111-D USTRIAL A N D ENGINEERING CHEMISTRY
April, 1924
ABSORPTION SYSTEM The curves of Fig. 1 show that the percentage recovery of sulfuryl chloride from a gas stream issuing from a reaction chamber a t 0" C. decreases much less rapidly when the temperature of the absorbent is 0" C. The divergence of the two curves would be a good deal more pronounced if the reaction chamber were a t 0" C . and the absorbent a t 25" C. Consequently, it is preferable to circulate the chlorosulfonic acid through this tower a t approximately the same temperature as the cold tower. The objects of the absorption tower (indicated in the flow sheet) in which 66" Be. sulfuric acid is circulated arefirst, to absorb the small amounts of chlorosulfonic acid carried over from the first tower; and second, to produce continuously, if desirable, a fresh supply of chlorosulfonic acid. Sulfur trioxide in the form of oleum or as a solution in sulfuryl chloridt: can be run into this tower. The sulfur trioxide will react with the traces of dry hydrogen chloride in the exit gases to produce chlorosulfonic acid. The object of the last tower is to absorb traces of hydrogen chloride, sulfuryl chloride, chlorine, and sulfur dioxide in order that a clean exit gas may be discharged into the atmosphere and any nuisance thereby avoided.
353
MATERIAL OF COJSSTRUCTION While no experience has as yet been had with plant operations, yet from experience with small-scale operations, the writers feel that the following construction materials will serve well: Towers should be of acid-resistant brick set in the customary acid-proof cement. Iron will serve for the coils, filter press, and other apparatus. The still for the distillation of the chlorosulfonic acid-sulfuryl chloride mixture should probably be of the type of the ordinary cast-iron nitric acid still, and the condensers should likewise be of one of the types customarily used for nitric acid. This still will be subject to some corrosion, but the amount should not be excessive. COSTESTIMATE The total cost of operation, which includes carbon a t 7 cents a pound, water, steam, air, repairs, refrigeration, labor a t $5.00 per man per day, overhead, interest on investment a t 6 per cent, and amortization over a period of three years, is estimated a t $12.00 per ton of sulfuryl chloride on a production of 10 tons per day. Taking the cost of sulfur a t $20.00, chlorine cell gas a t $40.00, and a yield of 92 per cent, the cost of sulfuryl chloride by this process should be about $42.00 per ton.
T h e Explosibility of Hydrogen Sulfide in Air' By G . W. Jones, W. P. Yant, and L. B. Berger BUREAUOF MINES,PITTSBURGH,
PA.
P
and Whitakert have shown REVIOUS tests made Apparatus and procedure for determining the limits of complete bY the Bureau of flammability of hydrogen sulfide in air haoe been described. that in tubes less than 5 cm. Tests made in a glass tube h cm. in diameter and 215 to 275 cm. in diameter the cooling Mines on high-sulfur long gave the following oalues for the limits of complete flammability effect of the walls exerts a crude Petroleums, Particufor hydrogen su&de in air (horizontalflame propagation): ( I ) lower marked influence on the re1arlY of ?&?xican origin, sults. The explosion chamshowed that during the dislimit of complete flammability, 5.9 per cent hydrogen sulfide; and ( 2 ) upper limit of complete flammability, 27.2 per cent hydrogen ber (tube) in experiments tillation and refining thereof in this laboratory was made the evolved gases may a t sulfide, times contain in excess of 40 extra long so as to be certain per cent of hydrogen sulfide. that a self-propagated flame This gas, when mixed with the proper proportion of air, is ex- was produced w-hich would not "die out" before reaching the plosive, and may thus create additional hazards in the refining end. A hot, intense spark was used for igniting the gaseous of crudes of this class. Hydrogen sulfide gas is also used in mixtures. This is very important when making experiments many chemical processes as a reducing and precipitating agent, a t the upper limit of complete flammability. All tests were and is liberated as a by-product in other reaction processes. It made by propagating the flame in a horizontal direction. is therefore desirable as a safety measure to know the propor'JhSTING APPARATUS AND PREPARATION OF HYDROGEN tions of hydrogen sulfide in air which are explosive, or, as SULFIDE termed in this report, are within the "limits of flammability"that is, mixturesof gas and air in which, if a flame is initiated, it The testing apparatus (Fig. 1) consisted of a glass explowill propagate itself through the entire space without the necessity of the continued presence of the original source of heat. sion chamber, a, 6 cm. in internal diameter and 275 cm. Exhaustive and careful investigations* have been carried (9 feet) (Position B ) long, joined in sections. For the preout on the flammable and explosive limits of most of the liminary experiments, an explosion chamber only 215 cm. common combustible gases and vapors, but for hydrogen (7 feet) (Position A ) long was used. The mixtures were ignited from the nickel electrodes, b and c. The electrodes sulfide no published values were available. I n general, the values reported by the different investi- were connected to the terminals of the secondary windings of gators for many of the common gases differ widely, depending a transformer, the primary connections in turn connected diupon the method and procedure employed, such as thesize rectly to a 110-volt alternating current lighting circuit. and shape of the explosion chamber, the direction of flame The transformer used gave a secondary voltage a t the elecpropagation, and the kind and intensity of the ignitions. trodes of about 5000 volts and 0.01 ampere of current; Profiting by the experiments of previous investigators, the d is an ordinary Kipp generator for the production of hydrowriters used a glass tube 6 cm. in diameter, because Wheeler gen sulfide, which for the preliminary experiments was made by the action of hydrochloric acid on ferrous sulfide. 1 Re( eived December 8, 1923. Published by permission of the Director, Hydrogen sulfide prepared in this manner was found to Bureau of Mines.
* See bibliography
at end of this report.
?' J . Chem. SOC.(London), ill, 267 (1917).
354
INDUSTRIAL AND ENGINEERING CHEMISTRY
contain 3 to 10 per cent of hydrogen, depending upon the quality of iron sulfide used and length of time it had been in the generator; but since the chemicals were cheap and the hydrogen sulfide was rapidly and easily prepared by this method, it was used for the preliminary experiments to get the approximate flammable limits. For the final results reported,.gas having a high degree of purity was prepared by the action of hydrochloric acid on cadmium sulfide. MANIPULATION OF THE Gas After the generator had been freed of contaminating air, connection was made to wash bottle e, containing water to remove acid vapors, and thence through the constant pressure device, f, the excess hydrogen sulfide bubbling out through the bottom of g and escaping. From the T-piece, 12, the gas passed through a capillary tube, i, which was of such a size and length to give a desired constant flow of gas. From here the gas passed into the mixing chamber, j. Compressed air from the laboratory air line was passed through the wet meter, and thence into the mixing chamber j. The mixed gases were then led into the right end of the explosion chamber, a, and the tube was swept out several minutes to displace all the air with hydrogen sulfide mixture. The left end of chamber a was then closed with a 2-hole, paraffin-coated rubber stopper. Through one hole in the rubber stopper extended a long capillary tube for sampling the mixtures before ignition. Through the other hole extended a long solid glass rod, I , with a coil, m, on the end for mixing the gases. After the mixture had been introduced, rod 1 was pushed back and forth through the tube k a t least fifty times before a sample was withdrawn for analysis. ANALYSIS OF THE GAS Analysis of the mixture was made in a 100-cc. Orsat gas buret, n, and a caustic pipet for absorbing the hydrogen sulfide. Water was used as the confining liquid, and was saturated with the hydrogen sulfide-air mixture used each time to sweep out the explosion chamber preliminary to each test. I n this way the water was approximately saturated with a hydrogen sulfide mixture of about the same concentration as the sample taken for analysis. Hydrogen sulfide mixtures a t or near the limits of complete flammability were further checked by passing a measured volume of the mixture through ammoniacal cadmium chloride solution, thus forming cadmium sulfide. This solution in turn was made acid and titrated with a standard iodine solution, from which the percentage of hydrogen sulfide in the
Vol. 16, No. 4
mixture was calculated. Both values are given in Tables I and 11. As previously stated, the flow of hydrogen sulfide into the mixing chamber j was constant, so that in order to prepare mixtures of different concentrations it was only necessary to vary the air flow through the meter into the mixing chamber, and then record the volume of air passed through the meter in a given interval of time. Just before the gas was ignited the stirring rod I and tube k were drawn out as far as possible through the rubber stopper; both stoppers were then loosened, and removed a t the instant of ignition.
IGKITION OF THE GAS Ignition was brought about by closing a key in the electric circuit. This produced a rapid series of sparks of high intensity across a 3-mm. (l/s-inch) spark gap. When the flammable limits were approached, the flame front traveled a t a slow rate, and could easily be followed with the eye to determine whether the flame died out or continued throughout the tube. The results of the test are given in Tables I and 11. TABLE I-THB LOWERLIMIT
OF
COXPLFTE FLAMMABII.ITY OB MIXTURES
OF HYDROGEN SULFIDE IN AIR
(Explosion chamber 6 cm. in diameter and 215 to 275 cm. long. Temperature approximately 25O C. Horizontal propagation of flame) -Per cent His-By K O H By Io- Ignited HzS Prepared Absorp- dine Ti- from Test bv tion tration Positiona REMARKS 1 FeS $ HCl 3.6 *. A 2 FeS HC1 4.8 A No flammation -~ 3 FeS HC1 A 4 FeS -r HC1 Slight sparkflammation gap at 5 FeS HC1 5.6 6 CdS HC1 5.7 No flammation 7 CdS HC1 6.0 5:$ 8 CdS HC1 6.0 5.8 B Complete flammation, ve;y quiet 9 6.0 CdS 4- HCl 6.3 Complete flammation,b CdS 6.7 HCl 6.5 10 quiet 6.8 6.7 11 CdS 4- HCl FeS 12 HCI 7.1 Complete flammation FeS 7.2 HC1 13 flame traveled s1owl)g FeS 7.8 ,HC1 14 Complete flammation 8.3 FeS HC1 15 flame traveled rapihly with slight noise a See figure. b Complete flammation is regarded as t h a t mixture in which the flame traveled t o within 15.2 cm. ( 6 inches) of the ends of the tube.
+ +
+ + ++ + + + + +
i:;
..
.. ..
1.
..
....
LIMITSOF EXPLOSIBILITY OF HYDROGEN SULFIDE I n a general way the tables give sufficient data so that an explanation of the different tests is not necessary. The limits of complete flammability, when the results of the two methods
I N D CSTRIAL A N D EXGINEERING CHEMISTRY
April, 1924
of analysis are averaged, give 5.9 per cent as the lower and 27.2 per cent as the upper limit. TABLEII.-THE UPPER LIMIT O F COMPLETE FLAMMABILITY O F MIXTURES O F HYDROGEN SULFIDE I N AIR (Conditions as in Table I) -Per cent HzSIgnited By KOH by Iofrom H?S Prepared Absorp- dine Ti- PosiTest bv tion tration tiona REMARKS -. B FeS HCI 91.0 1 B N o flammation FeS 75.5 2 HC1 62.0 FeS HC1 B 3 B 55.0 4 HC1 FeS N o flammation, deposit B 5 FeS HCl 44.0 of sulfur a t electrodes B FeS 32.0 HC1 6 B 29.0 2s: 0 7 HCl CdS N o flammation, slight B 28.2 CdS HC1 28.4 8 burning a t electrodes E 28.0 CdS HC1 9 P a r t flammation did 27.0 27: 8 B CdS 4- HCI 10 not extend throAghout chamber 11 CdS HCI 27.3 27.1 B Just perceptible, slow traveling flame through chamberb 12 CdS HCI 26.9 26.7 B Complete but quiet flammation 26.5 13 CdS f “ 2 1 Complete flammationc 14 CdS HC1 24.3 24:O 15 CdS HC1 21.0 20.6 B Complete flammation, fairly rapid 16 Fe3 HCl 19.8 A Complete flammation 17 FeS f HCI 18.5 B Complete flammntion, flame out both ends IS CdS HCl 18.0 .. B Complete flammation, HCl 17.5 B rapid out both ends 19 CdS a See figure. b A large amount of sulfur was liberated and was carried out of the tube suspensiori and burned a t the outlets of the chamber for a t least 1 minute. c Flame traveled a short distance, then back suction occurred and the flame receded then traveled short distance forward then another back suction. Thrhe back suctions occurred before flame keached the open end. I n this test the stopper was left in the end of chamber nearest the electrodes.
~-+
+ + + + + + + +
+ + + + + + +
.. .. .. .. ..
i i i ]
.. .. ..
1
I n Test 13 (Table 11) the rubber stopper in the end of the explosion chamber nearest the electrodes was purposely left in to observe its effect on the upper limit. As noted in the table, a uniform movement of flame was not obtained, but on the conlrary it surged forward and backward as it progressed through the chamber. I n this movement it would appear almost to die out a t the end of the back suction and then speed up a t an increased speed. This is what SVheelertt terms a “vibratory flame movement,” and is caused by the establishmenc of resonance within the chamber. As an outcome of the e:itablishment of resonance, the flame front acquires a periodic undulatory motion, which sooner or later leads to violent vibrations. These may vary in amplitude, but remain periodic. Wheeler has also shown that these undulations, which mainly occur when flammable mixtures are ignited from the closed end of the explosion chamber, agree closely with the periods calculated for organ pipes of the same dimensions as the explosion chamber used.. This one test clearly iihows that the nature of flame propagation in chambers ignited from the closed end is very different from that in which both ends are open.
BIBLTOGRAPH Y ON FLAMMABLE A N D EXPLOSIVE LIMITS OF COMMON COMBUSTIBLE GASESAND VAPORS 1-IIumbolt and Gay-Lussac, “Explosive Limits of Hydrogen and Oxygen,” Ann. phys., 20, 38 (1805). 2-’.Purner, “Experiments on the Application of Professor Doeberiner’s Recent Discovery t o Eudiometry,” Edinburgh Philos. J . , 11, 311 (1824). 3--13unsen, “Gasometry,” 1857 ed., p. 247. 4--Mallard, “Inflammable Limits of Gases,” Ann. m i n e s , 7, 355 (1875). 5--Wagner, “Hydrogen-Oxygen Mixtures,” Bayerisches Industrie und Gewengeblntt, 8, 186 (1876). 6--Ooquillon, “The Explosive Limits of Methane and Some New Propertier- of Palladium,” C o m p t . rend., 83, 709 (1876). 7-13roockmann. “Uber Erseheinungen beim Brennen von Gasgemischen,” J . Gasbel., 31, 189 (1889). 8-Le Chatelier, “Limits of Inflammability of Gases,” Ann. m i n e s , 19, 388 (1891). 9-Roszkowski, “Influence of Temperature on the Limits of the Explosion of Gaseous Mixtures,” Z. p h y s i k . Chem., 7, 499 (1891). I@-Malard and Le Chatelier, “Momentary Pressures Produced during Gaseous Explosions, Ann. mines., 4, 347 (1883).
--
t t Trans. Inpl. Mining Eng.,
60, 21 (1920).
355
11-Clowes, “Explosive Limits of Gases,” Proc. R o y . Soc. ( L o n d o n ) , 66, 2 (1894); 57, 353 (1S95). 12-Le Chatelier, “Combustion of Acetylene,” C o m p f . Yend., 121, 1144 (1895). l3-Clowes, “The Limiting Explosive Proportions of Acetylene and Air,” J . Soc. Chem. I n d . , 5, 701 (1896). 14-Le Chatelier, “Acetylene and Air Mixtures,” J . pharm. chim., 4 , 313 (1896). Id-Couriot and Meunier, “Explosion of Firedamp Mixtures by Electric Current,” Comfit. rend., 126, 750 (1898). le-Bunte, “Inflammability of Gases,” Bey., 31, 19 (1898). 17-Le Chatelier and Boudouard, “Inflammable Limits of Combustible Vapors,” Compt rend., 126, 1344 (1898). 18-Bunte, “Explosive Gaseous Mixtures,” J . Gas Lighting, 78, 1661 (1901). 19-Kubierschky, “Explosion of Mixtures of Combustible Vapors with Air,” Z . angew. Chem., 14, 129 (1901). aO-Eitner, “Explosive Limits of Combustible Gases and Vapors in Air,” J . Gasbel., 45, 21 (1901). 21-De Hemptinne, “Influence of Pressure on the Propagation of Explosions in Gases,” EtdZ. sci. acad. roy. Belg., 11, 761 (1902). 22-Bairstow and Alexander, “Explosive Mixtures of Coal Gases and Air in a Closed Vessel,” Proc. Roy. SOL.( L o n d o n ) , 76, 340 (1905). 23-Teclu, “Determination of the Explosive Limits OF Gas Mixtures,” J . prakt. Chem., 75,212 (1607). 24-Dixon and Coward, “The Ignition Temperatures of Gases,” J . Chem. Soc. ( L o n d o n ) , 95, 514 (1909). 2S-Burgess and Wheeler, “The Lower Limit of Inflammation of Mixtures of the Paraffin Hydrocarbons with Air,” I b i d . , 99, 2013 (1911). 26-Fischer and Wolf, “Experiments with Nonexplosive Mixtures of Oxygen and Hydrogen,” Ber., 44, 2969 (1911). 27-Taffanel and Le Floch, “Combustion of Gaseous Mixtures,” C o m p t . rend., 157, 593 (1913). 28-Parker, “Influence of Increase of Initial Temperature on the Explosiveness of Gaseous Mixtures,” J . Chem. SOC.( L o n d o n ) , 103,934 (1913). Zg-Thornton, “Electrical Ignition of Gaseous Mixtures,” Proc. R o y . SOC.( L o n d o n ) , 907, 272 (1914). 30-Parker, “The Lower Limits of Inflammability of Methane with iMixtures of Oxygen and Nitrogen,”J. Chem. SOC.( L o n d o n ) , 105,1002 (1914). 31-Terres and Plenz, “Influence of Pressure on the Combustion of Explosive Gas-Air Mixtures,” J . Gasbel., 57, 990 (1914). 32-Leprinz-Ringuet, “Inflammability of Mixtures of Methane,” Compt. rend., 158, 1793 (1914). 33--Coward, Cooper, and Jacobs, “The Ignition of Some Gaseous Mixtures by the Electric Discharge,” J . C h e m . SOC.( L o n d o n ) , 105, 1069 (1914). 34-Coward and Brinsley, “The Dilution Limits of Inflammability of Gaseous Mixtures,” I b i d . , 105, 1859 (1914). 35-Burgess and Wheeler, “The Limits of Inflammability of Mixtures of Methane and Air,” Ibid., 105, 2591 (1914). 3 G B u r r e l l and Oberfell, “The Explosibility of Acetylene,” B u y . M i n e s , Tech. P a p e r 112 (1915). 37-Burrell and Boyd, “Inflammability of Mixtures of Gasoline Vapor and Air,” Ibid., 115 (1915). 38-Burrell and Oberfell, “Explosibility of Gases From LMine Fires,” Ibid., 134 (1915). 39-Burrell and Robertson, “Effect of Temperature and Pressure on the Explosibility of Methane-Air Mixtures,” Ibid., 121 (1916). 40-Burrell and Gauger, “Limits of Complete Inflammability of Mixtures of Mine Gases and Industrial Gases with Air,” Ibid., 160 (1917). 41-Wheeler and Whitaker, “The Propagation of Flame in Mixtures of Acetone and Air,” J . C h e m . SOL.( L o n d o n ) , 111, 267 (1917). 42-Mason and Wheeler, “The Effect of Temperature and Pressure on the Limits of Inflammability of Mixtures of Methane and Air,” Ibid., 113, 45 (1918). 43-Coward, Carpenter, and Payman, “Limits of Inflammability of Gases,” Ibid., 115, 27 (1919). 44-White and Price, “The Ignition of Ether-Air and Acetone-Air Mixtures in Contact with Heated Surfaces,” Ibid., 115, 1462 (1919). 45-White, ‘‘Limits for the Propagation of Flame in Vapor-Air Mixtures,’’ I b i d . , 121, 1244 (1922); 121, 2561 (1922). 46-Payman and Wheeler, “The Effect of Pressure on the Limits of Inflammability of Mixtures of the Paraffin Hydrocarbons with Air,” I b i d . , 123, 426 (1923).
The Canadian Fellowship of Chemical Science, valued at 2600, has been awarded to Edward H. Boomer, a graduate of McGill University, Montreal. Dr. Boomer obtained the degree of Ph.D. in 1923 and was awarded the Ramsay Memorial Scholarship, under the terms of which he has been working at Cambridge University.
. . . .. . . .
The trustees of the Rockefeller Foundation have offered the University of Oxford a gift of $75,000 for the development of the department of biochemistry.
