Page images


Yields per hour

Duration of carbonization, hours...
Total cu. ft. of gas from charge.................
Av. cu. ft. gas per hour from charge..........
Av. cu. ft. gas per 1b. coal per hour........

Av. cu. ft. gas per 1b. coal, M & A free, per hr.......
Weight Balance-Products % by weight-

[blocks in formation]

Coke .......niin? Gas .......



66.0 16.3 8.8 5.5 3.4

Tar ......
Ammonia Liquor ...
Unaccounted for ...

65.7 15.9 8.2 5.3 4.8

65.3 16.3 8.2 5.9 4.2





[blocks in formation]

Mixed Monangah & Holden

Average composition

of coal

As Moist're Moisture || As Moist're Moisture | As Moist'rel Moisture charged free & Ash freelscharged free & Ash freelscharged free & Ash free

39.27 60.73


39.24 60.76

39.25 60.75


1.91 Volatile .....

36.40 37.11
Fixed Carbon ...

56.31 57.41

5.381 5.48

.88 .90 Calorific Value, BTU... /14088. J 14340.



[blocks in formation]



.96 15190.

1.25 t

1.18 14413.



Rate of EVOLUTION OF Gas From ONE RETORT. One of the striking facts shown by the curves of the five tests on the Hellier coal is the uniformity of the rate at which the gas is evolved from the retort, irrespective of the size of the charge of coal and the temperature of the retort. When conducting an experiment, meter readings are taken every five minutes, and the test is stopped when the rate of production of gas falls to about one-fourth of the average rate. The rate of evolution of the gas falls off very rapidly at he end of the test, and the time for closing the test is readily told.

With the Hellier coal, although the weight of charge varies from 300 to 500 pounds, and the retort temperature from 1841 to 2036 degrees F. the greatest variation from the mean rate of 447 feet per hour is only 7.0 per cent. The most discrepant figure is the low value of 416 feet on test 21, where the log of the test reports that the retort was badly coated with carbon. This fact seems to have a bearing as will be shown later. Rejecting this figure, and averaging the other four, the greatest deviation from the mean drops to less than five per cent. The seven tests on Pittsburgh coal, made at intervals during eighteen months, under whatever retort conditions happened to be present on the day of the test, show a mean rate of 441 cubic feet of gas per honr, with a maximum variation of 7.5 per cent. The conspicuously low figure of Test 19 may be accounted for by the memorandum in the log that the retort was badly coated with carbon, while the high figure of Test 20 may be explained by the memorandum that the retort had been recently decarbonized and was free from carbon. If these two tests are omitted from the average, the greatest variation from the mean of the five tests, including three 400, one 500 and one 300 pound charge, becomes two per cent. on an average of 443 cubic feet per hour.

In the same way the average rate of yield on the four tests with the LaFollette, Tennessee, coal was 465 cubic feet per hour, with a maximum deviation of five per cent., although the charges varied from 300 to 500 pounds. With the Harrisburg, Illinois, coal the average rate was 411 cubic feet per hour, with a maximum deviation of nine per cent. on four tests of charges varying from 300 to 500 pounds, including one 400 pound charge tested

Range in mean retort temperature

Maximum deviation

from average


in an extremely cold retort, which was badly coated with carbon. If this test were eliminated, the maximum variation from the average would be less than five per cent. The two closely related West Virginia coals, from Monangah and Holden, were tested six times, either alone or mixed with each other, on charges varying froin 300 to 600 pounds. The average rate of gas yield on these six tests is 491 cubic feet of gas per hour, with a maximum variation of eight per cent. The greatest deviation from the average is shown by the 600 pound charge in a rather cold retort. These results are summarized in Table VI.

Table VI.

Number of Range in Average

tests weight of gas yield COAL USED averaged coal charged per hour Hellier, Ky.

300-500 447 1841-2036 Scott Haven, Pa. 7 300-500 441 1844-1994 7.5 LaFollette, Tenn. 4 298-500 465

5. Harrisburg, Ill. 4 300-500 411 1777-1979 9. Holden and Monan

gah, W. Va. 5 300-600 491 1825-2188 8.

The variation in the mean rate of gas evolution as shown by these tests is so disproportionately small, when compared with the difference in the size of charge and the difference in temperature, and there is such a lack of systematic effect of these variables, that it must be concluded that the rate of gas evolution for a given retort is a constant quite independent of the size of charge and to a less degree independent of variations in retort temperature, within working liinits.

Let us consider the phenomena taking place within the retort, and see if there is any theoretical ground for expecting this result. A charge of coal, low in moisture (as all our coals were), is thrown into a retort where destructive distillation at once commences. It is held by some that the destructive distillation of coal is an exothermic process, and that the only function of the hot retort is to start the reaction. If that were the case, the rate of reaction, after it is started, should be entirely independent of the external temperature and practically independent of the size of the charge. There is evidence to support the view that, at certain stages of the distillation, heat is evolved, but it can hardly be said to apply to the process as a whole.

If we holil the more tenable theory that destructive distillation is accompanied by an absorption of heat, then it is evident that the rate of distillation must be regulated by the amount of heat transmitted through the wall of the retort. The transmission of heat should, for any given retort, be a function of the difference between the temperature outside of the retort and that prevailing within the retort at the point where distillation is taking place. It was shown in a paper presented before your Associaticn two years ago* that coal, when gradually heated, lost most of its volatile matter below a red heat. The process taking place in the retort niust be considered as a gradual heating, the external shell of coke transmitting heat slowly to the layer of raw coal next to it, which absorbs the heat and converts it into the potential energy of the gas. The temperature of this zone of coal undergoing distillation must, therefore, be approximately a constant, and its location must be steadily advancing from the neighborhood of the wall towards the center of the retort, so that the heat is transmitted to it through a constantly thickening wall of coke. The rate of progress would not be materially different for a 300 pound charge than for a 500 pound charge, and hence the rate of gas evolution will be independent of the weight of the charge.

The same line of reasoning demands that there should be a more rapid transmission of heat, and hence a more rapid evolution of gas, when the gases surrounding the retort are at a high temperature than at a low one. This only holds, however, for a retort which is always in the same condition. A retort in operation builds within itself an insulating layer of carbon, which is periodically stripped off. We have before alluded to the apparent effect of this insulating layer, and quote specifically the two tests on the Pittsburgh coal, which are the extremes of the series as regards rate of gas production. Test 19, with its retort badly coated with carbon, showed the low figure of 408 cubic feet of gas per hour. Test 20, with a freshly decarbonized retort, gave

* Destructive Distillation of Coal at Low Temperatures, by Alfred H. White, Fred E.

Park, and William A. Dunkley. Proceedings Michigan Gas Association, 1908.

the high value of 464 cubic feet. The temperature in the two tests was practically the same. The effect of temperature change would probably be almost proportional to its extent expressed in per cent. A variation of 200 degrees is a large one, but it is only ten per cent. of the average temperature, and we believe that a change of this magnitude would be readily masked by the effect of condition of the retort and also by the variable which we have not yet mentioned-pressure on the retort.

If it be granted that the rate of gas production is a function primarily of the rate of heat transmission to the charge, it becomes very evident why silica retorts, for example, give results differing from those with ordinary fire brick, and why some kinds of fire ciay may be better than others.

Though the rate of gas production is fairly constant for any one coal, it differs with different coals, the extremes in our tests being the Illinois coal, with its average of 411 cubic feet per hour and the West Virginia coal with 491 feet per hour. There does not seem to be any systematic variation in the chemical analysis of the coals to give the reason for this different rate of gasification. It niay lie in the physical properties of the coal.


It was shown in the preceding section that the rate of gas evolution from a retort was fairly constant for a given coal, independent of the weight of the charge. It follows as a conclusion that the rate of evolution of gas per pound of coal must be a direct function of the weight of the charge, being more rapid for light charges than for heavy. The variation is shown graphically for half-hour periods in Plate I for the Hellier, Kentucky, coal, where there are plotted under each other the yields by halfhour intervals per retort and per ton of coal. The rate per halfhour per ton of coal is lowest with the heavy charge, and highest with the light one. The shape of these curves is characteristic of all the tests.

The ash of the coal is an inert constituent, and the moisture exerts only a slight influence on the amount of gas made, so that an exact comparison is best made on a basis of coal free from moisture and ash. The relation between the rate of gas evolution

« ՆախորդըՇարունակել »