ThO(OCnH2n+1)2+2AH = 2(ACnH2n+1) + ThO2 + H2O. Thus if sulphuretted hydrogen were used as the acid thiols would be formed. This prediction has been verified, and the authors have prepared many thiols directly from the corresponding alcohols. The yield is good when the thiol is not too volatile, and can readily be condensed. Thiophenols can be prepared by passing a mixture of vapours of phenol and sulphuretted hydrogen over thoria.

--

2CH3CHO=2H+CO+CH3COCH3, propanone. 3CH3CHO=4H+2CO+C2H,COCH3, butanone. 2CH3CHO = 2H + CH3CO-COCH3, diacetyl. The acids are formed according to the following equations:2CH3CHO=C2H4 + CH3CO2H, acetic acid. 4CH3CHO=C2H4 +2C2H5CO2H, propionic acid. According to the experimental conditions one or other of these reactions may predominate. Water is formed by the condensation of the ketones with formation of mesityl oxide, phorone, and higher homologues, or by a more deep-seated destruction of the molecule :CH3CHO=H2O + C2H2.

Synthesis of Aromatic Nitriles.-F. Bodroux and F. Taboury.-Nitriles of formula C6H5-CR2-CN can be prepared from benzyl cyanide by the following reactions :C6H5-CH2-CN+2Na-NH2= = 2NH3 + C6H5—CNa2-CN C6H5-CNa2-CN+2R-X = 2NaX+C6H5-CR2-CN. The authors have thus prepared 2-ethyl 2-phenyl butane nitrile and 2-propyl-2-phenyl pentane nitrile.

Action of Hydracids on Glycidic Ethers.-Georges Darzens. Hydrochloric acid gives with 3-dimethylglycidic ether a-oxy-8-chlorisovalerianic ether. It is a crystalline substance, from which the dimethyl glycidic ether can readily be obtained by the action of sodium ethylate. When saponified with soda it is not possible to prepare the free acid, because of its instability in alkaline solution and its immediate transformation into isobutylic aldehyde. The action of hydrochloric acid on trimethylglycidic ether gives the corresponding ether, and similar results are obtained with hydrobromic acid. In all these reactions it is important to operate in absence of any trace of water. Hydriodic acid acts quite differently, the product with dimethylglycidic ether being dimethyl acrylic

acid.

Two New Isomers of Stearolic Acid.-A. Arnaud and S. Posternak.-The dihydriodic derivative of stearolic acid gives on reduction with zinc and acetic acid stearic acid, and on decomposition with alcoholic potash a com

The

plex mixture which contains the original stearolic acid and also two new isomers with the triple bond displaced to the next carbons on the left and right respectively. separation of the two isomers is effected by converting them into diiodo derivatives, and then subjecting these to fractional crystallisation. These reactions are not peculiar to stearolic acid, but apply also to other acids of the same series.

Hexahydrophenylglycolic Acid.-Marcel Godchot and Jules Frezouls.-This new compound can be prepared from hexahydrobenzoic aldehyde, which combines with hydrocyanic acid to give hexahydrophenylglycolic nitrile, from which the acid is obtained by saponification. The

MISCELLANEOUS.

City and Guilds of London Institute.-At a meeting of the Council of the City and Guilds of London Institute held on July 20th the diploma of "Associate of the City and Guilds of London Institute" was awarded to the following matriculated third year students of the Central Technical College who have completed a full course of instruction as prescribed by the Council:-Civil and Mechanical Engineering (50)-W. E. Bullock (awarded Bramwell Medal), H. R. Askew, H. G. Bambridge, R. E. Barton, R. O. Beit, C. St. J. Bird, M. A. Bulloch, R. Burton, E. R. A. Campbell, A. B. Chester, E. V. M. Crofton, C. J. Davis, W. R. Deuchar, A. P. Dicksee, D. M. Dinwiddie, C. L. Druitt, H. F. V. Ellison, E. P. Elworthy, W. J. Forsyth, F. R. Freeman, J. C. Gibson, A. C. Hartley, C. G. Hawes, E. S. Hoare, E. W. C. Jones, L. C. Kemp, L. A. E. Langlois, R. A. Mack, H. J. Masters, G. S. Mitchell, W. C. Mitchell, E. Moore, M. L. J. Morau, T. E. de Morsier, A. W. B. Pallister, P. G. Pappanicolaou, L. M. Paterson, H. G. Peake, G. T. Pound, J. H. F. Raper, G. A. N. Raymond, H. B. Richards, J. Richardson, H. G. Salmond, J. L. Sanders, E. H. G. Tomblings, D. C. W. Tonkin, A. D. Turner, N. L. Wallis, T. R. Young. Electrical Engineering (25) —J. D. Peattie (Awarded Siemens Memorial Medal and Premium), M. de Angelis, W. S. G. Baker, W. Barrs, P. R. Blamey, H. Clark, H. M. Dunkerley, D. H. Green, W. C. Griffith, G. E. W. Hitchcock, A. E. D. Kennard, G. Lambert, A. W. M. Mawby, P. A. McGee, D. J. McHatton, R. F. Niven, N. J. Perryman, L. C. Pocock, F. C. L. Racker, N. N. Roy, L. L. Ruderman, J. McL. Thornton, G. H. Tilly, F. J. Vinsen, G. Watts. Chemistry (5)-H. L. Armstrong, A. J. Daish, W. H. Franklin, F. Certificates have been awarded to 19 matriculated third W. Jackson, C. S. Mummery. In addition to the above, year students who have completed a full course of instruction at the Central Technical College, and to 56 students Technical College, Finsbury. who have completed a full course of instrucrion at the

NOTES AND QUERIES.

Our Notes and Queries column was opened for the purpose of giving and obtaining information likely to be of use to our readers generally. We cannot undertake to let this column be the means of transmitting merely private information, or such trade notices as should legitimately come in the advertisement columns.

Vacation Work. I should be greatly obliged for information as

to whether there is a possibility of working in organic chemistry during private laboratory.-Dr. R. V. the months of August and September, in London, either in a public or

IN a note communicated to the Society recently we described the preparation and properties of a polymeric form of carbon monosulphide. It was to be expected that carbon monosulphide would be a gas boiling at about -130° C.; yet the interaction of thiophosgene and nickel carbonyl at temperatures as low as -20° C. resulted in the formation of a polymeride, while when the same reaction was tried at -80° C. no change could be observed in the course of several hours.

The decomposition of carbon disulphide by light, studied by Sidot, also produces a polymeric form of carbon monosulphide, and we have shown that when this change

similar substance is appary means of ultra-violet light a

formed.

It seemed possible that carbon disulphide vapour at low pressures might be decomposed under the influence of the silent electric discharge or of the ultravlolet radiation associated with it. The conditions might then be favourable to the stability of the mono-molecular carbon monosulphide, provided a rapid current of the hypothetical vapour could be passed through the discharged tube; as the presumably unstable gaseous carbon monosulphide might remain unpolymerised, in the presence of sulphur and a large excess of carbon disulphide vapour, for a sufficient time to allow it to be condensed together with the unchanged carbon disulphide when cooled to the temperature of liquid air.

This anticipation was experimentally examined in the following way: - A bulb containing carbon disulphide was sealed on to an annular vessel through which the silent electric discharge could be passed, and the latter was sealed to a U-tube made of quill tubing and a bulb containing charcoal. While the bulb containing carbon disulphide was immersed in liquid air, the rest of the apparatus was thoroughly dried and cleaned by heating and exhausting by means of a good air-pump; it was then sealed off and the charcoal bulb cooled in liquid air. The apparatus made solely of glass was now so thoroughly exhausted that no discharge would pass through the ozoniser." The bulb containing carbon disulphide was now transferred from liquid air to a paste of solid carbon dioxide and ether, and thus maintained at a temperature of -80° C. In this way a rapid stream of carbon disulphide vapour at a steady pressure of about 1 mm. passed through the "ozoniser " and the U-tube, and was condensed in the charcoal bulb. On cooling the U-tube in liquid air, the carbon disulphide vapour was condensed and solidified as a white crystalline solid ring, When, after five or ten minutes, the U-tube was removed from the liquid air and allowed to rise in temperature, the carbon disulphide melted and quickly evaporated into the charcoal bulb, leaving no residue. The charcoal vacuum was maintained throughout the experiment, in order to ensure the absence of pressure due to any permanent gas.

While the U-tube was cooled in liquid air, the silent discharge was caused to pass through the vapour in the "ozoniser," causing a bluish light to fill the vessel: the discharge could be stopped at once by cooling the carbon disulphide bulb in liquid air. After the discharge had been allowed to pass for five minutes, a faint brown ring was noticed in the U-tube about a centimetre above the level of the surface of the liquid air. The liquid air was then removed, and the temperature of the U-tube allowed to rise, when a flash of light, accompanied by a detonation, was observed, and the surface of of the U-tube was fouud to be partly covered with a dark brown deposit. The detonation was usually gentle, but was sometimes so violent that the tube was shattered. A similar explosion may take place in the charcoal condenser.

When the conditions are arranged so that the temperature gradient in the U-tube is steep, as it is when the vacuum vessel surrounding it is filled to the top with liquid air, then the transformation may take place during the condensation, and no flash will occur when the temperature is allowed to rise. If, on the other hand, the temperature gradient is not steep, as when the vacuum vessel is only half filled with liquid air, the flash can always be observed in a darkened room and frequently also in daylight.

On examination, the interior of the "ozoniser" is found to be partly covered with a thin brownish deposit which consists partly of sulphur and can be dissolved away by carbon disulphide. When the potential of the discharge was increased until sparks passed through the air outside the "ozoniser,” then its inside surface became coated with a brown deposit similar in appearance to that which was found in the U-tube.

In order to show that the brown deposit in the U-tube was due to the change of a gaseous substance, a plug of

O'125

Hemingway's acid arsenate. Sol. As2O5

Per cent of moist arsenate.

Total CaO 4'380

A$205

O'012

The above results indicate that eight times as much arsenic is rendered soluble from the acid arsenate as from the neutral, or calculated to the original material this would be equivalent to 0.25 per cent of soluble As2O5 from the neutral and 1.98 per cent from the acid. Distinct losses of sulphur and lime have also taken place in the acid arsenate mixture, and it is evident that there is a mutual decomposition when lead acid arsenate is mixed with the limesulphur solution.

Analysis of the dried residues give further evidence of the breaking down of the acid arsenate under these

conditions::

Analysis of the Residue from the Mixture of Lime-sulphur and Lead Arsenate. Neutral arsenate residue.

Per cent.

0'70
I'47

Acid arsenate residue. Per cent.

20.80

14.80

10'40

Free S PbS CaO.. Sulphur was determined by extracting the dry residue with CS2; the lead sulphide by extracting the remaining residue with HNO3 1:10; the lime by precipitation as Oxalate after removing lead and arsenic with hydrogen sulphide.

A study of the reactions involved in the above mixtures indicates that there is a partial interchange of the lime and lead resulting in the formation of calcium arsenate and lead sulphide respectively, free sulphur being at the same time deposited. The soluble arsenic is then derived from partial solution of the calcium arsenate thus formed. Sulphides of arsenic are not formed, as these sulphides are soluble in the alkaline lime sulphur forming sulpho salts, and sulpho salts are not present in the solution. The reactions as shown above are much more pronounced with the acid than with the neutral arsenate, and it is therefore advisable to employ the neutral form when desiring to combine limesulphur and lead arsenate.

Having found that the alkaline lime-sulphur solution reacts quite readily with the acid lead arsenate, the tests were extended to determine the comparative solvent action of alkaline and saline waters on these arsenates.

Haedden has stated (Bull. 131, Col. Exp. Sta.) that lead arsenate is more soluble in waters containing alkali salts than in normal water. In the tests which are reported in the succeeding table a quantity of the moist arsenate, equivalent to 10 grms. of the dry arsenate, was treated in each case with 2 litres of solution, except in case of the CO2 test, in which 2 grms. of the moist arsenate were used. The solutions were kept at room temperature (about 17° C.) for twenty-four hours with occasional shaking,, then filtered, and the filtrate analysed for soluble arsenic by the modified method of Gooch and Browning (loc. cit.), 500 cc. equivalent to 2'5 grms. of the arsenate being used in each determination. The alkali water used in the tests was the extract from a natural alkali incrustation containing 50 per cent Na2SO4, 18 per cent NaCl, 13 per cent Na2CO3, and 2 per cent NaHCO3. The arsenic is reported as per cent of the original material.

This

It appears from the above results that both forms of the lead arsenate are more soluble in saline waters than in pure waters. Alkaline carbonate waters especially exert a solvent action on these arsenates, and the reaction is much more pronounced in the case of the acid arsenate. is, perhaps, due to the fact that lead carbonate is extremely insoluble, and that a base with which arsenic forms a soluble salt is present. It is evident that waters containing considerable quantities of alkali carbonates should be

Distilled water.. Distilled water + CO2 for two hours

0.23 per cent NaCl, 0.46 per cent CaCl2

o.10 per cent Na2SO4

0'201 0.106

Alkali water, 40 grains to gal. 0'536

0.108

0'310

0*155

1.833

THE following investigation was undertaken in connection with the preparation of an extensive series of a3-unsaturated ketones for the purpose of comparison by physical measurements. Many of these ketones are very easily obtained by the Schmidt-Classen (Ber., xiii., 2342) condensation of an aldehyde with a saturated ketone, but with aliphatic aldehydes, and with aromatic aldehydes in which the carbonyl group is in an aliphatic side-chain, the method generally gives aldols from which water is eliminated with difficulty, or else complex products not closely related to unsaturated ketones. We decided therefore to ascertain to what extent the Friedel and Crafts reaction can be used for preparing such ketones. The investigation led to the discovery of a number of facts of general interest in connection with the Friedel and Crafts reaction. These are published in this paper.

The literature on the subject under investigation is meagre. The reaction was first applied to unsaturated chlorides by Stockhausen and Gattermann in 1892 (Ber., xxv., 3536). These investigators started with the chlorides of cinnamic and phenylpropiolic acids, and obtained an excellent yield of the expected unsaturated ketone when they used anisol or phenetol, but could isolate no solid products when they used aromatic hydrocarbons. They attributed this result to the difference in the reactivity of phenol ethers and hydrocarbons.

Many years later Moureu studied the reaction between the chloride of acrylic acid and benzene in the presence of aluminium chloride (Ann. Chim. Phys., [7], ii., 198). He isolated a small quantity of a solid product that he assumed to be vinylphenyl ketone, but which was, in reality, a-hydrindone. It was shown a few years ago, that a satisfactory yield of unsaturated ketone can be obtained both with crotonyl chloride and benzene (Am. Chem. Fourn., xlii., 375), and with cinnamyl chloride and mesitylene (Ibid., xxxviii., 510).

We have found that the reaction between cinnamyl chloride and benzene takes place almost as readily as that between the same chloride and phenol ethers, but the product is a mixture of two saturated ketones - diphenylpropiophenone and phenylhydrindone :1. C6H5CH:CHCOCI+2C6H6=

= (C6H5)2CHCH2COC6H5 + HCl.

64

We

The reaction expressed by the second equation represents a new indene synthesis which differs from that devised by Kipping in that a benzene nucleus introduced by means of the Friedel and Crafts reaction, and not one already present in the acid chloride, forms the foundation of the bicyclic system. The difference is illustrated by the following equations, which represent the synthesis of a-hydrindone :— I. CH¿CH2CH2COCl → C6H4<CH2>CH2+HCI, II. CH2:CHCOCI CH2:CHCOC6H5 → C6H4CO

CH2 CH2.

The difference in the results obtained when cinnamyl chlorlde is condensed with benzene and with anisol may be due, in part, to the greater reactivity of phenol ethers, as assumed by Stockhausen and Gattermann; but this is certainly not the principal factor. With crotonyl chloride and benzene it is possible to conduct the reaction so as to get, at will, either saturated or unsaturated ketone. In concentrated solutions and at low temperatures the product is the unsaturated ketone, because under these conditions the aluminium chloride derivative of this ketone is so sparingly soluble that it separates quickly, and thus escapes further reaction. In more dilute solutions and at a higher temperature the aluminium chloride derivative remains in solution long enough to react with a second molecule of benzene, and this process takes place so readily that when only one equivalent of benzene is used nearly half of the acid chloride is left unchanged. The determining factor in this case is therefore the solubility of the aluminium chloride derivative of the unsaturated ketone.

The reactions with cinnamyl chloride cannot be controlled like those with crotonyl chloride because the double compound between the acid chloride and aluminium chloride is so slightly soluble that it is necessary to use higher temperatures and more dilute solutions. Phenol ethers give unsaturated ketones, nevertheless, because the alkoxyl groups hinder or prevent the secondary reactions that lead to saturated compounds. Thus indene synthesis does not take place because this would necessitate closure of the second ring in the meta position with reference to alkoxyl: C6H5CH:CHCOCI+C6H5OCH3 =

= C6H2CH: CHCOC6H4OCH3(f).

Our experiments show that the secondary reaction involving addition of a molecule of phenol ether to the unsaturated ketone first formed is likewise impossible probably owing to the stability of the double compounds that alkoxy ketones, as well as phenol ethers, form with aluminium chloride. For similar reasons it is just as easy to get unsaturated ketones by using mesitylene as it is by using phenol ethers.

C6H5CHBгCHBrCOCI+ C6H5=

CHC6H5 CHBr CO

+ HCl + HBr.

Similar products were obtained when the same acid chloride was used with other aromatic compounds. The yield was poor in all cases because aluminium chloride attacks the acid chloride itself at the lowest temperature at which it is possible to carry out the reactions.

With chlorides of the corresponding dibrom aliphatic acids, however, the Friedel Crafts reaction gives an excellent yield of dibrom ketones.

Experimental Part.

Cinnamyl Chloride and Benzene.-In the first experiments equimolecular quantities of chloride and hydrocarbon were dissolved in carbon bisulphide and treated with aluminium chloride, but we found that the product always contained a large quantity of unchanged chloride. This was the case, also, when the acid chloride was first transformed into the Perrier double compound with aluminium chloride, and this treated with one molecule of benzene dissolved in carbon bisulphide. In all subsequent experiments the hydrocarbon was therefore used in excess. reaction was studied under the following conditions:1. In carbon bisulphide cooled in a freezing mixture (20). No perceptible reaction took place in the dark. In direct sunlight the aluminium chloride dissolved, then the Perrier double compound separated in small yellow crystals which re-dissolved very slowly with evolution of hydrochloric acid.

The

2. In carbon bisulphide at the ordinary temperature. The aluminium chloride dissolved with brisk evolution of hydrochloric acid, and the reaction was complete after the addition of slightly more than one molecule of aluminium chloride.

3. The double compound between cinnamyl chloride and aluminium chloride was made in carbon bisulphide, the solvent removed by decantation, and the crystalline solid covered with benzene. The reaction that started in the dark was completed in the sunlight.

The orange-coloured liquids obtained in these reactions with ether, and the extracts washed with sodium carbonate were poured into iced acid, the organic products extracted and dried. The yellow liquids left after removal of the solvents in no case contained benzolacetophenone, for while they reduced potassium permanganate they did not combine with bromine. After standing in an ice-chest for several days, the residue obtained by the first procedure partially solidified. The solid was purified by crystallisation from methyl alcohol, from which it separated in large colourless prisms, melting at 78°.

Analysis.

01217 grm. substance gave o 3858 grm. CO2 and 0.0657 grm. H2O.

C H

Calculated for C15H12O.

86.6 5.8

Found.

86'4 5'9