@-AMINO-&PHENYLPROPIOPHENONE. 69 VI.-&Arnino-p-phenylpropiophenone. By ALEX. MCKENZIE and FRED BARROW. a-AMINO-KETONES. for example a-aminopropiophenone exhibit great instability in the free state on account of their tendency to condense in the presence of air to give substituted pyrazincs (Gabriel Ber., 1908 41 1127). On the other hand S-amino-ketones are much more stable; diacetoneamine is exceptional as it can even be dis-tilled but other ketones of this class are not nearly so stable; thus Gabriel (Ber. 1908 4 1 242) has shown that P-aminopropiophenone evolves ammonia when heated with aqueous potassium hydroxide, and it does not appear from his description that the compound is stable enough to be kept a t the ordinary temperature for any length of time.For the purpose of another research it was desired t o obtain a P-amino-ketone containing an asymmetric carbon atom and the preparation of B-amino-p-phenylpropiophenone was accordingly undertaken by the authors in the hope that this compound would be sufficiently stable to permit of its isolation. The preparation was carried out as follows. Cinnamic acid was converted into P-amino-& phenylpropionic acid from which the corresponding phthalimino-acid was obtained by heating with phthalic anhydride. The action of benzene and aluminium chloride on the chloride of the phthal-imino-acid led to ~-phthalimino-/3-phenylpropiophenone which was converted by means of aqueous alkali into the corresponding phthal-amic acid (compare Gabriel Ber. 1908 41 242 513 and other papers by the same author).T"he hydrolysis of the phthalamic acid was accomplished by a mixture of glacial acetic acid and concen-trated hydrochloric acid. The resulting hydrochloride on decom-position with ammonia gave P-amino-P-phenylpropiophenone as a solid melting a t 82-83O. The free amino-ketone is much more stable at €he ordinary temperature than is desylamine the related a-amino-ketone. An attempt was made t o obtain the ketone by a more direct method namely by the action of magnesium phenyl bromide on ~-amino-j3-phenylpropionanilide NH,*CHPh*CH,*CO*NHPh. The disruption of the additive compound by aqueous ammonium chloride led however to the regeneration of the anilide. The result would suggest that the additive compound may have been 7 HPh CH,*E*N Ph*MgBr NH*MgBr .0 J I . Ph MgB 70 MCKENZIE AND BARROW : in which case the rearrangement into a product giving the amino-ketone on hydrolysis did not occur. Now it has been found by McKenzie Martin and Rule (T. 1914 105 1583) that the action of Grignard reagents on allrylnted acid amides (for example, r-mandeloethylamide and I-P-hydroxy-/il-phenylpropionethylamide) does not always lead to the formation of the corresponding lretols, and it seems probable that the additive compounds formed in those cases may also have been of the oxonium type postulated above. v. Braun has in fact suggested the structure for the compounds obtained from dialkylated amides. The recent work of v. Braun Heider and Muller (Ber. 191'7 50 1637) and of v. Braun and Kirschbauin (Ber.1919 [B] 52 1725) on the action of Grignard reagents (prepared from bromoalkylsted anilines) on aldehydes and ketones respectively certainly lends support to the view that the first phase in the addition of a Grignard reagent to an aldehyde or a ketone consists in the formation of an oxonium compound where the reagent attaches itself to the oxygen atom of the carbonyl group of the aldehyde or ketone. Such an additive ccmpoiind would be labile enough to undergo as a general rule, iscmeric change with readiness prior t o its hydrolysis but appa-ren€ly it may sometimes happen that this rearrangement does not occur. T'he action of aluminium chloride on a solution of P-formylamino-,8-phenylpropionyl chloride in benzene took an unexpected course, the product being /3p-diphenylpropiophenone.Now it is shown in the present paper 1 hat P-phthalimino-/il-phenylpropionic acid decom-poses on heating into carbon dioxide phthalimide cinnsmic acid, and styrene the last-named substance being doubtless formed from the cinnamic acid (I). This result siiggested a similar course for R,N>,:o..,~~~f&~~lt: .R' c Ii *CIi.CH-CO('l 5 1 I .................................. . I i I b ; ; I I i NIC*CHO ................................. (11.1 the decomposition of P-f ormylamino-P-phenylpropionyl chloride into f ormamide and cinnamoyl chloride (11) aluminium chloride acting as a catalyst. The cinnamoyl chloride formed in this manner then reacts with benzene and aluminium chloride with the formation of Po-diphenylpropiophenone.The interaction of cinnamoyl chloride, benzene and aluminium chloride under the conditions described i 6-AMINO-6-PHENYLPROPIOPHENONE. 71 the experimental part gave a mixture of PP-diphenylpropiophenone and j3B-diphenylpropionic acid. B-Benzoylamino-P-phenylpropionyl chloride differed from the formyl compound inasmuch as it underwent the normal reaction with benzene and aluminium chloride the product being @-benzoyl-amino-& phenylpropiophenone . Some optically active derivatives of B-amino-j3-phenylpropionic acid are a t present under investigation by one of us (A. McK.). E X P E R I M E N TAL. @-A mino-8-phcnylpropionanilide. P-Amino-B-phenylpropionic acid was prepared by the action of hydroxylamine on cinnaniic acid according fo Posner's method (Ber., 1905 38 2316).Four grams of the well-dried and finely-powdered amino-acid were suspended in 20 C.C. of freshlydistilled acetyl chloride the mixt'ure was cooled in ice and treated gradually with 5.3 grams of finely-powdered phosphorus pentachloride. The mix-ture was shaken during the addition of the pentachloride (fifteen minutes) and the shaking was continued for half an hour longer at the ordinary temperature. The resulting solid 8-amino-P-phenyl-propionyl chloride hydrochloride was separated and washed with acetyl chloride and light petroleum. A solution of 2.5 grams of it were suspended in dry ether and treated a t the ordinary tempera-ture with a solution of 4.2 grams of aniline in ether. Water was then added the aqueous solution separated and made alkaline with sodium hydroxide.The anilide precipitated was crystallised from aqueous alcohol (Found C = 74.8 ; H = 6.9. Cl,Hl,0N2 requires C=75*0; H=6*7 per cent.). &Amino - fi - phenylpropio~aanilide NH,*CHPh*CH,*CO*NHPh, separates from ethyl alcohol in colourless plates melting a t 122O. It is readily soluble in ethyl alcohol or chloroform moderately so in benzene or ether and sparingly so in light petroleum. Attempts to prepare the amide and the piperidide of 8-amino-P-phenylpropionic acid by the action of dry ammonia and of piper-idine on a suspension of the chloride hydrochloride in dry ethereal solution gave amorphous products probably due to polypeptide formation. The anilide (1 mol.) was subjected to the action of magnesium phenyl bromide (6 mols.).On decomposing the additive compound with ammonium chloride the anilide was regenerated 72 McKENZIE AND BARROW : A ction of P-Formylamino-&phenylpropionyI Chloride on Benzene in the Presence of Aluminium Chloride. The formylation of the amino-acid was carried out as described by Fischer Scheibler and Groh (Ber. 1910 43 ZOZO) and the formyl acid then converted into its chloride by means of acetyl chloride and phosphorus pentachloride. The chloride prepared frcm 3.9 grams of the formyl acid was a viscid yellow oil; it was dissolved in 50 C.C. of benzene and acted on by 5 grams of alumin-ium chloride by heating on the water-bath for one hour. After decompdsition with hydrochloric acid and removal of the benzene, the product was crystallised twice from ethyl alcohol when lustrous needles (0.7 gram) separated.These melted a t 94-95O contained no nitrogen and consisted of PB-diphenylpropiophenone, CHPh,-CK,*COPh (Found C = 88.1 ; H = 6-5. Calc. C = 88.1 ; H=6.3 per cent.). Kohler (Amer. Chem. J. 1903 29 352) gives m. p. 96O. Action of Cinnamoyl Chloride on Benzene in the Presence of ,4 luminium Chloride. This action had already been investigated by Kohler Heritage, and Burnley (Amer. Chcm. J. 1910 44 60) under somewhat different conditions to those employed by us. They found that PP-diphenylpropiophenone was accompanied by 3-keto-1-phenyl-2 3-dihydroindene and P-chloro-6-phenylpropiophenone. The product obtained under the following conditions appeared t o consist entirely of PB-diphenylpropionic acid and p/3-diphenylpropiophenone.A mixture of 10 grams of cinnainic acid and 14.8 grams of phos-phorus pentachloride was heated on the water-bath for half an hour, the oxychloride removed under diminished pressure and the result-ing oil acted on by 50 C.C. of benzene and 10 grams of aluminium chloride. The action was completed by heating and the mixture allowed to remain overnight. Dilute hydrochloric acid was added, the product extracted with ether and the ethereal solution then shaken with sodium carbonate. On acidifying the alkaline extract with hydrochloric acid 5.1 grams of p/3-diphenylpropionic acid were precipitated which after crystallisation from ethyl alcohol separated in lusirous white needles melting a t 149-150°. Liebermann and Hartmann (Ber.1892 25 2124) who obtained the acid as one of the products of the action of sulphuric acid on cinnamic acid and benzene give m. p. 149O. (Equivalent =229.1. Calc. =226.) After being dried with sodium sulphate the ethereal solution on evaporation gave a red oil which solidified almost completely whe ,&AMINO-@-PHENYLPROPIOPHENONE. 73 kept in the ice-chest for two days. The solid was freed from a small quantity of oil and purified by crystallisation from alcohol. It amounted to 11.5 grams and consisted of PPdiphenylpropio-phenone. B- Benzoylamino-6-phenylpropiophenone. 8-Amino-P-phenylpropionic acid was converted into its benzoyl derivative which melted a t 194-195O whereas Posner (loc. cit.) gives 194-196O. Ten grams of the benzoylamino-acid were con-verted by means of acetyl chloride and phosphorus pentachloride into P-benzoylamino-P-phenylpropionyl chloride which crystallises in needles.The latter was dissolved in 50 C.C. of benzene and acted on by 10 grams of aluminium chloride. After the usual manipula-tion 4.1 grams of a solid were isolated. P-Benzoylamino- P-phenylpropiophenone, NHB z- CHP h CH2*COP h, separates from ethyl alcohol in lustrous slender colourless needles melting a t 152-154O (Pound C = 79.9 ; H = 6.1 ; N = 4.3. C2,H,,02N requires C = 80.2 ; H = 5.8 ; N = 4.3 per cent.). P-Ph thalimino-P-pheny Zpropionic A cid. A mixture of 11 grams of P-amino-p-phenylpropionic acid and 10.5 grams of phthalic anhydride were heated for two hours a t 155-165O a t the end of which time the phthalimino-acid com-menced to crystallise from the molten mass.On cooling the acid was dissolved in 20 C.C. of hot glacial acetic acid and 20 C.C. of benzene were added. On cooling 13 grams of the crude acid separated. Purification was effected by crystallisation from benzene. ~-P~thalimino-P-phenylpropionic acid, C,H,:(CO),:N*CHPh*CH,*C02H, melts a t 169*5-170*5O is very readily soluble in hot ethyl alcohol or hot glacial acetic acid moderately so in benzene and very spar-ingly soluble in hot water. It separated from alcohol in slender needles (Found C= 69.1 ; 13 = 4.7. C,,H,,O,N requires C = 69.1 ; H=4*4 per cent.). In a second preparation 35 grams of phthalic anhydride and 36 grams of the amino-acid gave 48 grams of the phthalimino-acid melting at 168-170° after one crystallisation from a mixture of 60 C.C.of glacial acet,ic acid and 60 C.C. of benzene. The acid chloride prepared by heating the acid with an excess of thionyl chloride for half an hour separates from benzene in needles melting a t 96-97O. For analysis it was dried over phos-D 74 McKENZIE AND BARROW : phoric oxide and paraffin wax (Found C1= 10.6. C,,H,O,NCl requires 11.3 per cent.). The anilide prepared by the interaction a t the ordinary tem-perature of aniline and the acid chloride in ethereal solution sepa-rates from glacial acetic acid in glistening prisms or rhomboidal plates melting at 238O. It is sparingly soluble in ethyl alcohol (Found N- 7.5. C,,H,,O,N requires N=7*6 per cent.). The methyl ester prepared by the esterification of the acid with methyl alcohol and hydrogen chloride separates from ethyl alcohol in plates melting a t 9 2 O (Found N=4-9.C18H,,04N requires N=4*5 per cent.). The action of heat on B-phthalimino-8-phenylpropionic acid was studied. Evolution of carbon dioxide began a t 240° and became vigorous at 300O. The temperature was maintained a t 320-350° for half an hour. 'The liquid which had a pronounced odour of styrene was then distilled under ordinary pressure. The residue was dissolved in ethyl alcohol and decolorised with charcoal ; phthal-imide was isolated from it. The distillate solidified to a white, crystalline mass which was crystallised twice from a mixture of ethyl alcohol and benzene; the product was identified as phthal-imide. The alcohol-benzene solutioii from which the phthalimide had been separated was extracted with aqueous sodium carbonate, and cinnamic acid was obtained from the alkaline solution by acidification and extraction with ether.B-P h t ha Zinzi?i.o -fl-phenylpropiophenone. The best conditions for the preparation of this compound were the following. Twenty-five grams of the phthalimino-acid were heated for half an hour on the water-bath with 20 grams of thionyl chloride until the evolution of hydrogen chloride ceased. The excess of thionyl chloride was removed by distillation under diminished pressure when the acid chloride remained as a very viscid yellow oil Benzene (100 c.c.) was added to the warm oil the greater part of which dissolved and on cooling separated in a mass of needles. Aluminium chloride (20 grams) was then added.Action soon started with evolution of hydrogen chloride. After remaining a t the ordinary temperature overnight the mixture was heated for twenty minutes when a further vigorous evolution of hydrogen cbloride took place. On cooling water and hydrochloric acid were added and the benzene removed by distillation in a current of steam. The ketone was left as a viscid pale yellow oil which solidified readily. It was separated and crystallised from ethyl alcohol. Yield 26 grams &AMINO -@-PHENYLPROPIOPHENONE . 75 B-Ph t halimino-&phenylpropiopheno ne, C,H,:C,O,:N*CHPh*CH~oC'BPh, separates from etlhyl alcohol in needles melting a t 116-117O. It is readily soluble in hot alcohol but only moderately so in the cold solvent.It crystallises sluggishly and tends to separate from hot concentrated solutions as an oil. B- Benzoy 1-a-ph enyle t hy l p h t halamic A cia?. B-Phthalimino-B-phenylpropiophenone (7.5 grams) was sus-pended in 20 C.C. of ethyl alcohol and heated on the water bath for ten minutes with 30 C.C. of N-sodium hydroxide. The solution was diluted with 200 C.C. of water and acidified with hydrochloric acid. The phthalamic acid was precipitated as an oil which on stirring coagulated to form a pasty mass and this on remaining overnight became quite solid and friable. It was purified by crystallisation from ethyl alcohol. B-Benzoyl-a-phenylethyl~~t~alamic acid, COPh*CH,-CHPh*NH*CO* C,H,-CO,H, separat'es from ethyl alcohol in needles melting a t 1 3 2 O (Found, N=3*T.C,,'H,,O,N requires 3.8 per cent.). It is readily soluble in acetone ethyl alcohol or glacial acetic acid moderately so in benzene and sparingly soluble in light petroleum. j3- A mino-6-pheny Zpropiophenone. A solntion of 10 grams of P-benzoyl-a-phenylethylphthalamic acid in 50 C.C. of hot glacial acetic acid was heated on the water-bath for one hour with 30 C.C. of concentrated hydrochloric acid; a small amount of a pale yellow oil separated on heating. On the addition of 150 C.C. of water a further quantity of oil was precipitated, which became pasty on shnkiug. The clear solution was separated and evaporated to about 50 C.C. under diminished pressure sepa-rated from a further small quantity of oil and finally evaporated to dryness under diminished pressure.The amino-ketone hydro-chloride was obtained in this way in slender colourless needles accompanied by a small amount of a viscid yellow oil. The aquecus soliition was extracted with ether and rendered alkaline with ammonia. The precipitated amino-ketone was extracted with ether the ethereal solution dried with sodium sulphate and the ether removed. The residual pale yellow oil solidified on cooling and was crystallised from a mixture of benzene and light petroleum. The yield was 0.75 gram. B-Amino-/3-phenylpropioyhenone NH,*CHPh*Cs*COPh crys-D* 76 P-AMINO-p-PHENYLPROPIOPHENONE. tallises in scft colourless leaflets melting a t 82-83O (Found, N=6.3. CI5Hl5ON requires N=6.2 per cent.). It is readily soluble in ethyl alcohol benzene ether or ethyl acetate and only sparingly so in cold light petroleum (b.p. 30-5OO). It is moderately stable differing in this respect markedly from desylamine; thus a specimen kept in a well-corked tube underwent n3 apparent change when kept for six days. It then became yellow the crystals adhered together and after a further four to five days it was transformed into a yellowish-brown semi-liquid mass. In another preparation where 15 grams of the phfhalamic acid were hydrolysed by heating for half an hour with 60 C.C. of hydro-chloric acid and 100 C.C. of glacial acetic acid the yield of amino-ketone (2.5 grams) was better than in the experiment just described. The picrate crystallises from ethyl alcohol in yellow leaflets (Found N = 12.4. C,,H,,ON,C6H30,N3 requires N= 12.3 per cent.) which commence to darken and sinter a t 175O and melt and decompose a t 193O. The platinichloride separates from dilute hydrochloric acid in brown needles (Found Pt = 22.5. (C,,H,,ON),H,PtCl requires Pt=22*7 per cent.). It shows no definite melting .point but com-mences to darken and shrink to a thin core a t 195O decomposition being complete a t 250O. The crude P-amino-j3-phenylpropiophenone hydrochloride result-ing from the hydrolysis of the phthalamic acid was shaken with benzoyl chloride and sodium hydroxide. The resulting solid was crystallised from ethyl alcohol from which it separated in needles melting a t 152-153O. This was identical with the benzoyl deriv-ative previously described (p. 73) obtained by the interaction of /3-benzoylamino-P-phenylpropionyl chloride and benzene in the presence of aluminium chloride. UNIVERSITY COLLEGE DUNDEE, UNIVERSITY OF ST. ANDREWS. BIRKBECK COLLEGE, LONDON E.C. [Received December 17th 1920.