U.S. Environmental Protection Agency Industrial Environmental Research
Office of Research and Development Laboratory
Research Triangle Park, North Carolina 27711
EPA-600/7-76-030
October 1976
FEASIBILITY OF PRODUCING
ELEMENTAL SULFUR FROM
MAGNESIUM SULFITE
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping WE.S consciously planned to foster technology
transfer and a maximum interface in related fields. The seven series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program. These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems. The goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
REVIEW NOTICE
This report has been reviewed by the participating Federal
Agencies, and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Government, nor does mention of trade names
or commercial products constitute endorsement or recommen-
dation for use.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
EPA-600/7-76-030
October 1976
FEASIBILITY OF PRODUCING
ELEMENTAL SULFUR
FROM MAGNESIUM SULFITE
by
Philips. Lowell, W.E. Corbett,
G.D. Brown, and K.A. Wilde
Radian Corporation
8500 Shoal Creek Boulevard
Austin, Texas 78766
Contract No. 68-02-1319, Task 31
Program Element No. EHB528
EPA Task Officer: Charles J. Chatlynne
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
A study was made to extend the potential applications
of MgO flue gas desulfurization processes by allowing the
sulfur to be recovered as elemental sulfur as well as sulfuric
acid. The particular question addressed in this study was the
feasibility of combining the exothermic SOa reduction reaction
with the endothermic MgS03 calcination.
Preliminary consideration of the reductants carbon
monoxide, hydrogen, methane and hydrogen sulfide showed that
the reaction with SOz can supply part, or in some cases all,
of the heat of decomposition of MgS03. Two cases were consider-
ed in detail: (1) A low temperature catalytic decomposition
using a commercially available low Btu syngas reductant mixture
and (2) A high temperature noncatalytic decomposition using
a medium Btu reducing gas from an oxygen-blown gasifier.
Complete heat and material balances for conceptual
process designs for the. above cases were made, in order to
identify problem areas. Recommendations for work required
to continue development of the process are given. Problem
areas identified include catalyst physical stability, catalyst/
MgO separation, dust carry-over, and noncatalytic SOa reduction
kinetics.
iii
-------
TABLE OF CONTENTS
Page
ABSTRACT ii:L
1.0 INTRODUCTION 1
1.1 State-of-the-Art 2
1.1.1 Mag-Ox Process 2
1.1.2 Elemental Sulfur Production from S02 8
1.2 Subtask Description 9
2. 0 TECHNICAL BACKGROUND 14
2.1 Process Chemistry 14
2.1.1 MgS03 Decomposition 14
2.1.2 Chemistry of Reduction of S02 to Elemental
Sulfur 15
2.1.2.1 S02 Reduction by Methane 16
2.1.2.2 Reduction of Sulfur Dioxide by Carbon Monoxide-•• 17
2.1.2.3 Reduction of Sulfur Dioxide with Hydrogen 17
2.1.2.4 Reduction of Sulfur Dioxide with CO + H2 18
2.1.2.5 Sulfur Dioxide Reduction by Coal 18
2.1.2.6 Sulfur Dioxide Reduction by Carbon 18
2.1.2.7 Sulfur'Dioxide Reduction by H2S 19
2.1.3 Other Potentially Important Reactions 19
2.1.3.1 Conversion of H2S to Elemental Sulfur 19
2.1.3.2 COS/CS2 Formation 20
2.2 Thermodynamic Screening 20
2.2.1 Selection of Cases 20
2.2.2 Results of Thermo Screening Calculations 23
2.3 Process Implications 30
2.4 Equipment Considerations 35
3.0 PROCESS DESCRIPTIONS 39
3.1 Catalytic Process Description 39
3.1.1 Catalytic Flow Sheet 41
IV
-------
TABLE OF CONTENTS (cont.) Page
3.1.2 Process Engineering Calculations 42
3.1.3 Catalytic Fuel Requirements 46
3. 2 Noncatalytic Process 48
3.2.1 Noncatalytic Flow Sheets 48
3.2.2 Noncatalytic Process Design 50
3.2.3 Noncatalytic Fuel Requirements 50
3.2.4 Noncatalytic-Air Blown Gasifier Case 52
4.0 RESULTS AND CONCLUSIONS ; 54
4.1 Summary of Results 54
4. 2 Conclus ions 59
5 . 0 RECOMMENDATIONS . 60
6 . 0 REFERENCES 63
APPENDIX A - Technical Note 200-045-31-Ola,
"Literature Survey on the Recovery
of Elemental Sulfur from Magnesium
Sulfite" 65
APPENDIX B - Technical Note 200-045-31-02,
"Thermodynamic Screening to Deter-
mine the Feasibility of Producing
Elemental Sulfur from Magnesium
Sulfite" . 180
v
-------
1.0 INTRODUCTION
The magnesium oxide (Mag-Ox) scrubbing process is a re-
generable flue gas desulfurization technique which has been demon-
strated on two major commercial- scale test facilities in this country.
The chemical basis of this process is the reaction of S02 with MgO
to form MgSO 3. The MgSO 3 crystals are calcined to regenerate MgO.
This produces an SOa -rich gas stream which is used as feed to a
sulfuric acid plant. That 1*250^ production is the only available
process option for treating the SO 2 rich gas produced by MgS03
calcination limits the potential for future application of the
MgO process .
An option other than HzSOn. production for the S02 is its
reduction to elemental sulfur. Commercial technology is already
available for this reduction, as will be discussed later. Combined
with the commercially demonstrated MgS03 calcination, there results
a complete regenerable flue gas desulfurization process. The
question addressed in this report is the feasibility of combining
the endo thermic calcination with the exothermic S02 reduction. This
combination does not appear to have been considered, but should
result in considerable simplification of process heat transfer and
better overall thermal efficiency. The objective of this study was
to make an assessment of the technical feasibility of this combination
option by means of a conceptual process design.
The study begins with a description of the state of the
art of Mag-Ox and other pertinent technology. Next, a literature
survey is presented that outlines what is known about MgS03 decom-
position and SO 2 reduction to elemental sulfur. Preliminary cal-
culations for the MgSO 3 -reducing gas system are then made to iden-
tify promising temperature, reducing gas, and stoichiometry
conditions. Based upon these preliminary calculations it was shown
that (1) a reducing calciner that would use the heat given off by
the exothermic SOZ reduction reaction to provide the heat required
-------
by the endothermic MgSOl calcination reaction was feasible and
(2) further treatment of the reducing calciner off gases was
required for more complete conversion to elemental sulfur.
Process designs were calculated for two schemes. The
first scheme requires a catalyst in the reducing calciner to
promote the gas phase SO2 reduction reaction. The calciner would
operate at 500°C. The second scheme would not require a catalyst
but would operate at 900°C. Both schemes require a low temperature
catalytic process (similar to a Glaus plant) for efficient con-
version of the gaseous mixture to elemental sulfur.
The most significant process problems identified are
the dusting of the MgSOa and its reaction products (with attendant
effects on catalysts) and separation of catalyst from MgO in a
catalytic reducing calciner. Information gaps are associated
primarily with the reducing calciner. They include decomposition
kinetics of MgSOs and SO2 reduction in the MgSOs-MgO system.
Process development recognizing the above problems is recommended.
1.1 State-of-the-Art
1.1.1 Mag-Ox Process
Commercial scale experience with the magnesium oxide
scrubbing process has been gained in the following two facilities.
A prototype demonstration unit on a 155 Mw oil-fired
boiler at Boston Edison's Mystic Station (tested from
April 1972 to June 1974).
A prototype demonstration unit on a 100 Mw coal-fired
boiler at Potomac Electric Power's Dickerson Station
(tested from September 1973 to September 1975).
-2-
-------
Additional Mag-Ox scrubbing units which are either cur-
rently under construction or in advanced planning stages include:
A 120 Mw installation at Philadelphia Electric's
Eddystone Station (coal-fired) undergoing start-up.
Planned expansions by Philadelphia Electric at their
Eddystone (576 Mw of additional scrubbing capacity)
and Cromley (150 Mw) stations.
Since the Boston Edison facility is representative of
this process technology, the Mystic scrubbing system is used here
as a basis for discussing important operating characteristics of
magnesium oxide scrubbing systems in general. A schematic view of
the Boston Edison MgO scrubbing system is shown in Figure 1-1.
In the Mag-Ox scrubbing process S02 sorption is accom-
plished by contacting hot flue gas with an aqueous alkaline solution
of MgS03. Although a venturi scrubber was used in the Boston
Edison system, this process can be used with any properly designed
vapor/liquid contacting device. Key chemical reactions which take
place in the Mag-Ox scrubbing system are the following:
MgO hydration/dissolution
MgO(s) + H20(aq) + Mg(OH)2(s) + Mg4+(aq) + 20H'(aq) (1-1)
S02 sorption
S02(g) •* S02(aq) (1-2)
S02 reaction
S02(aq) + 20H~(aq) -> S07 (aq) + H20(aq) (1-3)
-3-
-------
M,Q FROM ACID PLANT
ACID PLANT
FIGURE 1-1. PROCESS FLOW SHEET FOR BOSTON EDISON MgO-S02
SCRUBBING SYSTEM (OIL-FIRED BOILER)
Source: (KO-134)
-------
sulfite precipitation
ag) + S0;(aq) + xH20(aq)
-> MgS03.xH20(s) (1-4)
x = 3 or 6.
In addition to these main reactions, some oxidation
of product sulfite occurs. This results in the formation of
MgSO,, , an undesirable by-product, since it is more difficult
to regenerate than MgS03.
Net removal of sorbed sulfur species is accomplished
through the precipitation and continuous discharge of MgSO3
solids from the system. MgSOi* is removed in the adherent water.
Product MgSO3 crystals are dried at 200°C, a temperature which
is sufficiently high to drive off both free water and associated
waters of hydration (MgSO3 precipitates in a hydrated form
MgSOs'xH 0; where "x" equals either 3 or 6 depending on the
conditions of operation). At the same time, however, the exit
temperature of the MgSO3 solids leaving the drier must be kept
below the value at which thermal decomposition of MgSO3 begins
to occur (MgSO3 •* MgO + S02) . One aspect of the MgS03 drying
process which has a significant impact upon downstream processing
requirements is related to the dehydration step. As the MgS03
solids lose their waters of hydration, the thermal and mechanical
stresses which accompany this reaction tend to result in the
formation of product solids which are very finely divided. This
can create severe dust and reagent loss control problems. This
aspect of the regeneration problem is discussed in some detail in
a later section of this report.
The processing scheme which is currently being used to
regenerate MgO is shown in Figure 1-2. In the case of the
-5-
-------
SO2 GAS CLEANING
CONCENTRATED SO; GAS
"TO SULFURIC ACID PLANT
CONYEYOa
REGENERATED
MgO
SILO
MgO/S02 SCRUBBING
SYSTEM
(See Figure 1-1)
FIGURE 1-2. PROCESS FLOW SHEET OF EXISTING MgS03
REGENERATION FACILITY
Source: (KO-134)
-6-
-------
Boston Edison System and in the final phases of the Potomac
Electric Power program, MgO was regenerated in a facility
operated by the Essex Chemical Company in Rumford, Rhode Island.
In the Essex plant, MgSOs/MgSO^ solids were calcined
by heating the solids in a rotary kiln at essentially atmo-
spheric pressure. The mid-kiln temperature was about 680°C.
Coke was added to the calciner along with the feed solids to
act as a MgSCK reductant. A typical calciner feed composition
is shown in Table 1-1 (EN-316).
TABLE 1-1
TYPICAL CALCINER FEED COMPOSITION
MgSOs 63.9
MgS04 12.7
MgO 2.8
Water and Inerts 21.0
The principal reactions which take place in the
calciner are:
MgS03 + MgO + S02 (1-5)
MgSCU + %C ^ MgO + S02 + %C02 (1-6)
The existing process yields a solid product which is
987o MgO. The gas stream leaving the calciner is a dilute SO 2
mixture whose approximate composition is given in Table 1-2.
TABLE 1-2
TYPICAL CALCINER EXIT GAS COMPOSITION
N2 73
C02 6
02 5
H20 7
S02 9
-7-
-------
The residence time of the solid phase in the calciner is abouu
one hour (ZO-008).
In spite of some operational difficulties which have
been experienced with the MgSOa calcination process described
above, this technology has been shown to be a viable regenera-
tion process option. Likewise, the production of sulfuric acid
from SO2 is proven technology.
The direct reduction of MgSOs to produce MgO and
elemental sulfur does not appear to have been attempted on a
commercial scale. Available information on related process
technology, however, indicates that one approach to MgSOs
regeneration which does appear to be feasible is a two-step
process involving the decomposition of MgSOa to MgO and SO2
followed by a gas phase S02 reduction step. For this reason,
it is appropriate here to discuss the status of existing
technology for producing elemental sulfur from SOa.
1.1.2 Elemental Sulfur Production from SOa.
Three elemental sulfur production techniques which
have particular relevance to the present study are the
following:
Allied Chemical Process,
Asarco-Phelps Dodge Process,
Outokumpu Oy Process.
-8-
-------
General operating characteristics of each of these processes
Q >* ^ c? i irrrm *a T* -i rr ^ s3 -T T"\ T"1 o 1^ 1 /-^ T ^
are summarized in Table 1-3.
It is obvious from the information presented in the
table that all existing processes for producing elemental sul-
fur from SOa are similar with respect to their general principle
of operation. In each case, SO a is reacted with an appropriate
reducing gas such as CHi*, CO, Ha (or a mixture thereof) to form
elemental sulfur, COS, and HaS. Since equilibrium sulfur pro-
duction is maximized at low temperatures, there appears to be
an incentive for using a catalyst to promote this initial
reducing gas/SOa reaction since this option provides favorable
kinetics at lower temperatures. SOa conversion is controlled
so that an H2S:S02 mole ratio of 2:1 is obtained in the product
gas. With this approach, a tail end Glaus reactor can be used
to convert residual S02 and H2S to additional elemental sulfur
(2H2S + S02 -> 3S 4- 2H20) . A more detailed discussion of
these sulfur production technologies is presented in Technical
Note 200-045-31-Ola in the Appendix of this report.
1.2 Subtask Description
In this section, a general description is presented
of the technical approach which was followed in assessing the
feasibility of producing MgO and elemental sulfur from MgS03.
Basically, overall effort on this program was divided into the
five major subtask areas shown in Figure 1-3. A general
description of the work performed during each of these
-9-
-------
TABLE 1-3
COMMERCIAL PROCESSES FOR PRODUCING SULFUR FROM SO;
Process Developer
and Location
Allied Chemical -
Sudbury, Ontario,
Canada
Application
Treatment of tail
gas from sulfide
ore roasting facil-
ity.
Feed
Properties
Gas:
S02
i
i—1
o
Process Details*
Catalytic reduction of S02 with methane
Key reactions are:
CfU + 2SO2 + C02 + 2H20 + S2
4CH,, -f 6S02 + 4C02 + 41I20 + 4H2S + S2
Optimum reaction conditions are those which
yield;
(1) maximum conversion of SOz to elemental
sulfur (over 40%)
(2) a product gas containing H2S and S02 in
a 2:1 mole ratio.
After the product sulfur is condensed, a two-stage
Glaus reactor system is used to convert the re-
maining H2S and S02 to sulfur and water.
2H2S + S02 + 3S + 2H20
90+% conversion efficiency has been demonstrated.
Asarco-
Phelps Dodge -
El Paso, Texas
Pilot plant for
treatment of
Cu-Pb smelter
tail gas.
Gas: Catalytic reduction of SO2 with CO/H2 mixture
10-157. S02 produced by methane reforming. 70% conversion of
2-3% 02 S02 to S achieved in primary reactor. Residual
H2S/S02 reacted in single stage Claus reactor.
Total sulfur recovery obtained: 08-92%.
Outokumpu Oy
Company -
Finland
Production of
FeO, elemental
sulfur and sul-
furic acid from
pyrite ore
(FeS2).
FeS2 Solids Solids are suspended in hot reducing gas produced
by partial oxidation of fuel oil in a flash
smelter. CO and H2 react with S02 to produce S
and H2S as the gases are cooled (non-catalytic).
Reaction of H2S and S02 to produce more sulfur
occurs as the gases are cooled further. Sulfur
yield is optimized by passing residual H2S and'
S02 over an alumina catalyst at 270°C.
* More complete descriptions of these processes are presented in the Appendix
to this report in Technical Note 200-045-31-01.
-------
Current State-
of-the-Art and
Process Equip-
ment Performance
Data
LITERATURE
SURVEY
Process
Chemistry
Data
THERMODYNAMIC
SCREENING
Thermo
Feasibility
Data
SPECIFICATION
OF PROCESS
ARRANGEMENT
Process
Definition
PROCESS ANALYSIS
HEAT/MATERIAL
BALANCE
CALCULATIONS
CONCLUSIONS
AND
RECOMMENDATIONS
FIGURE 1-3. SUBTASK BREAKDOWN FOR STUDY TO DEFINE THE FEASIBILITY OF PRODUCING
ELEMENTAL SULFUR FROM MgS03
-11-
-------
subtasks is presented below.
Subtask 1 - Literature Survey
The goal of this subtask was the collection and
analysis of three different types of data:
Chemistry of the Mg-O-S System
These data were needed to define potential
methods of producing elemental sulfur from
MgS03.
Relevant Process Technology
This portion of the literature survey provided
background data for the state-of-the-art
discussion presented in Section 1.1. Equip-
ment performance data were also gathered as
part of this subtask for use in subsequent
phases of the program.
Kinetic Data
Two types of kinetic data were sought:
(1) MgS03 solids decomposition reactions;
(2) sulfur forming reactions involving such
gas phase components as CHi*, CO, Ha, tbO,
H2S, COS and S02. These data were needed
to define conditions under which the reac-
tions of interest would proceed at favorable rates
-12-
-------
Subtask 2 - Thermodynamic Screening
Equilibrium calculations for promising reaction sys-
tems were performed to determine the influence of temperature and
stoichiometry on gaseous and solid phase product distributions.
Subtask 3 - Possible Process Arrangements
Using information from the literature and guided by
the results of Subtask 2 above, conceptual processing strategies
utilizing various types of equipment and process arrangements
were developed. Factors considered here include the temperature
ranges of interest, energy transfer considerations, process sim-
plicity, and so on. One low temperature process using a low Btu
gas was investigated. The low temperature will involve a gas
phase catalyst so that the endothermic decomposition heat can be
supplied by the exothermic gas phase reduction reaction. Three
noncatalytic high temperature decompositions were investigated.
Reductants were a low Btu gas (air blown gasifier), a medium
Btu gas (oxygen blown gasifier), and H2S.
Subtask 4 - Heat and Material Balances
Heat and material balance calculations were performed
for the promising process arrangements and reductants of choice
devised in Subtask 3. These calculations, which were intended
to determine the performance characteristics and energy require-
ments of an integrated sulfur recovery system, were made by as-
suming that equilibrium was reached in all process reaction vessels
Subtask 5 - Conclusions and Recommendations
Based upon the results of the above four subtasks,
promising approaches to elemental sulfur production and gaps
in the existing process data base were identified. Also,
recommendations for future studies of this problem were proposed.
-13-
-------
2.0 TECHNICAL BACKGROUND
In this section justification is presented for the
selection of specific processing schemes for producing elemental
sulfur from MgS03 which appear to be technically feasible. The
information presented here is based mainly upon data collected
in the literature survey subtask.
2.1 Process Chemistry
This section summarizes the pertinent results of the
process chemistry portion of the literature survey subtask.
A more complete description of the results of this effort is
presented in Technical Note 200-045-31-Ola which is in the
Appendix of this report.
As a result of this subtask, it was concluded that a
two-step approach to the production of elemental sulfur from
MgS03 is required. No reports of a direct reaction between
MgS03 solids and a reducing agent to form elemental sulfur
were found. For this reason, effort on this subtask was directed
toward defining (1) the chemistry of MgS03 decomposition and
(2) general gas phase reactions of the form:
SO2 + Reducing Gas
-v Elemental Sulfur -1- By-Product (2-1)
The chemistry of MgS03 decomposition will be discussed first.
2.1.1 MgS03 Decomposition
Two hydrated forms of MgS03 exist: a tri- and a
hexahydrate. Both can be formed in MgO flue gas desulfuriza-
tion systems. Several investigators have studied the thermal
behavior of MgS03 solids. Starting with the hexahydrate, three
-14-
-------
waters of hydration are lost as the solid is heated between
60 and 100°C. By 200°C the last three waters are lost. Some
sulfite decomposition is also observed at this temperature.
MgS03 decomposition rates become significant at
temperatures in the 300-600°C range. Decomposition of MgS03 in
a nitrogen atmosphere at temperatures > 300°C yields measurable
quantities of MgO, S02, MgSOn, free sulfur and MgS203 (up to
550°C, magnesium thiosulfate is unstable at higher temperatures).
Experimental evidence suggests that the above products are
formed according to the following reactions:
MgS03 -" MgO + S02 (2-2)
2MgS03 + S02 •* 2MgSO., + %S2 (2-3)
MgS03 + %S2 -> MgS203 (2-4)
MgS03 decomposed very rapidly at temperatures > 600°C.
An empirical expression for the decomposition rate of MgS03
developed by Kim (KI-110) indicates reaction times of 38 seconds
for 90% decomposition at 700°C and 30 seconds for 9970 decomposi-
tion at 800°C.
2.1.2 Chemistry of Reduction of S02 to Elemental Sulfur
In this portion of the literature survey thermodynamic,
kinetic, and ,other pertinent information pertaining to the
chemistry of obtaining elemental sulfur from sulfur dioxide was
reviewed. The literature was searched from 1967 through the
present using Chemical Abstracts. Previous literature reviews
were relied on for access to key investigations conducted prior
to 1967.
-15-
-------
Reducing agents which were considered include:
methane,
carbon monoxide,
hydrogen,
CO + H2
carbon,
coal,
H2S.
Each of these systems is discussed in detail in Technical
Note 200-045-31-Ola in the Appendix. Important characteristics
of these systems which are pertinent to the overall scope of
this study are discussed in summary fashion below.
2.1.2.1 S02 Reduction by Methane
The use of methane as a reducing agent for sulfur
dioxide has been developed for commercial use by Allied Chemical
Corporation. A number of investigations have also been carried
out by several groups of Soviet scientists.
The overall reaction of interest in this system is:
2S02 4- CH^ -> S2 + C02 + 2H20 (2-5)
Side reactions which result in the formation of H2S, COS, and
CS2 are also significant.
In kinetic studies of this system, maximum elemental
sulfur yields were obtained at S02:CH4 ratios of 2:1 and at high
temperatures. Bauxite was shown to be an effective catalyst at
T > 800°C. In a non-catalytic system, this reaction is slow for
T < 1200°C.
-16-
-------
2.1.2.2 Reduction of Sulfur Dioxide by Carbon Monoxide
The reduction of SOa by carbon monoxide has been the
subject of numerous investigations. The main reaction of
interest here is
2CO + S02 = 2C02 + %S2 (2-6)
In the absence of a catalyst this reaction is slow for
T < 1000°C; therefore the emphasis of many studies of S02
reduction by CO is on catalysis. The performance character-
istics of a wide range of alumina based catalysts have been
studied experimentally. Catalytic systems show favorable
rates and first order kinetics at temperatures greater than
400°C. COS formation is a major problem.
2.1.2.3 Reduction of Sulfur Dioxide with Hydrogen
The main reaction of interest here is
S02 + 2H2 Z 2H2°(g) + %S2 (2-7)
however, H2S formation reactions are also significant in this
system. Thermodynamically, the formation of elemental sulfur
is favored by low temperatures (T < 400°C), however, the un-
catalized reaction proceeds too slowly at those temperatures to
be feasible from a commercial standpoint. With an appropriate
(bauxite or reduced alunite) catalyst, favorable kinetics were
obtained once temperatures in the 300-500°C range were reached.
The uncatalyzed reaction apparently proceeds slowly at tempera-
tures below 900°C.
-17-
-------
2.1.2.4 Reduction of Sulfur Dioxide with CO + H2
A number of studies of S02 reduction by converted
natural gas are reported in the literature. Both catalytic
and non-catalytic systems have been studied. Although this
system has not been studied as extensively as those discussed
previously, its behavior follows the same patterns which were
discussed for the pure component cases.
2.1.2.5 Sulfur Dioxide Reduction by Coal
One reference to the reduction of S02 with coal was
found in the recent literature - a patented gas purification
process involving the reaction of an S02-rich gas with coal
at temperatures >. 425°C (ST-322) . A high sulfur content coal
may be used. No additional information was available in the
abstract of this patent.
2.1.2.6 Sulfur Dioxide Reduction by Carbon
Studies of the reduction of S02 by various forms of
carbon have been reported in the literature. Mechanisms have
been suggested involving formation of carbon-sulfur and carbon-
oxygen bonds. Lepsoe (LE-175) reported that in the presence of
carbon, continuous reduction of S02 takes place through the
following reaction scheme:
S02 + C = C02 + %S2 (2-8)
C02 + C = 2CO (2-9)
2CO + S02 = 2C02 + %S2 (2-10)
-18-
-------
Above 650°C, these reactions proceed very rapidly, CS2 and
COS formation can occur with this system.
2.1.2.7 Sulfur Dioxide Reduction by H2S
This reaction is discussed below. Equilibrium distri-
bution will be similar to reduction of S02 with H2S. The
kinetics are similar to the Glaus reaction.
2.1.3 Other Potentially Important Reactions
Many of the reactions between reducing gases and
S02 produce significant quantities of undesirable by-products.
CO for example reacts with S02 to produce not only elemental
sulfur but also COS and CS2. Hydrogen (or H20) containing
reducing gases can react to form H2S.
Because some of these side reactions progress to
a significant extent under conditions which are favorable for
the production of elemental sulfur from S02, possible mech-
anisms for the conversion of these by-product species to ele-
mental sulfur were considered. A summary of the results of
this portion of the literature survey subtask is presented
below.
2.1.3.1 Conversion of H2S to Elemental Sulfur
The majority of the processes which are available to
accomplish this conversion step are based upon the Glaus
reaction.
S02 + 2H2S -> 2H20 +- Sx
A large number of commercially proven process variations based
upon this reaction system are available. These are discussed
in detail in the Appendix (see Technical Note 200-045-31-01 a) .
-19-
-------
2.1.3.2 COS/CS2 Formation
COS and CS2 formation occur as a result of gas phase
reactions between C02 , CO, S02, and elemental sulfur. This
problem area is best avoided by choosing reaction conditions
(temperature, catalyst, reducing gas, composition, and stoichi-
otnetry) which minimize the yields of these species. The use
of a reducing gas containing H2 also minimizes this problem.
2.2 Thermodynamic Screening
The objective of the thermodynamic screening subtask
was to examine the equilibrium sulfur yields which could be
obtained through the reaction of MgS03 with various reducing
gases. Equilibrium calculations were performed to determine
the influence of temperature, gas phase composition, and stoi-
chiometry on the gaseous and solid phase product distributions
which would be obtained if MgS03 were calcined in the presence
of a reducing gas. Overall heat balances for each reaction
system were also calculated.
2.2.1 Selection of Cases
Calculations were made for four reducing gases at five
stoichiometries and four temperatures (a total of 80 cases).
Reducing gases of CO, H2 , CO + H2 , and CH,, were selected on the
basis of their availability and their demonstrated capability
for reducing S02 to elemental sulfur. H2S will be similar to
H2 as far as equilibrium product distribution is concerned. The
stoichiometries of these reducing agents were varied both to
find the optimum conditions for elemental sulfur production and
to simulate ranges of conditions that would be found in
fluidized bed, co-current, or counter current reactors. A com-
plete discussion of the gas-solid flow conditions and tempera-
ture profiles obtained in each different type of reactor is
-20-
-------
included in the Appendix (Technical Note 200-045-31-02). The
four temperatures of 350, 500, 700, and 1000°C were chosen to
include the lower limit of 350°C for MgS03 decomposition and
the upper limit of 1000°C for non-catalytic reduction of S02 to
sulfur. Table 2-1 shows the selected equilibrium conditions.
TABLE 2-1
EQUILIBRIUM CASES
Four Temperatures (°C): 350; 500; 700; 1000.
Five Stoichiometries: 0.95; 1.0; 1.05; 1.5; 20.0.
Four Reducing Gases: 100% CEk ; 100% CO; 100% H2 ;
CO/Hz (50% each).
An energy balance around the overall process was made
possible by specifying the inlet temperatures of the gas and
solid streams involved in each case. A temperature of 25°C was
chosen for the MgS03 solids and the CH^ reducing gas. The other
three reducing gases, H2, CO, and the H2/CO mixture, were speci-
fied to be at 1000°C for purposes of computing reaction system
energy balances. H2, CO, and H2/CO were assumed to be hot since
these gases would most likely be produced on site using some
sort of reforming or gasification process.
The gas and solid species which were considered in
these calculations are listed in Table 2-2. These species
were selected on the basis of their potential for existence
in the reactor as determined by the literature survey. For
the majority of the species considered, thermodynamic proper-
ties were obtained from the well-known JANAF Thermochemical
Tables (ST-067). Some data for MgS203 were estimated.
-21-
-------
TABLE 2-2
CHEMICAL SPECIES CONSIDERED IN
THERMODYNAMIC SCREENING
Gas
H
H2
H20
H2S
02
CO
C02
COS
CS2
CH,
S
S2
S8
S02
S03
MgS
Condensed Phase
MgO MgS03
MgS MgS04
S (liquid) MgS203
C (graphite) MgC03
It should be noted here that, in the interests of
simplifying the computational procedures involved, S, S2 and
S8 were the only gas phase sulfur species which were considered
in the initial screening portion of the thermodynamic calcula-
tions which were performed as part of this subtask. In actual
fact, gaseous elemental sulfur with three through seven atoms
of sulfur per molecule can also exist at the conditions which
were of interest here. The total amount of elemental sulfur
vapor is still fairly well approximated by only S2 and S8. For
this reason, ignoring the S3_7 components does not affect the
results of these screening calculations. In subsequent process
design calculations, all potentially significant gas phase
species were included.
-22-
-------
2.2.2 Results of Thermo Screening Calculations
The results of the equilibrium calculations for the
80 different reaction cases considered are shown in Tables B3-1
through B3-8 of Technical Note 200-045-31-02 in Appendix B.
In all of the cases, the MgS03 was completely decomposed. The
main decomposition product was MgO. Some MgSOi* was formed in
certain low-temperature, low-stoichiometry cases. In none of
the cases were MgS203 or MgS thermodynamically stable.
When CO was used as a reducing gas close to a
stoichiometric amount, virtually all of the solid product
appeared as MgC03 at 350°C. This would not be a process prob-
lem area, however. Even if MgC03 were formed, it would decom-
pose during the scrubbing process to MgO and C02. MgC03 forma-
tion would be an economic penalty because of shipping costs
since it weighs almost twice as much as MgO.
The two reducing agents containing carbon (CH^. and
CO) showed large solid carbon formation tendencies at high
stoichiometries. This indicates that coke formation at the gas
inlet to a fluid bed or counter current calciner could be a
potential problem area. Coke deposition could cause problems
if the process is catalytic.
The highest sulfur yields occurred for the cases in
which CO was used as a reducing gas. At a CO stoichiometry
of 1.0 the elemental sulfur yield for this case was always
above 80%. At T < 450°C and 600°C < T < 800°C, a 90+% equilib-
rium elemental sulfur yield is obtained. This is shown in
Figure 2-1. With Ha, the equilibrium sulfur yield is consider-
ably less than that obtained with CO as shown in Figure 2-2.
The equilibrium sulfur yield for CHi* is intermediate between
those obtained with CO and H2 as shown in Figure 2-3.
-23-
-------
. 0
. 8
IH
I-H
en . 6
•U
O
H
M-t
0 .4
C
O
•H
4-J
O
(0
.2
Q
---- A
COS
200 400 600
Temperature,
800 1000
FIGURE 2-1. EQUILIBRIUM SULFUR DISTRIBUTION WITH CO
-24-
-------
1. Or
M-l
r-l
en
cfl
4J
o
4-4
o
o
•H
J_)
O
cfl
\ /
\ X
}-'<
/ v
/
01
i
i
i
,0 s
— • A SO;
200 400 600 800
Temperature, °C
1000
FIGURE 2-2. EQUILIBRIUM SULFUR DISTRIBUTION WITH H2
-25-
-------
1. or
* / \ --® s
/ \
4
c
•rl . 4V- v
u A
0 \
?! \ ^x
V
\
7 -~-- ------- a so
/
a
i _ i _ i
200 400 600 800 1000
Temperature, °C
FIGURE 2-3. EQUILIBRIUM SULFUR DISTRIBUTION WITH CH<
-26-
-------
Since the initial screening at four widely-spaced
temperatures indicated a number of apparent anomalies in the
equilibrium products if plotted, a closer look was taken at H2,
H2S, and H2 + CO at a stoichiometry of one. The results are
shown in Figure 2-4, where fraction of total elemental sulfur
is plotted versus temperature. Both a maximum and minimum
appear in all three cases, but are readily explainable in terms
of the reactions and thermodynamics involved. In the initial
low temperature region no MgSOs appears in the products, but
MgSCK does. As the temperature increases, the MgSCX decreases,
and the elemental sulfur increases until the MgSOi, disappears.
With no further increase of S02 due to MgSO^ decomposition, the
elemental sulfur declines with temperature in the range of 400-
550°G. In this range, the major sulfur species are the larger
ones, S5-8, for which the formation from S02 and H2S is exothermic.
The sulfur yield goes back up with temperature as
more S2 is formed. The reaction
2H2S + S02 = 2H20 + | Sx
is endothermic by 11.5 kcal at 298°K for x = 2, and exothermic
by 26 kcal for x = 8. In the higher temperature region, the
overall result for all S species is endothermic and the yield
increases with increasing temperature. For the simple gaseous
reaction, S02 + 2H2S, there is no MgO in the reaction system.
This eliminates the possibility of MgSCU formation and the
usual plot of elemental sulfur formation versus temperature
results as is shown in Figure 2-5.
A significant consideration is that the equilibrium
conversion to elemental sulfur is a minimum in the 550-700°C
range.
-27-
-------
OO
I.Or-
.9-
I
.8
.7
/
/ x
A
^N.
X-
CO
4-1
o
H
o
C
o
O
03
S-4
.6
.5
.4
.3
O
G
x S from 2H2S + MgS03
from CO + H2 + MgS03
S from 2H2 4- MgS03
.2
/
B.
Q
MgSOi, from 2H2 + MgS03
J
B
_L
300 400 ~500 600 700 800 900 1000 1100
Temperature, °C
FIGURE 2-4. REDUCING GAS/MgS03 REACTION SYSTEM
-------
1 Or
.8
M
cn
cd
4->
o
H
o
•H
4J
O
.6
.5
A
.3
.2
\
X
x—
S02 + 2H2S
2H20 4- i Sx
j I I I
j I
300 400 500
600 700 800 900
Temperature, °C
1000 1100
FIGURE 2-5. THE H2S/S02 REACTION SYSTEM
-29-
-------
In the cases where an excess of reducing gas was used,
all of the excess H2 tended to form H2S and all of the excess
CO to form COS. H2S formation tendencies were greater than
COS formation tendencies. Kinetic considerations would there-
fore provide the only incentive for operating the calcination
reactor with an excess of reducing gas.
The heat balance numbers which were generated along
with the equilibrium composition data are of interest because
they indicate whether it is necessary to provide or remove heat
from the reactor in order to maintain the specified steady
state temperature. It is obviously desirable to minimize the
need for transfer of heat. When using methane as a reducing
agent, the reactor would require the addition of heat. The
adiabatic operating temperature is between 700 and 1000°C for
CO and between 500 and 700°C for H2- Each reaction system is
exothermic below these temperatures. The heat requirements for
mixtures of H2 and CO fall between those of the pure components.
2.3 Process Implications
Based upon the process chemistry and thermodynamic
considerations discussed in previous sections, it is possible
to identify a variety of factors which are pertinent to the
conceptual design of a technically feasible sulfur-from-MgS03
process. Some of these factors are discussed in detail below.
Since the decomposition of MgS03 is endothermic and
the reduction of S02 is exothermic in the lower temperature
ranges, there is a large incentive for combining the calcina-
tion and SOrTreouct:ion steps in one reactor". The thermodynamic
screening results showed that with CO and H2 , the exothermic
reduction reaction can provide the necessary heat for the
-30-
-------
decomposition of MgS03 when operating below 700°C for CO and
below 500°C for H2. Therefore, if favorable kinetics exist at
those temperatures, the fuel required to produce the reducing
gas would be the only energy required by the calcination step.
More detailed heat balances were made for the
calcination step for four reductant cases, as shown in Table
2-3. It is seen that H2S may be eliminated as it must be heated
to 1325°C and the overall reaction is decidedly endothermic.
The high temperature non-catalytic option also requires an input
temperature higher than the output, in the kinetically viable
range of greater than 900°C.
The catalytic process is attractive for a variety
of reasons. First, there is no external heat requirement for
the catalytic process option. As shown in Table 2-4, CO and H2
require temperatures in excess of 900°C to reduce S02 to sulfur.
Methane will not reduce S02 to sulfur for T < 1200°C except
in the presence of a catalyst. With these high temperatures
in the calciner, there is the potential for "dead burning" of
the MgO. The "dead-burned" form of MgO, called periclase, is
unreactive. Another problem associated with the high tempera-
tures necessary for a non-catalytic process would be the need
to use exotic materials of construction. Above 600°C, special
materials of construction would probably be necessary.
A catalytic S02 reduction process can operate at
temperatures in the 400-450°C range. This eliminates many of
the problems related to high temperature operation. However,
with a catalytic process, there will probably be a reactor
effluent solid/solid separation problem. Several mechanical
options are available which are probably capable of solving
this problem. Some of these options are discussed briefly in
Section 2.4.
-31-
-------
TABLE 2-3
PROCESS OPTION TEMPERATURES. STAGE 2 OF FLUIDIZED BED
Reductant
Gas Effluent
Temperature
Gas Feed to Stage 2
Air-Blown Gasifier:
Catalytic
Noncatalytic
550
900
450
1130
Oxygen-Blown Gasifier:
Catalytic
Noncatalytic
900
800
1050
780
H2S Gasifier:
Catalytic
550
1325
-32-
-------
TABLE 2-4
SUMMARY OF MINIMUM TEMPERATURES FOR REACTION
Reaction
Non-Catalytic
Temperature
Catalytic
Temperature
Catalyst'
MgS03 Decomposition
550
450-500
Iron Bauxite
Iron and Chromium Oxides
Co
I
Reduction
CO Reduction
1200
Over 950
800
400
Bauxite
Reduced Alunite
Bauxite
Activated Alumina
Metal/Alumina
H2 Reduction
Over 900
300-400
Bauxite
Alunite
CO-H2 Reduction
No Information
400
Bauxite
A1203
* These are some of the catalysts for which a significant sulfur yield was
reported. This list is not complete.
-------
A noncatalytic process is attractive in its simplicity.
There are no gas phase catalyst and attendant solid/solid sepa-
rations problems. There is insufficient information available
at this time to choose between a catalytic and a noncatalytic
process, therefore, a process design was made for both.
For purposes of performing detailed process calcula-
tions , it was decided that a single catalytic reaction vessel
would be employed in Radian's conceptual sulfur-from MgS03
process. The single vessel design was chosen so that the heat
required for MgS03 decomposition reaction would be provided by
the exothermic S02 reduction reaction. The noncatalytic decom-
position of MgSOs was also carried out in a single vessel.
A brief investigation was made of the possibility
of using an initiator to start the homogeneous S02 reduction
at lower temperatures. For example, hydrogen peroxide could
provide OH radicals.
H202 + M = 20H + M
The known rapid reaction with H2 would provide H atoms.
OH + H2 = H20 + H
However, a consideration of the possible intermediate steps
indicates that there is not a low activation energy pathway
for overall reaction of S02 reduction to elemental sulfur.
The reaction
S02 + H = SO -I- OH
is central to such a scheme, but it is endothermic by 31
kcal. An activation energy of 36 kcal is thus indicated,
so that the homogeneous reaction could not occur in the
lower temperature range even if initiated by an external
-34-
-------
radical source. Rather, a temperature of the order of 900°C
would be required, just as with initiation by strictly thermal
radicals.
2.4 Equipment Considerations
In this section pertinent operating characteristics
of several different types of gas/solid contacting devices are
examined. This analysis emphasizes those features of each de-
vice which relates to its suitability for use as a MgSOs regen-
eration reactor. Key considerations in this analysis are the
ability of the contactor to promote efficient contact between
MgS03 solids and the reducing gas, heat transfer characteristics,
operability, and safety. The reactor types which were considered
include rotary kilns, multiple hearth furnaces, rotary grate
and flash roasters, and fluidized beds. These reactor types were
screened to determine which configuration was most suitable from
the point of view of accomplishing the calcination and reduction
steps in a single reaction vessel. The key features of each
reactor type are summarized below.
A rotary kiln is an insulated metal cylinder that
rotates upon suitable bearings and is slightly inclined to the
horizontal. Hot gases are used to heat the solid materials
and to carry away product gases from the decomposed solid. In
rotary kilns there is much less gas-solid contacting than in
fluidized units. In normal operation the kiln seals allow some
leakage of air, thus preventing operation under pressure or
vacuum. Special seals are required in processes which must
avoid problems associated with the entrance of outside air.
Multiple hearth furnaces consist of a number of
annular sloped beds mounted one above the other. Feed material
entering the top of the furnace falls from hearth to hearth as
a result of the movement of rabble arms. Hot gas flows counter-
currently upward through the hearths. Because of the poor
gas-solid mixing characteristics of this type of furnace, a
-35-
-------
high consumption of fuel and a low concentration of S02 in
the product gas is generally obtained when this type of contac-
tor is employed in the smelting industry.
Flash roasters are generally used when the gas/solid
reaction of interest is controlled by surface phenomenon or
when the solid particles are very small so that heat and mass
transfer from the interiors are very rapid. This type of fur-
nace usually consists of a brick lined cylindrical shell, which
encloses a relatively large combustion chamber, having one or
two collecting and desulfating hearths at the base of the cham-
ber and one or two drying hearths under these calcining hearths,
The operation of the flash roaster resembles the burning of
powdered coal in a furnace in that the solid concentrate is
normally injected into the combustion chamber through entrain-
ment in a stream of air.
Fluid bed reactors are often employed when using
reasonably small granular or powdery feed materials. However,
fluidizing MgS03 crystals may be a problem because of the
extremely small particle size involved - on the order of 10
micrometers. Fluid bed operation generally results in the
attrition of particles and high entrainment losses, in which
case recovery equipment such as cyclones would be needed. The
extremely small particle size of the MgS03 would probably lead
to severe entrainment problems. It is generally concluded
that particles distributed in size between 30-225 micrometers
are the best for smooth fluidization. Small particles (less
than 10 micrometers) frequently agglomerate. This can result
in the formation of large lumps in the bed. MgS03 particles
approach the lower limit for use in a fluid bed reactor.
-36-
-------
A newly developed circular grate pelletizer/roaster
which might avoid some of the problems associated with fluid
bed operation has been developed by McKee (IA-003). This cir-
cular grate apparatus features a horizontal, washer shaped
hearth which consists of a large number of metallic grate
elements. The entire structure, which is mounted on a single
center-support bearing assembly, slowly rotates through four
operating zones. Presently, as it is used in an iron-ore
pelletizing process, the four operating zones are feed/unload-
ing, drying, induration, and cooling. These operating zones
could probably be changed to accommodate a different type of
operation. In its existing application, a single air stream
flows successively through the four operating zones. This
promotes efficient heat utilization. The favorable character-
istic of this unit is its potential ability to pelletize the
MgSOs crystals before calcination. This may be a method of
avoiding the problems which would be associated with a fluid
bed reactor. Also the grate could carry a layer of fixed bed,
gas phase catalyst thus avoiding the solid/solid separation
problem.
In spite of these potentially severe particle carry-
over problems, the superior heat and mass transfer character-
istics of fluidized beds are significant factors which make
this contactor type attractive for this application. In
addition, a fluidized bed is one of the simplest models for
making process calculations. For this reason a fluidized bed
was the reactor type which was assumed to be employed in the
calcination/reduction section of Radian's conceptual sulfur
from MgS03 process which is described in Section 3.0. Even
through the use of this type of contactor is assumed here, it
must be recognized that potential solids carryover problems
do exist with this approach.
-37-
-------
The backmix characteristics of a fluid bed reactor
(relative to the solid phase) lead to non-uniform residence
times of solids in the reactor. For this reason, staging is
probably desirable from the point of view of achieving high,
uniform conversion levels. In the process design portion of
this study, the use of a two-stage flu.idized bed reactor was
assumed.
As discussed in the previous section, the use of a
catalytic process for the combined calcination/reduction step
is desirable because of the high temperatures which would be
associated with the non-catalytic process option. This leads
to potential MgO/catalyst separation problems, however.
The conclusions which were reached as a result of
this analysis can be summarized as follows. Conceptually,
it is desirable to accomplish the calcination and reduction
steps which must occur on the "front end" of this process
in a single catalytic reaction vessel. This approach appears
to be feasible, however, two potentially significant problem
areas are apparent. The extremely small sizes of the MgS03
feed particles involved will probably lead to severe dusting
problems. Also, if the catalytic process is used, a catalyst/
MgO solid/solid separation step will probably be required.
Although neither of these problems appears to be insurmountable,
they will have to be dealt with if further development of this
process is attempted.
Noncatalytic options have also been process engineered
because it is felt that it is too early to make a confident
decision as to the best course of action.
-38-
-------
3.0 PROCESS DESCRIPTIONS
Two processes were investigated, i.e., catalytic
and noncatalytic. For the catalytic process the reducing gas
was produced in an air blown gasifier. Two cases were con-
sidered for the noncatalytic process. They were reducing gases
from an oxygen blown gasifier and from an air blown gasifier.
Hydrogen sulfide was not considered in this step
because the heat requirement was judged to be too high. A
1000 Mw power plant burning 2.3% sulfur coal was used as the
basis for the design calculations. A flue gas sulfur removal
efficiency of 9070 was assumed. Given these assumptions, a
MgSOs/MgSOi, feed rate of 75.85 g-moles/sec (760 tons/day) was
determined.
The inlet MgSCh/MgSO^ feed rate was derived from the
design of an S02 removal system for Philadelphia Electric's
coal-fired Eddystone Station (120 Mw) burning 2.3% sulfur coal
(average) with a sulfur removal of 90% (EN-125). The flow rate
was linearly scaled to calculate MgSOa/MgSO^ production rates
for a 1000 Mw plant. The percent of MgS04 in the feed was
increased from 3.6% to 5.0%. The inlet MgSOs/MgSO* (and the
outlet MgO) flow rate was calculated to be 75.85 g-moles/sec.
This yields 264,211 kg/day of MgO.
3.1 Catalytic Process Description
Based upon the results of the literature survey and
thermodynamic screening subtasks, a process arrangement was
developed which appears to be a technically feasible method
of producing sulfur from magnesium sulfite. This process
arrangement, which incorporates the desirable features dis-
cussed in previous sections, is shown in Figure 3-1. A. low
Btu syngas is used as the reducing agent.
-39-
-------
200°C
<***-^-+
4SO°c
V&£°-£}<|>
120°C
-Uaa33_/$y_
r~^
\. /"
STAGE 2
X \
STAGE I
/ \
^r
/ 19.7 10' Bnyl
FI-UIDIZEO BED
EFFLUEMT
COOUEI2
FLUIDIZEO
BE.D
V — *— \
^6£D1^
6ED2
\^^ ^^s
•$iA°<.
jfoii-F
•*
CATAl
CONV
l»'c
COHV6PTEP
F££O
(3e3°FJ EFtUjENT
EJCWAN&EE
(435 -
O
I
(q-moles/sec)
Nz
CO
C0t
Hz
I-IJC
CH4
MM J
HIS
SOi
COS
• El«m S
MgSOi
/Uvg&Oil
MijO
TOTAL
/W3L WT.
k
72.54
3.31
75.85
IO5.06
7.970
I.SZO
<2>
155. 53
75 21
lfa.42
45. 3Z
24. 9fc
7.55
1-51
!_ 2 JO
3^810
2rt- IS
79
75.85
75.85
40.312
3.056
0.5821
<">
15599
O.O4
9817
uo8
6J.85
25.71
13. IZ
O.J7
3892
596.66
32.40
IZ 8Sfo
? 450
2.276
<5>
155.99
99 Ifc
8f oo
&.tA .
4.i4~^
o oz
tS.lfc
4W.il
11. O^
l2.8Sfa
2.150
t 9t?
fc3.oo
t.5.00
32 Ob
2.0fc
.397
1 OOO
155 99
99.lfc
61.00
8.4.4
1-M
o.oz
.It,
3<9.^l
30.48
io.b4fe
i.1S
7*8
<^>
IS5.99
»9- IS
66 .01
I.fo3
•8E
IO 69
354, .5Z
Z9.67
10. Mk
^.15
485
<9>
1053
10.55
JZ.Ofo
.J56
.Ofo44
J OO
V^
155.99
9<; '6
8poi
I.b5
-BE
-It
345.79
2961
'O 5O0
T.59
7.68
I. SrEEAW^ COftAPOSiT'OMS COMPUTE p TBOM EQL»IL/6(?(UAA
CALCULATIONS AT I.O Q+m
2. BASIS FoE rt/^SOj F6EO RATE. :
•IOOO MW POVVEK PiANT C^^NIMGi
2 i 7= SULFUE CLM.L
• MCIJOK 5CBU66INO SYSTEM SULFUI2
REMOVAL EFFICIEMCY : 90 ?„
5. DEDUCING Qfl5 STQicHIOAAETPY; I.O
. COMPLETE
TD
OP IIWUET
FIGURE 3-1. CONCEPTUAL PROCESS FLOW DIAGRAM FOR PRODUCING ELEMENTAL
SULFUR FROM MgSO3 IN A CATALYTIC DECOMPOSER
-------
The regeneration scheme which is proposed here
includes a two-stage fluid bed reactor followed by a two-bed
Glaus reaction unit. A Glaus unit is included in this design
since the conversion of S02 to elemental sulfur that occurs
in the fluid bed calcination reactor is only on the order of
50%. With the additional conversion, which is theoretically
obtainable through the use of a two-stage Glaus unit, a 97%
overall conversion efficiency of S02 to S. is theoretically
possible.
A sulfur recovery rate of 2.43 kg/sec (231 tons/day)
was calculated for the process scheme shown in Figure 3-1. If
the effluent gas stream from the process were to be recombined
with the power plant stack gas upstream of the scrubber, a
slight additional improvement in the overall sulfur recovery
efficiency of the overall sorption process would probably be
realized. In a typical 1000 Mw power plant, this gas stream
would increase the sulfur concentration in the stack gas by
about 40 ppm.
The process arrangement calculations discussed in
this section demonstrate that the calcination/reduction step
can be performed without the addition of heat from an external
source. The entire process is a net heat producer due to the
exothermic reactions in the Glaus reactor beds.
3.1.1 Catalytic Flow Sheet
The process flow sheet is shown in Figure 3-1. The
solids enter the top of a two-stage fluidized bed. In the
upper stage the majority of the calcination/reduction reac-
tions take place. The lower stage serves to complete the solids
reaction and to heat exchange the solids with the incoming
reducing gases.
-41-
-------
The gases laden with sulfur species are then cooled
and passed through a catalyst bed. The exothermic reaction
produces more elemental sulfur. Next, elemental sulfur is con-
densed and removed. The gases are reheated for additional
catalytic sulfur conversion. The sulfur is again condensed.
3.1.2 Process Engineering Calculations
This section discusses the assumptions made and
procedures used in developing the material and energy balance
for the process arrangement shown in Figure 3-1.
The reducing gas used in the process was a typical
air blown coal gasifier product gas (WA-199). The gas compo-
sition is shown in Table 3-1. This gas was chosen because
(1) it contained significant amounts of CO and H2 and (2) it
represented one of the more difficult cases for producing the
necessary heat in the calciner since 47.3% of the gas is
nitrogen.
Other reducing gas sources could also be used. A
typical composition for partial oxidation of Bunker C fuel oil
(3.5 wt % S) using the Shell- process (SH-200) is also listed
in Table 3-1 for comparison. The two gases are similar enough
so that no significant difference would be expected. One signi-
ficant aspect is that the sulfur need not be removed from the
reducing gas. Any high sulfur fuel may be used.
The first stage of the fluidized bed was assumed to
act primarily as a heat exchanger with no chemical reaction
between the species. The decomposition and reduction reactions
were assumed to take place in the second stage of the fluidized
bed. A temperature of 550°C was chosen for the second stage of
the reactor. At this temperature all of the MgS03/MgS04 will
decompose to MgO at equilibrium. The elemental sulfur yield in
-42-
-------
TABLE 3-1
TYPICAL AIR BLOWN REDUCING
GAS COMPOSITIONS
Mole Percent
Component
N2
CO
C02
H2
H20
CH,
NH3
H2S
Coal
47.
22.
5.
13.
7.
2.
0.
0.
3
9
0
8
6
3
4
7
Bunker C
56
21
2
16
3
0
0
0
.4
.6
.3
.1
.2
.4
Sources: Coal: (WA-199)
Bunker C: (SH-200)
-43-
-------
the reactor increases with decreasing temperature so that the
lowest possible temperature within operating constraints was
needed. Also, lower temperatures minimize material problems and
energy requirements for the reactor. Higher temperatures,
especially over 600°C, require the use of more expensive materials
of construction and increase the amount of heat that must be
supplied to the process by the reducing gas.
The temperature of the outlet MgO solids was calculated
to be 450°C. Inlet MgS03/MgS04 from a drier would be at 200°C.
A heat balance gave the inlet reducing gas temperature at 420°C.
The exit solids must be further cooled for ease of handling.
Should drying and dehydration-be done off site, the feed tem-
perature would be lower. A solids feed/effluent heat exchange
loop could solve both feed heating and product cooling problems.
In calculating the amount of reducing gas needed at a
stoichiometry of 1.0, it was assumed that HaS, NHs , and CHi*
would act as reducing agents even though it was noted in the
literature survey that CHi+ would not react with S02 to any
significant extent in a catalytic process operating below 800°C.
Since the CKU content of the gas was low it does not have any
appreciable influence one way or another. Methane was included
as a reactant for computational ease.
Another potential problem with this process which is
not indicated by the equilibrium calculations shown in Figure 3-1
is the formation of COS. At a 1.0 stoichiometry, equilibrium
calculations imply that the formation of COS is not a problem.
The kinetics of COS formation need to be investigated, however,
before it can be stated with certainty that COS is not a problem.
The literature survey reported that COS could be formed in large
or small amounts when S02 and CO are reacted depending on the
operating conditions of the reactor.
-44-
-------
Of the total sulfur in the gases leaving the reducing
calciner only about 5070 is in the form of elemental sulfur. A
lower temperature is required for further conversion. The gases
are therefore cooled and sent to a catalytic converter.
The catalytic converter beds were designed to operate
25° C above the dewpoint of sulfur since liquid sulfur condensate
will poison the converter catalyst (GA-155). In each case, the
converter inlet temperature was chosen such that the exothermic
heat of reaction in the converter beds was sufficient to keep
the gas temperature above the sulfur dewpoint.
The system pressure was assumed to be 1.0 atm for all
of the equilibrium calculations. In actual fact, the pressures
throughout the system would probably vary from 1-2 atmospheres.
The results of the equilibrium calculations would probably not
be significantly changed by pressure differences of this
magnitude.
The gas phase elemental sulfur must be removed to shift
the reaction equilibrium in favor of further formation of ele-
mental sulfur. This is accomplished by condensing molten sulfur.
Sulfur condensation calculations were made by
assuming that, except for the elemental sulfur species (S2 ,
83, ---Ss), the gas phase composition remained fixed. A sulfur
condenser temperature of 127°C (261°F) was chosen so as to be
7 C above the freezing point. Additional elemental sulfur
production by the Glaus reaction is favored thermodynamically
at 127 C; however, this will not happen because of the poor
noncatalyst kinetics at this low temperature. While the 7°C
approach in the sulfur condenser may have to be raised to avoid
freezing problems, the overall sulfur recovery will not be
significantly affected.
-45-
-------
After the first stage of sulfur condensation the
gases are heated so that the second catalytic bed exit tempera-
ture will be 224°C. This exit temperature was again selected
so as to be slightly above the elemental sulfur dewpoint.
The sulfur is again condensed at 127°C. Elemental
sulfur recovery is 97% of the sulfur entering the process. The
off gases contain minor amounts of S02, H2S, and Sx. These off
gases may contain too many noxious components to discharge
directly to the atmosphere. Incineration, additional catalytic
conversion, recycle, etc., are options to be investigated.
3.1.3 Catalytic Fuel Requirements
The heat requirement for the process is calculated
in two ways. The first way is intended to give an estimate of
the efficiency of the reducing calcining process. The calcula-
tion is carried out in the following manner. An "equivalent"
heat of combustion of the reducing gas is calculated for the
reducing gas at the fluidized bed inlet temperature (420°C)
going to combustion products of C02 , H20(v) , S(S-), and N2 at
the temperature of the second stage sulfur condenser (127°C).
Since the oxygen source is the MgS03, the oxygen for combustion
is calculated at 200°C, the MgS03 inlet temperature.
The equivalent heat of combustion calculated in the
above manner is distributed between the heat of calcination of
MgS03, the heat of the gas phase reactions, and process heat
rejection. The first two items are required and therefore
define the efficiency of the process. Table 3-2 summarizes
the process heat rejection loads. The reduction process ther-
mal efficiency is 77%.
-46-
-------
TABLE 3-2
ENTHALPY CONSIDERATIONS FOR THE PRODUCTION OF
ELEMENTAL SULFUR FROM MAGNESIUM SULFITE
^Equipment
Fluidized Bed Effluent
Cooler
Process Temperature
Inlet
550
Outlet
275
Heat Rate
(kcal/sec)
-1343
First Sulfur Condenser
Second Sulfur Condenser
TOTAL
300
224
127
127
-778
-300
-2421
Equivalent Heat of
Combustion of
Reducing Gas
-10392
Process Thermal Efficiency
-10392 + 2421
-10392
= 0.767
-47-
-------
The second way of looking at the heat requirement
is by comparing the S02 control process fuel requirement with
the power plant fuel requirement. Assuming a gasifier effi-
ciency of 75% the heat rate to the gasifier is 13,900 kcal/sec
This is 2.0 percent of the power plant heat rate of 700,000
kcal/sec (based on 10,000 Btu/kwh). If the calcination and
reduction steps were to be carried out separately, the fuel
requirement for calcination would be added to the present
reduction requirement. This additional heat would have to be
rejected or used in some manner. Compared to separate calcina-
tion and reduction facilities, the use of simultaneous solids
decomposition/S02 reduction will not only require less fuel
but also a potentially lower capital investment because of
less heat transfer surfaces.
The heat load for drying and dehydration will add
one percent or so to either one or two step processes. The
temperature level of the drying step is low (^200°C) so that
low level heat sources should be sought.
3.2 Noncatalytic Process
Two cases were investigated for noncatalytic
processes.
3.2.1 Noncatalytic Flow Sheets
The first, case was based upon a reducing gas from
an oxygen blown gasifier. This case is shown in Figure 3-2.
Because the gas phase reaction is noncatalytic, the
fluidized bed exit temperature is 900°C. As discussed in
Section 2, the formation of S2 in the gas phase reduces the
heat available for calcination of MgS03. The result is that
-48-
-------
FLUIDIZED 8EO
-P-
<03
I
COMPONENT
<^/ct*olM/set
N2
co
C02
HZ
HzO
CH«
NH3
HzS
S02
COS
•EtE/v\ S
MjS03
MgSO4
Mg O
TOTAL
/V\OL WT
K^/aec
Miu Ibs/J0>
^ c
240
76.20
&.70
74.0O
2010
O.9I
-
I.2O
-
-
-
-
-
-
18' .11
Kb 88
3.074
.585S
-
<£>
75.65
7S.S5
4o.3 12
3.056
0.5621
<€>
2. -JO
at3
Sojfc
l.8b
7717
-
-
18.29
IJ83
oai
•^601
-
-
-
Zli 77
3503
7.98fc
1 521
Z.OI
45 92
-
-
-
45 9£
32.0fa
I.472
0. 280
-
<$>
240
^t3
803fo
i.Sfa
7747
-
-
18 39
II 85
082
0091
-
-
-
1950S
33.2fa
6. 514
1 241
7&0
0
2.4O
-
8379
.002
9l Ol
-
-
6 .70
3.34
o.oi
21.06
-
-
208 33
31 Z7
fa.SM
' 245
668
<§>
2.40
-
8379
.002
9101
-
-
&.7O
3.M
O.OI
OOB9
-
-
-
187 34
31 17
5-84O
/-II2
7,feD
?4O
-
8360
-
95.85
-
189
0.93
-
8(4
-
-
-
19299
3040
-S.840
734
<8>
a 8n
-
21 8H
32. «,
O.fo99
0.153
-
<3>
2.1O
-
83.60
95 65
-
-
i 89
093
-
0037
-
-
164 94
3O.33
SoOS
lOtoe
/foo
€>
8053
-
-
-
305J
J20t
o t'58
OCW9
-
DPAWIMG MOIES •
CALCulATlotgs HT 10
. BASIS Foe ^ioj rtto
•IQOO MW POWCE Pt
2.SH SuLFoK tt»V4.
CtwuvPUTtO FKO«
RtWOVftl.
,/^, SO4
FIGURE 3-2. NONCATALYTIC REDUCTION, OXYGEN-BLOWN GASIFIER
-------
the inlet gas must be supplied at a higher temperature A
process alternate might be to supply extra reductant and air
for combustion to supply the extra heat.
The higher exit temperature for the noncatalytic
process has taken the process away from the minimum equilib-
rium sulfur formation range of 550-700°C. This results in the
need for a sulfur condenser and knockout ahead of the first
stage catalytic converter. The noncatalytic reaction temperature
was selected based on limited laboratory data. Further bench
scale studies should address this point. Reaction initiation
will be aided by the reducing gas feed to the calcining bed
being 150°C higher than the final bed temperature.
After the elemental sulfur has been condensed and re-
moved, the gas is reheated and sent to the first stage catalytic
converter. Equilibrium limits the conversion. Therefore,
elemental sulfur is again condensed and the gases returned to a
catalytic converter. The elemental sulfur formed is again
condensed. The waste gases are sent to final disposal.
3.2.2 Noncatalytic Process Design
All calculations were carried out at one atmosphere
pressure absolute. Overall removal of elemental sulfur is
about 9670. Design considerations of the catalytic recovery
units are similar to the catalytic fluidized bed system.
3.2.3 Noncatalytic Fuel Requirements
The heat requirements for the noncatalytic (fluidized
bed) process were calculated in the same manner as for the
catalytic process. The process heat rejection duties and equiv-
alent heat of combustion are given in Table 3-3. The reduction
process thermal efficiency is 80.3%.
-50-
-------
TABLE 3-3
ENTHALPY CONSIDERATIONS FOR THE PRODUCTION OF
ELEMENTAL SULFUR FROM MAGNESIUM SULFITE
NONCATALYTIC PROCESS
Process Temperature
^-Q Heat Rate
Equipment Inlet Outlet (kcal/sec)
First Fluidized Bed
Effluent Cooler 900 253 -1448
Final Fluidized Bed
Effluent Cooler 200 127 -163
Final First Bed
Effluent Cooler 286 127 -321
Second Bed Effluent
Cooler 239 127 -160
TOTAL -2092
"Equivalent" Heating Value of Reducing Gas -10640
Thermal Efficiency - -1064Q +092 = Q.803
-
-51-
-------
The fuel to the gasifier represents 2.07o of total
power plant heat rate if the gasifier is again assumed to
operate at 75% efficiency.
3.2.4 Noncatalytic-Air Blown Gasifier Case
Reducing gases from an air blown gasifier were also
considered. The process flow diagram for the reducing calciner
is shown in Figure 3-3.
In order to maintain the outlet temperature at 900°C
an inlet reducing gas temperature of 1130°C (2066°F) is required
This was considered to be unreasonably high so further calcula-
tions were not made in this feasibility study. If excess reduc-
ing gas and air were introduced to provide extra heat, a lower
temperature could be used and the process might look attractive.
At any rate, the process arrangement would be similar
to Figure 3-2 with compositions similar to those given in the
flow diagram of Figure 3-1.
-52-
-------
DPAWINQ
t STEEAIVA COWPOSJTIOWS OSMPIJTED feoflA
CALCULATIONS AT 1.0 o»m
2, BA5I5 FoR M^SOj FEED RATE :
•
I
FLUIDIZED
BED
H30 c
Air- Blotun
(?educ4ani
COM FOMENT
g /moles/sec
N2
CO
COZ
M2
M2O
CH«
MH3
NzS
SO^
C05
*ELEM 5
MgSOj
MgSCM
AAgO
TOTAL
MOL. W/T.
Kg /sec.
MILL, fbs/doy
*c, g-o^ogg
^ molecule
MW
28. oz
28.0!
44.OI
2.016
I8OI6
16.04
17.03
34.08
64.06
60.07
32.0fc
104.38
120 38
40.312
72.54
3.31
75.85
10506
7-970
<2>
155.33
75.2J
1642
<5.3Z
24.96
7.55
131
2.30
<5)>
75.85
75-85
40.312
3.058
0.58Z4
155.98
3. 7(
94.56
2 05
-7/.S4
16.05
11.37
O.9/
49.82.
405.99
31.667
i?.85£>
2.449
2.009
FIGURE 3-3 NONCATALYTIC REDUCTION AIR-BLOWN GASIRER
-------
4-° RESULTS AND CONCLUSIONS
In this section the significant results of this study
are summarized. The conclusions generated as a result of this
effort are also discussed.
4.1 Summary of Results
Subtask 1 - Literature Survey
From the literature survey it was concluded that
probably the reaction mechanism for producing elemental sulfur
from. MgS03 involved two steps. The first step was decomposi-
tion of MgS03 to MgO and S02. The second step was reaction
of S02 with an appropriate reducing gas. No evidence was found
to support the existence of a direct gas-solid reaction for
producing elemental sulfur from MgS03.
Descriptions of several commercial processes which
yield elemental sulfur as a product of reducing gas reactions
with S02 were found in the literature. CO, H2 , and methane
are all being used in this application currently. Since the
conditions for MgS03 calcination are approximately known, the
use of this two-step approach to elemental sulfur production
definitely appears to be technically feasible.
Subtask 2 - Thermodynamic Screening
In the thermo screening subtask, chemical reaction
equilibrium and heat balance calculations were used to define
the effects of temperature and reducing gas stoichiometry upon
the equilibrium sulfur yield which would result from the reac-
tion of MgS03 with four different reducing gases: H2, CO, CHu,
-54-
-------
and a H2/CO mixture. The highest equilibrium sulfur yields
were obtained in cases where CO was used as the reductant.
Significant (> 50%) conversion of S02 to sulfur was obtained
with all four gases under optimum conditions.
Among the significant conclusions which were reached
as a result of this effort are the following.
(1) At high stoichiometries, the two reducing
gases containing carbon (Cm and CO) showed
large solid carbon formation tendencies.
This indicates that coke formation at the
point of reducing gas injection to the cal-
cination reactor is a potential problem.
(2) At high stoichiometries, sulfur has a
thermodynamic tendency to react with excess
H2 to form H2S and excess CO to form COS.
This indicates that, from an equilibrium
point of view, no incentives exist for oper-
ating with a large excess of reducing gas.
It should be noted that this conclusion does
not take into account the kinetic factors that
may be involved in these reaction systems.
(3) The enthalpy data which were generated along
with these equilibrium results are also sig-
nificant. Since the MgS03 decomposition
reaction is endothermic and S02 reduction
reactions are exothermic, heat transfer con-
siderations make it desirable to accomplish
both of these conversion steps in a single
reaction vessel. The heat balance calcula-
tions which were performed as part of this
-55-
-------
subtask indicate that the adiabatic
operating temperature of the calcination
reactor is in the 700-1QOO°C range when
CO is used as the reductant and between
500 and 700°C for H2. Each reaction
system is exothermic (requires heat removal)
at lower operating temperatures. The heat
requirements for H2-CO mixtures fell between
those of the two pure component cases. The
methane system was endothennic at all tempera-
tures considered (T <^ 1000°C) .
Based upon the findings of the literature survey and
thermo screening subtasks, it was concluded that an attractive
conceptual approach to the production of elemental sulfur from
MgS03 appears to involve the use of CO or H2 in a catalytic
MgS03 decomposition process. This conclusion is based upon a
variety of factors.
(1) Noncatalytic process options require high
reaction temperatures and high .feed tempera-
tures for adiabatic operation. This presents
materials of construction and external heat
transfer problems.
(2) With methane, even when a catalyst is used,
high reaction temperatures and an external
source of heat would be required. This makes
CO and H2 more desirable as potential S02
reductants.
(3) Low calcination temperatures should maximize
MgO product reactivity.
-56-
-------
The problem of separating the gas phase SOa
reduction catalyst from MgO product -is an incentive for a
noncatalytic gas phase S02 reduction reaction. One of the
main objections, high inlet reducing gas temperatures, could
be overcome by introducing excess reducing gas and air to
obtain extra heat of combustion. There is not enough infor-
mation presently available to make a choice between either
catalytic or noncatalytic composition options.
The use of H2S does not appear to be feasible since
external heat transfer would be required. Burning excess H2S
to provide the extra heat is unattractive.
Subtask 3 - Specification of Process Arrangement
Based upon the results of the first two subtasks,
process arrangements which appear to represent technically
feasible approaches to the production of elemental sulfur from
MgS03 were developed. A major portion of this effort was con-
cerned with an evaluation of different types of solid-gas
contactors to determine which reactor type would be most suit-
able for use as a calcination/reduction reactor.
It was concluded that the superior heat and mass
transfer characteristics of fluidized beds make this contactor
attractive for use in this application. For this reason, a
fluidized bed was assumed to be used to accomplish the calcina-
tion step in the conceptual process arrangement which was
developed in this subtask. Potential problem areas which were
identified as being associated with this reactor design in-
cluded: (1) MgO solids entrainment problems and (2) catalyst
MgO solid separation problems.
Since the equilibrium sulfur yield in the calcination
reactor outlet gas was only on the order of 50%, a two-stage
catalyst reaction unit was also assumed to be used with this
process in order to obtain additional sulfur recovery.
-57-
-------
Process engineering calculations indicate that high
equilibrium sulfur yields are possible with reasonable fuel
requirements. Although no reports of this approach to MgO
regeneration were found in the literature, this approach appears
to be technically feasible.
5.0 RECOMMENDATIONS
Further study of the feasibility of producing
elemental sulfur from MgS03 should be directed mainly toward
the potential problem areas which were identified as a result
of this study. Basically, it appears that most of the problems
associated with the conceptual process which is proposed here
are concerned with the operation of the reducing calciner. The
technology associated with the treatment of the calciner ef-
fluent gas is well understood since the requirements of that
conversion step are similar to those which are currently being
handled by existing Glaus reaction units. Likewise, the pro-
duction of a suitable reducing gas is not anticipated to ba the
source of significant problems since a wide variety of reducing
gases are being produced on a commercial scale at present. In
theory, the effluent gas from any partial oxidation process
would be a suitable reducing gas for feed to the calcination
reactor.
The next step should be that of generating data
leading to the design and operation of a bench-scale reducing
calciner. The four major problem areas associated with the
design and operation of the calcination reactor which need to
be studied further are summarized below:
particle properties and size,
solid decomposition reactions,
-60-
-------
kinetics of the reducing gas/S02 reactions
with emphasis on heat transfer related problems,
evaluation of available catalysts for attrition
resistance and activity in the calcination
environment.
These phenomena must be understood and quantified
so that an engineering decision as to the type of reactor best
suited to this application can be made. Major mechanical
problems anticipated are:
handling of very fine particles (1-10 microns),
separation of MgO/catalyst mixtures.
After determining the reaction and heat transfer
kinetics, a selection of reactor type can be made to best
incorporate a solution to the above mechanical problems. This
reactor should be tested if possible at a bench-scale using
bottled reducing gases.
The above factors related to the noncatalytic process,
e.g. solid MgSOs decomposition and gas phase kinetic, should be
investigated in the temperature range of interest. There is
some possibility that a decision could then be made as to whether
a catalytic or a noncatalytic reducing calciner could be chosen.
Assuming the process still appears favorable, the
next phase should involve some type of pilot plant. A 20,000
nm3/hr (10,000 scfm) flue gas stream containing 2,000 ppm S02
could produce about one metric ton/day of elemental sulfur.
It is essential that a MgSOs stream of this size be treated
and elemental sulfur produced. It would be highly desirable
to coordinate this pilot study with an existing scrubbing
-61-
-------
facility to include magnesium sulfite drying and MgO recycle
On-site production of reducing gas is probably not necessary
and perhaps not even desirable.
The final step in this process development effort
would, of course, be a prototype unit in which all system
components would be tested in a closed loop operating mode.
-62-
-------
6.0 REFERENCES
EN-125 Environmental Protection Agency, Flue Gas
Desulfurization Symposium, 1973 , Proceedings,
EPA-650/2-73-038, Research Triangle Park, North
Carolina, 1973.
EN-316 Environmental Protection Agency, Flue Gas
Desulfurization and SuIfuric Acid Production via
Magnesia Scrubbing, EPA-625/2-75-007, Research
Triangle Park, North Carolina, 1975.
GA-155 Gathman, Wayne A., "Conversion of HaS to Elemental
Sulfur", Technical Note 500-006-03, Austin, Texas,
Radian Corp, 21 July 1975.
IA-003 lammartino, Nicholas R., Assoc. Editor, "Circular-
Grate Pelletizer Cuts Costs, Raises Quality",
Reprint. CEP 26., (May 1975).
KI-110 Kim, Yong K., R. Lindsey Dunn, and John D. Hatfield,
"Thermal Decomposition of Magnesium and Calcium
Sulfites", Muscle Shoals, Alabama, Division of Chemi-
cal Development, TVA, 1973.
KO-134 Koehler, George R., and Edward J. Dober, "New
England S02 Control Project Final Results", Flue
Gas Desulfurization Symposium, Atlanta, Ga., Nov.
1974.
LE-175 Lepsoe, Robert, "Chemistry of Sulfur Dioxide
Reduction: Thermodynamics", Ind. Eng. Chem. 30(1),
92 (1938).
-63-
-------
MC-076 McGlamery, G. G., Conceptual Design and Cost Study.
Sulfur Oxide Removal from Power Plant Stack Gas.
Magnesia Scrubbing - Regeneration: Production of
Concentrated Sulfuric Acid, EPA-R2-73-244, Muscle
Shoals, Alabama, TVA, 1973.
SH-200 Shell Development Company, The Shell Gasification
Process, brochure, Division of Shell Oil Company,
Houston, Texas 1974.
ST-067 Stull, D. R. , and H. Prophet, JANAF Thermochemical
Tables, Second Edition, NSRDS-NBS 37, Washington GPO,
1971.
WA-199 Waitzman, D. A., Evaluation of Fixed-Bed Low Btu Coal
Gasification Systems for Retrofitting Power Plants,
Palo Alto, Ca., EPRI, Feb. 1975.
ZO-008 Zonis, Irwin S., "The Production and Marketing of
Sulfuric Acid from the Magnesium Oxide Flue Gas
Desulfurization Process", Flue Gas Desulfurization
Symposium, Atlanta, Ga., Nov. 1974, Essex Chemical
Corp., 1974.
-64-
-------
APPENDIX A
TECHNICAL NOTE 200-045-31-Ola
"LITERATURE SURVEY ON THE RECOVERY
OF ELEMENTAL SULFUR FROM
MAGNESIUM SULFITE"
-65-
-------
TECHNICAL NOTE 200-045-31-Ola
LITERATURE SURVEY ON THE RECOVERY
OF ELEMENTAL SULFUR FROM
MAGNESIUM SULFITE
31 July 1975
Revised
2 December 1975
Prepared By:
Gary D. Brown
Nancy P. Phillips
Reviewed By:
William E. Corbett
David W. DeBerry
-66-
-------
TABLE OF CONTENTS
PAGE
1.0 INTRODUCTION 69
2.0 CHEMISTRY OF ELEMENTAL SULFUR FORMATION
FROM MAGNESIUM SULFITE 70
2.1 Magnesium Sulfite Decomposition 71
2.2 Chemistry of Reduction of S02 to
Elemental Sulfur 85
2.2.1 S02 Reduction by Methane 85
2.2.2 Reduction of Sulfur Dioxide
by Carbon Monoxide 96
2.2.3 Reduction of Sulfur Dioxide
with Hydrogen 104
2.2.4 Reduction of Sulfur Dioxide
with CO + H2 110
2.2.5 Sulfur Dioxide Reduction by Coal. . . 110
2.2.6 Sulfur Dioxide Reduction by Carbon. . 113
2.3 Conversion of Hydrogen Sulfide to
Elemental Sulfur 121
2.3.1 Gas Phase Glaus Process 121
2.3.2 Glaus Reactions in Liquid Media . . . 132
2.3.3 Other H2S Removal Processes ...... 133
2.4 Other Gas Phase Reactions 134
3.0 COMMERCIAL PROCESSES FOR PRODUCTION OF
ELEMENTAL SULFUR FROM MAGNESIUM SULFITE
OR SULFUR DIOXIDE 142
3.1 The Current Operation of the Magnesium
Oxide Recovery Process for the MgO
Scrubbing Plants 143
3.2 Allied Chemical S02 Reduction System .... 148
3.3 Asarco-Phelps Dodge Elemental Sulfur
Pilot Plant 151
-67-
-------
TABLE OF CONTENTS (Cont.)
PAGE
3.0 COMMERCIAL PROCESSES FOR PRODUCTION
OF ELEMENTAL SULFUR FROM MAGNESIUM'
SULFITE OR SULFUR DIOXIDE (Cont.)
3.4 Sulfur Production at a Pyrite Smelting
Plant 154
3.5 The Magnesium-Base Recovery Process
in the Pulping Industry 156
BIBLIOGRAPHY 159
-68-
-------
1.0 INTRODUCTION
This technical note describes the results of a
literature survey conducted to gather chemical and engineering
data pertaining to processes for producing sulfur from mag-
nesium sulfite. This information was needed to provide a basis
for subsequent thermodynamic screening studies, selection of
possible process arrangements, and calculation of process
heat and material balances.
The first major section of this document discusses
the chemistry of MgS03 decomposition and gas phase reactions
involving sulfur products and reducing agents. Emphasis is
on compilation and summary of the existing information rather
than critical evaluation. Relatively few data were found
concerning the direct reduction of MgS03. Considerable work
has been done on the thermodynamics and kinetics of the gas
phase reactions of interest.
The second major section of this document is
concerned with existing commercial processes for production
of sulfur from MgS03 or S02. Process schemes, equipment
types and operating conditions are given. It should be noted
that no existing process for direct conversion of MgS03 to
sulfur was found. Once again, emphasis in this section is on
description rather than critical evaluation.
-69-
-------
2-° CHEMISTRY OF ELEMENTAL SULFUR FORMATION FROM
MAGNESIUM SULFITE
In the first stage of this program, the potential
recovery of elemental sulfur from anhydrous magnesium sulfite
was examined considering both one- and two-step approaches.
A literature survey was conducted to identify the process
chemistry options involved in each case. The decomposition
and reduction of MgS03 will be important in either approach.
Information found concerning thermodynamics, kinetics, and
mechanisms of MgS03 reactions is presented in Section 2.1 of
this technical note. The literature was surveyed using
Chemical Abstracts from 1907 through June 1975.
It is likely that a two-stage mechanism will be most
feasible. This will be based on reduction of sulfur dioxide
resulting from MgSOa decomposition in the first step. There-
fore, identification of available information concerning thermo-
dynamics, kinetics, and mechanisms of S02 reduction chemistry
was the second goal in the literature search. Since formation
of hydrogen sulfide and other sulfur-containing gases may
occur through side reactions, this aspect is also of concern.
Because of the magnitude of information available in this
broad area, the scope of the summary presented in Sections 2.2
through 2.4 is limited to key literature covered by Chemical
Abstracts from 1967 to June 1975. Data contained in the
abstracts were relied on in many instances.
-70-
-------
2.1 Magnesium Sulfite Decomposition
Two hydrated forms of magnesium sulfite exist, the
tri- and hexahydrates. Both can occur in MgO flue gas desul-
furization systems.
Several experimental investigations of the heating
behavior of the hexahydrate have been reported. In early work,
Rammelsberg (2-1) reported a temperature of dehydration
slightly above 200°C. At the same time he noticed an evolution
of S02 due to the decomposition of the sulfite. This agrees
with the observation of Foerster and Kubel (2-2) who found a
loss of the water of hydration at 200°C with a simultaneous
decomposition of the sulfite. These authors also reported a
noticeable decomposition at 300°C without a complete loss of
the water of hydration. Hagisawa (2-3) found a continuous
transition from the trihydrate to the anhydrous salt.
Okabe and Hori (2-4) reinvestigated the decomposition
behavior in 1959 using differential thermal analysis, X-ray
diffractometry and infrared spectrometry. The DTA experiments
were performed with 500 mg samples and heating rates of 3 and
5°C/min. The atmosphere was not clearly defined in the article.
Evidently, the samples were heated in-air. The X-ray and
infrared experiments were performed after heating the samples
under the same conditions as in the DTA experiments. Figures
A2-1 through A2-3 show the results of this study. The first
three molecules of water of hydration are lost in two steps at
60 and 100°C. The X-ray pattern of the compound heated to 1003C
is remarkably different from the pattern at room temperature.
At 200aC, the last three molecules of water of hydration are lost
The DTA plot shows a strong endotherm at this temperature. The
-71-
-------
o
J;
o
«W
M
8 I
31
c <~
.? "
II
•-r c
FIGURE A2-1. DIFFERENTIAL THERMAL ANALYSIS OF
MgS03-6H20 (2-4)
.
'O 20 JO 40 &>
4000 JOOO t
-------
anhydrous phase is nearly amorphous as indicated by the X-ray
diffraction pattern at 200°C. The exothermic peak in the DTA
plot at about 480°C, as well as the endotherm at 560°C, are not
clearly interpreted by Okabe and Hori. The authors mention in
the article only that an oxidation and decomposition process
must be involved. They base this statement on the fact that
oxide and sulfate are" the decomposition products. They consider
the exotherm at 480°C as being the oxidation of magnesium sulfite,
whereas the endothermic reaction at 560°C is the dissociation of
the occluded sulfite. There are no reaction mechanisms proposed
in the article.
Foerster and Kubel used quantitative analytical tech-
niques to investigate the decomposition behavior of MgS03-6H20
The hexahydrate was dehydrated in a stream of nitrogen at 250°C,
and then heated to the desired reaction temperature. The gaseous
and solid reaction products were analyzed for S02, free sulfur,
sulfite, sulfate, and thiosulfate. The authors mention that no
sulfide was found. The results are given in Table A2-1 and Fig-
ure A2-4. The following decomposition mechanism was proposed:
(2-1)
(2-2)
MgO (2-3)
Parallel to these reactions, .they assumed a decomposition of
the sulfite,
MgS03 -> MgO + S02 (2-4)
and a decomposition of the thiosulfate,
MgS203 -»• MgS03 + S (2-5)
-73-
4MgS03 -
2MgS02 -
4MgS03 -
> 2MgS04 -
> MgS203 -
> 2MgS0lt -
1- 2MgS02
f- MgO
1- MgS203
-------
TABLE A2-1
DECOMPOSITION OF MSSO 3 AS FUNCTION OF TEMPERATURE
Heating Time One Hour
(2-2)
7
^
% Elemental
Total Sulfur
Temp . ,
°C
300
350
400
450
500
550
% S in
Undecomposed
Sulfite
84.7
75.1
69.5
67.5
52.5
3.6
% S in
Sulfate
7.3
13.7
17.0
17.0
23.3
27.9
% S in
Thiosulfate
2.5
5.2
6.7
6.7
5.7
—
Sulfur
in the
Residue
98.0
95.9
95.5
93.2
80.7
31.4
in the
Solid
Residue
3.5
1.9
2.3
2.1
0.8
-0.1
% S
as S02
in the
Effluent
Gas
2.9
3.7
5.0
6.4
12.9
54.7
•y
/Q
Elemental
Sulfur
in the
Effluent
Gas
-0.9
0.4
-0.5
0.4
6.4
13.9
% Total
Elemental
Sulfur
3.5
2.3
2.3
2.5
7.2
13.9
-------
-a
a)
w
o
a,
o
a
OJ
LOO
80
60
i C
H -H
300e
400(
500'
FIGURE A2-4
GRAPHIC PRESENTATION OF THE RESULTS SHOWN IN TABLE 2-1 (2-4)
-75-
-------
Similar techniques were used by Ketov and Pechkovskii
(2-5) to investigate the decomposition of MgS03. The magnesium
sulfite was placed in a porcelain boat and ignited in a tube fur-
nace in a stream of N2 gas. The results are given in Table A2-2
and Figure A2-5.
TABLE A2-2
EFFECT OF TEMPERATURE ON THE DEGREE OF THERMAL
DECOMPOSITION OF MAGNESIUM SULFITE IN A STREAM OF NITROGEN (2-5)
Temp . ,
°C
300
350
400
450
500
550
600
% Conversion of
SO?
0.7
2.1
5.7
16.3
36.8
68.5
88.2
MgS203
2.1
3.6
4.8
5.3
4.5
0.0
0.0
S2
2.0
2.6
3.1
3.8
5.2
8.4
3.9
S of MgS03
MgSO,
5.7
8.8
10.8
12.9
14.9
17.5
7.8
TOTAL
10.5
17.1
24.4
38.3
61.4
94.4
99.9
Gas Flow Rate: 3.0 £/hr
Test Time: 15 minutes
Sample Size: 0.5g
-76-
-------
500 600
Temperature, C
KEY
1 -
2 -
3 -
4 -
5 -
Sulfur Dioxide
Elemental Sulfur
Magnesium Thiosulfate
Magnesium Sulfate
TOTAL
FIGURE A2-5
DEPENDENCE OF THE DEGREE OF DECOMPOSITION OF MAGNESIUM
SULFITE ON TEMPERATURE IN A STREAM OF NITROGEN (2-5)
Ketov and Pechkovskii found the same decomposition products as
Foerster and Kubel. They proposed, however, a different reaction
mechanism. The sulfur dioxide is, according to Pechkovskii and
Ketov, formed by the dissociation of magnesium sulfite according
to the reaction:
MgS03 ->- MgO + SO2
(2-4)
The presence of magnesium sulfate and sulfur in the
solid products is explained by an oxidation of magnesium sulfite
by sulfur dioxide under the conditions of the experiment according
to the equation
2MgS0
S02 = 2MgSOt( + %S2
(2-6)
To prove this, magnesium sulfite was allowed to react
with 1007o sulfur dioxide for 15 minutes at 400°; it was established
that there was no dissociation of magnesium sulfite under these
-77-
-------
conditions, and the reaction products were magnesium sulfate,
magnesium thiosulfate, and sulfur, which were formed in considera-
bly greater amounts than in the decomposition of magnesium sul-
fite in a stream of nitrogen.
The sulfur liberated according to Equation 2-7 reacts
with magnesium sulfite below 500° with the formation of thio-8
sulfate according to the reaction
MgS03 + %S2 = MgS203 (2-7)
Experiments conducted on the ignition of mixtures of
MgSOa and sulfur in a stream of nitrogen showed that at 350°C,
8.97o of magnesium sulfite is converted to thiosulfate in 15 min-
utes while in the absence of sulfur, under the same conditions,
1.87» is converted.
At 550°C and above, magnesium thiosulfate is absent
from the decomposition products, since under these conditions it
is unstable. Consequently, the main sulfur-containing decomposi-
tion products of magnesium sulfite above 500°C in a nitrogen
atmosphere are sulfur dioxide, sulfur, and magnesium sulfate.
The effect of catalysts on MgS03 decomposition was also
investigated by Ketov and Pechkovskii. The presence of SiC>2 had
no measurable effect on either rate or product composition. Iron
and chromium oxides, however, increased, the reaction rate at 500°C
in a current of air. Also, higher levels of sulfate were mea-
sured in the product because of 1) enhanced direct oxidation
of sulfite to sulfate, and 2) catalyzed sulfur dioxide oxidation
to sulfur trioxide which subsequently reacted with available
MgO formed by sulfite dissociation.
-78-
-------
The decomposition of magnesium sulfite in a current
of hydrogen resulted in formation of hydrogen sulfide according
to the reaction
S2 + 2H2 j 2H*S (2"8)
The sulfur vapor is present due to two mechanisms, the oxidation
of magnesium sulfite by sulfur dioxide (Equation 2-6) and the
reduction of sulfur dioxide as shown below
S02 + 2H2 t 2H20 + %S2 (2-9)
The degrees of decomposition expressed as percent conversion of
S after 15 minutes, in nitrogen and hydrogen were not signifi-
cantly different. The addition of an iron bauxite catalyst
increased the degree of decomposition at 400° - 550°C and also
affected the gas phase composition, producing higher levels of
sulfur and hydrogen sulfide and lower S02 levels. In a hydrogen
atmosphere above 400°C the amount of MgSO^ formed decreases be-
cause S02 reduction by hydrogen is more kinetically favored than
reduction by MgS03. The amount of thiosulfate product is also
affected by a reducing atmosphere, catalyst, and temperature as
shown in Tables A2-2 and A2-3. This change is accompanied by
increased levels of H2S in the exit gases.
Kim, et al. (2-6) experimentally investigated the
kinetics of MgS03 decomposition in the range 500-600°C. The
effects of temperature, reaction time (2-22 minutes), and
concentrations of C02 (0-20%), H20 (0-20%), and 02 (-2 to 2%)
were examined. An oxygen concentration of -2% was established
by addition of 1% methane to an oxygen-free atmosphere. The
experiments were conducted by suspending a platinum reaction
dish containing ^ 300 mg of MgS03 in the reaction tube.
-79-
-------
TABLE A2-3
CO
o
i
TEMPERATURE
Temp . ,
°C
300
350
400
450
500
550
PURE AND
DEPENDENCE OF
WITH ADDITION
THERMAL
DECOMPOSITION OF MAGNESIUM SULPHITE,
OF 10% BAUXITE,
Pure
IN A CURRENT OF HYDROGEN
With Additive
% Conversion of
S0?
0.7
2.1
5.8
14.8
33.6
64.2
H2S
0.0
0.4
1.1
5.9
15.1
17.5
S2
2.1
2.5
3.1
3.1
3.4
0.3
MgS203
2.1
3.8
4.9
4.0
0.0
0.0
MgS03
5.7
9.0
10.9
11.3
10.8
7.5
Total
10.6
17.8
25.8
39.1
62.9
95.5
S02
0.6
1.8
3.2
5.1
17.1
58.0
S of MgS03
H2S S2
0.2 2.0
4.3 1.5
23.3 1.1
62.1 2.8
44.4 23.0
18.4 16.3
MKS203
2.0
2.4
0.5
0.0
0.0
0.0
MgS03
5.8
8.2
9.3
9.1
7.1
4.9
TOTAL
10.6
18.2
38.4
79.1
91.6
97.6
-------
Different temperature zones were maintained in the tube.
Following equilibration at 100°C for 30 minutes with the desired
gas mixture flowing at 500 cc/min, the sample was raised to the
200°C zone for 5 minutes where dehydration occurred. Then it
was quickly raised to the reaction zone for the specified time,
after which it was lowered to a lower-temperature zone. The
product was analyzed for Mg, total S, sulfite, and reduced
(thiosulfate, thionates, sulfide) sulfur. Sulfate was obtained
by difference.
The results of the chemical analyses of the solid prod-
ucts for each run are presented in Table A2-4. The following ob-
servations were made. The extent of decomposition increased as
the reaction time was increased, and the rate decreased as the
reaction neared completion. The rate increased with increasing
temperature over the experimental range. The decomposition was
adversely affected in an oxygen atmosphere, especially at lower
temperatures; a reducing atmosphere, however, failed to have a
significant positive effect. The effects of H20 and COa in the
atmosphere on MgS03 decomposition were small.
The kinetics of MgS03 decomposition from 500-600°C in
an atmosphere of 10% C02 , 10% H20, and 0% 02 were analyzed by a
least squares technique and expressed by Equations (2-10) and
(2-11).
g£ = (l-a)3/2 (2.338 x 10?) exp (-37.000/RT) sec"1 (2-10)
d-a)%
- 1
= 2.338 x 107 t exp (-37,000/RT) (2-11)
-81-
-------
TABLE A2-4
i
CO
K>
Reaction
time, min
7
IT
T
IT
T
IT
T
IT
T
IT
T
IT
T
IT
T
IT
22
2
12
12
12
12
12
12
12
12
12
12
Temp.
°C
526
526
5T5
5T5
526
526
5T5
5T5
526
526
5T5
575
526
526
5T5
5T5
550
550
602
501*
550
550
550
550
550
550
550
550
Compn. of
atmosphere, j>
C02 H20 02
5
5
5
5
15
15
15
5
5
5
5
10
10
10
20
0
10
10
10
10
10
10
5
5
5
5
5
5
5
5
15
15
15
15
15 15
15 15
15 15
15
10
15
10
10
10
10
10
10
20
0
10
10
10
10
1
-1
-1
1
-1
1
1
-1
-1
1
1
-1
1
-1
-1
1
0
0
0
0
0
0
0
0
2
-2
0
0
DISSOCIATION OF MAGNESIUM SULFITE
Composition of product, "b
Weight
loss, %
1*6.0
63.2
66.1*
62.8
1*5.6
1*1*. 9
61*. 2
66.1
51-9
1*9.6
62.5
61*. 9
51.2
62.9
61*. 5
65.1*
67.2
1*1.0
68.1*
3T-6
63-5
6l.O
1*9.2
1*7.0
55.1
63.1*
62.5
62.5
Mg
25-1
35-T
38.3
35.5
25.8
26.1
3T-2
3T-5
29-9
2l*.5
36.2
3T-7
26.3
32.7
35-2
36.7
3T-T
22. T
1*0.0
22.5
33-9
32.0
25.0
2l*.2
28.0
31*. 5
33.5
3!*. 8
Total
21.5
9-T
T.I
10.0
2l*.0
2l*.0
9.8
T.I
IT. 8
IT. l*
10.5
8.7
19-2
9.8
9.0
9-7
6.9
2l*.5
t*-9
22.1
10.8
11.3
IT. 8
18.7
15.5
6.1
10.1*
10.2
Sulfur
Sulfite
15.8
i*-3
1.7
1-3
19.6
16.1
2.3
l.l*
12.9
10.3
1.0
1.1*
9.9
2.0
2.1
1.0
1.1*
16.1*
o.i*
13.1*
2.2
1.5
0.9
1.2
1.7
1.6
1.8
2.1
Reduced
0.6
1.2
2.8
1.3
1.7
3.0
1.2
0.9
0.9
l.l*
l.l
2.6
1.1*
0.6
1.0
1.0
0.8
2.0
1.1
3.1
0.8
1.2
0.7
0.8
0.6
0.6
l.l
0.6
Fraction,, %, of sulfite
Decompd. Oxidized
to S02 Unchanged Reduced to S04
35-1
86^0
78.6
29.5
30.3
80.0
85.7
51*. 9
1*6.2
78.0
82.5
7T-3
80.6
80.0
86.1
18.2
• T
• 5
90.
25-
T5-9
T3.2
1*6.0
58.0
86.6
T6.5
TT-8
9-1
2.Q
57.6
1*6.8
2^8
32-T
31-9
2.1
2.8
28.5
i*.6
2.1
2.8
51*.8
0.8
1*5.2
•l*.9
3.6
2.T
3.8
l*.6
3-5
1.8
2.6
5-5
2.8
5.0
8.7
2-5
1.8-
2-3
2-3
5-2
l.l*
2.2
2.1
1.6
6.7
2.1
10.5
1.
2.
.8
,8
2.1
2-5
1.6
1-3
2-5
1.3
15.1*
8.9
5-2
15.8
7-9
11*.2
12.8
9.7
10.1
IT.6
IT.6
9
22
16
12
15-9
9.5
20.1*
6.1*
18.9
17.1*
20.1*
-J*9
52
35
.1
• 3
• 7
8.6
17-0
16.3
-------
In these equations a is the fraction of sulfite decomposed, R
is 1.987 cal/gmole °K, and T is the absolute temperature (°K).
The authors state that the reaction order of 3/2 can occur
when decomposition products interfere with the decomposition
reaction. Extrapolation of these results indicates a reaction
time of 38 seconds for 90% decomposition at 700°C or 30 seconds
for 99% decomposition at 800°C. There is some doubt as to the
efficiency of experimental gas-solid contact.
A mathematical model was developed for fluidized bed
thermal decomposition of commercial MgS03 (2-7). The effects of
coke ore gas concentration, air concentration, amount of MgS03,
and bed temperature on S02 yield were described by a set of
equations. The purpose of the investigation was to be able to
specify operating parameters of a cyclic method for sinter gas
desulfurization involving a boiling layer furnace to meet air
pollution regulations.
Schwitzgebel and Lowell investigated the thermodynamics
of the Mg-S02-02 and Ca-S02-0 systems (2-8). Predominance area
diagrams were constructed that explain the decomposition behavior
of the sulfites. Near 360°C MgS03 decomposes to MgO + SOa with
side reactions yielding sulfate, thiosulfate, and elemental S
as well up to 500° C. Although disproportion to MgS + MgS04 is
thermodynamically feasible, this reaction does not occur because
of the relatively low decomposition temperature of the sulfite,
slow disproportionation kinetics below 600°C, and the stability
of the thiosulfate up to 500°C.
Several additional references to thermal decomposition
of magnesium sulfite appear in the literature, although the
details are sketchy. In one case indirectly heating a slurry
of MgS03-6H20 (307=) plus Mg(OH)2 (0.5%) at an unspecified
-83-
-------
temperature yielded a solid phase consisting of 50% MgO and
457o MgSOi* ; the gas was a mixture of S02 and S03 (2-9) . Ninety
percent conversion of MgS03-6H20 from a flue gas desulfurization
system to MgO + S02 was achieved in an external heating rotary
kiln compared to only 10-157<, conversion in an internal heating
rotary kiln (2-10). In a similar application a mixture of an-
hydrous MgS03, MgSOi* and MgO is direct fired in a rotary kiln
or fluidized bed, again at an unspecified temperature, to pro-
duce MgO and S02. The calcination is carried out in the presence
of coke and carbon monoxide to reduce the sulfate to MgO + S02
(2-11). According to information contained in a patent, the
above calcination is carried out between 750 and 1300°C (2-12).
Berezina and Piraev determined by thermogravimetry that MgS03
decomposition begins at 380°C (2-13). The atmosphere was not
clearly identified in the abstract.
Several patents have been issued for the recovery of
elemental sulfur from MgS03-6H20 generated in flue gas scrub-
bing. In one case the MgS03-6H20 is first thermally decomposed
to MgO + S02, the latter subsequently reduced with carbon to
S, C02, and H20 at 850-900°C. The reductant is added in slightly
less than the chemical equivalent of S02 (2-10). According to
a second patent description the hydrated sulfite is dried at
200-300°C (or 70-380°F) in a gas with less than 5% oxygen con-
tent and then heated with a reductant (H2, CO, CH^ or C) at
800-900°C producing MgO + S (2-14). A possible one-step sul-
fur recovery process has also been patented wherein precipitated
MgS03 and CaSO^ is heated at 1200°C after addition of carbon.
The products are gaseous sulfur, MgO, and CaO (2-15).
-84-
-------
2.2 Chemistry of Reduction of S02 to Elemental Sulfur
In this section thermodynamic, kinetic, and other
pertinent information pertaining to the chemistry of obtaining
elemental sulfur from sulfur dioxide is summarized. Reducing
agents considered include:
methane
carbon monoxide
hydrogen
. CO + H2
carbon
coal
Each of these systems is dealt with in the following sections.
The literature was searched from 1967 through the present using
Chemical Abstracts. Literature reviews were relied on for access
to key investigations conducted prior to 1967.
2.2.1 SO2 Reduction by Methane
The use of methane as a reducing agent for sulfur
dioxide has been developed for commercial use by Allied Chemical
Corporation. A number of investigations have also been carried
out by several groups of Soviet scientists. The results of the
literature search for mechanistic and kinetic data relevant to
the reaction between S02 and CtU are presented in Table A2-5.
-85-
-------
TABLE A2-5
REDUCTION OF S02 BY METHANE
Scope of
Investigation
Reactions and/or Species
Considered Catalyst
Parameters
Method
Results and Conclusions
Temperature
700-1000°C
(1st reactor)
Ref.
Thermodynamic 2S02 + CH, = S9 '+ C09 -f 2H90
calculations up
to 1200°C.
Kinetics of two- Bauxite in
stage catalytic both reduc-
i reduction. tion steps
00
cr>
S02/CH4 = 2
Pressure:
1 a tm
Temperature:
up to 1200°C
Gas composi-
tion - 14.7%
so2
Space velocity
550 hr-1
Calculation of
equilibrium
constants up to
1200°C and equi-
librium gas mix-
ture at 800°
Experimental
two- stage
reactor
High values of the equilib- 2-16.
brium constant were calcu- 2-17
reaction will go essentially
to completion. Side reac-
tions are important. For a
S02: CH, ratio of 2 : 1 underQ
atmospheric pressure at 800 C
the possible yield of elemen-
tal S is 68.0% (2-16).
Sulfur yields were: 2-18
85% at 900°C
82% at 800°C
13% at 700°C
370"C (2nd
reactor)
Therraodynamic 2S09 -1- CH, = C09 + S9 + 2H90
calculations * z i
of SO 2 reduc-
tion with CH.,
Temperature -
25-800°C
CH, concentra-
tion - .005%
S09 concentra-
tion - .01 %
The free energy
equation for CH,
was used:
logK = ^fg^
5700
T
+ 4.83 log T
1.
T.°C
800
600
400
200
25
The following
constants were
Log K
12.0
13.1
14.8
17.9
24.3
equilibrium 2-19
calculated:
K
1 xlO-
1.2 x 10 \l
6.4 x 10!,
7.9 x 10^
2.0 x 10
- 0.0016T
+ 2.0 x 10"7T2
- 6.2
2. Side reactions that should
be considered include: the
action of H9O on Sy to form
S09 and H9S; formation of
COS and CS9.
-------
TABLE A2-5 - REDUCTION OF S02 BY METHANE (cont.)
Page 2
Scope of
Investigation
Reactions and/or Species
Considered Catalyst
Parameters
Method
Results and Conclusions
Ref.
oo
^j
i
Thermodynamic CH4 H~
calculations at cn „ _
727 and 1000°C b°2 H2°
and kinetics of H^S CS«
SOo reduction to ~ *
H,S in the 850- j£ Q
n t*Uo O/j
1000°C range L l
S02+CH4 = 1I2S+CO+H20
2 4 ~ T 2 2
S02+2H2S = 2S2+2H20
^S02+CH4 = |CS2+2H20+|C
PQ -I- 9 1-1 Q — 0 Q -l-f IJ
V*O « ~ £.i\f\ J ~ * £O rtTljn *
22 24
^S02+CH4 = |H2S+§H20+CC
S02+CH4 = is2+C02+2H2
S02+CH4 = C+is2+2H20
(1)
(2)
/"l\
V J/
(4)
:o2(5)
(6)
>2 (7)
(8)
(9)
Noncatalytic SO2/CH4 = n
and in the
presence of = 0.5-4
sum™ Pressure -latm
Temperature -
727, 1000°C
Experimental
temperature -
850-1000 C
Thermodynamics
Calculation of
equilibrium gas
phase with and
without excess
carbon present.
Experimental
Catalytic flowr
type reactor
1. The results of the equilib- 2-20
brium calculations showed
that at n=2 and 727 C, the
possible sulfur yield is
55.8%. At n=l and 727 or
1000°C zero sulfur yield
is predicted.
2. The rate of reduction of SCU
by methane in the tempera-
ture range studied is con-
trolled by the rate of
pyrolysis of methane.
Thermodynamic
calculations
Pressure - latm The calculations
or> ,„,, were based on
bUo/LH, = n
= 0.8-3.:
Sulfur yield is 1007. when the
n value is 2. At n=l the S
equilibrium con- yield is zero at t <_827°C, 3.97.
stants reported at 927 C —-* tri °°' — T"0"
in the literature.
2-21
and 60.37. at 1327°C.
Temperature -
627-1327°C
Kinetics of
catalytic
reduction at
800-1100°C
Activated SO0:CH, =10:6
Al O /I
23 Temperature -
800-1100 C
Catalyst parti-
cle size -
1.5-3 mm
Gas flow rate
Experiments were 1.
carried out in a
differential type
of reactor in a 5
ram catalyst layer.
O
Kinetically the initial 2-22
reaction rate is independent
of the flow rate, but depen-
dent on temperature and
catalyst amount. It is a
kinetic process below 950°C.
3.
(>950°) the reaction becomes
a diffusion-kinetic process.
Above 1050° it becomes a
diffusion process.
An expression was determined
to describe the kinetic phase.
-------
TABLE A2-5 - REDUCTION OF S02 BY METHANE (cont.)
Page 3
Scope of
Investigation
Reactions and/or Species
Considered Catalyst
Reduction of
industrial waste
gas with methane
over catalyst
Activated
A12°3
Parameters
Method
Results and Conclusions
Ref.
Catalyst sur-
face area -
40-60 m2/g
Temperature -
800°-900°C
Initial gas
composition -
10-12% S02
S0:CH= 10:6
Consult original
reference.
One m of catalyst yielded
300-350 kg S/hour under
experimental conditions
given.
2-23
I
oo
oo
I
Patent descrip-
tion for increas-
ing activity of
catalyst at low
temperatures
Activated
impregnated
with V20_
and K20
Catalyst com- Consult original Results were not presented
position reference. in abstract.
95.5-97% A1203
2.0-3.0% V205
1.0-1.5% K20
2-24
Mechanism of
catalytic reduc-
tion
6S02+4CH4=4COS+8H20+S2 Quartz
Also, COSTCO,S2.C02. and CS2
S02/CH4=2.2-3.4 Consult original
t = 820-1135°C reference.
Reaction shown takes place 2-25
first under conditions studied.
The COS then decomposes to CO
and S?, and CO2 and CS2. The
reaction occurs on the quartz
below 1050°C and in the gas
phase above 1050 C. A homo-
geneous-heterogeneous mechanism
was suggested.
Mechanism
and products
of catalytic
S02 reduction
8S02+6CH4=3CS2+S2+2CO+C02 Quartz
0.5-3.0 Consult original
t = 800-950°C
In contrast to Zavadskii's 2-26
results (2-25), it wag reported
that reduction takes place
with predominant formation of
CS2rather than COS.
-------
TABLE A2-5 - REDUCTION OF S02 BY METHANE (cont.)
Scope of
Investigation
Reactions and/or Species
Considered Catalyst
Parameters
Method
Thermodynamic
calculations
for the C-O-H-S
system in the
1000-1500°K
range
Species considered: Ten
CH4 CO
f2f\ U bU~
bUn HQ i
Sn COS
Prc
C00 CS0 r"
Results and Conclusions
Page 4
Ref.
H2S
Note
S, polyatomic
S, SO molecules, and
HS radicals are not
stable in the tem-
perature range con-
sidered. CS molecules
may be present in sig-
nificant quantities
above 1027°C in the
presence of excess
reducing agent.
Reactions considered:
oo
\£>
I
Temperature -
727-1227°C
=1.0-
2.5
Pressure -
0.15-l.Oatm
The equilibrium
constants of the
reactions consid-
ered were calcula-
ted based on
literature equilib-
rium constants for
dissociation of the
compounds into
their constituent
atoms. Based on
these results the
equilibrium compo-
sitions were calcu-
lated for the sys-
tem using ten equa-
tions, consisting
of the expressions
the K 's for the 6
reactions shown
plus 4 equations
Values of log K
2-27
(1)
(2)
3)
(4)
P)
(6)
Valuei of tog Ke at temperaturei 1"K)
1000
II. MM
1.5730
4.2607
-0.1543
1.6407
—0.8461
II
ixn 1 ina
1 1.2087
1.9091
3.4K8
0.0085
0.7508
—0.8262
10.7258
26358
21012
0.2193
—06078
— O.7956
10.2063
3.0834
I.I3G8
0.4184
—1.5944
-0.7729
Results wore also presented
in tabular form for the com-
positions of equilibrium mix-
tures, sulfur distribution,
and heats of reaction for the
S02-CH4 system considered.
Dependence of the equilibrium
yields of SOo, S , HoS, COS,
and CS2 on temperature SO-^CH,
ratios, and pressure were pre-
sented graphically.
Kinetics of
thermal reduc-
tion of concen-
trated S02 gases
by 'methane
2S02+CH/+=S2+C02+2H20
2S02+4H2S=4H20+3S2
2H2+S2=2H2S
C00+H0=H00+CO
222
2CO+S2=2COS
2COS=C02+CS2
Also,
CH4=C+2H2
Main reaction:
2S02+CH4=S2+C02+2H20
Intermediate and side
reactions :
CH4+H20=CO+3H2
2S02+2C=S2+2C02
2S02+4H2=S2+4H20
2S02+4CO=S2+4C02
2CO+S2=2COS
C+S2=CS2
2H2+S2=2H2S
(Ccmt . )
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(1)
/2\
v. *•/
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Tempera ture-
900-1250°C
Reaction Time-
.23-14.4 sec
Gas Feed Rate-
2.4 liters
hr'1
S02 content of
S02 containing
gas -1007. and
10-407.
Reducing gas
composition:
95-997. CH,
Balance:
hydrocarbons
S02/CHA=1.3-2.2
aerining partial
pressures of the
system.
Experiments were
conducted in quartz
flow-type reactors
in which the reac-
tion zone was de-
limited by two in-
serts formed by
sealed quartz tubes .
In general, the yield of
elemental S was maximized at
an S02/CH4 ratio of 2.0.
Sulfur yield also increased
with increasing temperature
over the range investigated
for ratios lower than 2.0.
S09^CH/.=1.9 and 1007. S00 Ras 2-28
1. The overall SO, conver-
sion and S yield were
strongly temperature
dependent:
Overall SOj S yield Reaction
t,°C Conversion, % % Time, sec
900 83 43 14.4
1100 94 81 3
1250 94 81 1.4
2. The experimental S yield was
slightly higher than the
equilibrium yield at 1100-
1250°C after 0.78-0.34 sec.
(Cont.)
-------
TABLE A2-5 - REDUCTION OF S02 BY METHANE (cont.)
Page 5
Scope of
Investigation
Reactions and/or Species
Considered
Parameters
Method
2S02+1.5CH4=CS2+0.5 CO.,
+3H20
(11)
VO
O
Results and Conclusions Ref.
2S02+1 . 5CH4=2H2S+1 . 5C
+H20
(10)
3. At 900°C the reduc-
tion is slow.
S02/CH4=1.27 and 100% S02
Gas
1. Overall SO2 conversion
and HoS yield were
strongly temperature-
dependent; see Reactions
(10) and (11).
2. The rate of reduction is
significantly higher than
at S02/CH/j ratio of 1:9,
but tne elemental S yield
is low.
S02/CH,=2.0 and Lower S02 Gas
1. Overall S02 conversion is
80-957o for reaction times
of 0.34-12.9 seconds.
2, The elemental S yield is
15-20% abs. less than SO,
conversion, and HoS yiela
is 10-207..
S02/CH4=1. 3-1.45 and Lower S02
Gas
1. SO,, conversions were 1007»,
ana S yields were SOilOI.
Kinetics
1. Satisfactory agreement was
found between experimental
and predicted values based
on the assumption that the
rate is controlled by the
rate of methane pyrolysis
up to 1250°C. Results are
tabulated below.
(Cont.)
-------
TABLE A2-5 - REDUCTION OF S02 BY METHANE (cont.)
Page 6
Scope of
Investigation
Reactions and/or Species
Considered . Catalyst
Parameters
Method
Results and Conclusions Ref.
Calculated reaction time tor the reduction of sulfur dioxida
by methane 4sec.)
or sot
Co
nctnuili
90
an of SO
.0
,.*
Temperature: 1 100* C
(K- 1.175)
0.!!
0.5
O.I.
0.9
0.05
0 'JO
3.0
7.0
10.1
13 1
O.OCi
3.0
7.2
10.3
13.4
0.97
3.1
7.1
10.7
M.I
0.9U
3.2
7.9
11.6
IS 1
2. The reaction
Contenliilwn of SO,. %
10
20
10
100
Temperaluie: 12SO*C
0.05
0.15
0.35
0.50
o.tr,
(K-
0.05
O.IS
0.35
0,51
0.67
36J)
0 05
0.15
0.37
0.53
0.70
0.05
o.ie
0.39
o.&a
0.76
time is ex-
pressed as:
t=| [(1+0 . 3N)/,Y^- -
where :
X
0 . 3Nx]
= degree of CH,
,
decomposition
fraction of
(S02+CH4) in
initial mixture.
Economics and
process descrip-
tions for SO2
reduction to S
for high temper-
ature (~1250b)
and catalytic
low temperature
(-800°) proces-
ses applied to
non-ferrous sul-
fide roaster
gases .
Dunite
Mugay Bauxite
High Clay
Bauxite
Gypsum
Alunite
Active Clay
SO,, Content of Consult
Gases - 6-100% original
Temperature- reference.
Catalytic
750-900
Noncatalytic
1250°C
CO- / pll . fn f--f fi _
O*JO/ \Jl\/. J- 43 1- JLU
1.3-2.0
Pressure - 1 atm
Catalyst contact
times -
0.07-0.97 sec.
3. If H2S is the main by-pro-
duct, a higher rate is
measured. A different
mechanism is probably
responsible .
1. Noncatalytic Tests - The 2-29
degree of SO? reduction 2-30
increased with temperature
and SOo/CH/j ratio. The
following products were
formed at S07/CHA=2.0,
1227°C, and I atm:
S 70.4%
COS 0 . 7
CS2 0.0
HoS 12.8
S02 12.1
2. Catalytic Process - Catalyst
efficiencies were determined
at various conditions listed
to the left. Activity in-
creased in the following or-
der at 900°C: dunite, Mugay
bauxite, high-clay bauxite,
gypsum, alunite, active clay.
Different orders were observed
at 800 and 850°C. Products
were: H2S, COS. CS2, and S.
-------
TABLE A2-5 - KEUUCTIOIJ OP S02 BY tJETHAHE (cont.)
Catalyst
Scope of
Investigation
Reactions and/or Species
Considered
Parameters
Method
Page 7
Results and Conclusions Ref.
Thermodynamic
calculations
for SO2 reduc-
tion by CHA in
presence of
carbon and
carbon + steam
Reactions (l)-(6) in Ref.
2-27
(7)
CVC(gr)+2H2
C(gr)+S2=CS2
(9)
Temperature -
727-1227°C
Pressure -
0.15-1.0 atm
S02/H20 -
6,12,100
1.0.2.5
Calculations
were based on
constants re-
ported in the
literature.
Methods used
were similar
to those re-
ported in Ref.
2-27.
1. The equilibrium con- 2-31
stants for Reactions
(7)-(9) are shown be-
low :
Ruction
No. (II
r
>
•
Value) of log KQ «l tcinpercurci CK>
HW
06921
S.M9B
I.OS23
,™
1.4215
S4628
o.wo
i*w
2-0884
521113
OHM}
tVM
25S2K
ioiso
O.H229
2. Equilibrium composi-
tions, sulfur distri-
butions, and heats of
reactions for the SOo-
CH4-H20-C and S02-CH4-
C systems were tabu-
lated. Sulfur distri-
butions in the first
system as a function
of temperature, pres-
sure, S02/H20 and S02/
CH, ratios are shown
graphically below.
Dapmdence of the equilibrium Jl.inbutlon of
•ulftlr tMtwven the components on temperature and the
8Ol: H^O ntflo U Zp( * I aim In reduction ot SO, by
mrtun. A) Yield (%); Bl Icmporature ("KJ. SO,:Hf)
ntl
-------
Scope of
Investigation
A.1-5 - REDUCTION OF S02 BY METHANE (cont.)
Catalyst
Reactions and/or Species
Considered
Parameters
Method
Page 8
Results and Conclusions Ref.
LO
I
3.
' Dependence of the equilibrium distribution of
•idfur between the components on temperature and the
SO,:CH, ratio at rpj = 1 atm In reduction of SO, by
carbon and methane. A) Yield (%); Bl temperature (°K>.
SO,:CH, ratio: 1) 1.0; 21 1.33; 3) 2.0; 4) 2.5. Sulfur
component!: 0 CS,; II) H,S; ml S,; IV) COS.
Although the thermo-
dynamics of elemental
S production do not
seem favorable, in-
dustrial coke plants
for S production are
feasible since they
operate under non-
equilibrium conditions.
Laboratory
investigation
of SOo reduc-
tion By natural
gas under cata-
lytic and non-
catalytic con-
ditions.
Bauxite
Reduced
Alunite
Temperature - Quartz Tube
700-1100°C Reactor
Gas Flow Rate -
12-70 ml /rain
Catalyst effect
Number of Stages -
1,2
1. The sulfur yield was
less than 40% in an
uncatalyzed system.
2. The catalytic system
was strongly tempera-
ture dependent . The
S yield with bauxite
2-32
from 28.1 to 83.3%
with temperature in-
creage from 800 to
1000 , then decreased
to 78% with further •
increase to 1100°.
With reduced alunite
an increase in yield
of 15.6 to 81.8% was
measured over the range
700 to 11OO°.
-------
Page 9
TABLE A2-5 - REDUCTION OF S02 BY METHANE (cont.)
Catalyst
Scope of
Investigation
Reactions and/or Species
Considered
Parameters
Method
Results and Conclusions
Ref.
3. Increase in gas flow
decreased sulfur
yields except for the
case of the bauxite-
catalyzed system at
1000-1100°.
4. Best results (98-997oS
yields) were achieved
with two-stage reduc-
tion under the follow-
ing conditions: 1st
stage - 900° reduced
alunite; 2nd. stage -
370° bauxite; SO,/CH4
=2.0; gas flow 10-13
ml/min.
Optimization
study of
catalytic
reduction
of S00
Bauxite
Temperature -
900°C
Flow Rate -
900-1000 hr
Fluidized
bed catalytic
reactor
-1
Sulfur yields in one- and
two-stage processes, res-
pectively, were 84% and
92%.
2-33
Catalyst Layer
Height -
50 mm (1st.
stage)
23 mm (2nd.
stage)
Catalyst par-
ticle siEe
0.5-1.0 mm
Optimization
study of
catalytic
reduction of
S00
Bauxite
S09 content—
5-30%
Bed temperature
Space velocity
Number of stages
Fluidized bed
catalytic
reactor
Optimum temperatures for all 2-34
SOo compositions were 900°
for first stage and 250° for
the second stage. The opti-
mum space velocities depended
on the S02 content:
57.
SO,
460 hr
534 hr
-1
-1
-1
30% S02 755 hr"
Total sulfur yields were 70-757.,
for one stage and 95-967» for
two stages.
-------
Page 10
TABLE A2-5 - REDUCTION OF S02 BY METHANE (cont.)
Scope of
Reactions and/or Species
Considered Catalyst
Parameters
Method
Results and Conclusions
Ref.
-. 1. - M . 1 .._-„—,.. — -._.-, _l. 1 1 — —
Literature Main Reactions :
review and 2<,Q +«j, _g +^,Q +2jj Q
thermodynamic 24222
calculations ,qf. i-jrH /.H c+Qfo +9H
for S02 reduc- tbt^-KJUi^-^b+jiA^-t-zi^
tion by methane Possible Side Reactions
2S02+2CS2=3S2+2C02
502+3H2=H2S+2H20
CH/+2S9=CS0+2H9S
4222
S02+2H2=2H20+%S2
CS0-f2H00=2H0S+C00
2222
S00H-2COS=2CO«+^SO
2. 2. L 2.
CH- +3C00=4CCH-2H00
if Z Z
, CH4+H20=CO+3H2
_n CH,-f2H90=4H9+C00
1 H Z Z Z
S02+2CO=2C02+H20
COS+H20=H2S+C02
ic H
H2+-sS2-H2S
CO+H20=C02+H2
2COS=CO2+CS2
3
2^ z^2= 2 2
co+%s2=cos
t\ \
V *•)
n (j)
u <.z,i
:
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(U)
(12)
(13)
fi/.\
V.IH;
(15)
(16)
/I 7\
(LI)
(18)
Temperature- The equilibrium constants 2-35
527-1527°C for reactions (1) and (2)
were calculated from 527
to 1527°C. Based on the
results, both reactions go
essentially to completion
above 1127-1227°C as can
be seen in the table below.
TempelKuri lx>R Kp * Kp
BOO 526.64 980.11 12.611 17.C18 4.09 1C12 .1; la"
900 626.84 1160.11 12.06S 15.289 1.16 10U .94 lt>"
1000 726.84 1340.11 11.025 33.421 4.22 111'1 .64 lo"
1100 126.84 1J20.11 11.260 11.089 1.82 1011 .71 10Jl
1200 926.84 1700.11 10.956 JO. 614 9.04 I0>0 .11 I0>0
1100 1026.64 1880.11 10.691 29.525 4.91 10>0 .35 lu"
1400 1126.64 2060.11 10.46} 28.593 2.90 10*° .92 101*
1500 1226.84 2240.11 .10.263 11.779 1.83 1010 6.01 lo"
1600 1326.114 2420.31 10.083 27.063 1.22 U10 1.16 10
1700 1426.84 2600.91 9.929 26.417 «.49 • 10 ' 2.71 > 10**
1800 1126.84 2780.11 9.785 21.871 t.10 . 10 ' 7.4* « 1«1!
* [ut« l«cul« Bp«cl«« tram JJUUiT t«bl««
Calculations of equilibrium
compositions for various feeds
show that the product mixture
contains no unreacted methane ,
neglibible CSo, and only small
amounts of COS. The balance
was elemental S (S2,Sft,S8)
H2S, C02, C.CO, H20, H2- Com-
parison of the results with
literature data show that equi-
librium can be achieved using
high temperature, low to mod-
erate space velocities, and/
or good catalysts (activated
alumina, silica gel, activated
bauxite, or Allied' s proprie-
tary catalyst as opposed to
quartz) . Formation of carbon
can be avoided with active
catalysts, or by high tempera-
ture noncatalytic conditions,
assuming space velocities which
are not too high.
Equilibrium constants for
several other reactions related
to reduction of SC>2 with CH,
were also calculated and plotted
as a function of temperature
(Exhibit 11-2 in original ref.).
-------
2.2.2 Reduction of Sulfur Dioxide by Carbon Monoxide
The reduction of SO2 by carbon monoxide has been the
subject of numerous investigations. The stoichiometry of the
main reaction was established in 1885 (2-36)
2CO + S02 = 2C02 + %S2 (2-12)
In the absence of a catalyst the reaction is slow; therefore
the emphasis of many studies of S02 reduction by CO is on
catalysis. Table A2-6 summarizes the reuslts of the literature
survey on chemistry of S02-C0 systems.
-96-
-------
TABLE A2-6
SULFUR DIOXIDE REDUCTION BY CARBON MONOXIDE
Scope of
Investigation
Thermodynamic in-
vestigation of S02
reduction systems
based on published
thermodynamic data.
Reaction and/or Species
Considered
2CO + SO2 = 2C02 + %S2(1)
CO -I- 3jsS2 * COS (2)
COS - %C02 + %CS2 (3)
S2 = VsSs = %S8 (4)
Catalyst
Parameters
Method
Results and Conclusions
Temperature;
350-1200°C
Log K and K for re- 1. Reaction (1) Calculations:
Ref.
2-19
action (1) were cal-
culated from the
free energy equation
in the range 350-
1200°C.
I
VO
t, °C
350
500
600
700
800
1000
1200
7.59x]012
1.95xl09
3.98xl07
1.74xl06
1.35xl05
2.29xl03
2.OOxlO2
Sinilar calculations were per-
formed for Reactions (2) and (3).
2. In a gas mixture of 2:1 C0/S02
(or excess SOz), the amount of CS2
formed will be negligible. Thus,
Reaction (3) was not considered in
further calculations.
3. The following gas composition was
calculated considering Reactions (1)
and (2).
TABLE IV. I'KRCKNTAGF OAK COMPOSITION AT TOTAL EQUI-
I.HIItlUM SuLFlii: COXVKKSIOX
Temp., % S
0 C. % CO: % CO % COS % SO, Conversion
1200
1000
£00
700
600
500
350
75.2
87. 0
9,1. «i
97 S
OS. j
99.4
fly . 94
10.2
7.9
2.0
0.7
0.2
00-'
<0.0l
0.3:1
0 80
i no
0 SO
o.so
0 40
0 01
s :>?
4.30
1 40
0 70
0 .id
0 . '.'0
0.02
90 5
94 5
97.5
9S S
tl'J . 4
4. In the presence of carbon, continuous
reduction of S02 by CO takes place.
Section 2.2.6 deals with this chemical
system.
Experimental in-
vestigation of
thermodynamics of
Reaction (1) in
1000-1500°C.
2CO + S02 - 2C02
Temperature:
Exp.: 1000-1200°C
Extrapolated: 1500°C
25 vol.% CO and S02
in feed.
Consult original
reference.
Equilibrium constants were experi-
mentally determined at 1000 and
1200°C. Additional constants were
calculated up to 1500°C.
2-37
-------
TABLE A2-6 - SULFUR DIOXIDE REDUCTION BY CARBON MONOXIDE (cont.)
Page 2
Scope of
Investigation
Experimental study
of catalytic
kinetics of S02
reduction.
Reaction and/or Species
Considered Catalyst
Parameters
Method
Results and Conclusions
Ref.
S02 + 2CO = 2C02
CO + %S2 = COS (2)
2COS + S02 - 2C02 + 3iS2
Pyrrhotite (FeS2)
Bochmite (hydrated
and acid-soluble
forms of alumina)
Guiana bauxite
(lightly calcined)
Activated alumina
Catalyst
Temperature: 250-
800°C.
Reactant Mixture:
63% CO
35% S02
CO generated from
coke
Contact time: 60 sec.
Quartz tube reactor. 1.
The uncatalyzed reaction proceeds 2-38
slowly even at 800°C. Pyrrhotite
is an efficient catalyst at 700°C.
At lower temperatures alumina in
slightly hydrated and acid-soluble
forms (boehmite) was efficient.
Bauxite and activated alumina were
also satisfactory catalysts.
The mechanism probably involves
formation of surface compounds of
sulfur dioxide and the catalyst.
In the 300-6QO°C range the reduc-
tion is apparently first order.
The temperature coefficient
Rd Hn k/d (1/T) varied from 14,000
kg-cal (bauxite) to 18,000 kg-cal
(pyrrhotite).
The heats of reaction are:
Reaction All
(1) 51,760 - 2.75 T + 0.0028 T2
(2) 22,500 kg-cal
(3) 6760 - 2.75 T + 0.0028 T2
The most efficient application is to
use CO (or COS) as S02 reductant as a
step subsequent to reduction by car-
bon. Optimum conversion to elemental
sulfur of 96% is predicted based on
97% catalyst efficiency and 0.994 de-
noting the conversion at equilibrium
at 500°C.
Thermodynamics of
the S02-C0 system
with respect to
combustion systems.
S02,.CO. C02, SO,
S2, O. 02, S.
Initial Reactant
Mixtures:
(1) 4% S02 - 4% CO
(2) IX S02 - 1% CO
Temperature: 1000-
2500°K
Calculations based
on JANAF data.
1. The character of the equilibrium 2-39
curves is determined by the ratio
of reactants rather than their
absolute concentration.
2. The species listed to the left
were generated. The species 0,
02, and S are abundant at the
higher temperatures.
3. The S2 mole fraction shows little
change as S02 varies from 1 to 4%.
-------
TABLE A2-6 - SULFUR DIOXIDE REDUCTION BY CARBON MONOXIDE (cont.)
Catalyst
Scope of
Investigation
Reaction and/or Species
Considered
Parameters
Experimental study
of catalytic
kinetics of S02
reduction.
2C02
2C02
2CO + S02 = %S? +
CO + %S2 = COS
2COS + S02 = V2S2
COS - %CO2 + %CS2
Proposed Reactions with
Transition Metal Catalyst:
lfxSx + %tt = %MS2
yCO + M = M(CO)y
M(CO)y + &1S2 •=
^\ + M(CO) _
CO + ^1S2 = %*S + SCO
2SCO + S02 = 3/xSx + 2C02
SCO
Iron
Alumina
Mixed Iron/Alumina
Mixed Iron/Silica
Mixed Iron Silicate/
Alumina
Red Bauxite
Surinam Red Mud
Commercial Catalysts:
Commercial alumina
catalysts containing
transition metals
Ca2SiOi, + Co
Graphite + Zn, Cu
Zeolite
Clay
Silica Gel
Iron Oxide
Diatomaceous Earth
Corundum
Laboratory-prepared
transition metal
oxide-alumina pellets
Catalyst composition
Pellet size
Temperature (350-
600"C)
Partial pressures of
S02, CO, C02, 02.
Method
Single-
pass
vertical
fixed-bed
catalytic
flow reac
tor.
Results and Conclusions
Page 3
Ref.
1. In absence of a catalyst the rate was ex-
tremely slow even at 950°C.
2. With pure iron or pure alumina present,
S02 conversion was immeasurable at 500°C.
- 3. Mixed iron/alumina catalysts result in 2^
significant conversion of S02 at low 2^
temperatures and low concentrations of 2^
S02 (usually 0-5%) and CO. Other transi- 2;
tion metal/alumina catalysts also were 2_;
effective. Synergistic effect explained
by a possible dual-site mechanism. Mixed
iron/silica catalysts were not effective.
4. Pellet size inversely affected reaction
rate for diameters greater than 1 mm.
5. At 350°C a measurable reaction rate was
observed with 41.2% iron; the rate doubled
approximately every 50° interval in the
range investigated.
6. The apparent activation energy was 18.3
Real per mole.
7. The Initial reaction rate was independent
of S02 partial pressure but was directly
proportional to the CO partial pressure;
formation of COS was also directly related
to the CO partial pressure, maximizing at
400°C. C02 concentrations less than 15%
did not affect the rate, but 02 levels
greater than 0.5% caused considerable rate
reduction.
8. Less COS was formed In the presence of an
Al20s-containing catalyst than with Fe or
Fe/Si02 materials. The COS was an inter-
mediate formed on the Fe (or other metal)
sites. It begins to form when all the iron
surface is sulfidized (i.e., converted to
FeS), with an apparent decrease in activity.
The intermediate migrates to AlaOj if
available where it reacts with chemisorbed
S02 to produce elemental S and C02.
9. Red bauxite and Surinam red mud showed
promise as commercial catalysts for re-
covery of S from waste gases.
10. Alumina catalysts were found to be activated
by sulfur and deactivated by pretreatment
with HF. This was explained by application
of Bronsted vs. Lewis acid site sorption
mechanism.
-------
TABLE A2-6 - SULFUR DIOXIDE REDUCTION BY CARBON MONOXIDE (cont.)
Page
Scope of
Investigation
Experimental investiga-
tion of catalytic con-
version of SOj by CO
O
O
I
Reaction and/or Species
Considered
Main Reactions:
2CO + S02 = %S2 + 2C02
CO -t- %S2 = COS
CO + %02 = C02
Side Reactions;
2COS + S02 = *6S? + 2C02
COS + 3/202 = C02 + S02
Possible Reactions if
H20 Present:
H20 + CO = Hz + C02
3H2 + SOZ = H2S + 2H20
2H2 + S2 = 2H2S
2H2S + S02 = 2H20 + S2
COS + HZ - CO + H2S
COS + H20 = C02 + H2S
Additional S Equilibria
S2 = Sx (x = 3,1,5,6,7,8)
Catalyst
Parameters
Screened:
S02/CO ratios
Copper on alumina*. Pressure - •*• 1 atra
cupric oxide on Temperature catalyst
alumina, silver on
alumina, molybdenum Catalyst
trioxide on silica,
alumina.
* Selected to inves-
tigate further
because it was
only one to exhibit
sustained activity
toward S02-C0 re-
action.
Method
Flow reac-
tor operat-
ing iso-
thermally
and near
atmospheric
pressure in
the absence
of oxygen
and water.
Results and Conclusions
Ref.
1. Temperatures greater than 390"C are
required to achieve 90% conversion
of S02, even at high CO levels. [Gas
composition was in flue gas concen- 2-52
tration range.]
2. At CO ratios >1 and temperatures
> 430°C reduction occurs rapidly;
e.g., contact time of 0.22 seconds.
[Note - CO ratio defined as (N^Q-
2No2)/NS02where NCO. N02 , and NS02
are upstream concentrations of CO,
02, and S02, respectively.] However,
in the range >430 to 525°C and for
>0.1 second contact time, 30% of
sulfur compounds formed are other than
elemental sulfur, mostly COS. To over-
come this, two possible approaches
exist.
a. Operate at temperatures >815°C
where thermodynamics of COS forma-
tion are unfavorable. This option
was not considered further.
b. Operate under conditions favorable to
reaction of COS and S02. [Catalytic
kinetics are higher at 5'iO°C than for
COS-CO reaction.) Possibly operate
with two beds; in first (480°C) con-
vert all S02 in two-thirds of g.-is
stream to COS and in second reactor
(315°C) COS would react with S02 in
remainder of feed stream to produce
S + CO2. Up to 97% conversion of S02
was achieved for contact times as low
as 0.18 seconds. Catalyst activity
remained constant over a continuous 30-
hour run.
Experimental study [Same as above.]
of kinetics of S02
reduction over a
copper on alumina
catalyst.
8% copper on alu- Inlet gas composi-
mlna (Harshaw Cu tion:
0803) 2000 p?m S02
3500-6500 ppm CO
14% C02
Balance Nj
Length of runs:
few - 45 hours
Contact time:
0.230 seconds
Fixed
catalyst
bed in
a tubular
flow re-
actor.
1. The kinetics of the reduction of S02 2-53
by CO can be adequately correlated
by a first order model for predicting
the dependence of S02 conversion on
process variables. The equation be-
low was derived for prediction of
ksO2 where kg<)2 represents the cata-
lyst activity expressed as
-*.n(l-Xs02)
KSU2 " 0
*so,
(CO ratio
(cont.)
(com . )
-------
TABLE A2-6 - SULFUR DIOXIDE REDUCTION BY CARBON MONOXIDE (cont.)
Page 5
Scope of Reaction and/or Species
Investigation Considered
Experimental study
of kinetics of S02
reduction over a
copper on alumina
catalyst, (cont.)
Catalyst Parameters
Space velocity;
15,000 hr~*
Temperature:
382-440°C
Method Results and Conclusions
Ref.
where:
XgQ2 = fractional conversion of inlet S02
concentration
let conditions and total catalyst
volume, sec.
ppm CO
C° rati° = S02)
= 0.9-1.6
(The effects of water,
oxygen, and nitric
oxide were not
studied.)
A" = 19.8
B (at 732°F) = 1.53
B (at 795°F) = 1.87
E/R = 2.16xlO'10R
2. The COS yield can be directly predicted
from SO2 conversion for a range of
temperatures, CO ratios, and contact times.
1
t-1
0
h-1
Investigation of [Same as above.]
the effect of
water on catalytic
S02 reduction.
8% copper on Inlet gas composition:
aluraina' 2000 ppm S02
- % CO (see CO ratio)
3.« 02
14 % CO2 •
0-9.6% H20
Balance N2
CO Ratio:
1.43-1.51 at 440°C
1.35-1.37 at 493°C
Space velocity.;
29.300-38.700 hr"1
Fixed
catalytic
bed in a
tubular
flow
reactor.
3. It was concluded that a single catalytic
bed can remove no more than 75-80% of in-
let S02 as elemental sulfur.
1. Based on thermodynaraic calculations, 2-54
at a given temperature higher CO ratios
are required in the presence of water to
convert a percentage of the feed S02 . H2S
production is favored by the presence of
water (compared to COS in a dry system) ,
lower temperatures, and higher CO ratios.
Hydrogen formation increases with in-
creasing CO ratio at a given temperature,
but decreases with decreasing temperature
at a given CO ratio. The water-gas shift
reaction is thermodynamically favored,
with resultant H2 reacting with sulfur to
produce II2S.
2. Based on experimental results, neither H2
nor HjS was detected. Thus, the water-gas
shift reaction did not proceed under the
test conditions. The water reduced the
activity of the catalyst and greatly re-
duced COS formation. The catalyst poison-
ing appeared to be reversible.
-------
TABLE A2-6 - SULFUR DIOXIDE REDUCTION BY CARBON MONOXIDE (cont.)
Page 6
I
t-1
o
Scope of
Investigation
Simultaneous re-
moval of SO2 and
NO by reduction
with CO
Reaction and/or Species
Considered
2CO + SOZ - 2C02 + %S2
CO + %S2 = COS
CO + NO - C02 * %N2
2COS + S02 = MiSa + 2C02
Catalyst
Parameters
Copper on alumina
Silver on alumina
Palladium on alumina
Manganese on silica
gel
Silver on silica gel
Copper on silica gel
Ruthenium on alumina
Iron on alumina
Chromium on alumina
Iron/chromium/alu-
mina
Simulated flue gas
Composition (2-57):
2000 ppm S02
•V6000 ppm CO
0-1000 ppm NO
14% C02
Balance N2
Temperature:
440°C
Space velocities:
^10* hr~l
Catalysts
Method
Fixed
catalyst
bed In
vertical
tubular
flow
reactor.
Results and Conclusions
1. Metals on alumina were the most active
catalysts for reduction of SO2 alone.
Copper on alumina was tested extensively.
2. Simultaneous reduction of NO and S02 by
catalyzed reaction with CO is possible.
Thermodynamlc calculations showed that
optimum initial ratio of CO to S02 Is
slightly less than 2 to minimize COS for-
mation. Essentially complete reduction
of NOX is possible.
3. The copper-alumina catalyst was less active
for the combined reduction than for either
(i.e., HO or SO2) separately.
4. Maximum S02 conversion (to either COS or
S) is limited to 75-80% in a one-bed
process.
5. The Cu-, Fe-, and Cr-alumina catalysts
were all active in the combination reduc-
tion, but an Iron/chromlum/alumina cata-
lyst was ten times as active.
Ref.
2-55
2-56
2-57
2-58
Laboratory inves- CO + NO = COz + %Nz
"Ration of dual- =
bed catalytic ' 2
reduction of CO + %S2 = COS
simulated flue 2COS + SQ^ __ 3^s^ + 2CQ^
gas.
Optimization study
for catalytic
reduction of S02
in one- and two-
stage processes.
Mechanistic study S02 + 2CO - 2CO2 + Sx
of catalytic
reduction of S02
Iron-chromia in Bed temperature:
flrst bed' 630-700-F
Activated alumina _ . . _ . ,
... Composition of inlet
in second bed .
gas to second bed
(i.e. , ratio of COS
to SOj)
Aluminum oxide Number of beds
Bed height
Gas flow rate
Catalyst particle
size
Temperature
Aluminum oxide Reaction time:
40-320 sec.
Vertical
tubular
flow
reactor
containing
two succes-
sive cata-
lyst beds.
Fluidized
catalyst
bed
reactor.
Flow
reactor
with cata-
lyst de-
posited
in thin
layer on
walls .
(cont.)
The dual-bed concept was successful in
achieving 90% conversion of S02 in simulated
flue gas to sulfur at temperatures <370°C
and high space velocities (at least 20,000
hr'"1 on the combined bed). Proper COS to S02
ratio entering the second bed (stoichiometric
ratio of 2) can be effected by controlling
the catalyst bed temperature.
The optimum parameters were: 600°C; flow
rate - 365-570 hr"1 ; bed height: first
stage - 65-70 mm, second stage - 20 mm;
catalyst particle size - 0.25-0.50 mm.
The sulfur yield in the one-stage process was
90-1% and 94% in the two-stage reactor.
1. Reaction rate becomes appreciable- at 450°C
and above. Retardation results with SOj
increase, while CO Increase causes faster
velocity.
2. At initial S02:CO ratio of 1:3 and 600°C,
30% conversion of react ants was measured.
(cont.)
2-56
2-59
2-60
-------
TABLE A2-6 - SULFUR DIOXIDE REDUCTION BY CARBON MONOXIDE (cont.)
Page 7
o
(jO
l
Scope of Reaction and/or Species
Investigation Considered
Mechanistic study
of catalytic
reduction of SOj
(cont.)
Experimental study Refer to Table 6 in
of catalytic re- original reference.
duction of SOj
Thermodynamics and SO2 + 2CO = %S2 + 2C02
kinetics of cata-
lytic S02 reduction
by CO in the system
SOz-CO-C02-N2 for
SO 2 removal from
combustion furnace
exhaust.
Catalyst
13 catalysts in-
cluding:
Cu-Al203 (various
ratios)
Cu-Cr203
Fe-Al203
Fe-Cr203
Co-Al203
Ni-Al203
Cu-Cr203-Al203
Cu-Fe-Al203
Cu-Cr203-Al2O3
Activated alumina
Alumina + iron
Alumina + silver
Alumina + copper
Alumina + calcium
Alumina + magnesium
Bauxite
Parameters
Reaction tempera-
ture:
300-550'C
Inlet gas composi-
tion:
2% SO2
4-8% CO
0-3.2% 02
0-6.6% H20
Balance N2
Temperature:
100-900"C
Contact time:
0.02-0.30 sec
S02 Concentration:
0.17-0.19%
CO concentration:
0.4-1.6%
C02 concentration:
0-15%
Gas flow rate:
0.25-1.5 H-min""1
Amount of catalyst:
0-20 cm3
Space velocity:
1000-3000 hr~'
Method
Chroraa tographic
analysis of
initial, inter-
mediate, and
final components:
CO, SOZ, C02,
COS, but not S.
ESR spectrum of
SO radical.
Flow-type reac-
tor with a fixed
catalyst bed.
Fixed-floor
apparatus con-
sisting of
vertical quartz
reaction tube
and gas mixing
apparatus.
Results and Conclusions Ref.
3. The SO radical was detected in the
heterogeneous-homogeneous reduction
of S02.
1. In the absence of HzO and 02, the 2-61
copper on alumina catalysts were
superior. Reduction in Cu content
led to reduced amounts of COS.
2. In the presence of HzO and Oz, the
following were observed 1
. temporary catalyst poisoning by
water vapor;
. catalysts readily deactivate due
to structural changes - possibly
formation of sulfides or sulfates;
. possibility of side reactions
(none actually proved) .
1. The maximum conversion of S02 was 2-62
obtained at 400-500°C using bauxite
as a catalyst.
2. The catalysts containing oxides of
metals such as Fe, Cu, Ca, and Ag
in each alumina were highly active.
3. The equilibrium conversion of SOj
increased with increasing temperature
and CO concentration, and decreased
with increasing COz concentration.
4. The reaction rate in the presence of
bauxite had first order dependence
with respect to SOz and CO concentra-
tions, but was adversely affected by
COz above 600°C.
-------
2'2'3 Reduction of Sulfur Dioxide with Hydro<
;en
An early explanation of the reduction of sulfur
dioxide by hydrogen is shown below (2-19) .
S02 + 2H2 2 2H20(g) + %S2 (2-13)
H2 + %S2 2 H2S (2-14)
+ 3/2S2 = 2H2S + S02 (2-15)
The author assumed S2, S6, and S8 to be the vapor phase sul-
fur species, although a later worker presented evidence for the
S3, Si,, S5, and S7 molecules as well (2-63) . The equilibrium
constants for reactions (2-13) , (2-14) , and (2-15) were calculated
from 300-1100°C based on free energies (Ki , K2 , and K3, respec-
tively). These are summarized in Table A2-7. The calculated
equilibrium ratio of S:H2S was approximately four at 325°C
assuming an initial mixture of two moles of water and one-half
mole of S2 .
Inclusion of all the gas phase sulfur species in
the model leads to eight independent reactions describing the
system (2-64) . The reactions and the respective equilibrium
constant equations are shown in Table A2-8. Murdock and
Atwood computed the equilibrium compositions and sulfur yields
over the temperature range 300-1100°C and at several nonstoichio-
metric feed compositions in the presence of nitrogen diluent.
Formation of elemental sulfur was favored at lower temperatures
although significant quantities of hydrogen sulfide were pre-
dicted even at low temperatures . The results are shown
graphically in Figure A2-6. Calculations of thermodynamic
equilibrium conversions of hydrogen and sulfur dioxide, and
hydrogen sulfide and sulfur yields at various feed compositions
-104-
-------
TABLE A2-7
EQUILIBRIUM CONSTANTS FOR THE
SYSTEM SOz-HzS-HzO-Hz-Sa
c."Loe K> fa " LotK' x* Lo*K>
1100
900
700
500
300
113
3.
5
7.
11.
,53
.51
.85
,82
10
3.4
3.2
7.1
6.6
1.3
X
X
X
X
X
10'
10'
10'
10"
O.S7
1.47
2.23
3.49
5. SO
7.4
29.5
191
3.1 X
4.0 X
10'
10*
-1
-1
-1
-0
0.
1
.77
.57
.29
.84
.00
.30
1.16 X 10-*
2.69 X 10-*
5.13 X 10-'
1.45 X 10-'
1.0
20.0
TABLE A2-8
EQUATIONS USED TO CALCULATE THE EQUILIBRIUM
CONSTANTS FOR THE REACTIONS IN THE
SYSTEM AT DIFFERENT TEMPERATURES
Reaction Equilibrium constant (T in °K)* Reference
2
3
4
5
6
7
8
9
1'AS, ss S,
2S, =iS,
2'AS. PS S,
3Si — S«
3'ASa ^ Sr
43,^3,
H,0 + V«S« ss H, + V,SO,
Hi + VsS, ri H,S
K, - exp(13.3 - O.Ol88T)/fiT
/f, - exp<23.2 - o.o;3G7T)/;er
K, - e:ep(47.8 - 0.0557T)/HT
Kt - exp(66.3 - 0.0751T).RT
Ki - exp(78.9 - 0.08937/)/fiT
^•, - exp(96.8 - 0.1103T)/.fir
/T, - exp(-7944/T - 0.50661nT + 1.75 +
1.525 X 10-jr - 2.648 X IQ-'T5)
K, - exp(19.4 - 0.00771 T In T + 1.30 X
10 -*T' + 0.0125T)
Detry, rt ai. (1067)
Detry, rt al. (1967)
Detry, et ai. (1967)
Detiy, et al. (1967)
Detry, et al. (1967)
Detry, et at. (1967)
Doumani, ef al. (1944]
Kelley (1937)
• R is 0.0019869 kcal/g niol °K.
-105-
-------
Is
35
i
Sol
•flELO
- CatcuUW. T>.\ -at. (2-64)
• t»w (2-19)
*oov 500' soo" roT soo 900
TEMPERATURE, -c
COO HOC
FIGURE A2-6 - EQUILIBRIUM GAS COMPOSITION FOR THE
REDUCTION OF SULFUR DIOXIDE WITH
HYDROGEN AT 760 mm Hg PRESSURE:
D, H2 (MURDOCK. 1973); 0, S02 ;
A, H2S; +, H20; x, SULFUR YIELD;
-, LEPSOE (1938).
-106-
-------
as presented in Table A2-9 indicated that sulfur dioxide
concentration should be maximized in order to achieve the
highest sulfur yields. Increasing temperature over the range
345 to 390°C was found to have a negative effect on sulfur
yield. Thennodynamically, almost complete conversion of sul-
fur dioxide is possible at temperatures below 400°C.
Kinetic investigations were conducted to define the
reaction mechanism and to determine if hydrogen sulfide forma-
tion could be suppressed to further increase sulfur yields
(2-64) . The effects of catalyst to feed ratio, flow rates, and
temperature were studied in the presence of an activated bauxite
catalyst. Significant levels of conversion of S02 to elemental
sulfur were observed at reaction temperatures in the 300-400°C
range. The rate of the primary reaction route (Equation 2-13)
was found to be independent of the sulfur dioxide concentration
and first order with respect to the hydrogen concentration, ex-
pressed as:
rS02
where :
kg = 0.014 ± 0.001 mol-hr"1 (g of cat.)'1 atja
at 375°C
The rate of H2S formation can be expressed as
H2S
where :
= 2.9x10"* ± 0.3xlO~' mole-hr'1 (g of cat.)"1
a tin l at the same temperature.
-107-
-------
TABLE A2-9
THERMODYNAMIC EQUILIBRIUM CONVERSIONS OF HYDROGEN
AND SULFUR DIOXIDE AND THE YIELD OF HYDROGEN
SULFIDE FOR THE H2/S02/SX/H2S/H20/N2 SYSTEM
•AT 375°C and 1 ATM
Peed
composition,
mole %
H,
2.67
5.33
8.0
2.67
5.33
8.0
2.67
5.33
8.0
SOj
1.
1.
1.
2.
2.
2.
4.
4.
4.
33
33
33
67
67
67
0
0
0
Product
Conversion
H,
100.0
74.9
49.9
100.0
100.0
99.9
100.0
100.0
100.0
SO,
90
100
100
48
90
100
32
63
91
.1
.0
.0
.2
.7
.0
.5
.8
.4
2
1
1
9
4
2
7
2
6
Mole
fraction
H,S
.73
.33
.33
,51
.87
.65
.07
.22
.90
X
X
X
X
X
X
X
X
X
10-
10-
10-
10-
10-
10-
10-
10-
3.
5.
5.
12.
3.
5.
17.
10.
10- 4.
Ratio
S,:H2S
33
18 X 10-'
79 X 10-'
5
98
81 X 10 -'
4
5
3
-108-
-------
Rate expressions for the other species were also derived.
Comparison of experimental and calculated SOz. conversions
agreed well over the entire range, 0 to 9870 conversion; how-
ever, the predicted yields of hydrogen sulfide were lower
than experimental yields for w/N values above 11.8 (w = grams
of catalyst, N = moles/hr of feed). The temperature dependen-
cies of the overall rate constants, k^ and k^ ^, were
experimentally determined and correlated using the Arrhenius
equation.
A reaction mechanism was proposed which was shown
to be consistent with observed initial kinetics. The mechanism
involved oxidation, reduction, desorption, and regeneration
of sulfur sites.
Non-catalytic reduction of sulfur dioxide reduction
by hydrogen was investigated at 700-900°C in a quartz tube
reactor (2-65). Sulfur yield increased from zero to 20.7%
over this range. The initial gas composition was 10 ml S02 ,
10 ml 02, 80 ml N2, and 50 ml H2. The sulfur yields improved
significantly in the presence of bauxite catalyst. At 800°C
and 160 ml/min flow rate the yields were 56 and 8670 respectively
for a one- and two-step reduction. The catalyst also prevented
the formation of hydrogen sulfide below 900°C; this is com-
pared to a 12% yield at 800°C in uncatalyzed tests. Additional
catalysts were later studied at lower temperatures (2-66) .
The best sulfur yield obtained was 98-100%> over reduced alunite
at 600°C and 54-90 ml/min gas flow rate.
Shakhtakhtinskii and co-workers (2-67) reported that
near complete conversion of S02 was possible by hydrogen at
600°C in a steel reactor in contrast to the results obtained
with the quartz reactor described above. Preheating the S02
and H2 separately to 350°C made much lower reactor temperatures
feasible (300°C) without a loss in sulfur yield.
-109-
-------
2.2.4 Reduction of Sulfur Dioxide with CO + H2
Shakhtakhtinskii and co-workers have reported a
number of studies of S02 reduction by converted natural gas.
Catalyzed and non-catalyzed experiments were conducted.
Information contained in Chemical, Abstracts since 1967 is
summarized in Table A2-10. No data on kinetics or mechanisms
were available.
2.2.5 Sulfur Dioxide Reduction by Coal
One reference to the reduction of S02 with coal
was found in the recent literature. This gas purification
process involves reaction of the humidified gas with coal at
temperatures >. 425°C (2-75) . A high sulfur content coal may
be used. No additional information was available in the
abstract of this patent.
-110-
-------
TABLE A2-10
REDUCTION OF S02 WITH CO + H2
Reactant Gas
Description
Roaster gas - S02
content not
specified.
Reformed natural
gas - chiefly H2
and CO.
Reducing gas :
roaster gas volume
ratio =2:1
Reducing gas :
H2- 71.20-73.52%
CO - 22. 72-26.407.
CH,, - 0.80-3.107.
S02 gas - composi-
tion not described
in abstract.
14 and 207. SOZ gas
mixtures.
Converted natural
gas.
7 and 207. S02 gas
mixtures
Converted natural
gas with 5-207.
residual CHi, .
757. S02 gas
Converted natural
gas.
No. of
Stages
Two
One
One
One
Two
Two
Two
Experimen-
tal tech-
nique was
not de-
scribed in
abstract.
(Not clearly
specified in
abstract.)
One
Experimental Conditions
Catalyst
Reduced alunite
in first stage;
bauxite in
second stage.
A1203
Bauxite
Bauxite
A1203
Bauxite
Bauxite (flu-
idized bed)
Bauxite (flu-
idized bed)
Bauxite
Temperature Space Velocity
(°C) (hr-1)
800 Not given in abstract.
600 See remarks .
600
800
450-700
450-700
400 434 (147. S02 gas)
(1st stage)
200-350 461 (207. S02 gas)
(2nd stage)
450-500
500-700
>700
500 1040-2075
500 >2075
600-700 2270
S Yield
Ci)
Not given
in
abstract.
87
< 78
82
97
97
97-98
Not given
in
abstract.
<85
85
80-81
86
<86
Optimal
Remarks
S yield decreased at
higher temperatures be-
cause of increased H2S
formation.
Optimal conditions -.
. Space velocity -
630-740 hr-1
. Catalyst bed depth -
60-80 mm
Sulfur yield decreased
with increased residual
CHu levels and increased
with increasing tempera-
tures .
Ref.
2-68
2-69
2-70
2-71
2-72
-------
TABLE A2-10 - REDUCTION OF S02 WITH GO + H2 (cont.)
Gas Composition
75% S02 gas
Converted natural
gas (cont.)
S02 gas - composi-
tion not described
in abstract.
Converted natural
gas.
50-1007. SO2 gas
mixture.
Converted natural
gas
Experimental Conditions
No. of Temperature Space Velocity
Stages Catalyst (°C) (hr~')
Two Bauxite 250-450 Not given in abstract.
One Steel reactor 800 Not given in abstract.
(optimum)
400
One A1203 (flu- 450-600 1100
idized bed)
One Natural bauxite 450 1100
Two Bauxite 450-500 1120
(1st stage)
450
(2nd stage)
Page 2
S Yield
(%) Remarks Ref.
96-97 The following reaction
(S yield occurs in 2-stage cataly-
from reac- sis in this temperature
tion in range:
Overall's 2HjS + SO, + 1.5 S2 + 2H,0
yield not
given . )
"Most" Gas flow rate: 0.0004 m/sec 2-73
96-97 Operation at lower temperature
possible with preheating of
initial gases to 400°C.
Not given 2-74
in ab-
stract.
Not given
in ab-
stract .
Not given
in ab-
stract.
-------
2.2.6 Sulfur Dioxide Reduction by Carbon
Studies of the reduction of S02 by various forms of
carbon have been reported in the literature. These studies
are summarized in Table A2-11. Mechanisms have been suggested
involving formation of carbon-sulfur and carbon-oxygen bonds.
Lepsoe (2-19) reported that in the presence of carbon, con-
tinuous reduction of S02 takes place through the following
reaction scheme:
2CO + S02 - 2C02 + %S2 (2-16)
C02 + C = 2CO (2-17)
S02 + C = C02 + %S2 (2-18)
Thus the chemistry of this system is related to that discussed
in Section 2.2.2.
-113-
-------
TABLE A2-11
-P-
i
REDUCTION OP S02 BY CARBON
Reducing Agent
Carbon (coke)
Carbon
Coke
Charcoal and
Coke
Medium
Activated
Coconut Shell
Charcoal
Scope of Reactions and/or Species
Investigation Considered
Kinetics of S02 S02 + C = C02 + %Sjt (1)
reduction by C,
CO, and COS. C02 + C = 2CO (2)
S02 + 2CO - 2CO2 + %SZ
(3)
Study of carbon-
oxygen system.
Investigation of CO, C02, S02, CS2 , COS
kinetics and
mechanism of SOj
reduction by
coke under con-
ditions similar
to copper smelt-
ing
Experimental in-
vestigation of
S02 reduction
to S and CS2
Experimental study SOz + C - COz + %S2 (1)
of chemisorption co „ 2CQ (2)
of SOj by and re-
generation of SO2 + 2CO = 2COa + %S2
Catalyst Parameters
Temperature:
850-1200°C
Contact time:
1.5-22.0 min
10 and 100% coke
in reduction layer.
Gas-coke contact
time: ^ 2 sec
Reactor tempera-
ture:
Initial - 1300'C
Final - 900°C
Sorption tempera-
ture: 50-650°C
Regeneration at
hieher tenioera-
Method
Vertical quartz
reaction tube.
Consult original
reference.
Analyzed gas pro-
duct composition
in apparatus sim-
ulating conditions
of sulfide ore
smelting.
Integral flow re-
actor with a
fixed charcoal or
coke bed .
Packed bed flow
reactor.
Results and Conclusions
An expression for the rate of C02
formation between 900 and J200°C
was derived:
(C02) = l.lKSOz)""' - (S02)]
where ( ) denotes moles in the re-
action product. Above 1200°C the
rate of SOi reduction was apparently
diffusion controlled.
The rate of reaction (2) above is
insignificant at temperatures of
250-30Q°C, even up to 700°C. There-
fore this mechanism cannot explain
CO formation observed (2-38) .
Under the conditions tested 78-80%
reduction of SO2 was measured.
Product analysis was not reported
in the abstract.
An induction period of about three
hours was observed before conver-
sion of S02 to CS2. Elemental sul-
fur also was produced.
1. Between 50 and 300°C, the
following observations were
made. In the absence of oxygen
and water, carbon has limited
Ref .
2-38
2-76
2-77
2-78
2-79
active carbon to
determine mechan-
ism and effects
of temperature and
CO.
(3)
tures (typically
9SO°C for 90 min
in He flow)
Effect of CO in
inlet gas (2%)
Gas flow:
200 cm3/min
Effective linear
flow:
153 cm/tnin
(cont.)
S02 sorptton capacity. Cheml-
sorption occurs on only 1% of
BET surface area, and is Inde-
pendent of temperature In this
range. Physical adsorption
decreases from 3% to 0.3Z be-
tween 50 and 150°C and is neg-
ligible above 250°C. Analysis
of regeneration effluent indi-
cated that regeneration ot char-
coal after sorption Jn the lower
range (50-300°C) occurs through
(cont.)
-------
TABLE A2-11 - REDUCTION OF S02 BY CARBON (cont.)
Page 2
Reducing Agent
Medium
Activated
Coconut Shell
Charcoal
(cont.)
Scope of Reactions and/or Species
Investigation Considered Catalyst Parameters
Method
Average residence
time: 3.5 sec
Inlet gas composi-
tion:
0.5% S02 in He
Ui
I
Results and Conclusions Ref.
reduction of chemisorbed S02 to
elemental sulfur. Results are
tabulated below.
2. Catalyst thermal regeneration was
'accompanied by weight loss attri-
buted to carbon conversion to
oxides, especially at temperatures
>500°C.
3. At 650°C the reduction between SO2
and carbon took place rapidly,
yielding CO, C02, and S. Signifi-
cant amounts of S left the bed and
condensed in cooler parts, leaving
active sites for further S02 reduc-
tion.
4. The presence of CO In Che inlet gas
had an inhibiting effect on oxygen
complex formation on the carbon
surface. At 550°C evidence for
reaction (3) was found, but at lower
temperatures, e.g., 350°C, exten-
sive formation of COS occurred. Re-
sults are shown below.
MATERIAL BALANCES ON REACTION SYSTEMS SO.-C AND SO.-CO-C
Time
product
Gaseous reactants, moles X 10-'/rnin sample
taken
Temp. SO, CO Min
Gaseous products, moles X lG~'/mm
SO,
CO
CO,
COS
650
600
550
5SO
500
350
4.5
4.5
4.5
4.5
4.5
4.5
nil
nil
17.0
17.0
17.0
17.0
420
130
15
195
240
330
nil
nil
0.4
0.4
nil
nil
i.O
0.5
8.9
8,2
6.G
4.0
3.8
1.8
8.6
7.7
10.6
9.6
nil
nil
0.2
0.4
1.2
J.4
-------
TABLE A2-11 - REDUCTION OF S02 BY CARBON (cont.)
Reducing Agent
1. Medium
Activated
Coconut
Shell
Charcoal
2. Bituminous
Coal Char
Scope of
Investigation
Experimental in-
vestigation of
simulated flue
gas and carbon
between 500 and
800°C.
Reaction and/or Species
Considered Catalyst
Side Reactions:
HaS + CO = COS + Hz (1)
HiS 4- C02 = COS + H20 (2)
Mass spectrometer analyzed
exit gases for C02, HjS,
CO, COS, SOZ, Hz, H20, and
He.
Parameters
Method
CTv
Run time:
350-500 min
Temperature:
500-800°C
Carbon forms
Inlet gas composi-
tion:
0.35% SOz
2.3% H20
15.8% C02
3.22 02
Balance helium
Quartz fixed-bed,
flow tube system
using simulated
flue gas. Gas
products analyzed
by mass spectro-
metry.
Page 3
Results and Conclusions Ret.
1. With the coconut shell charcoal 2-80
at 600 and 800°C the exit gas
was Cree of S compounds ioc a
certain period of time, which
was longer at the higher tempera-
ture. Then H2S and COS broke
through in both cases, although
their profiles were quite dif-
ferent. An explanation for this
phenomenon was based on Increasing
reaction rate between carbon with
H2S and COS with Increasing tem-
perature .
2. In the bed itself a C-S surface
complex was formed. This complex
has a high thermal stability.
3. Wliile the sulfur content of the
bed at tbS-COS break-through was
3.4 times higher at 800 than 600°C,
the residence time of the gas in
bed at breakthrough decreased by
40% going from 600 to 800.
A. S0j did not appear in the exit gas
until 200 mln at 500°C and 500 min
at 700°C. At 800°C, S retention on
the bed exceeds at least 11% before
SOj break-through occurs.
5. Reactions (1) and (2) were shown
to account for the observed HjS:
COS ratios during the experiment.
6. The results of the bituminous coal
char reaction were more complex.
Consult original article Cor detaila.
7. Carbon was found to be a poorer
catalyst for oxidation of COS to
elemental S than tor HzS oxidation.
8. When the exit gas of the reduction
reactor (containing slight excess O2)
was passed through a second carbon
reactor at 100°C, no sulfur com-
pound was measured in the effluent
even after four hours, Indicating
that conversion of COS and 1128 to
S was occurring.
-------
TABLE A2-11 - REDUCTION OF S02 BY CARBON (cont.)
Scope of Reaction and/or Species
Reducing Agent Investigation Considered Catalyst
Wood Charcoal
Pellets
Thermogravimetric
study of kinetics
of catalytic and
noncatalytic re-
duction of SO2 on
charcoal in range
615-940°C.
Na2C03
Parameters
Method
Temperature:
615-940°C
SOz feed concen-
tration
Catalyst
Therraogravimetric
apparatus consist-
ing of chainomatic
analytical balance
and tubular furnace.
Results and Conclusions
1. No single controlling step was
found for controlling the reac-
tion under all experimental
conditions. However, a promi-
nent influence by chemical reac-
tion was shown. An integral rate
equation, assuming chemical reac-
tion control, was proposed:
radat = A exp (-E/RT) V^3 t
where:
r0 = initial radius of pellet
do = density of pellet
f = fractional thickness of
pellet (related to frac-
tional conversion of carbon)
A = frequency factor
E = activation energy
T = temperature
P = partial pressure of S02
t = time
2. Values of A and E were determined:
Moncatalytic:
Page
Ref.
2-81
A = 104 g cm"2 min
E = 19,870 cal mol"
Catalytic:
A = 1.12 g cm 2 min
E = 10,200 cal mol~
Coke
Parametric study
of SO2 reduction
over hot carbon
surface, with
emphasis on CS2
yield.
NazC03
Bed temperature:
920 and 1000°C
Catalyst concentra-
tion:
Particle size -
8-11 British
sieve standard.
Pretreatment of
coke.
Reaction time:
200 min.
S02 concentration:
72%.
Not described in
abstract.
The CS2 yield increased with:
. increasing bed temperature,
. in presence of catalyst,
. decreasing particle size,
. coke pretreatment,
. increased reaction time.
2-82
-------
TABLE A2-11 - REDUCTION OF S02 BY CARBON (cent.)
Reducing Agent
Carbon
Scope of
Investigation
Thermodynamic
calculations of
S02 reduction by
carbon in presence
of HaO in range
727-1227°C, empha-
sizing CS2 yield.
Reaction and/or Species
Considered
Overall reaction at 1300-
1500"K:
5C + 2S02 = CS2 + 4CO
Catalyst
Parameters
Method
Temperature:
727-1227°C
S02:H20 ratio:
6-100
Pressure:
0.15-1.0 atm
Thermodynatiiic cal-
culations based on
method described in
Reference 2-27.
OO
l
Results and Conclusions
Equilibrium yield of CS2 was
greatest (70-80%) at 1027-1227°C.
The distribution of sulfur be-
tween components in the equili-
brium mixture at 1 atm is shown
below.
PageS
Ref.
2-31
Dependence of the equilibrium distribution of
sulfur between the components on temperature and the
SO2: H2O ratio at Epj = 1 atm in reduction of SOZ by
carbon. A) Yield K)\ B) temperature (°K), SO2:II2O
ratio: 1) 6; 2) 12; 3) 100. Sulfur components: r) CS,;
II) Sa; ni) COS; IV) H2S.
3. Equilibrium yields of elemental
sulfur and H2S were low.
Carbon
(several
commercial
samples)
Kinetics of reac-
tion system C-SO],.
Influence of compo-
nents of ash:
Si02
A1203
MgO
CaO
Fe20,
Pulsation reaction
chromatography in
a microreactor.
Results were not described in 2^-83
abstract.
Combination o£
thennogravimetry
and gas chromatogra-
phy to investigate
solid phase in relation
to ash composition.
-------
TABLE A2-11 - REDUCTION OF S02 BY CARBON (cont.)
Page 6
Reducing Agent
Carbon
Scope of
Investigation
Thermodynamics of
C-S02 system.
Reaction and/or Species
Considered
Reaction products con-
sidered :
CO
C02
COS
CS2
S2
Catalyst Parameters
Temperature:
527-1227°C
Effect of Hz, C02,
02, H20.
Method
Calculations based
on minimization of
free energy of the
system.
Results and Conclusions
1, Equilibrium composition of the
base reaction mixture (C~S02)
was calculated.
2. The equilibrium composition of
a simulated waste gas was pre-
sented. Removal of S02 as ele-
mental S was judged thermody-
namically feasible.
get .
2-84
3. The presence of N2O is highly
undesirable because of H2S
formation.
Metallurgical
Coke and Coal
Char
VO
I
Define chemistry
of S02 reduction
by carbon at
temperatures
near 1200°C and
space velocities
of 700-800.
Temperature:
VL200-1760°C
Space velocities:
700-800 hr~'
Inlet gas - S02
In N2
DTA
TGA
bed reactors.
1. In the fixed bed reactor the 2-85
following results were reported:
It'/mlTi'
0.50
0.0
0.0
1.0
CHSV
IIIJ
3'Jl
3li
013
l..pu,. %
SO,
100
HI'J
•s
N;
sr
A.fi.
tua
2 ion
22JO
25o0
2000
2-IW
Ontpur, %
COj
-------
TABLE A2-11 - REDUCTION OF S02 BY CARBON (cont.)
Reducing Agent
Metallurgical
Coke and Coal
Char (cont.)
Scope of Reaction and/or Species
Investigation Considered Catalyst
Parameters
Method
Results and Conclusions
3. The reaction was heat balanced
with addition of air to S02
stream and substitution of coal
char for metallurgical coke.
4. The results of rate studies
showed that 50% of the graphite
is reacted within 15 minutes at
1260°C. Faster rates were ob-
served with char than with graphite.
The kinetics are highly tempera-
ture dependent.
Page 7
Ref.
Low Volatile
Char
Experimental ln~
investigation of
SOj-carbon sys-
tem at 800-950°C.
O
I
2C + 2S02 - 2C02 + S2 (1)
C + S2 + C02 = 2COS (2)
C + 2COS = CS2 + 2CO (3)
C + C02 = 2CO (4)
C + S2 = CSj (5)
Feed gas - 10% SO2
in N2.
Feed gas flow
rate:
40-120 ml/min
Temperature:
800-950°C
Carbon mass:
4 and 10 g
Reactor pressure:
1 atm
Vitreous silica 1.
flow reactor in
tube furnace.
Effluent gas
analyzed for SOj, 2.
C02, CO, CS2, COS.
Elemental S obtained
by difference. 3.
The set of equations shown to 2-86
the left describes the kineti-
cally effective stoichiometry
occurring.
The two reactions yielding CS2
are similarly temperature-dependent
in the range 800-(J50°C.
Reaction (5) is more temperature-
dependent than (3).
4. A mechanistic argument was pro-
posed to remove the redundancies
in the kinetics of the proposed
stoichiometric scheme where CSg was
involved.
5. Reliable rate constants could not
be estimated based on the data.
-------
2.3 Conversion of Hydrogen Sulfide to Elemental Sulfur
The presence of excess hydrogen in the reducing
atmosphere for S02 reduction will result in H2S formation,
especially at high temperatures. For this reason the chemistry
involved in conversion of H2S to elemental sulfur is of interest
in this program. The majority of processes available for
accomplishing this are based on the Glaus reaction. Original-
ly developed as a gas phase process, many modifications have
been developed. In this section the chemistry of the gas phase
reactions will be discussed, and a brief summary of modifica-
tions in aqueous and organic media will be presented. Miscel-
laneous routes for conversion of H2S to elemental sulfur will
also be briefly addressed.
2.3.1 Gas Phase Glaus Process
This section contains a summary of the variations
of the gas phase Glaus process: the straight-through process,
the split flow process, and the direct oxidation process. The
three variations have been widely used to treat acid gas
streams having H2S concentrations ranging from 15 to 10070 H2S.
The Sulfreen process, which is a low temperature modification
of the basic Glaus process used in treating Glaus tail gases,
is also covered.
The Glaus process was designed for streams containing
H2S and C02. The main reactions involve partial oxidation of
the hydrogen sulfide to sulfur dioxide with subsequent conversion
of remaining H2S and S02 yielding elemental sulfur as shown below.
H2S + 3/202 = S02 + H20 (2-19)
2H2S + S02 = T S. + 2H20 (2-20)
-121-
-------
The overall Glaus reaction may be written as:
H2S + %02 = i S. + H20 (2-21)
J *J
The first detailed study of the reaction equilibria
between H2S and air was based on a stoichiometric 02/H2S ratio
(i.e., 0.5) in the initial mixture at total pressures of 0.5,
1, and 2 atm (2-87). The compounds assumed to be present in
the equilibrium mixture included S2, S6, S8, H20, H2S, S02,
and N2. In a later study these results were verified; also,
the influence of hydrocarbons and C02 in the starting mixture
was pointed out (2-88).
Basing calculations on more recently determined
equilibrium constants and considering additional compounds
(Sn , H2, 02, H2S2, SO, and SOa) in the equilibrium mixture,
similar sulfur yields at 550-650°K were predicted (2-89).
The calculated levels of the new compounds added were found
to be very low. Similar conclusions were reached in a dif-
ferent study for temperatures in the 400 to 1750°K range (2-90)
Bennett and Meisen (2-91) calculated equilibrium
compositions of mixtures resulting from reactions between H2S
and air at atmospheric pressure over the range 600-2000°K
(327-1727°C).. The 02/H2S ratio ranged from 0.05-1.0. These
parameters were chosen in order to simulate Glaus furnace
conditions. Compared to earlier studies additional species
were considered, particularly nitrogen compounds and free
radicals, the latter being potentially important at high
furnace temperatures. In this study C02, COS, CS2, NH3, and
hydrocarbons were omitted.
-122-
-------
The technique used consisted of calculating
equilibrium constants for each possible reaction at 100°K
intervals based on free energy data. Partial pressures of
the "key components," S2, H2S, H20, and N2, were guessed from
which the partial pressures of the remaining species could
be determined by applying the law of mass action to their
formation reactions. The total partial pressures, and the
atomic ratios of unbound oxygen to sulfur, hydrogen to sulfur,
nitrogen to oxygen, and carbon to sulfur were subsequently
evaluated. Using an iterative procedure, these were made to
converge, yielding characterization of equilibrium composi-
tion. The results of the calculations showed that 25 compounds
were present in concentrations >0.1 ppm for at least some
temperatures in the 600-2000°K range. Optimum sulfur yields
were predicted at 02/H2S ratios of less than 0.5 (stoichio-
metric) since this suppressed further oxidation of elemental
sulfur.
More recently equilibrium studies were performed
in which all Glaus reactions were considered by the same
researchers (2-92, 2-93). The following assumptions were
made:
1. All compounds behave as ideal gases,
2. Initial acid gas contains only H2S, C02, and
H20,
3. Air consists entirely of nitrogen (79%) and
oxygen (21%),
4. Total pressure of the system is one atmosphere.
-123-
-------
Equilibrium compositions were calculated for
mixtures resulting from reactions among H2S, C02 , H20, and
air at atmospheric pressure over the range 600-2000°K (327-
1727°C). The amount of air was varied from 20-300% of the
stoichiometric amount based on Reaction 2-21. The acid gas
composition included up to 10% H20 and up to 3070 COz . In
addition to the reactions already presented in the above dis-
cussions, the following were considered:
H20 + %S2 = %02 + H2S (2-22)
H20 = H2 + %02 (2-23)
C02 + %N2 + %H2 = HCN + 02 (2-24)
2HCN = C2N2 +• H2 (2-25)
C02 + 2H2S = CS2 + 2H20 (2-26)
2CS2 + H2 = C2H2 + 2S2 (2-27)
%C2H2 + 3/2H2 = CHu (2-28)
C2H2 + H2 = C2H4 (2-29)
C2H4 + %02 = CaH^O (2-30)
C02 = CO + %02 (2-31)
C02 + H2S = COS + H20 (2-32)
COS + H2 = CS + H20 (2-33)
The same calculation technique used in the previous study
(2-91) was employed.
-124-
-------
Of the 36 compounds considered, all except HCN,
C2N2, CHi,, CZH2, C2Ki>, and C2H40 had partial pressures greater
than 10~ atmospheres for at least some temperature in the
temperature range investigated. The partial pressures of the
remaining 30 compounds were plotted as a function of tempera-
ture. Sulfur yields were shown as functions of temperature,
C02 partial pressure, and 02 partial pressure. In order to
predict actual Glaus furnace compositions, the effect of
adiabatic conditions was also examined. Preheating of the
feed gases was recommended in order to achieve high sulfur
yields possible with less than stoichiometric amounts of air.
Another thermodynamic investigation of Glaus
combustion chambers was performed by Neumann (2-94). A
thermodynamic model modified by semiquantitative considera-
tions from reaction kinetics was employed to calculate
equilibrium compositions from waste gases containing H2S and
S02. Other species in the feed that were considered included
S2) H20, N2, H2, and/or C02.
Catalytic effects on oxidation of H2S by oxygen have
been the subject of numerous studies in recent years. Oxida-
tion on active coals gave 100% S yields compared to ^50% SOX
yield on "treated" coals (2-95). The catalytic properties
of the active coals were related to surface carbon oxides,
free radicals, and adsorbed cations. No further details were
available in the abstract. Bauxite, synthetic zeolite, and
other catalysts were tested in a low-temperature, two-stage
system (2-96). The pre-heated gas (350-460°C) was mixed
with a stoichiometric amount of air before passing through
the first reactor, cooled to 160°C to condense S, reheated to
260-270°C, and passed through the second reactor. Overall H2S
conversion was ^9070 with S the main product.
-125-
-------
Larin and Erofeeva reported that the heat of the
exothermic reaction
2H2S + 02 = 2H20 + S2 (2-21)
was 106 kcal (2-97). The reaction was carried out in a
fluidized bed packed with activated carbon.
Reaction kinetics and activation energies associated
with H2S oxidation by molecular oxygen over various catalysts
have been reported over the range 130-200°C (2-98). Catalytic
mechanisms for active C, molecular sieve 13X, and liquid sul-
fur were discussed. Cariaso also studied the oxidation over
porous carbons (2-99), while others investigated the catalytic
effectiveness of cobalt molybdate and related materials at
H2S levels below 4000 ppm (2-100). Bauxite catalysts of
hydroargillite structure were found to be effective for con-
version of COS resulting during production of S from H2S-
containing gases in the presence of C02 (2-101) .
Numerous investigations of the reaction between H2S
and S02 to produce water and elemental sulfur (Equation 2-20)
have been conducted. Mechanisms and kinetics of the alumina-
catalyzed reaction were determined using infrared spectrosc'opic
techniques (2-102). The relative rate constants for the
reactions below were 75:1:5 at 250°C in the presence of a
commercial cobalt-molybdate catalyst (2-103).
S02 + 2H2S = 2H20 + | Sx (2-20)
X
S02 + 2COS = 2C02 + | Sx (2-34)
COS + H20 = C02 + H2S (2-35)
-126-
-------
The HzS-SOz reaction appeared to be diffusion controlled with
an activation energy of 5.5 kcal/mole. Georgiev studied the
effect of steam concentration on the reaction (2-104).
An induction period T for the Glaus reaction was
measured by Kokochashvili and Labadze (2-105). The reactant
mixture composition and the state of the reactor walls were
shown to have an effect on the duration of T. Coating of
the walls with Cr oxide had a decreasing effect, while coating
with MgO, CuO, and Mn02 completely eliminated the induction
period. Adversely, treatment with KCl or KaB^O? inhibited the
reaction.
There are three basic arrangements of the gas phase
Glaus process: straight-through, split stream, and direct
oxidation. The main features of each are summarized below.
The Sulfreen process based on low-temperature gas phase reac-
tion is also described here.
Straight-Through Process
The straight-through process generally gives the
highest overall recovery of the three variations of the
Glaus process. It also allows maximum heat recovery at a
high temperature level.
In the straight-through process, the acid feed gas
is mixed with a stoichiometric quantity of air, determined
from Equation 2-21.
The combined stream is introduced to a furnace
where combustion occurs at about 2500°F. H2S is converted
to elemental sulfur to the extent of 30% to 69% (2-106).
-127-
-------
The remainder of the H2S is converted to S02, COS, and CS2
by the following reactions:
CIU H- 4S * CS2 + 2H2S (2-36)
C02 + H2S * COS + H20 (2-32)
H2S + 3/202 £ S02 + H20 (2-37)
Small quantities of C02 and cm are present in any H2S
stream.
The quantities of CS2 and COS generated are very
small in comparison to the S02 quantity; CS2 and COS become
important, however, since quantitative recovery of sulfur is
needed. The quantity of COS formed is usually close to its
thermodynamic equilibrium value for furnace flame conditions.
But CS2 concentrations are often hundreds or thousands of
times higher than the author's calculated equilibrium values
(2-107). The amount of CS2 formed is a strong function of the
hydrocarbon concentration in the acid gas feed.
The relative quantities of S02 and elemental sulfur
formed are determined by the combustion temperature of the
furnace. The reaction that forms S02 becomes faster at higher
temperatures than the reaction that forms sulfur. Therefore,
the furnace is generally maintained at a temperature which
will best compromise cooling requirements with sulfur conversion,
After the gas stream leaves the furnace, the
elemental sulfur is removed from the stream and the remaining
gas - a 2:1 mixture of H2S and S02 - is sent to a catalytic
-12S-
-------
converter at 218°C (2-108) . The HzS and S02 react to yield
H20 and S2 according to Equation 2-20. This reaction has been
found to give optimum conversion of HaS to sulfur at low
temperatures. However, the gas stream must be maintained
above the sulfur dewpoint at converter conditions, or the
liquid sulfur condensate will poison the catalyst.
Three catalytic converters are often used in the
Glaus gas phase process. The most common catalysts are
alumina and bauxite, but sometimes special catalysts are used
in one or more of these converters for the purpose of hydroly-
zing COS and CS2. Another common practice is operating the
first catalytic reactor at a relatively high temperature,
around 400°C, to hydrolyze COS according to Equation 2-35.
High temperature operation in this first converter must be
compensated for by operating the suceeding reactors closer
to the dewpoint if overall sulfur recovery is to remain high.
Each converter is followed by a condenser to remove sulfur,
driving the Glaus equilibrium closer to completion.
Tail gases from a Glaus plant usually contain HaS,
S02, elemental sulfur, COS, and CS2 . The total concentration
of sulfur is about 15,000 ppm, taken as S02 in an incinerated
tail gas (2-109).
Split Flow Process^
The split flow process, also called the modified
Glaus process, is the most widespread process for converting
concentrated H2S streams to elemental sulfur. In this form
of the Glaus process, one-third of the acid gas feed stream
is diverted and then completely oxidized to S02. The streams
are again combined and sent to a catalytic converter where
they react according to Equation 2-20. In this process, as
-129-
-------
in the straight-through process, the 2:1 stoichiometry must
be maintained by careful monitoring of the air supply.
The split flow process has greatest utility when
the HzS concentration in the acid gas feed is relatively low,
20 to 25 volume percent. In these cases the H2S concentration
may not be able to support combustion if the entire gas stream
is allowed to dilute the combustion products. If the acid
gas stream contains relatively high concentrations of hydro-
carbons, on the order of two to five percent, the split flow
process allows two-thirds of the gas stream to avoid the
furnace, producing less COS and C$2 (2-110).
The split flow process is subject to the same side
reactions and catalytic conversions as the straight-through
process. Typical conversions for the catalytic units are
(2-108, 2-111):
1 converter: about 80% recovery,
2 converters: 92-9570 recovery,
3 converters: 95-967, recovery,
4 converters: 96-9770 recovery.
Direct Oxidation
In this process the acid feed gas is burned directly
over a. bauxite catalyst. High temperatures, up to 1000°F, and
high space velocities are used. In this way the conversion
becomes controlled by kinetic factors; therefore, the limits
of the Glaus equilibrium do not apply to this combustion.
Stoichiometry is again important; the 2:1 H2S and SOz ratio
must be maintained.
-130-
-------
This process has its greatest utility with gas
streams having H2S concentrations of 1570 or less; higher
concentrations are more economically handled by the split
flow process (2-112). The direct oxidation step is generally
followed by conventional Glaus catalytic converters, which
convert the residual H2S and S02 to elemental sulfur.
Sulfreen Process
The Sulfreen process is essentially the Glaus
process made more efficient by operating at lower temperatures,
The H2S and S02 are reacted over an alumina or activated
carbon catalyst at 127-150°C - below the dewpoint of sulfur
in the reactor (2-113). These catalysts are very effective
adsorbents for sulfur. When the catalyst becomes saturated
with the liquid sulfur, hot gas is used to desorb the sulfur
and regenerate the catalyst.
While it is likely that little COS or CSz would be
generated at the Sulfreen reaction temperatures, these by-
products are not hydrolyzed to H2S over the Sulfreen catalyst
at these temperatures. Therefore, the loss of sulfur in the
form of COS and CS2 reduces overall recovery.
A variation of the Sulfreen process uses two stages.
The first stage uses the tail gas H2S and S02, adjusting the
stoichiometric balance so that all of the S02 is consumed. In
the second stage the residual H2S is oxidized directly to
sulfur.
-131-
-------
2-3.2 Glaus Reactions in Liquid Media
HaS removal processes based on the Glaus reaction
have been developed for liquid phase applications. Both
organic and inorganic media have been found practicable.
Two liquid phase processes have reached industrial
scale usage: the Bureau of Mines citrate process and the IFF
process. In the former an aqueous citric acid solution is
used to absorb SQz and HzS from the acid gas streams. The
basic Glaus reaction takes place in this medium although the
actual chemistry involved is more complex. The IFF process is
based on a catalyzed Glaus reaction in polyethylene glycol,
a relatively high molecular weight solvent. An activated com-
plex of HaS and SC-2 is formed with the metal catalyst.
Two experimental processes are currently in the
development stages: the Wiewiorowski liquid sulfur and the
ethylene glycol mono ether processes. Ethylenediamine and
other nitrogen compounds catalyze the Glaus reaction in the
liquid sulfur medium at 120 to 160°C. In the ethylene glycol
mono ether process, the,ether acts as the catalyst while the
amine provides nucleation or flocculation sites to minimize
colloid formation.
All of these are generally applicable to streams
of low HaS concentrations, e.g., Glaus tail gas streams;
although the Citrate process is a potential substitute for the
gas phase Glaus process.
-132-
-------
2.3.3 Other H2S Removal Processes
A number of other processes for HzS removal which
are not based on the Glaus reaction have been developed.
Those that employ liquid inorganic reaction medium involve
a coupled oxidation-reduction chemistry. Inorganic species
are used to oxidize the H2S, followed by air oxidation to
regenerate the inorganic oxidant.
Absorption of HzS by organic solutions and subsequent
oxidation are the bases of processes best suited to remove the
last traces of H2S before incineration and venting to the
atmosphere.
Several composite processes are also available on
an industrial scale. These are so classified because they
are composed of processes previously discussed.
In addition, hydrogen sulfide may he removed
electrochemically or by reaction with liquid SOz.
-133-
-------
2.4 Other Gas Phase Reactions
This section briefly summarizes recently published
information on the kinetics and chemistry of other gas phase
reactions of potential interest in this program. This category
includes possible side reactions in S02 reduction systems.
Literature reviews and data compilations (2-114, 2-115, 2-116)
should be consulted for information published prior to the
period covered in this literature survey, Chemical Abstracts
1967 through June 1975.
The results of the literature search summarized be-
low are grouped according to species involved. The order of
discussion is:
1) reactions involving sulfur dioxide and gaseous
species not dealt with in previous sections
(S02 oxidation is not within the scope of this
program) ,
2) reactions of sulfur vapor,
3) reactions of carbonyl sulfide,
4) reactions of carbon disulfide, and
5) reactions involving hydrogen sulfide not dealt
with in previous sections.
Reactions Involving S02
Gayen et al. calculated the free energy changes and
equilibrium constants for reactions in the C02-S02 system at
-134-
-------
one atmosphere, 600-1000°C, and C02/S02 mole ratios of 1.00-
0.25 (2-117). The two significant reactions considered are
presented below.
2S02 + C02 = CS2 + 302 (2-38)
CS2 + C02 = 2COS (2-39)
Their calculations indicate that CS2 formation in this system
is very small.
The interaction of S02 and COS has been the subject
of several recent studies. Pulse and semi-pulse methods were
used to study catalytic effects in one study (2-118). George
found no significant effect due to catalyst basicity on the
reaction below:
2COS + S02 = 2C02 + (-)S (2-40)
* x x
A mechanism was postulated for the reaction (2-119). Haas and
Khalafalla reported 90% conversion of reactants in the presence
of pure X-A1203 at 400°C. Inclusion of transition metals in
the catalyst decreased the interaction between COS and S02
(2-120).
Reactions of Sulfur Vapor
Information concerning the reactions of sulfur vapor
with atomic and molecular oxygen, C, CO, C02 and CH^ appeared
in the recent literature.
In the combustion of S vapors in jet conditions at
360-460°C, the presence of H2 or CHi* was found to produce an
inhibiting effect (2-121). The reaction rate of ground state S
-135-
-------
atoms with molecular oxygen to produce SO and 0 has been investi-
gated by flash photolysis-resonance fluorescence (2-122) and
spectroscopy in the vacuutn-uv region at 295°K (2-123) . Correla-
tion of the rate data using an Arrhenius type equation over the
range 252-423°K yielded, in cm3 mole'1sec") (2-122):
k = (2.24 ± 0.27) x 10"12 exp[(-O.CO
±0.10 kcal/mole)/RT] (2-41)
The absolute reaction rate measured at 295°K was 1.0 x 1012cm;
mole sec (2-12
earlier results.
mole sec (2-123); this value was reported to agree well with
Investigations of the interaction of sulfur vapor and
various forms of carbon have dealt with kinetics (2-124 - 2-126),
mechanisms (2-125, 2-127), and parameters such as pressure,
temperature, and bed height (2-124. 2-126, 2-128). The range of
temperatures included in these studies was 500 to iOOO°C. The
chief product was CSa, with some evidence of CS at high tem-
peratures (>1000°C) which subsequently underwent rapid poly-
merization .
Bechtold studied the kinetics of the reaction between
CO arid S on platinum wires in a flow system at 300-450°C (2-129),
The only reaction product detected was COS. Rate was dependent
on temperature, degree of coverage of the Pt surface with S,
and, depending on the first two parameters, CO and/or S pres-
sure.
The mechanism of sulfur formation in the flash photo-
lysis of carbonyl sulfide diluted in C02 was investigated
(2-130). The formation and removal of Sa takes place accord-
ing to the following scheme:
-136-
-------
2S2 + C02 = S^ + C02 (2-42)
S + S^ = S2 + S3 (2-43)
S + S3 = 2S2 (2-44)
The rate constant for the first of these reactions was deter-
mined to be ^1 x 10 cmsmole sec
The kinetics of the homogeneous reaction between sul-
fur and methane was reported to be first order with respect
to both reactants in the range 600-700° (2-131). The reaction
™ "3 ™ 1 ™ 0
rate constant k expressed as (mole €82) cm hr a tin is
k = 2.95 x 108 exp(-44.0 x 103RT): (2-45)
Leszczynski and Kubica calculated thermodynamic parameters for
the reaction
CH,, + 4S = CS2 + 2H2S (2-46)
in the range 25-1227°C (2-132). They determined the following
reaction rate constant:
k = 7.29 x 109 (P/RT)2 exp(-27,390/RT) (2-47)
The units were not available in the abstract. The catalyzed
synthesis of CS2 from S vapor and methane was the subject of
another experimental investigation (2-133) . An equation for
calculating the rate was proposed.
-137-
-------
Reactions Involving Carbonyl Sulfide
The gas phase chemistry of COS summarized below in-
cludes thermal decomposition and interactions with hydrogen,
sulfur vapor, water vapor, and oxygen.
The high temperature (2000-3200 °K) kinetics of COS
pyrolysis in argon atmosphere was studied in shock tube experi-
ments (2-134) . The principal species monitored were COS, CO,
S, S2, CS, and SO. Mechanisms, rate constants for the proposed
reaction schemes, and activation energies were reported. Haas
and Khalafalla studied the catalyzed decomposition in an inte-
gral reactor (2-120) . The apparent activation energy measured
in an uncatalyzed system was 28.7 kcal mole . This value was
reduced to 5.6 kcal mole when catalyzed with A120.3 or Si02.
Below 635°C the main products were C02 and CS2, while at higher
temperatures CO and S were produced. In another study supple-
mentary reactions of COS decomposition and catalytic effects
were investigated using pulse and semi-pulse methods (2-118) .
Donovan measured rate constants which represent the
sum of reaction and collisional relaxation for COS and H2 in
vacuum uv photolysis; the constants were 1.0 x 10 and 4.0 x
10~ cm3mole~ ^ec' , respectively (2-135) . The kinetics of
the reactions of hydrogen atoms with COS were measured at 25 °C
~llf
in a flow system (2-136) . The rate constant was 2.2 x 10~fcm3
particle" sec" . At all COS flow rates H2S is a major product,
CO production equals COS consumption, and 0.5 mole of COS are
consumed per H atom.
Reaction rates of COS with ground-state atomic sulfur
(3P) between 233 and 445°K, S(:D) atoms, and SO^) atoms in
the presence of SF6 have been studied (2-137, 2-138, and 2-139.
respectively) . The reaction shown below was reported to take
-138-
-------
place, and lower limits for the rate constant were established
(2-138).
SOD) + COS = S2 + CO (2-48)
The reaction between COS and H20 proceeds according
to the reaction below.
COS + H20 = C02 + H2S (2-49)
Pulse and semi-pulse methods were employed to examine roles of
supplementary reactions of the hydrolysis. Hydrated A1203 and
Si02 catalysts were both found to be active (2-118). George
reported that catalyst basicity increased the initial reaction
rate (2-119).
Rate constants were measured for the reaction of
ground state atomic oxygen with carbonyl sulfide.
0 + COS = CO + SO (2-50)
The rate constant in the Arrhenius form between 290 and 465°K
is given, in cm3mole"lsec~l, as (2-140) :
k = 1.2 x ID1* exp(-5800 cal mole'VRT) (2-51)
Correlation of rate data between 263 and 502°K yielded (2-141):
k = (1.65 ± 0.13) x 10"11exp(-4305 ± 55/RT) in cm3mole"'sec'1 (2-52)
An inhibiting effect of H2 and CH4 on combustion of COS vapor
in jet conditions at 360-460°C and 4-6 torr was observed by
Sarkisyan et al. (2-121).
-139-
-------
Reactions of Carbon Bisulfide
Hildebrand studied the gas phase equilibrium of the
reaction
CS2 + S = CS + S2 (2-53)
by mass spectrometry (2-142). The dissociation energy D0°(CS)
and heat of formation H0^ (CS) were determined: 166.1 ± 2
X 2 9 8
kcal and 70.0 ± 2 kcal mole"1, respectively.
The presence of H2 or CH^ produced an inhibiting
effect on CSz vapor combustion in jet conditions at 360-460°
and 4-6 torr (2-141). Methane produces a greater inhibiting
effect. The relationship between the concentration of 0 atoms
and the concentration of H2 in a CS2 flame was given.
Reactions of Hydrogen Sulfide
Kinetic information was found for the gas-phase reac-
tions involving H2S with hydrogen and oxygen.
Rommel reported the rate constants of the reaction
with hydrogen atoms at 298 °K to be 3.8 x 10" cm3particle~'sec
(2-136). At high flow rates of H2S 0.5 moles of H2S are con-
sumed per H atom originally present. The rate expression in
Arrhenius form over the range 190-464°K reported by Kurylo
et al. (2-143) in- cm.3mole~lsec~1 was:
-11
k = (1.29 ± 0.15) x 10 exp[-(1709 ± 60)/1.987T] (2-54)
This was obtained under conditions which favored only the H
atom-HzS reaction. Mihelcic and Schindler conducted an ESR
spectroscopic study of the reaction (2-144). In the -30 to
-140-
-------
95°C range the activation energy was 1680 cal mole" . Rate
constants for the following reactions were determined at 300°K:
1x3. US \»> w ii y h» i_»> 4. * v«>
Reaction (cm3mole~ sec )
H + H2S = H2 + SH 1.0 K 10"12 (2-55)
SH + SH = H2S + S 1.1 x 10"11 (2-56)
S + SH = S2 + H 4.5 x 10"11 (2-57)
The experimental frequency factor A was reported to be 1.7 x
10 cm3mole~ sec~ , which agreed well with the predicted
value.
Assuming a stoichiometry of 3.5 atoms 0 per mole HaS,
the specific rate constant for the reaction 0 + HaS = OH + HS
was (1.74 ± 0.40) x 1011 exp -(1500 ± 100/RT) in cm3mole"'sec'1
(2-145). Takahashi studied the chemiluminescent reaction
between HaS and atomic 0 in a flow system at room temperature
and 4-6 torr (2-146). The reaction scheme proceeded as follows:
HaS + 0 = HS + OH (2-58)
HS + 0 = SO + H (2-59)
H + HaS = Ha + HS (2-60)
The rate constant of the first reaction was evaluated as
3.52 x 10" "* cm3mole~1sec~1 .
-141-
-------
3.0 COMMERCIAL PROCESSES FOR PRODUCTION OF ELEMENTAL
SULFUR FROM MAGNESIUM SULFITE OR SULFUR DIOXIDE
In this section process schemes, equipment types,
and operating conditions as reported in the literature are given
for existing MgO regeneration and S02 reduction processes. In-
formation on the following processes is presented in this section;
MgO recovery from MgO flue gas
desulfurization systems
Allied Chemical S02 reduction system
Asarco-Phelps Dodge elemental
sulfur process
Elemental sulfur production from pyrite
Magnesium-base recovery process in the
pulping industry
No existing process for direct conversion of magnesium sulfite
to"elemental sulfur was found.
-142-
-------
3.1 The Current Operation of the Magnesium Oxide
Recovery Process for the MgO Scrubbing Plants
A commercial calciner is presently being used at the
Essex Chemical Company sulfuric acid plant in Rumford, Rhode
Island, to regenerate MgO for a stack gas S02 removal process.
Details of this process are shown in Figures A3-1 and A3-2. The
calciner is a direct fired, refractory lined, counter-current
rotary kiln (7'6" diameter by 120' long).
Temperatures at the middle of the calciner are kept
at approximately 677°C (1250°F). The calciner operates at
essentially atmospheric pressure. The calciner feed solids
are primarily MgS03 with some MgSOi* and unreacted MgO. A typi-
cal calciner feed composition is given in Table A3-1 (3-14).
TABLE A3-1
TYPICAL CALCINER FEED COMPOSITION
MgS03 63.9%
MgSO^ 12.77,
MgO 2.8%
Water and Inerts 21.0%
Impurities enter the system along with the make-up
MgO and water streams. A typical composition of commercially
produced make-up MgO is given in Table A3-2 (3-1).
TABLE A3-2
COMPOSITION OF'CALCINED MAGNESITE
MgO 97-99%
CaO 0.55-1.0%
Si02 0.2 -0.4%
Fe203 0.05-0.25%
A1203 0.04-0.20%
-143-
-------
SCHEMATIC PROCESS FLOW SHEET
««O FROM ACID PLANT
MgSOj TO ACID PLANT
FIGURE A3-1 (3-4)
MgO ADDITIVE SCRUBBER SYSTEM FOR
S02 RECOVERY, OIL FIRED BOILER
-144-
-------
SCHEMATIC PROCESS FLOW SHEET
SO2 GAS CLEANING
CONCENTRATED SO2 GAS
SULFURIC ACID PLANT
ELEVATOR
^Nt r CONVEYOR
TSw r
CONVEYOR
FIGURE A3-2 (3-4)
REGENERATION SYSTEM MgO RECYCLE PROCESS,
FOR PRODUCTION OF 98% SULFURIC ACID
-145-
-------
Calcination is performed by counter-current contact
with combustion gases produced by burning No. 6 fuel oil with 57«
excess air. Coke is added in the amount necessary to reduce
the MgSCK . The principal reactions in the calciner are (3-14):
MgSO* + C + %02 * MgO + S02 + C02 (3-1)
MgS03 -> MgO + S02 (3-2)
The process yields 987o MgO with 2% impurities remaining
in the solid product. The exit gas is a dilute stream of S02
whose approximate composition is given in Table A3-3 (3-_2) .
TABLE A3-3
TYPICAL CALCINER EXIT GAS COMPOSITION
N2 73
C02 6
02 5
H20 7
S02 9
The residence time of the solid phase in the calciner is about
one hour (3-2).
Operating problems which have been experienced with the
rotary kiln include:
(1) Formation of periclase, an unreactive,
"dead-burned" form of MgO at high
operating temperatures. Pulverizing
equipment was installed to activate the
MgO and proved satisfactory.
-146-
-------
(2) Excessive leakage at the seals of the
rotary calciner which makes it difficult
to maintain a reducing atmosphere . New
seals which were recently designed and in-
stalled have apparently corrected this problem
(3) Severe dusting in calciner which trips
flame safety controls . An extra operator
and flame scanner were necessary to solve
this problem
(4) Coke of low ash content (less than 10%) is
needed to prevent contamination of the calcined
solids and the product gas sent to the acid
plant (3^3).
In the course of the calciner operation upsets have
occasionally resulted in the formation of elemental sulfur.
Attempts to reproduce these upset conditions in the laboratory
have not as yet yielded any specific information on the formation
of elemental sulfur (3-4) .
-147-
-------
3•2 Allied Chemical S02 Reduction System
Allied Chemical Corporation has developed and commer-
cialized a process for direct, catalytic reduction of S02 to
elemental sulfur using natural gas as a reductant. The first
plant to use the process is located near Sudbury, Ontario, Canada.
A process flow diagram of this facility is shown in Figure A3-3.
The plant, which received a feedstock gas containing approximately
127o S02 from a sulfide-ore roasting facility, consists of three
main sections: gas purification, SOa reduction, and sulfur
recovery. The gas purification system, which is designed to re-
move excess water vapor and gaseous and solid impurities would be
optional in certain applications (3-5) .
The principle function of the catalytic reduction
section is to increase the H2S/SC>2 ratio in the gas stream to
approximately the ideal ratio of 2:1 required for the Glaus
reaction, while achieving maximum formation of elemental sulfur.
The primary reaction system may be summarized in the following
equations:
CH4 + 2S02 -» C02 + 2H20 + S2 (3-3)
4C1U + 6S02 -»• 4C02 + 4H20 + 4H2S + S2 (3-4)
In the reduction section, the combined, preheated natural gas and
SOa stream first passes through a four-way-flow reversing valve
and a final preheat reactor. The heated stream then enters the
primary reactor where over 40% of the total recovered sulfur is
formed (3-6). The reactor uses a catalyst developed by Allied
Chemical that is stable up to 1093°C (2000°F), achieves effi-
cient methane utilization, and provides minimum formation of
undesirable side products (3-5). Careful control of the reac-
tion conditions is necessary to achieve chemical equilibrium
in the single reactor.
-148-
-------
-P>
ALLIED CHEMICAL S02 REDUCTION TECHNOLOGY TYPICAL ROASTER GAS APPLICATION
S02 GAS
HOT GAS HEAT
-'.b^n
i--
NATURAL
GAS
MAIN BLOWER
M
umm TQ
HEUTRALIZATIOH
COOLING
fen
RECYCLE
./L
t s
_ i MspMJ >^
ACID emu
l
DRIP ACID
MIST TOWER
_J
COLO GAS BY-PASS
TWOSTAGE
CLAUS REACTOR
HEAT
REGENERATOR
INCINERATOR
&
COALESCER
SULFUR HOiniNG PIT
FIGURE A3-3 (3-8)
ALLIED CHEMICAL S02 REDUCTION TECHNOLOGY, TYPICAL ROASTER GAS APPLICATION
-------
The gas leaving the reduction reactor passes through
a second heat regenerator, where it gives up its heat to the
packing. Direction of flow through the two heat regenerators
is periodically reversed, to interchange their functions of heating
and cooling.
After condensing sulfur in a steam generator, the gas
stream enters a two-stage Glaus reactor system where HzS and
S02 react to produce elemental sulfur and water. At this point,
product sulfur is again removed from the gas by condensation.
Residual HiS in the Glaus plant effluent gas is oxidized to SOz
in an incinerator before it is exhausted through stack to the
atmosphere (3-6) .
According to the developer, this process can be applied
directly to SOz streams containing as low as 4% SOz where
oxygen content is not over 5.070. When higher oxygen concentrations
are encountered, provisions must be made to dissipate the excess
heat produced as a result of methane oxidation. The Canadian
plant has demonstrated that this process is capable of removing
better than 9070 of the SOz from the entering gas stream. Opera-
tion at one-third of design capacity with constant operating
efficiency has been established (3-7).
-150-
-------
3.3 Asarco-Phelps Dodge Elemental Sulfur Pilot Plant
A 20 tpd elemental sulfur pilot plant has been tested
by Asarco at their El Paso copper-lead smelter (3-9). The
Asarco process utilizes a feed gas stream containing a minimum
S02 concentration of 10-157o and a maximum 02 concentration of
2-370. A flow diagram for this process is shown in Figure A3-4.
A reforming process developed by Phelps Dodge Corporation was
used in the pilot facility to produce a reducing gas stream
containing a 48-50% mixture of CO and H2.
In the Asarco process, the reformed gas and S02 were
stoichiometrically combined and introduced into a primary
catalytic reactor at approximately 343°C (650°F). This reactor
was a vertical shell-and-tube heat exchanger with the tubes
filled with catalyst. Since the reduction of S02 is highly
exothermic, and an organic heat transfer fluid was circulated
through the shell side of the vessel to control reaction tempera-
ture. The heated coolant was used to generate steam and reheat
the process gas stream entering the second catalytic reactor.
The gas stream leaves the primary reactor essentially at equi-
librium. Approximately 69% conversion of S02 to sulfur vapor
was obtained when treating a 12% S02 stream.
Tail gases from the primary reactor were cooled in a
condenser to recover liquid sulfur and then reheated to about
204°C (400°F) and passed through a Glaus reactor. This step
utilized a catalytic, fixed bed reactor with no internal cooling
provided. In the second stage reactor, S02 and H2S react to
yield additional sulfur which is recovered in a second condenser
Total sulfur recovery averaged 88-92% over 91 days of
semi-continuous operation. One major problem involved in the
-151-
-------
ASARCO-PHELPS DODGE ELEMENTAL SULFUR PILOT PLANT
MOLTEN SULFUR
I
110 - r1- SULFUR
*"* ~H_ BURNER
i— •- STEAM
", 1 — | SOZ
pj ~
00ILER
PREHEATED AIR
NATURAL GAS ^ ,f*l
REFORMERREFC«MEO
GAS
t ALTERNATE SO, SOURCE
1
• ( STORAGE )
1 \ TANK /
1
1 LIOUIO
TAIL GAS RECYCLE
G\
PRIMARY ,'
REACTOR Hi
^
|
STFAU .__ .,.„..,.,,..
1 1 & .. " J^
/^S Cy
~7 \^y "*" V SECONDARY Pb\
i-^ HEAT TRANSFER A "6ACTOR \2/l
FLUID COOLER / \ i— j
x^x
V^/GAS STEAM
r-»- STEAM y pfi£HEATER | i
PR/MAHY 1 SECONDARY
CONDENSER CONDENSER
LIOUIO LIQUID
SULFUR SULFUR
TOeTJ f* if ^
AIR
INCINERATOR I
| """-—STEAM -r-" ^
J VAPORIZER
NATURAL
«A3 .
FIGURE A3-4 (3-9)
ASARCO-PHELPS DODGE ELEMENTAL SULFUR PILOT PLANT
-152-
-------
reducing operation was decrepitation of catalyst pellets in the
first few inches of the catalyst bed in the primary reactor.
No loss of catalyst activity was detected, but there was an
increased resistance to gas flow.
i)
Thermal decomposition of the organic coolant in the
primary reactor resulted in solid carbon deposits between the
tubes. This caused overheating in the blocked portions of the
reactor and subsequent warpage and burning through of the tube
sheet. The installation of a new primary reactor and further
test operation were planned as a future course of action.
-153-
-------
3.4 Sulfur Production at a Pyrite Smelting Plant
Outokumpu Oy Company is operating a pyrite smelting
plant in Finland that converts pyrite feed into elemental sulfur
and iron oxide pellets (3-10). The plant produces 90,000 tons/yr
of elemental sulfur and 250,000 tons/yr of iron oxide pellets.
A process flow diagram for this system is shown in Figure A3-5.
Pyrite concentrate (FeS2)is fed to a flash-smelting furnace.
The combustion of Bunker C fuel oil is carried out without any
soot formation in a specially designed furnace having three
high-efficiency burners. Incoming pyrites are suspended in
the hot reducing gases and fed to the smelter where thermal
decomposition takes place at about 1800°C (3270°F).
Reaction gases (C02, H20, N2, S02, H2S, CO, H2),
dust, and sulfur vapors are cooled in a radiation chamber
(high pressure steam boiler) and a convection chamber. This
cooling operation shifts the equilibrium of the gas components
to maximize sulfur content.
In the radiation chamber the gases are cooled to 600°C
(1110°F) where CO and H2 react with S02 to yield S2 and H2S. Dur-
ing a second cooling step down to about 300°C (570°F), H2S and S02
react to form elemental sulfur. Sulfur yield is optimized by
passing the gases at 270-277°C (518-530°F) over' an aluminum-based
catalyst in a catalytic reactor downstream of the low pressure
boiler. Separation of elemental sulfur from the combustion gases
takes place in a .cooling tower where molten sulfur circulates over
the gas stream. A second washing step with seawater removes small
amounts of sulfur remaining in the reaction gases.
-154-
-------
1IOKV
Pyrites, 70V. 200 mesh
(53% S, 46% Fe|
340,000 lon>/yr.
60 MW TURBOGENERATOR
ROTARY KIIN DRYER
»
FUEL OIL STORAGE
STEAM BOILER Sea water
Milk of lime
|572F-I ELECTROSTATIC
Oil burner. (3,270 F.|
FLASH-SMELTING FURNACE ,
WASTE HEAT BOILER
(high prenure)
Liquid FeS
1260,000 tons)
27% S »
CYCLONES ELECTROSTATIC
PRECIPITATOR
——fr- SO2 1,1 lulfuric acid plan!
ROASTING FURNACE
(1,B30-1,9JO F.)
FIGURE A3-5 (3-10)
PROCESS FOR CONVERSION OF PYRITE TO ELEMENTAL SULFUR AND IRON OXIDE
-------
3.5 The Magnesium-Base Recovery Process in the
Pulping Industry
Magnesium-base pulping with chemical recovery is
practiced in paper mills in North America and Europe. Most of
these mills use the magnesium bisulfite process ("magnefite").
In this study, the process step that regenerates magnesium oxide
is of interest. The details of this process are shown in Figure
A3-6.
Magnesium base liquor from the pulping operation,
containing 50-6570 lignosulfonate and the rest sugars and acids,
is concentrated to about 5570 total dissolved solids (3-12) .
Magnesium oxide ash is then recovered by burning the residue in
a furnace or a fluidized bed combustion chamber. The design of
the combustion chamber used to burn this material is important
in order to maintain the required properties of the MgO ash.
Combustion must be complete to produce a carbon free ash. At
the same time however, overheating of the MgO must be avoided
since this produces an unreactive, dead burned, form of MgO
called periclase. This crystalline form of the oxide cannot
be used in liquor regeneration.
The Babcock and Wilcox recovery units use a high
temperature furnace where fuel evaporation and combustion take
place. Hot liquor is sprayed into the furnace by means of
steam atomizing nozzles. Combustion temperatures .are controlled
so that the flue gases leave the combustion chamber at a tem-
perature above 1300°C (2372°F) to assure an essentially carbon-
free ash. The magnesium oxide ash is separated from the flue gas
with cyclonic collectors which have an 80-95% efficiency. The
remainder of the ash is collected in the cooking liquor prepara-
tion plant, where magnesium hydroxide and sulfur dioxide in the
flue gas are combined in gas-liquid contactors to regenerate the
-156-
-------
DIGESTERS 19)
120 tans/day pulp
WEAK RED
LIQUOR STORAGE (?)
263,00016s Steam/tir
ACID STORAGE
- S02
SEPARATION
MULTICLONES
RECYCU 7300
gaL'min
Magnefite chemical recovery cycle.
FIGURE A3-6 (3-11)
-157-
-------
cooking Liquor. In some installations, when the MgO ash is
finer, electrostatic precipitators or a wet separator is used
to recover the MgO (3-13).
An alternative system for decomposing waste cooking
liquor to MgO and S02 is used at the Wausau Paper Mills Company
in Brokaw, Wisconsin, where a fluidized bed combustion process
has been installed. Liquid concentrated to 4070 solids flows
into a fluidized bed furnace operating at about 930°C (1706°F).
This furnace contains a bed of granular magnesium oxide pellets
fluidized by the combustion air entering at the bottom. The
waste liquor decomposes to produce a flue gas containing S02-
MgO ash remains in the reactor and becomes part of the bed.
MgO pellets are continuously discharged from the bed to main-
tain a constant amount of bed material in the furnace. A
cyclone is used to remove entrained ash from the flue gas (3-13)
-158-
-------
BIBLIOGRAPHY
2-1 Rammelsberg, C., Ann. Phys. Chem. 67_(2) , 245 (1846).
2-2 Foerster, F., and K. Kubel,"Sulfurous Acid and its
Salts. II. Decomposition of Sulfites at High Tempera-
tures", Z. anorg. allgem. Chem. 139, 261-92 (1924).
2-3 Hagisawa, H., Sci. Rep. Tdhoku Imp. Univ. 123, 183
(1934/1935).
2-4 Okabe, Taijiro, and Shoichiro Hori, "Thermal Decomposi-
tion of Magnesium Sulfite and Magnesium Thiosulfate",
Technol. Repts. T6hoku Univ. 23^(2), 85-9 (1959).
2-5 Ketov, A. N., and V. V. Pechkovskii, "Thermal Decomposi-
tion of Magnesium Sulfite", Zhur, Neorg. Khim. 4, 272-
6 (1959).
2-6 Kim, Yong K., R. Lindsey Dunn, and John D. Hatfield,
"Thermal Decomposition of Magnesium and Calcium Sulfites",
Muscle Shoals, Alabama, Division of Chemical Development,
TVA, (1973).
2-7 Zhivopistsev, V. A., et al., "Planning of Experiments
for the Study of. the Dissociation of Magnesium Sulfite
in a Fluidized-Bed Furnace", Zavod. Lab. 37_(1) , 74-5
(1971).
2-8 Schwitzgebel, Klaus, and Philip S. Lowell, "Thermodynamic
Basis for Existing Experimental Data in Mg-S02-02 and
Ca-S02-02 System", Environ. Sci. Technol. 7_(13) , 1147
(1973).
-159-
-------
2-9 Nishimi, Fumio, et al., "Treating Magnesium Salt Slurry
from the Desulfurization of Flue Gas", Japan, Kokai 73
81,770 (Nov. 1973); C.A. 80: P-87321-d.
2-10 Akimoto, Shinkichi, "Recovery of Sulfur from Industrial
Waste Gas Containing Sulfur Dioxide", Japan, Kokai
73 09,995 (8 Feb. 1973); C.A. 79_: P-9422-c.
2-11 Shah, I. S., and Robert H. Quig, "Chemico-Basic Magnesium
Based Sulfur-Dioxide Recovery Scrubbing Systems", A.I.Ch.E.
Symp. Ser. 1973, 69_(127) , 84-5.
2-12 Shah, Indravadan S., "Hydrated Magnesium Sulfite-Mag-
nesium Oxide Conversion", Ger. Offen. 2,128,240, (27 Jan.
1972); C^A. 76: P-89728-h.
2-13 Berezina, M. I., and G. F. Pinaev, "Reaction of Metal
Oxides with Sulfur Dioxide", Vestsi Akad. Navuk Belarus.
SSR, Ser. Khim. Navuk 1971(3) . 61-5; C.A. 76_: 9997-g.
2-14 Ito, Hayami, and Koriaki Tahahashi, "Apparatus and
Method for Preparation of Magnesium Oxide and Sulfur",
Japan Kokai 74,118,699 (13 Nov. 1974); C.A. 82.: 128885-c.
2-15 Shah, Indravadan Shantilal, "Removal of Sulfur Dioxide
from Flue Gas", Ger. Offen. 2,127,133, (14 Dec. 1972);
C.A. 713: P-128117-w.
2-16 Yushkevich, N. F., et al., Zh. Khim. Prom., No. 2,
33 (1954), as cited in Ref. 2-27.
2-17 Lepsoe, R., Ind. Eng. Chem. 30, 32 (1938), as cited in
Ref. 2-27.
-160-
-------
2-18 Yushkevlch, N. F., et al., Zh. Khim. Prom. , No. 2,
33 (1934), as cited in Ref. 2-28.
2-19 Lepsoe, Robert, "Chemistry of Sulfur Dioxide Reduction:
Thermodynamics", Ind. Eng. Chem. 3J3(1) , 92 (1938).
2-20 Walker, Scott W. , "Hydrogen Sulfide from Sulfur Dioxide
and Methane", Ind. Eng. Chem. 38(9), 906 (1946).
2-21 Vilesov, N. G., and G. A. Gorbatykh, Khim. Prom. , No. 3,
27 (1966), as cited in Ref. 2-27.
2-22 Vilesov, N. G., et al., "Kinetics of the Natural-Gas
Reduction of Sulfur Dioxide to Elemental Sulfur",
Khim. Prom. Ukr. 1970(3) , 30-1; C.A. 73_: 68078-u.
2-23 Zal'tsman, S. L, et al., "Determination of the Produc-
tiveness of Active Aluminum Oxide in the Catalytic
Reduction of Sulfur Dioxide to Elemental Sulfur", Khim.
Tekhnol. (Kiev) 1974(3), 7-10; C.A. 81: 175435-v.
2-24 Otvagina, M. I., et al., "Catalyst for Converting
Sulfur Dioxide with Methane", USSR 362,632 (20 Dec.
1972); C.A. 7£: P-140907-m.
2-25 Zawadski, J. , et al., Przem. Chem. 2_2(11-12) , 558
(1938), as cited in Ref. 2-28.
2-26 Berkman, Ya. P., and S. V. Kushnir, Zh. Prikl. Khim.,
No. 6, 1211 (1959), as cited in Ref. 2-28.
-161-
-------
2-27 Averbukh, T. D., et al., "Thermodynamics of Reduction
of Sulfur Dioxide by Methane", Zh. Prikl. Khim. 43(1),
35-43 (Jan 1970).
2-28 Averbukh, T. D., et al. , "Production of Elemental
Sulfur by Reduction of Sulfur Dioxide with Natural Gas
Methane Process", International Chemical Engineering
9(2), 317 (1969).
2-29 Averbukh, T. D., "Methane Process for the Production of
Sulfur from Concentrated Sulfur Dioxides of Nonferrous
Metallurgy", Tsvetnye Metally 1970, 43(11), 26-30; C.A.
74: 23959-m.
2-30 Averbukh, T. D., et al., "Preparation of Elemental Sulfur
by the Reduction of Sulfur Dioxide by Natural Gas (Low
Temperature Catalytic Process)", Khim. Prom. (Moscow)
1971 47(3), 200-3; C.A. 74: 113870-e.
2-31 Averbukh, T. D., et al., "Thermodynamics of Sulfur
Dioxide Reduction. II. Reduction of Sulfur Dioxide by
Methane in the Presence of Carbon and by Carbon in the
Presence of Water Vapor", Zh. Prikl. Khim. (Leningrad)
1970, 43(2), 228-36.
2-32 Shakhtakhtinskii, G. B., et al., "Reduction of Sulfur
Dioxide by Karadag Natural Gas", Azerb. Neft. Khoz.
46(12), 40-2 (1967); C.A. 68: 88635-y.
2-33 Shakhtakhtinskii, G. B., et al., "Reduction of Sulfur
Dioxide by Natural Gas in a Fluidized Bauxite Catalyst
Layer", Issled. Obi. Neorg. Fiz. Khim. 1971(2), 75-80;
C.A. 77: 154524-a.
-162-
-------
2-34 Shakhtakhtinskii, G. B., et al., "Preparation of
Elemental Sulfur from 3-30% Sulfur Dioxide by Reduction
with Natural Gas in Catalyst-Bauxite Fluidized Bed",
Dokl. Akad. Nauk Azerb. SSR 1972 28(5) , 39-45; C.A. 78.:
161541-c.
2-35 Allied Chemical Corp., Applicability of Reduction to
Sulfur Techniques to_ the Development of_ New Processes
for Removing S02 from Flue Gases. Volumes !_ and II.
Washington, D.C., National Air Pollution Control Admin-
istration, 1969.
2-36 Hanish, H., and M. Schroder, Dinglers Polytech. J. 258,
225 (1885), as cited in Ref. 2-42.
2-37 Ferguson, J., Amer. Chem. Soc. 40, 1626-44 (1918), as
cited in Ref. 2-19.
2-38 Lepsoe, Robert, "Chemistry of Sulfur Dioxide Reduction:
Kinetics", Ind. Eng. Chem. 32., 910 (1940).
2-39 Jeffers, Peter, and S. H. Bauer, "Equilibrium Composi-
tions at High Temperatures in Selected Mixtures of
Sulfur Dioxide, Carbon Monoxide, Oxides of Nitrogen,
and Hydrocarbons", Combust. Flame 1971 17_(3), 432-6.
2-40 Haas, L. A., T. H. McCormick, and S. E. Khalafalla,
"Removing Sulfur Dioxide by Carbon Monoxide Reduction",
U. S. Bureau of Mines, Report of_ Investigations No.
7483, 1971.
2-41 Haas, L. A., and S. E. Khalafalla, "Removing Sulfur
Dioxide by Carbon Monoxide Reduction", Instrum. Contr.
Syst. 1972 45(3), 101-4, C.A. 76.: 144543-w.
-163-
-------
2-42 Khalafalla, S. E., et al., "Catalytic Reduction of
Sulfur Dioxide on Iron-Alumina Bifunctional Catalysts",
Ind. Eng. Chem. Prod. Res. Develop. 10(2), 133-7
(June 1971).
2-43 Haas, L. A., and S. E. Khalafalla, "Kinetic Evidence of a
Reactive Intermediate in the Reduction of Sulfur Dioxide
with Carbon Monoxide", Environ. Contr., Proc. Symp. 1972,
301-11; C.A. 79j, 80892-m.
2-4^ Khalafalla, S. E., and L. A. Haas, "Role of Metallic
Component in the Iron-alumina Bifunctional Catalyst
for Reduction of Sulfur Dioxide with Carbon Monoxide",
J. Catal. 1972, 24(1), 121-9; C.A. 7_6_: 37783-c.
2-45 Haas, L. A., and S. E. Khalafalla, "Kinetic Evidence of
a Reactive Intermediate in Reduction of Sulfur Dioxide
with Carbon Monoxide", J. Catal. 1973 29/2), 264-9.
2-46 Haas, L. A., et al., "Activity Patterns in the Catalytic
Reduction of Sulfur Dioxide on Some Transition Elements
in Alumina", U.S. Nat. Tech. Inform. Serv., P.B. Rep.
1972, No. 210988, 19 pp.
2-47 Khalafalla, S. E., "Abatement of Sulfur Dioxide Pollution
by Reduction with Carbon Monoxide", Int. J. Sulfur Chem.,
Part B 1972, 7_(1) , 41-50.
2-48 Khalafalla, S. E., and L. A. Haas, "Active Sites for
Catalytic Reduction of Sulfur Dioxide with Carbon
Monoxide on Alumina", J. Catal. 1972, 24(1), 115-20;
C.A. 76: 37782-b.
-164-
-------
2-49 Mohtadi, M. F., and H. B. Dingle, "Catalytic Conversion
of Sulfur Dioxide in Wet Stack Gases to Elemental Sulfur",
Adv. Chem. Ser. 1974, 133, 612-25; C.A. 81: 126333-b.
2-50 Querido, Robert, Removal of: Sulfur Dioxide from Stack
Gases by_ Catalytic Reduction t£ Elemental Sulfur with
Carbon Monoxide, University of Massachusetts, Ph.D.,
1971.
2-51 "Process Cleans S02 from Stack Gases", Chem. Eng. News
1970 (Dec. 14), 54.
2-52 Querido, Robert, and W. Leigh Short, "Removal of Sulfur
Dioxide from Stack Gases by Catalytic Reduction to
Elemental Sulfur with Carbon Monoxide", Ind. Eng. Chem.,
Process Des. Develop. 1973, 12_(1) , 10-18.
2-53 Quinlan, Charles W., et al., "Kinetics and Yields for
Sulfur Dioxide Reduction by Carbon Monoxide", Ind. Eng.
Chem., Process Des. Develop. 1973, 12(1), 107-11.
2-54 Okay, Vuranel C., and W. Leigh Short, "Effect of Water
on Sulfur Dioxide Reduction by Carbon Monoxide", Ind.
Eng. Chem. , Process Des. Develop. 1973, 12_(3) , 291-4.
2-55 Ryason, P. R., and J. Harkins, "A New Method of Simul-
taneously Removing Sulfur Dioxide and Oxides of Nitrogen
from Combustion Gases", J. Air Pollut. Contr. Ass. 17(12),
796-9 (1967).
2-56 Sood, Ajay, and J. R. Kittrell, "Dual Bed Catalyst for
Simultaneous Reduction of Sulfur Dioxide and Nitric
Oxide with Carbon Monoxide", Ind. Eng. Chem., Prod. Res.
Develop., 13(3), 180-185, (Sept. 1974).
-165-
-------
2-57 Quinlan, Charles W., et al., "Simultaneous Catalytic
Reduction of Nitric Oxide and Sulfur Dioxide by Carbon
Monoxide", Ind. Eng. Chem., Process Des. Develop. 1973,
11(3), 359-65.
2-58 Goetz, Vernon N., et al., "Catalyst Evaluation for the
Simultaneous Reduction of Sulfur Dioxide and Nitric
Oxide by Carbon Monoxide", Ind. Eng. Chem., Prod. Res.
Develop. 1974, 13_(2) , 110-14; C.A. 81: 29002-k.
2-59 Shakhtakhtinskii, G. B., et al., "Reduction of Sulfur
Dioxide by Carbon Monoxide in a Fluidized Aluminum Oxide
Catalyst Bed", Issled. Obi. Neorg. Fiz. Khim. 1971(2),
81-5; C.A. 77_: 154523-z.
2-60 Minasyan, V. T., and A. B. Nalbandyan, "Detection of
Sulfur Oxide (SO) Radicals in the Catalytic Reduction
of Sulfur Dioxide", Dokl. Akad. Nauk SSSR 1974, 218(2),
393-5.
2-61 Kasaoka, Shigeaki, et al., "Catalytic Reduction of Sulfur
Dioxide with Carbon Monoxide", Nippon Kagaku Kaishi
1973(2), 413-20.
2-62 Fukuzawa, Hisashi, and Yoshimi Ishihara, Catalytic
Reduction of_ Sulfur Dioxide by Carbon Monoxide. Part !_.
Effect o_f Gas Concentrations on the Conversion of Sulfur
Dioxide, PB-237725-T, Springfield, Va., NTIS, March 1972.
2-63 Meyer, B., Elemental Sulfur, New York, Interscience,
1965, as cited in Ref. 2-64.
-166-
-------
2-64 Murdock, David L., and Glenn A. Atwood, "Kinetics of
Catalytic Reduction of Sulfur Dioxide with Hydrogen",
Ind. Eng. Chem. , Process Des. Develop. 1974, 13_(3) ,
254-60.
2-65 Talybly, A. I., et al., "Reduction of 10% Sulfur Dioxide
to Elementary Sulfur", Mater. Konf. Molodykh Uch. Inst.
Neorg. Fiz. Khim. Akad. Nauk Azerb. SSR 1968 (Pub. 1969),
40-7; C.A. 74: 101164-q.
2-66 Shakhtakhtinskii, G. B., et al., "Reduction of Sulfur
Dioxide by Using Different Catalysts Under Steady-State
Conditions", Issled. Obi. Neorg. Fiz. Khim. 1971(2),
57-61; C.A. 77_: 154522-y.
2-67 Shakhtakhtinskii, G. B., et al., "Reduction of Sulfur
Dioxide by Hydrogen (as a Component of Converter Gas)
in a Steel Reactor", Issled. Obi. Neorg. Fiz. Khim.
1971(2), 62-6; C.A. 77_: 154526-c.
2-68 Shakhtakhtinskii, G. B., and A. I. Guliev, "Production
of Elemental Sulfur from Concentrated Sulfur Waste
Gases by Kivtset -(Sulfide B) Melting of Filizchai Poly-
metallic Ores", Issled. Obi. Neorg. Fiz. Khim. Ikh Rol
Khim. Prom., Mater. Nauch. Konf. 1967 (Pub. 1969), 25-7;
C.A. 73.: 100519-b.
2-69 Shakhtakhtinskii, G. B., et al., "Reduction of Sulfur
Dioxide by Converted Natural Gas on a Fluidized-
Catalyst Bed", Azerb. Khim. Zh. 1969(6), 125-8; C.A.
74: 5207-k.
-167-
-------
2-70 Shakhtakhtinskii, G. B., et al., "Reduction of 14-20%
Sulfur Dioxide by Converted Natural Gas in a Fluidized
Bauxite Catalyst Bed", Issled. Obi. Neorg. Fiz. Khim.
1971(2), 92-9; C.A. 77_: 154525-b.
2-71 Shakhtakhtinskii, G. B., et al., "Effect of Residual
Methane in Converted Natural Gas on the Reduction of
7-20% Sulfur Dioxide", Issled. Obi. Neorg. Fiz. Khim.
1971(2), 100-4; C.A. 78: 86583-n.
2-72 Shakhtakhtinskii, G. B., et al., "Reduction of 75% Sulfur
Dioxide by Converted Natural Gas in a Fluidized Bed of
a Bauxite Catalyst", Uch. Zap., Azerb. Univ.. Ser. Khim.
Nauk 1971(4) , 7-12; C.A. 78_: 48415-b.
2-73 Shakhtakhtinskii, G. B., et al., "Reduction of Sulfur
Dioxide by Converted Natural Gas in a Steel Reactor",
Issled. Obi. Neorg. Fiz. Khim. 1971(2) , 67-74; C.A. 7_8
86581-k.
2-74 Shakhtakhtinskii, G- B. , et al. , "Reduction of 5070
Sulfur Dioxide in a Fluidized Catalytic Bed Using Con-
verted Natural Gas as a Reducing Agent", Azerb. Khim.
Zh. 1974(3), 106-8; C.A. 82: 126987-p.
2-75 Steiner, Peter, "Reduction of Sulfur Dioxide to Sulfur",
Fr. Demande 2,195,584, 8 Mar. 1974; U.S. Appl 279,410,
10 Aug. 1972, 12 pp; C.A. 82: 34713-t.
2-76 Walker, P. L. , Jr., F. Rusinko, Jr., and L. G. Austin,
Advances in Catalysis, Vol. 11, New York, Academic
Press, 1959, pp. 133-221, as cited in Ref. 2-79.
-168-
-------
2-77 Ushakov, K. I., et al., "Reduction of Sulfur Dioxide
by Coke in Copper Smelting", Sb. Nauch. Tr. , Cos. Nauch.
Issled. Inst. Tsvet. Metal. 1967(26), 168-77; C.A. 69_:
88941-r.
2-78 Biswas, A. K, et al., "Conversion of Waste Sulfur Dioxide
Into Sulfur and Carbon Disulfide", Indian J. Techno1.
1968, 6(5), 157-8; C.A. 69: 60364-f.
2-79 Stacy, W. 0., et al., "Interaction of Sulfur Dioxide
with Active Carbon", Carbon 1968 6(6), 917-23.
2-80 Sappok, R. J. , and Philip L. Walker, Jr. , "Removal of
S02 from Flue Gases Using Carbon at Elevated Tempera-
tures", J. Air Pollut. Contr. Ass. 1969, 19/11), 856-61.
2-81 Haiti, B. R., et al, "Thennogravimetric Study on the
Kinetics of a Catalytic and Noncatalytic Gas-Solid
Heterogeneous Reaction Reduction of Sulfur Dioxide on
Charcoal Pellets", Trans. Indian Inst. Chem. Eng. 1969
(April), 60-5.
2-82 Biwas, A. K., et al., "Reduction of Sulfur Dioxide Over
Hot Carbon Surface", Indian J. Techno 1. 1974, 12_(4) , 161-
6; C-A- 81: 123717-u.
2-83 Macak, Juri, and Peter Pick, "Reaction System Carbon(s)-
Sulfur Dioxide(g)", Erdoel Kohle, Erdgas, Petrochem.
Brennst.-Chem. 1973, 26(6), 345-50; C.A. 79_: 96381-e.
2-84 Holub, Robert, et al., "Thermodynamic Equilibrium in
the Solid Carbon-Gaseous Sulfur-Dioxide System", Chem.
Prum. 1973, 23(1), 14-17; C.A. 7£: 128060-x.
-169-
-------
2-85 Kertamus, N. J., et al., "Process for Sulfur Dioxide/
Char Reaction", Hydrocarbon Process. 1974, 53_(2) , 95-6.
2-86 Blackwood, J. D., and D. J. McCarthy, "Kinetically
Effective Stoichiometry of Reactions in the Carbon-
Sulfur Dioxide System", Aust. J. Chem. 1973, 2_6(4) ,
723-31.
2-87 Gamson, B. W., and R. H. Elkins, Chera. Eng. Prog. 49,
203 (1953) , as cited in Ref. 2-91.
2-88 Valdes, A. R. , Hydrocarbon Proc. 43_(3) , 104 (1964),
as cited in Ref. 2-91.
2-89 Eriksson, G., and E. Rosen, Proc. Symp. on Recovery of
Pulping Chemicals, Helsinki, 1968, p. 325, as cited in
Ref. 2-91.
2-90 McGregor, D. E., Ph.D. Thesis, University of Alberta
(1971), as cited in Ref. 2-91.
2-91 Bennett, H. A., and A. Meisen, "Hydrogen Sulfide-Air
Equilibriums Under Glaus Furnace Conditions", Can. J.
Chem. Eng. 51(6), 720-4 (1973).
2-92 Meisen, Axel, and Howard A. Bennett, "Consider All
Glaus Reactions", Hydrocarbon Process., 53(11), 171-4,
(1974) .
2-93 Meisen, Axel, and Howard A. Bennett, "Hydrogen Sulfide-
Carbon Dioxide-Water-Air Equilibriums Under Glaus Furnace
Conditions", Can. Sulphur Symp. , (Pap.) 1974, (Pub. 1974),
13 pp.; C.A. 82: 90830-r.
-170-
-------
2-94 Neumann, Klaus Kurt, "Thermodynamic Calculations for
the Combustion Chamber of A Glaus Installation for Sulfur
Recovery", Erdoel Kohle, Erdgas, Petrochem. Brennst.-
Chem. 25(11), 656-9 (1972); C.A. 78: 161542-d.
2-95 Swinarski, A., et al., "Sulfur Oxides as Products of the
Catalytic Oxidation of Hydrogen Sulfide", Ann. Genie
Chim. 1967 (Pub. 1968), 3, 102-6; C.A. 70: 23259-a.
2-96 Adlivankina, M. A., et al., "Two-Step Catalytic
Reprocessing of Slightly Concentrated Hydrogen Sulfide-
Containing Gases", Khim. Prom. (Moscow) 46_(9) , 693-5
(1970).
2-97 Larin, E. T., and 0. B. Erofeeva, "Removing Sulfur-
Compound Impurities from Gases", Tr. Tashkent. Politekh.
Inst. 74, 8-9 (1971); C.A. 80: 147265-a.
2-98 Steijns, M., and P. Mars, "Role of Sulfur Trapped in
Micropores in the Catalytic Partial Oxidation of Hydrogen
Sulfide with Oxygen", J. Catal. 3J5(1) , 11-17 (1974);
C.A. 81:' 177024-j.
2-99 Cariaso, Onofre C., "Oxidation of Hydrogen Sulfide Over
Porous Carbons", Pennsylvania State Univ., Ph.D., 1972;
C.A. 80_: 72468-h.
2-100 Ross, Robert A., and Michael R. Jeanes, "Oxidation of
Hydrogen Sulfide over Cobalt Molybdate and Related
Catalysts", Ind. Eng. Chem., Prod. Res. Develop. 13(2).
102-5 (1974); C.A. 81: 29001-j.
2-101 Elistratova, L. I., "Effect of Carbon Dioxide on the
Process for Obtaining Sulfur from Hydrogen Sulfide Gases",
Gazov. Delo 1971(2), 34-7; C.A. 16: 47785-u.
-171-
-------
2-102 Dalla Lana, I. G., and C. L. Liu, "Catalytic Kinetics
in the Modified Glaus Process Use of Infrared Spectros-
copy to Measure Rates of the H2S/S02 Reaction", Paper
No. 2d, 79th National A.I.Ch.E. Meeting, Houston, March
1975.
2-103 George, Z. M., "Kinetics of Cobalt-Molybdate-Catalyzed
Reactions of Sulfur Dioxide with Hydrogen Sulfide and
Carbon Oxide Sulfide and the Hydrolysis of Carbon Oxide
Sulfide", J. Catal. 32_(2) , 261-71 (1974); C.A. 80:
74569-r.
2-104 Georgiev, B. P., "Equilibrium Yields in the Case of
Industrial Systems for the Production of Gaseous Sulfur
by the Glaus Method", Khim. Ind. (Sofia) 46(1), 30-2
(1974); C.A. 81: 123663-y.
2-105 Kokochashvili, V. I., and K. Z. Labadze, "Reaction of
Hydrogen Sulfide and Sulfur Dioxide in the Gas Phase",
Vses. Konf. Kinet. Mekh. Gazofazn. Reakts., 2nd. 1971,
10-11; C.A. 77: 156874-p.
2-106 Stecher, Paul G., Hydrogen Sulfide Removal Processes
1972, Park Ridge, N. J., Noyes Data, 1972.
2-107 Beavon, David K., and Raoul P. Vaell, "The Beavon Sulfur
Removal Process for Purifying Glaus Plant Tail Gas",
API Div. of Refining 1972, 267.
2-108 "Hydrogen Sulfide", in Kirk-Othmer Encyclopedia of
Chemical Technology, 2nd. ed., Vol. 19, N. Y., Wiley,
pp. 375 ff.
2-109 Beavon, David K., "Add-On Process Slashes Glaus Tailgas
Pollution", Chem. Eng. 78_(28) , 71-3 (1971).
-172-
-------
2-110 Goar, B. Gene, "Today's Sulfur Recovery Processes",
Hydrocarbon Proc. 47(9) , 248 (1968).
2-111 Beavon, David K., "Abating Sulfur Plant Tail Gases",
Pollut. Eng. 4(1), 34-5 (1972).
2-112 Grekel, Howard, "H2S to S...by Direct Oxidation", Oil
Gas J. 5_7_(30), 76 (1959).
2-113 Hyne, J. B., "Desulfurization of Effluent Gas Streams
- A Review and Comparison of Techniques", Proc. 51st
Annual Natural Gas Processors Assoc., 85.
2-114 Preston, K. F., and R. J. Cvetanovic, "Decomposition of
Inorganic Oxides and Sulfides", Compr. Chem. Kinet. 4,
47-141 (1972).
2-115 Schofield, Keith, "Evaluated Chemical Kinetic Rate
Constants for Various Gas Phase Reactions", J. Phys.
Chem. Ref. Data 1973, 2(1), 25-84.
2-116 Westley, Francis, National Bureau of_ Standards Special
Publication 362. "Chemical Kinetics in the Carbon-Oxygen-
Sulfur and Hydrogen-Nitrogen-Oxygen-Sulfur Systems:
A Bibliography 1899 through June 1971", GPO, Washington,
D.C., 62 pp, 1972.
2-117 Gayen, A. K., et al., "Thermodynamic Analysis of the
Carbon Dioxide-Sulfur Dioxide Reaction", Metals Miner.
Rev. 10(1), 41-3 (1970); C.A. 77: 106170-k.
2-118 Berezkina, L. G., et al., "Use of Pulse Chromatographic
Methods for Studying the Conversion of Sulfur-Containing
Gases", Eesti NSV Tead. Akad. Toim. , Keem. , Geol. 2_1(4) ,
321 (1972); C.A. 78: 62750-p.
-173-
-------
2-119 George, Z. M., "Effect of Catalyst Basicity for Carbonyl
Sulfide-Sulfur Dioxide and Carbonyl Sulfide Hydrolysis
Reactions", J. Catal. 35(2), 218-24 (1974); C.A. 81:
177048-v.
2-120 Haas, L. A., and S. E. Khalafalla, "Catalytic Thermal
Decomposition of Carbonyl Sulfide and its Reaction with
Sulfur Dioxide", J. Catal. 30(3), 451-9 (1973); C.A.
79.: 83886-y
2-121 Sarkisyan, E. N., et al., "Effect of Hydrogen-Containing
Substances on the Combustion of Carbon Disulfide,
Carbonyl Sulfide, and Sulfur Vapors", Dokl. Akad. Nauk
SSSR 203(4), 888-91 (1972); C.A. 77: 10087-n.
2-122 Davis, D. D., et al., "Flash Photolysis-Resonance
Fluorescence Kinetics Study of Ground-State Sulfur Atoms.
I. Absolute Rate Parameters for Reaction of S(3P) with
02(3S)", Int. J. Chem. Kinet. 4(4), 367-82 (1972); C.A.
77: 39659-m.
2-123 Donovan, R. J. and D. J. Little, "Rate of the Reaction
S(33Pj) + 02", Chem. Phys. Lett. 13(5), 488-90 (1972);
C.A. 77: 39636-b.
2-124 Wehrer, Pierre, and Xavier Duval, "The Reaction Between
Carbon and Sulfur at High Temperatures and Low Pressures'
C. R. Acad. Sci., Paris, Ser. C 265(7) , 432-5 (1967);
C.A. 68: 6776-h.
2-125 Agranovskii, I. N. , and A. I. Meos, "Physicochemical
Principles for the Formation of Carbon Disulfide from
Charcoal and Sulfur Vapor", Khim. Volokna 1968(3), 44-6;
C.A. 69: 70285-v.
-174-
-------
2-126 Levit, R. M., et al., "Effect of Gaseous Oxidants on
Organic Fibers", Nov. Khim. Volokna Tekh. Naznacheniya.
1973, 80-5; C.A. 80.: 49005-w.
2-127 Buri, Balwant R., and Ram S. Hazra, "Carbon-Sulfur
Surface Complexes on Charcoal", Carbon 9(2) , 123-34
(1971); C.A. 74: 150466-y.
2-128 Pedro, A. A., et al., "Determination of the Optimum
Height of the Charcoal Bed in Three-Stage Electric
Furnaces for Carbon Bisulfide Synthesis", Khim. Prom.
(Moscow) 49(7) 552 (1973); C.A. 8Ch 5322-t.
2-129 Bechtold, Ekkehard, "Reaction Between Carbon Monoxide
and Sulfur on Platinum", Z. Phys. Chem. (Frankfurt am
Main) 72(1-3), 99-108 (1970); C.A. 74_: 80290-k.
2-130 Langford, Robert B., and Geoffrey A. Oldershaw,
"Mechanism of Sulfur Formation in the Flash Photolysis
of Carbonyl Sulfide", J. Chem. Soc., Faraday Trans. I
6_i(Pt. 8), 1389-97 (1973); C.A. 79.: 72210-x.
2-131 Draganescu, A., et al.., "Kinetics of the Homogeneous
Reaction Between Methane and Sulfur", Rev. Roum. Chim.
18(11), 1859-63 (1973); C.A. 80: 137541-y.
2-132 Leszczynski, S., and S. Kubica, "Reaction of Methane
with Sulfur for Producing Carbon Bisulfide in an Adia-
batic System", Rev. Chim. (Bucharest) 24_(2) , 107-10,
(1973); C.A. 78.: 165125-t.
2-133 Cybulski, Andrzej, et al., "Rate of Sulfur and Methane
Catalytic Transformation to Carbon Bisulfide", Przem.
Chem. 50(4), 228-33 (1971); C.A. 75: 10776-v.
-175-
-------
2-134 Hay, Arthur J., and R. Linn Belford, "High-Temperature
Gas-Kinetic Study of Carbonyl Sulfide Pyrolysis Per-
formed with a Shock Tube and Quadrupole Mass Filter",
J. Chem. Phys. 47(10), 3944-60 (1967); C.A. 68: 33667-e.
2-135 Donovan, Robert J., "Direct Observation of S(31S0) Atoms
in the Vacuum Ultraviolet Photolysis of Carbonyl Sulfide",
Trans. Faraday Soc. 65_(6) , 1419-26 (1969); C.A. 71:
17482-x.
2-136 Rommel, H., and H. I. Schiff, "Reactions of Hydrogen
Atoms with Hydrogen Sulfide and Carbonyl Sulfide", Int.
J. Chem. Kinet. 4(5), 547-58 (1972); C.A. 77.: 106016-q.
2-137 Klemm, R. B., and D. D. Davis, "Flash Photolysis-
Resonance Fluorescence Kinetics Study of the Reaction
Sulfur (3p) + Carbonyl Sulfide", J. Phys. Chem. 78(12),
1137-40 (1974); C.A. 8J.:54818-p.
2-138 Fowles, P., et al., "The Reactions of Sulfur Atoms. IX.
The Flash Photolysis of Carbonyl Sulfide and the Reac-
tions of S(LD) Atoms with Hydrogen and Methane", ^J.
Amer. Chem. Soc. 89(6), 1352-62 (1967); C.A. 66: 80747-z.
2-139 Little, D. J., et al., "Relative Rate Data for the
Reactions of Sulfur (3^) Using the Thionitroso
Radical as a Spectroscopic Marker", Faraday Discuss.
Ghem. Soc. 1972(53), 211-16; C.A. 80: 41255-g.
2-140 Hoyermann, K., et al., "Rate Determination of the
0 + COS + CO + SO Reaction", Ber. Bunsenges. Phys.
Chem. n(6), 603-6 (1967); C.A. 67: 57584-f.
-176-
-------
2-141 Klemm, R. Bruce, and Louis J. Stief, "Absolute Rate
Parameters for the Reaction of Ground State Atomic
Oxygen with Carbonyl Sulfide", J. Chem. Phys. 61(11)
4900-6 (1974); C.A. 82: 90607-y.
2-142 Hildenbrand, D. L., "Thermochemistry of the Molecules
CS and CS+", Chem. Phys. Lett. 15(3), 379-80 (1972);
C.A. 77: 131498-f.
2-143 Kurylo, Michael J., et al.f "Absolute Rate of the
Reactions Atomic Hydrogen + Hydrogen Sulfide", J. Chem.
Phys. 54(3), 943-6 (1971); C.A. 74: 57690-g.
2-144 Mihelcic, D., and R. N. Schindler, "ESR Spectroscopic
Study of the Reaction of Atomic Hydrogen and Hydrogen
Sulfide", Ber. Bunsenges. Phys. Chem., 74(12) 1280-8
(1970); C.A. 74: 57711-q.
2-145 Hollinden, Gerald A., et al., "Electron Spin Resonance
Study of Kinetics of the Reaction 0(3P) Atoms with
Hydrogen Sulfide", J. Phys. Chem. 74(5), 988-91 (1970);
C.A. 72.'- 93701-d.
2-146 Takahashi, Saku, "Reaction of Hydrogen Sulfide with
Atomic Oxygen Studied by Emission Spectrum", Mem. Def.
Acad., Math., Phys., Chem. Eng., Yokosuka, Jap. 10
369-87 (1970); C.A. 75: 10804-c.
-177-
-------
3-1 McGlamery, G. G., et al., Conceptual Design and Cost
Study. Sulfur Oxide Removal from Power Plant Stack
Gas. Magnesia Scrubbing - Regeneration: Production
of Concentrated Sulfuric Acid, EPA-R2-73-244, Muscle
Shoals, Ala., TVA, 1973.
3-2 Zonis, Irwin S., et al., "The Production and Marketing
of Sulfuric Acid from the Magnesium Oxide Flue Gas
Desulfurization Process", Flue Gas Desulfurization
Symposium, Atlanta, Ga., Nov. 1974, Essex Chemical
Corp., 1974.
3-3 Koehler, G. R., "Operational Performance of the Chemico/
Basic Magnesium Oxide System at the Boston Edison
Company", Flue Gas Desulfurization Symposium, New Orleans,
May 1973.
3-4 Koehler, George R., and Edward J. Dober, "New England
S02 Control Project Final Results", Flue Gas Desulfuri-
zation Symposium, Atlanta, Ga., Nov. 1974.
3-5 Hunter, William D., Jr., and James P. Wright, "S02
Converted to Sulfur in Stack-Gas Cleanup Route", Chem.
Eng. 79(22), 50 (1972).
3-6 Hunter, William D., Jr., "Reducing S02 in Stack Gas to
Elemental Sulfur", Power 1973(Sept), 63.
3-7 Hunter, William D., Jr., and Aubrey W. Michener, "New
Elemental Sulphur Recovery System Establishes Ability
to Handle Roaster Gases", Eng. Mining J. 1973(June),
117.
-178-
-------
3-8 Hunter, William D., Jr., "Application of S02 Reduction
in Stack Gas Desulfurization Systems", Flue Gas Desul-
furization Symp., New Orleans, 14-17 May 1973.
3-9 Henderson, James M., "Reduction of Sulfur Dioxide to
Sulfur", Mining Cong. J. 59(3), 59-62 (1973).
3-10 Guccione, Eugene, "From Pyrite: Iron Ore and Sulfur
via Flash Smelting", Chem. Eng. 73_, 122-24 (1966).
3-11 Walther, J. E., H. R. Amberg, and H. Hamby, III,
"Meeting New Pollution Requirements at a Paper Mill",
CEP 69_(6) , 100 (1973) .
3-12 Blosser, Russell 0., and Isaiah Gellman, "Characteriza-
tion of Sulfite Pulping Effluents and Available Alterna-
tive Treatment Methods", TAPPI 56(9), 46 (1973).
3-13 Whittle, D. J., "Sulfite and Bisulfite Pulp Mill
Recovery Systems", TAPPI 54(7), 1074 (1971).
3-14 Environmental Protection Agency, Flue Gas Desulfurization
and Sulfuric Acid Production via Magnesia Scrubbing.
EPA-625/2-75-007, Research Triangle Park, North Carolina,
1975.
-179-
-------
APPENDIX B
TECHNICAL NOTE 200-045-31-02
"THERMODYNAMIC SCREENING TO DETERMINE THE
FEASIBILITY OF PRODUCING ELEMENTAL
SULFUR"FROM MAGNESIUM SULFITE"
-180-
-------
TECHNICAL NOTE 200-045-31-02
THERMODYNAMIC SCREENING TO DETERMINE THE
FEASIBILITY OF PRODUCING ELEMENTAL
SULFUR FROM MAGNESIUM SULFITE
4 September 1975
Prepared by:
Gary D. Brown
Philip S. Lowell
-181-
-------
TABLE OF CONTENTS
1. 0 INTRODUCTION 183
2. 0 DISCUSSION 184
2.1 Selection of Equilibrium Conditions 184
2. 2 Chemical Species Considered 197
3 . 0 RESULTS AND CONCLUSIONS 200
BIBLIOGRAPHY 211
-182-
-------
1.0 INTRODUCTION
The EPA is interested in the feasibility of
producing elemental sulfur in the regeneration section of
the magnesium oxide scrubbing process. This technical note
examines the influence of various reducing gases on the
decomposition of MgS03 to MgO and sulfur. The study will be
restricted to looking at the thermodynamics of the system.
The kinetics will be brought into the study later. Equili-
brium calculations were made to determine the influence of
temperature and stoichiometry on the gaseous and solid pro-
duct distributions.
The first section of the note discusses the
selection of the conditions and chemical species to be
included in the equilibrium calculations. The second section
contains the results and conclusions from the calculations.
The results of this technical note will be used as
the basis for selecting process configuration(s) and general
operating conditions.
-183-
-------
2.0 DISCUSSION
A thermodynamic screening was made by calculating
possible equilibrium concentrations of products from the
calciner. The influences of temperature and stoichiometry
using different reducing agents on gaseous and solid product
distributions were determined. An overall heat balance for
0
the calciner was also calculated.
A computer program was used on all equilibrium and
heat calculations. The program determined chemical equilibrium
by minimizing the free energy of the system. The input to the
system in the program was one mole of MgS03 with the amount of
the reducing gas determined by the desired stoichiometry.
The temperature of the system was specified and the equili-
brium composition was determined at 1.0 atm.
2.1 Selection of Equilibrium Conditions
The decomposition temperature of MgSOs is defined
as the temperature at which the equilibrium partial pressure
of S02 over solid MgS03 is 1 atm. Although this occurs at
360°C in an inert gas atmosphere, the reaction is fairly slow
at this temperature. MgSOa decomposes rapidly above 500°C.
Therefore, to insure that the lower temperature limit would
be within the investigation range, a lower limit of 350°C was
chosen.
From the literature search we see that the
noncatalytic gas phase reactions which produce elemental
sulfur using CO or H2 as reducing agents do not proceed
rapidly below 900°C (no significant conversion within several
minutes). Catalysts lower the gas phase reaction temperature
to the order of 400°C. In order to include noncatalytic
-184-
-------
temperatures of interest an upper limit of 1000°C was chosen.
The discrete temperatures used in the equilibrium screening
study are shown in Table B2-1.
Reducing gases of CO, H2, CO + H2, and CHu were
selected on the basis of.their relatively widespread avail-
ability and their previously known capability to reduce S02
to elemental sulfur. Coal and coke were not chosen because
their reactive intermediates are probably CO and H2. The
kinetic limiting step may well be the coke or coal gasifica-
tion step and is beyond the scope of this study.
The stoichiometries of the reducing agents were
varied to find the optimum stoichiometry for elemental sulfur
production and to simulate possible different conditions in
a reactor. Three major types of reactors were considered:
fluidized bed, co-current, and countercurrent reactors.
Figure B2-1 shows the directions of flow for the gas and solid
phase in each reactor. An important point is that for co-
current and fluidized bed reactors, the gas and solids leaving
the reactor are in intimate contact. For countercurrent
reactors they are not in contact.
Figure B2-2 shows the relative amounts of unreacted
species present (MgS03 and the reducing gas) over the length
of the reactor for the fluidiz-ed bed reactor. The solid com-
position is the same in all parts of the reactor because of
its mixing characteristics. The gas composition becomes lower
in unreacted reductant as it passes through the reactor from
inlet to outlet.
-185-
-------
TABLE B2-1
EQUILIBRIUM CASES
Temperature (°C) Stoichiometry Reducing Gases
350 .95 100% CHf
500 1.0 100% CO
700 1.05 100% H2
1000 1.5 50% CO, 50% H2
20.0
-186-
-------
A
TCT
•/ 4'
f'l0/
VoiV
5!3i/?
o
II"
0
o\l
)0|b
Solid
Gas Solid
FLUIDIZED-BED REACTOR
Gas
Solid
Gas
Solid
^/
^\
*J.
fe? -fc# -&• -&• •&• -&• •&• -&• -&• -Qi
COUNTERCURRENT REACTOR
COCURRENT REACTOR
FIGURE B2-1 - DIRECTIONS OF FLOW IN THREE TYPES OF REACTORS
-187-
-------
co
0)
•H
O
CD
"O
-------
In the countercurrent reactor depicted in Figure
B2-3, the unreacted solids decrease steadily from inlet to
outlet. The reacted solids at the outlet will "see" fresh
reducing gas. The "spent" reducing gas leaving "sees" fresh
solids. A cocurrent reactor is depicted in Figure B2-4.
The stoichiometries in Table B2-1 were chosen to
represent several different conditions. Obviously stoichi-
ometries with the minimum required to produce elemental sulfur,
i.e., near 1.0, are of most interest. Values of 0.95, 1.0,
and 1.05 are directed toward the gas phase and the product
composition attainable. The stoichiometry of 1.5 was chosen
to see what effect a significant excess of reducing agent
would have.
The value of 20 was chosen primarily to investigate
the solid effluent from a countercurrent reactor. Consider,
for instance, the case of the overall stoichiometry being one
and at a point near the solid exit. Choose the point at which
the unreacted solid is five percent. The gas therefore con-
tains twenty times the amount of reducing agent necessary.
The chemical potentials of solids are not dependent upon the
amount of various species present- In the gas phase the chemi-
cal potentials are dependent upon the amount present. For this
reason the twenty to one stoichiometry was used to predict sul-
fiding, coking, and other tendencies at the solid exit of a
countercurrent reactor.
A heat balance around the overall process was made
by inputing the gas and solid streams at specified tempera-
tures. A temperature of 25°C was chosen for the magnesium
sulfite, which corresponds to the case in which MgS03 is de-
hydrated at a site separate from the calcining operation.
If the MgS03 is dehydrated on site it would be at a higher
temperature because it would pass from a dryer to the calciner.
-189-
-------
Solids
Inlet
en
0)
•H
O
0)
OH
CO
-o
01
-U
u
cfl
QJ
Gas
Outlet
Gas
Inlet
Solids
Outlet
Reactor Length
FIGURE B2-3 - REACTION DRIVING FORCE IN A COUNTERCURRENT
REACTOR
-190-
-------
Inlet.
en
01
•H
o
ai
4-J
a
ca
-------
Three of the reducing gases, H2, CO, and the H2/CO
mixture, were specified to be at 1000°C. This temperature was
chosen as a rough estimate of the reducing gases produced from
a fuel combustion process. Methane was chosen to be 25°C
since it would be obtained from a pipeline. The gas and solids
leaving the reactor were assumed to be at the same temperature.
An enthalpy balance was made for the various
reactions. The simplified scheme shown in Figure B2-8 was
used. A major purpose of the calculations was to determine
the adiabatic operating temperature of the reaction, i.e.,
the condition at which Q = 0. For cases where Q < 0 heat
must be removed. For cases where Q > 0 heat must be added.
To extrapolate the cases to different inlet temperature con-
ditions, e.g., reducing gas at 500°C, the enthalpy difference
in Kcal per Kg of total reactants must be subtracted from
the calculated Q's to give values for the new conditions.
For the example given the number to be subtracted is negative
so the net result would be to make the reactions more endo-
thermic (less exothermic).
The temperature profile in a fluidized bed reactor
is shown in Figure B2-5. Fluidized bed reactors are character-
ized by isothermal operation because of high gas-solid heat
transfer rates and good mixing. The profile shows that the
gas and solid are at the same temperature everywhere within
the reactor. The temperature profile for a countercurrent
reactor is shown in Figure B2-6. The profile for this special
case has the gas and solid leaving the reactor at the same
temperature. Figure B2-7 shows a cocurrent reactor.
-192-
-------
u
o
0)
^
d
4-1
o;
a
I
25°C
(Inlet)
Solid Phase and Reducing Gas
1000°C
(Inlet)
Reactor Length
FIGURE B2-5 - TEMPERATURE PROFILE IN A FLUIDIZED BED REACTOK
-193-
-------
LOOO°C,
^ Solids
3 Outlet
cd
^
01
Gas
Outlet
Reactor Length
25°C
(Solids
Inlet)
FIGURE B2-6 - TEMPERATURE PROFILE IN A COUNTSRCURRENT REACTOR
-194-
-------
Gas
Outlet
Solids
Outlet
25°C
Solids.
Inlet
Reactor Length
FIGURE B2-7 - TEMPERATURE PROFILE IN A COCURRENT
REACTOR
-195-
-------
Q Heat
MgS03 @ 25°C
@ 25 °C
CO or H2 @ 1000°C
REACTOR
T = 350, 500, 700,
or 1000°C
Gaseous Products ^
@ T°C
Solids Products
FIGURE B2-8 - HEAT BALANCE SCHEMATIC
-------
The results may also be used to some extent for
investigating different exit conditions. The gas temperature
may not be changed because it would alter the equilibrium
mixture. The solid MgO temperature could be changed in a
fashion similar to the reactant sensible heat changes.
2.2 Chemical Species Considered
As the stoichiometry and temperature are changed,
the distribution of elements among the various species changes
Also, the solids in equilibrium with the gas can change.
The gas and solid species which were considered as
being potentially present in this system are listed in Table
B2-2. These species were selected on the basis of their
probability of existing in the reactor as determined by the
literature survey.
All of the gas species, with the exception of H, MgS,
and S, were commonly seen experimentally and were included in
thermodynamic calculations by investigators of S02 reduction
systems. The above three species were included to check their
possible presence at different conditions. The gas phase ele-
mental sulfur species were assumed to be 82 and S8. While one
investigator presented evidence that S3, S4, S5, and S7 also
could exit in this system (ME-121), the thermodynamic data for
these were not in the Radian data base.
The solid phase species MgO, MgSOs, MgSCK , and S
(liquid) have been found in the magnesium oxide calciner
used in Rumford, Rhode Island (KO-134). The MgS and
MgS203 species were a subject of study in the Mg-S02-03
-197-
-------
TABLE B2-2
CHEMICAL SPECIES CONSIDERED IN
THERMODYNAMIC SCREENING
Gas
H
H2
H20
H2S
02
CO
CO 2
COS
CS2
CHi,
MgS
S
S2
S8
S02
S03
Condensed Phase
MgO MgS03
MgS MgSO,
S (liquid) MgS203
C (graphite) MgC03
-198-
-------
system (SC-144). The two carbon containing species, MgC03
and C (graphite), were chosen to check their existence at
high concentrations of reducing agents containing carbon.
For the majority of the species considered the
thertnodynamic quantities used in this study were obtained from
the well-known JANAF Thermochemical Tables (ST-067). Data
were estimated only for MgSiOa.
-199-
-------
3.0 RESULTS AND CONCLUSIONS
This section presents a discussion of the results
of the thermodynamic screening. Calculations for the four
reducing gases at five stoichiometries and four temperatures
were made (a total of 80 cases) as shown in Tables B3-1
through B3-8.
The MgSOs was completely decomposed in all of the
cases. The main decomposition product was MgO. In the low-
temperature, low-stoichiometry cases some MgSOi* formation
was predicted. This indicates that a temperature above 350°C
but less than 500°C would be desirable from a solid product
point of view. In none of the cases were MgSaOa or MgS stable
There appears to be no danger of sulfide formation under re-
ducing conditions.
When CO is used as a reducing gas a large percentage
of the solid product appears as MgC03 at 350°C. The MgC03
compound is not a process problem since it would decompose
during the scrubbing process to MgO and C02- It could be a
shipping problem because MgCOs weighs twice as much as MgO.
The two reducing agents containing carbon, CH^, and
CO, have a large carbon formation tendency at high stoichiom-
etries. Coke formation might be expected at the gas inlet to
a fluid-bed or countercurrent calciner due to methane cracking
or CO decomposition. If the process is catalytic the coking
condition could be a problem.
-200-
-------
TABLE B3-1
MgS03 DECOMPOSITION WITH CH^ REDUCTANT
Case
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Stoic.
.95
.95
,95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
20.0
20.0
20.0
20.0
Temp.
( C)
350
500
700
1,000
350
500
700
1..000
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
(Fraction of tota]
S2
.004
.200
.554
.657
.005
.203
.554
.663
.005
.205
.548
.662
.001
.016
.125
.347
0
0
0
0
S3
.489
.354
0
0
.487
.348
0
0
.497
.336
0
0
.001
.0
0
0
0
0
0
0
S02
.002
.182
.184
.167
.014
.150
.150
.134
.014
.120
.120
.105
0
0
0
.004
0
0
0
0
. sulfur in each
H2S
.354
.263
.259
.171
.378
.297
.291
.196
.395
.337
.328
.225
.990
.959
.820
.623
1.0
1.0
.999
.999
COS
0
.001
.004
.006
0
.001
.004
.007
0
.002
.005
.008
.007
.025
.054
.025
0
0
.001
.001
species)
CS2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.001
0
0
0
0
0
MgS04
.151
0
0
0
.117
0
0
0
.089
0
0
0
0
0
0
0
0
0
0
0
-201-
-------
TABLE S3-2
MgS03 DECOMPOSITION WITH CHt» REDUCTANT
Case
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Stoich.
.95
.95
.95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
20.0
20.0
20.0
20.0
350
500
700
1,000
T cn
500
700
1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
/kcal\
\ kg J
I) + 101
+ 224
+ 319
+ 421
+ 107
+ 219
+ 314
+ 419
+ 112
+ 215
+ 309
+ 417
'+ 142
+ 188
+ 264
+ 428
+ 259
+ 507
+ 823
+1,432
Solid Vfractiony
MgO
.849
1.0
1.0
1.0
.883
1.0
1.0
1.0
.911
1.0
1.0
1.0
1.0
1.0
1.0
1.0
.283
.196
.194
.137
MgSO.,
.151
0
0
0
.117
0
0
0
.089
0
0
0
0
0
0
0
0
0
0
0
C(gr)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.111
.804
.806
.863
Gas (mole fraction)
Ctu
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.590
.393
.299
.087
H2
0
0
.002
.014
H20
.400
.392
CO;
.318
.270
.365! .249
i
.391, .229
0 .395 .317
0 ; .391
.002; .366
.016 .393
0
0
.002
.019
.001
.006
.032
.065
.174
.408
.396
.387
.365
.392
CO
0
0
.001
.015"
I
0
.278 0
.256
.234
.317
.284
.263
.237
.226 .330
.234 .319
.263 .278
.297
.154
.215
.002
.116 .003
.519 .057 .004
.764
.004
.001
.001
.017
0
0
.0011
i
.020
0
f
.002
.022
.083
0
.006
.059
.095
-202-
-------
TABLE B3-3
MgS03 DECOMPOSITION WITH CO REDUCTANT
Case
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Stoich.
.95
.95
.95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
Temp.
(°C)
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
20.0 350
20.0 ! 500
(Fraction of total sulfur in each species)
S2
.004
.271
.897
.837
.004
.277
.918
.863
.004
.284
.902
.883
.005
.069
.380
.811
.001
S8
.961
.607
0
0
.993
-612
0
0
.896
. .617
0
0
.388
.001
0
0
0
0 0
20.0 700 j .004 0
20.0 , 1,000
.146 0
S02
.001
.075
.071
.135
.001
.038
.033
.103
0
.001
.010
.077
0
0
H2S COS
-
-
-
-
-
-
-
-
-
-
-
-
-
0
.038 |
0
0
.001
.047
.031
.028
.002
.073
.048
.034
.095
.096
.087
.041
.518
.879
.596
.150
.978
.995
CS2
0
0
0
0
0
0
0
0
.004
.002
.002
0
.089
.051
.024
.002
.021
.005
0 j - .963 j .033
0 ; - .748 .105
MgSO.,
.033
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-203-
-------
TABLE B3-4
MgS03 DECOMPOSITION WITH CO REDUCTANT
Case
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Stoich.
.95
.95
.95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
20.0
20.0
20.0
20.0-
Temp.
(°C)
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
AH^
V kg y
- 386
- 154
- 56
+ 45
- 396
- 169
71
+ 33
- 398
- 184
- 83
+ 24
- 429
- 235
- 147
1
- 841
- 729
- 409
14
Solid (mole fraction)
MgO
0
1.0
1.0
1.0
0
1.0
1.0
1.0
0
1.0
1.0
1.0
.058
1.0
1.0
1.0
0
.055
.098
1.0
MgS04
.033
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
MgC03
.967
0
0
0
1.0
0
0
0
1.0
0
0
0
.779
0
0
0
.051
0
0
0
c, .
(gr)
0
0
0
0
0
0
0
0
0
0
0
0
.163
0
0
0
.949
.945
.902
0
f mole \
Gas \fractiotv
CO 2
.882
.847
.768
.705
.886'
.854
.776
.708
.825
.862
.773
.706
.679
.667
.630
.559
.949
.828
.366
.051
CO
0
.001
.005
.058
0'
.001
.007
.068
.001
.001
.013
.080
.002
-023
.117
.268
.002
.128
.602
.927
i
-204-
-------
TABLE B3-5
DECOMPOSITION WITH H? REDUCTANT
Case
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Stoich.
.95
.95
.95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
57 20.0
58
20.0
59 20.0
60 20.0
Temp.
(°C)
350
500
700
1,000
350
500
700
• 1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
(Fraction of total sulfur in each species)
S2
.004
.189
.393
.531
.004
.189
.388
.531
.004
.187
.379
.526
.001
.004
.031
.173
0
0
0
1,000 0
S8
.152
.128
0
0
.150
.117
0
0
.147
.104
0
0
0
0
0
0
SO 2
.002
.262
.238
.207
.002
.232
.206
.176
.002
.203
.176
.147
0
.002
.011
.047
0 0
i
0 0
0 0
0 0
H2S
-607
.422
.369
.262
.634
.463
.406
.293
.661
.506
.445
.327
.999
COS
-
—
-
-
-
-
-
_
CS2 MgSCK
.234
0
-
-
-
-
-
-
-
-
-
- -
-
.994
.958 | - -
.779
1.0
1.0
-
-
1.0
1.0
-
-
-
-
-
0
0
.210
0
0
0
.186
0
0
0
0
0
0
_0
0
0
0
0
-205-
-------
TABLE B3-6
MgS03 DECOMPOSITION WITH H2 REDUCTANT
Case
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Stoich.
.95
.95
.95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
20.0
20.0
20.0
20.0
1
Temp.
(°C)
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
M/kcal\
ml • •)
V k§ /
- 194
33
+ 49
4- 162
- 202
52
+ 30
+ 146
- 211
70
+ 11
+ 131
- 292
- 237
- 154
1,000 + 10
350
500
700
f mole N
Solid \fraction7
MgO
.766
1.0
1.0
1.0
.790
1.0
1.0
1.0
.814
1.0
1.0
1.0
1.0
1.0
1.0
1.0
• -1,116 1.0
- 873
1.0
MgSO,,
.234
0
0
0
.210
0
0
0
.186
0
0
0
0
0
0
0
0
0
- 542 • 1.0 0
1,000 - 28 1.0
,
0
i mole ^
Gas v fraction /
H20
.672
.650
.653
.668
.675
.656
.661
.675
.678
.661
.668
.679
.666
.664
.653
.608
.050
.050
.050
.050
H2
0
0
.003
.022
0
0
.003
.024
0
0
.003
.027
0
.004
.021
.101
.925
.925
.925
.925
-206-
-------
TABLE B3-7
HgSOa DECOMPOSITION WITH CO/H2 (1:1) REDUCTANT
Case
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
Stoich.
.95
.95
.95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
20.0
20.0
20.0
20.0
Temp.
(°C)
350
500
700
1,000
350
500
700
.1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
(Fraction of total sulfur in each species)
S2
.005
, ;233
.560
.653
.006
.237
.561
.661
.006
.239
.541
.631
0
.014 •
.121
.431
0 '
0
0
.001
S8
.499
.308
0
0
.510
.301
0
0
.520
.289
0
0
0
0
0
0
0
0
0
0
S02
.002
.186
.182
.178
.002
.154
.149
.147
.002
.124
.125
.164
0
0
.002
.009
0
0
0
0
H2S
.345
.270
.250
.158
.362
.305
.282
.180
.379
.345
.307
.174
.994
.940
.788
.520
.999
.995
.976
.965
COS
.001
.003
.007
,010
.001
.003
.008
.012
.001
,003
.025
.031
.006
.046
.089
.039
.001
.005
.024
.033
CS2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.001
.001
0
0
,0
.001
MgSCK
.147
0
0
0
.120
0
0
0
.092
0
0
0
0
0
0
0
0
0
0
0
-207-
-------
TABLE B3-8
MgS03 DECOMPOSITION WITH CO/H2 (1:1) REDUCTANT
Case
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
Stoich.
.95
.95
.95
.95
1.0
1.0
1.0
1.0
1.05
1.05
1.05
1.05
1.5
1.5
1.5
1.5
20.0
20.0
20.0
20.0
Temp.
(°C)
350
500
700
1,000
350
500
' 700
1,000
350
500
700
1,000
350
500
700
1,000
350
500
700
1,000
AH(^)
\ kg /
- 216
' - 103
17
+ 89
- 224
- 120
- 35
+ 75
- 232
- 137
50
+ 81
- 312
- 265
- 180
- 27
-1,132
- 983
Solid (mole fraction)
MgO
.853
1.0
1.0
1.0
.880
1.0
1.0
1.0
.908
1.0
1.0
1.0
1.0
1.0
MgSO^
.147
0
0
0
.120
0
0
0
.092
0
0
0
C, ,
(gr)
0
0
0
0
0
0
0
0
0
0
0
0
0 0
0
1.0 0
1.0
0
0
0 0
.082 0 .918
Gas (mole fraction)
C02
.482
.423
.398
.366
.483
.432
.407
.371
.483
.440
.406
.322
.498
.481
.429
.352
.247
.089 0 .911 .259
- 442 .197
j
16 I 1.0
0 ! .803 .110
0 0
.019
CO
0
0
.001
.025
H20
.308
-303
.295
.317
0 .308
0
.301
-002.294
.028
0
0
.318
.309
.296
H2
0
0
.001
.012
0
0
.002
.013
0
CHfe
0
0
0
0
0
0
0
0
0
0 JO
.005!. 295 .002
0
.073.327 .01310
i
0
.168 0 '0
.003J.184 .002 0
.032,. 224 .009 0
.100
.001
.032
.330
.48C
.262 .042 0
.501 .052 .153
.347 .213 .107
.091 ; .415. .024
.031 .444 0
-208-
-------
The amount of elemental sulfur that can be formed
is of interest. For CO at a stoichiometry of 1.0, the amount
of gas phase sulfur (82 + S8) is always above 8070 with tempera-
ture of 400°C and lower yielding 907o elemental sulfur. The
other sulfur containing gaseous species are COS and S02. The
remaining COS, S02, and CO are in the proper stoichiometry
to yield elemental sulfur.
For H2 at a stoichiometry of 1.0, the amount of gas
phase sulfur and H2S are considerably less than with CO.
Further processing with removal of sulfur and water would be
required to shift the equilibrium to favor production of more
elemental sulfur.
In the cases where an excess of reducing gas was
used all of the excess H2 tends to go to H2S and all of the
excess CO to COS. The only incentive to run an excess of
reducing gas would be for kinetic considerations.
The heat balance numbers are of interest because it
will be desirable to eliminate or minimize the transfer or
addition of heat. For the inlet conditions chosen (25 "C for
MgS03 and CHU, and 1000°C for CO and H2) it is seen that
methane (see Table B3-2) would require addition of heat even
for an outlet temperature of 350°C. The reactor would be
endothennic under all conditions investigated and would
require addition of heat.
For CO (see Table B3-4) the reactor is exothermic for
exit gas and solid temperatures in the range of 700°C and
below, and endothermic at 1000°C and above. The adiabatic
operating temperature is between 700 and 1000°C.
-209-
-------
For H2 (see Table B3-6) the exothermic operating
range is up through 500°C. The endothermic range includes
700°G and above. The adiabatic operating temperature is
between 500 and 700°C. Heat requirements for mixtures of
H2 and CO (see Table B3-8) fall between those of the pure
components.
The heat balance considerations discussed above
are for special cases. They do serve as a guide for con-
sidering what type process arrangement is feasible.
-210-
-------
BIBLIOGRAPHY
KO-134 Koehler, George R., and Edward J. Dober, "New
England S02 Control Project Final Results", Pre-
sented at the Flue Gas Desulfurization Symposium,
Atlanta, Georgia, November 1974.
ME-121 Meyer, B., Elemental Sulfur, New York, Wiley-
Interscience, 1965.
SC-144 Schwitzgebel, Klaus, and Philip S. Lowell, "Thermo-
dynamic Basis for Existing Experimental Data in
Mg-S02-02 and Ca-S02-02 Systems", Env. Sci. Tech.
7(13), 1147 (1973).
ST-067 Stull, D. R., and H. Prophet, JANAF Thermochemical
Tables, 2nd Edition, NSRDS-NBS 37, Washington,
GPO, 1971.
-211-
-------
TECHNICAL REPORT DATA
(Plccse resd Inurucuuns on ;i'e rsv?rsi before completing!
1. REPORT NO.
EPA-600/7-76-030
3. RECIPIENTS ACCESSION NO.
•i. TITLE AND SU3TITLE
Feasibility of Producing Elemental Sulfur from
Magnesium Sulfite
5. REPORT DATE
October 1976
S. PERFORMING ORGANIZATION CODE
7. AUTHOa(S)
Philips. Lowell, W.E. Corbett, G. D. Brown, and
K.A. Wilde
3. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Radian Corporation
8500 Shoal Creek Boulevard
Austin, Texas 78766
10. PROGRAM ELEMENT NO.
EHB528
11. CONTRACT/GRANT NO.
68-02-1319, Task 31
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE Of REPORT AND PERIOD COVERED
Task Final: 5/75-8/76
4. SPONSORING AGENCY CODE
EPA-ORD
is. SUPPLEMENTARY NOTES jERL-RTP task officer for this report is C. J. Chatlynne,
919/549-8411 Ext 2915, Mail Drop 61.
1o. ABSTRACT
The report gives results of a study to extend potential applications of MgO
flue gas desulfurization-processes by allowing the sulfur to be recovered as elemental
sulfur as well as sulfuric acid. The study considered the feasibility of combining the
exothermic SO2 reduction reaction with the endothermic MgSOS calcination. Prelim-
inary consideration of the reductants carbon monoxide, hydrogen, methane, and
hydrogen sulfide showed that the reaction with SO2 can supply part, or in some cases
all, of the heat of decomposition of MgSOS. Considered in detail were; (1) low-
temperature catalytic decomposition using a commercially available low-Btu
synthetic-gas reductant mixture; and (2) high-temperature noncatalytic decomposition
using a medium-Btu reducing gas from an oxygen-blown gasifier. Complete heat and
material balances for conceptual process designs for the above cases were developed
to identify problems. Recommendations for work required to continue process devel-
opment are given. Problems identified include catalyst physical stability, catalyst/
MgO separation, dust carry-over, and noncatalytic reduction kinetics.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATi Field/Group
Air Pollution
Flue Gases
Desulfurization
Magnesium Oxides
Sulfur
Sulfuric Acid
Sulfur Dioxide
Reduction (Chemis-
try)
Catalysis
Air Pollution Control
Stationary Sources
Magnesium Sulfite
Elemental Sulfur
MagOx Process
13B
2 IB
07A
07B
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (Tnis Report)
Unclassified
21. NO. OF PAGES
216
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |