March, 1941
REACTION OF ~-BROMO-~,@-UNSATURATED KETONESWITH AMINES
[CONTRIBUTION FROM THE AWRYLABORATORY OF CHEMISTRY OF
THE
837
UNIVERSITY OF NEBRASKA]
a,p-Unsaturated Aminoketones. IV.’ Mechanism Studies of the Reaction of a-Bromo-a,p-unsaturated Ketones with Secondary Amines BY NORMAN H. CROMWELL
In paper 112 of this series a mechanism was evidence for the mechanism previously proposed2 proposed for the reaction of morpholine with for these reactions and as negative evidence a-bromobenzalacetophenone, and with benzalace- against the following scheme. tophenone dibromide. The present C6HsCH=C-cOC& + R2NH +CsHsCH-cH-cOCeHs I I investigation was undertaken in order I R2N Br Br to check this proposal and a t the same 42R:NH time to eliminate certain other possible CeHbCH=C-COCaHs -+- C&,-cH--CH-cOC~Hj + R:NH2+Brways of explaining these reactions. I 1 I R2N NR: NRl It was found that a-bromo-a-piperidinobenzylacet~phenone~ reacted with RINH excess morpholine to give the two exA second study was conducted by using these pected products, a - piperidino - /3 - morpholinobenzylacetophenone I and a-piperidinobenzal- same two secondary amines in the reverse order. acetophenone 11. Approximately two moles of The results of this second experiment were not of product I to one mole of I1 resulted. The follow- the same nature as those of the first investigation, but they contributed materially to these studies. ing scheme seems to explain these results. When a - bromo - a - morpholinobenzylacetopheNCsHio none2 was treated with excess piperidine a much I CaHsCH=C-COCsHs + C~HICH*-C-COC~H~ more rapid reaction took place than in the first 1 I experiment described above. Almost two moles Br Br of unsaturated amino ketone product to one mole CsHi(r” 4 0 ” of the diamino ketone product resulted. Neither CsHsCH-CHCOCsHs C~H~CH=C-COC~HI of these two products was of a single chemical I 1 1 nature, but they were found to be mixtures that OHsCiN NCsHio NCsHio I I1 were almost impossible to separate because of the GHloON+Brsimilar solubilities of constituents. Mixed meltThe structure of I was shown by hydrolyzing ing point experiments with compounds differing it with dilute acid. Only w-piperidinoacetophe- only in the presence of a morpholino in place of a none and benzaldehyde were found in the reac- piperidino group gave only small depressions. tion mixture. Compound I1 was identical with Consequently, it was necessary to rely on carbon a-piperidinobenzalacetophenone prepared ac- and hydrogen analyses and decomposition prodcording to the method of DufraisseaJ ucts for the chemical identification of the products. HOH CaH6-CH-CH-COCeH6 + CsH6CHO f CiHsONH The analyses of the unsaturated product indiOHsCiNI NCsHio 1 cated that it was probably a mixture of a-morI CsHsCOCH&C6Iio pholinobenzalacetophenone IV and of a-piperi11. The analyses of the dinobenzalacetophenone It was not possible to obtain the unsaturated diamino ketone product also showed that it was amino ketone I1 from the diamino ketone I by long not the normal product 111, as expected, but that heating in alcoholic solutions. The diamino it more probably was a mixture of I11 with the ketone does not seem to be the precursor of the V of D u f r a i ~ s e . ~The hydipiperidino ketone unsaturated amino ketone. drolysis of this diamino ketone product gave a These experiments may be taken as positive mixture of w-morpholinoand w-piperidinoaceto(1) For paper I11 of this series see Cromwell, THISJOURNAL, ea, 3470 (1940). phenones. (2) Cromwell, ibid., 61,2897 (1940). From the experimental evidence it seems that (3) Dufraisse and Moureu, Bull. SOC. chim., (4’) 41, 466 (1927). (0 Dufraisse and Moureu, ibid., (4’) 41, 457 (1927). the relative basic strengths of the amines used
+
+
+
+
+
NORMAN H. CROMWELL
838
affect not only the rate but also the course of these reactions. It appears unlikely that the results of this second experiment in any way invalidate the conclusions drawn from the clear course of all previously described experiments. It has been stated by others6 that piperidine reacts more rapidly with halogen compounds and gives more by-products than does morpholine. These latter experiments not only bear out this contention, but, furthermore, indicate that possibly a morpholino radical already attached to carbon may be replaced by a piperidino radical under these conditions to give abnormal products.
Vol. 63
The residual red reaction mixture was evaporated and cooled to give a red product. This was recrystallized from absolute alcohol to give orange-red plates (2.4 9.) compound 11, m. p. 102-103". Compound I 1 was identical with a-piperidinobenzalacetophenone as prepared by Dufraisse. Anal. Calcd. for CZOH~INO: C, 82.46; H, 7.24; N, 4.80. Found: C, 82.80; H, 7.24; N, 4.90. Hydrolysis of IA and IB.-These products were hydrolyzed with dilute sulfuric acid (15%) in the usual way. Both gave w-piperidinoacetophenoneand benzaldehyde as the major products. The w-piperidinoacetophenone was recovered as the hydrochloride and was found to be identical with a synthetic sample prepared from w-bromoacetophenone.' w-Piperidinoacetophenone Hydrochloride.-IA ( 1 00 Br 9.) gave 0.60 g. (m. p. 225-227'); IB (1.44 9.) gave 0.65 g. I (m. p. 226-227"). A mixture of the hydrochlorides of CeHsCHz-C-COCsH6 CsHacH=C-coc~H6 w-piperidinoacetophenone and w-morpholinoacetophenone I I NCpHsO NC4HsO melted a t 212-219". IV + Anal. Calcd. for CiaHiflOCl: C, 65.12; H, 7.55. Found: C, 65.15; H, 7.65. excess C6HloNH CeHsCH-CHCOCsHa Attempted Decomposition of 1B.-A sample of the diI I HioC6N NC&O amino ketone (0.551 g.) was refluxed with absolute alcohol I11 (8 ml.) for five hours and then allowed to stand a t room normal products C~HK-CH=C-COC~H~ temperature for twenty hours. The light yellow solution I did not darken after this treatment. All of the starting NC~HIO material was recovered unchanged (0.54 g., m. p. 156Ir 1580). CeH6-CH-CH-COCeH6 Reaction of Piperidine with a-Bromo-a-morpholinoJ I benzy1acetophenone.-Freshly prepared a-bromo-a-morHioC5N NCKHIO pholinobenzylacetophenone2 (20.64 9.) was mixed with V, abnormal products absolute alcohol (37 mi.). To this suspension piperidine The reactions of a-bromo-a,P-unsaturated ke- (10.6 g., 2.2 molar amount) was added rapidly with stirtones with various pairs of secondary amines, as ring. A slow reaction set in, the solution slowly turning yellow and giving off a small amount of heat. After standwell as other studies of the mechanisms and prod- ing at room temperature for four days a mixture of orange ucts of these interesting reactions are in progress. and yellow crystals was filtered off and washed with water. Concentration of the filtrate gave more product, resulting Experimental6 in a total oE 14.8 g. of mixed product. The orange colored product was separated from the yelReaction of a-Bromo-a-piperidinobenzylacetophenone low one by washing with small amounts of 95% ethyl alcowith Morpho1ine.-A freshly prepared sample (15.7 9.) of a-bromo-a-piperidinoben~ylacetophenone~ (m. p. 160- hol. Several recrystallizations of the orange product from 162"), was mixed with absolute alcohol (30 ml.). To this absolute alcohol gave 5.0 g. of orange crystals, m. p. 96suspension was added a two molar amount of morpholine 97 O ; mixed with pure a-piperidinobenzalacetophenone, (7.35 8.). The mixture was then heated to boiling to m. p. 96-99 O , and with pure a-morpholinobenzalacetogive a red-orange solution which set solid while still hot. phenone, m. p. 93-96". Anal. Calcd. for C ~ ~ H ~ ~ NC,O 77.80; Z: H , 6.51; N, After standing a t room temperature for twenty hours a light yellow precipitate was filtered off and washed with 4.77. Calcd. for CzoH21NO: C, 82.46; H, 7.24; N, 4.80. 95% alcohol and then water to give light yellow needles Found: C, 78.90; H, 6.78; N, 5 08. (7.5 g.), m. p. 149-155". Fractional recrystallization from I t was recrystallized several times more from alcohol benzene and petroleum ether gave two products: I A and from benzene and petroleum ether (solutions seeded A with a-morpholinobenzalacetophenone2). (2.2 g.), m. p. 174-175"; I B (3.5 g.), m. p. 155-157'. mixture of I A and IB melted a t 153-158". Anal. Found: C, 80.30; H , 6.96; N, 4.95. Anal. Calcd. for Cs4HaoNzOz: C, 76.16; H , 7.98. Acid hydrolysis of the mixed unsaturated product in the Found: for IA, C, 76.41; H, 7.85; for IB, C, 76.22; H, usual way gave 8 0 4 5 % yields of benzyl phenyl diketone 7.98. The yellow product was recrystallized two times from absolute alcohol and then two times from benzene and pe(6) Harradence and Lions, J . Proc. Roy. SOC.N . S. Wales, 70, troleum ether to give 4.1 g. of pale yellow needles, m. p. 406 (1937). (6) Some of the microanalyses for carbon and hydrogen and 153-156"; mixed with IB, m. p. 148-151'; mixed with the
+
+
J.
+
D u m a s nitrogen are by E. E. Renfrew of the School of Chemistry, University of Minnesota.
( 7 ) Rabe and Schneider, Bor.. 41, 874 (1908).
March, 1941
COLORTESTSFOR ORGANOBISMUTH COMPOUNDS
839
Several recrystallizations from absolute alcohol and dry low melting form of a,@-dimorpholinobenzylacetophenone,2 ether did not change the melting point or composition of rn.p. 145-152’. Anal. Calcd. for Cz4Hao.Pu’zOz: C, 76.16; H, 7.98; N, this mixture appreciably. 7.40. Calcd. for C Z ~ H ~ ~ N C, Z O79.77; : H, 8.55; N, 7.44. Summary Calcd. for C28HZa203: C, 72.61; H, 7.41; N, 7.36. Found: C, 77.64; H, 8.39; N, 7.41. 1. Further evidence is presented for a It was recrystallized again from the same solvents. proposed mechanism2 of the reaction of QAnal. Found: C, 77.61; H, 8.20; N, 7.42. bromo-a,P-unsaturated ketones with secondary These experiments were repeated three times and the amines. above described products were obtained each time. 2. It has been shown that these reactions may Hydrolysis oE the diamino ketone product in the usual way gave a mixture of w-morpholino- and w-piperidino- take an abnormal course depending on the characetophenones, which was isolated as the hydrochloride acteristics of the secondary amine used. mixture. This mixture melted a t 216-220”. 3. A new a$-diaminobenzylacetophenone has Anal. Calcd. for C12H&TOZC1: C, 59.64; H, 6.66; N, been prepared. 5.79. Calcd. for C I ~ H ~ ~ N O CC, I :65.2; H, 7.57; N, 5.84. Found: C, 63.18; H , 7.21; N, 5.54. LINCOLN, NEBRASKA RECEIVED DECEMBER 30, 1940
[CONTRIBUTION FROM THE CHEMICAL
LABORATORY O F IOWA STATE COLLEGE]
Color Tests for Some Organobismuth and Other Organometallic Compounds1 BY HENRYGILMANAND H. L. YABLUNKY Different color tests have been proposed for organometallic compounds. The first of these, which is now being designated as Color Test I, is most comprehensive, and has been applied to a wide variety of organometallic compounds. This test uses Michler ketone and is applicable, in general, to all organometallic compounds which add to the carbonyl group in ketones.2 The second color test3 (Color Test 11) has a two-fold utility. First: it can be used to differentiate organolithium compounds from the corresponding Grignard reagents. Second, it is useful, with some limitations, for differentiating alkyllithium compounds from aryllithium compounds. The test about to be described, 111, appears to be characteristic not only for triarylbismuth dihalides but also for the more reactive arylmetallic compounds. Test 111is based on the observation of some years ago by Challenger4” that a transitory purple color developed when triphenylbismuth dibromide reacted with phenylmagnesium
bromide incidental to an attempted preparation of tetraphenylbismuthonium bromide. The reaction apparently did not proceed as expected (C6H&BiBra 4- CaHIMgBr -+ (CeH&BiBr MgBrz since the only products obtained were triphenylbismuth, diphenylbismuth bromide, phenylbismuth dibromide and bromobenzene. Later, Challenger and c o - w o r k e r ~reported ~~ the same intense purple color in attempting to prepare other arylbismuthonium chlorides and fluorides by this reaction. It is unlikely that this reaction can be used to prepare bismuthonium compounds inasmuch as triphenylarsenic dichloride and methylmagnesium iodide gave triphenylar~enic~~ and not the known stable methyltriphenylarsonium chloride.4d Two suggestions were offered4bto account for the transient purple color. One of these was the possible formation of tetraarylbismuthonium halides (RdBiX); and the other was the momentary formation of compounds of the type RsBi, analogous to the bright red triphenylmethyltetramethylammonium and benzyltetramethylammonium prepared by Schlenk and H o l t ~ . These ~ are the only RS types reported in Group V. The purple color is probably not due to the formation of a bismuthonium halide because R4AsX6 and R4SbX
+
(1) This is paper XXXV in the series “Relative reactivities of organometallic compounds.” Paper XXXIV is in THIS JOURNAL, 69, 3206 (1940). (2) Gilman and Schulze, ibid.. 47, 2002 (1925). The most recent study concerned with this color test is by Gilman and Jones, ibid., 62, 1243 (1940). The latter article has references t o several other reports on the range of applicability and degree of sensitivity of the color t e s t . (3) Gilman and Swiss, i b i d . , 69, 1847 (1940). (4) (a) Challenger, J. Chem. SOC.,106, 2210 (1914). (b) Chal( 5 ) Schlenk end Holtz, Bw., 49, 603 (1916); ibid., 60,274 (1917). lenger end Goddard, ibid., 117, 762 (1920); Challenger and Wilkinson, ibid., 121, 91 (1922). (c) Challenger and Allpress, i b i d . . 119, (6) Blicke and co-workers, Tars JOURNAL, 65, 3056 (1933); 67, 720 (1935); 60,423 (1938); 61,88 (1939). 913 (1921). (d) Michaelis, Ann., Sal, 166 (1902).
