SZARVASY : ELEGTROLYTIC PREPARATION OF INDULINE DYES. 207 XVII1.-Electrolytic Preparation of Induline Dyes. By EMERIQUE CHARLES SZARVASY, Ph.D. IT has long been known that different colouring matters are formed when the electric current is allowed to pass through solutions of aniline salts. The first experiments bearing on this subject were made by Letheby in 1862 (this Journ., 15, 161), who electrolysed a solution of aniline sulphate between platinum electrodes, the cathode being immersed in dilute sulphuric acid in a porous cell, and observed the formation of a bluish-green dye on the anode. aoppelsroder (Farbenelektrochemishe Mittheilungen, Muhlhausen) electrolysed neutral or acid solutions of aniline salts; the dyes formed varied according to the salts, solvents, electrodes, and current densities employed.On electrolysing a solution of aniline hydrochloride for several hours, he obtained aniline-black mixed with small quantities of aniline-violet and ‘6 anilein.” If the aniline con- tained tolnidine, then mauvaniline, rosaniline, and leucaniline were also formed, Besides these colouring matters, he prepared a large number, the nature of which was not determined. Voigt (Zed. angew. Chena., 1894, 107), on electrolysing concentrated solutions of aniline sulphate, obtained rosaniline, saffranine, chrys- aniline, and paraleucaniline. Complex mixtures of colouring matters are evidently obtained by these methods, but no explanation has been given of the reactions which lead to their formation. In all probability, the production of dyes from aniline in this way is traceable primarily to the formation of azo-compounds by the action of electrolytic oxygen which may result either in the dehydrogenation of the amino-group and sub- sequent condensation, or in oxidation of the base to nitrobenzene and interaction of this with the excess of aniline. This view is supported by Rotondi’s experiments (Jahres6er., 1884, 270), as azo-, diazo-, amidoazo-, and diazo-amido-compounds were found among the pro- ducts of the electrolysis of solutions of aniline.These secondary effects necessarily complicate the reactions, but by far the greatest influence on the nature and quantity of the final products is exerted by the solvent, which, being ionisable, takes part in the electrolysis introducing new factors, which cause further secon- dary reactions.On studying the electrolytic preparation of colonring matters, it occurred to me that the process could be essentially simplified by electrolysing the aniline salts in a fused state; the secondary reactions, due to the ionisation of the solvent, being eliminated, there208 SZARVASY : ELECTROLYTIC PREPARATION OF INDULINE DYES. was good reagon to hope for end products of a more homogeneous character. The present paper gives an account of the principal results I have obtained by electrolysing fused mixtures of aniline and aniline hydro- chloride, this salt being chosen as it has the advantage of melting comparatively easily and without decomposition. The preliminary experiments were made in glass tubes bent at right angles, the carbon electrodes being adjusted in the limbs of the tube, which was filled with aniline hydrochloride and kept a t tho desired temperature.When a current of 0.5-1 ampere was passed through the fused salt the mass soon turned blue. The colouring matter was formed on the anode, and could be better observed, when F I of suitable dimensions was connected loose asbestos, was packed in the bend of the tube forming a diaphragm. I n this case, the dye was only formed in the limb contain- ing the anode, that contain- ing the cathode remaining colourless. When the direc- tion of the current was re- versed, the other side be- came coloured, and a t the same time the blue colour on the cathode was slowly destroyed. I n the experiments on a larger scale, the substance was placed in a graphite crucible, A, which acted as the positive electrode.An- other graphite crucible, B, with the negative pole by a copper wire, fastened to the inner wall of the crucible and passing through the glass tube, C, which served as a rotating axle. The upper part of B was closed with plaster of paris, B, to fix the axle and to protect the metallic connections. The contact between i he two wire ends in P was effected by means of a mercury connection. With the aid of a motor and the wheel, E, the inner crucible was slowly turned round and the melted mass kept in motion during the electrolysis, so that fresh portions of the mixture came into contact with the electrode. The large crucible was heated with a Bungen burner, which was regulated t o keep the fused mass a t the desired temperature,SZARVASY : ELECTROLYTIC PREPARATION OF INDULINE DYES.209 The active surface of the anode is about twice as large as that of the cathode, so that a higher current density on the cathode is attained and also the reducing action of the latter, which naturally tends to diminish the yield, is restrained, The main advantage of this apparatus is that comparatively large quantities of the electrolyte form a thin layer between the two electrodes, and thus the resistance is very low. Graphite being a good conductor of heat, the temperature of the fused mass is uniform, this being an important condition in these reactions. I may mention that this apparatus is very suitable for electrolysing substances of high melting points, the only alteration to be made in this case being that the glass tube forming the axle is replaced by one of porcelain. After the electrolysis was completed, the product was poured into a flask, diluted with water, the excess of aniline hydrochloride decom- posed with sodium carbonate and the aniline removed by distillation with steam.The coloured substance remaining in the distillation flask was purified by subsequent washing with water and ether, and then extracted with methyl alcohol, in which it was very soluble. After extraction, a black powder remained, which consisted chiefly of aniline black, mixed with small quantities of graphite from the electrodes. The quantity of aniline black varied according to the conditions under which the electrolysis was carried out, as high current density and high temperature tend to increase the amount, whereas, under other conditions, very little is formed.This method, which merely separates the indulines from aniline black, was employed in determining the quantities of the products formed and ascertaining the most favourable conditions under which the current efficiency is the highest. The results obtained are arranged in the table on p. 210. I n columns D, and Do respectively, the current densities on the anode and cathode are given, the numbers being calculated in amperes per square decimetre of electrode surface. Experiments 1-7 were made to determine the influence of current density, 8-11 that of the duration of the action, a,nd 12-15 that of temperature. It will be seen that when z), is approximately 0.8, the efficiency of the current is at its highest, but at the same time the temperature must be maintained a t about 160'.The experiments also show that as the electrolytic action is prolonged, the product per unit of electric energy becomes less, so that only 20 per cent. of the original material can be profitably transformed into the dye, but aa the aniline is recovered almost quantitatively, when separating the products, the yield amounts to 86-90 per cent, of that actually used. The colouring matter extracted with methyl alcohol consists of o210 SZARVASY : ELECTROLYTIC PREPARATION OF INDULINE DYBSa ~- 2'9- 3'2 3 - 3'2 3 - 3.5 3.2- 5 4.5- 5 5.4- 7 LO -12 3 - 3-5 3 - 8 ' 5 3 - 3.5 3 - 3'5 5 - 7 4 - 5 2 - 3 2 - 3 Series.- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 16 0.57 0-7 0-8 1-03 1.12 2'2 3-2 0.8 0 '8 0.8 0.8 0.8 0.8 0.8 0 '8 1 '2 1'48 1.7 2-2 2 5 4.5 5.5 1.7 1.7 1.7 1.7 1 *7 1.7 1.7 1 *7 Duration of sxperiments in hours. 3 3 3 3 3 3 3 2 5 8 16 3 3 3 3 Cemperature, 160" 160 160 160 160 160 160 160 160 160 160 120 150 180 210 Product in grams per 1 amp. hour. 1 *64 2'0 2'5 2 *1 1 *86 1.54 1 *45 2'1 1.9 1 *3 0.61 1.8 2'3 1.67 1.1 mixture of hydrochlorides of bases of the induline class, and in acid solution dyes silk or wool, but has no affinity for cotton. It may be remarked that this mode of preparation resembles the well known " induline process," inasmuch as a mixture of indulines is produced, the nature of which, as well as the proportions existing between the constituents obtained, depends largely upon the temperature at which the reaction takes place and the manner in which it is conducted.Separation of the Dyes. When electrolysing with low currenh density a t a comFaratively low temperature, several colouring matters are formed, which are soluble in water ; from these I succeeded in separating (' induline " and '' B,,4 anilinoinduline," first described by 0. Fischer and Hepp (Artmlen, 256, 262, 266, 272, 286). The colouring matter was dissolved in 50 per cent. acetic acid, and concentrated hydrochloric acid was added to the violet solution. After some time, a crystalline precipitate, A, separated, which was removed by filtration. Sodium chloride was then added to the filtrate, and a precipitate, B, obtained consisting of a mixture of several dyes which dissolved in hot water forming a dark blue solution. Ppecipitccte B.-The aqueous solution, on the addition of caustic soda, gave a precipitate which was soluble in benzene, and from the solution after concentration small crystals were obtained.After recry st allisiag several times from a mixture of light petroleum and benzene, the product wag obtained in reddisb-brown needles ; itSZARVASY : ELECTROLYTIC PREPARATION OF INDULINE DYES. 211 melted at about 206O, and dissolved in concentrated sulphuric acid with a violet colour. On analysis, it proved to be induline. Found C = 79.37 ; H = 4.21 ; N = 15-35 per cent. Cl,Hl,N, requires C = 79.70 ; H = 4-79 ; N = 15-49 per cent. Precipitate A.-This precipitate was examined in the same way, the crystalline powder which separated from the solution in benzene being recrystallised several times from this solvent.When pure, the product formed small prisms with a metallic lustre, and melted at about 150’. It dissolved in strong sulphuric acid with a dark blue colour, which, on the addition of water, became violet. On analysis it gave numbers agreeing with those required for “B2,4 anilino- induline.” Found C = ‘79.73 ; H = 5.30 ; N = 15.22 per cent, C,,H,,N, requires C = 79.55 ; H = 4.98 ; N = 15.47 per cent. When the conditions of experiment were varied by increasing the current density, and the temperature was maintained at 160-170’ during the electrolysis, it was found that only small quantities of soluble indulines were formed, and that the ‘( induline 6B,” described by Witt and Thomas (Trans., 1883, 43, 112), occurred among the products.This separated in a crystalline form when the fused mass was allowed to cool very slowly, and was freed from soluble indulines by filtration and subsequent washing with alcohol, in which it is almost insoluble. It was then treated with alcoholic caustic soda, and the resulting base, after washing with water and then with dilute alcohol, was purified by repeated recrystallisation from aniline, from which it separated in greenish, glistening crystals. On analysis, it gave numbers agreeing with those required for ‘‘ induline 6B.” C,,H,7N, requires C = S1.66 ; H = 5-1 1 ; N = 13.23 per cent. Found C = 81-78 ; H = 5-23 ; N = 13.10 per cent. Theory of the Process-Etectrolytic Pveparation of Axophenine. Induline dyes are formed by the action of azo- and amidoazo-com- pounds on the hydrochlorides of aromatic amines; as an intermediate product, azophenine is formed, which plays a most important part in the formation of indulines.The formation of this compound is in all probability due to the oxidising effects of the azo-compounds, which decompose into p-phenylenediamine and aniline with the elimination of ammonia. With these bases, the azophenine reacts yielding indulines. Only small quantities of the azo-compounds are found in the pro- ducts of electrolysis, as immediately after their formation interaction with the aniline hydrochloride takes place. However, it is certain212 SZARVASY : ELECTROLYTIC PREPARATION OF INDULINE DYE13. that azo-compounds are the primary products, and are present in the early stages of the reaction.Their formation may be explainec! as follows. By the electrolytic decomposition of the aniline hydrochloride, chlorine is liberated a t the anode, and reacting with the aminic hydrogen of the aniline effects dehydrogenation with the production of hydrogen chloride, hydrazobenzene, and azobenzene. So far as I am aware, only one instance of a similar reaction has yet been re- corded, namely, the formation of azobenzene by the action of bleaching powder on aniline dissolved in chloroform (Schmitt, J. pr. Chem., 1878, [ ii], 18, 196). As a rule, the halogens, when reacting with aniline, give rise to the formation of substitution products. Azophenine can be detected in the product of the electrolysis of a mixture of aniline hydrochloride and aniline, and may be obtained in considerable amount if the temperature is kept at 70-90° during the reaction.In one experiment, after the electrolysis had continued for 2-3 hours, the product, after treatment with sodium carbonate and re- moval of the excess of aniline, formed a brownish powder, which, after being washed with hot water and alcohol, was dissolved in aniline. From this solution, brownish-red crystals separated, which, after re- crystsllisation from benzene, melted a t 239O. On analysis, it gave numbers agreeing with those calculated for azophenine. Found C = 82.05 ; H = 5.67 ; N = 12.73 per cent. C,,H,,N, requires C = 81.36 ; H = 5.46 ; N = 13-18 per cent. The substance formed a violet solution in concentrated sul phuric acid, gave a violet mass when heated with p-phenylenediamine hydro- chloride, and, on reduction with tin and hydrochloric acid, formed a mixture of bases, among which aniline and p-phenylenediamine were detected. Briefly summarised, the results prove that (1) the chlorine produced by the electrolytic decomposition of aniline hydrochloride effects de- hydrogenation of aniline ; (2) under certain conditions, the azo-com- pounds thus formed interact with the aniline hydrochloride, forming (3) primarily azophenine, and (4) ultimately indulines. I propose to extend these investigations to the homologues of aniline and some other aromatic amines, and the results will, in due course, be communicated to the Society. This research has been oarried on in the Dnvy-Faraday Research Laboratory of the Royal Institution, and I take this opportunity of tendering my thanks to the Directors and to Dr, 4. $cott far the facilities they have given me. DAVY-FARADAY RESEARCH LABORATORY, ROYAL INSTITUTION.