89 XIII.-Oi% the Luminosity of Bewzol when twrnt with noidurninow Cornbustib le Gases. By E. FRANKLAND, F.R.S., and L. THORNE, Jodrell Scholar. As early as the year 1852 it was pointed out by one of us that hydro- gen, carbonic oxide, and marsh-gas practically contribute nothing to the light of coal-gas, and that the only constituents of this gas having any substantial value as light-giving agents are those hydrocarbons which combine with fuming sulphuric acid.* The chief of the lumini- ferous hydrocarbons are benzol, ethylene, propylene, butylene, and acetylene ; but although a knowledge of the intrinsic individual luminosity of these and other similar bodies is of very great importance in connection with the subject of artificial light, no successful attempts have yet been made to determine it.We have therefore endeavoured to supply the necessary data, and the following pages contain t,he results of our experiments upon benzol. They will be followed by a similar series upon each of the other chief illuminating constituents of coal- gas. All our attempts to determine the illuminating power of benzol when burnt alone were futile, on account of the extreme difficulty of obtaining a smokeless flame b r the combustion of the pure hydro- carbon. In our efforts to evercome this difficulty we were most kindly assisted by Mr. A l b e r t S i l b e r , who constructed for us several lamps specially designed to consume this liquid ; nevertheless, as the utmost light obtainable without smoke, even with these lamps, did not exceed that of one candle, we were compelled to abandon the attempt as hope- less, and to confine our experiments to the determination of the luminous effect of benzol vapour when diffused in hydrogen, carbonic oxide, and marsh-gas.The illuminating power of benzol, as thus determined, is probably a more trustworthy guide to the value of this compound in coal-gas than would have been any direct determinations with the liquid alone, had such been possible. Moreover, these experiments with benzol vapour have brought to light some points of considerable interest in connection with the relative fitness for their office of the three non-illuminating ga,ses, or diluents, of coal-gas. The determinations were made in the following manner :-Each non-illuminating gas to be charged with the vapour of benzol was collected in sufficient quantity in a holder capable of containing 20 cubic feet, and to which a pressure of 8 feet of water from a ball- * Memoirs of the Manchester Literary and Philosophical Society (2nd Series), vol.x, p. 71. VOL. XXXlII. I90 FRANKLAND AND THORKE ON THE cock cistern could be conveniently applied. From this holder the gas was led through a &-inch tube to a photometric room, where it was received in a floating bell-gasholder of 1 cubic foot capacity, which served to reduce the pressure of the gas delivered from the large holder, so as to bring it within the cont'rol of an experimental gcjvernor, whence it passed to a test-met,er of the usual construction, showing the hourly consumption of gas by observations of one minute's duration. On leaving the meter the gas entered a benzolizer consisting of a brass cylinder 6% inches long, and 3 inches internal diameter, filled with sponge saturated with pure benzol, and so arranged as to compel the whole of the gas to pass through the sponge.The benzolizer was entirely immersed in a large vessel of water, the temperature of which was maintained constant during the con tinuance of the experiments. Each determination of luminosity was the mean of at least ten separate observations. At the close of the photometric readings, and whilst the current of benzolized gas was maintained, a sample was collected for eudiometric determination of the percentage of benzol vapour, which was effected by fuming sulphuric acid in the usual manner .* A.Beizzolized Hydrogem. The hydrogen was prepared from commercial zinc and dilute sul- phuric acid, and contained the usual traces ofimpurities. It is almost needless to say that a flame of it, burning a t the rate of 5 cubic feet per bouy, possessed no measurable amount of illuminating power. After pwsage ihrough the benzolizer, the gas was burnt froin a fish-tail Iiurner, when the following photometric results were obtained :- Bate at which hydrogen passed through meter, 4.95 cubic feet per hour. Temperature of water in which beiizolizer was immersed.. .. Temperature of room. ......... Barometer .................. 29.6 inches. Luminosity corrected to 120 grs. sperm, and 5 cubic feet hydro- gen per hour .............. Ditto, and t o 30 inches barometric 14.8" C .(58-6" F.). 16.0" C. (60.8" F.). 28.13 candles. pressure and 60" F. ........ 28-58 ,, Percentage of benzol vapour in gas 7%; or 100 vols. of hydrogen passing meter, took up 8.22 vols. of benzol vapour. * We find that ordinary sulphuric acid (S02H02), absorbs benzol-vapour rather rapidly, and may be used for its eudiometric determination.LUMINOSITY OF BENZOL, ETC. '31 B. Benzolized Carbonic Ozide. The carbonic oxide was prepared by heating a mixture of potassic ferrocyanide and concentrated sulphuric acid, and was freed from traces of carbonic anhydride by passage through concentrated solution of caustic soda. The gas was then manipulated in the manner just described for hydrogen. It was, however, found impossible t o burn carbonic oxide from a fish-tail burner, under suitable conditions, at R greater rate than 4.22 cubic feet per hour.The following results were obtained :- Rate at which carbonic oxide passed through meter 4.22 cubic feet per hour. Temperature of water in which Temperature of room .......... Barometer .................. 28.88 inches. Luminosity corrected to 120 grs. sperm, and 5 cubic feet of car- bonic oxide per hour ...... 22.66 candles. Ditto, and to 30 inchcs barometric benzolizer was immersed ... 13.0' C. (55.4" F.). 15.0" C. (59.0" F.) . pressure and 60" F. ........ 23.48 ,, Percentage of benzol vapour in gas 6.0, or 100 vols. of carbonic oxide took up 6.38 vols. benzol vapour. C . Bemolixed Marsh-gas. The marsh-gas used in these experiments was made by heating a mixture of sodic-acetate and soda-lime in an iron mercury bottle.So prepared it was far from pure. as the following results of its analysis show ; but we were unable, by this process conducted upon a sufficiently large scale, t o obtain a purer product, and we shrank from the labour necessary t o prepare so large a quantity of pure marsh-gas from zinc-methyl. Indeed had we resorted to this latter method the diffusion of atmospheric air into the pure gas could scarcely have been prevented. Composition of the crude Marsh-gas used in the experimeizts. Marsh-gas .................. 70.08 Hydrogen.. .................. 20.60 C,H,,. ....................... -55 Nitrogen .................... 8.1 7 Oxygen,. .................... -60 100-00 7- 1 292 FRANKLAND AND THORNE ON THE Burnt a t the rate of 5 cubic feet per hour this gas gave the light of 1.4 standard candle.After passing t,hrough the benzolizer it was found impossible to consume the gas from a fish-tail burner, under suitable conditions for the development of light, a t a greater rate than 2.96 cubic feet per hour. The following numbers were recorded in two series of observations :- Hate a t which marsh-gas passed through meter . Temperature of water in which benzolizer was immersed. ........... Barometer ............ Imminosity corrected to 120 grains sperm, and 5 cubic feet marsh-gas Ditto, and to 30 inches Ditto, after deducting 1.4 candle for luminosity of C,H,, in marsh-gas Percentage of benzol- 100 vols. of marsh-gas took up of benzol-va- pour ................ Temperature of room.... per hour ............ bar. press. and 60" F.. . vapour in gas.. ...... I. 11. 9-96 cub. ft. per hour 2.85 cub. f t . per hr. 13.0" C. (55.4" F.) 25.0" C. (77.0" F.) 30.2 inches 29.75 candles 30.56 ,, 29.16 ,, 6.1 6.49 12.0" C. (53%" F.) 22.0" C. (71.6" F.) 30.4 inches 22.92 candles 23.10 ,, 21.70 y , 4-44 4.66 The results of the foregoing experiments may be thus sum- niarised :- 5 cubic feet of hydrogen after benzolization a t 14.8" C., gave for one hour the light of 5 cubic feet of carbonic oxide after benzo- lization, a t 13.0" C., gave for one hour the light of ........................ 23.48 ,, 5 cubic feet of marsh-gas after benzolization at 134", gave for one hour the light of 5 cubic feet of marsh-gas, after benzolization at 12'0" C., gave for one hour the light of ............................21.70 ,, But the volumes of benzol vapour taken up by 100 vols. of the three gases a t the different temperatures jush specified were :- 28.58 candles. 29.16 ,,LUhIINOSITY OF BEKZOL, ETC. 93 For hydrogen . . . . . , . . . . . , . , 8.22 vols. a t 14.8" C. ,, carbonic oxide . . . . . . . , . . 6.38 ,, 13*O0 C. ,, marsh-gas (1st experiment) 6.49 ,, 13.0" C. 5 , ,, (2nd ,, ) 4-66 ,, 12.0" C. Otherwise expressed it may be stated that, measured at 60' F. and 30 inches barometric pressure- ,410 cubic foot of benzol rapour buimt with H gave for one hour the light . . . . . . . . -320 cubic foot of benzol vapour burnt with GO gave for one hour the light of . . . . -3154 cubic foot of benzol vapour burnt with CH4 gave for one hour the light o f .. . . .231 cubic foot of benzol vapour burnt with CH, gave for one hour the light of . . . . 1 cubic foot of benzol vapour burnt with H gave for one hour the light of . , . , . . . . 1 cubic foot of benzol vapour burnt with CO gave for one hour the light of . . . . . . . . 1 cubit foot of benzol rapour burnt with CH, gave €or one hour the light of . . . . . . . 1 cubic foot of benzol vapour burnt with CH4 gave for one hour the light of . . . . . ,. . 28.58 candles. 23.48 29.16 21.70 .. ,, ,, Hence at the standard temperature and pressure- 69-71 candles. 73.38 92.45 93.94 > 7 \, ,, Or if N cubic feet of benzol vapour burnt with H give for one hour the n cubic feet of benzol vapour burnt with CO will give for one hour 1% cubic feet of benzol vapour burnt with CH, will give for one hour TL cubic feet of benzol vapour burnt with CH, will give for one hour light of 1 candle, then the light of 1.053 candle ; the light of 1.326 candle ; the light of 1,347 candle.Now 1 cubic foot of benzol vapour a t 60" F. and 30 in. bar. press. weighs 1,444 grains, and therefore 1 lb. (7,000 grains) of benzol burnt with hydrogen gives a light equal t o that of 337.9 sperm candles for one hour, or 5,793 lbs. of spermaceti. Hence 1 lb. aroirdupois of benzol gives when burnt with- H the light yielded by 5.793 lbs. of spermaceti. CO ,, ,, 6.100 9 , ,? CH4 > 1 9 , 7.682 ,, 7 , CH4 ,> > > 7.803 ,> 7 )94 FRASELAND AND THORNE ON THE LUMINOSITY OF BENZOL. The light evolved by the luminiferous constituents of coal gas is therefore not altogether independent of the proportion of the diluents (hydrogen, carbonic oxide, and marsh-gas) with which they are mixed; for the foregoing experiments show that a given weight of benzol produces 5.3 per cent, more light when it is diluted with car- bonic oxide than when it is diffused in hydrogen, and between 32% and 34.7 per cent.more in marsh-gas than in hydrogen, and tlie last numbers would doubtless have been still higher had the marsh-gas been pure. This difference in the luminosity of benzol when burnt in different media is probably due, in part at least, to tlie different pyrometric t Iiermal effects of the three gaseous media employed. The actual pyrornetric effects of hydrogen, carbonic oxide, and marsh-gas when burnt in atmospheric air have never been determined, but calculated from their absolute thermal effects as measured by F a v r e and Silbermann, the temperatures of their flames would be- Hydrogen.................. 2,080" C. Carbonic oxide ............ 2,828" Marsh-gas ................ 1,935" The actunZ pyrometric effects of hydrogen and marsh-gas are pro- bably nearly equal, whilst that of carbonic oxide is considerably higher. H e r t helot, however, estimates the average pyrometric effect of car- bonic oxide burnt in air (1,975' C:) to be only 75" C. higher than that of hydrogen (1,900" C.). If the actual pyrometric effects of the three gases bear the same relation to each other as the calculated values, the greater amouut of light emitted by benzol when burnt with car- bonic oxide, receives a satisfactory explanation, but the still higher illuminating effect obtained by burning the benzol in marsh-gas is not accounted for. It has been shown by one of us (JozwnuZ of Gus Light- k q , 1867) that marsh-gas burns with a flame of' considerable luminosity when it is heated to about 300' C. along with the air necessarg for its combustion, and it is certain that the pyrometric effect of marsh-gas would be augmented by the presence of 5 cr 6 per cent. of benzol vapour, but whether this is the sole cause of the increased light must be determined by further investigation.