TUTIN : SYNTHESES Ih' THE EPINEPHRINE SERIES. PART 11. 2495CCLVIL- Sptlwses in the h''ineph&ae Xeiies. Part 11.The Formation and Pr-operties of Some 2 : 5- and2 : 6-Substituted Pyq-axines and their Convemiorbinto Amino - ke t ones a.nd h i n o - d ike t ones.By FRANK TUTIN.IN a recent communication (Tutin, Caton, and Ham, Trans., 1909,95, 2113) it was shown that the action of ammonia on o-chloro-p-hydroxyacet,ophenone did not result in the formation of o-amino-phydroxyacetophenone, but yielded only resiiious products. Thisresult was considered somewliat, remarkable, inasmuch ::.Y theanalogous chloremp-dihydroxy-ketone readily yields the correspond-ing amine (D.R.-P. 155632). The behaviour of a number o2496 TCTTIN : SYNTHESES IN THE EPINEPHRINE SERIES.PART 11.o-chloroacetophenone derivatives on heating with ammonia hastherefore been invcstigated, with the result that i t has been renderedevident that these compounds may be divided into three classes,according to the products which they yield on this treatment.Thus, w-chloro-mpdihydroxyacetophenone, on treatment withammonia, behaves in a normal manner, yielding the correspondingamine. Only amorphous products result from the interaction ofammonia and o-chloro-p-hydroxy ace top henon e, w-c hloro-o-met hox y-acetophenone, or o-chloro-op-dimethoxyacetophenone. When, how-ever, either w -chloroacetop henon e , w-c hloro-p-methoxy acetophenone,or o-chloro-mp-dimethoxyacetophenone is heated with alcoholicammonia, the principal product of the reaction is a mixture of2 : 5- and 2 : 6-substituted pyrazines, in about equal proportions.The formation of 2 : 5-diphenylpyrazine (11) from o-bromoacetcFphenone and ammonia was studied by Gabriel (Ber., 1908, 41,1127), who showed that, after replacement of the halogen, 3 : 6-di-hydro-2 : 5-diphenylpyrazine (I) was formed, and that this thenunderwent' spontaneous oxidation to the diphenylpyrazine, asfollows :QOPh*CH,\N~I, -+ <CPh*C$>jy -+ N/ CPh'CH,N, .* NH,b H;COPh C H, CP h' \CH.CPh/The last-mentioned author, however, overlooked the fact that2 : 6-diphenylpyrazine is also formed in this reaction, and the modeof production of this compound therefore remains to be explained.Gabriel (Zoc.cit .), however, identified diphenacylamine,(Ph*CO-CH,),NH,as a product of the interaction of o-bromoacetophenone andammonia, and the present author has similarly obtained this base,as a minor product, from o-chloroacetophenone.It is now shown that diphenacylamine (111) and its derivativesare intermediate compounds in the formation of 2 : 6-substitutedpyrazines, f o r they pass into the latter on heating with ammonia.The series of changes which results in the formation of 2 : 6-di-phenylpyrazine (IV) from o-chloroacetophenone and ammonia maytherefore be represented as follows :CE-I,Cl*COPh CH,* COPhNH3 CB,Cl*COPh --+- NH<CH,*COPb -+-H(1.) (11.1(111.)CH:CPh*OH NH, ---f NH< CK:CPh>NH -~N"<C€€:cFh*oH CH:CPhiv. 1y,CH:CPh,N,* \C'H:CPh/(IV.TUTIN : SYNTHESES IN THE EPINEPHRINE SERIES.PART 11. 2497According to this scheme the action of ammonia on diphenacyl-amine first results in the production of 1 : 4-&hydro-2 : 6-diphenyl-p!pa:i?be (V), which then passes into 2 : 6-diphenylpyrazine byspontaneous oxidation. The change might, however, conceivablytake place as follows:(VI.)If this be the case, the intermediate compound will be 3 : 4-di-hydro-2 : 6-dipheqlpyrazine (VI).The change which is here shown to occur on heating compoundsof the type (R*CO*CH,),NH with ammonia does not appear to havebeen observed before, and it therefore seems to afford a new,general method €or the production of 2 : 6-substituted pyrazines.The interaction of ammonia and o-chloro-p-methoxyacetophenoneproceeds similarly t o that of w-chloroacetophenone and ammonia,yielding, as principal products, ppl-dimethoxy-2 : 5-diphenylpyrazine(m.p. 2 2 3 O ) and ppf-dimethoxy-2 : 6-diphenylpyrazine (m. p. 137.5O).The former of these two compounds is of particular interest, as, onfusion, it passes into the “ liquid-crystalline ” state, and this phasepersists over an exceptionally large range of temperature, namely,4 1 ’ 4 O . ppl-D-imethoxy-2 : 5-diphenylpyrazine therefore represents anew addition to the already considerable list of “ liquid-crystalline ”p-anisyl derivatives, but it appears to be the first compound of thisclass in which the anisyl group is attached to a ring. A furtherinteresting property of pp’-dimqthoxy-2 : 5-diphenylpyrazine is thatits solutions exhibit a violet-blue fluorescence, a behaviour whichhas not previously been observed amongst pyrazine derivatives.Furthermore, on the addition of a drop of concentrated hydrochloricor sulphuric acid t o a chlorofcrm solution of this base, a mostbrilliant green fluorescence is produced.ppl-Dimethoxy-2 : 6-di-phenylpyrazine behaves in marked contrast to its 2 : 5-substitutedisomeride, as it fluoresces but slightly, and only in neutral solution,and it does not pass into the “ liquid-crystalline ” state.w-Ch Zoro-mp-dime thoxyac etoph enone yielded minlppl-t e trame t hoxy -2 : 5-diphenylpyrazine (m. p. 208O) and mm1pp’-tetramethoxy-2 : 6-di-phenylpyrazine (m. p. 160O) on treatment with ammonia, neitherof which passes into the “ liquid-crystalline ” state.The formercompound is, however, strongly fluorescent, but only in neutralsolution.It is furthermore shown in the present communication that theseries of changes which result respectively in the formation o2498 TUTIN : SrNTHESES IN THE EPINEPHRINE SERIES. PART If.2 : 5-substituted pyrazines from o-aminoacctophenone or itsderivatives, and in the conversion of diphenacylamine or itsderivatives into 2 I 6-substituted pyrazines, may he reversed bymeans of hydriodic acid. Thus, when 2 : 5diphenylpyrazine isheated with hydriodic acid, reduction followed by hydrolysis occurs,resulting in the formation of two molecules of o-aminoacetophenonehydriodide. Similarly, 2 : 6-diphenylpyrazine, when analogouslytreated, is converted into diphenacy Zarnine hydriodide andammonium iodide.Of course, when employing the pyrazinederivatives containing methoxyl groups, the methyl group is alsoeliminated by the hydriodic acid. This reaction therefore hasafforded a new method of preparing o-amino-p-hydroxyacetophenoneand o-amino-mp-dihydroxyacetophenone, two bases which are ofinterest on account of their physiological activity. The former ofthese bases was previously prepared by the present author in con-junction with Messrs. Caton and Hann (ZOC. cit.) from o-chloro-p-acetoxyacetophenone, whilst the latter base is of special importanceon account of its near relationship t o epinephrine.ppr-Dihydroxydiphenucylamine and mmtppr-tetrahydroxy&phen-ncylamine have been prepared by the action of hydriodic acid onthe previously-mentioned methoxy-2 : 6-diphenylpyrazines.It willreadily be seen from a comparison of the formula: given below thatppl-dihydroxydiphenacylamine (VII) and, especially, mm’ppr-tetra-hydroxydiphenacylamine (VIII) are closely related to the ketonederived from epinephrine (IX), as also to the above-mentioned twoo-aminohydroxyacetophenones :HO/-\CO*CH;NH.CH,*CO/-\OH \-/ \-/( v I r .HO OH(VI TI.)KOHO/-\CO~CH,~NH*CI-I,. \-/(IX.)It was therefore to be expected that these two diphenacylaminederivatives would be possessed of physiological activity, and theirpropert‘ies have accordingly been investigated in the WellcomePhysiological Research Laboratories by Dr. H. H. Dale, to whomthe author is indebted for the following and the subsequentlymentioned physiological experiments.It was found that each ofthese compounds, in the form of salts, when injected intravenouslyinto cats, caused a rise in blood-pressure, pp’-dihydroxydiphenacylTUTIN : SYNTHESES IN THE EPINEPHRINE SERIES. PART 11. 2499aniine (VII) having an action similar to that of the related com-pound, w-amino-p-hydroxyacetophenone (Tutin, Caton, and Ham,Zoc. cit.), whilst the corresponding tetrahydroxy-base (VIII) had agreater activity, more resembling that of the ketone derived fromepinephrine (IX).It has already been mentioned that o-chloro-o-methoxyaceto-phenone and w-chloro-opdimethoxyacetophenone yield only amor-phous products when heated with ammonia, whereas the analogouscompounds containing the methoxyl groups in the rn- and p-positionsreadily yield substituted pyrazines.It theref ore appears that thepresence of a methoxyl group in the o-position with respect to theside-chain precludes the formatim of pyrazines from w-chloroaceto-phenone derivatives, although the reason for this is not apparent.On account of the above-mentioned property of the o-sub-stituted w-chloroacetophenone derivatives here described, it wasimpossible to obtain from them the corresponding o-aminohydroxy-acetophenones in the way which has already been noted in connexionwith the preparation of o-amino-p-hydroxyacetophenone fromw-chloro-p-methoxyacetophenone. Recourse was therefore had tothe use of potassium phthalimide, and by this means derivativesand salts of o-amino-o-hydroxyacetophenone and o-amino-op-di-hydroxyacetophenone have been obtained.When examinedphysiologically, the hydriodide of the o-hydroxy-base was found tobe practically inactive, whilst the corresponding salt of the op-di-hydroxy-base had no greater activity than the analogous p-hydroxy-compound. It is therefore seen that hydroxyl groups in theo-position with respect to the side-chain are devoid of physiologicalactivity in the class of compounds under consideration, a resultwhich is in harmony with a previous observation of Dr. Dale, whofound o-h ydr ox y-fl-p hen ylet h y lamine, /-\CH,*CH2*NH,, to beinert, whilst the analogous pcompound is strongly active (Barger,Trans., 1909, 95, 1123).o-Chloro-o-methoxyacetophenone is formed, together with thecorresponding p-compound, by the action of aluminium chloride onchloroacetyl chloride and anisole. The further action of aluminiumchloride on o-chloro-o-methoxyacetophenone results in the formationof w-chloro-o-hydroxyacetophenone.The latter substance differsfrom the corresponding p-compound, inasmuch as it is quite insolublein aqueous sodium carbonate, thus showing how the relative positionsof the groups in the benzene nucleus affect the acidity of thehydroxyl group.The above-mentioned substituted o-aminoacetophenones, contain-ing a hydroxyl group in the 0-position with respect to the side-chain,OH\-2500 TUTIW : SYNTHESES IN THE EPINEPHRIKE SERIES. PART 11.differ markedly in their properties from the previously-mentionedanalogous compounds which are substituted in the m- or p-position,inasmuch as they condense and oxidise, when dissolved in neutralsolvents, to form 2 : 5-substituted pyrazines.oo'-Dihydroq-2 : 5-dipheltylpprazine and oo'ppt-tetrahydroxy-2 : 5-&iphenylpyraainehave thus been prepared.The two o-substituted o-aminoacetophenones described also showa singular behaviour when benzoylated, either by the Schotten-Baumann method or in pyridine solution, for, when thus treated,they yield benzoyl derivatives of internal anhydrides. It wouldappear possible that these condensation products are 1-b enzoyt-indoxyl aad 6-benzoyloxy-1-benzoyLindoxy1 respectively.Gabriel (loc. cit.), from his work on w-aminoacetophenone, con-cluded that a-amino-ketones of this type were incapable of existencein the free state, but always underwent condensation when liberatedfrom their salts. It is evident, however, from the results given inthe present paper that this is not invariably the case.Thus,o-aminoacetophenone, o-amino-p-methoxyacetophenone, w-amino-mp-dimethoxyacetophenone, and w-amino-o-hydroxyacetophenonecondense spontaneously, yielding pyrazine derivatives in the mannershown by Gabriel. o-Amino-phydroxyacetophenone and o-amino-mp-dihydroxyacetophenone, on the other hand, cannot be causedto condense ; whilst o-amino-op-dihydroxyacetophenone possesses pro-perties between those of these two groups, for it can be obtainedin the free state, although it condenses somewhat readily.EXPERIMENTAL.Interaction of w-Chloroacetophenone and Ammonia.Chloroacetyl chloride was dissolved in an excess of benzene, andone molecular proportion of aluminium chloride added.A violentreaction ensued, and, when this had subsided, ice and hydrochloricacid were added. The aqueous layer was then separated, and, afterwashing the benzene solution with water, the greater part of thesolvent was removed from it,. On adding light petroleum t o theconcentrated liquid thus obtained, o-chloroacetophenone separatedin glistening plates, melting at 59O. The yield was nearlyquantitative.Fifteen grams of o-chloroacetophenone were heated for one anda-half hours at looo in sealed tubes with an excess of alcoholicammonia.After allowing the contents of the tubes to cool, thesolid which had separated was collected, washed with alcohol, andthen extracted repeatedly with boiling xylene. The materialundissolved by this treatment consisted entirely of ammoniumchloride, but on concentrating the xylene extracts, a compounTUTIN: SYNTEESES IN THE EPINEPHRINE SERIES. PART 11. 2501separated in plates, melting at 194O. As thus obtained, this sub-stance possessed a dark bluish-green colour, and was only obtainedcolourless after being treated, in acetic acid solution, with a smallamount of potassium permanganate dissolved in t.he same solvent.When crystallised from xylene after this treatment, it formed large,colourless plates, melting at 194O, and was identified as 2 : 5-diphenyl-pyrazine (Found, C = 79.5 ; H = 5.1.Calc., C = 79.3 ; H = 5.0 percent.)This compound was first prepared by Staedel and Rugheimer(Ber., 1876, 9, 563), who described it under the name of ‘‘ isoindol.”As subsequently obtained by Staedel and Kleinschmidt (ibid.,1880, 13, 836), it was observed to exhibit diverse colours, and theyregarded it as being “ idiochromatic.” Pure 2 : 5-diphenylpyrazineis, however, quits colourless, as ha? been shown by Gabriel (Ber.,1908, 41, ll27), who prepared it by the interaction of w-bromo-acetophenone and ammonia.The original alcoholic filtrate from the 2 : 5-diphenylpyrazine andammonium chloride was evaporated to a low bulk and largelydiluted with benzene. The filtered liquid was then again evaporatedas far as possible, and the residue dissolved in alcoholic hydrogenchloride, when the mixture rapidly became dark brown, but noblue colour was developed (see below).The solution was con-centrated somewhat, and hot ethyl acetate added, when, on coolingthe mixture, a crystalline substance separated in needles, whichwere collected and washed with a mixture of ethyl acetate andalcoholic hydrogen chloride. The product so obtained was dissolvedin the minimum amount of absolute alcohol, and a little alcoholichydrogen chloride added, when it immediately separated in soft,almost colourless needles, melting at about 189O :0.2020 gave 0.1060 AgCl. C1= 13.0.C16HlzNz,HCl requires C1= 13.2 per cent.This substance was identified as 2 : 6-diphenylpyrazine rnono-hydrochZode,* since it yielded 2 : 6-diphenylpyrazine, which formedcolourless needles, melting at 90°.(Found, C= 79-3 ; H = 5.2.Calc., C=79-3; H=5*0 per cent.)2 : 6-Diphenylpyrazine monohydrochloride is almost insoluble inbenzene or ethyl acetate, but it dissolves fairly readily in alcohol,owing to the fact that it becomes, for the most part, dissociated.It is not stable in moist air, and is instantly dissociated whenbrought in contact with water.Gabriel (loc. cit.) did not note the formation of 2 : 6-diphenyl-pyrazine when he investigated the interaction of o-bromoaceto-* It has been found that the pyrazines are tliacidic bases, and yield two series ofsalts (compare following paper).VOL. XCYlI. 8 2502 T’UTIN : SYNTHESES IX THE EPINEPHRINE SERIES.PART 11,phenone and ammonia, but it would appear certain that it musthave been present in the reaction mixture examined by him.The original filtrate from the 2 : 6-diphenylpyrazine hydrochloridewas dark brown, and contained considerable resinous matter. Itwas largely diluted with water, filtered from the precipitated resin,concentrated somewhat, and treated with animal charcoal. Onallowing the clear liquid to cool, a somewhat sparingly solublecompound separated, which melted at about 235O, and was sub-sequently identified as diphenacylamine hydrochloride, a compoundwhich has been described by Gabriel (Zoc. cit.).In a subsequent preparation of the above-described 2: 5- and2 : 6-diphenylpyrazines, a quantity (40 grams) of o-chloroaceto-phenone was heated in an autoclave with an excess of alcoholicammonia, the mixture being subsequently kept for fourteen daysbefore it was worked up.After separating the ammonium chlorideand 2 : 5-diphenylpyrazine in the manner already described, theresidual solution containing the 2 : 6-base, which was of a muchmore pronounced red colour, and appeared to be freer from resinousmatter than that obtained in the previous preparation, was mixedwith a large volume of ether and extracted several times with amixture of concentrated hydrochloric acid (1 part) and water(2 parts). This caused the separation of some brown, resinousmatter, which was removed. The ethereal liquid was thenevaporated, and the red residue dissolved in absolute alcohol, anda solution of hydrogen chloride in the same solvent added.Theliquid then became deep blue, and, on cooling the mixture afteradding some ethyl acetate, a solid separated, which, when collected,was seen to consist of a mixture of white and deep blue needles, theformer predominating. The separation of these two products wastedious, but was eventually effected by taking advantage of thefact that the blue hpdrochloride was somewhat more sparinglysoluble in a boiling solution of hydrogen chloride in absolutealcohol than was the white one, which consisted of 2 : 6-diphenyl-pyrazine hydrochloride. The blue compound crystallised in smallneedles, which had no definite melting point, and were only stablein dry air or in an anhydrous solvent in the presence of a moderateexcess of hydrogen chloride.The amount obtained was only about0-5 gram, and consequently the formula could not be established :0-3506 gave 0-4791 AgCl. C1=33*8 per cenb.The base obtained from this deep blue hydrochloride crystallisedfrom alcohol in small tufts of brilliant scarlet crystals, melting at195O. It was readily soluble in chloroform, ethyl acetate, orbenzene, but only moderately so in alcohol. On exposing a solutionof this scarlet-coloured base in chloroform or benzene to direcTUTIN : SYNTHESES IN THE EPINEPBRINE SERIES. PART 11, 2503sunlight, the colour was discharged in half-an-hour, a compoundcrystallising in yellow needles being formed.Preparation of w-Chloro-0- and -p-methoxyacetophenones.o-C hlorep-met hoxyacetophenone was prepared by Eunckell andJohannssen (Ber., 1897, 30, 1715; 1898,31, 170) by the interactionof anisole and chloroacetyl chloride in the presence of aluminiumchloride.Mr. F. W. Caton, who conducted this operation for thepresent author, found it important to avoid the use of any excessof aluminium chloride and not to employ heat, as the methyl groupis very easily eliminated. With the object of avoiding this hydro-lysis, experiments were made with the use of sublimed ferric chloride,but the yield of condensed product so obtained was only small.One molecular proportion of anisole was mixed with rather morethan an equivalent amount of chloroacetyl chloride, and, afterdiluting the mixture with three times its volume of carbondisulphide, one molecular proportion of powdered aluminiumchloride was cautiously added, the flask being kept cool during thisoperation.After three hours the carbon disulphide was decanted,the residue being decomposed with ice and hydrochloric acid andthe product extracted with ether. The ethereal liquid was thenshaken with aqueous sodium hydroxide, which removed smallamounts of hydrolysed product and red resin, after which the solventwas evaporated. On fractionally crystallising the residue fromalcohol, it was found to consist, for the most part, of w-chloro-p-methoxyacetophenone (m. p. lOZ0), which formed long needles,but the more soluble fraction contained a second substance. Thiscompound formed large, colourless, diamond-shaped plates, which,after being separated mechanically from the greater part of thep-compound and submitted to several recrystallisations, melted at6 9 O :0.2154 gave 0'4571 CO, and 0.0993 H,O.0'2288 ,, 0.1778 AgCl.C1=19*2.CQH,O,C1 requires C = 58.5 ; H = 4'9 ; C1= 19'2 per cent.This substance was evidently o-chloro-o-m ethoxyacet ophenonte,since it readily yielded salicylic acid on fusion with potassiumhydroxide.This appears to be the first time that the formation of an 0-mono-substituted ketone by means of the Friedel and Crafts' reactionhas been noted, although phenyl 0-tolyl ketone has been stated tobe formed by ?,he interaction of toluene and benzoic acid in thepresence of phosphoric oxide (Kollarits and Merz, Ber., 1873, 6,538).o-Chloro-o-methoxyacetophenone is slightly volatile at theC=57*9; H=5'1.8 A 2504 TUTIN : SYNTHESES IN THE EPINEPHRINE SERIES PART 11.ordinary temperature, and sublimes readily on heating.It is morevolatile in steam than is the corresponding p-compound, and maybe approximately separated from the latter by taking advantage ofthis property. When brought into contact with the skin, it causesconsiderable smarting, and it has an extremely irritant action onthe eyes.Attempts to prepare o-met hoxydip hen ylpyrazines by heatingw -c hlor 0-0-me thoxy acet o phenone with alcoholic ammonia in sealedtubes resulted only in the formation of resinous products.o-Chloro-o-h ydroxyacet ophenone.o-Chloro-o-methoxyacetophenone was dissolved in carbon di-sulphide, one molecular proportion of powdered aluminium chlorideadded, and the mixture heated for two hours under a reflux con-denser.The solvent was then removed, and the residue heated at looofor ten minutes, after which ice and hydrochloric acid were added,and the product extracted with ether. On shaking the etherealliquid with a solution of sodium carbonate, nothing was removed,but subsequent treatment with aqueous sodium hydroxide extracteda relatively small proportion of oily matter. The ethereal liquid, onevaporation, yielded a considerable quantity of unchanged u-chloro-o-methoxyacetophenone, this compound being evidently much morestable towards aluminium chloride than is the correspondingpderivative. The oil which had been removed by sodium hydroxidewas dissolved in ether, and light petroleum added, which caused theseparation of a viscid, red product, whereupon the mixture wasshaken with animal charcoal, and filtered.After concentrating thefiltrate, a substance separated in yellow, flattened needles, which,after recrystallisation from alcohol, melted at 101O :0.1238 gave 0.2546 CO, and 0.0490 H20.This substance was therefore o-chloro-o-h ydrox yace toph enone,HO*C6H,-CO*CH,C1. It differed from the corresponding p-com-pound in being insoluble in aqueous sodium carbonate (compareTutin, Caton, and Hann, Trans., 1909, 95, 2118).C=56.0; H=4*4.C,H,O,Cl requires C = 56.3 ; H =4*3 per cent.Intemction of w-Chloro-p-methoxyacetophenone and Am.mnia.o-Chloro-pmethoxyacetophenone was heated in an autoclave forthree hours at l l O o with a large excess of alcoholic ammonia.Whencool, the solid contained in the dark-coloured reaction mixture wascollected and washed, first with alcohol, and subsequently with water.The residue was crystallised from xglene, when it separated inlarge leaflets, melting at 2 2 2 O . The substance, as thus obtainedTUTIN : SYNTHESES IN THE EPINEPHRINE SERIES. PART 11. 2505could not be rendered colourless by recrystallisation, but differentpreparations of it exhibited diverse tints, such as dull green,purplish, or greenish-yellow. It was, however, rendered colourlessby the means previously found useful in the case of 2 : 5-diphenyl-pyrazine (p. 2501), but the melting point was practically unchangedby this treatment.On crystallising the purified substance fromglacial acetic acid or xylene, it formed large, coiourless leaflets, butwhen crystallised from chloroform or ethyl acetate it separated inhexagonal plates :0.1088 gave 0.2947 CO, and 0.0556 H20.0.3246A molecular-weight determination by the cryoscopic method gave0.3153, in 33.2 of phenol, gave At = - 0*30°.C18H,,02N2 requires M.W. = 292.Several attempts were made to estimate the number of methoxylgroups in this substance by Perkin’s modification of Zeisel’s method,but accurate results could not at first be obtained, owing to thegreat stability of the compound. It wits eventually ascertained,however, that the methyl groups axe rapidly eliminated if someglacial acetic acid be added to the hydriodic acid employed:C = 73.9 ; H = 5.7.N=9.8.,, 29.0 C.C. N2 (moist) at 20° and 728 mm.C,,Hl,O,N, requires C = 73.9 ; H = 5.5 ; N = 9.6 per cent.the following result :M.W. =243.0.2096 gave 0.3366 AgI. OMe=21.1.Cl,Hl,N,(OMe), requires OMe = 21.2 per cent.The compound was evidently ppl-dimethoxy-2 : 5-dip72.e~l-pgraaine, CkH,N,(C~H4*OMe),, and its constitution was subsequentlyconfirmed by its conversion by hydriodic acid into oamino-p-hydroxyacetop?Lenone hydriodide and methyl iodide (p. 2520).On heating pp’-dimethoxy-2 : 5-diphenylpyrazine, fusion occurs at223O, and the substance passes into a (‘ liquid-crystalline ” state.This phase persists until a te.mperature of 265’4O is reached, whenthe “ crystalline ” liquid phase instantly passes into the normalliquid state.A t the point of change it can easily be observed thatthe two liquid phases are immiscible, and the “ liquid-crystalline ”product appears to possess the greater density. The reverse change,from the normal liquid to the “ liquid-crystalline ” phase, occursat precisely the same temperature, and is exhibited in a strikingmanner when viewed through crossed Nicol’s prisms. The pointof change from the “ liquid-crystalline ” to the normal liquid phase,and vice versa, of pp’-dimethoxy-2 : 5-diphenylpyrazine is a muchmore delicate criterion of the purity of this substance than is itsmelting point, as a mere trace of impurity causes a very appreciablelowering of the temperature of transition from one liquid phas2506 TUTIN : SYNTRESES IN THE EPINEPHRINE SERIES.PART 11,to the o t h r , whilst an amount of extraneous substance sufficientto cause a depression of the melting point by about 3O completelyextinguishes the ‘‘ liquid-crystalline ” phase.pp’-Dimethoxy-2 : 5-diphenylpyrazine is practically insoluble inether or alcohol, very sparingly soluble in chloroform, benzene, orethyl acetate, moderately soluble in boiling xylene, and more readilyso in glacial acetic acid. Its dilute solution in chloroform exhibitsa violebblue fluorescence, and when a drop of concentrated hydro-chloric acid is added, a yellow colour is produced, accompanied by amost brilliant green fluorescence.The original, dark-coloured, alcoholic filtrate from the ammoniumchloride and pp’-dimethoxy-Z : 5-diphenylpyrazine was evaporatedto dryness, the residue extracted with benzene, the solutionevaporated, and the residue dissolved in absolute alcohol.A solu-tion of hydrogen chloride in absolute alcohol was then added, when,after concentrating the solution, it was mixed with hot ethyl acetate.On cooling the mixture, a compound separated in yellow needles,which were collected, washed with a mixture of alcoholic hydrogenchloride and ethyl acetate, and recrystallised by dissolving themin absolute alcohol, adding alcoholic hydrogen chloride, con-centrating the mixture, and then diluting it with ethyl acetate.Soft, yellow needles were thus obtained, which melted at about0.2420 gave 0.1003 AgC1.C1= 10.3.c,sH,60,N2,Hc1 requires c1= 10.8 per cent.This salt proved to be ppl-dimethoxy-2 : 6-diphenylpyrazine mono-hydrochloride, C4H,N2(C6H,*OMe),,HC1. It dissolves sparingly inethyl acetate or chloroform containing an excess of hydrogenchloride, but is unstable in moist air, and is dissociated by alcoholor water.C,R2N2 (C6H,- OMe),, ob-tained from the above-described salt by treatment with water oralcohol, crystallised from the latter solvent in colourless needles,melting at 137.5O:178-180’ :pp ‘-Dim e t ho x y-2 : 6-d iph en y l p y razine,0.0987 gave 0.2670 CO, and 0-0505 H,O.C,8H,g02N, requires C = 73.9 ; H = 5-5 per cent.ppr-Dimethoxy-2 : 6-diphenylpyrazine is very readily soluble inchloroform, ethyl acetate, benzene, or xylene, but only moderatelyso in alcohol.Its neutral solutions exhibit a slight blue fluorescence,but this is destroyed by the addition of concentrated hydrochloricacid. It does not pass into a “liquid-crystalline” state on fusion.The constitution of pp’-dimethoxy-2 : 6-diphenylpyrazine was subse-quently proved by its conversion by means of hydriodic acid intoC=73*8; H=5.7TUTIN : SYNTEESES IN THE EPINEPHXINE SERIES. PART 11. 2507methyl iodide, ammonium iodide, and pp1-dihydroxydiphenacyl-amine hydriodi.de (p. 2522).The filtrate from the crude ppl-dimethoxy-2 : 6-diphenylpyrazinehydrochloride was dark brown, and contained considerable resinousmatter. It was digested with aqueous hydrochloric acid, filtered,and the filtrate treated with animal charcoal.After concentratingthe liquid thus obtained, it deposited a relatively small amount ofa sparingly soluble hydrochloride. This was recrystallised fromwater, when it formed leaflets, melting at 256O:0.2459 gave 0.1003 AgCl. C1=10*0.This salt was doubtless ppl-dimethoxyd@&enacyZamine hydro-chloride, (MeO*C,H,-CO*CH,),NH,HCl, as it was obtained in amanner analogous to that which resulted in the formation ofdiphenacylamine hydrochloride from w-chloroacetophenone, and itsproperties are strictly analopus to those of the latter salt. More-over, from evidence given in-the latter part of this communication,it is evident that ppl-dimethoxydiphenacylamine must have beenformed during the interaction of ammonia and w-chloro-p-methoxy-acetophenone, since the former base is an intermediate compoundin the production of the above-described ppl-dimethoxy-2 : 6-di-phenylpyrazine.It has already been shown in connexion with the preparation ofthe 2 : 5- and 2 : 6-diphenylpyrazines that if the reaction mixturewere kept for some time before it was worked up, a highly-coloured\y-product was formed, together with these bases.This is also theGse when working with the pmethoxy-derivatives, but in the latterintance several other compounds were also obtained in smallamunh, possibly owing to the fact that the reaction mixture wasexaLined much more fully than in the former case.o-(hloro-p-methoxyacetophenone was heated with alcoholicammohia as above described, but the reaction mixture was kept forthree \eeks before being examined.The ppl-dimethoxy-2 : 5-di-phenylpyazine was isolated as before described, but with the useof chloroorm instead of xylene. The mother liquors then yieldeda small abount of a compound, which formed soft, colourlessneedles, meking at 232-233O. On working up the original filtratefrom the MI-dimethoxy-2 : 5-diphenylpyrazine and ammoniumchloride in tb manner previously described, a mixture of pp'-di-methoxy-2 : 6-dphenylpyrazine and another salt was obtained. Thelatter compound was evidently the p-methoxy-derivative of the bluehydrochloride pnviously described ; it was dark green, and wasseparated from tb3 salt of the pyrazine derivative in a, manneranalogous t o that employed is connexion with the previously-C18H,,0,N,HC1 requires C1= 10.1 per cent2508 TUTlN : SYNTHESES IN THE EPINEPHRINE SERIES.PART IT.described blue compound. The mother liquors from these liydroechlorides yielded, together with traces of other compounds, a sub-stance which formed yellow leaflets, melting at 213--214O, but didnot fluoresce when treated in chloroform solution with hydrochloricacid. The deep green-coloured hydrochloride melted quite in-definitely, owing to dissociation, and this change was also readilybrought about by treatment with any solvent which did not containan excess of anhydrous hydrogen chloride. It yielded a deepcrimson-coloured base, crystallising from alcohol in small, lustrousprisms, which were so dark red as to awear almost black, andmelted at about 165O.This compound, like the corresponding phenylderivative previously described, is decolorised by exposure to directsunlight when dissolved, yielding a yellow substance, which formedneedles (m. p. about 255O) from xylene. The amounts of thesevarious by-products obtained was small, and their investigation wasnot further pursued.Derivatives of o-Ami~to-p-methoxyacetop7Lertone.It would appear that the above-described ppf-dimethoxy-2 : 5-di-phenylpyrazine must have been formed by the condensation of twomolecules of w-amino-p-methoxyacetophenone, followed by spon-taneous oxidation of the resulting ppf-dimethoxy-3 : 6-dihpdro-2 : 5-diphenylpyrazine in a, manner analogous to that which hasbeen shown by Gabriel (Zoc. cit.) to result in the formation of2 : 5-diphenylpyrazine from o-aminoacetophenone.With the object,theref ore, of verifying this conclusion, o-amino-pmethoxyacetcphenone was prepared, in the form of its hydrochloride, as folio+.o-Chloro-p-methoxyacetophenone was heated for some time i T anickel crucible with rather more than one molecular propopionof potassium phthalimide. The reaction mixture was then extflctedwith boiling xylene, and the product which crystallised fr@ thissolvent after concentration was repeatedly boiled witl- largequantities of water for the removal of unchanged phthalimde. Onrecrystallising the residue from xylene, glistening leaflets melting.at 164--165O, were obtained :0.1437 gave 0.3650 CO, and 0.0604 H2,0.o-Phthalimino-p-methoxyacetophenone,C = 69-2 ; P= 4.6.C,,HI3O,N requires C = 69.2 ; H =4*4 per cezt.MeO* C,€€4*CO*CH,*N<CO>C,H co 4,is very sparingly soluble in alcoho'l, ethyl acetat, or chloroform,but dissolves more readily in glacial acetic acid 01 boiling xylene.The above-described phthalide derivative w*s boiled for eighthours with concentrated hydrochloric acid, whea it gradually passeTUTIN : SYNTHESES I N THE EPINEPHRINE SERIES.P A R T 11. 2509into solution. The mixture was then deprived of phthalic acid bymeans of ether, and evaporated to dryness under diminishedpressure. On crystallising the residue from alcohol, o-amino-p-methoxyacetophenone hydrochloride,MeO*C6H4* C'O*C'H,*NH,,HCl,was obtained in small, colourless prisms, which melted and decom-posed at 204O, after having become red:0.2121 gave 0.1408 AgCl.C1=16'4.C,H,,O,N,HCl requires C1= 16.6 per cent.When an alkali is added to an aquepus solution of o-aminep-methoxyacetophenone hydrochloride, no immediate separation ofbase occurs. The mixture, however, rapidly darkens somewhat, and,after some time, a dark-coloured, semi-crystalline product separates.On purification, this yielded pp/-dimethoxy-2 : 5-diphenylpyrazine(m. p. 223O), thus proving that a change analogous to that observedby Gabriel (Zoc. cit.) had occurred.o -A mino-p-me t ho xy ace to ph enone Pla tinichlorid e,(Me0 *C6~4*C;O*CH,oNH2),H2PtCl,.--This derivative crystallised very readily in deep yellow leaflets,and melted and decomposed at 225-228O:0.1434 gave 0.0373 Pt.o-Amino-p-methoxyacetophenone Aurichloride,Pt =26-0.(C,H,102N)2H,PtCI, requires Pt = 26.3 per cent.MeO*C,H4*CO*CH2*NH2,HAuC14.-The aurichloride did not crystallise readily, but was eventuallyobtained in handsome, golden-coloured leaflets, which melted at74O, and evidently contained water of crystallisation :0.2024 gave 0.0762 Au.CgH1102N,HAuC14,H20 requires Au = 37.6 per cent.o - Amino - p - methoxyacetophenone picrate, C9Hl1O,N,C,H,O7N3,formed small, bright yellow leaflets, which? like the preceding com-pound, contained water of crystallisation.It melted and decomposedat 185O.The mercum'chlom'de crystallised very readily in long, colourlessneedles, which melted at 171O.Au = 37.6.o-Chloro-mp-dimetioxyacetophenone.Catechol was methylated by means of methyl sulphate," and theThe veratrole was then resulting veratrole purified by distillation.* Perkiii and Weizmann (Trans., 1906, 89, 1649) state that an almost quanti-tative yield of veratrole may be obtained by treating 100 grams of catechol with75 grams of methyl sulphate and 150 grams of potassium hydroxide.The figure2510 TUTIN : SYNTHESES IN THE EPINEPHRlNE SERIES. PART 11.dissolved in carbon disulphide, an equivalent amount of chloroacetylchloride added, and then one molecular proportion of powderedaluminium chloride introduced. The mixture was heated on awater-bath for two hours, but the reaction which ensued was byno means violent. The carbon disulphide was then removed andthe residue decomposed by ice and hydrochloric acid, the productbeing extracted with ether.On shaking the ethereal liquid withaqueous potassium hydroxide, a small quantity of demethylatedproduct was removed. The ether was then evaporated, and theresidue deprived of a fairly large proportion of unchanged veratroleby means of steam. The non-volatile product was crystallised fro-&alcohol, when it yielded o-chloro-mp-dimethoxyacetophenone,(MeO),C,H,*CO-CH,~Ci, which formed small, colourless prisms,melting at 101O:0.1172 gave 0.2394 CO, and 0.0538 H,O.o-Chloro-mp-dimethoxyacetophenone is moderately soluble inalcohol, but much more readily so in ethyl acetate or chloroform.When in the dry state, it occasions violent sneezing.C=55*8; H=5.1.C1,HllO3C1 requires C = 55-9 ; H = 5.1 per cent.The Imt eraction of o-Chloro-mp-dimethoxyacetophenone andA mmonia.o-Chloro-mp-dimethoxyacetophenone was heated in an autoclavefor three hours at llOo with a large excess of absolute alcoholicammonia.When cool, the solid contained in the reaction mixturewas collected, washed with alcohol, and then extracted many timeswith boiling xylene. The xylene extracts, on cooling, deposited adark red, crystalline powder, melting at 208O. After treatmentwith a small amount of potassium permanganate in glacial aceticacid solution, in the manner previously described, it sep-arated fromglacial acetic acid in light grey needles, which melted at the sametemperature as before this treatment:0.1130 gave 0.2834 CO, and 0-0590 H20.C20HdO4N2 requires C = 68.2 ; 33 = 5.7 per cent.This compound was evidently mmrppr-t etramethoxy-2 : 5-dipherqLpyrazi'lte, C,H4(OMe),*C4H2N2*C,H3(OMe)2, and its constitutionwas subsequently confirmed by its conversion into o-amino-mp-dikhydroxgacetophenone hydriodide and methyl iodide by the action ofhydriodic acid (p.2520). It' is insoluble, or nearly so, in all the usualsolvents with the exception of glacial acetic acid and boiling xylene,and in the latter solvent it dissolves but sparingly. Its much greatergiven are, however, obviously incorrect, since the amount of catechol mentionedwould require theoretically 229 grams of methyl sulphnte and 102 grsnis of thealkali.C=68*4; H=5*8TUTIN : SYNTHESES IN THE EPINEPHRINE SERIES.PART 11. 2511solubility in glacial acetic acid than in any other liquid employedappeared to be due to salt formation, as the solution was orange-yellow, and it was subsequently found that the tetramethoxy-diphenylpyrazines are more strongly basic than the other compoundsof this class described in the present communication. A very dilutesolution of mmfpp/-tetramethoxy-Z : 5-diphenylpyrazine in chloro-form exhibits a strong blue fluorescence, but’ this phenomenondisappears on the addition of a drop of concentrated hydrochloricacid, a non-fluorescent, deep yellow liquid being produced. Onfusion, this pyrazine derivative does not pass into a, “liquid-crystalline” state, as is the case with the corresponding ppl-di-methoxy-compound.The original alcoholic filtrate from the mmlppf-tetramethoxy-2 : 5-diphenylpyrazine and ammonium chloride was evaporated todryness, the residue extracted with benzene, the solution againevaporated, and the product so obtained dissolved in a small amountof absolute alcohol and a solution of hydrogen chloride in the samesolvent added.On cooling the dark brown mixture, a compoundseparated in deep yellow needles. This was collected, washed withalcoholic hydrogen chloride, and recrystallised from absolute alcoholby the addition of a solution of hydrogen chloride in this solvent,when long, deep yellow, soft needles were obtained, which melted atabout195-200°:0.2030 gave 0.0730 AgC1. C1=8-9.C,,H,,04N2,HC1 requires C1= 9.1 per cent.This compound was mm1ppf-tetramethoxy-2 : 6-diphenylpyraainemonohydrochloride, C6H,(OMe),~C4H2~-C,H,(OMe),,HC1.It wasreadily dissociated by water, or by alcohol, unless the latter con-tained an excess of hydrogen chloride. It yielded mmlppl-tetra-metAoxy-2 : 6-diphenyZpyra&e, which, when crystallised fromalcohol, formed long, almost colourless needles, melting at 160° :0.1079 gave 0.2734 CO, and 0.0563 H,O.CmH2,O4N2 requires C = 68.2 ; H = 5-7 per cent.This base was rather sparingly soluble in alcohol, but readily soin benzene, xylene, chloroform, glacial acetic acid, or ethyl acetate,and differed from the corresponding 2: 5-compound, inasmuch asits solutions exhibited no fluorescence. Its constitution was sub-sequently confirmed by its conversion into mm’ppl-tetra~~drozy-diphenucylanzline Iqdhodide, methyl iodide, and ammonium iodideby means of hydridic acid (p.2523).C=68-0; €€=5*82512 TUTIN: SYNTHESES IN THE EPlNEPHRINE SERIES. PART 11.w -Chlo ro-op-dime t hoxyac e tophenone.Resorcinol dimethyl ether was prepared from resorcinol by theaction of methyl sulphate and potassium hydroxide, and purifiedby distillation. The dimethyl ether was then dissolved in carbondisulphide, and the requisite amount of chloroacetyl chloride added.One molecular proportion of powdered aluminium chloride witsthen introduced, when a violent reaction ensued. After removingthe solvent, the residue was treated with ice and hydrochloric acid,the resulting solid being collected and crystallised from .alcohol.A very good yield of small, colourless, prismatic needles, melting at96O, was thus obtained:0.1233 gave 0.2518 CO, and 0.0577 H,O.C1,H1,0,C1 requires C = 55.9 ; H = 5.1 per cent.w -Chloro-op-dime t ?LOX yace t ophenone, ( Me0)2C6H3*C'O-CH2Cl, issomewhat sparingly soluble in alcohol, but much more readily soin ethyl acetate or chloroform.It was formed in much better yieldthan the corresponding mp-compound.Attempts to convert w-chloro-opdimethoxyacetophenone intopyrazine derivatives by heating with alcoholic ammonia resultedonly in the formation of brown resins, just as was the case whenw-chloreo-methoxyacetophenone was employed. It therefore appearsthat the presence of a methoxyl group in the ortho-position withrespect to the side-chain precludes the formation of substitutedpyrazines from w-chloroacetophenone derivatives.Attempts were made to prepare an w-chlorotrimethoxyaceto-phenone by the interaction of chloroacetyl chloride and pyrogalloltrimethyl ether, but without success.C=55.7; H=5.2.w -A mino-op-dih ydro xyace t o p h enone and i t s D erivu tives.It is subsequently shown that the methoxy-2 : 5-diphenylpyrazinesreadily yield w-aminohydroxyacetophenones, the formation of whichwas the primary object of this research.Since, however, nopyrazine derivative could be obtained from w-chloro-opdimethoxy-acetophenone, other means had to be devised for the conversion ofthis compound into the desired dihydroxy-mine.w-Chloro-op-dimethoxyacetophenone was heated in a nickelcrucible at about 160° with rather more than one molecular pro-portion of pot.assium phthalimide until the reaction mixture, whichwas at first fairly fluid, became almost solid.The product w mthen extracted several times with boiling xylem, and the combinedfiltered liquids concentrated to a small bulk. The product whichseparated on cooling wils collected and repeatedly boiled with largTUTIN : SYNTHESES IN THE EPINEPHRINE SERIES. PART Ir. 2513quantities of water until free from phthalimide, after which it wa,srecrystallised from xylene o r glacial acetic acid, when it formedacicular crystals, melting at 188O :0.0903 gave 0.2209 CO, and 0.0394 H,O.w-Phthalimino-op-dimethoxyacet ophemne,C= 66.7 ; H =4%CI8H,,O,N requires C = 66.5 ; H = 4.6 per cent.is insoluble, or very sparingly soluble, in all the usual solvents, withthe exception of glacial acetic acid and boiling xylene, in which itis moderately soluble.The above-described phthalide derivative was boiled with con-centrated hydriodic acid containing some glacial acetic acid, whenit very gradually passed into solution.The mixture was thendiluted with water, and repeatedly extracted with ether for theremoval of the phthalic acid, after which it was evaporated t odryness under diminished pressure. The solid residue was thendissolved in alcohol, the solution concentrated, ethyl acetate added,and the mixture again evaporated somewhat, when o-amino-op-di-hydroxyacetophenone hydriodide, C6H,(OH),*CO*CH,*NH,,HI,separated from the boiIing mixture :0.2196 gave 0.1733 AgI.I = 42.7.0'4535 ,, 0.3405 AgI. I=42*8.o-Amino-opdihydroxyacetophenone hydriodide forms nearlyIt is readily solubleC8Hg03N,HI requires 1 = 43.0 per cent.colourless needles, which decompose a t 25SO.in water or alcohol, but dissolves only sparingly in ethyl acetate.o-A mino-op-dih ydro xyace t o ph enone Hydro ch2orid e,C,H,(OH),*CO*CH,*NH2,HCl.-This salt was prepared by the addition of concentrated hydro-chloric acid to an alcoholic solution of the corresponding hydriodide,when the new derivative immediately separated in needles. Whencrystallised from water or dilute alcohol, it yielded small, hardprisms, which melted at 2SOo, darkening previously :C1= 17.2. 0.3297 gave 0.2292 AgCl.o-A mino-op-dih ydroxyac et opli enone A urichloride,C8Hg03N,HC1 requires c1= 17.4 per cent.C6H3(OH)2*C'O*CH,*NH2,HAuC1,.-The gold salt of o-amino-op-dihydroxyacetophenone was readilysoluble in water, but crystallised from its concentrated solution inorange-coloured leaflets, which, on heating, gradually darkened,and finally melted at 283O.0*1201 gave 0.0467 Au.The dried salt was analysed:Au = 38.9.C8H,0,N,HAuCl, requires Au = 38.9 per cent2514 TUTIN: SYNTHESES IN THE EPINEPHRINE SERIES.PART 11.o-Amino-op-dihydroxyacetophenone Platinichloride,[C6H,(OH)2*C00CH2*NH2]2,H,PtC16.-This derivative was rather readily soluble in water, and crys-tallised from this solvent in fawn-coloured needles, which meltedand decomposed at 247O:0.1037 gave 0.0294 Pt.Pt=28*5.(C,H,03N),H2Pfc16 requires Pt = 28.5 per cent.o -,4 mino,op-dihy dro x yac e toph enone pkra t e, CaHg0,N,C6H,0,N,,crystallised from water in bright yellow needles, which melted anddecomposed at 222O.o - Amino-op-dihydroxyacetop7zenone, C6H,(OH),*CO*CH2=NH,,was prepared from the above-described hydriodide or hydrochlorideby the addition of a hot concentrated solution of sodium carbonateto a similar solution of the respective salt, both liquids havingpreviously been deprived of dissolved air by boiling. The newbase then immediately separated in small, pinkcoloured plates,which, on heating to 310°, suffered some decomposition, but did notmelt :0.1065 gave 0.2240 CO, and 0.0558 H,O.C,Hg03N requires C = 57.5 ; H = 5.4 per cent.w-Amino-op-dihydroxgacetophenone is soluble in both acids andalkali hydroxides, but is insoluble, or practically so, in all the usualsolvents with the exceptiop of pyridine, although when dissolvedin the last-mentioned liquid it suffered change.Attempts to prepare o-amino-op-dimethoxyacetophenone by heat-ing o-phthalimineopdimethoxyacetophenone with hydrochloricacid were unsuccessful, as the methyl groups were partly eliminatedby this treatment, the resulting product being a mixture.C=57.3; H=5.8.oorppr-Tetrahydroxy-2 : 5-diphenylpyrazine.o -Amino-opdih ydroxyace t op henone was boiled with p yridine,when it slowly dissolved, the solution acquiring a yellow colour.The liquid was then concentrated and cooled, when a substanceseparated in yellow needles.This was collected, but when washedwith ethyl acetate, or when dried, it lost pyridine of crystallisation,and became bright orange-coloured. It was unchanged at 326O,but at a higher temperature sublimed in yellow leaflets:C=64.6; H=4*3.C,,H,.O,N, requires C = 64.8 ; H = 4.1 per cent.0.1033 gave 0.2447 CO, and 0.0400 H,O.This substance was evidently oorppr-tetrahydroxy-2 : 5-diphenyEpyrazine, C,H,(OH),.C,H,N,°C6H3(0H)2, and had been formed bythe condensation of two molecules of the original keto-base followedby spontaneous oxidation of the resulting oo’ppl-tetrahydrozyTUTIN : SYNTHESES IN THE EPINEPHliJNE SERIES. PART 11, 25153 : 6-dil~ydro-2 : 5-diphenylpgra:ine. It yielded unstable salts withthe mineral acids, of which the monosulphate wi19 bright orangeand the disulphate intense purple.oo’pp’-Tetrahydroxy-Z : 5-di-phenglpyrazine is very sparingly soluble in glacial acetic acid, morereadily solub!e in pyridine, and insoluble in all the other usualsolvents.oo’ppr-Tetrab enzoyloxy-2 : 5-diphenylpyrazine,C,H,( OBz)zDC*H2N2.C6H3(OBZ)2.-The above-described oo’pp’-tetrahydroxy-2 : 5-diphenylpyrazinereadily underwent benzoylation when treated according to theSchotten-Baumann method, yielding a product which crystallisedfrom ethyl acetate in glistening, colourless leaflets, melting at 2 1 2 O :0.0976 gave 0.2643 C’O, and 0.0356 H,O.C44H2808N‘2 requires C = 74.1 ; H = 3.9 per cent.oorppf-Tetrabenzoyloxy-2 : 5-diphenylpyrazine is somewhat spar-ingly soluble in ethyl acetate and in alcohol, but dissolves readily inchloroform.With the object of preparing the benzoyl derivative of o-amino-op-dihydroxyacetophenone, the hydriodide of this base was dissolvedin water, benzoyl chloride added, and then excess of aqueouspotassium hydroxide introduced, and the mixture shaken for sometime.The pasty product which separated was collected, dissolvedin boiling absolute alcohol, and then submitted to steam distillationfor the removal of the ethyl benzoate which had been formed fromthe occluded excess of benzoyl chloride. The non-volatile residue wasdissolved in alcohol, when, on keeping, a crystalline benzoyl derivativeseparated, but by no means in quantitative yield. The motherliquors from this solid contained an uncrystallisable oil, which, froma subsequent observation, would appear to have been the compoundwhich it was sought to prepare, namely, u-b enzoylamino-op-dib enzoyl-oxyacetophenone, C6H,(OBz),*~o0CH,*N~Bz. The crystallinesolid which was obtained formed small prisms, melting at 136-137O,and, on analysis, was found to be the benzoyl derivative of acondensation product of the base : *C = 73.8 ; H = 4.0.0.1532 gave 0.4106 CO, and 0.0624 H20.C2,H,,0,N requires C = 73.6 ; H = 4-2 per cent.0.2970, in 24 of benzene, gave At = - 0*165*.CZ2Hl5O4N requires M.W.= 357.This cornpound waa therefore the dibenzoyl derivative of aninternal anhydride of o-amino-op-dihydroxyacetophenone, and sincean analogous compound was obtained from a-aminGo-hydroxyaceto-phenone (pa 2518), but not from the related bases containing* The same compound was obtained when w-amino.opdihydroxyacetoy,henonehydriodide was benzoylated in pyridine solution.C = 73.1 ; H =4-5.M.W. = 3752516 TUTIN : SYNTHESES IN THE EPINEPHRINE SERIES.PART I1hydroxyl groups in the m- or p-positions, it would seem likely thatthe o-hydroxyl group was concerned in the anhydride formation. Inview of this consideration, it would appear probable that the sub-stance melting at 136-137O is a dibenzoyl derivative of 6-hydroxy-indoxyl (X):B~o*c,H,<~~>cH, H O * C , H ~ < ~ ~ ~ ~ > C H .(X-) (XI.)6-Hydroxyindoxyl, however, might have been expected to reactin its enolic form (XI), yielding a tribenzoyl derivative.That one of the benzoyl groups was attached to nitrogen wasproved by the conversion of this dibenzoyl derivative int'o o-benzoyl-amino-op-dihydroxyacetophenone by the action of alkali hydroxides.C,H,( OH),*CO*CH,-NHBz.-A quantity of the dibenzoyl derivative melting at 136-137O wasboiled with concentrated alcoholic potassium hydroxide for onehour, when water was added, and the mixture acidified with hydro-chloric acid.A compound then separated in slender, glisteningprisms, melting and decomposing at 260-265O :w-Benzoylamino-op-dih ydroxyacetopkenome,0.1339 gave 0.3267 CO, and 0.0590 H,O.C,,HI30,N requires C = 66.4 ; H = 4.8 per cent.w-Benzoylamino-op-dihydroxyacetophenone is very sparingly solublein alcohol, ethyl acetate, chloroform, or benzene, moderately so inglacial acetic acid, and readily so in pyridine. On prolongedheating with concentrated hydrochloric acid, it yielded o-amino-op-dihydroxyacetophenone hydrochloride and benzoic acid, and onbenzoylation it yielded a compound which appeared to be thetrib enzoyl derivative of the corresponding base.This compoundwas a liquid, and was doubtless identical with the similar productwhich was obtained together with the dibenzoyl derivative meltingat 136-13707 as previously noted.C = 66.5 ; H = 4.8.w -Ph tha Zimko-op-dih ydroxyac eto phenone.During the hydrolysis of w-phthalimineop-dimethoxyacetophenoneby means of hydriodic acid, it was observed that the reaction pro-ceeded in two stages, the methyl groups being much more rapidlyeliminated than was the phthalyl radicle.In one experiment,therefore, the reaction was stopped as soon as the evolution ofmethyl iodide had ceased, the mixture being diluted with waterand cooled. A solid then separated, which was collected andwashed. When recrystallised from acetic acid, this substance formedsmall tufts of short, colourless prisms, which gradually darkenedabove 2550°, and fused at 270°TUTIN : SYNTHESES IN THE EPINEPHRINE SERIES. PART 11. 25170.1835 gave 0.4350 CO, and 0.0637 H;O.w- Pht halimino-op-dih ydrox yacetophenone,C=64*6; H=3.8.C16H,,0,N requires C = 64.6 ; H = 3.7 per cent.is rather sparingly soluble in most solvents. When heated withconcentrated hydrochloric or hydriodic acids, it yielded the corre-sponding amine.w-1'1~ t hdamino-op-dihydroxyacetophenone,C,H,( OH),*CO*CH,*NH*CO*CGH4*C0,H.-The above-described pht halimino-derivative was dissolved inaqueous potassium hydroxide, and the solution boiled for some time.The mixture was then acidified with hydrochloric acid, boiled withanimal charcoal, and the filtered liquid concentrated to a, smallbulk and cooled.The solid which separated consisted largely ofpotassium chloride, but also contained crystals of an organic com-pound. The latter was isolated by extraction with boiling xylene,after which it was finally purified by crystallisation from water.Long, glistening leaflets were thus obtained, which melted a t 227O:C,GHl,O,N requires C = 61.0 ; H =4.1 per cent.0.1372 gave 0.3064 CO, and 0.0519 H,O.C = 60.9 ; H = 4.2.De~ivatives of w-Amino-o-hydroxyacetophenone.Since w-chloro-o-methoxyacetophenone gave only resinous productswhen heated with ammonia, it was necessary to employ potassiumphthalimide for the conversion of this chloro-ketone into the corre-sponding amine, just as was the case with the andogous opdi-methoxy-compound (compare p. 2512).w-Chloro-o-methoxyacetophenone was therefore converted into thecorresponding phthalimino-derivative in a, manner precisely similarto that employed in the case of the op-dimethoxy-derivative. Theresulting o-phtlbalimino-o-methoxyacetophenone,MeO*C,H,*CO*CH2*N<CO>C6H4, cowas very sparingly soluble in most solvents, but was readily purifiedby crystallisation from slightly diluted acetic acid.It formedcolourless, diamond-shaped plates, melting a t 200'5O :0.1553 gave 0.3972 CO, and 0.0632 H,O.This derivative was boiled for three hours with a mixture ofglacial acetic acid and concentrated hydriodic acid. After freeingthe liquid from phthalic acid by extraction with ether, the mixturewas evaporated to dryness under diminished pressure, and theresidue crystallised from a, mixture of ethyl acetate and alcohol.C = 69.5 ; H =4*5.C1,H,,O,N requires C = 69.2 ; H = 4.4 per cent.VOL. xcvir. 82518 TUTIN : SYNTHESES IN THE EPINEPHRINE SERIES. PART 11,Very lustrous, colourless plates were thus obtained, which melted at255O :0.3846 gave 0-3230 AgI.o-S mino-ehydrox yacet ophenone hydriodide,I =45-4.C,H,O,N,HI requires I = 45.5 per cent.HO*C6H4*CO- CH,*NH,,HI,is very readily soluble in water or alcohol, but only sparingly soin ethyl acetate.It does not tend to become discoloured, as is thecase with salts of the analogous bases containing a hydroxyl groupin the met% or parzLposition with respect to the side-chain.A quantity of this hydriodide was dissolved in pyridine andbenzoylated by means of benzoyl chloride. The product crystallisedreadily from ethyl acetate, forming colourless plates, which meltedat 133O :0-1277 gave 0.3538 CO, and 0.0530 H20.C,,H,,02N requires C = 75.9 ; H = 4.7 per cent.This compound was therefore evidently the benzoyl derivative of acondensatdon product of o-amino-o-hydroxyacetophenone, and isdoubtless constituted analogously to the corresponding derivativeof the op-dihydroxy-base (p.2516). It may therefore be l-benzoyl-indoxyl, a compound which does not appear to have been preparedpreviously.ooI-Dih ydroxg-2 : 5-diph eny Zpy razine.A quantity of o-amino-o-h ydrox y acet op henone h ydriodide wasdissolved in water and aqueous sodium carbonate added, the result-ing precipitate being collected, and crystallised from xylene. Asubstance was thus obtained in yellow needles, which melted at259--262O, and were insoluble in dilute acids. The same compoundwas obtained if the solution of the hydriodide were renderedalkaline by means of sodium hydroxide, and then acidified, theresulting precipitate being recrystallised from xylene :C = 75.6 ; H=4-6.0-0904 gave 0.2420 CO, and 0*0400 H20.This substance was evidently a condensation product, and itsproperties indicated it to be oof-dihpdroxy-2 : 5-diphenyZpyrazine,C,H2N,(C6H4*OH)2.It is very sparingly soluble or insoluble innearly a.11 solvents, and forms unstable salts of a bright red colourwhen treated with mineral acids in an anhydrous solvent. Whenheated above its melting point, it sublimed in yellow leaflets.It appears from this result that o-amineo-hydroxyacetophenone,when dissolved, behaves in a manner strictly analogous to thatexhibited by the opdihydroxy-base. That is to say, that twomolecules condense with the formation of 00'-dihydroxy-3 : 6-dGC = '73-0 ; H = 4.9.C,,H,,0,N2 requires c = 72'7 ; H = 4.5 per centTUTIN : SYNTHESES IN THE EPINEPHRINE SERIES.PART 11. 2519hydro-2 : 5-diphenylpyrazine, which then undergoes spontaneousoxidation to the corresponding pyrazine derivative.oot-Bib ensoyloxy-2 : 5-diphenylpyrasine, C,H,N,( C,H4*OBz),, wasprepared by benzoylating the above-described pyrazine derivativein pyridine solution, It formed small, almost colourless prisms,which melted at 185O, but the amount available was not sufficientfor analysis.A ction of Hydriodic Acid on pp/-Dimethoxy-2 : 5-diphenylpgrazine.As previously shown, the methoxy-2 : 5-diphenylpyrazinesdescribed in the present paper, which contain the substituent ethergroupings in the met% and para-positions, are formed by the con-densation of two molecules of an w-aminomethoxyacetophenone,followed by spontaneous oxidation of the resulting dihydropyrazinederivative, as follows (Gabriel, loc.cit.) :The corresponding o-aminohydroxyacetophenones, however, couldnot be caused to condense under any conditions.This behaviour is the reverse of that shown by the ortho-substituted o-aminoacetophenones, for o-amino-o-methoxyaceto-phenone and w-amino-op-dimethoxyacetophenone will not yieldpyrazines, whilst the corresponding hydroxy-derivatives spon-taneously pass into such compounds.With the object therefore of preparing pp'-dihydroxy-2 : 5-di-phenylpyrazine, the action of hydriodic acid on ppl-dimethoxy-2 : 5-diphenylpyrazine was investigated. It was found, however,that this acid alone had only an extremely slow action on thecompound in question, but that if a quantity of acetic acid wereadded to the mixture, a change ensued with moderate rapidity.The product obtained, however, was not the expected hydroxy-diphenylpyrazine, but the reaction proceeded further, fission of thepyrazine nucleus taking place, resulting in the formation of twomolecules of o-amine-p-hydroxyacetophenone (Tutin, Caton, andHann, Zoc.cit.). It was thus shown that the series of reactionswhich result in the formation of pyrazine derivatives from w-amino-acetophenones can be quantitatively reversed by means of hydriodicacid.A quantity of ppl-dimethoxy-2 : 5-diphenylpyrazine was boiled fortwo hours with a mixture of concentrated hydriodic and glacialacetic acids. The liquid was then diluted with water and extractedwith ether for the removal of iodine, after which it was evaporatedto dryness under diminished pressure.On crystallising the residue8 ~ 2520 TUTIN : SYNTHESES IN THE EPINEPHRINE SERIES. PART 11.from ethyl acetate, colourless, prismatic needles were obtained, whichmelted at 230O:0.3435 gave 0.2885 AgI. 1=45*4.w- A mino-p-l~y droxyac e t oph enone h y driodide,C8ff,0zN,HI requires I = 45.5 per cent.HO C6H,*CO*CH,-NHz,HI,is much more soluble in organic solvents than is the correspondinghydrochloride (Tutin, Caton, and Ham, Zoc. c i t . ) . On benzoylation,it yielded o-benzoylaminep-benzoyloxyacetophenone, melting a t173-174O.d ction of Eydriodic A.cid on mmfppf-Tetramethoxy-2 : 5-ifiphe.nyl-pyrazine.mmfpp’-Tetramethoxy-2 : 5-diphenylpyrazine was boiled for twohours with a mixture of glacial acetic and concentrated hydriodicacids, after which the liquid was diluted, extracted with ether, andevaporated to dryness under diminished pressure.The residue wascrystallised from a mixture of ethyl acetate and alcohol, when itformed small, nearly colourless prisms, melting at 247-248O :0-2175 gave 0.1717 AgI. 1=42.7.C8H,03N,HI requires I = 43.0 per cent.This salt was therefore evidently o-amino-mp-dihydroxyaceto-plhenone hydriodide, C6H3(OH)z*CO*GH,*NHz,HI. On renderingits solution alkaline with sodium carbonate, o-amino-mp-dihydroxy-acetophenone separated in nearly colourless leaflets, which graduallydecomposed and melted above 235O. This base has previously beenprepared by another method during the synthesis of epinephrine(D.R.-P.1556321, and the above-described reaction therefore affordsa new means of obtaining this important compound.Action of EydAodic Acid o n 2 : 6-Diphenylpgrazine.As the action of hydriodic acid on the 2 : 5-substituted pyrazineswas found to result in the complete disruption of the pyrazine ring,it was considered of interest to investigate the effect of this reagenton the analogous 2 : 6-substituted bases. It was then found thatthe nitrogen-containing ring was broken in this case also, in thefollowing manner :A quantity of 2 : 6-diphenylpyrazine was heated for several hourswith a mixture of concentrated hydriodic and glacial acetic acids.On allowing the liquid to cool, a sparingly soluble hydriodideseparated. This was collected, and recryst,allised from glacial acetiTUTIN : SYNTHESES IN THE EPINEPHRINE SERIES.PART 11. 2521acid, when it formed flattened needles, which melted and decom-posed at 211O. This salt proved to be diphenacylamine hydriodide,( C6H5* CO CH2),NH, H I :0.1981 gave 0,1217 AgI. 1=33.2.C1,Hl5O2N,HI requires I = 33.3 per cent.This hydriodide was converted into the corresponding hydro-chloride by treatment with hydrochloric acid in alcoholic solution,when glistening leaflets were obtained, which melted at 235O, afterpreviously becoming red. (Found, C1= 12.4. Calc., C1= 12.3 percent.) This salt had all the properties of diphenacylamine hydro-chloride, as described by Gabriel (Zoc. cit.), and it yielded gold andplatinum salts, in agreement with the similar derivatives preparedby him.The original acid filtrate from the diphenacylamine hydriodidewas evaporated to dryness under diminished pressure, and theresidue crystallised from a mixture of ethyl acetate and alcohol.A colourless salt was thus obtained, which dissolved easily in water,and was readily recognised by the usual tests as ammonium iodide.Conversion of Biphenacylamine into 2 : 6-IAphenylpyrazine.Both Gabriel (Zoc.c d . ) and the present author (p. 2502) haveobtained diphenacylamine by the interaction of w-bromo- or chloro-acetophenone and ammonia, and the present author has shown that2 : 6-diphenylpyrazine is also formed in this reaction (p. 2501).Now, since diphenacylamine results when this pyrazine derivativeis heated with hydriodic acid, it appeared to the present author thatthe former base might be the intermediate compound in theformation of the latter by the reaction mentioned.This has beenfound to be the case, for, when one of the above-described salts ofdiphenacylamine was heated with ammonia, 2 : 6-diphenylpyrazinewas regenerated. It is thus seen that the change which results inthe formation of 2 : 6-diphenylpyrazine is capable of reversion bymeans of hydriodic acid, just as has been shown to be the case withthe analogous 2 : 5-substituted pyrazines.A quantity of diphenacylarnine hydrochloride was heated in a,sealed tube for three hours at looo with a large excess of a solutionof ammonia in absolute alcohol.The reaction mixture was thenevaporated to dryness, the residue extracted with benzene, and thebenzene liquids evaporated. The dark-coloured residue so obtainedwas dissolved in a small amount of ethyl acetate, and a solutionof hydrogen chloride in absolute alcohol added, when 2 : 6-diphenyl-pyrazine monohydrochloride (m. p. lS9O) separated. On treatmentwith alcohol, this salt dissociated, yielding 2 : 6-diphenylpyrazine,melting at 90°2522 TLTTIN : SYNTHESES 1N THE EPLNEPHRINE SERIES: PART 11.It is, of course, evident that the interaction of diphenacylamineand ammonia must first result in the formation of a dihydro-2 : 6-diphenylpyrazineY the la-tter then undergoing spontaneousoxidation.Action, of Hydriodic A cid on pp1-Dimethoxy-2 : 6-diphenylp~razine.A quantity of pp'-dimethoxy-2 : 6-diphenylpyrazine wits boiledfor several hours with a mixture of concentrated hydriodic andglacial acetic acids.On allowing the mixture to cool, a verysparingly soluble hydriodide separated in long, colourless needles,which melted and decomposed at 2 5 1 O :0*1050 gave 0.1805 CO, and 0.0392 H20.ClGHl,04N,HI requires C = 46.5 ; H = 3.8 per cent.This salt theref ore was ppl-dihydroxydiphenacyZamine hydriodide,( HO*C,H,*CO0CH2),NH,HI. It was very sparingly soluble inwater, and rather more soluble in alcohol, but was insoluble in coldsolvents in the presence of an excess of hydriodic acid. pp'-Bi-hydroxydiphemcylamine, prepared from this salt, formed dark redcrystals, but as it was very unstable it was not further investigated.ppf-BihydroxydiphenacyZamine Hydrochloride,( HO*C6H,-CO* CH2),NH,HC1.-This salt was prepared by the addition of concentrated hydrechloric acid to an alcoholic solution of the corresponding hydriodide.It crystallised from alcohol in colourless leaflets, or from water inneedles, and melted at 279O.It is less soluble in alcohol than thehydriodide, but dissolves in water more readily than the latter :0.2093 gave 0.0914 AgCl.ppI-Bih y drozydiph enac ylamine Yla tinichloride,C=46*8; H=4*1.C1= 10.8.CuH150,N,HC1 requires C1= 11.0 per cent.[ ( HO°C,H4*C'OoCH2)2NH]2H2PtC16.-This derivative crystallised very readily in buff-coloured needles,which melted and decomposed at 230O:0.1210 gave 0.0241 Pt.Pt=19*9.pp'-BihydroxydiphenacyZamine AurichZoride,(C,,Hl,O,N),H,PtCI, requires Pt = 19.9 per cent.(HO*C,H4*CO*CH,),NH,HAuC1,.-This salt crystallised readily in bright yellow needles, whichmelted at 259O after undergoing some decomposition :0.1012 gave 0.0319 Au. Au=31.5.pp' - Di~~yd?.oxydi~?~enucyla.ntil.2.e @crate, C,,H,,04N,C,H,0,N3,C,,H,,O,N,HAUC~~ requires Au = 31.5 per cent.forms long, bright yellow needles, which melt at 169OTUTIN : SYNTHESES IN THE EPINEPHRINE SERIES. PART 11. 2523Conversion of ppl-Dihydroxydiphenacy2amine into ppf-Dihydroxy.2 : 6-diphenylpyrazhe.ppl-Dihydroxydiphenacylamine hydrochloride was heated for twohours in a sealed tube at looo with a large excess of a solution ofammonia in absolute alcohol. The mixture was then evaporated,and the residue extracted with boiling xylene. On crystallisingfrom glacial acetic acid the material dissolved by the xylene, small,pale yellow prisms, melting at 3 0 5 O , were obtained:0.1210 gave 0.3192 CO, and 0.0496 H,O.This compound wi19 evidently ppl-dihydroxy-2 : 6-diphenyl-pyi-uaine, C4H2N2(C6H4*OH),. It yielded unstable salts with themineral acids, the monohydrochiloride and monosalphate beingorange-coloured, whilst the diszclphute was deep reddish-purple.C=72*5; H=4*5,C,,H,,0,N~2 requires C = 72.7 ; H = 4.5 per cent.Action of Hydriodic Acid on mmfpp~-Tetrumethoxy-2 ; 6-diphenyl-pyrazine.nzmIppl-Tetramethoxy-2 : 6-diphenylpyrazine was heated withhydriodic acid in a manner similar to that described in connexionwith the corresponding dimethoxy-compound. A hydriodide wasthus obtained, which crystallised from acetic acid in colourlessleaflets, and melted and decomposed at 236O:0.1725 gave 0.0918 AgI. I=28*3.rnmlppl-T etrahydroxyd iph enacylamine hydriodide,C,,H,,O,N,HI requires I = 28.5 per cent.[C,H3(0H),*CO*CH2J2NH,HI,is somewhat more soluble in water than the corresponding di-hydroxy-compound. On treating its aqueous solution with alkalis,a yellow colour is produced, but oxidation very rapidly ensues, withthe development of a brown colour.mrnlppf-T e t rahydro x ydiph enac ylamin e I1 ydrochl oride,[C,H3(OH),*CO*CH,],NH,HCl.-This saIt was prepared by the addition of concentrated hydro-chloric acid to an alcoholic solution of the abovedescribedhydriodide, when the new derivative immediately separated. Itcrystallises from water in colourless leaflets, which melt and decom-pose at 264O:0.2687 gave 0.1075 AgC1. C1=9.9.C,,H,,O,N,HCZ requires C1= 10.0 per cent.On treating a solutian of mrnlppl-t e t ra?hydroxydiphenci cy Zami~i r:hydrochloride with auric o r platinic chloride, the respective metalwas quickly deposited. A rnercurichloride was obtained from th2524 TUTIN AND CATON: THE ABSORPTION SPECTRA OFhydrochloride in tufts of small, white needles, but it was unstable,and, on warming its solution, sercurous chloride soon separated.mml p p 1 -Te t rahydrox ydiph enac y Zarni~~e picrat e,[C,H,( OH),-CO *CH,],NH,C6H3O7N3,crystallised readily in tufts of yellow needles, which containedwater of crystallisation, and, when air dried, melted at 112--115°.I n conclusion, the author wishes to acknowledge his indebtednessto Mr. F. W. Caton, B.A., B.Sc., for the preparation and purificationof the o-chloro-o- and p-methoxyacetophenones employed in thisresearch.THE WELLCOME CHEMICAL RESEARCH LABORATOKIES,LONDON, EC