January 25, 19 12] 



NATURE 



431 



temperature distillation as if the coal had been carbonised 

 at the temperature existing on the retort surface. 



It is quite clear, however, that, coal being a bad con- 

 ductor of heat, and coke a worse one, it is only the layer 

 of probably less than an inch thick that is carbonised 

 at anything like the retort temperature, and that the 

 remainder of the charge is distilled at a slowly rising 

 temperature, which attains its maximum only after the 

 volatile products have been practically all driven off. 



The real distinction between high heats and lower flue 

 temperatures is that the higher the temperatures employed, 

 the thicker and hotter will be the layers of coke which 

 the gases and vapours have to traverse in their escape 

 from the inner portions of the charge, and the greater 

 will be their exposure to radiant heat and contact with 

 the highly heated surfaces of the retort in their outward 

 passage from the carbonising mass ; the products of the 

 primary action are, in fact, being subjected to secondary 

 decomposition under conditions we neither know nor can 

 control, and this is one of the weakest points in our 

 methods of carbonising for the production of illuminating 

 gas.^ 



We make elaborate tables of the composition of gases 

 and tars produced at various distillation temperatures, 

 but the 'only information that they give us is what is left 

 undecomposed under unknown and varying conditions, the 

 only certain factor being that the heat "was nowhere above 

 that which we are pleased to call the temperature of 

 distillation. 



It is evident that if these variations exist in the tem- 

 perature at which the coal is distilling in the comparatively 

 small charge in the gas retort, they must be accentuated 

 when one comes to deal with carbonisation in bulk as 

 practised in oven and chamber settings, as not only is 

 the travel of the gases and vapours through the red-hot 

 coke much longer, but the rate at which the heat is 

 conducted through the carbonising mass becomes slower 

 as the bulk of the charge increases, whilst the tempera- 

 ture in the crown of the oven during the first half of 

 the time is higher than is found in the gas retort, and 

 this also applies to the temperature in the top layer of 

 the coke. 



If the coal is carbonised in a 6-inch diameter tube 

 filled so that the heat shall be penetrating from every 

 side, there is an almost immediate rise in temperature 

 throughout the mass, owing to the hot gases and vapours 

 passing through the interstices between the pieces of coal, 

 and the coke attains its maximum temperature at the rate 

 of about one inch per hour, so that in three hours, with 

 a wall temperature of 1000° C, the centre of the mass 

 would be at about 950° C, and the carbonisation would 

 be finished ; if, however, the tube be increased to 12 inches 

 in diameter, the rate of conduction is reduced to 05 inch 

 per hour, and the same thing takes place with a flat 

 chamber retort heated from the sides, so that it would 

 take about twelve hours to complete the carbonisation ; 

 whilst with further increase in the width of the chamber the 

 rate of travel of the heat grows still less, the passage of 

 the heat being still slower as the distance between the 

 walls of the chamber gets greater. 



The result of this is that in by-product recovery coke 

 ovens and large chamber retorts the period of carbonisation 

 becomes very long, and the gas has to pass through so 

 much hot coke that the illuminating power is reduced to 

 nine or ten candles. 



These rates of passage of heat apply only to vertical 

 r'torts or chambers, the sides of which are heated, as 

 bottom heat penetrates the mass rather more quickly owing 

 to convection coming to the aid of conduction, and the 

 upward flow of heated gases raising the temperature in 

 iilvance of the conducted heat. 



Moreover, the rate at which the heat travels in the 

 carbonising- mass depends to a groat extent on the initial 

 temperature employed, the figures given being attained 

 only when the flues and outer walls of the retort or 

 chamber are heated to about 1100° C. (2012° F.), but if 

 the flue temperature is lowered, the transmission of the 

 heat becomes lower, and a longer period, therefore, is 

 rrquirrd for the complete carbonisation, the time taken 

 hi int; luariy inversely proportional to the temperatun' ; 

 sn that if in a 6-inch tube with a wall temperature of 



NO. 2204, VOL. 88] 



1000° C. (1832° F.) it takes three hours to complete car- 

 bonisation, it would take six hours to do the same work 

 with a wall temperature of 500° C. (932° F.). Conse- 

 quently, in making low-temperature coke, such as coalite, 

 in tubular retorts 5^ to 6^ inches diameter, it takes four 

 hours to drive off two-thirds of the volatile matter that 

 is in the coal. 



The temperature of the coke or coal through which the 

 gas and tar vapours have to pass, and the length of travel 

 they have in reaching the exit from the retort or chamber 

 in which carbonisation is proceeding, are two of the most 

 important factors in determining their decomposition, as 

 it is these which give rise to the secondary reactions that 

 largely determine the final composition of the gas and tar. 



Valuable pyrometric observations on the temperatures 

 existing in charges of varying size have been made by 

 Mr. Bond, of Southport, and other observers, from whose 

 work we can deduce the following results as typical : — 



If an ordinary D-shaped horizontal retort, 18 to 20 

 inches wide and 15 inches high, has a 6-lnch charge 

 fed into it, the space from the apex of the crown to the 

 top of the charge will be 9 inches deep. If now thermo- 

 couples properly protected are placed (i) at the bottom 

 of the charge, (2) in the centre, and (3) at the top of 

 the charge, we can gain a good idea of the way in which 

 the heat is acting on the coal. 



With full heats the coal at the bottom of the retort 

 rapidly heats up, and in fifteen minutes has reached 700° C. 

 (1292° F.), after which its rise in temperature slows down, 

 and it takes two hours to reach 800° C. (1472° F.) ; after 

 this it heats more rapidly, and attains 1000° C. (1832° F.) 

 at the end of four hours, and then there is practically no 

 rise in the last two hours of carbonisation. The tempera- 

 ture at the top of the charge rises more slowly, and by the 

 end of the second hour is only 740° C. (1364° F.), or 60° 

 cooler than the bottom, and remains at a lower tempera- 

 ture throughout the whole carbonisation. This is not to 

 be wondered at, as although the top flue of the setting is 

 1150° C. (2102° F.), and the bottom flue barely 1100° C. 

 (2012° F.), the coal at the top of the charge is being 

 heated largely by radiant heat acting across a considerable 

 gas space, whilst the bottom of the charge is in direct 

 contact with the heated bottom, and is taking in heat by 

 conduction. 



The thermo-couple in the centre of the charge throws 

 most light upon the course the distillation is taking, and 

 we discover that so great is the heating effect of the gases 

 and vapours passing up from the hot zone at the bottom 

 of the retort that at the end of the first hour the tempera- 

 ture is only 30° below that of the bottom, 730° C. 

 (1346° F.), whilst in t%vo hours it is at the same tempera- 

 ture, and then falls slightly below it for the rest of the 

 time, the rush of hot gases from the bottom having ceased, 

 and the temperature of the top of the charge equals the 

 centre only after the fourth hour. 



Now the fact that differences in temperature are so small 

 throughout the mass, and that during the whole of the 

 period when the bulk of the gas is being evolved the centre 

 of the charge is hotter than the top, points to the gas 

 forcing its way through the pasty mass of distilling coke 

 upwards into the space below the crown of the retort, 

 where it is baked by radiant heat from the mass of fire- 

 clay at 1050° C. (1922° F.), and the retort walls at 1050" C. 

 and the coke at from 700° to 1000° C. (1292° to 1832" F.) 

 are also in surface contact with it. 



The passage of the gas through the pasty coke causes 

 considerable swelling during the first hours of distillation, 

 and when the shrinkage in the charge of coke takes place 

 during the last two hours, the top portion, presumably 

 carbonised by radiant heat from the top of the retort, 

 shrinks over a smaller depth than the bottom and large 

 portion, so that when the charge comes to be drawp there 

 is found to be a fissure running horizontally between the 

 upper and lower portions, but nearer to the top, from 

 which vertical cracks branch to the top and bottom of the 

 charge. 



We are at present dealing only with the thermal con- 

 ditions existing during carbonisation ; but when we come to 

 siudv the more chemical side of the actions taking place 

 \\. shall see that such methods are the most brutal form 

 of distillation— high heats and small charges certainly mean 



