380 THE COLLIERY GUARDIAN. February 22, 1918. EFFECTS OF LOW-TEMPERATURE OXIDATION OF COAL.* By S. H. Katz and H. C. Porter. Introduction. Bituminous coal, when exposed to the air at ordi- nary temperatures, undergoes various changes that reduce its heating value. These changes affect the coal substance both chemically and physically. The molecular changes during the alteration of coal are unknown, but the changes among the different elements of the coal substance have been the subject of considerable experiment. In order to gain infor- mation regarding the possible changes of the hydrogen of the coal substance during the alteration of coal by the air, the work described in this paper was under- taken. At ordinary temperatures oxygen combines with coal by an addition reaction, and for the most part remains chemically combined with the coal substance. As a result, carbon, in the form of carbon dioxide and carbon monoxide, is evolved from the coal. The amount of monoxide is much smaller than that of To trap preventing influx of air PzO* PjO» PtOi CaCh 3 3 I I 1 I 2 1 Fig. 1.—Diagram of Apparatus. dioxide, and together they remove from the coal a small quantity of oxygen as compared with that added during the same time. Hydrogen makes up a portion of the coal substance which has been less investigated as regards changes during low-temperature oxidation of coal. Freshly mined coal is generally preferred for gas making or other industrial purposes. For these pur- poses any loss or changes in the chemical combinations of hydrogen would affect the value of the coal. In the standard method of analysing coal, a sample is air- dried before analysis, so that there is some change through oxidation. Hence there is the question as to the effect of the oxidation on the hydrogen of the coal, and its bearing on the determination of water and hydrogen. Nearly all of the recorded work of previous investi- gators of the oxidation of coal as related to the hydrogen of the coal substance was at the temperature of boiling water and above. Water was invariably produced in readily determined quantities. Three different investigators have studied the subject with regard to changes at ordinary temperatures. The results obtained by these investigators vary widely— from the statement that “ quite a large amount of water is produced in oxidation ” to those of the authors who can find no water produced by their method of experiment. The third investigator found that in some cases very small amounts of water were produced, and in others none whatever. The authors of this paper conclude that at ordinary temperatures coal undergoing oxidation produces no water. The authors have determined, too, that the change in weight of the coal is less than that of the water removed from the coal when drying takes place in an inert atmosphere. Others have noted this discre- pancy. A possible explanation is that the discrepancy results from the absorption of gas bv the coal on drying. Previous Investigations. Porter and Ovitz have shown that hydrogen escapes from coal in the form of methane when the coal is exposed after mining. Later experiments by Katz show that this gas is absorbed by the coal, so that the manner of absorption is physical rather than chemical, and that hydrogen should be differentiated from that present as part of the coal substance. Consequently, this change need not be considered here. Taffanel exposed freshly-made coal dusts to mine air. Analyses showed that the percentages of hydrogen, nitrogen, and volatile matter changed very little when computed on an ash-free basis. The oxygen increased and the carbon decreased. There was no evidence of changes in the absolute amount of hydrogen in the coal. In a further study of changes in natural dusts, as they occur in the mines, Taffanel says: — The carbon and hydrogen contents are diminished by the alteration; one must particularly see in this diminution the percentage compensation of the increases of the oxygen and the nitrogen content. The degree of accuracy of these calculations does not permit appreciating the magnitude of the loss of hydrogen and carbon referred to a determined quantity of fresh coal. Mahler heated 200 grms. of powdered coal in a cur- rent of dry nitrogen to about 100 degs. Cent, till no water was given off, as shown by the weight of a tube of sulphuric acid. The temperature was then lowered to 20 to 25 degs. Cent., and dry air passed through the coal at a rate of a litre an hour for 30 hours. The * United States Bureau of Mines Technical Paper 98. water liberated was collected and weighed, with the following results: — Water Evolved from Coal. H2<> produced, T Per cent. grammes per Goal* volatile. 200 grms. of coal. A ............ 33 0-018 B ............ 27 ..... 0-00 C ............ 25 0-034 D ............. 8 ..... 0-00 E ............. 2 0-183 Here are trustworthy data which show that in low- temperature oxidation hydrogen is evolved from coal in the form of water, but the increase of water pro- duced with the decrease of hydrogen in the coal seems anomalous. Graham passed a stream of dried air through desic- cated coal at 30 degs. Cent., and analysed the coal before and after the experiment. He says : — From the analyses of the moisture content of the coal dust samples before and after “ dry ” oxidation, it is evident that quite a large amount of water is produced during oxidation at low temperatures. ... It is not, however, possible to base any definite conclusion on these results, owing to the fact that anhydrous calcium chloride was placed in the oxidation bottles. . . . Since for the pro- duction of 1 grm. of moisture, 620 c.c. of oxygen are neces- sary, it would seem that at low-temperature oxidation (that is, below 50 degs. Cent.) of dry coal dust, a large amount of water is formed compared with the amount of oxygen absorbed. Graham’s method differed from that of Mahler in involving the heating of the coal after the oxidation, with the possible decomposition of the coal-oxygen complex formed and a consequent formation of water. eight of H2O Formed per Litre of Gas, Milligrms. 12 O Fig. 2.—Relation of Moisture Content of Gas to Volume of Gas Aspirated. 10 8 4 6 2 200 400 600 800 t/3 o Nitrogen aspirated x Air aspirated Volume of Gas, Litres. 1000 1200 1400 1600 1800 2000 Ph O Ft O O However that may be, the presence of calcium chloride in the system next to the coal lessens the reliance to be placed on his work. Method and Apparatus. The apparatus used by the authors was similar to that used by Mahler and by Graham. From a drying train the gases passed through a large quantity of coal at room temperature; thence they passed through another drying train which caught the water given up by the coal. This water was. periodically weighed. The corresponding change of weight of the coal during the same time was also determined. However, a different principle than that used by Mahler or by Graham was adopted by the authors. The Sub-Committee on Standardisation of Methods on Determining Water in Coals have shown that coal dried in a vacuum at room temperature loses water rapidly at first, and then loses very small amounts for an extended period.of time, when the curve for time versus water lost by the coal becomes almost parallel to the time axis. Also, Porter and Ralston have shown that, as coal dries, the relation of vapour pressure to water content is such that plotting the total water content against the total vapour pressure exerted by the coal produces a smooth, even curve, without sharp changes of direction. The authors did not attempt to obtain an absolutely dry coal, which at best is an uncertain quality in view of the report of the committee mentioned, but dried the coal long enough to reach that part of the time- water curve parallel to the time axis. Thereafter the method employed was to observe vapour pressure of the coal through the weight of water removed by a regulated stream of dry gas. Nitrogen and air were alternately employed. A change in the direction of the drying curve would indicate a change in the vapour pressure due to formation of water by the oxidation of the coal. Description of Apparatus. Fig. 1 shows the arrangement of the apparatus. The two large bottles on the left were arranged to form a gas holder with a capacity of 9 litres. One bottle was graduated for measuring the gas. The gas from the holder was dried by passing in succession through a calcium chloride tower a wash bottle containing sulphuric acid and two U-tubes containing phosphorus pentoxide on glass wool. The ground glass stoppers were arranged so that on being turned they closed the tubes. One of these stoppers was set to allow the gas to pass through at an average rate of a litre an hour. This settling was not disturbed during the entire experiment. The coal container was made long and narrow, to ensure the gas passing all parts of the coal. The mer- cury manometer and the small vessel holding phos- phorus pentoxide were used during the preliminary drying of the coal in a vacuum. The two U-tubes, following the coal container, held phosphorus pentoxide on glass wool for collecting the water removed from the coal. These two tubes, though shown as in a straight line, were placed side by side, so as to form a single compact piece for weighing. Beyond the weighing tubes were two guard tubes con- taining phosphorus pentoxide and calcium chloride. Coal Used. The coal was from the Pittsburg seam of the Mans- field mine at Carnegie, Pennsylvania, and had been mined only a few days before the experiment. Selected lumps were crushed to pass a 40-mesh screen; the coal that lodged on a 60-mesh screen was used for the experiment. A sample was taken for analysis. Crush- ing and screening were done as quickly as possible. The holder was filled with coal about 270 grms. The air was immediately pumped out, and the drying in a vacuum began. Gas Used. Nitrogen and air were used. The nitrogen was generated from commercial nitrogen, as sold in cylinders, by passing it from a cylinder through a Van Brunt nitrogen generator. Analyses of 10 different lots gave a maximum oxygen content of 0-25 per cent. Former experiments by the authors had shown that the amount of oxygen absorbed from this atmosphere by the coal would be negligible. Procedure in Experiment. Much of the water in the coal was removed by drying in a* vacuum. Then the coal was dried further by pass- ing a stream of the dried nitrogen through it. When 550 litres of nitrogen had been used, the data showed that each litre of the gas was continuously removing about 1-6 mgrm. of water from the coal. Air was then passed through the coal, and the change in weight of the coal and the water evolved were recorded. The two gases were alternately used in this way throughout the experiment. When the experiment was concluded, the coal was brought to equilibrium with the moisture in the air—a reverse air-drying process—and then analysed. Accuracy of the Gas Volumes. The volumes of gas used were corrected to atmo- spheric pressure and room temperature, any water vapour present being neglected. The small and con- stant error this introduced does not affect the value of the results as a basis for conclusions. Otherwise the error of measurement was less than 2 per cent. Accuracy of Weight of Coal. The large surface of the coal container presented a source of error in weighing. Before each weighing the surface was carefully wiped with a clean cloth. The coal was put into the balance and allowed to stand half an hour before weighing. The total weight of coal and container was 500 grms. The balance had a capacity of Ikilog., was sensitive to a milligramme, and was glass enclosed. An accurate set of weights were used. Errors in weighing are undoubtedly within 10 mgrms. Accuracy of Weight of Water. The U-tubes containing the water were weighed on a sensitive analytical balance. Before each weighing .the glass was carefully wiped with a clean cloth. The