1136 THE COLLIERY GUARDIAN. June 16, 1916. little, it is not of the least use to. reduce the oxygen percentage in the mine air to a point at which lights will still burn, in the hope thereby to reduce in any marked degree the occurrence of spontaneous heatings. Whilst the connection between the oxygen percentage and the rate of absorption is not one of simple proportion, there is a simple relation between the two—namely, the rate of oxidation is proportional to the -square root of the oxygen percentage in the air. If A = the rate of absorp- tion, P = the percentage of oxygen, and K = a constant : then— A = K a/P; or = K. That this relation expresses the results fairly well is shown by the values of K as tabulated below :— Percentage of oxygen. K. 5 54 10 57 15 57 21 56 Percentage of oxygen. 40 . 60 80 100 54 54 56 58 For a. simple chemical reaction the rate of absorption would probably be proportional to the oxygen percentage or to some higher power of it. No definite meaning can be given to the fact that the absorption is proportional to the -square root of the oxygen percentage. Since, how- ever, the absorption is complicated by the number of substances in the coal which possibly take part in the reaction, and also by the physical structure of the coal itself, it is probable that the relation A = K P is only an approximation to a more complex equation. It never- theless expresses the results with some accuracy, and, whatever the actual mechanism of the reaction may be, from a practical point of view the measured rate1 of absorption is the important factor, since this controls the rate of heat production by the coal. The importance of keeping goaves, or other areas liable to. spontaneous heating, airtight becomes very evident. A small leak into such an area which suffices to keep the oxygen percentage at 5 will keep the spontaneous generation of heat at half what it would be in normal air. No place (in a mine liable to gob-fires) in which the oxygen ever rises above 2 per cent, should bo considered properly stowed off and not liable to heating. So long as the ■oxygen is below 2 per cent, it would seem that no coal has a sufficiently high rate of oxidation to- cause any serious rise in temperature. Part IX.—Comparison of Rates of Absorption of Oxygen by Different Varieties of Coal. Introduction.—So far the experiments were concerned only with the atmospheric oxidation of coals from the Barnsley -seam; but since it was important to ascertain how far a similar explanation of spontaneous heating in that seam would hold good, samples were taken from a considerable number of pi'ts in different parts of the country, some liable to- fire, and others not liable. Results of Analyses.— A complete analysis of each sample was made as soon as possible after grinding, usually within 24 hours, and the results were all calcu- lated on the basis of the coal containing the natural percentage of moisture, but free from occluded gases. Where the percentage of sulphur is high, some allowance has to be made for the replacement of the sulphur in the ash by oxygen. If all the sulphur were present in the coal as pyrites, a simple correction would suffice, viz., the percentage of oxygen, as determined by difference, throe-eighths of the percentage, of sulphur found should be added. There is no doubt, however, that all the sulphur in coal is not present as pyrites, though how much is present in other forms it is difficult to decide. In some coals containing 1 per cent, or less of sulphur, only very -small traces of iron have been found. Further, pyrites when burnt leaves a weight of ferric oxide five- fourths as large as the weight of sulphur present. In some cases the total ash of the coal sample is less than five-fourths of the -sulphur content (of. especially Table VI.), which cannot therefore be due to pyrites alone. As an approximation, no addition to the percentage of oxygen by difference has been made when the sulphur content does not exceed .1 per cent.: above this amount three-eighths of the excess has been added to the oxygen percentage in order to correct for the replacement of sulphur by oxygon in the ash. Measurement of Oxygen Absorbed.—The oxygen absorbed was measured, as already ’described. The figures given in the following tables all refer to the rate of absorption measured in cubic centimetres per 100 gr. of coal in one hour. No allowance has been made, in calculating the absorptions, for ash or moisture in the coal, -so that the figures relate directly to the analyses given for the sample. To serve as a basis for comparison, the rates of oxida- tion of the coals from the Barnsley seam in the Doncaster area may be considered -first. In this district there remains little doubt that the atmospheric oxidation of the coal itself is the cause of gob-fires, and that pyrites in the form in which it occurs there has no influence in promoting combustion. Any other coals, therefore, which have an oxidation similar to the Barnsley coals should also be liable to fire from similar causes. As to whether two similarly oxidisable coals from different pits do fire equally readily in the pit depends, of course, on the conditions under which the coal is found and is worked. It has been shown that those factors that particularly favour spontaneous heating are the regula- tion of the air supply and the state of division of the coal. Conditions which cause breaks in the coal, falls at the face, and the leaving of any considerable amount of small coal, or coal which becomes crushed small, in the goaf, will assist the starting of a fire; whilst, if all these conditions are absent, the initiation of a fire will become difficult or impossible. Two coals having equal oxida- tions will be equally liable to fire if the conditions are identical. Where in practice two equally oxidisable coals do not both fire in the pit, it will mean that in the one case working conditions are such as' to prevent or avoid circumstances which favour spontaneous heating. In Table I. are set forth the rates of oxidation of Barnsley Hard coal from five different pits, together with the analyses .of the coals. Table I.—Rate of Oxidation of Barnsley Hard Coal AT 30 DEOS. AND 60 DEGS. CENT. Mine, and sample number. ( A. B. C. D. E. A. 24 25 26 27 28 24 at 60° Rate after hours k— y > Cent. 2 20-0 ... 18’0 ... 15’0 ... 66’0 4 149 ... 11’3 ... 10’1 ... 38’2 6 11-8 ... 7’5 ... 7’2 ... 27’7 8 9-7 ... 6’4 ... 5’7 ... 22’7 12 7-2 ... 5’1 ... 5’3 ... 17’7 18 5’6 ... 3’9 ... 4’2 ... 13’7 24 4’8 ... 3’2 ... 3’6 ... 11’4 30 4T ... 2’9 ... 3’2 ... 9’9 36 3’7 ... 2’7 ... 2’9 ... 8’9 48 3’2 ... 2’3 ... 2’6 ... 7’5 ‘ 60 2’7 ... 2’0 ... 2’25 ... 6’6 72 2’3 . ... 1’8 ... 2’00 ... 6’0 96 1’9 ... 1’5 ... 1’65 ... 5’2 Total absorbed in 96 hours, in c.c. 453 ... 327 ... 340 ... 1,150 No sulphate was produced in any of the above oxidations, hence pyrites played no part in them. Analyses of Coals. Mine, and sample number. A. B. C. D. E. 24. 25. 26. 27. 28. P. cent. P. cent. P.cent. P.cent. P. cent. Moisture 9’55 ... 8'39 ... 8’41 ... 5’06 ... 4'60 Ash 3’40 ... 3’92 ... 2’59 ... 3’21 ... 5’91 Carbon 71'80 ... 72'87 ... 73'69 ... 75'67 ... 74'82 Hydrogen 4’48 ... 4'50 ... 4'52 ... 4'69 ... 4'67 Nitrogen 1'30 ... 1 37 ... 1'29 ... 1'06 ... 0'95 Sulphur 0’83 ... 0’70 ... 0’78 ... 0’84 ... 0'78 Oxygen* 8’64 ... 8’25 ... 8’72 ... 9’47 ... 8’27 * By difference. In addition to the Hard coal, there are also -soft coals, cannels, and jacks, associated with the Barnsley seam. The rates of oxidation of these are very similar to those for the Hard coal.* Samples of the Barnsley Softs have been taken as representing these other parts of the seams, with the results recorded in Table II. Tables I and II. show that all these coals have an oxidation of very much the same type, but that the most oxidisable coals, both Softs and Hards, are from pits A, B, and C, and that the samples from pits D and E are distinctly less oxidisable-. Of these five pits, the one least troubled with fire is D, despite the fact that the oxidation of the coal is as large as that from E. The reason for the freedom from trouble in the former case lies in the different working conditions. Less coal is left behind, and the roof is said to be less liable to fracture, making falls less frequent and the goaf more airtight; in addition, fewer faults are met with. The presence of faults in a seam liable to- fire very readily sets up just those conditions most favourable to heating, owing to the difficulty of getting the goaf airtight on the fault side, both on account of the uneven subsidence and of the occurrence of falls. Of the three pits A, B, and C, A and B have had many more fires than C, probably owing to the fact that in the setting-out of C features which were known to have caused trouble at A and B were largely avoided. In faulty ground heating occurs as .readily in C as in A or B. Table II.—Rate of Oxidation of Barnsley Softs and Cannfl at 30 degs. Cent. Mine, and sample number. A : A. 29. C. 30. D. 31. E. 32. A. 23? Rate after hours 2 21’7 ...19 0 ... 15’0 ... 15’0 ... 17’3 4 *16’8 ... 15’0 ... 9’5 ... 9’5 ... 12T 6 14’0 ... 12’2 ... 7’7 ... 7’7 ... 9’6 8 12’0 ... 10’5 ... 5’0 ... 5’0 ... 8T 12 9’3 ... 8’5 ... 3’5 ... 3’5 ... 6’2 18 7’0 ... 6’5 ... 3’0 ... 3’0 ... 4’8 24 5’5 ... 5’2 ... 2’6 ... 2’6 ... 4’0 30 4’6 ... 4’3 ... 2’4 ... 2’4 ... 3’5 36 4’2 ... 3’8 ... 2’3 ... 2’3 ... 3’1 48 3’6 ... 3’4 ... 2T ... 2T ... 2’6 60 3’2 ... 2’9 ... 2’0 ... 2’0 ... 2’3 72 2’8 ... 2’6 ... 1’9 ... 1’9 ... 2’0 96 2’3 ... 2’1 ... 1’7 ... 1’7 ... 1’7 Total absorbed in 96 hours, in c. c. 516 ... 475 ... * Cannel. 311 ... 311 ... 374 No pyrites was oxidised in any of the above samples. Analyses of Coals. Mine, and sample number. A Moisture ... ( A. 29. P. cent. ... 9’79 .. C. D. 30. 31. P. cent. P. cent. . 10’00 .. 4’34 . E. A. 32. 23.* P. cent. P. cent. 3’30 ... 6’60 Ash 2’07 .. .. 2’15 ... 2’29 . .. 132 ... 4’39 Carbon ... 70’38 . .. 70’72 ... 76’51 . .. 79’67 ... 72’22 Hydrogen... ... 4’81 . .. 4’77 ... 5’01 . .. 5’10 ... 5’46 Nitrogen ... ... 1’33 . .. 1’38 ... 1’32 . .. 1’16 ... 1’24 Sulphur ... 1’01 . .. 108 0’98 . .. 1’03 ... 1’27 Oxygenf ... ... 10’61 . .. 9’80 ... 9’55 . .. 8-42 ... 8’82 * Cannel. f By difference. In these five cases, then, the oxygen absorptions of the coals are in keeping with the liability to fire, as found by practical experience underground; these absorptions may therefore serve as a rough basis of comparison for other seams. As examples of coals known to be free from the liability to spontaneous heating the anthracites and steam coals of South Wales may be taken. The oxida- tions of two samples of anthracite are set forth in Table III. * See Trans. Inst. M.E., 1913, vol. xlvi., p. 563; and 1914, vol. xlviii., p. 521. Table III.—Rates of Oxidation of Anthracite under Varying Conditions. Samples Nos. 14 and 14a from the Big Vein, South Wales; sample No. 13 from the Peacock seam, South Wales. Sample. 13 at 14 at 14a at 30° 14a at 30° C. 30° C. C. passing 60° C. passing passing 2- and over passing 60-mesh 60-mesh 10-mesh 60-mesh Rate after hours sieve. sieve. sieve. sieve. 2 ... 18’1 .. . 23’0 ... 2’07 ... 50’0 4 ... 10’0 .. . 15’2 ... 1’72 ... 18 0 6 ... 7’25 .. . 12’0 ... 1’5 12’2 8 ... 6’0 .. . 9’8 ... 1’3 9’9 12 ... 4’5 .. . 7’0 ... IT 5’7 18 ... 3’2 .. . 4’7 ... 0’9 3’9 24 ... 2’6 .. . 3’4 ... 0’8 3’2 30 ... 2’2 .. . 3 0 ... 0’7 2’8 36 ... 1’9 .. . 2’6 0’65 2’4 48 ... 1’5 .. . 2’2 ... 0’60 ... 1’7 60 ... 1’3 .. 1’8 — 1’4 72 ... 11 .. . 1’55 0’55 IT 96 ... 0’8 .. IT ... 0’50 ... 0’4 Total absorbed in 96 hours, in c. c. 272 .. . 381 75 450 No pyrites was involved in the above oxidations. Analyses of Coals. Sample No. <— A 13. 14. Per cent. Per cent. Moisture 1’58 ... 1’80 Ash 1’96 ... 1’74 Carbon 89’20 ... 88’85 Hydrogen 3’51 ... 3’34 Nitrogen 1’16 ... 1’16 Sulphur 0’75 ... 0’81 Oxygen (by difference) ... 1’81 ... 2 30 These two anthracites have’ an oxygen absorption 96 hours at 30 degs. Cent, of as follows :— C.c. Big Vein 381 Peacock 272 m The first of these absorptions is larger than, those of the Barnsley Hards from pits D and E, which are liable to fire, and yet anthracite has never been known to.fire spontaneously. If, then, the oxygen absorption is a measure of the liability to fire, that of anthracite must be more considerably modified by different conditions than is the case for the Barnsley coals. The conditions which may affect the absorption most are : (1) The fineness of division; and (2) the variation of temperature. If, therefore, the rate of absorption were more reduced for coarse anthracite dust than for coarse dusts of other coals, this might afford some explanation of how it is that a high oxygen absorption does not cause the anthracite to fire. From Table III. it is clear that variation of absorption with fineness is much the same for anthracite as for other coals, and that it does not give any explanation of the anomalous oxygen absorption. Through Through a 2- and a 60-mesh over a sieve. 10-mesh sieve. Oxygen absorbed, in 96 hours by Barnsley Hards ................ 453 ... 126 Oxygen absorbed in 96 hours by anthracite ................... 381 ... 75 The explanation lies with the variation in oxidation with temperature. From Table III. it appears that the quantity of oxygen absorbed by anthracite increases very little with rise of temperature, in comparison with other coals. Barnsley Hards at 30° Cent., 453 c.c. ; at 60° Cent., 1,150 c.c. Anthracite at 30° Cent., 381 c.c.; at 60° Cent., 450 c.c. In 96 hours there is a slight increase in the absorption at 60 degs. Cent., but this increase does not indicate that any new part of the coal substance is undergoing oxidation. After about the twentieth hour, the absorp- tion at 60 degs. Cent, is lower than that at 30 degs. Cent., falling at the 96th hour to one-third the rate at the lower temperature. The meaning of these figures is clear. The anthracite has a definite capacity for oxygen, and -though this capacity is satisfied more quickly as the temperature rises, the total absorption is unaltered. In these circumstances the coal cannot possibly fire spontaneously. The total capacity for oxygen is little more than 450 c.c., and in absorbing this amount the coal would produce sufficient heat to raise its tempera- ture about 30degs. Cent., say, from 30degs. to- 60 dogs. Cent., provided no heat were lost. When the coal had attained this temperature its capacity for oxygen would be completely exhausted; it could produce no more heat, and consequently would cool off. Whilst the anthracites -show at 30 degs. Cent, an oxygen absorption not very different from that of a coal liable to gob-fires, the samples described in Table IV. show a very much smaller absorption—so small, in fact, that the quantity of heat produced by the oxidation would appear to be far too little to give rise to any serious increase in temperature. In addition, they -seem to be subject to a limitation in the quality of oxygen absorbed similar to that already discussed in the case of the anthracites. If sample No. 8 be taken as typical o-f this set of samples, Table IV. shows that the quantity of oxygen absorbed in 96 hours is larger at 60 degs. Cent, than at 30 degs. Cent., but that at the end of this time the rate of absorption -at- the-higher temperature is half that at the lower—indicating that, although oxygen is being absorbed faster at 60 degs. Cent., the total quantity absorbed will be the same as at 30 degs. Cent. Such a coal will be quite free from the liability to ■spontaneous heating. Sample No. 8 would appear to have a capacity for absorbing perhaps 150 c.c. of oxygen, and this would under conditions of perfect heat insula- tion raise the temperature of the coal 10 degs. Cent. The capacity for further heat production would then be lost, and the- coal would cool to the temperature of its surroundings.