KO\-., 1 0 2 0
T H E J O U R N A L OF I J D U S T R I A L AiVD E N G I N E E R I N G C H E M I S T R Y
order t o decompose completely t h e calcium picrate during t h e steam distillation. I t was obviously desirable t o know what happened t o this excess of bleaching powder during t h e steam distillation. T h e following experiments establish t h e approximate rate a t which active chlorine disappears during steam distillat*on, first, in t h e residue left after t h e preparation (of chloropicrin, and, second, in a more concentrated solution containing t h e same concentration of bleaching powder as t h a t originally present in t h e chloropicrin stills. Finally, analyses were made t o determine t h e concentration of active chlorine remaining in each of t h e t e n plant stills a t t h e end of t h e distillation which removed t h e chloropicrin from t h e sludge. TABLEI-PERCENTAGES After Distilling off Chloropicrin 0.0330 0,0505 0.0459
OF ACTIVII CHLORIKE After 0.5 H r . Steam After 1 5 Hrs. Steam Distillation Distillation 0.0262 0.0157 0.0172 0.0218 0.0160 0.0242
1069
proximate d a t a for the rate a t which bleaching powder is decomposed by steam. When higher initial concentrations of t h e hypochlorite were employed, t h e rate of decomposition with steam was noticeably higher. The initial concentrations shown in Table I1 correspond t o those employed in t h e plant. TABLE11-PERCENTAGES INITIAL 6.98 6.18
O F ACTIVE CHI.ORINB After 1 Hr. Steam After 2 Hrs. Steam Distillation Distillation 2.52 ’ 0.714 3.21 0.747
Samples of t h e sludge left in t h e chloropicrin stills a t t h e end of t h e distillat.ion were analyzed. T h e conditions under which the various runs were made and t h e concentrations of t h e active chlorine based upon the weight of t h e sludge are shown in Table 111. The results show t h a t t h e concentration O E active chlorine is very low in all cases. TABLE I11 No. of Max. Temp. during Still Distillation 1 . . . . . . . . . . . . . . . . 102.0 2 . . . . . . . . . . . . . . . . 103.5 3 . . . . . . . . . . . . . . . . 104.0 4 . . . . . . . . . . . . . . . . 105.0 5 . . . . . . . . . . . . . . . . 100.0 6 . . . . . . . . . . . . . . . . 104.0 7 . . . . . . . . . . . . . . . 104.0 8 . . . . . . . . . . . . . . . . 105.0 9 . . . . . . . . . . . . . . . . 101.0 10 . . . . . . . . . . . . . . . . 104.0
The residue left in the flask when chloropicrin was prepared as described in t h e preceding paper was analyzed for available chlorine b y t h e iodometric method The sludge was then subjected t o steam distillation for periods of one-half and one and onehalf hours, respectively. T h e analyses furnish ap-
Percentage Active Cl in Sludee 0.28 0.24 0.28 0.28 0.285 0.32 0.29 0.29 0.26 0.22
ORIGINAL PAPERS A STUDY OF T H E REACTIONS OF COAL SULFUR IN T H E COKING PROCESS1~2
cluded t h a t a somewhat greater percentage of t h e organic sulfur was volatilized t h a n t h e inorganic, but no very sharply cut difference was noticeable. By Alfred R. Powell J. R. Campbell’ stated t h a t most of t h e coal sulfur EXPERIMENT STATIOX,BUREAUOF MINES, PITTSBURGH, PA. was present as pyrite, t h a t 42 per cent of this was PREVIOUS INVESTIGATIONS volatilized during t h e coking process, and t h a t t h e The sulfur of coal has now been definitely established remainder was left in t h e coke as pyrrhotite. He also t o exist as pyrite or marcasite, FeS2, organic sulfur, stated t h a t most of t h e organic sulfur was retained and a rather small amount of sulfates, a n d accurate in t h e coke. methods have been devised for t h e determination of Some tentative conclusions have been drawn by S. these different forms.3 T h e behavior of each form W. Parr2 concerning sulfur in t h e coking process during t h e destructive distillation of coal is of theoreti- from his work on t h e low temperature carbonization cal interest, as well as of t h e greatest technical impor- of coal. He states t h a t “the organic sulfur in t h e tance, since t h e reactions of t h e coal sulfur will deter- raw coal and half of t h e sulfur of FeS2 is for t h e most mine t h e percentage of sulfur left in t h e coke and t h e part discharged a t relatively low temperatures.” This nature and t h e amount of t h e sulfur compounds in occurs at about 500’ C. At about 7 o o o , t h e sulfur the by-products. of t h e FeS formed from t h e FeSz is taken up by carbon, The percentage of coal sulfur expelled during t h e leaving free iron in t h e coke. coking process varies over wide limits, and this variaI t therefore appears t h a t different investigators tion ha:; always been supposed t o be due t o different have obtained widely divergent results, and have relative percentages of sulfur forms in t h e ~ 0 a l . 4 advanced many different theories as t o t h e reactions LcI’Calliim states6 t h a t he separated coal into different undergone by t h e coal sulfur during carbonization. fractions b y specific gravity methods, thereby securing The present investigation was undertaken with t h e a partial separation of t h e organic and t h e inorganic idea of carbonizing a variety of coals under carefully sulfur. From coking tests on these fractions he con- controlled conditions, and studying t h e character and ’ Published b y permission of the Director of the U. s. Bureau of Mines. amount of t h e various sulfur compounds formed. 2 Presented by title a t the 59th Meeting of the American Chemical Since these analyses were t o be made over every Society, St. Louis, Ma., April 12 t o 16, 1920. period of t h e carbonization process, a complete and 3 A . R . Powell and S . W. Parr, “A Study of the Forms in Which Sulfur Occurs in Coal,” University of Illinois, Engineering Experiment Station, detailed picture could be obtained of t h e changes occurBuZZrlin 111 (1919). ring in t h e coal sulfur. Fulton, “Treatise on Coke,’’ International Textbook Co., Scranton, Pa., 1905,pp, 38-40. 5 Chent. Eng., 11 (1910), 27.
1 2
Bull A m I m t Mznzng Ens., 1916, 177 I b i d . , 1919, 1807.
1070
T H E J O U R N A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
12,
No.
II
Tor filfer and
Ncohoh"
FIG.I-APPARATUS USED IN
THE
STUDY OF COALSULPUR CONVER~ION IN THE COKING P~ocSss
APPARATUS
The apparatus was designed t o coke about 5 g. of coal, and was so arranged t h a t the temperature of t h e coking chamber could be controlled a t all times t o within a few degrees. A t first thought, criticism might be made of this procedure, in t h a t the quantity of coal used was so small. However, when i t is considered t h a t t h e primary coking reaction was desired, without the secondary effects produced by t h e travel of the gases through t h e hot coking mass, it is seen t h a t t h e use of such a small quantity is essential. Furthermore, uniform temperature and close temperat u r e control are impossible with a large quantity of coal. Details of the coking apparatus as finally perfected are shown in Fig. I . The coking chamber consisted of a five-eighths inch fused silica tube. The coking charge of j g. of the powdered coal was evenly distributed over 4 in. of t h e middle portion, with asbestos plugs a t each end of the charge. A cotton plug was placed a t t h e outlet end of the t u b e t o act as a t a r filter. T h e charge was heated by a nichrome-wound tube furnace, which consisted of an alundum tube, of a somewhat greater diameter t h a n the tube containing the charge, around which was wound nichrome wire of such a size and length t h a t I O O O O could be easily attained. The resistance wire was set in a mixture of alundum cement and water glass, and heat insulation was obtained by wrapping around t h e furnace several layers of heavy asbestos paper. Temperature readings were taken by means of a Hoskins thermocouple, the end of which rested against t h e outside of the coking tube a t the point where t h e coal charge had been placed. Since t h e temperature of the run was maintained for 2 hrs. in t h e furnace, the temperature on the outside and inside of t h e silica tube must have been t h e same. A slow stream of d r y carbon dioxide was passed into the inlet end of t h e coking tube during certain periods of the run t o act as a rinsing agent. This carbon dioxide was generated by t h e action of hydrochloric acid on calcium carbonate, and was passed through sodium bicarbonate and calcium chloride.
This procedure was followed only a t the beginning of a run t o insure a non-oxidizing atmosphere over t h e coking charge, and a t the end of the run t o rinse o u t t h e last traces of volatile sulfur compounds. The outlet and t a r filter end of the tube was surrounded by a copper water bath of the design used by t h e Steel Corporation for the same purpose. By means of this device t h e temperature of t h e t a r filter was a t all times kept a t goo. The gas from t h e outlet end of the tube passed directly into a flask containing ammoniacal cadmium chloride which removed t h e hydrogen sulfide. T h e gas then passed through a n absorption bulb containing absolute alcoholic potash, by means of which t h e presence of carbon bisulfide could be established. This last absorption bulb was replaced in certain runs by a Referee's total sulfur apparatus. METHOD O F P R O C E D U R E
After t h e charge of coal had been placed in t h e t u b e and t h e steam bath around the t a r filter brought up t o the required temperature, a stream of carbon dioxide was started through t h e tube, and a t the same time t h e heating current was turned on. After a few minutes t h e carbon dioxide was shut oft', and the electric current was carefully regulated by means of a rheostat so as t o bring the temperature slowly up t o t h a t required for t h e run. I n t h e case of t h e higher t e m perature runs this gradual increase sometimes required as much as one hour. S U L F U R I N TAR-The run was continued for 2 hrs. a t t h e temperature desired, this length of time being considered necessary t o bring t h e reactions t o equilibrium. The current was then shut off and carbon dioxide again run through for a few minutes. T h e coking tube was removed from t h e furnace, and as much of t h e tar as possible taken out for a sulfur analysis, which was made by the use of Eschka mixture. This determination must be considered only approximate, but, since i t involved such a small portion of t h e coal sulfur, was sufficiently accurate for this investigation. SULFUR I N coKE-The coke was shoved out of t h e tube by means of a long iron rod, separated as thoroughly as possible from the asbestos plugs, and ground
Nov., 1920
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
1071
TABLEI-SOURCE A N D PROXIMATE ANALYSISOF COALSUSED IN EXPERIMENTS
No. 21 100
20507 18847 23066 33945
'
I
NAME Tennessee Pocahontas Pittsburgh Upper Freeport Joliet
Mois- Volatile Fixed ture Matter Carbon
SOURCE
Ash
Sulfur
10,'OO 7.27
1.38 1.18
,
coal. As charged a t coke ovens, Illinois Steel Co., Joliet 33819 Raw Vandalia Vandalia, Ind. 33820 Washed Vandalia Vandalia, Ind. Same as above h u t washed All samples were in an air-dry condition.
2.35 2.48
*
in a porcelain mortar t o loo-mesh size. I t was then placed in a flask so arranged t h a t a current of hydrogen could be passed through and t h e outgoing gas bubbled through ammoniacal cadmium chloride. One hundred cc. of a I : I mixture of concentrated hydrochloric acid and water were poured over t h e coke and t h e contents of t h e flask heated t o boiling, for about 1 5 min., while a brisk stream of hydrogen was passed through. The hydrogen sulfide given off was a measure of t h e metallic sulfide content of t h e coke, t h a t is, the ferrous sulfide and pyrrhotite. It was estimated by treating t h e cadmium sulfide precipitate with a n excess of hydrochloric acid and standard iodine solution, and titrating back with thiosulfate. A sulfur analysis of t h e hydrochloric acid extract filtered off from t h e coke gave t h e sulfate content of t h e coke. The coke was then treated with I : 3 mixt u r e of concentrated nitric acid and water, and allowed t o stand a t room temperature for 24 hrs. The extract was filtered, t h e nitric acid evaporated off, and sulfur a n d iron determined in t h e resulting residue. By this means t h e undecomposed pyrite of t h e coke was figured. The iron value proved t o be a good means of checking t h e sulfur, since t h e iron-sulfur ratio proved whether or not t h e nitric acid had selectively extracted pyrite. I n most cases t h e residual coke was discarded, t h e sulfur in i t being calculated b y difference, b u t in a few runs t h e sulfur was determined directly. I n all such cases t h e sum of t h e sulfur determinations checked very closely with t h e total sulfur of t h e coal. S U L F U R I N V O L A T I L E MATTER-The sulfur which had been given off from t h e coal in t h e form of hydrogen sulfide had been caught in t h e ammoniacal cadmium chloride flask of t h e coking apparatus. It was estimated as just described under t h e determination of metallic sulfide in t h e coke. Carbon bisulfide in the volatile matter was tested for by boiling t h e contents of t h e alcoholic potash absorption bulb, acidifying with acetic acid, and adding copper acetate. The presence of a reddish brown precipitate would indicate carbon bisulfide. D E S C R I P T I O N O F COALS U S E D
Eight coals were used during these investigations. Table I gives t h e laboratory number, the name under which t h e coal will be referred t o hereafter, a general description of t h e source, and t h e proximate analysis of each coal. II--As.%LYsIsOF S K L F c X FORMS (Values given in per cent, air-dry basis) No. Coal, Total Pyritic Sulfate 2 1 IOO--Tennessee. . . . . . . . . . . 4.25 1.75 20507--Pocahontas. ......... 0.56 0.08 18847-Pittsburgh. .......... 1.72 0.79 1.21 23066-Upper Freeport. 0.47 33945--Jdiet. . . . . . . . . . . . . . . . 0 . 8 2 0.26 33819--Raw Vandalia. ....... 1.38 0.70 3382O--Washed Vandalia.. ... 1.18 0.25 TAI3l.G
.....
Organic 1.79 0.47 0.70 0.67 0.56 0.65 0.90
38.05 39.52
49.60 50.73
A complete analysis, b y t h e method of Powell and Parr,' was made on all t h e coals for t h e different forms of sulfur present. The results are given in Table 11. E X P E R I 11E KT A L R E S U L.T S
PYRITE-Since pyrite is one of t h e most important sulfur constituents of coal, i t was necessary t o study its behavior a t different temperatures before going ahead with t h e study of coal itself. Very little pyrite was decomposed up t o joo', but a t I O O O O t h e decomposition was complete. Table I11 shows t h e changes in t h e forms of sulfur in mineralogical pyrite between 0' and 1000'. EFFECT O F TEMPERATURE ON
TABLE111-ANALYSIS
OF
PYRITEBEFORE AT
loooo c.
Pyritic Sulfur.. . . . . . . . . . . . . . . . . . Sulfate Sulfur.. . . . . . . . . . . . . . . . . Free Sulfur. Sulfide Sulfur.. Sulfur as HzS..
.................... ................. ................. TOTAL .......................
AND
00 c. 4R. 52 0.16 0.00 0.00 0.00 48.68
AFTER
DBCOMPOSITION 1000*
c.
0.00 0.00
21.88 24.24
7.56
48.68
These results point pretty clearly t o t h e following reaction: FeSz = FeS 4-S The sulfide content of t h e decomposed pyrite is practically in quantitative accordance with this reaction. Owing t o t h e presence of moisture or other hydrogen yielding bodies, a small part of t h e free sulfur has been converted into hydrogen sulfide. After pyrite has been heated t h e residue is almost invariably magnetic. This is not due t o t h e presence of free iron but t o t h e fact t h a t t h e residue is not plain ferrous sulfide, but pyrrhotite, or magnetic sulfide of iron. It has been shown t h a t pyrrhotite is not a definite compound, but is a solid solution of sulfur in ferrous sulfide.2 The amount of sulfur which is present in t h e sulfide is determined b y t h e temperature and t h e partial pressure of free sulfur over it. I n a n atmosphere of hydrogen sulfide, t h e maximum amount of sulfur retained in ferrous sulfide is 6.0 per cent at 6 0 0 ' . At 1000' t h e percentage is 3.5 per cent, and a t 1300'it is 2.0 per cent. A t 1000' t h e partial pressure of free sulfur in an atmosphere of hydrogen sulfide is 7 0 mm., or a little less t h a n one-tenth of an atmosphere. This amount of free sulfur in t h e space above t h e pyrrhotite holds its content of dissolved sulfur a t 3 . j per cent, as has been stated. If, however, as is true in actual practice, t h e amount of hydrogen sulfide, and therefore t h e amount of free sulfur, present is very small, t h e percentage of sulfur dissolved in t h e ferrous sulfide is very much less t h a n 3.5 per cent. When t h e residue from t h e heated pyrite was dissolved in acid, the sulfur left undissolved and in t h e free state was denoted by a very slight cloudiness of the solution. The 1 LOG. 2
cit.
E. T. Allen, J. L. Crenshaw and John Johnston, "The Mineral
Sulfides of Iron," A m . J . Sci., 33 (1912), 169
I072
T H E J O C R N A L O F I Y D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
indications were t h a t i t would amount t o only a very small fraction of one per cent. Under these circumstances, it can be neglected for quantitative purposes, although it explains t h e weakly magnetic character of t h e pyrite residue. Because of these considerations, all t h e calculations of this investigation have been based on t h e decom-position of pyrite t o form ferrous sulfide and free sulfur, b u t i t must be remembered t h a t a small p a r t of t h e sulfur remains in t h e ferrous sulfide in t h e forin of a solid solution. N o free iron existed in t h e residue from decomposed pyrite or in finished coke, as shown by microscopic tests. When a n acid solution of a copper salt was added t o t h e finely powdered substance no metallic copper could be detected under t h e microscope, thus proving the absence of free iron. D E C O M P O S I T I O N O F P Y R I T E - C O A L MIXTURE-A further study of t h e decomposition products of pyrite was made, in which t h e pyrite was incorporated with a coal of known composition, so t h a t all secondary reactions could be noted. The mixture contained 50 per cent of mineralogical pyrite and 50 per cent of Tennessee coal. T h e results of this test are given in Table IV. TABLEIV-DECOMPOSITION OF PYRITE-COAL MIXTURE 00
Pyritic Sulfur. . . . . . . . . . . . . . . . . . . 25.14 Sulfate Sulfur.. . . . . . . . . . . . . . . . . . 0.39 Organic Sulfur.. . . . . . . . . . . . . . . . . 0.90 Free Sulfur.. . . . . . . . . . . . . . . . . . . . 0.00 Sulfide Sulfur.. . . . . . . . . . . . . . . . . . 0.00 Sulfur as HtS. . . . . . . . . . . . . . . . . . . 0.00 Tar Sulfur.. .................... 0.00 TOTAL ....................... 26.43
-
AT
~ o o o oc . 1000~
0.00 0.00 1.93 .5.22 13.12 6.08 0.08 26.43
I t may safely be assumed t h a t the sulfate form will be reduced t o t h e sulfide. Subtracting t h e sulfide sulfur from this source from t h e total sulfide sulfur leaves 12.73 per cent as t h e sulfide sulfur coming from t h e pyrite. This figure is very close t o 12.57 per cent or one-half t h e pyrite sulfur. The complete decomposition of pyrite in coal must therefore yield ferrous sulfide and free sulfur, which later changes t o hydrogen sulfide if an excess of hydrogen yielding matter is present. T E N K E S S E E COAL-The results of a complete study made upon t h e Tennessee coal of sulfur distribution a t different temperatures are given in Table V. I t will be noted t h a t t h e sulfur not found in any other type of compound is placed under t h e head of "organic sulfur." T h a t t h e coal sulfur other' t h a n t h a t 'of t h e pyrite a n d sulfates is organic in nature has been proved.' This organic sulfur persists almost unchanged in t y p e up t o 400'. Between 400' and 500' a decided change i n its characteristics takes place. This is shown b y t h e fact t h a t treatment with nitric acid and subsequent treatment with ammonia does not t a k e into solution t h e cornpounds containing t h e organic sulfur. T h a t organic sulfur is present, even in t h e finished coke, and t h a t its percentage is higher in t h e coke t h a n i t was in t h e original coal has been well proved by Wibaut and Stoffel,2 who have produced a form 1 2
Powell and Parr, Lac. cit. Rec. fi'aa. cltim., 38 (1919), 132.
12.
N,.
II
of organic sulfur much resembling t h a t of t h e coke b y heating together sulfur and sugar. From t h e results given in Table V, i t is possible t o calculate t h e nature a n d magnitude of t h e probable reactions undergone b y t h e various forms of coal sulfur during t h e coking process. This is simplified b y t h e preliminary study of pyrite decomposition. I n t h e presence of a large excess of coal substance, pyrite decomposes quantitatively into ferrous sulfide and hydrogen sulfide. Also i t may safely be assumed t h a t any sulfate present will be reduced t o sulfide. The pyrite decomposed will give a measure of t h e hydrogen sulfide from this source, b u t in every case t h e actual hydrogen sulfide produced is greater t h a n this. A probable source of this excess hydrogen sulfide is t h e organic sulfur, which makes a third probable reaction. T h e sulfur of t h e t a r , being entirely of an organic nature, must find its origin in t h e organic sulfur of t h e coal. This makes a fourth measurable reaction. It may be noticed t h a t t h e , sulfide sulfur of t h e coke is generally present in smaller amount t h a n would be formed from t h e decomposition of t h e pyrite and t h e reduction of t h e sulfate form. This seems t o be accounted for b y a rather peculiar reaction, namely, t h e transference of a portion of t h e sulfur combined as sulfide t o a sulfur-carbon combination. This fifth reaction has been noted b y other investigators.' T h e evidence for this last reaction is simply t h e fact t h a t during t h e latter stages of t h e coking process there is quite a decided increase in t h e sulfur held in t h e carbon-sulfur combination, with a decrease in t h e amount of sulfides present. The carbon-sulfur combination formed b y this reaction has all t h e properties of t h e remainder of t h e sulfur existing in this form in t h e coke. TABLB V-DISTRIBUTION OF SULFURIN TENNESSEECOAL No. 21100 (Values given in percentage by weight of original air-dried coal) 0" 300' 400" 500' 600' 1000' Pyritic Sulfur. .............. 1 . 7 5 1.75 1.42 0.31 0 . 0 0 0.00 Sulfate Sulfur. .............. 0.71 0.55 0.44 0.01 0.01 0 . 0 0 Organic Sulfur. . . . . . . . . . . . . . 1.79 1.63 1.51 1.70 1.87 1.81 Sulfide Sulfur.. 0.00 0.13 0.44 0.93 0.82 0.84 Sulfur as HzS.. . . . . . . . . . . . . . 0.00 0.19 0.39 1 . 2 0 1.39 1.44 Tar Sulfur.. . . . . . . . . . . 0.00 0.00 0 . 0 5 0.10 0.16 0.I6 0 . 0 0 0.00 0.00 0.00 0 Sulfur as CSz.. . . . . . . . 2 0.00 TOTAL.. ........... 4.25 4.25 4.25 4.25 4 . 2 5 4 . 2 5
.............
- - - ._
TABLEVI-REACTIONS OCCURRING I N CARBONIZATION O F TENNESSEG COAL(Results expressed in per cent sulfur on basis of air-dried coal) Organic Organics MS = Temperature FeSz = FeS MSOq Range O C. -I- HzS = MS S = HoS = Tar S OmanIcS 0-300 0.00 0.16 0.19 0.00 0.00 300-400 0.33 0.11 0.05 0.00 0.05 400-500 1.11 0.43 0.25 0.05 0.49 500-600 0.31 0.00 0.03 0 06 0.26 0 01 0.05. 0 2 -0,02 600-1000 __ TOTAL 1.75 0.71 0.57 0.16 0.73
000
-
I n addition t o t h e five reactions or classes of reactions enumerated above, there are several others which have been noted b u t have not been measured, owing- either t o their complexity or t o difficulties of analysis. Prominent among these is t h e formation of pyrrhotite, already noted, and t h e decided transformation in t h e characteristics of t h e organic sulfur compounds between 400° and j o o o . Other reactions, of a secondary nature, also occur in retorts. Reactions between t h e organic sulfur compounds a n d hydrogen of t h e gas t o form hydrogen sulfide, and between 1 S. W. Parr, A m . Inst. Mi?zing Eng., 1919, 1807, J. P. Wibaut and A . Stoffel,Lac. cit.
NOT.,
$15
T H E J O U R N A L O F I N D U S T R I A L AiVD E N G I N E E R I N G C H E M I S T R Y
1920
I073
-
2 :
G
I
-
3
400
FIG 2-sERICS
800 0
400
800 0
400 800 0 400 800 0 TEMPERATURE OF R E T O R T
OF C U R V C S S H O W I N G V A R I A T I O N S O F S U L F U K C O N S T I T U E N T S A T S U C C E S S I V E NO.
21 100
REACTIONS CAUSING
hydrogen sulfide and red-hot coke t o form carbon bisulfide may be mentioned as typical. T h a t carbon bisulfide is not a primary product of coal distillation is shown b y t h e results obtained above. This fact has been known for some time.' Following t h e facts given above, t h e magnitude of each of t h e five measurable reactions has been calculated for each temperature range. A summary of these calculations is given in Table VI, t h e results being expressed in t h e per cent of sulfur undergoing t h e reaction on t h e basis of t h e original air-dried coal. It was necessary t o express results in this manner i n order co get comparable figures. The courses of these reactions are represented graphically in Fig. 2 . The decomposition of t h e pyrite begins a t 300') is complete a t 600°, and reaches its maximum between 400' and 500'. T h e reduction From of t h e sulfates is practically complete a t 500'. one-quarter t o one-third of t h e organic sulfur is decomposed with t h e production of H2S, this decomposition occurring for t h e most p a r t below joo'. At t h e lower temperatures of coking, a small p a r t of t h e organic sulfur finds its way into t h e tar. Not more t h a n one-tenth of t h e coal organic sulfur is so distributed. From 400' t o 500' a large p a r t of t h e 1
Lewes, "The Carbonization of C o a l , " John Allan & Co., 1912, p. 274.
400
400
800 0
TEMPERACURES O F DISTILLATION OF TGN NESSEE:
800 COAL,
T H E V A R I A T I O N S .&RE h'0TED
ferrous sulfide is decomposed, t h e sulfur apparently entering into combination with t h e carbon. P O C A H O N T A S COAL-The values for sulfur distribution in t h e Pocahontas coal are given in Table VII, a n d a summary of its reactions in Table VIII. TABLEVII-DISTRIBUTIONOF SULFUR IN POCAHONTAS C O A L , No. 20507 (Values given in per cent of original air-dried coal) 0' 300° 400° 500' 600' 1000' Pyritic Sulfur.. . . . . . . .. 0.08 0.07 0.09 0.01 0.00 0.00 Sulfate Sulfur . . . _ . . . , 0 . 0 1 0.02 0.02 0.00 0.00 0.00 Organic Sulfur.. . . . .,. . 0 . 4 7 0.47 0.44 0.43 0.35 0.27 Sulfide Sulfur ... . . . . . , . 0.00 0.00 0.00 0.04 0.05 0.09 Sulfur as HzS.. . . , . , 0.00 0.00 0.01 0.06 0 . 1 3 0.17 Tar Sulfur , . . . . . , , . . , .. 0 . 0 0 0.00 0.00 0.02 0.03 0.03 Sulfur as C S ? ... , , , , ... . D.00 TOTAL . . . . . _ . . . . . . . . 0.56 0.56 0.56 0.56 0.56 0.56
.
. .
o.00 o.00 o.00 o.00 o.00
TABLEVIII-REACTIONSOCCURRINGIN CARBONIZATION OF POCAHONTAS COAL(Results expressed in per cent sulfur on basis of air-dry coal) Organic S Organic S MS = Temperature FeSz = PeS MSOd Range, O C . 0-300 300-400 400400 500-600 600-1000
TOTAI,
-I-H2S
= MS
0.00 0.00
0.00
0.07 0.01
0.00 0.08
0.00 0.01 0.00
0.00
0.01
= HB
0.00 0.01 0.01 0.07 0.04 0.13
-
=
Tar S
0.00
0.00 0.02 0.01 0.00
0.03
Organic S 0.00 0.00 0.00 4 . 0 1
-0.04 -0.05
This coal differs from t h e Tennessee coal in t h a t it is of a semi-bituminous t y p e and also in t h e fact t h a t its sulfur content is considerably lower. Furthermore, a very large proportion of t h e total sulfur is organic, with very little pyrite a n d practically no sulfate. This may account for some of t h e differences in t h e reactions between t h e two coals.
T H E J O U R N A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
1074
12,
NO.I I
a8
0.7
06
0.5
0.4
03
0.2
0. I
0
400
8OU 0
400
800 0
400 800 0 400 800 0 TEMPERATURE OF RETORT
400
BOO
0
400
800
FIG.3---SERI@SOF CURVES S H O W I N G VARIATIONS O F SULFUR CONSTITUENTS A T SUCCRSSIVE~TEMPERATURES OF DISTILLATION O F POCAHONTAS COAL, NO. 20507.
REACTIONS CAUSING THE VARIATIONS ARE NOTED
The reactions of t h e sulfur in t h e Pocahontas coal are shown graphically in Fig. 3. I n two particulars, these reactions differ materially from those of the Tennessee coal. T h e decomposition of t h e organic sulfur t o form hydrogen sulfide is more pronounced at t h e higher temperatures t h a n a t t h e lower degrees of heat, Instead of a tendency of t h e sulfide sulfur t o be converted into organic sulfur, t h e reverse of this seems t o be true above j o o o . This can be explained by t h e fact t h a t very little sulfide sulfur can be formed from t h e small percentage of pyrite and sulfate, whereas organic sulfur is present in fairly large quantities. The reaction would seem t o be reversible and an excess of one form of sulfur causes t h e equilibrium t o shift toward t h e other form. PITTSBURGH coAL-In Tables I X and X are given t h e results obtained for t h e Pittsburgh coal. This was carbonized a t only two temperatures, 500' and I0OO0. IN PITTSBURGH COAL, No. 18847 TABLE IX-DISTRIBUTION OF SULFUR (Value given in per cent of original air-dried coal) 00 500' 1000~ 0.00 0.32 Pyritic Sulfur.. . . . . . . . . . . . . . . . 0.79 0.00 0.00 Sulfate Sulfur.. . . . . . . . . . . . . . . . 0 . 2 3 0.98 0.74 Organic Sulfur.. . . . . . . . . . . . . . . 0.70 0.16 0.23 Sulfide Sulfur. . . . . . . . . . . . . . . . . 0.00 0.53 0.38 Sulfur as HzS, 0.00 0.05 0.05 Tar Sulfur. . . . . . . . . . . . . . . . . . . . 0.00 0.00 0.00 Sulfur as CSz.. . . . . . . . . . . . . . . . 0.00 1.72 1.72 TOTAL..................... 1.72
................
-
-
-
TABLBX-REACTIONS OCCURRING
IN
CARBONIZATION OF PITTSBURGH
!!i COAL(Results: expressed in:ber cent sulfur on basis of air-dried coal) Temperature FeSz = FeS Range O C. HIS, 0-500 0.47 0.32 500-1000 TOTAL 0.79
+ -
MSOa = MS 0.23
Organic S Organic S MS = = HzS = Tar Organic S 0.15 0.05 0.24 0.00 0.00 0.23 ___ 0.15 0.05 0.47
-
__ 0.00 0.23
A study of these reactions reveals t h e fact t h a t they do not differ essentially from those undergone by t h e Tennessee coal. UPPER FREEPORT coAL-The Upper Freeport coal gave t h e results and reactions shown in Tables X I and X I I . TABLEXI-DISTRIBUTION OF SULFUR IN UPPERFREEPORT COAL,No. 23066 (Values given in per cent of original air-dried coal) 00 500' 1000~ 0.00 0.33 Pyritic Sulfur., 0.47 0.00 0.01 0.07 Sulfate Sulfur.. 0.66 0.67 0.58 Organic Sulfur.. 0.12 0.09 0.00 SulfideLSulfur. 0.40 0.17 0.00 Sulfur as HzS.. 0.03 0.03 Tar Sulfur.. . . . . . . . . . . . . . . . . . . 0.00 0.00 0.00 Sulfur as CSz.. . . . . . . . . . . . . . . . 0.00 1.21 1.21 1.21 TOTAL
............... ............... .............. ................ ...............
.....................
-
-
-
TABLEXII-RBACTIONS OCCURRING I N CARBONIZATION O F UPPER FREEPORT-COAL(Results expressed in per cent sulfur on basis of airdried coal) Temperature FeSz = FeS MSOa Organic S Organic S MS = Ranee C. H2S = MS HIS = T a r S Organic ----0-50' 0.14 0.06 0.10 0.03 0.04 0 14 0.33 __ 0.01 0.06 0 2 500-1000 TOTAL 0.47 0.07 0.16 0.03 0.18
+
-
-
-
s
-
I n general, these reactions resemble those of t h e coals already described.
Nov., 1920
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
JOLIET COAL-Through t h e courtesy of Mr. J. V. Freeman, chief chemist of t h e central laboratory of t h e Illinois Steel Co., a sample of t h e coal as it is charged into the by-product ovens a t t h a t plant was obtained. 4 sample of t h e coke from t h e ovens was also obtained for comparison. This comparison is given later in t h e paper. The results obtained on this by-product coking coal are given in Tables XI11 a n d XIV.
TABLEXIII-DISTRIBUTION
OF SULFUR IN JOLIET COAL N O . (Values given in per cent of original air-dried cdal) 00 500' Pyritic Sulfur.. . . . . . . . . . . . . . . . 0.26 0.12 0.00 Sulfate Sulfur.. ............... 0.00 0 . 5 6 0.44 Organic Sulfur.. . . . . . . . . . . . . . . 0.08 Sulfide Sulfur.. . . . . . . . . . . . . . . . 0.00 0.16 Sulfur as HrS.. 0.00 0.02 T a r Sulfur.. . . . . . . . . . . . . . . . . . . 0 . 0 0 Sulfur as CSz.. 0 2 0.00 TOTAL. .................... 0.82 0.82
............... ...............
33945 1000~ 0.00 0.00 ....
0.49 0.06 0.25 0.02
0.00
0.82
XIV-REACTIONS OCCURRING I N CARBONIZATION O F JOLIET COAL (Results expressed in per cent sulfur on basis of air-dried coal) MS. = Temper$ure FeSz = FeS MSO4 Organic S Organic S Range C. HrS = MS = HrS = T a r s Organic S 0.00 0.09 0.02 0.00 0-500 0.14 500-1000 __ 0.12 0.00 0.03 0.00 0 2 TOTAL 0.26 0.00 0.12 0.02 0.07
TABLE
+
-
-
The general course of reactions in this mixed coking coal resembles t h a t of t h e majority of coals studied. R A W A N D W A S H E D V A N D A L I A COAL-Comparative tests were made between a coal in the raw state and t h e same coal after passing through a washery. The results of these tests are given in Tables XV and XVI. TABLEXV-DISTRIBUTION OF SULFUR I N R A W VANDALIA COAL, NO. 33819, AND WASHEDVANDALIACOAL, No. 33821 (Values given in per cent of original air-dried coal) R -AW-WASHED? 00 5000 1 0 0 0 ~ 0' 500' 1000° 0.25 0.12 0.00 Pyritic Sulfur.. . . . . . . . . . 0 . 7 0 0.32 0.00 Sulfate Sulfur.. . . . . . . . . . 0.03 0 . 0 0 0.00 0.03 0.00 0 . 0 0 Organic Sulfur. . . . . . . . . . 0 . 6 5 0.50 0.70 0 . 9 0 0 . 6 3 0.66 0.00 0 . 0 4 0 . 0 3 Sulfide Sulfur.. . . . . . . . . . 0.00 0.12 0.11 Sulfuraq H B . . . . . . . . . . . 0 . 0 0 0 . 4 1 0 . 5 4 0.00 0.36 0 . 4 6 T a r S.. . . . . . . . . . . . . . . . 0.00 0.03 0.03 0.00 0.03 0 . 0 3 Sulfur acl CSz ........... 0 . 0 0 0.00 0 2 0.00 0.00 0.00 TOTAL . . . . . . . . . . . . . . . 1.38 1.38 1.38 1.18 1.18 1 . 1 8 TABLEXVI-REACTIONS OCCURRINGIN CARBONIZATION OF VANDALIA COALS(Results expressed in per cent sulfur on basis of air-dried coal) TemperaFeSr = ture Range FeS MSOa Organic S Organic S MS = COAL C., HzS = MS = HeS = T a r S Organic Raw . . . . . . . . 0-500 0.38 0.03 0.22 0.03 0.10 Raw ......... 500-1000 0.32 0.00 - 0 2 0.00 0 2 TOTAL 0.70 0.03 0.19 0.03 0.27 Washed . . . . . . 0-500 0.13 0.03 0.29 0.03 0.05 Washed . . . . . . 500-1000 022 0.00 024 0.00 0.07 TOTAL 0.25 0.03 0.33 0.03 0.12
--
-
+
.
Some rather interesting comparisons can be made from these coals. I n t h e raw coal, where t h e inorganic sulfur predominates, a larger quantity of metallic sulfides are converted into t h e organic form t h a n in t h e washed coal. I n t h e washed coal, where t h e organic sulfur predominates, more of t h e organic sulfur is decomposed into hydrogen sulfide t h a n in t h e raw coal. Therefore, as might be expected, t h e extent t o which a sulfur reaction will proceed depends very largely on t h e amount of t h e reacting constituent present. The results obtained from the tests on these various coals check up very closely t h e conclusions arrived a t from t h e series of tests on t h e Tennessee coal. One very noticeable exception was t h e Pocahontas coal in which organic sulfur was present in relatively very small quantity. Because of this there was no conversion of metallic sulfides into organic sulfur, but t h e
I075
reaction seemed t o be t h e reverse of this. Exceptions of this kind may be expected where t h e sulfur constituents are in abnormal relation t o each other. DISTRIBUTION
OF C O A L S U L F U R I N T H E C O K I N G PROCESS
A question which has long been of great interest t o t h e coal and coke chemists of t h e country is t h a t of t h e behavior of t h e different forms of coal sulfur during the coking process. Which is eliminated i n t h e volatile matter in the greatest proportion-the organic sulfur or t h e pyritic sulfur? What conditions, if any, affect the distribution ratio of t h e two forms between t h e coke and t h e volatile products? The conclusions of previous investigators have varied widely on this point. Campbell believes t h a t most of t h e organic sulfur remains in t h e coke. Parr, on t h e other hand, states t h a t the organic sulfur is discharged for t h e most part at relatively low temperatures. M'Callum concludes t h a t a somewhat greater percentage of the organic sulfur t h a n of t h e inorganic was volatilized. From t h e standpoint of coal washing, t h e question is an important one, since this process selectively removes t h e pyrite without touching the organic sulfur. If organic sulfur is volatilized in greater proportion t h a n t h e pyritic sulfur, coal washing for coking coals is more efficient t h a n the sulfur reduction in the coal would indicate. The facts are t h a t washery men have always evaluated their practice b y t h e sulfur and ash reduction secured, without any attention t o t h e forms of sulfur removed. Table X V I I i s d e s i g n e d t o bring out t h e d a t a concerning t h e relation of t h e sulfur in t h e coal t o t h a t in t h e coke. TABLE XVII-RELATION
OF THE
SULFUR FORMS OF
. . . . .. .. .. ..
COALTO SULFUR containing only
THE
COKE (Data based on sulfate-free coal, organic and pyritic sulfur forms) Organic Inorganic Sulfur Sulfur Sulfur Per cent Per cent of of in Coal Total S Total S Per cent COAL 14.5 0.55 Pocahontas. , . , , , . , , . 8 5 . 5 78.2 21.8 1.15 Washed Vandalia.. 68.3 31.7 0.82 Joliet Coking.. 41.2 1.14 Upper Freeport. . . . . . . 5 8 . 8 49.4 3.54 Tennessee. . . . . . . . . . . . 5 0 . 6 51.9 1.35 Raw Vandalia. . . . . . . . 48.1 53.0 1.49 Pittsburgh. ........... 4 7 . 0 I N THE
Sulfur in Coke Per cent 0.45 1.14 0.75 1.09 3.22 1.30 1.51
Sulfur in By-product Oven Coke Per cent
.... ....
0.64
.. .. .. .. .... ....
It will be noted t h a t all t h e d a t a above have been calculated on t h e basis of sulfate-free coal. This is necessary since it is known t h a t t h e sulfate form is completely retained in t h e coke and its presence in t h e calculations would obscure t h e essential d a t a regarding t h e pyrite and organic sulfur. I n freshly mined coal sulfates are normally absent. The coals have been arranged in t h e table in t h e order of their relative content of organic sulfur, beginning with t h e highest proportion of organic sulfur. A careful scrutiny of t h e percentages of sulfur in t h e coals and t h e percentages of sulfur in t h e corresponding cokes will not reveal any constant difference. Furthermore, t h e differences between t h e two percentages do not vary as t h e proportion of organic sulfur, b u t seem t o be entirely independent of this. These figures would seem t o indicate t h a t t h e total sulfur of t h e coal is t h e most important factor affecting t h e sulfur content of t h e coke, t h a t t h e relative amount of sulfur forms present do not affect i t materially, and t h a t
1076
T H E J O U R A ' A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y Vol.
certain other factors, particularly t h e nature of t h e coal, will vary t h e amount of sulfur in t h e coke t o a limited extent. T h a t another factor, not considered u p t o this time, materially affects t h e sulfur content of coke will be shown later. A comparison between t h e raw and t h e washed Vandalia coals shows t h a t t h e sulfur reduction in t h e coke is not even as good as t h e corresponding reduction of sulfur in t h e coal. For all practical purposes, however, t h e sulfur in t h e two cokes is roughly proportional t o t h e sulfur in t h e two coals, which strongly bears out t h e statement just made, namely, t h a t t h e total sulfur of t h e coal is t h e main determining factor and not t h e relative quantity of organic and inorganic sulfur. At t h e time t h e coking coal was obtained from t h e Joliet by-product plant, a sample of coke, made from this coal in t h e ovens, was also furnished. The sulfur analysis of this coke is given in t h e last column of t h e table. The reason for t h e difference in t h e sulfur content of the coke made in t h e laboratory and t h a t made in t h e by-product ovens is found in a very interesting secondary coking reaction, which has been well proved, and will be described in a later paper. This reaction is one between t h e hydrogen of the by-product gases and t h e red-hot coke, which results in t h e formation of hydrogen sulfide and its ultimate removal from t h e oven. This decrease in t h e coke sulfur due t o this secondary oven reaction would apply t o all t h e cokes in t h e above table, but this increased difference between t h e percentage of sulfur in t h e coal and t h a t of the coke would in no wise affect t h e conclusions reached as t o t h e effect of t h e relative quantities of sulfur forms present. The comparison of these tests with actual operation is interesting, however, as showing the difference between t h e effect of t h e primary sulfur reactions, as worked out in t h e laboratory, and t h e combined effect of primary and secondary reactions as shown in t h e by-product coke oven. The nature of t h e sulfur compounds in coke has always been a problem difficult t o solve. The present investigation has brought out very little additional information on this subject. It still seems probable t h a t t h e majority of t h e coke sulfur is organic, with a smaller quantity of iron sulfide present. The nature of this iron sulfide in coke has already been described. Proof t h a t a magnetic sulfide of iron existed in coke was established b y t h e following experiment: Coke, made in t h e laboratory from t h e Joliet coking coal, was pulverized in a porcelain mortar. It was found t h a t about one per cent of this coke could be separated from t h e remainder by means of a large horseshoe magnet. The magnetic portion contained 2.4 per cent, and t h e nonmagnetic portion 0.12 per cent of sulfide sulfur. I n other words, t h e magnetic portion contained twenty times t h e percentage of sulfide sulfur t h a t was in t h e remainder. This would point t o t h e presence in t h e coke of magnetic sulfide of iron or pyrrhotite. Campbell has previously indicated t h e presence of this substance in coke and this investigation confirms his statements, b u t exception is t a k e n t o his
12,
So.
11
claim t h a t pyrrhotite has t h e formula Fe,Ss, t h e t r u e nature of this substance having been indicated earlier in this paper. Free iron has not been detected i n coke prepared under careful conditions, b u t has been found in large quantities in coke which has been ground in an iron or steel crusher, due solely t o t h e erosion of t h e machinery. CONCLUSIONS
Previous work has indicated t h a t sulfur exists i n coal in three typical forms-pyrite or marcasite, sulfates, and organic sulfur. A study of t h e changes which these forms undergo during coking has been made on a variety of coals, and t h e five following classes of reactions established: I-Complete decomposition of the pyrite and marcasite t o ferrous sulfide, pyrrhotite, and hydrogen sulfide. This reaction begins at 300' C., is complete at 600' C., and generally reaches its maximum between 400' and 500' C. 2-Reduction of sulfates t o sulfides. This reaction is complete a t j o o ' C. ,?-Decomposition of t h e organic sulfur t o form hydrogen sulfide. From one-quarter t o one-third of t h e organic sulfur is so affected in t h e primary decomposition, but t h e by-product gases traveling through t h e coking mass increase this reaction t o as much a s one-half of t h e organic sulfur present. Primary decomposition is most active below joo' C. 4-Decomposition of a small part of t h e organic sulfur t o form volatile organic sulfur compounds. The greater portion of these find their way into t h e tar. This decomposition occurs a t t h e lower t e m peratures of t h e coking process. ;-Disappearance of a portion of t h e ferrous sulfide and pyrrhotite, t h e sulfur apparently entering i n t o combination with t h e carbon. This reaction seems t o be most active in t h e neighborhood of j o o o and higher. The organic sulfur not accounted for by t h e above reactions undergoes a decided change in character between 400' and j o o o and shows none of t h e properties of t h e original coal sulfur. This investigation indicates t h a t t h e total sulfur of t h e coal is t h e most important factor affecting t h e sulfur content of t h e coke, t h a t t h e relative amounts of sulfur forms present do not affect i t materially, a n d t h a t certain other factors, particularly t h e nature of t h e coal, will vary t h e amount of sulfur in t h e coke t o a limited extent. T h e secondary reaction between t h e sulfur of t h e red-hot coke and t h e hydrogen of t h e by-product gases traveling through it causes a more marked reduction in t h e amount of sulfur left in t h e coke t h a n t h e above primary reactions would indicate. A magnetic sulfide of iron, probably pyrrhotite, has been proved t o be present in coke. ACKNOWLEDGMENTS
The author wishes t o express his appreciation a n d gratitude for suggestions and other valuable assistance during this investigation to Mr. A. C. Fieldner, supervising chemist of t h e Pittsburgh Station, Bureau of
Nov., 1920
T H E J O U R N A L OF IiVDUSTRIAL A X D ENGINEERING CHEMISTRY
Mines, Mr. J . D. Davis, assistant supervising chemist, Mr. J. R.Campbell, chief chemist, H. C. Frick Coke Co., Mr. J. V. Freeman, chief chemist, Central Laboratory, Illinois Steel Co., and t o others who have been interested in t h e work as it progressed. THE DESULFURIZING ACTION OF HYDROGEN ON COKE122 By Alfred R. Powell FXPERIMENT STATION,
BUREAUOF kIINES, PITTSEL‘XGIX’,P A .
Sulfur has always been a n objectionable constituent of coke, and, as t h e supply of low sulfur coal becomes less, t h e presence of more a n d more sulfur in metallurgical coke is a real problem. Coal washing has been resorted t o in many regions where high sulfur coal is mined, b u t even efficient methods of washing often will not solve t h e problems, since this treatment removes a part of t h e sulfur which is combined as pyrites, b u t does not affect t h e finely disseminated iron pyrite or t h e sulfur in organic combination.3 In most cases, one-fourth t o one-half of t h e sulfur in coal can be removed b y washing, which means a corresponding reduction in t h e coke sulfur, as has been shown previously b y t h e a ~ t h o r . ~ PROCESSES F O R THE
DESULFURIZATIOX
OF
THE
IO77
process is rather expensive, and furthermore tends to destroy the coke and by-products. Fingerland’ passes chlorine through the hot coke after adding certain catalyzers, and claims that the sulfur passes off as sulfur dichloride. Carbon monoxide-Several patents call for the use of carbon monoxide, but no data as to their efficiency are available. PROCESSES INVOLVING THE ADDITION OF COMPOUNDS TO THE
England the Calvert process2 made use of sodium chloride added to the coal before coking. The purpose was to form volatile compounds of sulfur and phosphorus with the chlorine of the salt, but later experiments have demonstrated that sulfur is actually increased in the coke by the addition of salt. The Rowan process, patented in 1868, made use of the addition of salt, with subsequent washing of the coke by immersion in water. It is claimed that this process gave good results, but it has never been applied on a commercial scale. Sodium carbonate-This process, patented by Spurrier,3 is not in reality a desulfurizing process, but is simply the addition of sodium carbonate in excess, to prevent the sulfur of the coke from uniting with the iron in the smelting operation. Manganese dioxide-The addition of manganese dioxide to coal before coking has been patented by Franck.4 The claim is made that the oxygen liberated effects a rapid combustion of the organic sulfur compounds, which are then removed with the gases. No mention is made of the simultaneous oxidation of the coal substance proper, which must be excessive. COAL BEFORE COKING-Salt-In
COKE
I n addition t o methods for removing a portion of t h e sulEur in coal by washing, many schemes have been proposed for t h e removal of sulfur from t h e coke itself. All of these processes involve either t h e elimination of t h e sulfur as volatile compounds or its conversion into compounds which may later be leached o u t with water, and which, b y their nature, are harmless for t h e uses t o which t h e coke is t o be put. A general summary of desulfurizing processes which have been patented or proposed follows. PROCESSES INVOLVING THE PASSAGE OF GASES THROUGH THE
Steam-Scheerer5 in 1854 passed high pressure steam into an oven before drawing the coke, and claimed a resultant loss of 0.4per cent sulfur in the coke. A steam desulfurizing process was patented by Claridge and Roper in 1858. U‘olterecke has a process combining the use of air and steam at not over 400’ C., by which it is claimed the sulfur is driven out as the dioxide. The disadvantage of steaming processes is that an excessive amount of the coke is used up to secure desulfurization. Air-The passage of air through red-hot coke was investigated by Philippart.’ Elimination of a part of the sulfur as sulfur dioxide was secured, but only a t the expense of a prohibitive portion of the coke. The use of air under high pressure with the coke at 300’ C. did not give very efficient desulfurization. Chlorine-Stoners has a process in which coke is treated at the close of the coking operation with chlorine or chlorinated gases, and then washed to remove the soluble salts. This COKING :MASS.
Published by permission of the Director of the U. S. Bureau of Mines. Presented b y title a t the 59th Meeting of the American Chemical Society, St. Louis, Mo , April 12 to 16, 1920. a T. Fraser and H. F. Yancey, BulE. Am. I n s t . M i n i n g Eng., 1919, 1817. 1 Alfred R . Powell, THISJOUR”., 12 (1920), 1069. 6 Groves and Thorp, “Chemical Technology,” 1, p. 123. J. e n d A. Churchill, London, 1889. 6 D. K. P. 261,361, May 2, 1912. 1 Groves and Thorp, LOC. cit. a U.S. Patent 887,145, May 12, 1908.
PROPOSED METHOD
Kone of t h e coke desulfurization processes just described have ever found extensive application, and t h e author knows of no large-scale operation based on any of these processes in this country. Coke desulfurization must of necessity be cheap, must remove a large percentage of t h e sulfur, and must involve little change in existing equipment. In addition, a successful process must not affect t h e quality or quantity of coke produced. I n the preceding paper t h e reactions which coal sulfur undergoes during carbonization have been described. From these experiments it was concluded t h a t pyrite decomposed t o form ferrous sulfide or pyrrhotite and hydrogen sulfide, t h e ratio between t h e residual sulfur and t h e volatile sulfur being about I : I . Secondary reactions a t t h e higher temperatures of t h e coking process cause t h e ferrous sulfide t o change over partly t o what is apparently a “carbon-sulfur” compound.j T h e organic sulfur completely decomposes, more t h a n one-half being retained in t h e coke in a n altered form, while t h e remainder is evolved as hydrogen sulfide, together with small quantities of thiophene or other volatile organic compounds. T h e resulting sulfur of t h e coke will consist, therefore, of iron sulfide, either a s ferrous sulfide or as pyrrhotite, and a larger quantity of a very stable organic sulfur substance. The percentage of organic sulfur in t h e coke is higher t h a n t h a t of t h e
I
2
D . R . P. 270,573, June 7 , 1913. Groves and Thorp, LOG.cit. a U. S. Patent 1,007,153, October 31, 1911. 4 D. R . P. 274,853, April 12, 1912. 6 References in this paper to a “carbon-sulfur” compound and to organic sulfur in coke are simply convenient terms t o designate a combination or existence of sulfur in coke as yet unknown. Indications point t o the association of this sulfur with the carbon of the coke, but whether this is a physical or a chemical association is as yet undetermined. 1
2
