PRODUCTION O F NITRO- AND AMIDO-OXYLUTIDINES. PART 11. 235 XVIII.-Productio?z o f some Nitro- and Amido- oxylutidines. Purt 11. By Miss L. HALL (University College, London) and J. NORMAN COLLIE, Ph.D., F.R.& (Professor of Chemistry at the Pharma- ceutical Society of Great Britain, London). THE production of nitro- and amido-derivatives in the pyridine series has been the subject of two former papers communicated by one of the authors to the Society (Trans., 1897, 71, 838 and 1898, 229). In236 HALL AND COLLIE: PRODUCTION OF SOME the first of these papers, it was shown that, when dioxypicoline is dissolved in nitric acid, nitration a t once occurs, and from the nitro- dioxypicoline produced, several amido- and oxy-picoline derivatives could be prepared. The present communication is a continuation of that work.The oxypyridine derivative chosen was lutidone, or aa’-dimethyl-y oxy pyridine. NH CH,* ;ci! / ‘C*CH, Hc,co/8H Lu t id one. It was found that lutidone, however, was not nitrated when treated with nitric acid alone, as was the case with the di-oxypicoline, but was only converted into the nitrate of lutidone, and a mixture of fuming nitric and sulphuric acids had to be employed before it was changed into nitrolutidone. This was the case also with pseudoliitidostyril (Trans., 1898, 230). The nitrolutidone is a very pale yellow, crystal- line compound which dissolves in alkalis with an intense yellow colour, has a strong acid reaction, and is not volatile with steam. When treated with t i n and hydrochloric acid, it is easily reduced, forming an amidolutidone.NH CH3. fi / ‘fi*CH3 HC \ ,c*m GO hmidolu tidone. This amidolutidone, unlike amidodioxypicoline and amidopseud 0- lutidostyril, does not give characteristic colours when treated with various oxidising agents, t h i s being probably due to the fact that the amido- and the oxy-group are in the ortho-position relatively to one another. It does, however, yield a brownish-red coloration with ferric chloride, but with strong sulphuric acid and potassium dichromate or with nitric acid no markedcoloration is produced. It has, however, strong reducing properties ; with nitrate of silver, it gives first a white, semi- crystalline precipitate which is entirely reduced to metallic silver on warming. It also forms two hydrochlorides, a mono- and a di-deriva- tive, and its platinochloride belongs to the class of double salts where 1 mol.of the dihydrochloride unites with 1 mol. of platinic chloride. This platinochloride is very unstable, and when dissolved in water and warmed undergoes reduction, but if its solution be warmed with hydrochloric acid, another kind of change ensues and an exceedinglyNITRO- AND AMIDO-OXYLUTIDINES. PART 11. 237 insoluble platinochloride separates, which, from its analysis, seems t o be the salt of propine-diamine, CH3*f?NHz,H2PtCI, HC*NH, The decomposition having been brought about by hydrolysis, C7HIoN2O + 3H20 = C3H,N, + 2CH3*COOH. This breaking down of the pyridine ring is one of considerable interest, and in a substance like lutidone was hardly to be expected, for the nitrogen atom is bound to two carbon atoms neither of which are united to oxygen, and lutidone is not the anhydride of an amido-acid as is the case of pseudolutidostyril. E XPE R I M E N T A L.aa'-DimethyL y-ox y - P-nitrop yridine. The lutidone used in the following experiments was prepared from dehydracetic acid by heating it in sealed tubes at 130' with excess of strong aqueous ammonia, the product being a mixture of lutidone and ammonium lutidonecarboxglate. By evaporating the contents of the tubes to dryness and subsequent distillation crude lutidone was ob- tained ; this was recrystallised from water until pure. A large number of attempts were made t o nitrate lutidone, at first with nitric acid alone, it having been found in the case of dioxypicoline to give almost quantitative yields of the nitro-derivative.A very soluble nitro-compound mas obtained which decomposed rapidly at 85'. After it had been recrystallised from water, several analyses were made, but although these agreed amongst themselves, yet, as soon a8 the substance was recrystallised from alcohol or acetic acid, the analytical numbers showed that, with each solvent, a diff erent change was being effected, and it was not until the compound was purified by recrystallisation from nitric acid that any results were obtained which could be relied upon. Found C = 45.4 and 44.9 ; H = 5.7 and 5.8 ; N = 15.2 and 15.2. Calculated for C7H,,N20,, C = 45.2 ; H = 5.3 ; N = 15.1 per cent. The substance melted at 120Owhen quickly heated to that tempera- ture, but if kept at 100' it decomposed.Attempts were made also to reduce it with tin and hydrochloric acid, and a crystalline hydrochloride was obtained, but this hydrochloride did not seem to possess the properties of an amido-oxypyridine derivative, neither did the original nitrolutidone yield yellow salts with soda or potash, Ultimately, it was found that the supposed nitro-compound was merely the nitrate of lutidone, for when neutralised with the proper quantity of caustic soda and the solution evaporated, unchanged238 HALL AND COLLIE : PRODUCTION OF SOME lutidone was recovered together with sodium nitrate ; moreover, the original substanco gave, with strong sulphuric acid and ferrous sul- phate solution, the ordinary nitrate test, whilst the supposed hydro- chloride of amidolutidone gave results agreeing with lutidone hydro- chloride, and lutidone was actually prepared from it by treatment with soda solution, Found N = 8.7 ; C1= 22.3 per cent.Calculated for C7H,N0,HCl,H,0, N = 8.7 ; C1= 22.0 per cent. The supposed nitrolutidone was, therefore, merely lutidone nitrate, Not having been able to prepare a nitro-derivative by the action of nitric acid alone, a mixture of nitric and sulphuric acids was next tried. Twenty grams of lutidone was dissolved in 30 C.C. of strong sulphuric acid and 60 C.C. of a mixture of equal volumes of fuming nitric and sulphuric acids was added ; no nitrous fumes were evolved. It was then warmed for a few moments on a water bath and when cold poured into water, and the aqueous solution nearly neutralised with sodium carbonate, pale yellow crystals at once began to separate ; these, which were nearly soluble in cold water, were washed and recrystallised from dilute acetic acid. When pure, they are nearly colourless, having only a faint yellowish tinge.The compound melts at about 290-300° with considerable decomposition. It has a strong acid reaction, and when dissolved in soda yields a brilliant yellow solution ; the yellow solution obtained by dissolving the substance in ammonia gradually loses its colour when boiled, the un- changed nitro-derivative ultimately crystallising out after all the am- monia has been driven off. This ammoniacal solution gives a brilliant yellow precipitate with silver nitrate solution. CVH,JY,O* = C,H,NO,HNOp On analysis, the following numbers were obtained.Fbund C = 49.9 and 49-$ ; H = 5.1 and 5.2 ; N = 16.7 and 16.9. Calculated for C7H,N20,, C = 50.0 ; H = 4.7 ; N = 16.6 per cent, C7H,N0 + HNO, = C7H,N,0, + H20. The compound is therefore the true nitrolutidone, ad-dime thy Z-y-oxy-P-umidopyridine. When nitrolutidone is added to n mixture of granulated tin and strong hydrochloric acid, it at once dissolves and the temperature rises considerably ; the reduction is very rapidly effected, and after the tin has been precipitated by hydrogen sulphide, the filtered liquid on evaporation yields crystals of amidolutidone hydrochloride. This salt is best recrystallised from hydrochloric acid ; it then forms a granular maas of cryatals, which, in solution, have a strongly acid reaction.OnNITRO- AND AMIDO-OXYLUTIDINES. PART XI. 239 keeping, they lose both hydrochloric acid and water. substance was dried between filter paper and analysed as soon as dry. Found C = 33.6 ; H = 6.9 ; N = 11.6 ; C1= 28.4 per cent. Calculated for C7Hl,N20,2HC1,2H,0, C = 34.0 ; H = 7-3 ; N = 11.4 C1= 28.7 per cent. An estimation was also made of the water of crystallisation, but as it loses hydrogen chloride also when heated, the result is possibly not a correct one. Some of the Found (after heating quickly a t 100') H20 = 15.0 per cent. Calculated, 2 mols. of water, H20 = 14.5 per cent. The salt melts at about 275-280O (corr.),losing a considerable amount of hydrogen chloride. When its solution is treated with only one molecular proportion of sodium hydrogen carbonate, a monohydrochloride melting at 186" (corr.) can be obtained; this, when heated at 100' for some time, decomposes and turns brown. It was recrystallised from alcohol and a nitrogen estimation made.Found N= 16e7 per cent. Calculated for C,FI,,N,OHCl, N = 16.1 per cent. Both these hydrochlorides when entirely neutralised by either soda or sodium carbonate, yield amidolutidone which is not very soluble in cold water, and can be easily purified by recrystallisation from water. It crystallises in long, needle-shaped crystals, and when analysed gave the following results. Found C = 52.7 ; H = 7.9 ; N = 17.8 per cent. Calculated for C7H,,N20,H20, C = 53.8 ; H = 7.7; N = 17.9 per cent. The substance was also dried at 100° and analysed. Found C = 60.1 ; H = 7.7.Calculated for C7H10N20, C = 60.9 ; H = 7.3 ; N = 20.3 per cent. Two determinations of the water of crystallisation were also made. Found H20 = 11.7 and 11.5. Calculated for 1H20, 11.5 per cent. This amidolutidone, as has already been pointed out, does not yield such a brilliant series of colour tests, when treated with various oxidising reagents, as either the amidodihydroxypicoline or the amido- pseudolutidostyril. It does, however, give a brownish-red coloration with ferric chloride, and with strong sulphuric acid and solid potassium dichromate a green colour. Its chief characteristic property is its reducing power ; when silver nitrate is added to its aqueous solution, a voluminous, white precipitate is formed at first, but on warming complete reduction occurs, and metallic silver is produced as a N = 20.6 per cent.240 PRODUCTION OF NITRO- AND AMIDO-OXYLUTIDINES.PART 11. grey metallic deposit, in this respect it resembles the amidodihydroxy- picoline. If amidolutidone is persistently boiled with carbonate of soda solution, it is partially decomposed, and some acetate can be detected in the solution, and instead of giving the platinochloride of amido- lutidone, another platinum salt was obtained containing 37.5 per cent. Pt. This agrees with C,HloN20,H2PtC16 ; Pt = 37.2 per cent. The de- composition may possibly have occurred as follows, C7Hl0N,O + 2H,O = C,H,,N,O + C2H40,. The true platinochloride of amidolutidone is a very soluble salt, and is best prepared by pouring a strong solution of platinic chloride on to solid amidolutidone hydrochloride.It then crystallises in small, very characteristic, microscopical, shield-shaped crystals ; these cannot be recrystallised from hot water, as they entirely decompose when warmed in aqueous solution. Found Pt = 32.3 and 32.7; water of crystallisation, H,O = 5.9 per cent. Calculated for C7HloN,0,H,0,H,PtCI, + 2H,O, Pt = 32-4 ; 2H,O =5*9 per cent. When this platinum salt is dissolved in dilute hydrochloric acid and the solution is warmed, a granular platinum salt begins to separate after a short time ; this is only sparingly soluble, even in hot water, its hot aqueous solution also, if not acid with hydrochloric acid, undergoes reduction on boiling, and the dry salt, when heated, chars, but does not melt. When dried, it contains as much as 40.2 per cent.of platinum, thus showing that it must not only be the platinum salt of a base with a very small molecular weight, but also that the base must belong to the class of diamines. Found C = 7-2 ; H = 2.2 ; N = 5.9 ; Pt = 40.2, 40.3, 40.5 per cent. Calculated for C,H,N2,2HCI,PtCl4, C = 7-53 ; H = 2.4 ; N = 5.6 ; The undried salt contained H20 = 3.8. Calculated lH,O = 3.6 per This curious platinum salt was, therefore, evidently the salt of They were analysed. The salt was dried at 100' and analysed. Pt=40*2 per cent. cent. ; also 39.1 per cent. Pt ; calculated, Pt = 39.0 per cent. propinediamine, CH3* $ "H2,H2PtCl,, and this diamine had been pro- H.C *NH, duced by the decomposition of the amidolutidone, C7HIoN,O + 3H,O = C,H,N, -I- 2C,H402. But it is remarkable that such a change should occur, especially as lutidone itself is such a stable pyridine compound, and is not liable to be reconverted into open chain compounds. An attempt was made, but without success, to prepare the free base, propinediamine, fromNOTE ON THE ACTION OF BROMINE ON BEKZENE. 2.11 the platinum salt by precipitating the platinum with sulphuretted hydrogen and subsequently evaporating ; the residue of the hydro- chloride was excessively soluble in water, and when treated with soda gave some ammonia gas, and the solution at once reduced silver nitrate solution, but owing to the small quantity of the platinum salt at our disposal, we were unable to isolate the free base.