TERRY G. SELINAND ROBERTWEST
1860 [CONTRIBUTIOS FROM
THE
voi. a4
DEPARTMENT OF CHEMISTRY, UNIVERSITY OF WISCONSIN, MADISON 6, WISC.]
Stereochemistry and Mechanism of Silane Addition to Olefins. I. Addition of Trichlorosilane to 1-Methylcyclohexene Under Free Radical and Thermal Conditions1 BY TERRY G. S E L I NAND ~ ROBERTWEST RECEIVED OCTOBER19, 1961 T h e addition of trichlorosilane t o 1-methylcyclohexene initiated with peroxides or ultraviolet light proceeds stereoselectively yielding principally cis-1-methyl-2-( trichlorosily1)-cyclohexane. Thermal additions with and without free radical inhibitors were less selective and yielded some (cyclohexylmethy1)-trichlorosilanein addition to cis- and trans-l-methyl-2(trichlorosily1)-cyclohexane. Based on the observed stereochemistry, mechanisms are proposed for the various reactions.
Recently the stereochemistry of trichlorosilane addition to acetylenes was studied by Benkeser and Hickner3 for both peroxide-initiated and platinum-catalyzed additions. The resolution and characterization of the various cis- and transmethylsilylcyclohexanes4 have now made it possible to extend stereochemical studies to the addition of trichlorosilane to the olefin l-methylcyclohexene. I n these experiments trichlorosilane was allowed to react with 1-methylcyclohexene usually in the presence of a catalyst or a free radical initiator, and the resulting alkyltrichlorosilanes were reduced with lithium aluminum hydride to the methylsilylcyclohexanes. These alkylsilanes were then resolved and analyzed by vapor phase chromatography. Preparative vapor phase chromatography provided samples of the products for final characterization by comparison with known isomers. Free Radical Additions.-Previously i t has been shown that the free radical addition of hydrogen bromide to 1-methylcyclohexene proceeds with trans addition resulting in cis-1-bromo-2-methylcy~lohexane.~This stereospecific trans addition is explained by assuming an intermediate (I) in which the bromine atom is centrally located between the ethylenic carbon atoms similar t o the bro-
/c\T/c\ Br
one might expect the trans isomer to predominate on purely thermodynamical grounds,? these results indicate that the mechanism involves a preferred trans stereochemical course as observed in hydrogen bromide additions. The preponderant formation of the cis isomer could be rationalized in terms of an analogous cyclic intermediate (11) followed by addition of silane hydrogen from the opposite side of the bridge. H H
/
.Si,
I1
'+
/
.Si,
However, unlike the additions of hydrogen bromide to 1-methylcyclohexene, a relatively large amount of the trans isomer resulting from cis addition was also obtained. T o explain the presence of this isomer i t is necessary to suppose the reaction can also proceed to a lesser degree through an open free radical intermediate such as 111. The incoming trichlorosilyl group would m ,A
qcH3 Si,- B
I11
I be expected to assume the less hindered equatorial monium intermediates involved in ionic additions position making paths A and B equally likely for of bromine t o olefins.6 The second step would the chain propagation step. Path A would result then involve abstraction of a hydrogen atom from in cis isomer, while path B would yield the observed hydrogen bromide from the opposite side of the trans isomer. The intermediate I11 would allow bromine bridge. the bulky trichlorosilyl group to assume an equaThe addition of trichlorosilane to l-methylcyclo- torial configuration rather than one above or below hexene in the presence of acetyl peroxide proceeded the plane of the ring as in 11. Since intermediate in quantitative yield. The product after reduction I11 would result in both cis and trans products, with lithium aluminum hydride was analyzed somewhat less than the observed 85y0cis-l-methylby vapor phase chromatography and found to 2-(trichlorosilyl)-cyclohexane would be due to consist of 1570 trans-1-methyl-2-silylcyclohexane reaction by means of the cyclic intermediate 11. and 85% cis-1-methyl-2-silylcyclohexane. Since However, an alternate mechanistic interpretation is possible in which the cyclic intermediate I1 (1) This research was supported by the United States Air Force through the Air Force O5ce of Scientific Research of the Air Research need not participate a t all. In the chain propaend Development Command, under Contract No. AF 49(638)-285. gation step hydrogen has to be abstracted from Reproduction in whole or part is permitted for any purpose of the trichlorosilane. A possible transition state for United States Government. this step is represented by IIIa in which the bulky (2) General Electric Co.,Silicone Products Department, Waterford, N. Y. trichlorosilyl group assumes an equatorial position (3) R . A. Benkeser and R. A. Hickner, J . A m . Chcm. S o c , 80, 5298 (1968). (4) T. G. Selin and R. West, ibid., 84, 18513(1962). (5) H. L. Goering, P. I. Abell and B. F. Aycock. ibrd.. 74, 3588 (1952). ( 6 ) I. Roberts and G. E. Kimball, ibid., 69, 947 (1937).
(7) The more thermodynamically stable isomer should be the one containing fewer skew interactions in the molecule. This stability could not be demonstrated experimentally for the methylsilylcyclohexanes for lack of a means for cis-trans equilibration. See reference 14 in preceding paper.4
May 20, 1962
TRICHLOKOSILANE ADDED TO 1-METHYLCYCLOHEXENE
1861
view is consistent with the low energy of the siliconhydrogen bond and is supported by the report that thermal cross-linking of vinyl siloxanes by disilylbenzene is inhibited in the presence of chloranil.ll However, kinetic studies by BarryI2 indicate that these reactions are bimolecular suggesting an ionic mechanism as opposed to a radical chain process. The stereochemistry of the thermal addition of trichlorosilane to l-methylcyclohexene was determined in an effort to shed further light on the mechanism. Trichlorosilane added quantitatively to l-methylcyclohexene a t 300’ (sealed tube) in the absence of added catalyst. After reduction, gas chromatrichlorosilane molecule, interactions with the axial tographic and infrared analysis of the silane mix1,3-ring hydrogens may hinder approach from the ture indicated that three isomers were present. top. The more facile approach of trichlorosilane In addition to 23y0 trans-1-methyl-2-silylcyclofrom the bottom would give the observed higher hexane and 58% cis-l-methyl-2-silylcyclohexane, yields of the cis isomer. I n general, the observed 19% (cyclohexylmethy1)-silanewas also observed. ratio of cis to trans isomers would be a measure of This rearranged adduct is the principal product the degree with which each approach is favored. obtained from addition of trichlorosilane to 1Stereoselectivity has also been observed in the methylcyclohexene in the presence of chloroplatinic free radical addition of trichlorosilane to acetylenes3 acid13 (Table I). The presence of this terminal hlthough a direct comparison between additions adduct as well as the higher proportion of trans to olefins and acetylenes cannot be made, i t is isomer distinguish the thermal addition products interesting that both proceed selectively, but not from those of peroxide and light-induced reactions. entirely, by trans addition. The free radical The increased amount of trans product could be addition of trichlorosilane to 1-methylcyclohexene rationalized in terms of a free radical mechanism. is somewhat more stereoselective (cis-trans ratios The possible bridged intermediate I1 might be of 6: 1) than additions to acetylenes (cis-trans thermally unstable, so that a greater proportion of ratios of 3 : 1).9 the reaction would proceed through the open interSimilar stereoselective results were obtained mediate I11 a t high temperature, leading to inwhen the reaction was initiated with ultraviolet creased amounts of trans isomer. Alternatively, light. The products from ultraviolet light-in- if an open intermediate such as IIIa were involved duced reactions consisted of 11% trans-l-methyl-2- exclusively, the approach of trichlorosilane might (trichlorosilyl)-cyclohexane and 89% cis-l-methyl- be less selective a t higher temperature. However, 2 - (trichlorosily1)-cyclohexane. However, the total the presence of (cyclohexylmethyl)-silane is quite yield of adducts was much less than for compara- difficult to explain by a free-radical mechanism. tive reactions utilizing peroxides (cf. Table I). This product does not appear to result from rearrangement followed by addition, because i t was TABLE I found that 1-methylcyclohexene underwent no YIELDS AXD PRODUCTS OBTAINEDFROM TRICHLOROSILANE isomerization to methylenecyclohexane when heated ADDITIONSTO ~-METHYLCYCLOHEXEKE to 280’ for comparable lengths of time. ReacProduct composition, % I t therefore seems probable that both freetion Tertime, Yield, 1,21,2minal radical and ionic mechanisms are operating simulConditions hr. Yo ris &am adduct taneously in thermal addition. If ionic addition Acetyl peroxide 9 100 S5 15 .. of trichlorosilane to olefins exhibits the Sam? Ultraviolet light 44 49 s9 11 ,. stereochemistry as observed for acetylenes,3 an Chloroplatinic acid S is .. .. 98“ increased amount of cis addition to give trans Thermalb 24 100 58 23 19 product would result if ionic catalysis took place. Thermal, chloranil 36 100 57 25 18 In fact, an ionic reaction involving carbanion T h e r m a l , F e a n d F e C l s 36 93 44 35 21 intermediates would provide a pathway both to I n addition t o t h e product listed, trace amounts of increased amounts of trans isomer and to terminal various methylsilylcyclohexane isomers were observed. * Thermal additions were carried out in sealed Pyrex tubes adduct. and the free radical carbon becomes trigonal in structure.* Because of the size of the incoming
at a temperature of 300”.
Thermal Addition.-It is generally acceptedlo that thermal addition of silanes to olefins takes place through free radical processes. Such a (8)There is some question as t o t h e structure of carbon free radicals. A planar structure is assumed for simplicity. (e) R. A. Benkeser, M. L. Burrous, L. E. Nelson and J . V. Swisher. J. Am. Chem. SOC.,83, 4385(1962). (lo) (a) C. L. Agre and W. Hilling, ibid., 74, 3895 (1952); (b) G. Fritz, 2. Kalurfof‘sch., 7B, 507 (1952); (c) R. S. Haszeldine and R. J. Marklow. J . Chcm. SOL., 962 (1956); (d) M . G. Voronkov, h’. G. Ramanova and L. G. Smirnova, Chcm. Lisly, 64, 040 (1958); C. A , , 62, 13615 (1958); (e) D. G. White and E. G. Rochow, J. A m . Chcm. Soc., 7 6 , 3897 (1954).
CH,
(5 t
HSlCI,
I-;
- - -f
g2sic13
+ jL
-
t ISlCI,I (11) C. L. Segal and J. B. Rust, “Abstracts of Papers of 137th Meeting, American Chemical Society,” 1960, p. 37M. (12) A. J. Barry, “Abstracts of Papers of 137th Meeting, American Chemical Society ” 1960, p 19M. (13) The stereochemistry and mechanism of additions catalyzed by chloroplatinic acid will be t h e subject of the third paper in this series; T. G. Selin and R. West, J. Am. Chcm. SOC.,82, 1863 (1962).
TEKRV G . SELIS.\NU KOHEKTL~-EST
1862
A similar mechanism has been proposed previously for platinum-catalyzed silane additions. l 4 < l 5 In an attempt to check the duality of mechanism proposed for the thermal reactions, several thermal additions were carried out in which free radical inhibitors were added to the reaction mixture. This approach served a twofold purpose since it not only illustrated the presence of a free radical path but also indicated that the trans isomer was arising in part by a non-radical mechanism. Chloranil did not change the product composition, perhaps because i t reacted with the trichlorosilane a t the high reaction temperature. However, the use of iron and ferric chloride16 as a free radical inhibitor did alter the isomeric coniposition of the product. From Table I it can be seen that increased amounts of both trans1 - methyl - 2 - silylcyclohexane and (cyclohexylmethyl)-silane were obtained a t the expense of the cis isomer. This suggests that the free radical path is partially inhibited, allowing more efficient ionic competition for the olefin. Other Catalysts.-Several other catalysts were used for the addition of trichlorosilane to 1methylcyclohexene without success. Thus, while pyridine has been used successfully with simple olefins,l 7 no reaction was observed between trichlorosilane and 1-methylcyclohexene a t temperatures up to 180' in the presence of this amine. Similarly, the use of iron pentacarbonyl was fruitless a t temperatures up to 120'. Triethoxysilane would not add to 1-methylcyclohexene under mild coiiditions in the presence of peroxides or chloroplatinic acid. Experimental .I11 vapor pliasc chrornatography was carried out in a 25 ft. X 0.25" copper column packed with 30Yc DC-550 Silicone oil on 30 mesh firebrick a t 100' with a flow rate of 60 ml. of helium per minute. Pure samples of each of the silane products werc obtained by preparative gas chromatography. These purified products were then characterized by comparison of their physical properties with those of aut lientic samples. Starting Materials.-The 1-methylcyclohexcnc was prepared according t o the method of Bartlett and Rosenwald18 in 51yo yield. Gas chromatography showed t h a t the product was contaminated by less than 1% of isomeric inipurities. Trichlorosilane was purchased from Union Carbide Co., and redistilled prior to use. Triethoxysilane was preccording to the method of Havill, Joffe and Postlg yield and cyclohexylmethyl bromide was prepared from cycloliexylcarbinol and plinsphorus tribromide in 53%) ;\+Id. The properties of all nf these compounds agreed with those reported in the l i t e r a t ~ r e . 5 ~ ~ ~ ~ ~ ~ Preparation of Standards.-Both cis- aiid trans-l-niethyl2-silylcyclohexane were prepared by hydrogenation of otolyltriethoxysilane followed by lithium aluminum hydride reduction. Cyclohexylmethylsi1ane.-A Grignard reagent was prepared in ether from 18 g. (0.10 mole) of cyclohexylmethyl hroniide. This reagent mas added slowly with stirring t o ,51 g . (0.3 mole) of silicon tetrachloride dissolved in 300 ml. ~~
~~~
~-
(14) J C. Saam and J, I, Speier. J . A m . Chem. S O ~ 80, . , 404 (1958). (I.-)) J I.. Speier, J . A . Webster and G. H. Barnes, ibid., 79, 974 i 1!1.-)7) r l l i l J 1.. Speier and J . A . Webster, J . O l e . ( ' h e m . , 21, 1014 (1956). f l i i S Nozakura and S. Konotsune, Bull. Chem. S u r . J a p a n , 29, :!2P IlSriS) ( 1 8 ) P I ) Bartlett and R . H. Rosenaald, J . A m . Chen?. Soc., 5 6 , l!J!lO , 19:34) ( 1 0 ) 31 I < I i a v i l l , 1 Joffe and H W. Post, J . Ovg. C ' h m z , , 13, 2x0 f 1!!481
120)
(;
S IIiers ilnd I< Ad,
