THE DIRECT COMBINATION OF ETHYLENIC HYDROCARBONS ETC. 307 XLVII1.-The Direct Combination of Ethylenic Hydro-carbons with Hydrogen Xzclphites. By ISRAEL KOLKER and ARTHEX LAPWORTH. AGENTS such as ozone which attack the simplest compounds, have been distinguished (as " Class I ") by Lapworth and MeRae (J. 1922 121 2741 ; compare also Lapworth &fern. fManchester Phil. Xoc. 1920 64 iii 11) from others (" Class I1 ") which M 308 KOLKER AND LAPWORTH THE DIRECT COMBINATION OF are inert towards ethylenic hydrocarbons but which nevertheless are additive to the ethylenic group when the latter occurs in the a-position with respect to a carbonyl cyano or similar group. It is significant that all the known agents of the second class are either (IIa) metallo-compounds or (IIb) compounds of the general type H-Z in which the hydrogen atom is capable of being replaced by the action of metals.Examples of sub-class IIa are Grignard reagents potassium cyanide and sodio-derivatives of ketones esters and nitriles . Examples of IIb are HCN HCH(CO,Et), HNR,R, the last includ-ing amines hydroxylamines and hydrazines . The electronic theory of metallic compounds suggests that the radical of each agent of class IIa is capable of existence in all cases as a negatively charged ion Z-. In some cases the free ion Z- is identical in consti-tution with the radical as it occurs in the compound H-Z (example H-NH,) ; in other cases it is probably not ; (example : H-CH,.NO,) but the structural difference is such that intracon-version readily takes place in either direction.As is well known, agents of sub-class IIb also form addition products with many saturated carbonyl compounds. Not all compounds which have the characters defined in the preceding paragraph can be referred to Class IIb ; thus powerful acids are by definition excluded from this class as they attack ethylenic compounds of all types. It is further necessary to observe that in referring any agent to Class IIb there are restrictions as to the experimental conditions prevailing while the agent is applied. For example hydrogen cyanide may properly be referred to Class IIb when applied in presence of an alkaline catalyst but not in presence of an acid catalyst unless it has been shown that under the same conditions the combination is inert or nearly so toward ethylenic hydrocarbons.Within the range occupied by IIb which must be limited a t one extreme by acids too weak to attack ethylenic compounds at measurable speed and at the other extreme by very feebly ionisable compounds such as ammonia there does not at present appear to be any direct relation between the additive efficiency of the agent and the electro-affinity of the anion. Thus many very weak acids appear almost or quite unable to form stable addition products with carbonyl compounds. With reference to Class I1 as a whole it is probably true that no compounds of which the ion Z- has a very high affinity for the charge can be included in either sub-class (compare Lapworth, Zoc. cit.). There are for example no cases recorded where the potassium or sodium salts of powerful acids form additive corn ETHYLENIC HYDROCARBONS WITH HYDROGEN SULPHITES.309 pounds with carbonyl compounds. The powerful acids have already been excluded for the reasons above stated. From the preceding considerations it is clear that any agent which forms addition products with the carbonyl group of aldehydes and ketones or with the ethylenic linking in ap-unsaturated carbonyl compounds may safely be referred to sub-group IIb only when it is known that the agent does not attack ethylenic hydrocarbons under comparable conditions; in many cases as in the instance of ammonia experimental evidence is already extensive and <o uniformly negative that this indifference may confidently he inferred whilst in other cases further investigation is required.The correct classification of agents additive to ethylenic com-pounds is of the utmost importance in studying the influence of atoms and groups on the properties of others in the same molecule. The authors have made a careful study of two series of agents previously referred in the papers above specified (Zoc. cit.) to Class 11. The experiments carried out by the authors and by other workers in these laboratories on the possible addition of hydrogen cyanide and metallic cyanide to ethylenic hydrocarbons including cycio-hexene and styrene have given uniformly negative results. As the addition products in these cases would have been nitriles easily convertible into carboxylic acids and so capable of detection even in traces it may be concluded that metallic cyanides and hydrogen cyanide in absence of acid catalysts are indeed highly selective ant1 that when they do attack an ethylenic linking the latter is almost certainly affected by conditions similar to those which ohtr~in iri @-unsaturated ketones.The other series of reagents tested were sulphites and iuore especially hydrogen sulphites. The sole instance hitherto recorded (so far as we have been able to discover) of a hydrocarbon combining directly with hydrogen sulphites is that of styrene and even in that, case the published evidence was inconclusive (Miller dnticrlctr, 1877 189 340; LabbB Bull. SOC. chim. 1893 [iii] 22 1077; Dupont and Labaune Sci. Ind. BzZZ. 1912 [iii] 7 3). In thr: case of styrene moreover there was an element of doubt whether phenyl can exercise the same influence on a double bond as carbonyl can.Similar uncertainty existed in the theoretical interpretation of the observations of Dupont and Labaune (loc. cit.) nho obtained additive products of hydrogen sulphites with unsaturatctl alcohols. The results obtained by the present authors remove all doubt on many of these moot points. Hydrogen sulphites combine directly with ethylenic hydrocarbons and the failure of previous investi 310 KOLKER AND LAPWORTH THE DIRECT COMBINATION OF gators fully to establish this fact is attributable to the unsuitable experimental conditions used. Special precautions to ensure intimate contact between the hydro-carbons and the aqueous solutions are usually but not invariably necessary.They may consist in emulsifying with the aid of purified kieselguhr. Contrary to expectations dilution of the sulphite solution possibly by an effect on the solubility of the hydrocarbons, favoured the interaction. A not less important practical feature in the later experiments was the use of ammonium hydrogen sulphite instead of the sodium or potassium salt. This device admits of the ready removal of any excess of the reagent by boiling with sufficient barium hydroxide, when insoluble barium sulphite is formed ammonia is expelled, and such organic addition products as are not destroyed during this process are easily isolated as soluble barium salts. Pinene dipentene cyclohexene crude " amylene " and even the completely symmetrical ethylene combine readily under favourable conditions with hydrogen sulphites usually in the cold yielding products which for the most part appear to be the salts of the expected saturated sulphonic acids ; but these are usually mixed with a variable proportion of other salts apparently isomeric with the sulphonates.These secondary products unlike the true sul-phonic acids are hydrolysed by dilute acids or alkalis and are probably salts of the alkyl hydrogen sulphites though some slight doubt attaches to this view of their constitution. These salts appear to be considerably more stable than the salts of alkyl hydrogen sulphites prepared by union of sulphur dioxide with sodium alkyl-oxides but this may be due to stabilisation by the other salts present ; indeed the authors have observed (1) that the hydrolysis of the salts of methylethionic acid SO,H*CH(CH,)*CH,*SO,H (which normally is readily effected by dilute acids) takes place much more slowly if excess of salts of methylisethionic acid, SO,H *CH (CH,) *CH,* OH , are present and (2) that pure barium methylethionate decomposes a t loo" but in presence of 9-10 times its weight of isethionate does not decompose appreciably below 150".It is just possible however, that two series of alkyl sulphurous acids exist. It seems probable that carbonyl compounds react with bisulphite, as with metallic cyanide by first capturing the anion. Ethylenic hydrocarbons evince no definite preference for anions and may react with bisulphites either by first capturing the hydrogen ion or by uniting with the unsaturated centres of the sulphite molecules much as they unite with ozone and possibly in both ways.The gross result of the process of addition of hydrogen sulphite ETHYLENE HYDROCARBOKS WITH HYDROGEN SULPHITES. 31 1 to ethylenic hydrocarbons may thus be represented by the following scheme as adapted for sodium hydrogen sulphite : CHR,R,*CR,R,-SO,Xa A CHRIR,*CR,R,*O*SO,Nat. CRIR,:CR,R + NaHSO / Hydrogen sulphites must now therefore be omitted from the list of reagents of Class 11. E x P E R I ni E N T A L. Throughout this section the term “ molar ” and the symbol M applied to a hydrogen sulphite solution denote a solution containing one gram-molecule per litre. Inorganic sulphite in solutions of the reaction products was assumed to be absent when mineral acids in the cold caused no evolution of sulphur dioxide and when iodine was not decolorised.General Method of !Treatment of Insoluble Fluid Compounds with ,4mmonium Bisulphite Xolzction .-Usually the fluid unsaturated compound was shaken vigorously at intervals during several days with M/4-ammonium hydrogen sulphite and kieselguhr. The kicdguhr was then separated by filtration wished with boiling water and the united filtrates were heated with excess of barium 1 1 droxide until ammonia ceased to be evolved. The whole was then neutralised with dilute sulphuric acid the precipitated barium sulphite and sulphate were filtered off and washed with boiling water, the filtrates being finally evaporated to dryness on the steam-bath, Xtudies of the Influence of Conditions on Speed of Addition oj’ Ammonium Bisulphite to cycloHexene .-Owing to adsorption of solutes by the kieselguhr used as emulsifying agent i t was found necessary in comparative experiments to convert the addition product into barium salt and then to isolate and weigh this.I n one series of four experiments the following results were obtained using 5 C . C . of cgclohexene in mch instance with N j 4 -ammonium hydrogen sulphite solution. Volume of f1~/4-NM,HSO3 solution used ............ 240 C . C . 480 C . C . Excess of NH,HSO more than theory ............ 25yA 1500,b Yield (yo of theory) without kieselguhr ............ 16.8 40.11 Yield (yo of theory) with kieselguhr ............... 19.3 5 7 .o I n another series of three parallel experiments using the same quantity namely 5 c.c of cyclohexene with the same weight of ammonium bisulphite in each experiment but dissolved in different weights of water all with shaking a t intervals during 10 days the following results were obtained.In this series no kieselguhr was 11sed 312 KOLKER AND LAPWORTH THE DIRECT COMBINATION OF ............ Strength of bisulphite solution 2M M 12 C.C. of bisulphite solution .................. 60 120 240 Weight of crude barium salt obtained ... 1-29 1-51 2.26 Yield of crude product yo of theory ... 10.3 12.1 18.0 It is evident from the f i s t series that kieselguhr has a very f avourable influence on the yield especially when a considerable excess of hydrogen sulphite is used. The two series taken in conjunction show that the yield improves with the dilution of the hydrogen sulphite at least up to M / 4 ; the volumes of fluid to be evaporated however set a limit to the practical application of this fact.Examination of Products from cyclol3exene.-( a) With sodium hydrogen sulphite. When cyclohexene (30 c.c.) was shaken at intervals with N/2-hydrogen sulphite solution (1800 c.c.) for 8 days the resulting aqueous solution separated from unchanged cyclohexene (found 20 c.c.) evaporated to dryness and extracted with 96-98% spirit a white powder (1.9 g.) was obtained (Found Na = 12-3. C,H,,*SO,Na requires Na = 12.3y0). This powder gave no sulphuric acid when boiled with excess of dilute hydrochloric acid, and was free from inorganic sulphite; on hydrolysis with boiling sodium hydroxide however it yielded a little inorganic sulphite, estimated from titrations with iodine to represent 6-7.5y0 of sodium cyclohexyl sulphite.cycloHexene (36 c.c.), M/4-ammonium bisulphite solution (3360 c x.) and kieselguhr (120 g.) were shaken together at intervals during 10 days. On working up the product as on p. 31 1 a brown solid was obtained which on recrystallisation yielded white hexagonal plates of barium cyclo-hexanesulphonate identical with that obtained by the method of Borsche and Lange from cyclohexyl chloride through the sulphinic acid (Bey. 1905 38 2766) [Found H20 = 13.5; Ba (in dried salt) = 29.6. (C6Hl1*SO,),Ba,4H2O requires H,O = 13.4; Ba (in anhydrous salt) = 29.7y0]. The identity was established by direct comparison of the barium salts and of the sulphanilides (m.p. 85").* * The following new derivatives of cyclohexanesulphonic acid were pre-pared and examined during the course of the work. Sodium salt prisms, readily soluble in water and in dilute alcohol (Found Na = 11.1. C 6H,,.SOsNa,H20 requires N a = 1 1.3y0). Ammonium salt hygroscopic and exceedingly soluble in water ; very readily soluble in 98% spirit crystallising therefrom in granules. Magnesium salt rhombic plates by spontaneous evaporation of an aqueous solution. Copper salt small light green rhombic plates moderately soluble in water [Found : H,O = 15.4. (C6Hll.SOs)2Cu,4H,0 requires H20 = 15.6%]. SuZphonyZ chloride analysed by Borsche and Lange and described by them as an oil, separates from ether in rhombic plates m.p. 106'. Sulphonamide may be crystallised from water and melts at 93-94O. (b). With ammonium hydrogen sulphite. Very readily soluble in water ETHYLENIC HYDROCARBONS WITH I-IYDROGEN SULPHITES. 313 The nature of the crude brown solid product was investigated. When boiled for 4 hours with a large excess of 12% sodium hydroxide it yielded sulphite corresponding with 2% of alkyl sul-phite. When boiled with 10-150/ hydrochloric acid it did not >ield any sulphate but some sulphur dioxide was detected. When 3-5 g. were boiled for 6 hours with 15% sulphuric acid (50 c.c.), some siilphur dioxide was evolved the solution turned light yellon-in colour and a slight aromatic odour was observed The product of this hydrolysis could not be isolated but after removal of con-stituents soluble in ether and neutralisation of the solution with barium carbonate evaporation yielded 3.1 g.of pure bariuni r,!/~.lohexanesulphonate. It woulcl thus seem that the material contains a t least 90 % of ammonium cgclohexanesulphonate together with a small quantity of an organic compound which yields sul-pliurous acid or sulphite on hydrolysis.* This compound could not be isolated but it was observed that the mother-liquor remaining after separation of cyckohesanesulphonate from the crude product a t first gave with ferric chloride a deep red coloration similar to that obtained with sulphites and sulphinates; later this test gave a negative result so that the compound causing this coloration was evidently unstable either to water or to air or to both.Ethylene and Ammonium Hydrogen Xulphite .-Ethylene wab hrought into contact with M/4-ammonium sulphite ( 2 litres) in a large bottle a t about atmospheric pressure with occasional shaking. At first the solution absorbed nearly its own bulk of the gas during each interval of 24 hours when the residual gas with any gaseous impurities in it was allowed to escape and was then replaced by fresh ethylene. After this process had been repeated daily during a fortnight about 5 litres of ethylene had been absorbed and the original speed of absorption reduced by about four-fifths. The liquid was then worked up in the usual way and 13 g. of crude barium salts were obtained. * It is difficult to reconcile the properties of this secondary constituent n-ith tho assumption that it is sodium cyclohexyl sulphitc as it has resisted thi operations used in preparing the crude product.For comparison, sodium cyclohexyl sulphite was made by passing sulphur dioxide into (a) A solution of sodium in cyclohexanol and precipitating with alcohol ( b ) zt. suspension of sodium cyclohoxyl oxide in benzene and then draining thG solid on porous earthenware. The solid obtained by either process was, like other salts of alkyl sulphurous acids previously described extremely unstable losing sulphur dioxide on exposure to air and being immediately ligdrolysed in aqueous solution with liberation of inorganic sulphite (Found in sodium cyclohexyl sulphite made by process ( b ) Na = 13.4. C,H,,~SO,Na rc-quires Na = 12.4%. 0-346 G.dissolved in water required 39-5 C.C. of N/lO-I, while one molecule of sulphite from C,H,,-SO,Na requires 37.2 c.c.). ( For coniments on these points compare introductory section.) ix 314 THE DIRECT COMBINATION OF ETHYLENIC HYDROCARBONS ETC. The salts consist mainly of barium ethanesulphonate but also, as in the case of cyclohexene of small quantities of other salts, which give a red coloration with ferric chloride and are somewhat stable towards alkalis but unstable towards boiling mineral acids, which cause evolution of sulphur dioxide but no liberation of sulphuric acid [Found in barium salt H20 = 9.2; Ba (in anhydr-ous salt) = 38.6. (C2H5-S03)2Ba,2H,0 requires H20 = 9.2; (C2H5*S03)2Ba requires Ba = 38.7y0]. water, and easily recrystallised from ether forming prisms m.p. 59-GO” (James J . p r . Chem. 1882 [ii] 26 384 gives the melting point of ethanesulphonamide as 56”). Commercial “Amylene ” and Ammonium Hydrogen Sulphitc-Combination was here so rapid that in 7 days with occasional shaking 5.3 C.C. of “ amylene ” were almost completely absorbed by M/4-ammonium hydrogen sulphite solution without kieselguhr. On working up in the usual way 9.4 g. of crude soluble barium salt (86% of theory) were obtained [Found Ba = 31.5. (C,H,,~SO,) Ba requires Ba = 31-3y0]. The crude salt did not decolorise permanganate nor absorb bromine it gave no sulphur dioxide or sulphuric acid when boiled for several hours with 15% hydrochloric acid and 95% of the original product was recoverable. “ Amylene ” was exceptional among the hydrocarbons examined in yielding nothing but true sulphonic derivatives.As the ‘‘ amylene ’’ used was the mixture of several isomerides obtained from fuse1 oil the products were not further examined. Dipentene andAmmonium Hydrogen Su1phite.-Dipentene (15 c.c.), ammonium hydrogen sulphite (M/4 ; 2000 c .c .) and kieselguhr (50 g.) shaken at frequent intervals during 14 days gave 12.1 g. of barium salts at least 90% of which consist of barium menthne-disulphonate [Found H,O (lost at 160O) = 4.0. requires H20 = 4.0%. In the anhydrous salt ; found Ba = 31.4, S = 15.1; theory requires Ba = 31.6 S = 14-7y0]. This salt is readily soluble in water forms clusters of needles, and is stable to boiling mineral acid. The mother-liquors obtained on recrystallising the salt from water decolorise a little bromine or permanganate and slowly evolve some sulphur dioxide when boiled with dilute sulphuric acid.Pinene and Ammonium Hydrogen Su1phite.-From pinene (32 c.c.), ammonium hydrogen sulphite (M/4 ; 2000 c.c.) and kieselguhr (70 g.) after 12 days 7.4 g. of mixed soluble barium salts were isolated but retained an odour resembling pinene after repeated The sulphonamide made from this salt was soluble in C~~H18(S03)2Ba>H?, METHOD OF MEASURING THE DIELECTRIC CONSTANTS OF LIQUIDS. 315 extraction with absolute alcohol. The crude product differed appreciably from that obtained from any other hydrocarbon inas-much as the proportion of hydrolysable salt was nearly 45% of the whole. When 2 g. were boiled with 15% sulphuric acid much sulphur dioxide was evolved and appreciable quantities of an oil having a terpene-like odour appeared.After removing the oil with ether and the sulphuric acid with barium carbonate only 1.1 g. of stable barium salt were recovered. The crude product, which was free from inorganic sulphate also discharged the colour of an unusually large proportion of bromine. ORUANIC CHEXICAL LABORATORIES, THE UNIVERSITY MANCRESTER. [Received November 7 t h 1924.1 ADDENDuM.-Throughout this paper the terms (‘ hydrogen sulphite ” and (( bisulphite ” are used interchangeably and without regard to the question of the constitution of salts t o which these names have commonly been applied in the past THE DIRECT COMBINATION OF ETHYLENIC HYDROCARBONS ETC. 307 XLVII1.-The Direct Combination of Ethylenic Hydro-carbons with Hydrogen Xzclphites.By ISRAEL KOLKER and ARTHEX LAPWORTH. AGENTS such as ozone which attack the simplest compounds, have been distinguished (as " Class I ") by Lapworth and MeRae (J. 1922 121 2741 ; compare also Lapworth &fern. fManchester Phil. Xoc. 1920 64 iii 11) from others (" Class I1 ") which M 308 KOLKER AND LAPWORTH THE DIRECT COMBINATION OF are inert towards ethylenic hydrocarbons but which nevertheless are additive to the ethylenic group when the latter occurs in the a-position with respect to a carbonyl cyano or similar group. It is significant that all the known agents of the second class are either (IIa) metallo-compounds or (IIb) compounds of the general type H-Z in which the hydrogen atom is capable of being replaced by the action of metals.Examples of sub-class IIa are Grignard reagents potassium cyanide and sodio-derivatives of ketones esters and nitriles . Examples of IIb are HCN HCH(CO,Et), HNR,R, the last includ-ing amines hydroxylamines and hydrazines . The electronic theory of metallic compounds suggests that the radical of each agent of class IIa is capable of existence in all cases as a negatively charged ion Z-. In some cases the free ion Z- is identical in consti-tution with the radical as it occurs in the compound H-Z (example H-NH,) ; in other cases it is probably not ; (example : H-CH,.NO,) but the structural difference is such that intracon-version readily takes place in either direction. As is well known, agents of sub-class IIb also form addition products with many saturated carbonyl compounds.Not all compounds which have the characters defined in the preceding paragraph can be referred to Class IIb ; thus powerful acids are by definition excluded from this class as they attack ethylenic compounds of all types. It is further necessary to observe that in referring any agent to Class IIb there are restrictions as to the experimental conditions prevailing while the agent is applied. For example hydrogen cyanide may properly be referred to Class IIb when applied in presence of an alkaline catalyst but not in presence of an acid catalyst unless it has been shown that under the same conditions the combination is inert or nearly so toward ethylenic hydrocarbons. Within the range occupied by IIb which must be limited a t one extreme by acids too weak to attack ethylenic compounds at measurable speed and at the other extreme by very feebly ionisable compounds such as ammonia there does not at present appear to be any direct relation between the additive efficiency of the agent and the electro-affinity of the anion.Thus many very weak acids appear almost or quite unable to form stable addition products with carbonyl compounds. With reference to Class I1 as a whole it is probably true that no compounds of which the ion Z- has a very high affinity for the charge can be included in either sub-class (compare Lapworth, Zoc. cit.). There are for example no cases recorded where the potassium or sodium salts of powerful acids form additive corn ETHYLENIC HYDROCARBONS WITH HYDROGEN SULPHITES.309 pounds with carbonyl compounds. The powerful acids have already been excluded for the reasons above stated. From the preceding considerations it is clear that any agent which forms addition products with the carbonyl group of aldehydes and ketones or with the ethylenic linking in ap-unsaturated carbonyl compounds may safely be referred to sub-group IIb only when it is known that the agent does not attack ethylenic hydrocarbons under comparable conditions; in many cases as in the instance of ammonia experimental evidence is already extensive and <o uniformly negative that this indifference may confidently he inferred whilst in other cases further investigation is required. The correct classification of agents additive to ethylenic com-pounds is of the utmost importance in studying the influence of atoms and groups on the properties of others in the same molecule.The authors have made a careful study of two series of agents previously referred in the papers above specified (Zoc. cit.) to Class 11. The experiments carried out by the authors and by other workers in these laboratories on the possible addition of hydrogen cyanide and metallic cyanide to ethylenic hydrocarbons including cycio-hexene and styrene have given uniformly negative results. As the addition products in these cases would have been nitriles easily convertible into carboxylic acids and so capable of detection even in traces it may be concluded that metallic cyanides and hydrogen cyanide in absence of acid catalysts are indeed highly selective ant1 that when they do attack an ethylenic linking the latter is almost certainly affected by conditions similar to those which ohtr~in iri @-unsaturated ketones.The other series of reagents tested were sulphites and iuore especially hydrogen sulphites. The sole instance hitherto recorded (so far as we have been able to discover) of a hydrocarbon combining directly with hydrogen sulphites is that of styrene and even in that, case the published evidence was inconclusive (Miller dnticrlctr, 1877 189 340; LabbB Bull. SOC. chim. 1893 [iii] 22 1077; Dupont and Labaune Sci. Ind. BzZZ. 1912 [iii] 7 3). In thr: case of styrene moreover there was an element of doubt whether phenyl can exercise the same influence on a double bond as carbonyl can.Similar uncertainty existed in the theoretical interpretation of the observations of Dupont and Labaune (loc. cit.) nho obtained additive products of hydrogen sulphites with unsaturatctl alcohols. The results obtained by the present authors remove all doubt on many of these moot points. Hydrogen sulphites combine directly with ethylenic hydrocarbons and the failure of previous investi 310 KOLKER AND LAPWORTH THE DIRECT COMBINATION OF gators fully to establish this fact is attributable to the unsuitable experimental conditions used. Special precautions to ensure intimate contact between the hydro-carbons and the aqueous solutions are usually but not invariably necessary. They may consist in emulsifying with the aid of purified kieselguhr.Contrary to expectations dilution of the sulphite solution possibly by an effect on the solubility of the hydrocarbons, favoured the interaction. A not less important practical feature in the later experiments was the use of ammonium hydrogen sulphite instead of the sodium or potassium salt. This device admits of the ready removal of any excess of the reagent by boiling with sufficient barium hydroxide, when insoluble barium sulphite is formed ammonia is expelled, and such organic addition products as are not destroyed during this process are easily isolated as soluble barium salts. Pinene dipentene cyclohexene crude " amylene " and even the completely symmetrical ethylene combine readily under favourable conditions with hydrogen sulphites usually in the cold yielding products which for the most part appear to be the salts of the expected saturated sulphonic acids ; but these are usually mixed with a variable proportion of other salts apparently isomeric with the sulphonates.These secondary products unlike the true sul-phonic acids are hydrolysed by dilute acids or alkalis and are probably salts of the alkyl hydrogen sulphites though some slight doubt attaches to this view of their constitution. These salts appear to be considerably more stable than the salts of alkyl hydrogen sulphites prepared by union of sulphur dioxide with sodium alkyl-oxides but this may be due to stabilisation by the other salts present ; indeed the authors have observed (1) that the hydrolysis of the salts of methylethionic acid SO,H*CH(CH,)*CH,*SO,H (which normally is readily effected by dilute acids) takes place much more slowly if excess of salts of methylisethionic acid, SO,H *CH (CH,) *CH,* OH , are present and (2) that pure barium methylethionate decomposes a t loo" but in presence of 9-10 times its weight of isethionate does not decompose appreciably below 150".It is just possible however, that two series of alkyl sulphurous acids exist. It seems probable that carbonyl compounds react with bisulphite, as with metallic cyanide by first capturing the anion. Ethylenic hydrocarbons evince no definite preference for anions and may react with bisulphites either by first capturing the hydrogen ion or by uniting with the unsaturated centres of the sulphite molecules much as they unite with ozone and possibly in both ways.The gross result of the process of addition of hydrogen sulphite ETHYLENE HYDROCARBOKS WITH HYDROGEN SULPHITES. 31 1 to ethylenic hydrocarbons may thus be represented by the following scheme as adapted for sodium hydrogen sulphite : CHR,R,*CR,R,-SO,Xa A CHRIR,*CR,R,*O*SO,Nat. CRIR,:CR,R + NaHSO / Hydrogen sulphites must now therefore be omitted from the list of reagents of Class 11. E x P E R I ni E N T A L. Throughout this section the term “ molar ” and the symbol M applied to a hydrogen sulphite solution denote a solution containing one gram-molecule per litre. Inorganic sulphite in solutions of the reaction products was assumed to be absent when mineral acids in the cold caused no evolution of sulphur dioxide and when iodine was not decolorised.General Method of !Treatment of Insoluble Fluid Compounds with ,4mmonium Bisulphite Xolzction .-Usually the fluid unsaturated compound was shaken vigorously at intervals during several days with M/4-ammonium hydrogen sulphite and kieselguhr. The kicdguhr was then separated by filtration wished with boiling water and the united filtrates were heated with excess of barium 1 1 droxide until ammonia ceased to be evolved. The whole was then neutralised with dilute sulphuric acid the precipitated barium sulphite and sulphate were filtered off and washed with boiling water, the filtrates being finally evaporated to dryness on the steam-bath, Xtudies of the Influence of Conditions on Speed of Addition oj’ Ammonium Bisulphite to cycloHexene .-Owing to adsorption of solutes by the kieselguhr used as emulsifying agent i t was found necessary in comparative experiments to convert the addition product into barium salt and then to isolate and weigh this.I n one series of four experiments the following results were obtained using 5 C . C . of cgclohexene in mch instance with N j 4 -ammonium hydrogen sulphite solution. Volume of f1~/4-NM,HSO3 solution used ............ 240 C . C . 480 C . C . Excess of NH,HSO more than theory ............ 25yA 1500,b Yield (yo of theory) without kieselguhr ............ 16.8 40.11 Yield (yo of theory) with kieselguhr ............... 19.3 5 7 .o I n another series of three parallel experiments using the same quantity namely 5 c.c of cyclohexene with the same weight of ammonium bisulphite in each experiment but dissolved in different weights of water all with shaking a t intervals during 10 days the following results were obtained.In this series no kieselguhr was 11sed 312 KOLKER AND LAPWORTH THE DIRECT COMBINATION OF ............ Strength of bisulphite solution 2M M 12 C.C. of bisulphite solution .................. 60 120 240 Weight of crude barium salt obtained ... 1-29 1-51 2.26 Yield of crude product yo of theory ... 10.3 12.1 18.0 It is evident from the f i s t series that kieselguhr has a very f avourable influence on the yield especially when a considerable excess of hydrogen sulphite is used. The two series taken in conjunction show that the yield improves with the dilution of the hydrogen sulphite at least up to M / 4 ; the volumes of fluid to be evaporated however set a limit to the practical application of this fact.Examination of Products from cyclol3exene.-( a) With sodium hydrogen sulphite. When cyclohexene (30 c.c.) was shaken at intervals with N/2-hydrogen sulphite solution (1800 c.c.) for 8 days the resulting aqueous solution separated from unchanged cyclohexene (found 20 c.c.) evaporated to dryness and extracted with 96-98% spirit a white powder (1.9 g.) was obtained (Found Na = 12-3. C,H,,*SO,Na requires Na = 12.3y0). This powder gave no sulphuric acid when boiled with excess of dilute hydrochloric acid, and was free from inorganic sulphite; on hydrolysis with boiling sodium hydroxide however it yielded a little inorganic sulphite, estimated from titrations with iodine to represent 6-7.5y0 of sodium cyclohexyl sulphite.cycloHexene (36 c.c.), M/4-ammonium bisulphite solution (3360 c x.) and kieselguhr (120 g.) were shaken together at intervals during 10 days. On working up the product as on p. 31 1 a brown solid was obtained which on recrystallisation yielded white hexagonal plates of barium cyclo-hexanesulphonate identical with that obtained by the method of Borsche and Lange from cyclohexyl chloride through the sulphinic acid (Bey. 1905 38 2766) [Found H20 = 13.5; Ba (in dried salt) = 29.6. (C6Hl1*SO,),Ba,4H2O requires H,O = 13.4; Ba (in anhydrous salt) = 29.7y0]. The identity was established by direct comparison of the barium salts and of the sulphanilides (m.p. 85").* * The following new derivatives of cyclohexanesulphonic acid were pre-pared and examined during the course of the work. Sodium salt prisms, readily soluble in water and in dilute alcohol (Found Na = 11.1. C 6H,,.SOsNa,H20 requires N a = 1 1.3y0). Ammonium salt hygroscopic and exceedingly soluble in water ; very readily soluble in 98% spirit crystallising therefrom in granules. Magnesium salt rhombic plates by spontaneous evaporation of an aqueous solution. Copper salt small light green rhombic plates moderately soluble in water [Found : H,O = 15.4. (C6Hll.SOs)2Cu,4H,0 requires H20 = 15.6%]. SuZphonyZ chloride analysed by Borsche and Lange and described by them as an oil, separates from ether in rhombic plates m. p. 106'. Sulphonamide may be crystallised from water and melts at 93-94O.(b). With ammonium hydrogen sulphite. Very readily soluble in water ETHYLENIC HYDROCARBONS WITH I-IYDROGEN SULPHITES. 313 The nature of the crude brown solid product was investigated. When boiled for 4 hours with a large excess of 12% sodium hydroxide it yielded sulphite corresponding with 2% of alkyl sul-phite. When boiled with 10-150/ hydrochloric acid it did not >ield any sulphate but some sulphur dioxide was detected. When 3-5 g. were boiled for 6 hours with 15% sulphuric acid (50 c.c.), some siilphur dioxide was evolved the solution turned light yellon-in colour and a slight aromatic odour was observed The product of this hydrolysis could not be isolated but after removal of con-stituents soluble in ether and neutralisation of the solution with barium carbonate evaporation yielded 3.1 g.of pure bariuni r,!/~.lohexanesulphonate. It woulcl thus seem that the material contains a t least 90 % of ammonium cgclohexanesulphonate together with a small quantity of an organic compound which yields sul-pliurous acid or sulphite on hydrolysis.* This compound could not be isolated but it was observed that the mother-liquor remaining after separation of cyckohesanesulphonate from the crude product a t first gave with ferric chloride a deep red coloration similar to that obtained with sulphites and sulphinates; later this test gave a negative result so that the compound causing this coloration was evidently unstable either to water or to air or to both. Ethylene and Ammonium Hydrogen Xulphite .-Ethylene wab hrought into contact with M/4-ammonium sulphite ( 2 litres) in a large bottle a t about atmospheric pressure with occasional shaking.At first the solution absorbed nearly its own bulk of the gas during each interval of 24 hours when the residual gas with any gaseous impurities in it was allowed to escape and was then replaced by fresh ethylene. After this process had been repeated daily during a fortnight about 5 litres of ethylene had been absorbed and the original speed of absorption reduced by about four-fifths. The liquid was then worked up in the usual way and 13 g. of crude barium salts were obtained. * It is difficult to reconcile the properties of this secondary constituent n-ith tho assumption that it is sodium cyclohexyl sulphitc as it has resisted thi operations used in preparing the crude product.For comparison, sodium cyclohexyl sulphite was made by passing sulphur dioxide into (a) A solution of sodium in cyclohexanol and precipitating with alcohol ( b ) zt. suspension of sodium cyclohoxyl oxide in benzene and then draining thG solid on porous earthenware. The solid obtained by either process was, like other salts of alkyl sulphurous acids previously described extremely unstable losing sulphur dioxide on exposure to air and being immediately ligdrolysed in aqueous solution with liberation of inorganic sulphite (Found in sodium cyclohexyl sulphite made by process ( b ) Na = 13.4. C,H,,~SO,Na rc-quires Na = 12.4%. 0-346 G. dissolved in water required 39-5 C.C. of N/lO-I, while one molecule of sulphite from C,H,,-SO,Na requires 37.2 c.c.).( For coniments on these points compare introductory section.) ix 314 THE DIRECT COMBINATION OF ETHYLENIC HYDROCARBONS ETC. The salts consist mainly of barium ethanesulphonate but also, as in the case of cyclohexene of small quantities of other salts, which give a red coloration with ferric chloride and are somewhat stable towards alkalis but unstable towards boiling mineral acids, which cause evolution of sulphur dioxide but no liberation of sulphuric acid [Found in barium salt H20 = 9.2; Ba (in anhydr-ous salt) = 38.6. (C2H5-S03)2Ba,2H,0 requires H20 = 9.2; (C2H5*S03)2Ba requires Ba = 38.7y0]. water, and easily recrystallised from ether forming prisms m. p. 59-GO” (James J . p r . Chem.1882 [ii] 26 384 gives the melting point of ethanesulphonamide as 56”). Commercial “Amylene ” and Ammonium Hydrogen Sulphitc-Combination was here so rapid that in 7 days with occasional shaking 5.3 C.C. of “ amylene ” were almost completely absorbed by M/4-ammonium hydrogen sulphite solution without kieselguhr. On working up in the usual way 9.4 g. of crude soluble barium salt (86% of theory) were obtained [Found Ba = 31.5. (C,H,,~SO,) Ba requires Ba = 31-3y0]. The crude salt did not decolorise permanganate nor absorb bromine it gave no sulphur dioxide or sulphuric acid when boiled for several hours with 15% hydrochloric acid and 95% of the original product was recoverable. “ Amylene ” was exceptional among the hydrocarbons examined in yielding nothing but true sulphonic derivatives.As the ‘‘ amylene ’’ used was the mixture of several isomerides obtained from fuse1 oil the products were not further examined. Dipentene andAmmonium Hydrogen Su1phite.-Dipentene (15 c.c.), ammonium hydrogen sulphite (M/4 ; 2000 c .c .) and kieselguhr (50 g.) shaken at frequent intervals during 14 days gave 12.1 g. of barium salts at least 90% of which consist of barium menthne-disulphonate [Found H,O (lost at 160O) = 4.0. requires H20 = 4.0%. In the anhydrous salt ; found Ba = 31.4, S = 15.1; theory requires Ba = 31.6 S = 14-7y0]. This salt is readily soluble in water forms clusters of needles, and is stable to boiling mineral acid. The mother-liquors obtained on recrystallising the salt from water decolorise a little bromine or permanganate and slowly evolve some sulphur dioxide when boiled with dilute sulphuric acid. Pinene and Ammonium Hydrogen Su1phite.-From pinene (32 c.c.), ammonium hydrogen sulphite (M/4 ; 2000 c.c.) and kieselguhr (70 g.) after 12 days 7.4 g. of mixed soluble barium salts were isolated but retained an odour resembling pinene after repeated The sulphonamide made from this salt was soluble in C~~H18(S03)2Ba>H?, METHOD OF MEASURING THE DIELECTRIC CONSTANTS OF LIQUIDS. 315 extraction with absolute alcohol. The crude product differed appreciably from that obtained from any other hydrocarbon inas-much as the proportion of hydrolysable salt was nearly 45% of the whole. When 2 g. were boiled with 15% sulphuric acid much sulphur dioxide was evolved and appreciable quantities of an oil having a terpene-like odour appeared. After removing the oil with ether and the sulphuric acid with barium carbonate only 1.1 g. of stable barium salt were recovered. The crude product, which was free from inorganic sulphate also discharged the colour of an unusually large proportion of bromine. ORUANIC CHEXICAL LABORATORIES, THE UNIVERSITY MANCRESTER. [Received November 7 t h 1924.1 ADDENDuM.-Throughout this paper the terms (‘ hydrogen sulphite ” and (( bisulphite ” are used interchangeably and without regard to the question of the constitution of salts t o which these names have commonly been applied in the past