410 MELDOLA AND HOLLELY :XLV.-Acylation as InJEuenced by Xteric Hindrance :the Action of Acid Anhydrides orb 3 : 5-Dinitro-p-aminophenol.By RAPHAEL MELDOLA and WILLIAM FRANCIS HOLLELP.IN compounds of the type of isopicramic acid (I) the protection ”of the hydroxyl is so effective that methylation by methyl sulphateand alkali completely fails to attack this group, the aminqroupalone being attacked, with the final formation of those quinone-ammonium- derivatives whichpapers formerly communicatedOHN0,()N02\/N=2(1. )*have been described in a series ofto, and published by, the Society :OH/\NO,j,NO, INH2(11.)The isomeric 3 : 5-dinitro-paminophenol (I1 above : Reverdin,Arch. Sci phys. mt., 1904, [iv], 18, 342; 1905, 19, 353) shows anequally “ protected ” amino-group, so that methylating agents inthis case fail to attack this group.The corresponding dinitro-anisidine can be readily prepared by the action of methyl iodide onthe silver salt of the monoacetyl derivative and hydrolysis of theproduct (Meldola and Stephens, T., 1905, 87, 1206), but themethylation of the amino-group cannot be effected by methylsulphate alone, whilst alkaline methylating agents are inapplicableon account of the tendency of the free alkali to cause completedecomposition with removal of a nitregroup. I n these circumstancesit became of interest to study the influence of configuration on theintroduction of acid radicles, it being well known that underordinary conditions unprotected amino-groups are much morereadily acylated than similarly unprotected hydroxyl groups.Thusisopicramic acid quite readily gives a monoacetyl derivative (acetyl-amino-compound) when boiled for a few minutes in acetic acidsolution with acetic anhydride. The hydroxyl in this compound isonly attacked when the monoacetyl derivative is dissolved inundiluted acetic anhydride mixed with a little concentratedsulphuric acid. From this point of view, the action of aceticanhydride on 3 : 5-dinitro-paminophenol ha5 been studied under* This formula is assigned to the free compound; in solution it may have aquinonoid structure (Meldola and Hewitt, T., 1913, 103, 880)ACYLATION AS INFLUENCED BY STERIC HINDRANCE. 411various conditions, and with results which appear of importancefrom several unexpected points of view.In the first experiments the inertness of the compound towardsacetic anhydride was revealed by the fact that acetylation in aceticanhydride a t 1000 only takes place after heating for several hours.Under these conditions the product proved t o be mainly the knowndiacetyl derivative.The same diacetyl derivative is formed imme-diately on dissolving the dinitro-compound in acetic anhydride withthe addition of sulphuric acid. From this diacetyl derivative theknown monoacetyl derivative (m. p. 182-183O) can be most readilyprepared by dissolving the diacetyl compound as rapidly as possiblein cold dilute sodium hydroxide and precipitating by acid. Thephenolic character of the product indicates that the 0-acetyl isremoved by the alkali.These first experiments resulted, therefore,simply in the production of the diacetyl derivative, and from thelatter the N-monoacetyl derivative by quite normal methods,Acetylation in Acetic Acid Solutwn.In order to modify the action of the acetic anhydride, the latterwas diluted with glacial acetic acid, and the experiments continuedwith this mixture. Under these conditions it wit^ found that anon-phenolic substance wi~g formed, the further investigation ofwhich showed that the compound was. isomeric with the knownmonoacetyl derivative. After many experiments the followingmethod was found the most effective for the preparation of thenew compound :The dinitroaminophenol is dissolved in a small quantity of hotglacial acetic acid, and two to three times the theoretical quantityof acetic anhydride necessary for the introduction of one acetylgroup added to the solution.The latter is then boiled for half-an-hour, and when cold poured into water so as to precipitate theproduct. I f the heating is prolonged beyond the time specified, theyield is diminished; three or four hours' boiling appears t o destroythe compound completely. The crude product is collected, washed,and purified by first washing with a little dilute alkali, then withwater, and finally crystdlising from alcohol or acetic acid. Thepure compound crystallises in flat needles or scales of a golderi-yellow colour; the melting point is 185-186O. A mixture of thesubstance with the isomeric monoacetyl derivative (m. p.182-1 830)melted a t 151-156O:0-1026 gave 0-1500 CO, and 0.0282 H,O. C=39.87; H=3*05.0'0898 ,, 13.7 C.C. N, (moist) a t 21° and 762.8 mm. N=17.43.0.1090 ,, 0.1588 CO, a, 0.0284 HZO. C=39.73; H=2.89.C,H,O,N, requires C=39*82; H=2.93; N=17.43 per cent.E E 412 MELDOLA AND HOLLELY :Them results leave no doubt as to the isomerism between the twoacetyl derivatives.3 : 5-Dinitro-Oa;cetyI-pami~pl5.e?tot.I n discussing formuh for tho new compound, its properties ascompared with those of its isomeride had to be considered. TheN-acetyl derivative when pure crystallises in nearly white needles,extremely soluble in alcohol, and distinctly phenolic in character.The new compound is fairly stable towards alkali, and not veryreadily soluble in alcohol. It is very stable under the influence ofheat, and can be partly sublimed without decomposition whenheated in a dry tube.The most striking visible difference betweenthe isomerides is the intense yellow co'lour of the new compound.The chemical properties of the latter are at first sight quite inharmony with the view that it is the 0-acetyl derivative:On further acetylation with acetic anhydride in presence of alittle sulphuric acid it gives the known diacetyl derivative identicalwith that obtained by the direct acetylation of the dinitro-com-pound- When hydrolysed by sulphuric acid it gives 3 : 5-dinitrepaminophenol; it imparts a transitory violet colour to an alkalihydroxide solution, this colour being due to the dinitro-compoundliberated by hydrolysis.The free dinitro-compound has the sameproperty, the transient violet colour being due to the alkaline saltwhich is a t first formed, but decomposes in presence of excess ofalkali. Boiling with sodium carbonate solution also effects hydro-lysis, and the violet solution of the sodium salt on acidificationgives 3 : 5-dinitro-paminophenol. Attempts to methylate the amino-group by methyl sulphate under various conditions gave negativeresults. The compound dissolves much more freely in methylsulphate than the isomeric N-acetyl derivative, but the solutionafter being kept at looo for an hour wi18 found to contain only theoriginal dinitroaminophenol, the methyl sulphate acting simply asa hydrolysing agent.These properties all point to the conclusion that the new com-pound is the expected 0-acetyl derivative ; nevertheless, its d o u rhnd the extreme readiness with which it is formed seem to indicatethat the simple formula given above does not adequately representthe whole of the ascertained facts.The acetylation of a truephenolic hydroxyl group by simply boiling for a short time witACYLATION AS INFLUENCED BY STERIC HINDRANCE. 413acetic anhydride diluted with acetic acid is, as already stafed, mostunusual. Moreover, such phenolic acetatee, even when containingseveral nitro-groups, are never very highly coloured. The colour ofthe present compound suggested, theref ore, that the amino-groupwas implicated, and the formulz given below have been considered :0 OAc OAc/\ /\or cross-linking) I I:NO2H NO, I ‘No“\/NH,Ac g HA\/IN0\()N ’.Hf- /NO,\/NO, I 1 (..(1.1 (11.1 (111.)Of these formula?, I is disposed of by the fact that no evidenceof interconvertibility of the isomerides has been observed; allattempts to convert the N-acetyl derivative into the isomeride andvice versa, have-led to negative results. Between I1 and I11 it isnot so easy to decide, but the absence of acidic character andgeneral analogy to similarly constituted nitroamino-compoundsleads us to adopt the “inner salt” formula (111) as the mostprobable.It follows from these conclusions that the isomerides are ofdiff went types of benzenoid structure, the N-acetyl derivative beingincapable of forming an “inner nitrolic salt ” on account of theattachment of the acid radicle to the 4N-atom, whilst the 0-acetylderivative containing a nitregroup orthol to the 4NH2-group iscapable of (‘ inner salt ” formation, with consequent orthoquinonoidstructure.Specimens of the isomerides were kindly examined forus spectroscopically by Dr. J. T. Hewitt, by whom the accompany-ing curves represehting the absorption spectra have been prepared.It will be seen that the absorption spectrum of the 0-acetylcompound is quite different in type from that of the N-acetylderivative.We may add that in deciding in favour of the above formula(111) we have had also under consideration the results of certainexperiments devised with the express object of getting evidence asto paraquinonoid structure (I) and nitrolic acid structure (11) ;thus the non-phenolic acetyl derivative is not profoundly affectedby such reagents as hydroxylamine, hydrazine, or phenylhydrazine,all of which simply act as hydrolysing agents.An attempt to forma nitrolic salt by adding sodium ethoxide to the toluene solutionof the compound also led to a negative result. The whole of thecompound is precipitated by this means as a brown, amorphoussubstance, which is the same product as that which results fromthe decomposition of the origind dinitro-compound by the samemethod. Sodium ethoxide does not, therefore, form a nitrolic salt414 MELDOLA AND HOLLELY:but simply removes the 0-acetyl, and then decomposes the dinitro-compound with the formation of the usual indefinite products.The Colour of Nitroaminop?benols and their Derivatives inRelation to their Constitution.This conclusion concerning the constitution of the 0-acetyl deriv-ative warrants the extension of the same type of structure t o theSca Ze of oscillation frcqiscncics.-- Non-ph,enolic acctyl derivative in alcohol.- - - Phenolic acetyl derivative in alcohol..... .. ). f ? 9 ) ,, alcoholic KOH.original (unacetylated) compound, and, generally, to all the nitro-aminophenols and their derivatives which have been under investi-gation in our laboratory for many years. The condition essentialfor the production of highly-coloured compounds appears t o be theortho-position of a nitro-group with respect to an amino-group or ACYLATION AS INFLUENCED BY STERIC HINDRANCE.415substituted amino-group. The presence of additicmal nitro-groupsin the nucleus has generally the effect of increasing the intensityof the colour. From this point of view 3 : 5-dinitro-paminophenolwould have the formula (I) * :OH OH O*CH,(1.1 (11.1 (111.)O*CH, OH OoCH,/\ N02/\N0 /\NO, I Red l"(j\ / * \o \o .. /oNH2-- GH2--/ NH,---/NO,! Bed I* NO Bsd 1.~0NO;\/* \(IV.) (V. 1 (VI. 1The above compound is of a deep red colour, all analogouslyconstituted compounds being more or less red and far more highlycoloured than the mere presence of nitro- and amino-groups in thenucleus would account for unless some more profound change ofstructure is allowed; thus, isopicramic acid, in which there is nonitro-group cirtho to an amino-group, is very feebly coloured (inthe free state) as compared with the above isomeride.On theother hand, 2 : 3-dinitro-paminophenol (I1 above) is red (Meldolaand Hay, T., 1907, 91, 1482), and the corresponding 2:3-dinitro-panisidine deep orange (Meldola and Eyre, T., 1902, 81, 990).2 : 6-Dinitroanisidine (111) is ochreous (Meldola and Stephens, T.,1905, 87, 1204) and 3:5-dinitroanisidine (Tv) deep red (ibid.,p. 1206). 2:3:6-Trinitro-p-aminophenol (V) is also deep red(Meldola and Hay, T., 1909, 95, 1380), as is the correspondingtrinitroanisidine (Meldola and Kuntzen, T., 1910, 97, 456). Theisomeric 2 : 3 : 5-trinitro-p-anisidine (VI above) is also red (Reverdinand de Luc, Arch. sci.phys. nat., 1909, [iv], 27, 383). A detailedcomparative study of the absorption spectra will be necessary tocomplete the evidence from the optical side. So far as observationswith these compounds have hitherto been carried out, it is instruc-tive to compare the curves representing the absorption spectra of2 : 3 : 5-trinitrvanisidine and of its acetyl derivative (in whichno inner salt formation is possible) given in a former paper(Meldola and Hewitt., T., 1913, 103, 884) with those given in thepresent communication.* This structural formula may explain the inertness of the amino-group towardsacylating agents, although "steric hindrance " must also be allowed for, since the.5-nitro-group would act as a " protecting" group416 ACYLATION AS IEFLUENCED BY STERIC HINDRAKCE.Preparation of the 0-Propionyl Derivative.The extreme readiness with which the hydroxyl group of t'lredinitro-compound can be acetylated suggested the application of theforegoing method for the introduction of other acid radicles. Itwill no doubt be found possible to acylate the hydroxyl group of3 : 5-dinitro-paminophenol by means of the higher homologues ofacetic anhydride generally.A preliminary experiment with pro-pionic anhydride diluted with propionic acid carried out inprecisely the same way-as with acetic anhydride showed that the0-propionyl derivative could be prepared with equal readiness.This compound has all the properties of its lower homologue; itcrystallisee from alcohol in flat, golden-yellow needles, melting at172-173':0.1022 gave 0.1588 GO, and 0.0354 H20.0.1274 ,, 18.5 C.C. N, (moist) a t 23O and 760.5 mm. N=16*37.C,H,O,N, requires C= 42.34 ; H = 3-55 ; N = 16.47 per cent.The compound is therefore 3 : 5-dinitro-0-propionyl-p-amino-cT=42*38; H=3.85.phenol,NO,\ /:NO\KH*- /O'Attempt to Prepare the 0-Benzoyt Derivative.An attempt to introduce benzoyl by the same method led to aninteresting result. The dinitro-compound was boiled as before withbenzoic anhydride dissolved in glacial acetic acid. The productproved, however, to be the 0-acetyl derivative previously described.I n this case, therefore, the benzoic anhydride simply promotes theacetylation, or else the benzoyl derivative is first formed and thenconverted into the acetyl derivative by interchange of radicles.*Whether the protection of the amino-group by ortho-substituentsis exerted equally by radicles other than the nitro-group yet remainsto be investigated. The corresponding 3 : 5-dihaloid-p-aminophenols,so far as we have been able to ascertain, do not appear to be known,and their preparation will be attempted in order to throw furtherlight on this question.FXNSBURY TECHNICAL COLLEGE, E.C.* The dinitro-compound is not acetylated by boiling for many hours with aceticacid alone