Modern Gasworks Practice

'WATER GAS 739 wire-drawn, the vesicles being made to agglomerate and form a rain. The gas is then deflected upwards on to the underside of the top cone, which. again deflects it dowawards on to the top of the lower cone. This lower cone receives the deposit of tar rain, which trickles from the edges and runs away to the seal. There is no doubt that wire-drawing, if effectively carried out, can be used successfully for the removal of tarry vesicles. It would appear, however, that the principle of bubbling as embodied in washers of the Livesey type is difficnlt to improve upon. A washer of this description (into which a small stream of liquor from the water gas scrnbber is constautly permitted to run), fitted prior to the carburetted water-gas purifiers, will usually eliminate almost the whole of the suspended particles in the gas. It must not be forgofrten, moreover, that efficient condensation is an important item in this respect, and if the gas is permitted to leave the condensers at an abnormally high temperature, trouble must be expected. In the second direction water-gas tar is in many cases an annoyance in that, being of practically the same specific gravity as water, it admixes with the latter and forms an emulsion of a very permanent character. The consequence is that the tar as made ready for sale may contain so much as 40 per cent, of water. Various means, including centrifugal treatment, heating, etc., have been experimented with for the purpose of effecting separation, but the matter still remaiiis something of a problem. The main characteristics of water-gas tar are low specific gravity of the tar itself and the creosote obtained from it, low proportion of free carbon, and practical absence of naphth alene and phenols. Also in all fractions a large amount of paraffin bodies are found. On fractioaation an average sample of carburetted water-gas tar shows the follow-ing result:— Distillates j’rom Wateb-Gas Tar Yield^per 1,000 gallons of tar. Water ......... 35-0 gallons. Benzol, 90 per cent, at 100° C. . . . 0-75 „ Toluol, 90 „ „ 120° C. . . . 13-5 Naphtha, 90 „ „ 160° C. . . . 28-0 Heavy naphtha, 90 „ ,, 190° C. . . . 14-5 „ Creosote ......... 539-0 ,, Pitch (sp. gr. 1-199) ....... 1 ton 18 cwt. Löss ........ about 12 gallons. The proportion of benzene to toluene varies considerably and is mainly dependent upon the temperature at which. the carburettor and superheater are operated, also upon the point at which the tar is collected. W. Kirby, for instance, shows by the folio wing analysis that while tlie yield of benzol and toluol together is practically identical with the sum of the two in the above figures, the proportion of benzol is very .much higher.