400 THE COLLIERY GUARDIAN. February 21, 1913. Table 4b.—Showing Percentage ( — ) Lost by Coal and ( +) Water Gained by Drying Tube when Coal is Dried in a Current of “Inert” Gas at a Temperature of 100—110 degs. Coal Number. Worker. Coal weighed in Dried in current of 1. 2. 3. 4. 5. 6. - + - + - + - + - + - + II. Porcelain 2’93 10’12 7’46 5’88 • 7 31 7’95 boat. CO2. £ 2‘84 — 9’78 — 7’30 — 5’89 — 7T0 — 7’67 — 2’96 — 9’84 — 7T0 — 5’58 — 6’65 — 7’50 — IV. H2. £ 2‘88 — 9’46 —- 7’50 — 5’62 — 6’90 7’54 — U tube at Coal gas — 3T5 — 10’46 — — V. 100 degs. (purified). (CaCLJ vCaCl2) Boat at n2. 2’65 — 10’18 — 7’73 — 6’24 — 7’38 — 7 61 — VIII.* llO.degs. — Special Coal gas 2’52 — 7’87 — 6’66 . — 6’26 5’98 — 6’40 — drying (dried). — XII. tube. Nickel co2. 2’87 2’85 10’44 10’20 8’03 7 83 6’51 6’62 7’85 7’87 8T2 7’89 boat. h2so4 h2so4 H2'SO4 H2SO4 h2so4 h2so4 XV. Platinum N2. ( 2’72 2’62 10’61 10’54 7’66 7’91 6’33 6’41 7’59 7’98 7’53 7’94 or porce- I 2’87 3’03 10’54 10’62 7’77 8’18 6’27 6’36 7’54 7’90 7’62 7’72 lain boat. CaCl2 CaCl2 CaCl2 CaCl2 CaCl2 (CaCl2) Maximum 2’96 3T5 10’61 10’62 8’03 8’18 6’51 6’62 7’85 7’98 8’12 7’94 Minimum 2’52 2’62 9’46 10’20 7’10 7’83 5’58 6’36 6’65 7’87 7’50 7’72 ‘Average ... 2’84 2’91 10’12 10’40 7’57 7’97 6’04 6’46 7’29 7’92 7’69 7’85 * Not included in average for reason stated in text. is sufficient) no appreciable difference in results of moisture determinations need be anticipated, provided the coal is distributed evenly over the bottom of the containing vessel, and a concentration of not more than 0 3 gramme per square centimetre of exposed surface is allowed. The worker suggests that 1 gramme would in practice be too small a weight, for the reason that the samples are not generally prepared with the same care as those submitted in the present research. This is undoubtedly the case, and it might be found convenient by those who have two samples of each coal prepared (i.e., a quickly ground sample for moisture and a finely-ground sample for analysis and determination of calorific power) to determine the moisture in the coarse sample on 5 grammes, and in the fine sample on 1 gramme. It is worthy of especial note that, in the above experi- ments, the heating was continued for a quarter of an hour only. 4d.—The same workers as investigated the effect of varying the amount of coal taken, also studied the effect of time of drying on the results. Their figures are given in the table below :— From these results, it would appear that the short period of a quarter of an hour possesses some advan- tages over any longer period, in so far as the loss of water appears to be as great, and the gain by oxidation temperature of 100 degs., viz., 0’21 per cent.) its effect is not great, renders the use of this appliance undesirable for exact moisture determinations, and we cannot recommend it. We feel that the mere fact that the range of temperature between, say, 80 degs. and 98 degs., appears to be a particularly insensitive one, and therefore allows very concordant results to be obtained under varying conditions, induces a false sense of security which a few experiments at 105 degs. will completely shatter. 46.—The method of drying at 104-107 degs. in a One Gramme of Coal Dried at 105 Degs. 1. 2. 3. 4. 5. 6. (a) For 15 min. (mean see Table 2, i 7’51 1 XIV.) 1 305 10’03 7’52 6’19 7’78 1 3 04) 9’92) 7’33) 6’16) i 7’46) 7’63) (6) For another 20 min 3’04 >3’03 9’95 >9’97 7"34 > 7’34 6’09 >6’17 : 7’41 > 7’44 7’65 > 7’64 (c) For 1 hour (mean see Table 1, 3’00) 10’04? 7’34? 6’26? ! 7’44) 7’61) XIV.) ! 2’79 9’77 7’13 6’00 6 99 7’13 (d) For 2 hours (mean see Table 1a, XIV.) ■ 2’82 9’49 6’78 5’69 6’74 6’89 current of gas free from oxygen appears to possess a considerable advantage over any other process of drying at an elevated temperature, in that the necessary compromise between the desire to dry completely on the one hand and the fear of oxida- tion on the other is reduced to a minimum. It may be said, at any rate, when “ fixed ” gases such as hydrogen, nitrogen, or carbon dioxide are used, that the only reasonable possibility of error is the loss of some" thing other than water due to the elevated temperature adopted. That this is small, for at least some coals, has been shown by Mahler (Comptes Rendus, 1910, 1,521, 2). The use of coal gas unless carefully purified would seem to be very risky, as is shown by the results marked VIII. It is, perhaps, not unreasonable to suppose that dry coal may possess a certain power of ad- or ab-sorbing some of the constituents of coal gas. The possibility is suggested by Dr. Lessing, whose figures seem to confirm this view ; Mr. Hills, who uses a direct method, i.e., actually weighs the water in calcium chloride tubes, informs us that, since the amount of CO2 in coal gas has become considerable, owing to the abandonment of purification with lime, he has found it necessary to take special precautions for the removal of this gas (CO)2> since calcium chloride usually contains sufficient free lime for it to absorb CO2 readily. His two results by this positive method and those of Drs. Constam and Pollard are particularly interesting, as they show that at least the amount of water indicated is actually given off from the coal, since calcium chloride, unlike sulphuric acid, is not an absorbent of olefines. It is worthy of note that in, four coals out of the six sent out, the highest result obtained was by this method. Two series in which the coals were dried at 105 degs. in a current of dry air and the loss of weight compared with a gain of a drying tube were intended to show the differences between the direct and indirect methods when the influence of oxygen is not excluded. The reporters feel that these results, especially those in which one of them was concerned, and in fact, all comparisons of the direct and indirect methods, would be more convincing if accompanied by a state- ment that a drying tube, placed between the apparatus used to dry the gas and the coal which is being dried, Coal number. Worker. 1. 2. 3. 4. 5. 6. - + - + ■ - + - + - + - + XI. 3’15 3’25 10’31 10’68 7’78 8’18 6’09 6’34 7’54 8’01 7’84 8’28 XII. 3’19 3’23 10’22 10’38 7’52 8’29 6'42 6’56 7’47 8’37 7’92 8’25 and exactly similar to that in which the evolved water is collected, neither gained or lost in weight during the experiment. Without this information it is futile to theorise as to the difference between the results of the' two methods, as they may be altogether independent of 1 the coal itself. It is, however, only fair to state that! Dr. Pollard found with coal No. 2 that the difference between loss and gain in a current of hydrogen at i 105 degs. was only 0’07 per cent. j Dr. Constam, who ascribes the differences between | his results by this method and by the vacuum method to the fact that three weeks elapsed between the two sets of determinations, says: “ Diese methode ist eine ausserst subtile und schwer zu handhabende.” We endorse this statement in so far as it refers to the direct method, but see little difficulty in an accurate deter- mination of loss in a current of inert gas. 4c.—Profs. De Koninck and Huybrechts have experi- mented with very varying weights of coal dried in weighing bottles of differing areas of cross-sections for a quarter of an hour at 105 degs. Their results are as follow :— 1. 2. 3. 4. 1 5’ 6. 3’04) 9’98) | 10’03 > 10’03 i 7’51) 6’20) 7’56) 7’71) 1 gramme* ) 3T0 >3’05 7’53 > 7’52 | | 6T3>6’19 7’47 >7’51 7’87 >7’78 0’19 gramme per square centimetre ) 3’00 ) 10’091 7’52 ) | 6’24) 7’51) 7’77) 1’5 gramme* ( 0’28 gramme per square centimetre (. B-o?}301 1O‘l6ho-14 1 10’13) , 7;58}7’63 7’53)™ ' 7’60 ) ' 56 SIR83 5 grammesf > 0’16 gramme per square centimetre ) 2’98 9’94 | 7’53 6’08 7’53 7’77 7’5 grammesf ) 0’25 gramme per square centimetre ) 3’00 10’02 ’ 7’76 6’15 7’60 7’85 10 grammesf 1 2’95 9’65 1 7’27 Not determined. 0’33 gramme per square-centimetre ) * Weighing bottle 50 to 70 millimetres high and 27 millimetres diameter = 5’7 square centimetres cross-section, f Weighing bottle 30 millimetres high and 60 to 63 millimetres diameter = 30’4 square centimetres cross-section. This very interesting series points the moral that it is not so much the amount of coal taken for an experiment as its surface of exposure that matters in determining the amount of moisture in a sample. The significance of the results obtained may be summarised in the statement that: At a temperature of 105 degs. (provided the air-current of the coal certainly seems to be much less. The period of two hours certainly seems to be long. 4e.—One group of ’workers (XII.) dried the coals, in watch glasses, in an ordinary desiccator, the results obtained being 1. 2. 3. 4. 5. 6. Percent. Percent. Percent. Percent Percent. Percent. 2 96 .. 9’95* ... 7’18 ... 5’20 ... 7*28 ... 7'53 * A series of tests with this coal was carried out by Dr. Pollard using various desiccants. The results will be given in the section on appliances. The method, which required several days for a relatively constant weight to be reached, is obviously too slow for technical purposes; but, taking the highest results obtained by any worker as representing the correct values, it will be seen that the following supposed percentages of the total water were lost by mere exposure to dry air. 1. 2. 3. 4. 5. 6. Percent. Percent. Percent. Percent. Percent. Percent. 78 ... 95 ... 90 ... 80 ... 93 .. 90 These workers found that the coal experimented with by them, which contained 13 90 per cent, of water (lost at 105 degs), could, when dried, be rehydrated up to within 1’5 per cent, of its original weight by exposure to a moist atmosphere. Dr. Dudley, a member of the American committee, considered that for coals containing about 2 per cent, of moisture satisfactory results could be obtained by drying in a watch glass for 24 hours at atmospheric pressure and temperature over sulphuric acid. The method is not recommended by him for lignites and coals with high moisture. We do not consider that the method possesses any advantages over other methods which excel it in both accuracy and celerity. 4f.—Dr. Constam gives some results obtained by his colleague, Dr. Schlapper, with a modification, not hitherto published, of a method described by Marcusson for the estimation of water in oils, fats, soaps and resins.