704 ________________________________________________________________________________________________________ THE COLLIERY GUARDIAN. April 14, 1916. Influence of Incombustible Substances on Coal Dust Explosions.* By A. S. BLATCH FORD, M,Sc. It has been known for some years past that, by mixing incombustible solid matter with coal dust, the inflammation of the dust in a coal-dust-and-air explosion has been retarded. The present experimental work, which is a development of Dr. Bedson’s previous investigations,! was undertaken to observe the quench- ing effect of different substances, to find the most efficient of these substances, and to arrive at a possible explanation of their preventive action. Apparatus. The explosions were carried out in a form of vessel previously describedwhich is a strong spherical glass vessel of a capacity o>f about 120 cu. in. It is provided with three tubulures; the upper one carries the means of ignition, the second serves for the introduction of the coal dust, whilst to the third can be attached some appliance for ascertaining the impulse produced by the explosion. The means used for this latter purpose was a device described by Teclu, and employed by him in his investigation of the purity of illuminating gas. The essential part of the apparatus consists of a pendulum suspended from a brass frame. To the end of the pendulum is attached a small aluminium basin, which fits lightly but well over a brass tube, 1 in. in diameter, attached by a rubber collar to the end of a glass tube, 15 to 16 in. long and 1| in. in diameter, attached at its other end by means of a rubber collar to the third tubulure of the explosions vessel. The pendulum moves over a graduated arc, and to it is attached a simple arrangement fitted into a ratchet on the upper part of the graduated arc, so that the pendulum is arrested in the position to which it is forced by the impulse acting on the concave side of the basin. A quenched explosion would be indicated by the stationary position of the pendulum. A Nernst filament, giving a temperature of 1,500 degs. to 2,000degs. Cent., was used as a means of ignition. The filament was heated by a Bunsen burner to a temperature sufficiently high for it to conduct the electric current, and then introduced into the explosions vessel through the upper tubulure. Experimental Method. The coal dusts experimented with were such as passed through a 100-mesh sieve (10,000 holes to the sq. in.), and were obtained by grinding the coals. The quench- ing substances were used in as fine a state of division as possible. A weighed quantity (1 gramme) of the mixture of quenching substance and coal dust, in known proportions, was placed in the tubulure of the vessel, which was closed by a tightly fitting rubber stopper carrying a tube connected with a compressed air supply (atmosph. + 12/z — 14/z Hg). The filament was then placed in position, and the coal projected by the blast of air over the glowing filament, the impulse communicated by the explosion being indicated by the deflection of the pendulum. By using the same mixture and igniting it at practically the same temperature, fairly concordant results have been obtained. About four to six separate explosions were necessary in the case of each different mixture. The explosions were carried out in groups of three, the filament being extin- guished, the apparatus allowed to cool, and then cleaned out between each three. The ratio of the quenching substance to coal was varied, and Table I. records the least percentage of quenching substance in the mixture which prevents an explosion. Table II. records the specific heats of the materials used; Table III. gives the thermo-chemical data; whilst Table IV. contains mis- cellaneous information regarding the behaviour of the quenching materials at high temperatures. Table I.—The Least Percentage of Quenching Material in the Mixture which Prevents an Explosion. Coal _______ ............... Boiler-ashes ............... Quicklime ................. Ground shale............... Chance mud ............... Gypsum ................... Magnesia................... Magnesia alba (levis) ....... Anhydrous sodium carbonate Soda-crystals............... Sodium bicarbonate ....... Glauber salts............... a ... b ... c ... D 57 ... 50 ... 47 ...58-60 50 ... 45 ...42-44... 55 43 ... 37 ... 35 ... 46 38-40... 30-33... 29-30... 39-40 33-35...26-28... 29 ... 35 28-30... 28 ...25-26...32-33 22 ...17-19... 15 .. 22-23 12-13... 10+... 12 ... 15 10... 10-... 9+... 11 9-10... 7 ... 7 ... 8 — 8 ... 8 ... 7—... 8- Table II.—Specific Heat of Materials Used. Quicklime........................... 0 19 ... Chance mud ......................... 0'22..... Gypsum ............................. 0’24 Magnesia ........................... 0'23 Magnesia alba (levis) ................. 0'26— Anhydrous sodium carbonate .......... 0’27 Soda-crystals (5H2O)................... 0'35 Sodium bicarbonate................... 0'30 Glauber salts (6H2O) ................. 0’34 Table III.—Thermo-chemical Data. CaC03 _______ Requires 42'52 K for grm.-mol. decom- position. CaSO4j,2H2O ... Requires 4'84 K to drive off 2H2O per grm.-mol. Na2CO3, 5H2O Requires 12'364 K to become anhydrous. Na2SO4, 6H2O Requires 11'52 K to become anhydrous. 2NaHCO3..... Requires 30'78 K to become sodium carbo- nate. * Paper read before the North of England Institute of. Mining and Mechanical Engineers. f “ Experiments Illustrative of the Inflammability of Mix- tures of Coal Dust and Air,” by Prof. P. Phillips Bedson and Mr. Henry Widdas, Trans. Inst. M. E., 1906, vol. xxxii., p. 529; 1907, vol. xxxiv., p. 91; and “ Experiments Illustrative of the Inflammability of Mixtures of Coal Dust and Air,” by Prof. P. Phillips Bedson, ibid., 1910, vol. xxxix., p. 719; and 1911, vol. xli., p. 235. J Tbid., 1906, vol. xxxii., p. 529. Table IV.—Miscellaneous Information Regarding the Behaviour of Quenching Materials at High Temperatures. Gypsum _____ Loses 2H2O at 120 degs.-130 degs. Cent., becoming anhydrous. CaCO3 _______ Commences to decompose at 550 degs. Cent. MgO ......... Melts at 2,250 degs. Cent, (no decom- position). Lime ......... Melts at 1,900 degs. Cent, (no decom- position). Magnesia alba Converted to magnesium carbonate by (levis) 200 degs. Cent., afterwards decomposes, giving off CO2. NaHCO3 ..... At a dull-red heat is converted to Na2CO3. Na2CO3....... Melts at 1,098 degs. Cent, (no decomposi- tion.} Na2CO3, 1* H2O Loses 5 molecules at 12'5 degs. Cent. Loses 9 molecules at J-8 degs. Cent. Becomes anhy< rous at 87 degs. Cent. Na2SO4, xH20 Loses all its water of crystallisation by 100 degs. Cent. Na2SO4 ...... Melts at 863 degs. Cent. The quenching materials used were gypsum, dried Chance mud, quicklime, magnesia, magnesia alba (levis), anhydrous sodium carbonate, sodium bicar- bonate, soda crystals, Glauber salts, ground shale, and boiler ashes. The anhydrous sodium carbonate was obtained by strongly heating the bicarbonate. The boiler ashes were ground from the ashes of a boiler fire burning coke. The soda crystals and Glauber salts were used with as much water of crystallisation as was con- sistent with a fine state of division. They were obtained by making a saturated solution of the ordinary variety in warm water, rapidly cooling the solution from a moderate temperature, drying the crystalline product, and grinding it. The friction produced during grinding tended to give heat, and liberated some of the water of crystallisation. A specimen of the salt as used in the experimental work was afterwards subjected to analysis, in order to determine the proportion of water of crystallisation present in it. Influence of CO2 from Decomposition of a Quench. Both calcium and magnesium carbonates are decom- posed by heat (under the experimental temperature), with liberation of carbon dioxide, yet the critical percen- tages of these carbonates are greater than that of sodium carbonate, which may fuse, but is not decomposed (except to a negligible extent by the carbonaceous matter present). Although the Second Report of the Royal Commission on Explosions in Mines suggests that carbon dioxide may be a determining factor in the quenching of an explosion, it * is suggested from this experimental work that the liberation of carbon dioxide from a quench has but small influence on the explosive character of a mixture. Sodium bicarbonate gives a low critical percentage, and is decomposed by heat, yielding the carbonate (2NaHCO3 = CO2 + Na2CO3 + H20), ’ with liberation of carbon dioxide. The temperature of the decomposition is comparable with that of the decom- position of calcium carbonate; but a possible explana- tion—other than the carbon dioxide hypothesis—of the much greater efficiency of sodium bicarbonate will be given below. A comparison of calcium carbonate and gypsum will show that, so far as decomposition with liberation of an incombustible gas is concerned, water (as steam) is more effective than carbon dioxide (44 per cent. CO2 in calcium carbonate; 20-9 per cent. H2O in CaS04, 2H2O). Influence of Specific Heat. A direct comparison of the action of quicklime and magnesia will show that the difference between the behaviour of these compounds is explained most satis- factorily on the hypothesis that the specific heat of the quenching substance is the important factor in its efficiency. A comparison of the action of sodium bicar- bonate (7 per cent.) and that of anhydrous sodium carbonate (12 per cent.) also supports the specific heat hypothesis—assuming that the carbon dioxide production has little effect. On the basis of efficiency due to carbon dioxide liberation, calcium carbonate (yielding 44 per cent, of its weight as carbon dioxide) would be almost an ideal quenching substance. Influence of Water of Crystallisation. The hydrated sodium carbonate has greater specific heat than the anhydrous variety; but the greater efficiency of the soda crystals may be due to the water liberated by dehydration during the rise of tempera- ture. In experiments with one coal, mixtures of (a) 85 per cent, of poal and 15 per cent, of anhydrous car- bonate, and (5) 89 per cent, of coal and 11 per cent, of soda crystals, proved incombustible. The 11 per cent, soda crystals were resolved into 6 per cent, of sodium carbonate and 5 per cent, of water vapour. These soda crystals contained 46 per cent, of water of crystallisa- tion. It is evident that there may be some preventive action in the liberated water vapour. Increase of specific heat accompanies increase of the number of molecules of water of crystallisation, so that it is necessary to determine which is the essential factor —the specific heat or the water. The experiments with sodium bicarbonate show it to be as efficient as Glauber salts. The bicarbonate remains stable up to about 600 degs. Cent, (specific heat 0-30), whilst the Glauber salt has a specific heat of 0'34, which falls at 100 degs. Cent, to 0-24. From a comparison of the properties of these two compounds, it is suggested that specific heat is the factor which determines efficiency, and that water of crystallisation is to be considered mainly in that it gives an increase in specific heat to hydrated compounds above the anhydrous compounds from which they are derived. The soda crystals contained 46 per cent, of water of crystallisation; the Glauber salts 43 per cent.; yet the latter substance is slightly more efficient than the former (8 per cent, and 10 per cent, respectively). The specific heats of the hydrated varieties are about the same (0-35 and 0-34), that of anhydrous sodium sulphate 0-24, and that of anhydrous sodium carbonate 0-27. The specific heat hypothesis seems, therefore, to break down; but it is suggested that during the rise of temperature to the firing point the sodium sulphate remains hydrated for a longer time than the sodium carbonate, and then the influence of the high specific heat of the hydrated Glauber salts is noticed. Influence of the Heat of Reaction. In reviewing the conclusions, it is of interest to note the range of temperature during which the quenching action seems to take place. In the case of one coal, the effect of 28 per cent, of magnesia is equal to the effect of 8 per cent, of Glauber salts. After losing water of crystallisation by 100 degs. Cent., Glauber salt has a specific heat of 0-24. Neglecting for the moment the heat effect during the rise to dehydration temperature of Glauber salts from 100 to 800 degs. Cent, (about the firing temperature of the coal), we have two mixtures behaving alike, namely :— (a) 28 parts of magnesia of specific heat 0-23, and 72 parts of coal of specific heat 0-24. (5) 4 parts of anhydrous sodium sulphate of specific heat 0-24, and 92 parts of coal of specific heat 0-24. The specific heat of coal may be taken as about 0-24. The suggestion arises that the heat absorption from 0 to 100 degs. Cent, (that is, during the dehydration of the Glauber salts), is a most effective factor in deter- mining the low percentage of this material as a quench- ing substance. X x Na2C Na2SO jO3.' 51 4,'6H2< i2O ) X *JaHCC 3 v 2CO3 X x MgC O3Alba x. x C< \ Mg< tSO4?2 H2O X ^CaCO j X, NCaO —N— . O 10 20 3p 40 50 60 I CRITICAL PERCENTAGES. Fig. 1.—Curve showing the Relation between the Specific Heats and the Critical Per- centages FOR ONE OF THE COALS TESTED. The heat of reaction is of the same order of magnitude as the heat absorbed by specific heat requirements :— 250 grammes Na2S04, 6H2O require 11-52 K for dehydra- tion. 100 grammes Na2S04, 6H2O require 4-61 K for dehydra- tion. 100 grammes Na2SO4,6H2O require 3-5 K for a rise through 100 degs. Cent. 100 grammes Na2S04, anhydrous, require 2-4 K for a rise through 100 degs. Cent. 56 grammes Na2SO4, anhydrous, require 1-4 K for a rise through 100 degs. Cent. The specific, heat hypothesis now modifies itself into one of “ heat absorption,” in which both (1) heat required for decomposition, and (2) heat required for rise in temperature, are to be taken into account. The heat required, thermo-chemically, by the soda crystals is also notable :— 100 grammes Na2CO3, 5H2O require 6-3 K for dehydra- tion. 100 grammes Na2CO3, 5H2O require 3-55 K for a rise through 100 degs. Cent. 53 grammes Na2CO3 require 1-43 K for a rise through 100 degs. Cent. The effect of a quench is probably due to the rate— as implied in the conception of specific heat—of heat absorption over a particular range of temperature, and depends not merely so much upon the total heat effect. The fact of there being an explosion of a dust—as well as, the force produced by an explosion—would depend partly on the velocity of the rise of temperature; a quenching material would be most efficient when it prevents a rapid rise of temperature by large heat absorp- tion comparatively early in the rise of temperature— say, in the rise from 0 to 100 degs. if the coal fired at 800 degs. Cent. Chance mud requires heat for heat of reaction, but not until 550 degs. Cent., which is too near the explo- sion temperature of the coal : for, by the time that this point has been reached, the velocity of the rising temper- ature is too great for the thermo-chemical effect of the decomposing calcium carbonate to be of much value. A reaction involving liberation of carbon dioxide may be of use in quenching the explosive character of a dust if the carbon dioxide is all given off by 200 degs. Cent. From 130 degs. Cent, upwards anhydrous calcium sul- phate (specific heat 0-20) is apparently more efficient than Chance mud (calcium carbonate; specific heat 0'25), thus supporting the suggestion of the action of a quench taking place during the first 200 degs. of rise. It may be suggested that the action of solids which liberate a large percentage of water of crystallisation is due to the mechanical effect of the free gases blowing the dust away from the igniting medium, and keeping the coal out of the sphere of action. Reference to the