229 BY JOHN J. SUDBOROUGH, PERCY G. JACKSON, AND LORENZO L. LLOYD, IT has been previously shown by Victor Meyer and one of us (Bey., 1894, 27, 510, 1580, and 3146) that diortho-substituted benzoic acids yield no ethereal salts when their cold methyl alcoholic solutions arc saturated with dry hydrogen chloride; these salts, however, can be obtained by the action of alkylic iodides on the silver salts of the acids, or by the action of alcohols on the acid chlorides. Victor Meyer has shown that these ethereal salts are characterised by the fact that they are extremely difficult to hydrolyse (Bey., 1895, 28, 188). The acid chlorides of diortho-substituted benzoic acids, further, are remarkably stable, being only completely decomposed on prolonged boiling with potassium hydroxide solution (Sadborough, Trans., 1895, 6'7, 587 ; v.Meyer, Rev., 1894, 27,3153 ; Liitzens, Be?*., 1896, 29,2837). It has also been pointed out by one of us (Sudborough, Trans., 1895,67, 601) that probably diortho-substituted benzamides exhibit a similar degree of stability. Thns it has been shown that 2:6-dibromobenz- amide remains practically unacted on when heated a t 170" with SO per cent. sulphuric acid, whereas the isomeric 2 : 4-dibromobenzamide, under similar conditions, is completely hydrolysed. With the purpose of further investigating the properties of sub- Ortho-, meta-, and para-bromobenzamides, 2 : 4-, 2 : 6-, and 2 : 5-dibromobenzamides, 2 : 4 : 6- and 3 : 4 : 5-tribromobenzamides, 2 : 4 : 6-Trichlorobenzamide, 2 : 4 : 6-Trimet hylbenzamido, Mesitylacetamide, stituted benzamides, we have prepared the following amides.and have in each case determined the amount of amide converted into acid when heated with different strengths of sulphuric acid a t given temperatures. The general method adopted was as follows. 0.5 gram of the amide was heated with 20 or 25 C.C. of the sulphuric acid to a given temperature for different lengths of time, generally in sealed tubes. On cooling, the acid was diluted with water and the mixture extracted three times with ether ; the ethereal solution was extracted twice with dilute caustic potash, in order to remove any acid which might have been formed, and the combined alkaline extracts were acidified with hydrochloric acid and then extracted twice with ether. This ethereal extract was washed with water, dried over calcium chloride, transferred to a tared flask, and the ether slowly evaporated.VOL. LXXI, It230 SUDBOROUGH, JACKSON, AND LLOYD : During the course of this work, the following new compounds have been prepared. 3 : 5-Di~onzobenxc~micle, C,H,Br,* CONH, [ CONH, : Br, = 1 : 3 : 51, was prepared by the action of aqueous ammonia on 3 : 5-dibromobenzoic chloride. It crystallises from hot, aqueous alcohol in colourless needles, melts at 187", and is soluble in alcohol, ether, benzene, and chloroform, but more sparingly in boiling water. 0.2608 gave 0.3505 AgBr. Br = 57.19. 0.3005 ,, 12.8 C.C. moist nitrogen a t 11" and 760 mrn. N = 5.07. C,H,Br,*CONH, requires Br = 57-31 ; N = 5.02 per cent. This amide was completely hydrolysed on treaticg it with 75 per cent.sulphuric acid for 7 hours at 160". The acid thus formed melted at 209". Synametriccd tyibronzo6enxonit.l.ile, C?,H,Br,.CN [CN : Br, = 1 : 2 : 4 : 61. -This nitrile is really prepared from symmetrical tribromaniline by the Sandmeyer Reaction (as described in Bey., 1894,27, 512). It is advisable, after diazotising, to filter off from any unaltered tribromaniline before adding to the cuprous potassium cyanide solution. The reddish-coloured product which is thus obtained is best boiled in alcoholic solution with animal charcoal, and then recrystallised from dilute alcohol, when it is obtained in small, colourless needles, readily soluble in alcohol and ether, but only sparingly in boiling water. It melts a t 127", and very slowly volatilises in a current of steam.0.1452 gave 0.241 AgBr. 0.4 ,, 14.2 C.C. moist nitrogen a t 16" and 756 mm. N=4*116. Br = 70.62. Theory requires Br = $0.59 ; N = 4.12 per cent. Xymmet&cd t~ibromoben,xc6mide, C,H,Br,* CONH,, is readily obtained when the nitrile is heated in sealed tubes with 80 per cent. sulphuric acid at 160" for several hours; the nitrile is not hydrolysed when heated to the same temperature with 60 per cent. acid. Ten grams of the nitrile were heated with 200 C.C. of 80 per cent. sulphuric acid at 160" f o r 6 hours, and on cooling, the acid was diluted with water, and the solid residue collected and recrystallised from dilute alcohol or from boiling water. It crystallises in small, colourless needles, or from dilute solutions in hard prisms, melts a t 193-194", is readily soluble in alcohol, ether, and chloroform, moderately in boiling water, and almost insoluble in cold water.I. 0.1675 gave 0.264 AgBr. Rr = 67.06. 11. 0-401 HI. 0,2156 ,, 14.2 C.C. moist nitrogen at 12" and 759 mm. N=4.19. ,, sufficient NH, to neutralise 6 C.C. N/lOH,SO,. N= 3-89. Theory requires Br = 67.04 j N = 3.91 per cent,DITHIO-SUBSTITUTED BENZOIC ACIDS. 231 As regards its hydrolysis, this amide exhibits a remarkable degree of stability as compared with its isomeride, [COXH, : Br, = 1 : 3 : 4 : 51, see appended table. Unsynnzetiical tribi.onzobe,z~~cp,z~~~e, [CONH, : Br, = 1 : 3 : 4 : 51.-3 : 4 : 5- Tribromobenzoic acid was prepared by the method previously des- cribed (Bey., 1894, 27, 513), a slight modification being made by using hydrochloric instead of sulphuric acid for liberating the nitrous acid ; the yield of tribrom-acid is very good.The acid was converted into its chloride by warming it with phosphorus pentachloride, the oxychloride distilled off under diminished pressure, and the residue treated with concentrated ammonia ; the solid mass thus obtained was washed with water, and recrystallised from dilute alcohol. It is readily aoluble in ether, chloroform, and alcohol, but only sparingly so in boil- ing water. It crystallises from its dilute alcoholic solution in colourless, slender, silky needles, and melts a t 199-200", and not a t 210' as pre- viously stated (Trans., 1894, 65, 596). 0-3008 gave 0.3117 AgBr. Br = 67.32. 0.1961 gram, when heated with soda lime, gave sufficient ammonia t o neutralise 5-3 C.C. of N/10 sulphuric acid.N = 3.18. C,H,Br,* CONH, requires Br = 67.04 and N = 3.91 per cent. Xynmet&al t~ic~loi~obe~aaonitl.ile, C,H,Cl,* CN [CN : C1, = I : 2 : 4 : 61.- Trichloraniline was prepared by saturating an acetic acid solution of aniline with chlorine, filtering OR the crystalline precipitate, and decom- posing it with water. The trichloraniline was then diazotised in the cold, and poured into a warm (60") solution of cuprous potassium cyanide (10 grams of 98 per cent. potassium cyanide to 9.5 grams of copper sulphate). The precipitated nitrile, after being allowed t o stand for 12 hours, was extracted with ether, the ether evaporated, and the residue dissolved in alcohol, boiled with animal charcoal, and recrystal- lised several times from dilute alcohol. The yield is not very good, in most cases not more than 20-30 per cent.of the theoretical, and it is still worse when the diazotised solution is poured into boiling cuprous potassium cyanide solution. The nitrile crystallises in colourless, silky needles, melts at 75", and is extremely soluble in alcohol and ether, but only sparingly in boiling water. C,H,Cl,*CN requires C1= 51.57 per cent., and N=6.78 per cent. Found, C1=51*53 per cent., and N = 6.9 per cent. The nitrile is practically unacted on when heated a t 160" for 3 hours with 60 per cent. sulphuric acid, but is readily hydrolysed when heated with 80 per cent. acid ; thus 0.5 gram of the nitrile, when heated with 20 C.C. of 80 per cent, sulphuric acid a t 160" for 3 hours, gave 0.43 gram of the amide and 0.06 gram of trichlorobenzoic acid.2 : 4 : 6-~ricl~Zol.obenxan2~~e, obtained by heating the nitrile (10 grams) P 2232 SUDBOROUGH, JACKSON, AND I,T,OYD : with 80 per cent. sulphnric acid (200 grams) for 3 hours a t 160°, crys- tallises from boiling water in colourless, glistening plates, and melts at 177". It is readily soluble in most organic solvents, and moderately so in boiling water. 0.1865 gave 0.3565 AgC1. c11= 47.29. 0.1 909 gram, when heated with soda lime, gave sufficient ammonia t o neutralise 8.3 C.C. of NjlO sulphuric acid. N = 6.09. C,H,CI,* CONH, requires C1= 47.43 ; N = 6.23 per cent. This arnide is also remarkably stable when heated with 75 per cent. sulphuric acid. iClesit~la;ceto~nzicle,C,H,Me,.CH,*CONH,[CH,*CONH,:Me,=1:2:4:6]. This compound was obtained by converting mesitylacetic acid into its chloride, and pouring the chloride into ammonia; the action is violent, and the product, after recrystallisation from dilute alcohol, forms long, slender, silky needles, melting a t 209-210'. It is iden- tical with the compound obtained by Claus (J.p ~ . Clmz., 1890, [ 21, 41, 507), which melted a t 208". Results of HydYolyses. II No. - 1 2 3 4 5 6 7 8 9 10 11 12 1 3 14 15 16 17 18 19 20 21 - Amide taken. Urthobromobenzamide . . . . . , Do. Me tabromobenza.micle . . . . . . , . . Do. Do. Do. Do. Do. Do. Do. Do. Parabromobenzamide ......,.. 3 : 5-Dibromobenzamide.. . .. 2 : 4-Dibromobenzamide.. , . . . 2 : 6-Dibroniobenzan~ide . . . . I 3 : 4 : 5-Tribronio~enzamide..2 : 4 : 6-Tribromobenzaniide. 2 : 4 : 6-Trichlorobenzamide.. Mesitylacetamide . . . . . . . . . . . . . Do. D O . - 0" 3 cu 5 ba g* 4 u ) m 9 'er cent. 30 50 30 30 30 30 30 50 50 75 75 75 75 75 75 75 75 75 75 30 30 - 20 20 20 20 20 20 20 20 20 25 25 25 25 25 25 25 25 25 25 25 5 - QJ + M 2 i ~~ Boiling Do. Do. Do. Do. Do. Do. Do. Do. 160" 160" 160" 160" 160" 60-17F 160" 160" 160" 160" Boiling Do. - 15 mills 30 9 J 5 9 > 15 9 9 15 ? ? 30 9 9 30 9 2 7 9 9 7 9 9 7 9 ) 3 9 3 3 > 9 7 9 9 30 1 9 2 hour J J J 7 J 9 J 3 15 mins 2 hour - :ram, 0.5 0.5 0.5 0.5 0 -5 0-5 0.5 0 5 0.5 0.5 0.5 0.5 0.5 0 *5 0 -5 0 -5 0 *5 0.5 0.5 0.5 0.1 - gram. 0'126 0.411 0'335 0'4985 0'499 0'4056 0'4986 0.125 0.255 0'487 0'47'39 0.4718 0'498 0.058 0.0626 0'4703 0.023 0'022 0'242 0'1752 0.073 25-07 81-79 66'66 99 *20 99-30 80.71 99'22 24.91 50'81 97'05 94'44 94'02 99'24 11.56 12'47 93.80 4 '59 4'39 48'18 34-85 72'59 -DITHIO-SUBSTITUTED BENZOIC ACIDS.233 The behaviour of these acid-amides towards hy drolysing agents is extremely characteristic. Of the three monobromobenzamides, the ortho-compound is somewhat more stable than the other two. This is only what was to be expected, since V. Meyer has shown that, not only are the ethereal salts of ortho-substituted-benzoic acids more slowly formed than those of isomeric acids, but also that these ortho-substi- tuted salts are more stable than the isomeric meta- and para-substituted compounds. It has further been shown by one of us that orthobromo- benzoic chloride is somewhat more difficult to hyclrolyse- than the meta- and para-compounds.Of the di- and tri-bromobenzamides, those which have bromine atoms in the two ortho-positions are remarkably stable, yielding only 4-12 per cent. of acid when heated with 75 per cent. sulphuric acid at 160" for several hours, whilst the isomeric amides in which the two ortho- positions are not substituted yield over 90 per cent. of acid under exactly similar conditions. This, again, is in complete accord with the behaviour of the corresponding ethereal salts and acid chlorides. It is interesting to note that 2 : 4 : 6-tribromobenzamide is even more stable than 2 : 6-dibrsmobenzamide7 although both are ortho-substituted ; the introduction of a third substituting group, even when the two ortho- positions are already occupied, seems to render the amide more difficult to hydrolyse.A similar phenomenon has already been noticed in the case of substituted benzoic chlorides. 2 : 4: 6-Trichlorobenzamide proves to be difficult to hydrolyse, but not nearly so difficult as the corresponding bromine derivative. This is in accord with the results obtained with substituted benzoic acids ; it has been shown by V. Meyer t h a t the raterat which diortho-substituted benzoic acids are converted into ethereal salts depends mainly on the atomic or molecular weights of the atoms or groups which occupy the ortho-positions, Diortho-substituted acids, in which the substituting groups are methyl groups (CH, = 15), are slowly converted into ethe- real salts a t the boiling point of the alcoholic solution, whereas di- ortho-substituted acids, in which the substifuting groups are bromine atoms or nitro-groups (Br = 80, NO, = 46), do not yield a trace of the salt under similar conditions, We have also attempted t o hyilroly se mesitylformamide-2 : 4 : 6-trimethylbenzamide-by heating it with '75 per cent.sulphuric acid, in the hope of being able t o show that it mas more readily hydrolysed than the corresponding trichloro-derivative (C1= 35.5). It was, however, found impossible to carry out the pro- cess of hydrolysis, as the amide was charred and completely destroyed by the sulphuric acid. An interesting point in connection with the diortho-substituted benzoic chlorides is the fact that, although they are so dificult to hydrolyse by treatment with water or sodium hydr- oxide solution, they are yet readily acted on by aqueous ammonia,234 SUDBOROUGH, JACKSON, AND LLOYD : DITHIO-, ETC.yielding the corresponding amides. In brder t o prepare 2 : 6-dibromo- or 2 : 4 : 6-tribromo-benzamide, it is only necessary t o warm the corres- ponding chloride with aqueous ammonia for a few minutes. Some of the isomeric chlorides react with cold, aqueous ammonia ; the diortho-substi- tuted compounds, however, require heating with the smmoniacal solu- tion for a short time. There is thus a difference between the diortho-substituted chlorides and their isomerides as regards their behaviour towards aqueous ammonia, but this difference is not nearly so marked as the difference between the two series of compounds when treated with hydrolytic agent 5.Wegscheider (Momtsh., 1895,16, 75) has' suggested t h a t the reason why diortho-siibbtituLed acids do not yield ethereal salts on treatment with alcohol and hydrogen chloride, is to be sought for in the fact t h a t the substituting groups prevent the formation of a n additive compound of the acid with the alcohol, this additive compound always being an intermediate product in the preparation of these salts by the method mentioned above. It may be that, in the hydrolysis of acid chlorides and amides, an additive compound with water is also formed, and t h a t this is afterwards decomposed into the acid and hydrogen chloride, or into the acid and ammonia. The substituting groups in the ortho- positions would have a tendency to prevent the formation of such an additive compound, and would thus retard the hydrolysis. In the action of ammonia on the acid chlorides, we must suppose that no such intermediate additive product is formed, but t h a t the chlorine is at once withdrawn and replaced by the amido-group. I n conclusion, we intend taking up the study of the amides of a few methyl- and chloro-substituted benzoic acids, and also, if possible, the preparation of certain diortho-substituted benzsldehydes and benzylic alcohols, in order to determine whether they also have such well-defined characteristics. The qnestion a s t o whether a diortho-substituted benz- aldehyde can form additive conipouncls with hydrogen cyanide, &c., or whether it can enter into t'he usual aldehydic condensations ( V . Meyer, Bey., 1895, 28, 1267), seems to us to be one worthy OF closer study. UNIVERSITY COLLEGE, NOTTINGRAM,