April, 1944
INDUSTRIAL AND ENGINEERING CHEMISTRY
(22) Hinchley, J. W., Ure, 5. G . M., and Clarke, B. W., Trans. Inst. Chem. Engrs. (London), 3, 24 (1925); J . SOG. Chem. I d . 45, 1T (1926). (23) Jewett, E. E., and Montonna, R. E., Chem. & Met. Eng., 34,86 (1927). (24) Larian, M. G., Trans. A m . Inst. Chem. Engrs., 35, 623 (1939). (25) McMillen, E. L.. and Webber, H. A., IND.ENG.CHEM.,30,708 (1938). (26) Munning, thesis, Mass. Inst. Tech., 1921. (27) Olin, H. L., Morrison, F. W., Rogers, J. S., and Nelson, G. H., Trans. Am. Inst. Chem. Engrs., 18,379 (1926). (28) Phillips, J. M., Trans.Ind. Chem. Engrs., (London), 3, 35 (1926). (29) Pickard, J. A., Ind. Chemist, 4, 186 (1928); "Filtration and Filters", London, Ernest Benn Ltd., 1929. (30) Porter, D. J., IND. ENQ.CHEM.,ANAL.ED.,15, 269 (1943). (31) Rhodes, F. H., IND. ENG.CHEM.,26, 1331 (1934). (32) Ruth, B. F., Ibid., 27, 708, 806 (1935). (33) Ruth, B. F., Montillon, G. H., and Montonna, R. E., Ibid., 25, 76, 153 (1933).
RECOVERY OF SULFUR
329
(34) Ruth, B. F., and Kempe, L. L., Trans. Am. Inst. Chem. Engrs., 33. 34 (1937). (35) Spyry, D. R.,'Chem. & Met. Eng., 15, 198 (1916). (36) IbuZ., 17, 161 (1917). (37) Sperry, D. R., J. IND. ENG.CHEM.,13, 986, 1163 (1921); 18, 276 (1926); 20, 892 (1928). (38) Tattersficld, thesis. Mass. Inst. Tech., 1922. (39) Underwood, A. J. V., Trans. Inst. Chem. Eng7s. (London), 4, 19 (1926); J . SOC.Chem. Ind., 47, 3251' (1928); Ind. Chemist, 4, 463 (1928); in Pickard's "Filtration and Filters", p. 87, London, Ernest Benn Ltd., 1929; Proc. World. Eng. Congr Tokyo, 1929,245,258. (40) Walker, W. H., Lewis, W. K., McAdams, W. H., and Gilliland, E. R., "Principles of Chemical Engineering", New York, McGraw-Hill Book Co., 1937. (41) Waterman, H. I., and Gilse, J. P. M. van, Rec. trav. chim., 43, 757 (1924); 45, 628 (1926). (42) Webber, H. C., and Hershey, R. L., IND.ENQ.CHEM.,18, 341 (1926). (43) Woodward and Edmunds, thesis, Mass. Inst. Tech., 1922.
.
T. F. DOUMANI, R. F. DEERY, AND W. E. BRADLEY
sulfur dioxide in waste gases
Union Oil Company of California, Wilmington, Calif.
Elementary sulfur has been recovered from gases containing sulfur dioxide by a two-step catalytic process operating at atmospheric pressure. Wet gases containing relatively low percentages of sulfur dioxide can be treated. In the first step the sulfur-dioxide-containing gas is mixed with hydrogen and passed through a catalyst bed at temperatures above 300' C. A gel-type catalyst comprising about 25% iron suEde and 75% alumina is effective. In the second step complete conversion to elemental sulfur is accomplished by passing the mixture of hydrogen sulfide and sulfur dioxide from step one over activated alumina at temperatures of 100" to 200"C. The sulfur tends to deposit in the catalyst bed and eventually deactivates the catalyst. The reactivation of the latter is readily accomplished by raising the temperature of the catalyst bed above 500" C.
T
i
HE waste gases from certain types of operation, such as the processing of sulfur-containing crude oils, have appreciable percentages of sulfur dioxide. Owing to the nuisance and consequent possibility of litigation arising from the indiscriminate venting of such gases to the atmosphere, number of industries have devised methods for recovering the sulfur dioxide. Normally the process selected depends on the composition of the gas to be treated, particularly its sulfur dioxide and water content. The methods previously reported for the reclaiming of elernental sulfur from sulfur-dioxide-containing gases involve either concentration of the sulfur dioxide and subsequent reduction, or direct conversion t o sulfur. The following materials have purportedly been used for the conversion step: coke, carbon mbnoxide, hydrocarbons, hydrogen sulfide, and hydrogen. Essentially, the commercial process employing metallurgical coke (9) comprises passing the sulfur dioxide with pure oxygen through an incandescent coke bed to produce carbon dioxide, sulfur vapor, carbon monoxide, and carbon oxysulfide. The latter compound, formed by the apparently rapid reaction of sulfur vapor with carbon monoxide, is catalytically reacted with additional sulfur dioxide to form elemental sulfur. The gas containing sulfur vapor is partially cooled and the sulfur collected by Cottrell precipitators. The thermodynamics of the reduction of sulfur dioxide with carbon, carbon monoxide, carbon oxysulfide, hydrogen, hydrogen sulfide, and methane were reported by Lepsoe (6). H e also gave data (6) on the kinetics of the reduction of sulfur dioxide with coke, carbon monoxide, and carbon oxysulfide.
Carbon monoxide is used commercially in the Imperial Chemical Industries and Bolidens Gruvaktiebolag processes for the direct treatment of dry smelter gases of about 8% sulfur dioxide content, according to Appleby (1). Essentially, these processes comprise the production of carbon monoxide, catalytic reduction of sulfur-dioxide-cdntaining gases by the carbon monoxide, and cooling of the gases and separation of the sulfur by electrostatic precipitation. Coke is the source material for the carbon monoxide, and the latter is reacted with sulfur dioxide a t 250-300" C. in catalyst chambers containing mixed ferric and aluminum oxides. The use of hydrocarbons for the reduction of sulfur dioxide was studied by Yushkevitch and co-workers (IO). They found that methane and sulfur dioxide react at 900" C. in the presence of bauxite to form sulfur, hydrogen sulfide, water, and carbon dioxide, with a yield of sulfur of 89 to 95%, based on the methane consumed. At higher temperatures hydrogen and carbon monoxide are also formed and at lower temperatures the methane reacts incompletely. Light hydrocarbons (benzine) react with sulfur dioxide at 700-800" C. to give 60-9070yields of sulfur. Some work has been reported on the reduction of sulfur dioxide with hydrogen when employing dry gases (9). A! 500800' C. hydrogen was reacted almost completely with sulfur dioxide over various catalysts, principally bauxites, to form sulfur in 40-60% yields; the remaining sulfur formed hydrogen sulfide.
330
INDUSTRIAL AND ENGINEERING CHEMISTRY
A commercial process for the reduction of sulfur dioxide with hydrogen sulfide operates at 282' C., using a catalyst whose composition is not given (4). Sulfur is also produced by passing sulfur dioxide gases into hydrosulfurous acid solutions (Leahy, 8). The experimental work reported in this paper deals with the reduction of sulfur dioxide with hydrogen and the interaction of sulfur dioxide and hydrogen sulfide to form free sulfur. Both wet and dry gases are used. THERMODYNAMlC CONSIDERATlONS
Chemical and thermodynamic equations for the reduction of sulfur dioxide with hydrogen and hydrogen sulfide are given in Table I. Because of the variable composition of sulfur vapor with temperature and concentration, and the difficulty of determining the amounts of each form present, equations are given for possible sulfur forms. Free energies and equilibrium constants for chemical equations that might be involved are given in Table 11. Reactions 1 to 5 (Table I) can be involved in the first step of th? recovery process. Calculations of the percentage conversion for the stoichiometric amounts of reactants shown in these equations indicate that essentially complete conversion to products would be obtained if equilibrium were reached at the reported temperatures of 150" to 500" C. Reactions 6, 7, and 8 (Table I) are probably involved in the second step of the recovery process. The equilibrium constants for these equations indicate that the percentage conversion for the stoichiometric amounts of reactants shown decreases with increase in temperature. Inspection of the free energy value for reaction 6 seems to show that sulfur is not seriously involved in the second step of the process. APPARATUS AND MATERIALS
The apparatus employed for the reduction of sulfur dioxide with hydrogen is shown in. Figure 1. Synthetic sulfur dioxide gases were prepared with nitrogen as a diluent. The oxygen in the commercial nitrogen was removed by reaction with finely divided copper which was held at d$l red heat by a small tube furnace. Theeffect of water was studied by saturating the metered nitrogen with water vapor by passage through 10 feet of S/s-inchdiameter copper coil containing water held at a predetermined temperature. This nitrogen was then freed of entrained water by
TABLIO I.
Vol. 36, No. 4
CHEMICAL A N D THERMODYNAMIC EQUATIONS"
-
*
4Hz (Q) sz ( 8 ) f 4 8 1 0 (8) AFe -63,120 1.74Tln T (1.22)(10-2)T2 (2.1)(10-8)Ta 13.9T 13,800 ' log K = __ 0.88 log T (2.65)(10-3)T (4.6)(10-')2'2 - 3.04 T
2soz (8) f
(1)
-
+
-
+
+
+
(2) %On (Y) 4- 4H2 (Q) I/asa ( 8 ) 4Hz0 ( 8 ) AF' = -86,170 3.48T In T (8.0)(10-3)Tz
log K
+
-
-
-+ 1.76 log T 18,830
(3) 2502 (8) 4Hz (Q) l/4& (8) -k 4Hz0 (g) APo = -888.700 - 3.48Tln T (8.0) (10-9T2 19,400 log K = - 1.75 log T (1.75) (10-3)T T
+
- (2.1)
(lO-')Ta
+ 54.3T
+ (1.75) (10-a)T + (4.60) (l0-7)Tz + 10.88
*
+
+
-
-
- (2.1)
(lO-e)T3
+ 57.9T
+ (4.60) (10-7)TZ - 12.68
( 8 ) -I- Sz ( 8 ) S 2HzS ( 8 ) AFO = -38,400 4- 1.88Tln T 4- (3.30) (10-a)Tz
(4) 2Hz
- (7.4) (10-7)TS + 3.3T
- 9.5 log T - (7.2) (10-4)T + (1.62) ( l O - 7 ) T Z - 0.72 ( 5 ) SO1 (8) + 3Hz (8) HiS (Q) + 2Hz0 -50,760+0.07TIn T + (7.75) (1O-a)TJ - (1.42) (10-e)Ta + 9.352' 8 400 T
log K =
(8)
=
log K =
11 100 - 0.03510g T T
- (1.69) (1O-3)T + (3.1) (lO-')T2 - 2.04
*+
+
+
(6) SOz (8) 2HzS ( 8 ) 2Hz0 (B) s/zSz (Q) AFO = 6870 - 2.7621 In T (2.8) (10-a)Tz - (3.1) (10-7)Ta log K =
- 1503 ~
+ 1.38 log T - (8.13) (10-4) T + (6.78) (10-8)
+
+
(7) SO2 (0) 2HzS (Q) 2H20 ( 8 ) */zSa (0) A F " = -27,730 5.36Tln T (3.5) (1O-Z)TS log K =
(8)
-
6070
+ 2.70 log T 'p (8)
+ 2HzS (Q)
A F O = -31,530 log a
K
-
+ 3.4T T'
- 0.74
- (3.1) (lO-7)Ta f B4.2T
+ (7.67) (lO-')T + (6.78) (lO-s)T* - 14.04
Sa - 5.36TIn2Hz0 T - (3.5) (10-3)~z - (3.1) (10-7)~a+ 69.5T (8)
'18
(Q)
+ 2.69 log T + (7.66) (10-4)T + (6.78) T
= 6880
(10-8)TZ
- 15.2
Free energies of formation from International Critical Tables, Vol. VII.
P. 236 (1930).
Figure 1. Diagram of the Apparatus for Reduction of Sulfur Dioxide w i t h Hydrogen
-
+a
1, 28. Nitrogen 2. Furnace 3. Copper 13 4, 11, 26, 32. Flowmeters 5. 9. Water 6. Copper coils 7. .. Acetone 8. Condenser 10. Electrically jheated and insulated 12. Hydrogen 13. Sulfur dioxide 14. Catalyst 15. Catalyst tube 16. Thermocouple well 17. Glass wool 18. Polythionic acids 19. Iz-KI solution 20,21.' 1-liter flasks 22. Aqueous KOH eolution 23. Ice bath 24. Concentrated HzSOd
18
-
- I
7 U
1 6 -
33
25, 27. Calcium chloride 29. Hz oxidation furnace 30. Cupric oxide 31. Dehydrite absorption tube 33. To vacuum line 34.35. &liter flasks
INDUSTRIAL A N D ENGINEERING CHEMISTRY
April, 1944
TABLE11. FREEENERGIES AND EQUILIBRIUM CONSTANTS Temp.,
C.
150 200 250 300 350 400 450 500 150 200 250 300 350 400 450 500
AF' -Equation -59.700 -59,100 -58,500 -57,900 -57,200 -56500 -55:800 -55,000 ----Equation -45.340 -44.550 -43,730 -42,880 -41.970 4 1,080 -40,140 - 39.100
-
K
IX X X X X X X X 53.6 X 4.0 X 1.9 x 2.2 x 5.5 x 2.2 x 1.4 x 1.3 X
6 8 2 2 3.2 1.3 1.1 2.3 7.6 3.5
AF'
108' 102' 1024 1021 lo*@ 10" 10'8 10'6 lO*a
1020 101' 101' 1014 iola 101s 10"
-Equation -70,800 -69,030 -67,270 -65,480 -62,700 -61,810 -60,020 -57,400 --Equation 1,740 1,060 370 320 1,040 1,720 2,430 - 3,080
-
K
2X lOs@ X lO*l X lOzS X 102' X 1022 X 10s X 1018 X 101' 6--8.0 -3.2 -1.4 1.3 2.3 3.7 5.5 7.8
4 8 1.3 1.0 1.0 1.2 1.4 1.7
AFo K -Equation 3-71,800 1.3 X -69,900 2 X -67,900 3 X -65,900 1 4 X -63,900 3 X -61,900 1.3 X -60,000 1.4 X -58,000 2.5 X ---Equation 715,000 5 . 6 X - 13,850 2 . 3 X 12,760 2 . 0 x -11,640 3.0 x 10,660 5 . 8 X 1.5 x -9,710 -8 880 4 . 8 X 1.8 x -7:760
-
passage through another copper coil held a t the boiling point of acetone, a stopcock a t the bottom of the coil being employed to remove any liquid water condensed from the nitrogen. The nitrogen, hydrogen, and sulfur dioxide were then introduced through an electrically heated section of stainless steel pipe into the reaction furnace containing the catalyst. The exit end of the catalyst tube was heated electrically to prevent condensation of sulfur. The condensation of sulfur took place almost quantitatively in a removable 2-inch section of glass tubing, the remaining sulfur being collected in the glass wool. The rest of the apparatus pictured in Figure 1 was used to determine the hydrogen sulfide, sulfur dioxide, and hydrogen in the exit gases. The apparatus employed for the reaction of sulfur dioxide with hydrogen sulfide was simpler than for the reduction of sulfur dioxide with hydrogen. Equipment for the determination of hydrogen was eliminated (Figure I), and the following modification was used: Hydrogen sulfide was admitted into the catalyst tube 0.25 inch below the catalyst bed through four small holes in the thermocouple well. This was found to be necessary to avoid reaction of the hydrogen sulfide with the sulfur dioxide before the mixture reached the catalyst. This thermal reaction is much slower than the desired catalytic one. The compressed gases used were obtained from the following sources : hydrogen sulfide (Ohio Chemical and Manufacturing Com any), refrigeration grade sulfur dioxide (Dow Chemical Companyp, and the nitrogen and hydrogen (Linde Air Products Company). METHOD OF ANALYSIS
Analyses for sulfur dioxide and hydrogen sulfide were usually made simultaneously. A definite volume of the gas containing hydrogen sulfide, sulfur dioxide, hydrogen, and inert gases was bubbled through a definite volume of standard iodine solution, and the excess iodine was titrated with standard sodium thiosulfate solution. After subtracting a blank correction (obtained by bubbling into an e ual volume of the standard iodine solution a volume of air e q u j to the volume of inert gases i s the sample taken for analysis), the equivalents of hydrogen sulfide plus sulfur dioxide could be determined. Since twice as much acid is formed when sulfur dioxide is titrated with iodine as when hydrogen sulfide is titrated, the amount of each can be calculated from the two simultaneous equations; one involves the e uivaacid lents of iodine consumed and the other the e uivalents formed (titrated with standard sodium hydroxile solution). The equations involved are as follows:
,
lo*' 10'2 10'8 1096 10'2 1020 10" 1016 10' 10' 106 104 108 io8 102 102
AFo -Equation -31,600 -30,700 -29,700 -28,700 -27,700 -26600 -25:600 -24,600 -Equation 16,500 15,150 13.760 - 12,440 11,160 -9,910 8,680 -7,560
--
SOs
+ Is + 2H20 HzSOI + 2HI + Is S + 2HI
4-X X X X X X X X
101'
1014
1012 101' 109 10' 10'
The reduction of sulfur dioxide with hydrogen was first carried out with a bauxite which analyzed as follows:
IO'
8--
3.4 5.0 3.8 5.8 8.3 1.7 4.2 1.4
x 108 X 10' X 10' X 10' x loa
X 108 X 108 x 102
TABLE 111. REDUCTION OF SULFURDIOXIDEWITH HYDROGBIU AT ATMOSPHERIC PRESSURE
-+
-*
Analysee for hydrogen were executed by oxidation of a definite volume of the feed and product to water by means of copper oxide which was maintained a t a dull red heat. Both the hydrogen sulfide and sulfur dioxide were first removed by means of an alkaline wash solution followed by partial deh dration of the sample by cooling to O b C., followed by further z y i n g with concentrated sulfuric acid, and finally with calcium chloride. The amount of water formed was determined gravimetrically by a b sorption in tubes containing magnesium perchtorate. When the amount of hydrogen sulfide was low in comparison with the amount of sulfur dioxide, it was determined directly by titration with standard iodine solution after converting the sulfur dioxide to a sulfurous addition compound with formaldehyde in aqueous zinc acetate solution, according to the improved method of Wollak (8).
K
2,2 1.6 2.6 8.5 5.4 4.6 5.6 9.3
Below 325' C. the rcduction of sulfur dioxide (2.5%) with hydrogen (5.0%) in the presence of nitrogeB (92.5%) was not detectable. At 370" C. the reduction proceeded slowly, and a t 425' to 480' C., quite rapidly. Furthermore, the catalyst improved considerably after several hours of use, as evidenoed by the increased conversions with time and by the fact that the used catalyst was effective for certain space velocities and temperatures at which the fresh catalyst effected no detectable conversion. Inspection of the used catalyst showed that a portion a t the exit end of the catalyst tube had darkened considerably owing to the formation of iron sulfide, which was proved by analysis. A negative sulfide test was obtained with a portion of the catalyst a t the entrance end of the catalyst. The observation that the sulfide, and not the oxide, of the metal is the catalyst is of practical value, for regeneration of the sulfide catalyst is not necessary; if the metal: oxides were catalysts, regeneration would be necessary because of the transformation of the oxide to the sulfide. This farmation of iron sulfide from the iron oxide of the catalyst explains why, at first, the iron oxide catalysts show no activity and gradually increase with passage of the gas a t temperatures of about 325' C. This initial period of no activity a t these low temperatures occurs when no iron sulfide is present, and the period of slowly increasing activity represents the time when the iron oxide is being transformed to iron sulfide. This same effect was noticed with all the iron-oxide-containing catalysts tried. To convert the iron oxide in the catalysts to the sulfide, prior t o tkeir use for the reduction, they were pretreated with hydrogen sulfide a t about 500' C. The reaction between ferric oxide and hydrogen sulfide was studied by Sayce (7). At 300" C. iron disulfide (FeSz) is obtained exclusively as the product; between 400"and 500" C. a product is
01
H2S
331
REDUCTION OF SULFUR DIOXIDE WITH HYDROGEN
sox HI
HIS
B. 25% FezOa-750/ Ale08 Gel Catalysta Pret:eated with He8 At 316O C. A . Bauxite Catdysta Gaseous Gaseous Gaseous roductec~ Gaseous Eom roducte wet feedb more % feedb, feed$, mole % Dry feed Wet feedd mole % mole % REACTION TmMP., REACTION TEMP.,482' C. (900° F.) 316' C. (600' F.) 3.10 1.10 1.41 3.0 0.2 4.87 0.16 0.42 9.3 0.07 0.00 0.65 1.05 0.0 2.4 97.12 92.03 98.09 87.7 97.33 100.00 100.00 100.00 100.0 1oo.00~
-
-
-
REACTION TEMP R~ACTION TEMP..482' C. (gooo F.) 482' C. (900° F.7 so1 2.80 0.00 0.24 3.1 0.7 HI 9.85 0.25 0.56 7.1 0.1 HIS 0.00 2.37 2.66 0 0 1.3: Nl 87.85 97.38 96.64 __ 89.8 97.9 100.00 100.00 100.00 100.0 loo 0 12-20 standard Tyler screen. 6 Space velocity, 592 volumes of dry feed per unit volume of catalyst per hour at standard conditions. Also contained some free sulfur and polythionic acids in aqueous cond Feed contained 13.3% water vapor. densate.
-
-
-
INDUSTRIAL AND ENGINEERING CHEMISTRY
332
Vol. 36, No. 4
composition of certain catalytic regeneration gases from the TABLE IV. REDUCTION OF SULFUR DIOXIDEWITH HYDROGEN treatment of high-sulfur hydrocarbon feed stocks. The gases PRESSURE SULFIDEAT ATMOSPHERIC were sufficiently dilute so that the reaction conditions and cataGaseous Product, lysts reported should be effective for the reduction of other sulfur Feed, % b y Vol. Mole % Temp., dioxide gases. HIS SOz Hz0 NQ F. (" C.) HnS 902 To recover free sulfur from a sulfur-dioxide-containing gas Bauxite Catalysta, 55O-Vo:. Space Velocityb 2.0 1.0 . . . . 97.0 200( 93.3) 0.02 0.01 when no hydrogen sulfide is present, or insufficient for reduction, 300(148.9) 0.01 0.02 a two-step process can be used. The reactions involved may be 400(204.4) 0.20 0.22 500(260.0) 0.58 0.24 represented by the following equations: 600(315.6) 0.64 0.36 Activated Alumina Catalyst" (Alorco), 650-Val. Space Velocityb 0.75 16.8 80.95 ZOO( 93.3) 0.00 0.00
1.5
*
300(148.9) 350(176.7) 400(204.4) 450(232.2) 500(260.0) 550(287.8) 600(315.6)
0.00 0.00 0.00 0.18
0.00 0.00 0.00 0.09
0.50 0.68
0.28 0.35
0.91
0.43
12-20 standard Tyler mesh. b Of feed per unit volume of catalyst pet hour a t standard conditions.
obtained which is soluble in hydrochloric acid with liberation of sulfur. This material was regarded as a mixture of FeS and FeSz, although i t is believed that it might be FeaSr. CATALYSTS. Various metal oxides and salts were tried as catalysts; those containing iron sulfide were found to be most economical and effective. Gel catalysts, consisting of ferric and aluminum oxides and pretreated with hydrogen sulfide to convert the iron oxide to the sulfide, are particularly valuable for the reduction of sulfur dioxide with hydrogen. These catalysts may be used at temperatures as low as 325" C. a t atmospheric pressure, with space velocities of over 600 volumes of gas per volume of catalyst per hour (Table IIIB). Both wet and dry gases containing low percentages of sulfur dioxide can be catalytically treated, so that all of the sulfur dioxide is reduced to hydrogen sulfide and free sulfur, or the amount of hydrogen can be regulated so that the exit gas contains the required amount of hydrogen sulfide for its stoichiometrical interaction with sulfur dioxide to form free sulfur (discussed later). At temperatures above 325' C., less active catalysts than the iron-alumina gel types can be used, such as iron sulfide on various aluminas. Bauxites of high iron content are quite effective after converting the latter to the sulfide (Table IIIA). REDUCTION OF SULFUR DIOXIDE WITH HYDROGEN SULFIDE
Many types of catalysts were used for the reduction of sulfur dioxide with hydrogen sulfide. Those containing aluminum oxide were superior to all others tried, being effective even with wet gases. Bauxite (Table IV) and precipitated alumina were used a t 100" to 200" C., but these catalysts were not so good as highly purified and specially treated aluminas, such as Alorco, grade A (Table IV). For all these catalysts complete conversion was obtained below about 210" C. a t atmospheric pressure. The activity of these catalysts decreases with the reaction time as a result of sulfur deposition. For example, when a gaseous mixture was passed through Alorco, grade A, under the conditions described in Table IV, the exit gases analyzed 0.02 to 0.04% of sulfur dioxide and hydrogen sulfide, respectively, for the first 30 hours; then the reaction was quite incomplete. At the end of this time about 90% of the catalyst was covered with sulfur, the bottom portion of the alumina being more densely covered than the upper part. The following method was effective in regenerating the catalysts from deposited sulfur: The reaction tube with the spent catalyst was raised in temperature to 500" C. and held there for about 2 hours. By this means the sulfur was vaporized from the catalyst and collected as molten sulfur. The last traces of sulfur can be stripped from the catalyst by means of an inert gas. DISCUSSION
In this work the sulfur dioxide, hydrogen sulfide, and water vapor contents of the gases reported were of the approximate
+ 4Hz e Sz + 4HzO 2H2 + Sz 2HzS 3/2S2 + 2H20 e 2HzS + SO2 2HB + SO2 35 + 2H20 2SOz
(1)
(2)
(3) (4)
Step 1 (Equations 1, 2, and 3) involves the reduction of all the sulfur dioxide to hydrogen sulfide with hydrogen a t 315" C. (600" F.) or higher, using a catalyst containing iron sulfide; or it involves the reduction of sufficient sulfur dioxide in the gas so that the stoichiometric quantity of hydrogen sulfide is formed for reduction of sulfur dioxide (Table I I I B a t 482" C.). Step 2 (Equation 4) involves the passage of a gaseous mixture of hydrogen sulfide and sulfur dioxide in the molar ratio of 2 to 1 over activated alumina at temperatures below about 200' C. (Table IV). The sulfur tends to deposit in the catalyst bed and eventually deactivates the catalyst. Reactivation of the latter is readily accomplished by raising the temperature of the catalyst bed above 500" C. It was found impossible to reduce sulfur dioxide with hydrogen directly to sulfur without simultaneous formation of hydrogen sulfide. Whensulfur dioxideis reactedwithsufficienthydrogen to form sulfur (Equation l), a mixture of sulfur dioxide, hydrogen sulfide (Equation 2), and sulfur is obtained. Furthermore, when wet gases are reduced, the interaction of sulfur with the water vapor originally present, plus that produced in the reduction, causes the formation of hydrogen sulfide and sulfur dioxide (Equation 3). This is of no serious consequence in the two-step process since these two compounds are formed in the molar proportions required for the second step (Equation 4). Any oxygen associated with the sulfur dioxide gas should be avoided as much as possible, since i t reacts with the iron sulfide in the catalysts and converts it to the oxide. If the amount of oxygen is small, the hydrogen sulfide formed in the reduction will convert the iron oxide to the sulfide; however, this is at the expense of a correspondingly greater amount of hydrogen. When sulfur dioxide gases which contain relatively large amounts of water vapor are reduced with coke, the consumption of the latter substance can become very large due to the reaction of water vapor with the coke. ACKNOWLEDGMENT
The writers wish to thank the Union Oil Company of California for permission to publish this paper. LITERATURE CITED
(1) Appleby, M. P., J . Soo. Chem. Ind., 56, 139T (1937). (2) Kirkpatrick, 9. P.,Chem. & Met. Eng., 45,483 (1939). (3) Leahy, M.J., Refiner Natural Gasoline Mfr., 15,276 (1936). (4) Lee, J. A., Chem. & M e t . Eng., 49,80 (1942). (5) Lepsoe, R.,IND.ENG.CHEM.,30,92 (1938). (6) Ibid., 32, 910 (1940). (7) Sayce, L.A., J . Chem. SOC.,1929,2002. (8) Wollak, R.,8. anal. Chem., 77, 401 (1929). (9) Yushkevitch, N. F.,Karzhavin, V. A., Avdeeva, A. V., and Kreohemov, T. F., J . Chem. Ind. (U.S.S.R.), 1933,No.8,60. (10) Yushkevitoh, N. F., Karrhavin, V. A., Avdeeva, A. V., and Nikol'skaya, Yu. P., Ibid., 1934,No. 2,33. P B E ~ ~ N Tbefore E D the Division of Petroleum Chemistry at the 106th Meeting of the AMBRICAN CH~MICA SOCIBTY, L Pittsburgh, Pa.
