Organoselenium chemistry. Conversion of cyclic ketones and .beta

Jul 1, 1974 - J. Org. Chem. , 1974, 39 (14), pp 2133–2135. DOI: 10.1021/ ... A Cascade Route to the Lycopodium Alkaloid (−)-Huperzine A ... Accoun...
7 downloads 0 Views 351KB Size
J. Org. Chem., Vol. 39, No. 14, 1974 2133

Communications J. F. Coetzee and G . R . Padmanabhan, J , Amer. Chem. SOC., 87, 5005 (1965). I . M. Kolthoff and M. K. Chantooni, J . Amer. Chem. SOC.,95, 8539 (1973). On the other hand, acetonitrile has two (surmountable) disadvantages: anthrone tautomerizes spontaneously in this solvent and anthranol hydrogen bonds weakly to acetonitrile. See also P. W. Arana, C. W. Su, and J. W. Watson, Chem. Commun., 363 (1970). No statistical correction for Dabco was made. F. G. Bordwell and W. J. Eoyle, d . Amer. Chem. SOC.,93, 511 (1971). P. Haake, R. D. Cook, and G. H. Hurst. J . Amer. Chem. SOC., 89, 2650 (1967). A qualitative explanation for the reactivity of weak oxygen bases in terms of Linnett theory may be found in the Ph.D. thesis of R. F. Williams (see footnote 7 ) . The magnitude of 1.47 pprn may be gauged from the 2.57-ppm shift of sym-tetramethylguanidine relative to that of N.N-dirnethylacetamide (the strongest and weakest bases recorded in ref 3) Phenoxide is a stronger base than triethylamine in acetonitrile by more than 8 pKa units: J, F. Coetzee. Progr. Phys. Org. Chem., 4, 45 (1967). H. Baba, A. Matsuyama, and H. Kokubun, Spectrochim. Acta, 2 5 A , 1709 (1969). have found that p-nitrophenol and triethylamine form solvent-separated ion pairs in acetonitrile. However, anthranol (with a pK. in acetonitrile probably six units greater than that of p-nitrophenol)17wouldnot be expected to form ion pairs. F. M. Menger and A. C. Vitale, d. Amer. Chem. SOC., 95, 4931 (19731. Recipient of a Camille and Henry Dreyfus Foundation TeacherScholar Grant and a National institutes of Health Research Career Development Award.

Department of Chemistry Emor)) University Atlanta, Georgia 30322

F. M.Menger*20 R. F. Williams

Scheme I

6

0 1. O,, -78'

1. LiNR, A

2 PhSeBr

h P e S & ,

___3

U

2 . 25'

1

0

2. 48%

3.17%

1 O,, -78'

25"

- 3

1

51%

2 . LiNR,, -78' 3. PhSeCl

5

tions we have tried. If the oxidation is carried out using sodium metaperiodate buffered with sodium bicarbonate, 33 and 48%, respectively, of the vinyl selenides analogous to 3 are formed. The occurrence of a facile Pummerer reaction depends on the acidifying effect of the carbonyl group on the a proton. Hence it is not surprising that the ketal 6 undergoes oxidation** and eliminationsb to enone ketal 7 in

Received February 20, 1974

Organoselenium Chemistry. Conversion of Cyclic Ketones and P-Dicarbonyl Compounds to Enones

Summary: The selenoxide syn elimination method for the synthesis of enones has been extended to the preparation of a-dicarbonyl enones, cyclobutenones, and enone ketals; and an important limitation to the method has been found.

6,81%

7. 8@$

good yield. 2-Phenylselenocycloheptanonecan similarly be converted to the ethylene ketal of cycloheptenone in 68% yield. Table I shows several examples of the preparation of cyclopentenonesga and cyclobutenes. The great facility with which cyclobutanones undergo Baeyer-Villiger oxidation necessitates the use of ozone as oxidantlo for the preparation of 3-phenyl-2-cy~lobutenone.~~ Scheme I1 presents two examples which illustrate the ability to trap copper enolates with PhSeBr for the synthesis of @-substituted enones. The extension of the selenoxide elimination to the synthesis of enediones from P-dicarbonyl compounds is an important one, since such transformations are difficult using classical methods.12 The dehydrohalogenation in particular often fails because of instability of halodicar-

Sir: The syn elimination of selenoxides has been shown to be a convenient, mild, and high-yield method for the preparation of a$-unsaturated carbonyl compounds.lg2 The precursor a-phenylselenocarbonyl compounds can be prepared from ketones, aldehydes, and esters,lJ as well as from enol acetateslbX2b and acetylenes.Ib We describe here some limitations of the method not heretofore recognized, as well as extensions to four-membered rings and P-dicarbony1 compounds. The necessity for achieving a cyclic transition state in Scheme I1 the selenoxide elimination3 may impose conflicting con0 0 formational demands on cyclic systems, and in fact only a limited range of cyclic enones (five- and six-membered rings) have been prepared. Our inability to achieve a high yield transformation of 2-methyl-6-phenylselenocylohexanone (1) to the enone (2)la led us to examine this reaction in more detail (Scheme I). The formation of by-products 3 and 4 can be rationalized as resulting from a Pummererlike reaction of the ketoselenoxide. 2-Phenyl-6-phenylsele1 hfeZCaLir H,O; nocyclohexanone6 also gives only a fair yield of enone, but + Ph Ph Ph the isomeric 2-phenylseleno compound, in which the phe2 PhSeBr" nyl substituent prevents the Pummerer reaction, gives SePh 88% enone in high yield (Table I ) . Scheme I also presents an 8.3% alternate synthesis of the vinyl selenide 3 by selenenylaa T o t a l q u a n t i t y of P h S e B r used was 10% excess over RLi used tion of the ketoselenoxide. in t h e p r e p a r a t i o n o f t h e cuprate. A s m a l l a m o u n t of PhzSez was Both 2-phenylselenocycloheptanone and -cyclooctanone a d d e d t o suppress f o r m a t i o n of a - h a l o ketones. B o t h cis a n d give only small amounts of enone (5-15%) under all condit r a n s isomers (1:4.5)appeared t o give enone. Reference 8a.

+

2134 J. Org. Chem., Vol. 39, No. 14, 1974

Communications

Table I Preparation of a,p-Unsaturated Carbonyl Compounds Compd

Olefin

bPh %" d"'" &Ph

bPh &f

-----Yield,a Selenide

Ph

aPh flo

Ph

86'

CO,CH,

0

& bPh bPh

Mb"

95d

5544

J-ph

87'

go

47'

n =!

96'

n = i

90'

&x F

'

h

93d 93d 281 81."1,

(CH,),

v

ad

Ph

ovcodcHJ h

94d

Ph

CO,cn,

P

Olefin

6Ohd

Ph

&

%-----

'

90

v R= =H n-Pr

93 96'

EOd 93"

a All compounds were fully characterized by spectral methods. * Overall yield. Selenide was prepared by the reaction of lithium enolate (LiN-i-Pr2)with PhSeBr or PhSeCl a t -78". Oxidation of selenide with H202in CH2C1,; see ref Sa. "elenide was prepared by reaction of lithium enolate (from cleavage of enol acetate with MeLi) with PhSeBr a t -78". One pot procedure described in ref la. 0 Selenide prepared by reaction of enol acetate with PhSe02CCF3; see ref lb. Oxidation by ozonolysis in CH2C12at -78", followed by warming to 25". Selenide prepared by the reaction of sodium enolate (NaH) in T H F with PhSeCl or PhSeBr. j A 95: 5 mixture of geometric isomers is formed.

bonyl compounds or enones under the reaction conditions. We have found the method to work superbly for the eight, seven and six-membered 2-carboethoxycloalkenones,a result which underscores the conclusions reached above that reactions involving the acidic a hydrogen, were responsible for the failure to achieve high yield syntheses of cylooctenone or cycloheptenone itself. Hydrogen peroxide cannot to be used as oxidant for the five-I3 and six-membered cyclic ketones, since rapid epoxide formation and further degradation occurs. Here ozonolysis a t -78" followed by warming is the best procedurelo (elimination occurs at or below -10"). An important consequence of the mild reaction conditions is that in all cases exclusively nonenolized P-dicarbonyl enones are formed, even though a number of these systems are known to be significantly or even predominantly enolic a t equilibrium.12 Other synthetic methods invariably give a mixture of keto and enol forms. The preparation of a-phenylseleno-/3-dicarbonyl compounds is conveniently carried out a t room temperature by the addition of ketone to a suspension of NaH (excess) in THF. When hydrogen evolution is complete (