Aug., 1945
STOBBE CONDENSATJON WITH 2-ACETYLNAPHTHALENE
[ CONTRIBUT1ON
FROM THE
LABORATORY O F ORGANIC CHEMISTRY OF THE UNIVERSITY
1357
O F WISCONSIN]
The Stobbe Condensation with 2-Acetylnaphthalene. A Synthesis of 2,4-Dimethylphenanthrene BY WILLIAMS. JOHNSON, ARTHURGOLDMAN-‘ AND WILLIAM P. SCHNEIDER thy1)-butyrolactone, 11, was produced in 57-6970 yields. In contrast when the Stobbe condensation was conducted in t-butyl alcohol containing potassium t-butoxide, the reaction was very much more rapid (forty minutes instead of several days), the product was much cleaner, and the yield of practically pure lactone I1 was 82%.* It was found that the decarbethoxylation step could be accelerated considerably by the use of hydrobromic in place of hydrochloric acid. Robinson and Slaterg have prepared the lactone I1 by the interaction of P-naphthylmagnesium bromide and ethyl levulinate. The reported yield of crude material was 3870. For the synthesis of l14-dimethylphenanthene, VIII, the reactions indicated in the accompany7 0 7 ing flow sheet were employed. This hydrocarbon has been synthesized previously by two different methods. Akin, Stamatoff and Bogert,’O using the Pschorr process, obtained it in about 25% over-all yield in six steps from 2,5-dimethylbenzyl chloride. Later Papa, Perlrnan and Bogertl’ Included in accompanying papers is a descrip- in the same laboratories announced a successful tion of the application of this chain-lengthening synthesis by the Bogert-Cook method. The picprocess to the ring ketones tetralone-l4 and 1- rate of VI11 thus was obtained in five steps from keto- 1,2,3,4-tetrahydr0phennanthrene.~ The pres- bromo-p-xylene (and o-methylcyclohexanone) in ent work deals with the application of the process an over-all yield of about 21%. I n the present to an open-chain ketone, 2-acetylnaphthdene. work VI11 was obtained in 54% over-all yield in six The preparation of 2,4-dimethylphenanthrene steps from 2-acetylnaphthalene, I. The lactone from the resulting lactone I1 is also described as I1 was cleaved with warm dilute sodium hydroxtypifying a use of the chain-lengthening process ide, and the aqueous solution of the salt of the hyin polycyclic aromatic synthesis. The general droxy acid I11 was hydrogenated easily over plan bears some resemblance to Cook and Robin- copper-chromium oxide catalyst. In this way son’s6 approach to 8-methyl-l,2-benzanthracene,the crystalline 7-2-naphthylvaleric acid, IV, was but differs fundamentally a t the decarboxylation obtained in 90% yield. From this point the synstep. thesis was conventional. Cyclization with anThe condensation of 2-acetylnaphthalene with hydrous hydrogen fluoride proceeded smoothly diethyl succinate in the presence of sodium ethox- to give 4-keto-l-methyl-l,2,3,4-tetrahydropheide according to conventional procedures has al- nanthrene, V, in 93% yield. The addition of ready been described.’ When the crude mixture methylmagnesium iodide to this ketone resulted of isomeric half-esters produced in this way was in a good (86%) yield of the crystalline 1,4-ditreated with a refluxing mixture of acetic and hy- methyl - 4 - hydroxy - 1,2,3,4 - tetrahydrophedrochloric acids, carbon dioxide was evolved nanthrene, VI, which on dehydrogenation over slowly and the crystalline ymethyl-y-(2-naph- palladium-charcoal catalyst afforded ld4-dimeth(1) Present address: National Mineral Company, Chicago, Illiylphenanthrene, VIII, in a yield of 91%. T h e nois.
I n this communication we wish to introduce procedures for (1) the use of potassium t-butoxide in the Stobbe condensation2 and (2) the acidcatalyzed decarbethoxylation of the resulting half-ester (B) to give the y-lactone or derivative (C). The combination of these two steps followed by reduction (step 3) affords a method of executing an important synthetic process : namely, the introduction of a propionic acid residue a t the site of the carbonyl group (A) of a ketone.a The general scheme is represented as
(2) The alkoxidekatalyzed aldol type of condensation of succinic ester with a ketone (or aldehyde) was discovered and extensively studied by Stobbe and his collaborators, and is appropriately referred to as the “Stobbe condensation.” (3) For a brief summary of previous methods for effecting this transformation see Haberland and Heinrich, Bn.,7¶B, 1222 (1939). (4) W. S. Johnson, H. C. E. Johnson and J. W. Petersen, T m s JOURNAL, 67, 1360 (1945). (5) Johnson and Petersen, ibid., 67, 1,366 (1945). (6) J. W. Cook and Mrs. A. M. Robinson, J . Chem. Soc., 505 (1938). (7) Johnson and Goldman, THIS JOURNAL, 66, 1030 (1944).
(8) An exhaustive study of the modified Stobbe condensation and the decarbethoxylation reaction is now in progress. The.results of experiments with a number of ketones indicate that potassium f butoxide is s superior reagent for the condensation. For example, Gilbert Stork and Chester E. Davis have found that with cyclohexanone 84% yieIds of the condensation product can be obtained: whereas by the conventional procedure the yield is reported to be about 40%: Alder, Pascher and Schmitz, Bcr., 76B, 27 (1943). For other examples see refs. 4 and 5. (9) Robinson and Slater, J . Chcm. Soc.. 376 (1041). (10) Akin, Stamatoff and Bogert. THISJOURNAL, 65, 1268 (1937). (11) Papa, Perlman and Bogert, ibid., 60, 319 (1938).
1358
WILLIAM
s. JOHNSON,ARTHURGOLDMAN AND WILLIAM P. SCHNEIDER
melting points of the hydrocarbon and picrate were in agreement with those previously reported.losll
Vol. 67
cation of the process described in the present work. The reduction step was carried out with the unsaturated acid instead of with the lactone. Experimentali4
-&.--
~-Methyl-r-(2-naphthyl)-butyrolactone(11). (aj B y the Sodium Ethoxide-Cutalyzed Stobbe Coiulensation.-The best results were realized when the Stobbe condensation was conducted in ether solution for five days with alcohol free sodium ethoxide according to a procedure already described in deI I1 tail.' The crude oily mixture of half-esters obtained OH from 12.5 g. (0.25 mole) of 2-acetylnaphthalene and 43.5 g. (0.25 mole) of diethyl succinate and 0.5 mole o'H of sodium ethoxide, was treated with 250cc. of acetic and 150 cc. of concentrated hydrochloric acid. --., acid After refluxing for fifty-two hours (mercury trap) the solution was concentrated under reduced pressure. The residue was taken up in benzene, washed with 111 (not isolated 111 11. water, followed by saturated sodium bicarbonate svn thesiq of i171 solution and dried over anhydrous sodium sulfate. CH? Acidification of the bicarbonate extracts gave 10.0 g. of intractable brown acidic material which melted over a wide range, and which was largely unaffected by further acid hydrolysis. The neutral fraction was distilled rapidly in a two-bulb flask, and the small forerun (0.5 g.) was discarded. The main fraction distilled at 185-195" (2-3 mm.) giving a yellow oil which rapidly set to a light yellow solid, m. p. 74-77 ",yield 39.0 g. (69%). Recrystallization I I from benzene-petroleum ether (60-68 ") gave colorless plates, m. p. 76-77' ( r e ~ o r t e d ,77"). ~ That this was truly ~-methyl-y-(2-1iaphthyl)-butyrolactone follows from the reactions described below in connection with the synthesis of 1,l-dimethylphenanthrene. When the time for the Stobbe condensation was shortened to three days the yield of crude lactone, 111. p. 72-75', was 65%. In another experiment the VI 11 VI I condensation was carried out a t refluxing temperature for nine hours and the yie1.d dropped to %yo. When a refluxing (nineteen hoitrs) alcoholic solution of Additional evidence for the structure of the sodium ethoxide was used for the condensation according ketone V was afforded by dehydrogenation over to a procedure already described,' the yield of lactoile, palladium-charcoal in a tube sealed under nitro- ni. p. 74-76', was 57%. gen according to a procedure of Fieser and Joshel.'* ( b ) By the Potassium t-Butoxide-Catalyzed Stobbe ConThe product was shown to be the expected 1- densation.-To a refluxing solution of 2.15 g. (0.055 mole) of potassium in 45 cc. of &butyl alcohol in an atmosphere methylphenanthrene, VI I, by comparison with of nitrogen, was added a mixture of 8.50 g. (0.05 mole) of an authentic specimen. 2-acetylnaphthalene and 13.1 g. (0.075 mole) of diethyl Since the modified Stobbe condensation seems succinate, an additional 10 cc. of t-butyl alcohol being used to be generally applicable to the synthesis of lac- l o complete the transfer. After refluxing for forty minutes the orange-colored solution was chilled. 10 cc. of contones like I1 (or the corresponding unsaturated centrated hydrochloric acid in 50 cc. of water was added, acids) from aryl ketones like I,* the scheme of re- and the alcohol was removed under reduced pressure. The ctctions described above promises to be useful in light yellow oily residue was taken up in ether, and was the synthesis of certain types of polycyclic com- washed with 5% potassium carbonate solution. Acidification of the aqueous layers liberated the half-ester as a light pounds which hitherto have not generally been yellow oil which was extracted with ether, washed with easily obtained through the conventional cycliza- water followed by saturated salt solution, and dried over tion process, because of difficulty experienced in anhydrous sodium sulfate. The half-ester obtained after obtaining the requisite y,y-disubstituted butyric complete removal of the ether by warming under reduced was treated with 45 cc. of acetic acid (distilled acid (like IV). In preparing such acids for use in pressure from potassium permanganate), 30 cc. of 48% hydropolycyclic' syntheses, previous workerst3 have bromic acid aiid 15 cc. of water. After refluxing for three usually eniployed the reactions of the Grignard and one-quarter hours a total of about 1200 cc. of gas was reagent with a y-keto ester, an approach which is collected (over saturated sodium chloride solution), and rate of evolution had become quite The acid not generally satisfactory, and which in a t least the solvents were then removed under reduced pressure and one instance has been reported to fail altogether.l k the residual oil was taken up in ether and washed with 5% The synthesis of lJ2,3,10-tetrahydroperinaph- potassium carbonate solution to remove acidic material. thanone-74 affords another example of the appli- Evaporation of the dried (over anhydrous sodium sulfate), pale yellow ether solution afforded 8.12 g. of almost color(12) Fieser and Joshel. TEXSJOURNAL, 61, 2958 (1939). less crystalline lactone, ni. p. 65-72", (13) See for example (a) R. D. Haworth, et a [ . , J . C h o n . Soc.. The acidic material which was separated by ether ex1784, 2248 (1932); i b i d . , 454, 864 (1934). fb) Fieser and Johnson. traction of the acidified carbonate wtrarts amounted to Tms JOURNAL, 61, 1847 (1939). (e) Fieser and Joshel. i h i d , 61, ~
,.
407.
39.5s i l w l r r i
( A ) O r n M atid R w w t ;hid
63, l ? i (1041!
(1
:
411 mt.llinr [,,,inti
.iT'.
n~r,vv..
