1286 THE COLLIERY GUARDIAN December 24, 1915. A B C D E F G Benzol in 10,000 cubic feet unscrubbed gas Benzol in return gas Benzol in rich oil Benzol in debenzolised oil from still Temperature of oi‘ passing to still Temperature of oil issuing from still 2’5 to 2’7 gals. 0’4 to 0’75 „ 3’6 p.c. 0’6 p.c. ? 110° C. 2’5 0’5 2’5 p.c. ’03 ? 105° C. 2’32 0’07 4 5 p.c. 0’2 to 0’3 p.c. 108° C. 112° C. 3’0 0’09 to 0’15 4’5 p.c. 0’8 to 1’2 p.c. 120° C. 9’5° C. 3’3 0’5 to 0T5 3’5 p.c. 0’5 p.c. 85° C. 105° C. 2’52 0'60 3’ 1 p.c. 0’9 p.c. ? 3’0 0’25 to 0’40 2'75 p.c. 0’25 to 0’9 p.c. 15’25° C. 104° C. more or less completely into its constituents, these last to conform with commercial specifications. The chief impurities to be removed are :— Carbon bisulphide CS2, thiophene C4H4S, thiotolene C5HflS, paraffins Cn H(2n + 2), phenols Cn H(211 — )7OH, pyridine C5H5N, aniline C6H5NH2, naphthalene CJ0H8, creosote oil and resinous matters (such as cumarone, indene, etc.) The customary procedure for the removal of impurities is to wash with 168 degs. Twaddle sulphuric acid, then with water; acids are next washed out with 20 degs. Twaddle caustic soda solution, and the benzol is re-washed with water. This washing may take place on the benzol, as recovered, or the naphthalene and creosote may be separated by a primary dis- tillation with wet steam, known as “ blowing over,’’ and the purer distillate washed as above. A further alternative is to separate into crude fractions, 'which are washed when sufficient of any one is obtained. The last method presents numerous advantages, amongst which are economy of acid, smaller washing losses, and a reduction of the work to be performed by the rectifying still. Since it is probable that, in future, commercial pure products will be in greater demand than formerly, this advantage should be considered., The plant necessary for washing is a specially-shaped vessel of boiler plate, usually lined with lead, of two or three thousand gallons capacity, and provided with mechanical means by which the benzol can be thoroughly agitated with the washing material. Basic impurities are easily removed by acid washing, but if it is necessary to remove thiophene also, the acid washing must be repeated and prolonged until the laboratory sample gives a negative result by the isatin test. In any case, the completion of the washing should be ascertained by distilling a sample from the washer, and treating the distillate corresponding to 90 per cent, benzol with 10 per cent, of pure sulphuric acid. A well-washed benzol will not colour the acid darker than decinormal potas- sium bichromate solution. The .acid washing is followed by a water washing. Here care must be taken to avoid violent >agitation, the water being preferably used to rinse the sides of the washer, as some benzols have a tendency to emulsify with water and cause large washing loss. The acid washings treated with water separate into solid resinous matter and an acid liquor, from which pyridine may be by treatment with ammonia liquor, the resulting sulphate solution removed being used in the sulphate house. The benzol is then washed with an excess of 20 degs. Twaddle caustic soda solution, which removes the traces of sulphuric and sulphurous acids and any phenols that are present. The caustic solution may be used until spent. The quantity of acid used is generally about 6 to 10 per cent, of the weight of benzol, and washing losses may be from 6 to 16 per cent, on the same. The still employed for “ blowing over ” the benzol is of three to four thousand gallons capacity, has 12 to 15 trays in the fractionating column, and is provided with direct and indirect steam heating. The charge may be considered as worked off when the distillate boils at 200degs. Cent., or when naphthalene becomes evident. The once-run benzol must then be separated into commercial fractions by dis- tillation in a rectifying .still. Many types of these have been erected, the following being the essential features :— The body of the still should be of the short horizontal, cylindrical shape, the heat being supplied by a large number of steam pipes, running through the still, and by a wet steam coil. The fractionating column should be of ample capacity, the two main varieties now at work being the cap-and-plate, introduced into benzol distillation by Heckmann, and the perforated-plate originally adopted in Coffey’s still. Differences in detail are found in the size and shape of the caps, depth of seal, position, and shape of overflow; also in the number and size of holes in the perforated plate. It is worth noting that whilst newcomers in this field erect complicated plant, with cap-and-plate rectifying columns, benzol distillers of old standing show a preference for simple plant and perforated-plate column. As a number of stills of the latter kind are successfully producing pure benzene and toluene, it would appear that the opinion prevailing in some quarters, that only the cap- and-plate column will produce a pure product, is scarcely correct. There are also two main types of analyser; one being a cylindrical vessel about half the size of the fractionating column, and filled with copper balls or other condensing material; whilst the other is the well-known multitubular condenser supplied with a water feed, the benzol vapours passing around the tubes. Differences are also found in the point at which the analyser discharges its condensate into the fractionating column. The cooler is usually of the worm type. The varieties of apparatus for separating the benzol and water all work on the principle of syphoning the water from beneath the benzene, which overflows directly, and is controlled by the height of the syphon. Some Typical Results. The chief products obtained by distillation of benzol are shown in the following table :— Comm er ial name. Per cent, of pure Specific gravity i5f;c. Range of distillation at barometric pressure 760 mm. Benzene. Toluene. Xylene. Cumene. 90 per cent, benzol 84 13 3 — ’880 90 per cent, between 81° and 100° C. 50 per cent, benzol ... .. 43 46 11 — ’875_ 90 per cent, between 89° and 120° ; 50 per cent, at 100° C. Commercial toluol 15 75 10 — •870 ’ 90 per cent, between 107° and 120° C. Solvent naphtha — 5 70 25 •875 90 per cent, between 135° and 160° C. Heavy solvent naphtha — 5 80 •920 90 per cent, at 190° C. Benzol for pure benzene 95 5 — — — 90 per cent, at 90° C. A coke oven benzol with retort test (specific gravity, 0-887) Degs. Cent. Percent. Degs. Cent. Percent. 100 ............ 51 I 160 .............. 82| 120 ............ 72£ | 180 .............. 87i gave on primary distillation in laboratory without washing :— 100 degs. Cent.......... J00 to 120 degs. Cent.... 120 to 160 160 to 200 Residue ................ Loss on distillation ... 70’6 WO 7*6 6*3 4'6 0’9 crude 90’s. ,, 50’s. „ solvent naphtha. ,, heavy. creosote oil, etc. 100’0 by volume. Further fractionation gave the following results after washing :— Washing losses 8’2 Specific gravity. Distillation 2’8 — CS2 fraction 2’0 0'890 Pure benzene 54’0 0’884 ,, toluene 11’8 0 872 Solvent naphtha 11’4 0’875 Heavy „ 5’0 0’930 Creosote oil 4’8 — 1000 Works distillation of 610 6'5 5’0 a similar benzol gave 90 per cent, benzol. ,, toluol, solvent naphtha, heavy ,, Loss (washing, creosote, etc.) being 15 per cent. The points at which the distillates should be changed into different receivers may be ascertained in two ways. The approximate quantities of each distillate are usually known. If, therefore, the tank into which the distillate is running gives the correct distillation test, the receiver should be changed. This necessitates the constant attendance of some- one able to make and interpret the tests. Alternatively, a litre sample may be fractionated in the laboratory, and, the results being compared with the works results, it is possible to forecast the yields from future distillations with a fair degree of accuracy. In working for pure products, special cuts are made, and the progress of distillation must be care- fully watched, since towards the end of distillation changes take place with rapidity. It is perhaps needless to remark on the great risks of explosion and fire in the distillation house; but it is interest- ing to note that air in contact with benzene at ordinary temperature contains over 7 per cent, of benzene vapour at the surface of contact, and as the explosive range of benzene vapour in air has been found to lie between 2-6 and 6-5 per cent., it is evident that precautions against explosion cannot be too complete. The Future. As to how far coke oven owners should extend their activities in chemical manufacture is a matter of opinion. Tar distillation and the production of intermediate com- pounds, amongst which benzol derivatives occupy a highly important place, have been profitable industries in the past; still, little of the profit has been devoted to the extension of existing processes, and less to the 'acquisition of the technical knowledge necessary for an incursion into the fine organic chemical and colouring matter industry, now chiefly in the hands of alien enemies. If, then, the manufacture of inter- mediate compounds becomes a branch of the coking industry, the present producers will be compelled to turn their atten- tion to processes which really require technical skill and ■scientific direction, so reviving what was until recently a moribund industry, and thereby justifying coke oven enterprise. DISCUSSION. Mr. Stanley Cooper agreed with the author’s con- clusions that there was room for improvement in the working of the majority of the benzol plants in the country. There were other points in the paper, how- ever, with which he could not agree, and upon which he would like further information. The question of the falling off in calorific value and loss in illuminating power of the gas resulting from the removal of benzol was one to which he (Mr. Cooper) had given some atten- tion, and it would be of interest to know how far the carburetting influences of the creosote oil compensated for the loss of benzol. It would be found that the theoretical loss did not agree with the practical loss. Although he had had no experience of the rotary scrubbers, he was told they were not quite as cheap to maintain as they were claimed to be, and some particulars as to relative costs would be valuable. He did not think there was much advantage in the matter of space, because a tower scrubber did not occupy a great deal of space, and there was no limit to its height, and the absorption area could be increased in that way. The table of results given by Mr. Taylor certainly gave material for careful thought and study. With regard to the use of superheated steam, as suggested, he would like to know what steam pressure one might go .to without getting superheated steam. The usual working pressure on a benzol plant was 100 lb.; but if one could get 100 1b., was it necessary to superheat beyond that, or would the 100 lb. answer the same purpose? He did not think that, with ordinary steam pressure, the results should be 5 per cent, below those with superheated steam—provided other circumstances were equal, and there were sufficient area in every case. Information with regard to continuous rectification would also be welcome. In the petroleum industry there was a con- tinuous type which certainly had some advantages. In certain plants the dephlegmatory action was not con- fined to a column at the top of the still, but a separate condenser was employed thau brought off a fraction usually above 90 degs. Cent., and in the majority of cases that fraction was allowed to crystallise. That fraction he had tested, and in several cases it was found to contain a portion of toluene. If that went back into the oil, did they eventually get an oil which was more or less saturated with toluene? If so, then in the 65 per cent, they would get an excessive proportion of toluene. If they made 65 per cent, benzol with this particular type of plant they would recover this so-called toluene fraction, and run it back in the scrubber. It would be interesting to hear whether it would not be more advantageous, in present circumstances, to make 60 per cent, benzol, and to get for it the same price as they were getting for the 65 per cent. He maintained they would get the same proportion of benzene and more toluene, which was worth more than 8d. per gallon; therefore, they would get the same price for 60 per cent, benzol as they were getting for 65 per cent., with a corresponding larger revenue to the country and a bigger supply of toluene. Another point on which Mr. Taylor might) have touched was the method of testing benzol sent out to large buyers. Most producers made the old retort tests, fin., or Jin. from the bottom, which was absolutely unscientific. He thought that associa- tion might take up that question, and see if they could not secure some absolutely standard method of testing, with standard apparatus and standard conditions. At the present time tests were very haphazard. He wondered whether the development of the recovery of benzol would proceed along the same lines as the recovery of ammonia, and whether there would be a synthetic benzol. There were possibilities, he thought, in that direction. Mr. Taylor, in reply, agreed that the theoretical and practical results obtained in extracting benzene from gas did not correspond. It largely depended upon the wash- ing plant. As to the carburetting action of the oil, in his experience it was safe to strike out about 10 British thermal units for every gallon of benzol per ton of coal. With regard to the rotary scrubbers, these did occupy less ground space than tower scrubbers. As a rule, for 60 ovens three tower scrubbers would be required, whereas the same plant could be dealt with by a rotary scrubber of 8 ft. diameter, and about 10 washing chambers. Moreover, with tower scrubbers, more pumps were required than with the rotary type, and pumps took more upkeep than the rotary scrubber engine. He was not aware of any place that was work- ing the continuous process for the rectification of benzol. Primary distillation had been effected in a continuous manner, and there was no reason why the continuous rectification should not be adopted, although it might be rather difficult to control fractions. He agreed with Mr. Cooper with regard to 60 per cent, benzol, and thought it would pay coke producers to make 60 per cent, benzol rather than 65 per cent.; but in some cases people specified for 65 per cent., and would have it, whether they lost by it or not. He hardly agreed with Mr. Cooper about standard tests for benzol, for, in the majority of cases, the retort test would give a result correct to about 0-75 per cent. Mr. D. V. Hollingworth did not think there was quite as much doubt as to the composition and forma- tion of benzene and its homologues as Mr. Taylor sug- gested. The formation had proceeded on fairly well understood lines, and the syntheses were almost cor- rectly known and tabulated. With regard to the loss of calorific value and illuminating power of gas from which benzol had been extracted, he could give a personal experience. Like other gas works chemists, he had to put up with the inconvenience of a loss of benzol from coal gas in the winter,' and in order to ascertain how much loss in illuminating and calorific power was to be expected in a range of miles from the works, a particular research was carried out. One chemist determined the illuminating power and calorific composition at the works, day and night, while another did the same thing three miles away. The result established that, without doubt, coal gas did lose its benzol content to some extent by transit through the pipes : it lost some of its calorific value, and conse- quently some of its illuminating power. Concerning the question of the proportions, he believed that the quan- tity of hydrocarbon in coal gas depended upon several factors, one of which was related to naphthalene and the constituents of creosote. The greater the percentage of benzol in the gas, the greater the quantity of naphtha- lene and similar hydrocarbons that gas would carry. Consequently the amount of creosote which could be carried after benzol extraction depended on the amount of benzol which had been left in the gas. It had been found that a certain proportion of benzol in the gas would carry a definite quantity of naphthalene or hydro- carbon, so that the matter had long since passed out of