432 



NATURE 



[January 25, 19 12 



high makes of flas, but got at the expense of all the other 

 by-products, and bringing in their train all the troubles of 

 •topped ascension pi|K-.<:, carbonised retorts, ruined tar, and 

 poor coke. If instfud of using a 6-inch charge in such 

 a retort it bo filled with coal as nearly as a horizontal 

 retort can be filled, that is, to within 3 or 3 inches of 

 the apex of the crown of the retort, and the same tempera- 

 tures as before are used in the setting, you approach to 

 the same conditions that obtain in vertical retorts, which 

 are entirely filled, and quite different results are obtained. 



The improved results are by most observers ascribed to 

 the doing away with the baking chamber above the charge, 

 radiant heat being credited with the formation of free 

 carbon and naphthalene, which have been the chief curses 

 that have accompanied the raising of temperatures with 

 small charges to increase the gas volume obtained. 



The determination of temperatures as before in the 

 charge shows that filling the retort does a deal more than 

 this, however ; in the first place, in order properly to 

 carbonise the charge, the heating has to continue "from 

 eight to twelve hours instead of six, and the heat is poured 

 into the charge in a much more even manner, the rise in 

 temperature at both top and bottom being equal through- 

 out the distillation, and with the top temperature always 

 al^ut 100° C. more than the bottom, trulv reflecting the 

 difference in temperature between the top and bottom flues 

 of the setting. Owing to the thickness of the charge 

 havmg been doubled, the heat advances more slowly into 

 the mass, rendering the coal semi-fluid and pasty in front 

 of It, and so forming the whole centre of the charge into 

 a tube, through which a large proportion of the gas finds 

 Its way down to the mouth of the retort without having 

 been overheated in the crown. 



That this is the true explanation is shown by the thermo- 

 couple \n the centre of the charge, which during the first 

 SJ hours of the distillation rises slowlv in temperature at 

 about the same rate as the outer portions, but on a plane 

 500 C. (932° F.) below them, whilst after this period the 

 central heat rises rapidly, and the same temperature as 

 exists at the bottom of the charge is attained in the eichth 

 hour. '' 



The gas is improved because onlv a small portion has 

 to run the gauntlet of overheating; the tar is improved 

 because a considerable proportion is made at a lower 

 temperature; the coke is improved because in the earlier 

 half of the distillation a good deal of tar has condensed 

 in the cool zone, and as the temperature rises the more 

 volatile portions redistil, but the least volatile remain and 

 get decomposed only in the last hour of carbonisation, 

 increasing the gas yield, the factor for which the gas 

 manager at the present time seems prepared to make any 

 sacrifice. 



It must be clearly borne in mind that the possibility of 

 filling a horizontal retort to within 3 inches of the crown, 

 although suggested by Kunath in 1885, is a comparatively 

 new condition, which could be achieved only bv the intro- 

 duction of discharging machinery that would push the 

 charge of coke out of the retort, and that, so long as we 

 were dependent upon hand labour to draw the charge, it 

 was impossible ; now, however, the introduction of vertical 

 settings allows a more complete filling of the retort by 

 gravity and easy discharge, so that the question of the 

 route taken by the gas in its escape from the carbonising 

 mass is receiving its full share of attention ; and opinions 

 on the subject are by no means unanimous. 



Dr. Bueb, to whom we owe the inception of the Dessau 

 vertical retorts, is strongly of opinion that the largest bulk 

 of the escaping gas finds its way through a cool core in 

 the same way as it undoubtedly does in a well-filled 

 horizontal retort; others favour the theory that the major 

 portion passes up the sides, whilst Dr. Harold Colman and 

 many others agree that, as the pasty area of the decom- 

 posing coal works its way inwards from the walls of the 

 retort to the centre, gas escapes on both sides of it, so that 

 a portion goes up the centre and the remainder through the 

 hot coke or up the walls of the retort. What proportions 

 exist between the gas that takes the cool route and that 

 which is forced to pass through the hot coke cannot be 

 exactly determined, and would vary with every kind of 

 coal ; but what we do know, beyond' doubt, is that a large 

 volume of the richest gas comes off from all coals when 



NO. 2204, VOL. 88] 



first heated, so that a large evolution of primary ^ 

 be on the inside of the tube of semi-fused coal, 

 another factor which would tend to make the gas take ti 

 inner route is that any passing back into the hot zot 

 would be cxpandfxl by the temperature, whilst the St! 

 pasty coke would offer more resistance to the passa^ . 

 the gas than the coal, so that pressure would be h'r'H- 

 on the hot side and lower on the cool, the prob.-i! 

 therefore being that, so long as the central passage 

 impeded, the major portion of the gas passes up the cenii 

 The analyses of the gases, however, all show that far t< 

 much cracking of the rich hydrocarbons has taken place. 



If w^e distil coal at a fixed temperature, as has be. 

 pointed out, we know that in the distillates we ha. 

 the volatile products yielded at all temperatures below tl 

 one employed, and such secondary products as have be< 

 formed by further decompositions or by internctio- 

 between the primary products at or below that !• 

 so that if we distil coal in bulk at the lowest 

 at which volatile matter can be driven out from mv ro.' 

 the analyses of the products will give us the maximum < 

 primary products with the minimum of secondary, and v 

 shall obtain the best possible data for forming an opinif 

 as to the composition of coal, on the one hand, and tl 

 course taken by the interactions and decompositions in ti: 

 secondary actions. 



In the experiments that led to the introduction of coali' 

 it was found that 420° C. (800° F.) was the lowest temper, 

 ture that could be employed successfully, and mai 

 thousand tons of coal have been distilled at about th: 

 point, say, from 400° to 500" C. 



Under these conditions a fair gas coal containing 30 '■ 

 33 per cent, of volatile matter will yield 68 to 70 per cer 

 of low-temperature coke containing 10 to 14 per cent. > 

 volatile matter, and a volume of gas varying from 30" 

 to 5000 cubic feet, according to the coal ••'"' 1... :..rt , 

 composition approximating to — 



Hydrogen ... ... ... ... ... - -, 



Saturated hydrocarbons — methane ... 48 



higher members lO'i 



— 58 I 



Unsaturated hydrocarlions... .. ... 30 



Carbon monoxide ... ... 7 ; 



Carbon dioxide ... ... ... ... - 5 



Nitrogen ... ... ... ... ... 10 



The gas before iron purification contains about 1-8 per cent, 

 of sulphuretted hydrogen. 



The low-temperature gas, of course, varies with the co 

 used, and the hydrogen is often below 18 per cent.. an<i 

 the saturated hydrocarbons up to 68 per cent. ; but the 

 marked feature of the gas is that it contains hardly anv 

 benzene vapour, and owes its illuminating power to a sm; 

 percentage of ethylene, and chiefly to the gaseous memly 

 of the paraffin series (mostly ethane), and also that ti 

 percentage of hydrogen is low, very rarely reaching t! 

 amount present in the above sample, whilst bisulphide . 

 carbon likewise is very low. 



This I pointed out in the last course of Cantor lectur 

 which I gave in March, 190S, and it has since been civ 

 firmed by the work of Porter and Ovitz, White Park ai 

 Dunkley, and Burgess and Wheeler, 



The tar also varies with the coal used, and is sometim 

 as high as 23 gallons per ton, and in most cases averag' ' 

 20 gallons. 



The low-temperature tar is as distinctive in its character- 

 istics as the gas. It has a specific gravity of about 1-07; 

 is very liquid, and contains an abundance of light solv( i 

 oils, very low aromatic hydrocarbons, very little phem 

 but large quantities of cresol, no naphthalene and vr: 

 little anthracene, whilst the free carbon is, as a rule, belou 

 2 per cent. 



The ver}' low percentage of benzene in the light oils i- 

 made up for by the presence of paraffins, such as hexari< 

 heptane, and octane, whilst there are also present co' 

 siderable quantities of that curious group of hydrocarbo; 

 known as naphthenes or hexahydrobenzenes, which play - 

 important a part in Russian petroleum. 



.As before mentioned, carbolic acid occurs in ver}' smr. 

 quantities, but its higher homologues, such as cr^^svi; 



i 



