock is withdrawn
from the regenerator, via an aerated standpipe which extends
through the grid to a slurry tank. This prevents a buildup
of rock which would result from the constant addition of
fresh limestone. In the slurry tank a small stream of water
is mixed with the withdrawn rock to make a pumpable mixture
containing 10 wt% solids. Alternatively, the solids could
be withdrawn using a dry system. Gases from the reduction
process pass up the regenerator vessel a distance of 30 ft.
(equal to the transport disengaging height). Gas, containing
residual solids, then passes from the regenerator to an
external cyclone where solids are removed and returned to the
dense bed. An external cyclone is possible in the low
pressure regenerator case due to the relatively low pressure
differential across the cyclone walls.
Regenerated rock is withdrawn from the regenerator via
an aerated standpipe which drops lime into the regenerated
lime hoppers H-2 and H-2A. These are lock hoppers which
serve to build up pressure between the regenerator and
combustor and operate in the manner previously described.
In addition to receiving regenerated lime, the hoppers also
receive a small flow of fresh limestone which helps to main-
tain a constant sorbent activity level. It is likely that
some of the fresh limestone will calcine when mixed with
2000°F lime from the regenerator. Gas from the calcining
reaction would pass through a vent to the atmosphere. From
H-2 and H-2A, regenerated lime is fed to the combustor via
rotary feeders at the bottom of the hoppers. From here air
transports the lime in dilute phase to the combustor bed.
19
-------
B. One Step Regeneration - high pressure case
Figure 6 illustrates the principle of operation of this
case. The essential difference between this case and the
previous low pressure case, as far as solids handling is
concerned, is the elimination of one set of lock hoppers.
Since the combustor is in no way different in this case
from the low pressure case, it will not be discussed again.
Sulfated lime is removed from the combustor, D-101, via
an aerated standpipe which extends through the combustor
grid and the combustor plenum chamber directly below the
grid. Solids from the standpipe, at a density of 40#/cf,
pass through a slide valve which has a pressure drop of 2
psi. Downstream of the valve, solids are transported by
steam at a density of 10#/cf to the regenerator D-102
which operates at 153 psia. Reducing gas from the partial
oxidation of CH4 is introduced below the grid and passes
through the fluidized bed where CaSO, is reduced to CaO at
4
2000°F. From the bed, gas passes up a distance of 29 ft.
(equal to the transport disengaging height) to an internally
mounted cyclone which separates residual solids. An internal
cyclone is used in this case to avoid the high pressure
difference which would result from the cyclone being mounted
externally. A small slipstream of rock is removed from the
regenerator via an aerated standpipe to a slurry tank which
is vented to the regenerator. Water is ar'ded to the slurry
tank to make a 10% slurry. Slurry flows from the tank, through
a valve, to an evaporation pond. Alternatively, the solids
could be withdrawn using a dry system.
Regenerated solids are withdrawn from D-102 via an aerated
standpipe. Solids pass through a slide valve, which takes
a 2 psi pressure drop, and are picked up by air which trans-
ports solids back to the combustor at a density of 10#/cf. A
measured amount of fresh limestone is added to the combustor
via lock hoppers H-2 and H-2A. Fresh limestone passes through
20
-------
rotary feeders and is introduced to the combustor bed via
dilute phase transport with air.
C. Two Step Regeneration
The two step regeneration scheme is diagrammed in Figure 7.
Only pressurized regenerators were considered in the two step
regeneration process and therefore, only one solids handling
diagram is presented. Here, as before, the principle of
operation of the combustor remains the same as was described
in (A), and therefore will not be discussed. Sulfated lime
is withdrawn from the combustor via an aerated standpipe and
transported to the first stage regenerator (D-102) , in the
manner that was discussed for the one step regeneration-
pressurized regenerator case (B). Reducing gas from the
partial oxidation of CH. enters the bottom of the first
stage below a distribution grid and passes through a fluidized
bed operating at 1340°F and 153 psia where CaSO. is reduced
to CaS. Gas then passes through an internally housed cyclone
where entrained solids are separated and returned to the
bed via the cyclone dipleg.
Rock from the bed of the first stage is withdrawn via
an aerated "lateral" to the upper bed of the second stage
regenerator, D-103, which operates at 980°F and 150 psia.
Regeneration gas (containing CO- and H_O) enters the bottom
of D-103 below a grid, and passes up through the bottom of
the first of two beds, then through another distribution grid,
and finally through the top bed. After passing through the
top bed, gas enters an expanded section of D-103 where entrained
solids settle out. No cyclone is provided for the 2nd stage
regenerator since a relatively small diameter (3 ft) expanded
section will allow settling of all but the finest particles.
Solids from the upper bed descend to the lower bed via an
aerated downcomar. Solids from the bottom bed of the second
stage are returned to the combustor via a standpipe which
extends through the bottom grid. Solids pass through a slide
21
-------
valve and are transported to the combustor in dilute phase
by air. A small slipstream of rock is withdrawn from the
bottom bed of D-103, slurried, and sent to an evaporation
pond in a procedure similar to that previously discussed.
Alternatively, the solids could be withdrawn using a dry
system as mentioned in the 1-step process description.
Fresh limestone is added to the combustor via lock hoppers
H-2 and H-2A in the same way as described in Section (B).
22
-------
VII. Process Design Notes
A. Combustor Design Bases
In order to perform a process design, certain basic
assumptions had to be made regarding the size of the boiler
(and thus the size of the lime regeneration and sulfur
recovery plants) and the S0_ absorption capacity of the
lime bed. In doing this, data and information were taken from
Esso Research and Engineering Interim Report, January 1,
1971-June 1, 1971) contract CPA 70-19 (3). Definition regard-
ing the scope of the entire study was taken from M.W. Kellogg
letter No. K-9-3 dated February 8, 1972.
Conforming with the above documents, a boiler size of
15 MW was taken as a basis for calculations in both the
1-step and 2-step regeneration schemes. Using a heat rate
of 10,000 Btu/hr-KW as a first approximatation, a feed rate of
11,539 #/hr of 4.5% sulfur coal was determined assuming a
coal heating value of 13,000 Btu/lb. Using the coal feed rate
initially determined, a heat balance on the boiler was made
assuming 15% excess air for combustion and a flue gas outlet
temperature of 200°F. In this way a required heat rate of
9,607 Btu/hr-KW was calculated. Table 3 shows the results of
the heat balance for the 1-step regeneration process.
Although the 2-step process would have a slightly different
heat balance as a result of the regenerated rock being cooler
than in the 1-step process, and the recycle gas being a
smaller volume, the effect would be negligible in the total
balances, and not worthy of consideration. In addition,
it is advantageous for evaluation purposes to keep the coal
feed rate, and thus the S0_ release rate, the same for the
two process schemes.
23
-------
In both regeneration schemes, 90% removal efficiency of
all sulfur entering the combustor was assumed for design
purposes as this was the standard removal referred to in
the Esso report. This means that when there is a recycle
stream containing SO., such as the sulfur plant tail gas,
the net removal efficiency based on sulfur in the entering
coal is lower than 90%. Thus as a result of the recycle of
sulfur plant tail gas, only 88.6% of the sulfur in the entering
coal is recovered.
In order to determine the required circulation of rock
between the combustor and regenerator, it is necessary to
know the CaO utilization at 90% SO removal defined as:
mols CaSO.
(CaO) = - = - x 100
y mols (CaO
The (CaO) _Q has an inverse relationship to particle size;
as particle size increases more solid is required on a weight
basis to perform a given desulfurization. It was originally
intended to use 20% utilization of CaO; however, this required
a relatively small size lime particle (1000 u) which in turn
would have required a very low superficial velocity to avoid
enormous entrainment rates. As an alternative, the lime
particle size was increased to 2000 p and assumed to have a
size distribution similar to that of lime N-1359 on page
15 of the Esso report (3). The chemical composition of the
lime was taken from page 12 of the above report. Associated
with this, a utilization of 7.6% was used, based on the Esso
report (3) , page 46, Table 6. A limestone makeup of 10%
of the stoichiometric requirement was used. A superficial
gas velocity of 6 ft/sec was chosen for the combustor to
avoid unreasonable entrainment rates.
24
-------
Calculation of the flue gas composition and volume was
made using the Pittsburgh bituminous coal composition given
in the Esso report (3) Pg. 14, Table 2 - ultimate analysis,
adjusted to 4.5% sulfur.
B. Regenerator Calculations
For both the 1-step and 2-step regeneration schemes,
the composition of the rock withdrawn from the bed and the
gas leaving the bed for all regeneration reactions was
determined by assuming that the reactions proceeded to
chemical equilibrium. This was made necessary by complete lack
of kinetic data. In doing this an internal computer program
which predicts the equilibrium composition of reacting
systems was modified and debugged for use with the solid-gas
reactJon systems. This program, which was developed by
Kellogg, solves for equilibrium compositions by minimizing
the total free energy of the reacting system. This circumvents
the necessity of identifying individual reactions in a complex
system and has the advantage of being completely rigorous.
In making equilibrium calcualtions for the 1-step process
a temperature of 2000°F was found to be desirable from the
standpoint of suppressing the CaS formation. Atmospheric
pressure and 150 psia were examined for effect and, at the
ratio of reactants used, no significant difference in
equilibrium composition could be seen between the two pressures.
Table 4 presents the pertinent reaction conditions for the
1-step process and the composition of the reducing gas and
the product gas. Interestingly, past laboratory investigations
at Esso Research and Engineering (1) have shown that the re-
25
-------
duction of CaSO. with CO and H. is limited by equilibrium
at 2000°F in a batch operated fluidized bed.
For the 2 -step process it was found that the pressure
effect was very important in the reaction:
CaS(s) +C02(g) + H2°(g) * CaC°3(s) + H2S(g)
Increasing pressure increases the conversion. For this reason
the 1st stage and 2nd stage regenerator pressures were chosen
to approximate 150 psia, consistent with the combustor pressure.
Tables 5 and 6 present the pertinent reaction conditions
and the feed gas and product gas compositions for the 1st
and 2nd stage regenerators .
C. Fluidized Vessel Calculations
Basically, the fluidized combustors and regenerators
are sized to conform with the superficial velocities mentioned
in the Esso report (3) . Table 7 summarizes the dimensions
and the design parameters which were calculated for all of
the fluidized vessels. Only one combustor is listed since
this vessel is the same size for both regeneration schemes.
The calculation procedures used to generate the data are
discussed below:
1. Minimum Fluidization Velocity - Care was taken to see that
the minimum fluidization velocity was less than one-half
the normal superficial velocity on all vessels, thereby allowing
for a 50% turndown ratio. The value of the minimum fluidization
velocity was determined by the method of Leva (4) :
GM
MF
0.88
u
UMF
26
-------
where ,
o
G = fluid mass velocity for minimum fluidization , Ib/hr-ft
= minimum fluidization velocity, ft/sec.
D = particle diameter, ft.
fififi- = fluid, solids densitv , lb/ft3
L S
u = fluid viscosity, Ib/ft-sec.
2. Slugging Height - The slugging height is defined as the
bed height at which gas bubbles passing up the bed grow to
the size of the containing vessel diameter. Since such
large bubbles can set up vibrational forces, bed heights
were kept below the slugging height. The correlation of
Yagi and Muchi (5) , developed for particle sizes greater
than 100 microns, was used to calculate the slugging height:
Lsh = 1.18 Dt
-------
4. Solj^ds^ Entrainment - Considering 2000p lime particles
alone, there should be no entrainment above the TDH since
the lime particle terminal velocities for pressurized and
atmospheric operations are 19 ft/sec, and 72 ft/sec, compared
to superficial velocities of 6 and 7 ft/sec. Considering,
however, that smaller sized coal and fly ash will be present
in the combustor bed and possibly the regenerator beds (not
to mention attrited lime), it is evident that entrainment
can occur above the TDH. Esso Research and Engineering (3)
was able to calculate some entrainment rates (pg. 78) by
assuming a bed composition of 80% lime, 17% fly ash and
3% coal. With this as a basis, they calculated an entrainment
rate of 5 Ib/min at 50 ft. outage for their 15 in. combustor
at a superficial velocity of 7 ft/sec. The combustor entrainment
rate reported in Table 7 was calculated using the above Esso
number but corrected for a higher pressure. This rate of
entrainment results in a combustor cyclone loading of 669
gr/ft which is acceptable. It was not possible to calculate
entrainment rates for the regenerator vessels although it is
likely that there would be some as a result of some coal
and fly ash being present in the sulfated lime transported
from the combustor to the regenerator(s).
As can be seen from Table 7 the combustor and 2nd stage
regenerator (2-step process) are provided with expanded
upper sections. In the case of the combustor, this was
done in order to make room for the cyclone. In the case
of the 2nd stage regenerator, it was done to reduce the velocity
such that no cyclone would be necessary. Two beds are shown
for the 2nd stage regenerator on Figure 3 as a result of the
low maximum allowable bed height imposed bv slugging limitations.
This was entirely arbitrary and would or would not be necessary
depending on the kinetics of the 2nd stage regeneration
reaction.
28
-------
D. H2S Plant
The production of H_S from molten sulfur and CH. as shown
i 4
on Figure 2 represents no new technology. The following
two reactions form the basis of the process as used in the
1-step regeneration scheme:
CH4(g) + 4S(g) ?CS2(g) + 2 «2S (g)
CS2(g)+ 2H2°(g) 1 C02 (g) +2 H2S (g) (2>
CS_ is produced commercially in large quantities via reaction
(1) and the conditions used for it in this study were taken
from the literature (7). The pertinent reaction conditions
are as follows:
Temperature 1200 °F
Space Velocity 560 (SCFH gas/CF
catalyst)
Catalyst Silica Gel
CH .. Conversion 90%
4
Excess Sulfur 10%
The reaction is slightly exothermic, so that the effluent
temperature is 1229°F.
The hydrolysis step, reaction (2), is also well established
and is used to increase the yield of H2S by hydrolyzing CS-.
The following reactor conditions were taken from the literature
(8) :
Temperature 600°F (outlet)
Contact Time 0.4 sec.
Catalyst Activated Alumina
CS2 conversion 100%
Excess Steam 32.6%
29
-------
Reaction (2) is strongly exothermic (29,500 Btu/lb mol CS_)
and for this reason heat must be removed via steam generation
to prevent an excessive temperature rise.
E. Sulfur Plant
The sulfur plant of the 1-step regeneration process shown
in Figure 2 is a normal sulfur plant with some slight modifi-
cations. A typical sulfur plant feed gas contains sulfur
as H2S only, and for this reason, the plant must contain a
furnace to oxidize 1/3 of the entering H S to SO, as required
^ «
by the Glaus reaction stoichiometry :
2H2S(g) +S°2(g, l2H0 + 3S(g)
Since the sulfur plant feed in the 1-step regeneration process
already contains SO- and H_S in the proper proportions , the
furnace is not necessary.
The sulfur plant for the 1-step process therefore
consists of only the two Claus reactors, two condensers,
and one reheat exchanger. Technical information and proce-
dures for calculating the conversions and temperatures around
the sulfur reactors of Figure 2 were taken from the Pan
American Petroleum Corporation sulfur plant design manual.
Since details of this process are of a proprietary nature,
the calculation procedures cannot be discussed.
The acid gas feed to the sulfur plant of the 2 -step
regeneration process contains only H2S and for this reason
a conventional sulfur plant is applicable. Details are
not shown on Figure 4 since this is a well known process
not requiring special discussion.
30
-------
A sulfur removal efficiency of 90% was assumed for
purposes of material balance calculations.
In both the 1-step and 2-step processes it was assumed
to be desirable to have the entering acid gas as dry as
possible in order to increase the Claus reaction equilibrium
conversion. For this reason, the regenerator off gas was
cooled to 100°F in both cases.
31
-------
VIII. Process Appraisals
A. Comparison of the processes
Since no cost estimating has been done, and no attempt
at process optimization has been made, meaningful compari-
sons between the 1-step and 2-step regeneration processes
are difficult to make. Despite this, some general comments
can be made in surveying the two processes.
1. Utilities - Listed below are the utilities which were
calculated for the two regeneration processes. There was
essentially no difference in utilities between the low
pressure and high pressure cases of the 1-step regeneration
process.
1-step 2-step
Steam generation, Ib/hr 5,394 13,050
Steam consumption, Ib/hr 605 2,647
CTW consumption, gpm
(exclusive of compressors) 241 861
Electrical power consumption, KW
(for compressors only) 7,556 7,461
Natural gas consumption, SCFH
(exclusive of sulfur plants) 10,700 34,400
32
-------
The higher steam generation rate for the 2-step process
results from having to cool the large volume of 1st stage
reducing gas in order to maintain the proper heat balance
in the 1st stage regenerator, D-102. In addition, the con-
ventional sulfur plant generates more steam than the modified
sulfur plant because it has a furnace for oxidation of H S
in the feed gas with an attendant waste heat boiler. The
net steam export, represented by steam generation minus
steam consumption, is about twice as high for the 2-step
process versus the 1-step process. The higher natural gas
consumption for the 2-step process results from the large
amount of reducing gas required in the 1st stage regenerator.
Power consumption for compressors is nearly equal, even though
the 2-step process requires three compressors and the 1-step
process requires two.
2. Solids Handling and Operability - In comparing the low
pressure and high pressure cases of the 1-step regeneration
process it can be seen that the low pressure case requires
one more set of lock hoppers and solids feeders than the
high pressure case. In addition to this, a larger regenerator
and a larger cyclone are required for the low pressure case.
In comparing the 1-step and 2-step regeneration processes,
it should be noted that the lime particles will be cycled
over wider temperature extremes between the combustor and
regenerators in the 2-step process (1700°F to 980°F) than
in the 1-step process (1700°F to 2000°F). This could have
detrimental effects on the maintenance of lime reactivity
in the combustor. In addition to this, the CaCO_ coming
from the 2nd stage regenerator in the 2-step process will
be subjected to flash calcining in the combustor. This
also could have implications with regard to lime reactivity
in the combustor.
33
-------
3. Processing Steps - The advantage of the 2-step process
lies in its ability to remove sulfur as H S, obviating the
need for the H_S generating facility. Balancing this, however,
the 2-step process has a more involved regeneration scheme,
additional equipment in the sulfur plant (furnace) , an amine
system for recovery of C0_, and higher consumption of natural
gas for regeneration.
B. Areas Requiring Further Investigation
During the course of the study it became evident that
certain potential problems regarding the combustor and
regenerator would need to be resolved before a final process
design could be performed. In most instances this information
would need to be obtained through laboratory investigation
on a continuously operating combustor/regenerator. Listed
below is a discussion of the various areas which are considered
to be problematical.
1. Fly Ash - An important topic for investigation concerns
the fate of the fly ash formed from combustion of coal.
Presumably some of the f]y ash initially elutriated from the
bed contains unburned carbon which must be returned to
the bed for burning in order to obtain full heating value
from the coal. With this in mind two possibilities exist;
(1) fly ash will elutriate from the bed, be captured by
the cyclone, and be returned to the bed over and over until
the heating values have been removed from the fly ash and
it is such a size that it exits from the cyclone with the
flue gas and, (2) fly ash recycles from the bed to the
cyclone and back to the bed at such a rate that the cyclone
becomes overloaded and relatively large fly ash particles
containing unburned carbon exit with the flue gas. If the
second situation developed, a high temperature carbon
burnup cell might be needed to cut down the amount of recycle
by insuring that fly ash would be captured only once by
the cyclone, and subsequently would be burned in the cell
34
-------
to a size which would permit it to escape the cyclone on
the next pass.
Another area for concern regards the fly ash dust load
on the turbine blades. Since for maximum power extraction,
the high pressure gas entering the turbine should be hot
(1700°F), it will not be possible to remove all entrained
fly ash by conventional means (electrostatic precipitators
and bag filters require temperatures no higher than approximately
800°F and 300°F respectively). It should be possible to
remove some of the fly ash upstream of the turbine with
high efficiency externally mounted cyclones downstream of
the internally mounted, "rough cut" cyclone. Technology
however, would need to be developed for turbine blades
which could handle gas at 1700°F which still contained some
dust.
Another problem related to fly ash in the 1-step process
regards its possible action in the lime regenerator. Since
the combustor bed will contain a certain steady-state amount
of fly ash, especially in the upper part of the bed, it is
likely that some fly ash will be withdrawn with the sulfated
lime and be sent to the regenerator. Here, at 2000°F, there
might be a possibility of slag formation. It is not known
what the effect of this would be, but it is likely that
proper fluid bed operation would be interfered with. In
the event of slag formation, provisions for removing it would
be required; otherwise regenerator temperatures below the
ash fusion point would be needed.
Another area for concern regarding fly ash in both the
1-step and 2-step processes is that the carbon and hydrogen
remaining on it would be reformed in the regenerator to
form CO and H2; this is an endothermic reaction which could
have a large effect on regenerator heat balance depending
35
-------
on the amount of carbon present and the extent to which
reforming occurred. To counter this heat effect, more air
would be introduced with the same amount of natural gas
indicated on the flow sheet thereby increasing the reducing
gas inlet temperature to the regenerator (1st stage regenerator
in the case of the 2-step process).
2. Lime Attrition - In the process description a method for
purging lime from the system was described in which reactivated
lime was withdrawn from the regenerator (2nd stage regenerator
in the case of the 2-step regeneration process), slurried
with water in a tank, and sent to an evaporation pond. In
actual practice, attrition of lime particles and subsequent
loss of fines through the cyclones may make the purge
unnecessary and indeed, the fresh limestone makeup rate
jnight need to be increased over that indicated to maintain
a constant solids inventory in the system. Data on lime
attrition, therefore, need to be obtained.
3. Kinetics of Regeneration - The most obvious lack of
data is in the area of lime regeneration kinetics. It
has been mentioned that regenerator effluent streams shown
on Figures 1 and 3 were computed by assuming that the reactions
occurred at equilibrium. At the high temperature (2000°F)
of the 1-step regeneration process this may be a good assump-
tion; however, at the lower temperatures of the 2-step process
(1340°F and 980°F) such an assumption could be far from
correct. Therefore, kinetic data on the three reactions
which were used in this study need to be obtained.
The regenerator on Figure 1 and the 1st stage regenerator
on Figure 3 are indicated to be single beds. In actuality
it may be necessary to provide staged beds (as shown on
Figure 3 for the 2nd stage regenerator) in order to obtain
higher conversions. The 2nd stage regenerator is shown as
having two beds, not for kinetic reasons, but because of
the relatively low bed heights which can be used before the
36
-------
onset of slugging.
Of interest regarding the 2-step process is a patent
(2) describing a method of catalyzing the 1st stage regenerator
reaction:
CaSO., . + 4
4 (s)
CO
H2
, . -»- CaS, . + 4
(g) *• (s)
C02
H20
(g)
According to the patent, the above reaction is accelerated
when a small amount (as low as 0.1 vol. %) of gas containing
sulfur as S^, SO?, or a sulfide such as H?S is added to
the system. According to the patent, practice of the invention
at a temperature as low as 1300°F is possible. In the 2-step
regeneration process the patent could be applied by recycling
a small amount of gas containing approximately 28% H2S from
F-201 on Figure 4 through a booster compressor and into the
bed of the 1st stage regenerator D-102 on Figure 3. The
H_S accelerator would remain in the system since after
leaving the regenerator it would be absorbed by amine
solution in D-301. Here again, experimental work would be
required to check the applicability of the patent.
4. Elemental Sulfur in Regenerator Effluent - Calculations
made with the free energy minimization computer program showed
that, theoretically, traces of elemental sulfur could be
present in the effluent gas from the 2nd stage regenerator
of the 2-step process and also the regenerator of the 1-step
process. The reason for concern regarding elemental sulfur
is that when the off gas from the regenerators is cooled
to 100°F for water condensation, any sulfur present in the
gas would condense and solidify thereby causing plugging
problems in the condensers. Laboratory investigations will
be required to determine if elemental sulfur will actually
be present.
37
-------
5. Trace Elements - The capture from coal of trace elements
(such as chlorine) by lime in the combustor could lead to
deactivation of the lime and consequently higher fresh
limestone makeup rates or higher regenerated line recycle
rates. In order to evaluate this possibility, extended
runs should be made on a continuously operated laboratory
combustor/regenerator to check for contaminant buildup.
Using the free energy minimization computer program it
was possible to theoretically investigate the fate of
chlorine in the regenerators assuming that under the conditions
existing in the combustor, the lime would capture the chlorine
as CaCl_. It was found for the 1-step process (at both
high and low pressure) that CaCl2 would not exist in the
regenerator and that chlorine would exit with the regenerator
off gas as HCl. For the 2-step process at 1340°F however,
it was found that both CaCl2 and HCl could exist.
38
-------
IX. References
Bertrand, R.R. et al, Fluid Bed Studies of The Limestone
Based Flue Gas Desulfurization Process, Interim Report,
October 15, 1967 - February 15, 1969, submitted to
National Air Pollution Control Administration under
Contract No. PH 86-67-130.
Smith, Jay Charles, and Reinhardt, James R., Increasing the
Rate of Reaction In Reducing Calcium Sulfate to Calcium
Sulfide, U. S. Patent 3,640,682, (February 8, 1972).
Hammons, G. A., et al, Studies of NOx and SOx Control
Techniques In a Regenerative Limestone Fluidized Bed Coal
Combustion Process, Interim Report January 1, 1971 -
June 1, 1971, submitted to Office of Air Programs under
Contract CPA 70-19 by Esso Research and Engineering Company.
Leva, M., Fluidization, McGraw-Hill Book Company, Inc.
New York, 1959 (pg. 64).
Frantz, J. F., Design for Fluidization, Chemical
Engineering, September 17, 1962.
Zenz, F. A., and Weil, N. A., A Theoretical-Empirical
Approach to the Mechanism of Particle Entrainment from
Fluidized Beds., A.I.Ch.E.J. 4,472 (1958).
Folkins, H. 0., et al., Carbon Disulfide from Natural Gas
and Sulfur, Ind. Eng. Chem., 42, 2202-2207 (1950).
Bacon, R. F. and Boe, E. S., Hydrogen Sulfide Production
from Sulfur and Hydrocarbons, Ind. Eng. Chem., 37, 469-474
(1945).
39
-------
X. Appendices
40
-------
Appendix A
41
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
Appendix B
43
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
Appendix C
ENC AOM a 13-83
»^ 1-step reqen.
LOCATION.
DATt_
•Y
TABLE 1
Stream Summaries
component
CaS04
CaO
CaS
CaCOi
Inert
Total mph
Total lb/hr
MW
Temp °F
Press, psia
Transport Stm.
Transport Air
Str. 1
mph
16.40
199.39
_
_
1700
-------
ENG ADM 3 12-63
FOR
PAGE.
1-step regen.
EST No
LOCATION.
TABLE 1 (Cont.)
Stream Summaries
DATE.
BY
Component
CaS04
CaO
CaS
CaC03
Inert
Total mph
Total Ib/hr
MW
Temp. °F
Press, psia
Transport Stm.
Transoorti Air
Str. 5
mph
-
-
-
1.64
77
(Ho?
U.S)
Ib/hr
—
-
-
164
2
166
(•m )
46
-------
ENC ADM 9 12-A9
FOR-
IKEUOGGI
.-.-^, 1-step regen.
DATI
LOCATION.
BY.
TABLE 1 (Cont.)
Stream Summaries
component
CH4
07
H2
H50
C02
N7
SO2
H2S
CO
CS5
S?
Sfi
S8
Liq. sulfur
Total mph
MW
Total lb/hr
Temp. °F
Press, psia
Str.
6 *
mph
_
_
8.61
42.45
11.48
120.16
-
-
8.61
-
_
_
_
—
L91.31
25.57
4,892
2,680
diii)
* compos
reduci
Str.
7
mph
—
_
0.41
62.77
19.76
120.17
16.35
.02
.31
-
Trace
_
-
—
>19.79
29.21
6,418
2,000
(&)
ition a
ng gas
Str.
8
mph
—
_
0.41
6.51
19.76
120.17
16.35
.02
.31
—
_
-
—
-
163.53
33.06
5,407
100
(J^4)
nd flow
above r
Str.
9
mph
-
_
_
56.26
—
—
-
-
-
—
_
—
—
—
56.26
18
1,011
100
(?44)
rate g
sgenera
Str.
10
mph
0.91
_
0.41
7.95
27.95
120.07
16.35
32.78
0.31
-
_
-
—
-
206.73
33.47
6,920
100
32
ven is
or gri<
Str.
11
ir.ph
0.91
_
0.41
7.95
27.95
120.07
16.35
32.78
0.31
-
_
-
-
-
206.7:
33.47
6,920
425
28
for nel
Str.
12
mph
0.91
_
0.41
34.17
27.95
120.07
3.24
6.56
0.31
-
0.53
3.78
1.95
-
199.88
34.62
6,920
743
28
Str.
13
mph
-
_
_
M ,
-
-
-
-
-
-
_
-
-
37.31
37.31
32
1,194
375
26
.
i
47
-------
ENG ADM 3 12-63
FOR
PAGE-
[KEUOCCl
EST No.
SUBJECT 1-step regen
LOCATION.
DATE-
BY
TABLE 1 (Cont.)
Stream Summaries
Component
CH4
°?
H?
H00
CO,,
N^
so0
tus
CO
cs2
s,
sf
S8
Liq. sulfur
Total mph
MW
Total Ib/hr
Temp . ° F
Press, psia
Str.
14
mph
0.91
—
0.41
34.17
27.95
120.07
3.24
6.56
0.31
—
—
.069
0.20
—
L93.89
29.53
5.726
375
26
Str.
15
mph
0.91
^
0.41
39.42
27.95
120.07
0.61
1.31
0.31
_
0.06
0.47
0 .87
-
192.39
29.76
5.726
491
26
Str.
16
mph
_
_
—
-
.
_
.
_
—
^
^
^
^
9.84
9.84
32
315
300
25
Str.
17
mph
0.91
_
0.41
39.42
.27.95
120.07
0.61
1.31
0 31
_
_
.002
.006
_
191.0
28.33
5.411
300
25
Str.
18
mph
H
1.39
_
60.58
37.98
207.41
1.98
_
_
_
_
_
_
309.34
28.25
8.740
1450
25
Str.
19
mph
_
_
36.49
_
_
_
^
_
_
_
_
_
_
36.49
18
657
120
17
Str.
20
mph
—
1.39
-
24.09
37.98
207.41
1.98
^^
_
_
272.85
29 .62
8.083
120
17/15^
48
-------
EN6 ADM 3 12-63
Fen.
1-step regen.
Bin. NO-
DATE.
L.OCATION.
BY.
TABLE 1 (Cont.)
Stream Summaries
component
CH4
Oo
H2
H20
CO 2
N2
S02
H2S
CO
CS2
89
Sfi
SR
Lia. sulfur
Total mph
MW
Total lb/hr
Temp °F
Press, psia
Str.
21
moh
—
_
-
-
_
_
-
-
_
-
-
-
_
32.77
32.77
32
1,049
300
30
Str.
22
mph
—
—
-
-
—
_
-
-
_
-
-
-
_
40.05
40.05
32
1,282
500
44
Str.
23
mph
9.10
_
-
-
-
_
-
—
_
-
-
-
_
—
9.10
16
146
760
43
Str.
24
mph
0.91
—
-
-
—
_
-
16.38
_
8.19
2.88
0.24
0.01
-
28.61
49.87
1,427
1.229
41
Str.
25
moh
0.91
—
-
-
-
_
-
16.38
_
8.19
-
-
_
-
25.48
46.86
1,194
248
37
Str.
26
mob.
_
-
-
-
-
_
-
-
—
-
-
-
_
7.28
7.28
32
233
248
30
Str.
27
mph
-
—
-
21.45
-
_
-
-
_
-
-
-
—
-
21.45
18
386
300
35
Str.
28
mph
0.91
_
-
5.07 ,
8.19 '
"~ I
1
32.76
_
-
-
-
_
-
46.93
33.67
1,580
600
33
i
49
-------
ENG ADM 3 12-63
FOR
PAGE-
EST NO.
SUBJECT.
1— steo recren,
LOCATION-
DATE.
BY
Str
TABLE 1 (Cont.)
Stream Summaries
Str.
Component
CHd
o,
H,
H00
C00
N,
SO.,
H^S
CO
cs2
S^
se
sn
u
Liq. sulfur
Total mph
MW
Total Ib/hr
Temn. °F
Prpss. nsia
29
mph
—
^
^
1.7T
—
^
^
—
_
_
_
—
_
^
3.73
18
67
100
32
30
mph
0.91
_
^
1 _T4
8.19
^
_
32.76
_
_
—
_
«.
^
43.20
35.02
1,513
100
32
50
-------
Appendix D
51
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
Appendix E
53
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
Appendix F
ENG ADM 3 12-63
FOR
PAGE_
IKEUOCG]
EST NO.
2-step reqen.
LOCATION.
DATE.
BY
TABLE 2
Stream Summaries
Component
CaS04
CaO
CaS
CaCO-i
Inert
Total mph
Total Ib/hr
MW
Temp. °F
Press, psia
Transport Stea
Transport Air
Str. (1)
mph
16.05
195.22
—
_
1700
153
n 8.15
_
Ib/hr
2,185
10,948
—
_
284
13,417
147
-
Str.
mph
_
195.29
15.98
_
1340
150
1.00
—
55
(2)
Ib/hr
_
10,952
1,153
_
284
12,389
18
_
Str. (3)
mph
_
195.29
_
15.97
980
150
1.00
_
Ib/hr
_
10,952
_
1.598
284
12,834
18
_
Str. (4)
mph
_
1.52
_
.12
980
150
1.00
Ib/hr
_
85
_
12
2
99
18
-------
ENG ADM 3 12-63
FOR.
IKCUOGC]
PAGE-
Esr No.
suBjrrr 2-step reaen.
LOCATION-
DATE.
BY
TABLE 2 (Cont.)
Stream Summaries
Component
CaS04
CaO
CaS
CaCO,
Inert
Total mph
Total Ib/hr
MW
Temp . ° F
Press, psia
Transport Stea
Transport Air
Str. (5)
mph
-
193.78
-
15.84
980
150
n
12.10
Ib/hr
-
10,867
-
1,585
281
12,733
351
Str. (6)
mph
-
-
-
1.60
77
150
4.35
56
Ib/hr
-
-
-
160
2
162
126
-------
ENC ADM 3 12-63
FOR
PAGE-
EST NO.
2-step regen.
DATE.
LOCATION.
BY.
Component
H0S
C0?
CO
HjO
H0
SO2
S9
Liq. Sulfur
No
°2
Total mph
Total Ib/hr
MW
Temp . ° F
Press, psia
Str.
(7)
mph
-
24.08
67.42
161.16
86.16
_
M
-
410.27
_
749.08
17,506
23.37
2680
155
TABI
Stre<
Str.
(8)
mph
0.08
71.22
20.27
178.14
69.28
_
_
-
410.25
—
749.25
18,536
24.74
1340
152
,E 2 (Cc
am Summc
Str.
(9)
mph
0.08
55.91
-
55.91
—
_
_
-
—
_
111.90
3,469
31.00
ann
154
57
>nt.)
tries
Str.
(10)
mph
15.86
39.90
0.03
39.93
0.11
0.02
0.05
-
_
_
95.90
3.019
31.48
qpn
149
Str.
(11)
mph
0.08
71.22
20.27
14.43
69.28
_
—
-
410.25
—
585.53
15.593
26.63
ISO
148
Str.
(12)
mph
-
15.31
20.27
4.45
69.28
_
—
-
410.25
_
519.57
12.947
24.92
IIS
144
Str.
(13)
mph
0.08
55.91
-
9.98
_
_
_
-
_
_
65.97
2.643
40-Ofi
1 7n
20
Str.
(14)
mph
15.86
39.90
0.03
0.43
0.11
0 .02
0 .05
-
_
_
56.40
2.308
40 . 92
4^n
35
-------
ENG ADM 3 12-03
FOR.
PAGE.
EST NO..
SUBJECT.
2-step regen,
LOCATION.
DATE-
BY
TABLE 2 (Cont.)
Stream Summaries
Component
H2S
CO,
CO
H20
H,
SO?
s2
Liq. Sulfur
No
o.
Total mph
Total Ib/hr
MW
Temp. °F
Press, psia
Str.
(15)
mph
_
43.72
_
24.07
_
1.60
^
66 70
0.58
136.66
4,344
31.79
1400
25
Str.
(15a)
mph
_
_
H
^
_
_
_
1 4 44
14.44
462
32
_
—
Str.
(16)
mph
_
43.72
..
18.77
1.60
66 70
0.58
131.36
4,249
32.35
120
17/153
58
-------
Appendix G
59
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
Appendix H
TABLE 3
COMBUSTOR HEAT BALANCE
Enthalpy Basis: 75°F
IN
Recycled Rock
MM Btu/HR
2.020
Combustion Heat
0.870
SO. Absorption Heat
TOTAL 153.007
OUT
Flue Gas*
Sulfur Plant Tail Gas 0.117 Sulfated Rock
Heat Losses
TOTAL
6.192
2.290
150.000 Heat to Steam 144.100
0.425
153.007
*After being cooled in energy recovery
63
-------
Appendix I
TABLE 4
Equilibrium Yields - One Step Regeneration
Temp . :
Press. :
Mais (C0+H0)
2000°F
14.7 psia
1.05
Mol CaS04
Reducing Gas
Component Mol %
CO 4.50
H2 4.50
C02 6.00
H20 22.19
N0 62.81
TOTAL 100.00
Product Gas
Component Mol %
H2 0.20
H20 24.38
N2 57.86
S02 7.88
Sj trace
H2S 0.01
CO 0.15
9.52
TOTAL 100.00
64
-------
Appendix J
TABLE 5
Equilibrium Yields - Two Step Regeneration
1st Stage
Temp . :
Press. :
Mols (C0+H0)
Mol CaSO. ~
4
1340°F
150 psia
5.58
Reducing Gas
Component Mol %
CO 9.00
H2 11.50
C02 3.21
H20 21.51
N0 54.78
TOTAL 100.00
Product Gas
Component
CO
H2
C0_
2
H2°
N2
H,S
2
Mol %
2.71
9.25
9.51
23.78
54.75
trace
TOTAL 100.00
65
-------
Appendix K
TABLE 6
Equilibrium Yields - Two Step Regeneration
2nd Stage
Temp.
Press.
980°F
150 psia
Inlet Gas
Component
co
TOTAL
Mol %
50.00
50.00
100.00
Product Gas
Component
Mol %
H.S
2
CO.
2
CO
H20
Ho
2
SO.
2
S2
TOTAL
16.54
41.61
trace
41.64
0.21
trace
trace
100.00
66
-------
Appendix L
TABLE 7
Combustor & Regenerator Dimensions
and Design Parameters
Combustor
1-Step 1-Step
Low Pressure Pressurized
Regenerator Regenerator
2-Step
1st Stage
Regenerator
2-Step
2nd Stage
Regenerator
Bed Temp.: °F
Dilute Phase Pressure: psia
Superficial Velocity: ft/sec
Vessel I.D.i ft.
Expanded Section I.D.: ft.
Transport Disengaging Height: ft.
Entrainment at Cyclone Inlet: Ib/hr
Minimum Fluidization Velocity:
ft/sec
Sluffing Height: ft.
Overall Height: ft.
1700
150
7.0
6.4
7.4
43.5
78,500
2.4
17.0
70
2000
40
6.0
3.7
-
30.0
2.6
9.85
43
2000
153
6.0
1.5
-
29.0
2.4
4.0
39
1340
153
6.0
2.4
-
28.7
2.4
6.3
44
980
150
6.0
0.8
3.0
4.5
2.4
2.1
20
-------
Appendix M
Reduction of CaSO^-Summary of Literature Search
1. General Literature Review
a. Comments on Review of Literature
Since the literature search did not find any directly
pertinent information on the reduction of CaSO. or CaSO.-MgO
under pressure, the literature and experience at atmospheric
pressure (1,2) must be used to analyze the problems which
may occur under pressure.
An Esso patent, U.S. 2,970,893 (6) is completely lacking
in detailed experimental information. Pressures noted
are atmospheric to 1000 psig (claim 3) and the statement
is made that 10 to 200 psig may be used advantageously
in fluidized bed operation. In essence, gypsum is converted
to lime by reduction with methane-steam at 850-1500°F
(no 0 ), preferably 1200-1400°F, using cobalt molybdate
on alumina catalyst and an excess of theoretical methane
and steam. Ratio of CaSO. to catalyst is between 1 to 1
and 10 to 1, and steam to methane (or hydrocarbon)
must exceed the theoretical requirement. Novel (if true)
is the use of solid catalyst with gypsum to lower the
temperature of reaction and the liberation of hydrogen
sulfide rather than sulfur dioxide.
Wheelock's thesis (1) has a good review of the literature
to 1958 on CaSO. decomposition and reduction (pp.21-29).
Carbon monoxide starts to react at 1300°F but the rate
increases rapidly with temperature up to 1650°F (maximum
temperature investigated). The reaction is:
CaSO. + 4CO + CaS + 4CO_
4 2
Using dry and wet hydrogen led to some H S in the exit gas.
One author (Riesenfeld-7) studied the action of dry and wet
methane. At 1470 to 1830°F dry methane quantitatively
reduced CaSO. according to the reaction:
4
CaSO. + CH. •* CaS + CO. + 2H.O
44 22
*Numbers in () refer to references listed at the end of this section,
68
-------
With water present, gypsum was completely converted
into calcium oxide and sulfur dioxide, given sufficient
time, by using a large ratio of water to methane and a
temperature of 2200°F or higher.
Zawadzki (3) found dry H- reduced CaSO. to CaS quanti-
tatively at a temperature range of 1110-1470°F. Above
1470°F it was noted that CaO formed and at 1920°F it
predominated. Sulfur came off as SO. and free S. He
found CO did about the same thing but the rate was not
as rapid. In 1956, Burwell patented a process (U.S. 2,740,691)
for converting finely divided gypsum to lime and hydrogen
sulfide by reduction with hydrocarbon gas and steam at
a temperature range of 1630 to 1830°F. He apparently used
a large excess of steam (8).
The following reaction is of interest because it is
involved in some of the previous observations:
3CaS04 + CaS + 4CaO + 4S02
The literature indicates that this reaction occurs at 1830°F
and higher. The following reaction has been shown to occur:
CaS + HO ->• CaO + H-S
However the equilibrium constants indicate it would not
be practical (1, p. 29).
In reviewing this and other literature, an exact
equation illustrating the reaction of calcium sulfate,
methane and steam has not been found. The following
equation can be written assuming the products shown:
CaSO. + 2CH. + H.O •+ CaO + H_S + 2CO. + 4H.
442 2 22
69
-------
Excess steam would decrease the CO concentration. As
is evident from the equation, 3 volumes of gas are changing
to 7 volumes; this is definitely in the direction where
pressure is working against the reaction. The work of
Wheelock(l) shows that by keeping the concentration of CO
and H_ low, the formation of lime instead of calcium sulfide
predominates at high temperatures.
Depending on temperature, the C02 and H2 may react and
reach water gas equilibrium:
C02 + H2 "* C0 + H2°
The cobalt molybdate catalyst, a hydrocarbon desulfuri-
zation catalyst, is stated in the Esso patent(6) to lower the
temperature for the CaSO., CH4, H_0 reaction. It may be
that the cobalt molybdate catalyst does some steam reforming;
namely, it catalyzes the formation of hydrogen and carbon
monoxide:
CH4 + H20 -* CO + 3H2
The CO and H_ formed then reduce the calcium sulfate at
temperatures below 1500°F to calcium sulfide:
CaS04 + CO + 3H2 -»• CaS + C02 + 3H20
The use of excess steam helps to convert some of the CaS to calcium
oxide and hydrogen sulfide:
CaS + H20 •* CaO + H2S
Although the equilibrium constants show this reaction is
unfavorable, it may occur to some extent by keeping the
pressure of hydrogen sulfide low with excess steam. The
cobalt molybdate catalyst may catalyze the following
reaction as well:
70
-------
CaS + 3H0 + CO •* CaO + H0S + CH.
2 24
The catalyst may aid in the removal of sulfur as hydrogen
sulfide by some mechanism which operates as it does in de-
sulfurization of organic compounds. Pressure would be
favorable here to diminish carbon formation on the catalyst
(coke on catalyst gets hydrogenated to methane).
The memorandum of February 25, 1971 by E.L. Plyler
of EPA indicates Esso RED is using 1900-2000°? in
a fluidized bed combustor to regenerate lime and produce
sulfur dioxide. At this temperature the following reaction
becomes significant:
3CaSO. + CaS ->• 4CaO + 4SO_
4 2
This is a considerable temperature departure from the Esso
patent. Wheelock's equilibrium constants for this reaction
are (p. 33) :
Temperature;°F log.QK AH°;KcaI/gram-mole
1700 -4.67 240
2060 0.96 211
2420 4.95 198
Thus temperatures above 1800°F are needed and the heat duty
is high. Other literature is cited in reference (2). None
of this was done too critically and it adds little to the survey.
If Esso is doing what is indicated by Plyler's memorandum
and is still using methane and steam,then another equation
must be written, namely:
CaSO. + CH. + H.O -* CaO + SO. + CO. + 3H-
442 22^
71
-------
The high temperature also would favor the gas phase reaction:
Plyler's memorandum indicates SO- is the product and not
H_S; thus it is concluded that methane and steam probably
the reactants. The combined reaction is shown below:
CaS04 + CH4 •* CaO + CC>2 + H2S + H20
The effect of pressure is unfavorable since three gaseous
volumes are obtained from one gaseous volume.
b. Pertinent Information in June 14, 1971 Memorandum of
EPA-OAP
On page 3 of this memorandum, reference is made that
Esso has shown high S0_ removal rates from a flue gas by
using a fluidized bed reactor at about 1600°F and a gas
velocity of 6 fps. The use of limestone as the fluidized
bed material and its regenerability have been demonstrated.
On page 4, a two-stage FBC unit burned a high sulfur residual
oil (Esso Res., Ltd.). Continuous regeneration of limestone
and recovery of sulfur values is part of this Esso concept.
These British Esso workers mentioned five problems:
1. Metal erosion in the fluid bed
2. Carbon loss by solids carryover
3. Distribution of coal and air uniformly
4. "Turndown" capability of bed
5. Control of steam temperature and pressure
over range of boiler conditions.
Effort by GAP is concentrating on regeneration of limestone
sorbent beds for combustors and gasifiers , both atmospheric
and pressurized, for coal and oil.
72
-------
Esso RED has evaluated the combustion of finely ground
coal in a fluidized bed of coarse limestone. They explored
the regenerability of the limestone. They are seeking a
method of reducing NO and SO emissions by interaction between
X X
NO and SO . They are designing a 12" ID continuous pressurized
X X
FBC using regenerable limestone.
As Zielke et al (4) point out, MgO in the dolomite
derived product has no ability for S0_ absorption due to
equilibrium limitations. After regeneration the most probable
form will be the oxide.
What will be the composition of the sulfated limestone
or dolomite from the combustor-absorber? Zielke (4) shows
that in 7 cycles the dolomite went from 77% to 28% efficiency
of sulfation, i.e. (CaSO. x 100)/(CaO + CaSO.). They used
a regeneration temperature of 19508F which is high enough
to cause sintering which may account for the loss of efficiency
during cycling.
If Esso intends to employ cobalt molybdate de-
sulfurization catalyst and allows much lower regeneration
temperatures as indicated by the patent, then the efficiency
could be maintained at a much higher level. Economic evaluation
should be brought in here to determine the justification for
use of the catalyst. Other difficulties are apparent as well.
Since the SO. acceptor is coarse, what size should the catalyst
be? Work is necessary to allow adequate separation of catalyst
from the lime-containing product. The Esso patent requires carbon
burnoff from the cobalt molybdate catalyst. In addition,
at 1400°F the cobalt in the desulfurization catalyst may
react irreversibly with the alumina to form a cobalt
aluminate which could be inactive; thus the catalyst could
lose activity as well.
73
-------
c. Problems in Regeneration of Limestone
Many assumptions will be made in the following in view
of the lack of knowledge and data on the pressurized regeneration
of the sulfated limestone. First, it is assumed that a super-
ficial gas velocity of about 5-10 fps will be used; this will
fix the limestone size on the coarse side. Wheelock(l) used
7/8 mesh and Zielke et al (4) used a dolomite of 14/35 Tyler
mesh.
The limestone or dolomite to use can be answered in
part by the Consolidation Coal Company which has done a large
amount of evaluation of these substances for use in their
coal gasification process. The fact that they used one of
the best attrition resistant dolomites in their work (4),
namely a Tymochtee dolomite from Western Ohio, is a good
start. A program on easily obtained limestones in plentiful
supply may involve a study of strength and attrition rates
before activity evaluation. No doubt many of these studies
have been done already.
The use of the word limestone means calcium carbonate and
dolomite means calcium magnesium carbonate. However, the
experimental conditions of temperature and carbon dioxide
pressure will determine whether the limestone remains as
such or is converted to lime in use and whether the dolomite
is converted to CaO'MgO.
The obvious advantages of pressure to reduce equipment
size, etc., may outweigh any negative mass action effect
and may even improve kinetics through concentration effects,
compositional changes (CaO to CaCO , etc.) and a change in
mechanism.
Attrition rates of catalyst and limestone must be low
enough to be economical.
74
-------
In summary, lack of exact knowledge concerning the method
of regeneration has limited this discussion. It does appear
that high temperature atmospheric processes (above 1800°F)
give good rates of reaction and have established feasibility.
Pressure may have little effect except for some of the reasons
noted. Below 1800°F, e.g., 1200 to 1500°F, if a catalyst
such as cobalt molybdate is used, it is important that it
maintain its activity. Much laboratory scale experimentation
can be done to answer many of these questions. In this latter
temperature region, the use of coarse sulfated lime may be
a deterrent to reasonable efficiencies as well as reasonable
absorption rates.
2. Specific Literature Review
Since A.M. Squires is well known for his work in this
area the recent literature has been searched for his pub-
lications (5). Not all papers were obtained, but probably a
sufficient number have been found to complete this investi-
gation. The CCNYU library was contacted concern-
ing the reference of Narayanan. They stated they had noth-
ing as he was not in the graduating class of 1970 or 1971.
However, in a recent publication (JAPCA for May 1971), his
name appears in the acknowledgement as having performed
experimental work for Squires and he may not have graduated
as yet. In summation of this review, no direct experimental
work is available to define the kinetics for pressure reduc-
tion of calcium sulfate.
In reviewing Squires' work from 1966 to now, a pattern
appears: (1) the early work was concerned with utilization
of fully calcined dolomite to absorb H S, and then the main
theme of Squires was used to liberate H_S by reaction of the
formed CaS with steam and C02 followed by calcination of the
CaCO_'MgO to CaO-MgO for reuse; (2) the recent trend is
to absorb S0_ with half calcined dolomite, (i.e., CaCO *MgO),
reduce the CaSO. to CaS with H_ + CO from a steam methane
75
-------
reformer, liberate H S as before and return the half calcined
dolomite for reuse. The early publications assumed fuel
desulfurization under gasification-type conditions where
H_S would be the sulfur compound available for reaction with
the calcined dolomite. The later publications recognize that
S0_ may be liberated during combustion or partial combustion
and a panel bed filter with half-calcined dolomite is proposed
for S02 scrubbing of the flue gas.
The 1967 paper (Paper No. 1) by Squires presents the
original concept with relevant thermodynamics and chemistry
and only a minor amount of feasibility type experimental
data. It is a lengthy paper in which he proposes the use
of calcined dolomite to abstract sulfur from the fuel (9):
H2S + CaO-MgO •* CaS-MgO + HO (1)
Then Squires' main theme, the liberation of H S by steam
plus carbon dioxide follows:
CaS-MgO + H20 + C(>2 -»• CaCO -MgO + H2S (2)
The regeneration of the calcined dolomite is the next reaction:
CaCO.-MgO + heat -»• CaO-MgO + C02 (3)
Feasibility of reaction (2) was established at 1000°F and
220 psia. An inlet gas with a composition of 82% CO., 9%
H2, and 9% CO was blended with steam to give a C0_/H_0 ratio
of 1.75 which yielded a dry gas of 20-24% H S (about equilibrium)
Feasibility for reactions (1) and (3) was also established.
Squires did state kinetics are needed for process development.
Squires, soon after the above papers were given (September,
1966), became a professor at CCNYU where he has been working
to obtain the kinetic data and to further develop this and
other concepts. Although a copy of the series paper (Abstract
76
-------
labeled No.l) was not obtained, the second paper (No. 2) includes
additional kinetic data on the CaO plus H.S reaction (1).
Actually, the use of half-calcined dolomite as well as fully
calcined is now considered.
CaO-MgO + H.S ->• CaS-MgO + H_0 (1)
CaC03-MgO + H2S -» CaS -MgO + H2
-------
In the paper labeled No. 4, only the first page was copied
as it mainly is a mathematics paper and not extremely pertinent.
However, it does indicate Squires' new concern with scrubbing
the flue gases for removal of fly ash and S02 and the use of
panel filters which is also mentioned in Abstract No. 2. The
last paper (labeled No. 5) contains most of the experimental
information generated at CCNYU (Note that Squires is now
Chairman of the Chemical Engineering Department at CCNYU) .
However, this paper is concerned with scrubbing stack gas
from power stations.
The stack gas process scheme considers converting dolomite
at 1300 °F with 25% CO2 in N2 to half-calcined dolomite,
CaCO. -MgO, which is used in a panel bed filter at about
1100°F to absorb S0_ from the flue gas and yields CaSO. -MgO (10)
The spent sulfated dolomite is reduced with H_+CO (ex CH4-H_0)
at about 1400°F with 25% HO, 25% H2 , 25% CO, 25% C02 to
produce CaS-MgO and unconverted CaCO_-MgO. Treatment with
H_0+C02 at 1200 °F under pressure springs out the H_S and
regenerates the CaCO.-MgO for reuse.
The results of J.R. Coke (11) were confirmed by Esso and by
Squires on the reactivity of half calcined dolomite with S02
at 1100°F, i.e. :
CaCO_-MgO + S02 + 1/202 -»• CaSO -MgO + CO2 .
Coke, in fact, using a fixed bed of usually 1/8 x 1/4 inch
dolomite did a fair job of investigating the variables and
arrived at a kinetic expression. Squires found the kinetic
expression did not satisfy his smaller size dolomite, 10/28
mesh. Coke's results usually showed about 50% conversion
whereas Squires obtained 50 to 90% conversion depending on
size. However, in a panel bed filter Squires assumes 8/40
mesh stone and 50% conversion for calculation. Thus 100 Ibs
of CaCO.-MgO could take up 22.8 Ibs S02 (comparable to Bureau
of Mines alkalized alumina) . Squires also indicates space
78
-------
velocities in the panel bed filter which seem quite low.
Lack of data on reduction of the sulfated limestone
(CaSO.-MgO) still is evident. Thus, nothing further can be
added to what has been said before. The concepts of Squires
seem to be quite good, but supporting process information is
lacking. The M.W. Kellogg molten salt work substantiates
the liberation of H2S by H2
-------
(9) Squires, A.M., Graff, R.A., Pell, M., "Desulfurization of
Fuels with Calcined Dolomite; I. Introduction and First
Kinetic Results", CEP Symposium Series, No.115, Vol. 67,
pp. 23-34.
(10) Squires, A.M., Graff, R.A., "Panel Bed Filters for
Simultaneous Removal of Fly Ash and Sulfur Dioxide; II.
Reaction of Sulfur Dioxide with Half-Calcined Dolomite",
Journal of Air Pollution Control Association, Vol, 21
(5), pp. 272-276, May 1971.
(11) Cokes, J.R., "The Removal of Sulfur Oxides from Waste
Gases by a Dry Method", Doctoral Thesis, University
of Sheffield, England, May, 1960.
80
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO
EPA-650/2-74-012
2.
3 RECIPIENT'S ACCESSION-NO.
4 TITLE AND SUBTITLE
Evaluation of the Regenerative Pressurized
Fluidized-Bed Combustion Process
5 REPORT DATE
February 1974
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
T. E. Johnson, A. G. Sli ger,
P. A. Lefrancois , and D. O. Moore
8. PERFORMING ORGANIZATION REPORT NO
9 PERFORMING ORGANIZATION NAME AND ADDRESS
M. W. Kellogg Company
1300 Three Greemvay Plaza East
Houston, Texas 77046
10. PROGRAM ELEMENT NO.
1AB013, ROAP 21ADE-10
11 CONTRACT/GRANT NO.
CPA 70-68 (Task 9)
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. ABSTRACTTne report gives results of an in-depth literature search into the regener-
ation chemistry of CaSO4 and CaSO4- MgO. It presents an evaluation of this chemis-
try and recommendations for potential processing schemes. Included are process
flow sheets, material balances, process descriptions, and solids handling techniques
for one- and two-step regeneration. It delineates areas where additional experimen-
tal data are needed to confirm assumptions. Esso Research and Engineering has
conducted extensive studies toward developing a regenerative pressurized fluidized
limestone bed coal combustion process using either limestone or dolomite as the
SO2 acceptor. This process requires a regeneration step to convert the sulfated
acceptor back to lime for recycle to the combustor.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c COSATI I icId/Group
Air Pollution
Reviewing
Evaluation
Coal
Combustion
Desulfurization
Fluidized-Bed
Regeneration (Engineer-
ing)
Pressurizing
Limestone
Dolomite (Rock)
ProcessindAir Pollution Control
Stationary Sources
Regenerative, Press-
urized Fluidized-Bed
Combustion
13B
17 B
14A
18 DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
21 NO OF PAGES
90
20 SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (9-73)
81
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to mam title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a dale indicating at least month and year Indicate the basis on which it was selected (eg , date of issue, date of
approve!, dale of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R Doe. J. Robert Doe, etc ) List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code List no more than two levels of an organizational hirearchy
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with. Translation of. Presented at conference of.
To be published in. Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSAT1 FIELD GROUP • Field and group assignments are to be taken from the 1965 COSATI Subject Category List Since the ma-
jority of documents are multidisciplmary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary postmg(s)
18. DISTRIBUTION STATEMENT
Denote releasabtuty to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price. .
19. & 20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |