THE ACTION OF WATER ON DIAZO-SALTS. 19 V.-The Action of Water on D,iazo-salts. By JOHN CANNELL CAIN and GEORGE MARSHALL NORBIAN. DURING an investigation carried out by one of us (J. C. C.) and Nicoll on the rate of decomposition of diazo-compounds, it was noticed (Trans., 1902, 81, 1440) that the products of decomposition of the tetrazo-salts prepared from oo-dichlorobenzidine and dianisidine were not the corresponding dihydroxy-compounds, as in the case of benzidine and tolidine, although the nitrogen mas evolved as usual. It was accordingly of much interest to investigate the nature of the sub- stances formed, and the result of this work was t o show that in each case the chief, if not the only product formed was of a quinonoid character (Trans., 1903, 83, 688). The very great difference in behaviour of these two tetrazo-salts as compared with those from benzidine and tolidine was apparently to be attributed to the presence of a chlorine and a methoxyl group respectively in the ortho-position to the diazonium group.With the object ol testing this view a search was made in the literature for any abnormal cases of decomposition of ortho-substituted c 220 CAIN AND NORMAN: diazo-salts belonging to the benzene series, and, curiously enough, all the instances found answered this description. It was therefore decided to make a thorough examination of these compounds with the object of discovering whether the alleged non-formation of phenols was to be explained by the presence of an ortho-substituent or otherwise. Methods of carrying ozct the Decomposition.What may be regarded as the normal method of decomposition of such diazo-salts is to diazotise in moderately strong mineral acid solu- tion (preferably hydrochloric or sulphuric acid) by adding a solution of sodium nitrite and then either heat to boiling the diazo-solution thus obtained or pass a current of steam through it. When, however, no phenol was found by this means, the method adopted by Heinichen (Artnalen, 1889, 253, 281) was tried. This consists in heating the strong diazo-solution with strong sulphuric acid, whereby the boiling point becomes raised to 150'. Heinichen, by this method, was success- ful in obtaining oo-dibromophenol from the corresponding diazo-salt, after the usual method had failed. Heinichen's method has been applied by various other workers in attacking similar problems, but often without success; thus Orton (23*oc., 1905, 21, 170) failed to obtain s-tribromophenol from the diazo-salt of tribromoaniline.This method of decomposition is, however, not a rational one owing to the fact that by using more concentrated sulphuric acid in order to reach a higher temperature a very considerable retarding influence is intro- duced. This retarding influence can be measured by determining the " coefficient of decomposition G " according to the method described by one of us and Nicoll (Zoc. cit., p. 1412). Thus, in the case of benzene- diazonium chloride the value of C is 0.0298 (Zoc. cit. p. 1420). A corresponding experiment with benzenediazonium sulphate (solution containing 1 per cent.H,SO,) gave nearly the same number, namely, 0.0302. When, however, the amount of sulphuric acid is increased until the solution contains about 35 per cent. of the acid, the value of C diminishes to 0.0197 (for details of this work, see Cain, Ber., 1905, 38, 2511). It follows, therefore, that an increase in the quantity of sulphuric acid produces an apparent increase in the stability of the diazo-compound, This is thought to be due (Zoc. cit.) to the withdrawal of water from the reaction by the sulphuric acid. A third method, which yields by far the best result with refractory substances, is that described in the English Patent No. 7233 of 1897 (Kalle and CO. i9.R.-P. No. 95339), which consists in dropping the diazo-solution into a mixture of dilute sulphuric acid and sodium sulphate heated to 136-145" and allowing any volatile products to distil over. This method, applied in the Patent Specification especially to theTHE ACTION OF WATER ON DIAZO-SALTS.21 production of guaiacol from the diazo-salt of o-anisidine, is successful where others have failed. Although by using this method we have been successful in obtain- ing phenols where previous observers have failed, many of the sub- stances examined have given good yields of the corresponding phenols under the ordinary conditions. This applies particularly to the cases of the diazo-salts from dibromoaniline, dibromo-p-toluidine, and bromo- and chloro-p-toluidines, which were described by Wroblewski in 1874 as yielding no trace whatever of phenols, but only the substituted hydrocarbon. More recently (in 1884), he attributes this abnormality to the presence of a minute quantity of alcohol in the diazo-derivative, left in during the preparation. This explanation is possibly correct, although it is evident that the quantity of alcohol contained in the diazo-salt must have been considerable.I n every case which has been examined, we have been able t o obtain the corresponding phenols, and therefore prove that there is, so far as we know, no case in the benzene series where an ortho-substituent hinders or diverts the course of the reaction, as is, apparently, the case in the diphenyl series. I n addition to this we have been able to throw some light on the course of the diazo-reaction as applied to the sub- stances here described.EXPERIMENT A L. o- Anisicline. Limpach (.Bey., 1891, 24, 4136) prepared the diazo-salt from this substance and passed steam through the solution, but (‘ es t r a t voll- stiindige Verharzung ein ” and no guaiacol could be detected. Gatter- mann (Ber., 1899, 32, 1136) attempted to prepare the hydroxy-com- pound, and remarks (( dass auch dieses (namely, o-anisidine) eine ungewohnlich bestandige Diazoverbindung bildet, die selbst nach dem Erhitzen in einer Bombe auf uber 100’ noch nicht zersetzt war.” As we have been able to isolate guaiacol in the products of decomposition as carried out in the usual may, and apparently by the same method as Limpach used, we give the full details of the experiment. o-Anisidine (24.6 grams) was dissolved in water and 60 C.C.of concentrated hydrochloric acid, and diazotised with addition of 200 C.C. of normal sodium nitrite at 20-25‘. Concentrated sulphuric acid (60 c.c.) was now added and steam passed into the diazo-solution in a large flask arranged for steam distillation. The decomposition proceeded very slowly and there was no violent evolution of nitrogen as in the case of more unstable diazonium salts. The colourless solution gradually turned pink, which was obviously due to the formation of an azo-colour- iug matter, this being found to dye wool direct from an acid bath.22 CAIN AND NORMAN: The colouring matter is therefore most probably produced by the com- bination of the diazonium salt with the guaiacol formed, and con- sequently possesses the formula CH3*O*C6H4*N:N*C6H3(OH)*O*CH,.A t a r gradually collected in the flask and an oil was seen to distil with the steam. The distillation was carried on for four to five hours, and the contents of the flask, when tested the following day, gave no colour with ‘‘ R salt,’’ showing that the diazonium salt had been com- pletely decomposed. Examination. of Residw-The solidified tar was filtered, dried, powdered, and in one experiment boiled out with water, which dissolved the foregoing azo-colouring matter. In other experiments this treatment with hot water was omitted. The dry powder was then extracted with ether, leaving a tarry residue. The ethereal solution was dried over calcium chloride, filtered, and the ether distilled off ; a viscid oil was left which distilled above 300O.The distillate was left for some days in a desiccator over sulphuric acid, when large, flat, square tables had crystallised; these were dried on porous porcelain and recrystallised from dilute acetic acid, when fern-like needles separated (m. p. 88-89’), The boiling point was a little above 310’. The substance was soluble in alkalis, being reprecipitated by acids. The high boiling point suggested the formation of a con- densation product, and analysis showed that probably the mono-methyl ether of o-dihydroxydiphenyl, OH(2)C,H,*C,H4( 2)0*CH,, had been formed. 1 2 5’ 9t 1’ The yield was very small. 0.1074 gave 0,3057 CO, and 0.0609 H,O. Cl3H1,O2 requires C = 77.96 ; H = 6.05 per cent. Examhaation of the Steam Distillate.-The aqueous distillate con- taining drops of oil was saturated with salt and extracted with ether, After evaporating off the ether from the dried solution, an oil was left which was distilled. The temperature rose gradually to nearly 300O.On rectifying, a large fraction,was collected at 195--205O, when the distillation was stopped. This oil was found to be gusiacol (b. p. 2 0 5 O , m. p. 33O) ; it solidified in a desiccator, gave a green colour with ferric chloride, and was identical with a sample of guaiacol prepared accord- ing to the above-mentioned English Patent. The residue in the distilling flask was extracted with ether and the oil left on evaporating off the ether solidified to large, square, tabular crystals. These were recrystallised from dilute acetic acid and melted at 88-89O. A further quantity of the methyl ether of o-dihydroxydiphenyl had thus distilled over with the guaiacol during the steam distillation.The diazo-salt from o-anisidine was also decomposed exactly as described in the fore- going English Patent and a satisfactory yield of gusiacol was obtained. C = 77-63 ; H = 6.30.THE ACTION OF WATER ON DIAZO-SALTS. 23 Dichloronniline (NH, : C1: C1= 1 : 2 : 5). Schlieper (Bw., 1893.26, 2465) was unable to prepare the diazo-salt of this substance, obtaining only the diazoamino-compound. This was also confirmed by Zettel (Rer., 1893, 26, 2471). As no mineral acid was used by these chemists, but the diazotisation attempted by the use of amyl nitrite, we were confident of being able to prepare the diazo- solution in the usual way, and a clear solution was easily obtained.I n one experiment the base was dissolved in hydrochloric acid, sulphuric acid added, and sodium nitrite solution dropped slowly into the boiling solution. A red solid formed in the flask was dried a.nd recrystallised from benzene, the solution yielding shining, bronze-coloured plates (m. p. 195O). 0.1330 gave 14.8 C.C. nitrogen at 18.5' and 755 mm. C,,H7N,Cl, requires N = 12.54 per cent. N = 13.0. By heating with glacial acetic acid and acetic anhydride, an acetyl compound was obtained, melting at 226'. The substance formed in the reaction was therefore t'he aminoazo-compound having the above formula; it dissolved with a red colour in concentrated sulphuric acid. During the progress of these experiments, a paper appeared by Noelting and Kopp (Ber., 1905, 38, 3506) describing a number of derivatives of this dichloroaniline and the preparation of the above aminoazo-compound by heating a mixture of the hydrochloride of the base, the diazoamino-compound, and the base itself. I n this way they obtained the aminoazo-compound corresponding exactly with the substance above described.They describe also the preparation of the dichlorophenol by decomposition of the diazo-salt, and therefore there was no necessity for us to carry on our work further in this direction. DibromoaniZine (NH2 : Br : Br = 1 : 2 : 4). Wroblewski (Bey., 1874, 7, 1061) obtained only dibromobenzene by the decomposition of the diazo-salt and detected no trace of the phenol. By using the method of the above-mentioned English Patent, the dibromophenol (m.p. 39') was isolated. Br = 63.0. C,H,OBr, requires Br = 63.5 per cent. 0.224 gave 0.3332 AgBr. s- T&hZoroaniline. Hantzsch (Ber., 1895, 28, 685) prepared a solution of the diazo- chloride of this substance, and was unable to detect the formation of the corresponding phenol by the action of heat. He says '' Trichlor-24 CAIN AND NORMAN: diazobenzolchlorid lasst sich mit Wasser, ja selbst mit salpetersalzsaure kochen, ohne Stickstoff zu entwickeln oder sich uberhaupt zu verandern.” Under the usual conditions we were also unable t o isolate any trichlorophenol, but by using the patented method a small quantity of this substance ci%lled over with the steam, and melted correctly, namely, at 68’. 0.2511 gave 0.5441 AgC1. Cl=53*4.C,H,OCl, reqliires C1= 53.8 per cent. s- T T ~ 5romoani line. Silberskein ( J. p. Chsm., 1883, 27, 98) was unable t o obtain any phenolic derivative by boiling s-tribromobenzenediazonium sulphate or nitrate with dilute acids. Ihntzsch also (Bey., 1900, 33, 2517) obtained ‘‘ gar kein Tribrompienol.” Orton (Proc., 1905, 21,170), by using Heinichen’s method, was unsuccessful in his attempt t o prepare this subRtance. We ca11 confirm the work of these chemists, as under ordinary conditions no phenol can be isolated. When, how- ever, the patented method is applied, a small quantity of s-tribromo- phenol (m. p. 92’) can be obt&md. 0.2113 gave 0.3590 AgBr. Br = 72.31. C,H,OBr, requires Br = 72-51 per cent. Chloro-p-toluidine (NH, : CH, : C1= 1 : 4 : 2). Wroblewski (Bey., 1874, 7, 1061 ; Ann., 1873, 168, 147) states that the diazonium salt of this substance on decomposition by boiling yields m-chlorotoluene and no phenol.By decomposing the diazo- solution according to the patented method and by extraction of the distillate with ether, an oil was obtained boiling at 191’. C1= 24.47. C,H,OCl requires C1= 24-86 per cent. 0.1621 gave 0.1 607 AgC1. The substance is therefore chlorocresol (OH : CH, : C1= 1 : 4 : 2). Brmo-p-toluidine (NH, : CH, : Br = 1 : 4 : 2). Wroblewski in this case also (Zoc. cit.) failed to obtain the bromo- cresol, but described the production of the m-bromotoluene ; due, no doubt, as indicated above, to the presence of alcohol. When steam is passed through the solution containing the diazonium salt, an oil (b.p. 214’) distils over, which is identical in every way with the bromocresol described by Schall and Dralle (Bw., 1884, 17, 2530). 0.2209 gave 0.22053 AgBr. Br = 42.48. CiH70Br requires Br = 42-78 per cent.THE ACTION OF WATER ON DIAZO-SALTS. 25 Dibromo-p-tohidime (NH, : CH, : Br : Br = 1 : 4 : 2 : 6). Here again Wroblewski obtained ‘ I nicht die geringste Spur von By the patented method we have Kresol,” but only dibromotolueno. had no difficulty in obtaining the expected dibromocresol (m. p. 48O). 0-211 gave 0.297 AgBr. Br = 59-90. C7H,0Br, requires Br = 60.15 per cent. It is to be concluded from the above experiments that the supposed cases of abnormal behaviour on boiling certain ortho-substituted diazonium salts with water have no foundation in fact.We do not claim to have examined every alleged exception to the general rule, but think that the number of examples shown in this paper is sufficient to indicate that any remaining instances which may possibly have been overlooked by us are to be regarded with suspicion. I n conclusion, we wish to express our grateful thanks to the Chemical Society for the grant from the Research Fund by means of which the cost of this work was partly defrayed. NOTE BY JOHN CANNELL CAIN.-with the completion of the fore- going work it may be of interest to summarise the general results of a series of researches on the diazo-reaction carried out by me during the past four years. The action of water on diazonium salts (mostly in presence of a mineral acid) has been investigated both quantita- tively and qualitatively, and the main conclusions are as follows : 1.Diazonium salts of the benzene and naphthalene series decom- pose according to the equation expressing a unimolecular reaction, namely : 1 A -log-- = C (a constant) t A - x (Trans., 1902, 81, 1412 ; 1903, 83, 206). Only one tetrazo-salt (from dichlorobenzidine) was found to conform to this rule. By the deter- mination of the value of “0” in the above equation, the relative stability of diazoniurn salts is obtained. 2. The rate of decomposition increases rapidly with the temperature, the values of “ C ” obtained being in accordance with Arrhenius’ formula for the temperature-coefficient, namely : (Trans., 1903, 83, 470). 3. The rate of decomposition (in the case of benzenediazonium salts) is independent of the quantity of mineral acid present (except sulphuric acid, which tends to withdraw water from the sphere of action), and Ct, = Ct,.e 4 Ti - TO) TI TO26 CROSSLEY AND RENOUF : DIHYDROLAUROLENE, DIHYDRO- is independent of the nature of the acid. Equivalent solutions of benzenediazonium chloride, sulphate, nitrate, and oxalate decompose at the same rate (Ber., 1905, 38, 2511). 4. The primary action in the decomposition (except possibly in the diphenyl series) is that phenols are formed. A number of apparent exceptions to this rule have been proved to be groundless (this paper). Unless special precautions are taken in the decomposition of very stable diazonium salts, the unchanged diazo-salt condenses with the phenol formed (probably yielding either a diazo-oxy-compound or an azo-colouring matter), and consequently the latter is not isolated (Zoc. I n the diphenyl series, the tetrazo-salts from dianisidine and dichlorobenzidine yield as chief products quinones and not phenols ; Thereas benzidine, tolidine, &c., behave normally (Trans., 1903, 83, 688). The tetrazo-salt from dichlorobenzidine behaves normally in the various other diazo-reactions (Trans., 1904, 85, '7). 5. The influence of a substituent in the ortho-position with respect to one diazonium group contained in a tetrazo-salt is very great, and it has been possible in such a case to decompose one diazonium group completely and leave the other intact, the hydroxydiazonium salt thus being obtained in the crystalline condition from boiling water (Trans., 1905, 87, 5). cit.). MUNICIPAL TECHNICAL SCHOOL, BURY, LANCASHIRE.