Dec. 5 , 1958
PYROLYSIS OF XANTHATE AND SULFITEESTERS OF BUTANOLS [CONTRIBUTION FROM THE CHEMICAL
6353
LABORATORY OF NORTHWESTERN UNIVERSITY]
Elimination Reactions. X. Pyrolysis of Xanthate and Sulfite Esters of erythro- and thueo-3-p-Tolylthio- and 3-p-Tolylsulfonyl-&butanols BY F. G. BORDWELL AND P. S. LAND IS^ RECEIVED MAY26, 1958 Pyrolysis of the S-methyl xanthates of dl-threo- and dl-erythro-3-p-tolylthio-2-butanols gave cis- and trans-2-p-tolylthio-2butenes, respectively (stereoselective cis eliminations). I n contrast, pyrolysis of the S-methyl xanthates of dl-threo- and dl-erythro-3-p-tolylsulfonyl-2-buta1iols gave the same product, cis-2-tolylsulfonyl-2-butene (the more stable isomer). The latter results are interpreted in terms of a stepwise mechanism involving a dipolar intermediate, which can undergo rotation both gave cis-2-P-tolylprior t o forming olefin. The methyl sulfites of dl-threo- and dl-erythro-3-p-tolylsulfonyl-2-butanols sulfonyl-2-butene on pyrolysis, probably by a similar mechanism. The methyl sulfites of dl-threo- and dl-erythro-3-p-tolylthio-2-butanols also gave essentially the same pyrolysis product, but this is believed to be due to an acid-catalyzed rearrangement of one of the olefins.
It is generally agreed that the Chugaev reaction oxidized to the corresponding sulfones which were and related ester pyrolyses usually proceed by way identified by infrared analysis.6 The product from of a concerted path involving simultaneous removal pyrolysis of the threo isomer I after oxidation was (111)) whereas trunsof the ester grouping and a cis-P-hydrogen.2 cis-2-p-tolylsulfonyl-2-butene However, only one example demonstrating the 2-p-tolylsdfonyl-2-butene (IV) was obtained from stereoselective cis nature of ester pyrolyses in the erythro isomer I1 under similar conditions. The acyclic systems appears to have been recorded thus analyses indicate that no more than 5y0of I V can farS2s3 In preceding papers in this series we have be present in the product from I. Similarly, I1 apshown that cis-2-p-tolylthiocyclohexylS-methyl H\ xanthate and methyl sulfite undergo pyrolysis t o give predominantly 3-p-tolylthio- 1-cyclohexene, the expected cis elimination p r ~ d u c t . ~I n contrast, cis-2-p-tolylsulfonylcyclohexylS-methyl xanthate U and methyl sulfite undergo first-order decompositions on heating to give l-p-tolylsulfonyl-l-cycloI hexene, the truns elimination product. It was sugSCH3 I11 gested that the latter reactions occurred by a stepI (threo) wise process involving a dipolar intermediate in pears to have given only IV, but i t would be difipreference to a concerted mechanism. I t was of cult to detect the presence of even as much as 10% interest to extend this study to an acyclic system, of I11 in this product. The eliminations are theresince it would be anticipated on the basis of these fore strongly stereoselective although they may not results that in comparable open-chain compounds be stereo~pecific.~ the analog containing the C7H7S group would The stereoselective cis pyrolytic eliminations pyrolyze by the usual concerted mechanism (stereofor I and I1 are to be expected on the basis of conselective cis elimination), whereas the analog concerted mechanism involving a quasi six-membered taining the C7H7S02 group would decompose by a ring.2 They resemble Cram's3 results on the pyrolstepwise mechanism in which stereoselectivity ysis of the S-methyl xanthates of 3-phenyl-2-butawould not be mandatory. no1 except that the pyrolysis of the threo isomer was The synthesis of the erythro and threo forms of less stereoselective in his system, and about 20% CH3CHOHCH(CH3)SGH7 and CH3CHOHCH(C- of 1-olefin was formed in each pyrolysis. The H3)S02C7H7,and establishment of their structures absence of bands in the 10.1 and 11.2 p region of the has been described p r e v i ~ u s l y . ~The same method infrared spectra of I11 and IV indicate that little or was used except that in the first step of the syn- none of the 1-olefin is formed in the pyrolysis of I thesis i t was found that the cis- and trans-2-butenes or 11. It would appear that the acidity of the acould be converted in high yield to the correspond- hydrogen and conjugative effects promoting foring bromohydrins by reaction with N-bromosuc- mation of the 2-olefin are stronger in these pyrolcinimide in an aqueous suspension a t OO.6 The S- yses than for the S-methyl xanthates of 3-phenylmethyl xanthates and methyl sulfites were pre- 2-butanol. pared in the usual manner. In contrast to the highly stereoselective pyrolyThe S-methyl xanthates of dl-threo- and dl- tic eliminations of I and 11, the corresponding sulerythro-3-p-tolylthio-2-butanol(I and 11, respec- fones, the S-methyl xanthates of dl-threo- and dltively) each decomposed smoothly when heated erytkro-3-p-tolylsulfonyl-2-butanol(V and VI, reunder vacuum a t 200-220O. The olefins were spectively), both pyrolyzed to give predominantly the cis-olefin 111. For the threo isomer V this is a (1) Socony-Mobil Predoctoral Fellow, 1955-1958. cis elimination which may be concerted, but for ( 2 ) See the discussion by D. J. Cram in Chapter 6 of "Steric Effects in Organic Chemistry," edited by M. S. Newman, John Wiley and Sons, which we prefer to write the stepwise mechanism Inc., New York, N. Y.,1966. shown.4 For the erythro isomer VI the over-all reJOURNAL, 71, 3883 (1949); D. J. Cram and (3) D. J. Cram, THIS sult is trans elimination. F. A. Abd Elhafez, ibid., 1 4 , 5828 (1952). (4) F. G. Bordwell and P. S. Landis, ibid., 80, 2450, 6379 (1958). (5) F. G. Bordwell and P. S. Landis, ibid., 79, 1593 (1957). ( 6 ) C. 0. Guss and R. Rosenthal, ibid., 77, 2549 (1955).
(7) It seems advisable t o reserve the term stereospecific for reactions which can be demonstrated t o be nearly completely stereoselective (i.e., 99 =k 1%).
F.G.BORDWELL AKD P.S.LANDIS
(i:W
Yol.
so
support the previous suggestion that wheii the phydrogen is activated by a strong electron-withdrawing group (C~H~SOZ) a dipolar ion intermediate may be formed. A comparable dipolar intermediate (VIa) from VI could undergo direct decomposition to give IV (which might be preceded by inversion and 60" rotation to give VIb) or could rotate 120" to VIc prior to elimination which would give 111. The observed products therefore can be explained in terms of a carbanion intermediate which rotates prior to elimination. The intermediates from V and VI are apparently not identical since the proportions of I11 and IV V (threo isomer) IV (5%) The formation of 111from VI occurs by a primary from the two reactions are somewhat different. reaction and not by rearrangement of the cis elim- This may mean that carbanions are formed from ination product IV, since IV fails to rearrange to both V and VI9 (with several paths each to I11 and I11 when treated for as long as 108 hr. with concd. IV), or it may mean that V decomposes by a conhydrochloric acid in aqueous dioxane and also re- certed mechanism and V I by a stepwise mechmains essentially unchanged when mixed with the anism. The similarity in the rates of decomposithreo-xanthate V and the mixture pyrolyzed. It tion of I, 11, V and VI and of the corresponding was also established that rearrangement did not cis and trans isomers in the cpclohexyl series4argues occur during xanthate formation, since i t was shown for the first possibility. -4lthough attempts to crystallize I, 11, V and VI that the threo and erythro isomers could be converted to potassium salts and recovered unchanged were unsuccessful, rate measurements on the crude xanthates gave reasonably good first-order plots. on hydrolysis. Heating IV with one-twentieth equivalent of These data support the postulated mechanism potassium t-butoxide in t-butyl alcohol a t SO" for 30 since they show that the decompositions must be min. converted i t to a mixture of about 807p 111 intramolecular. The first-order rate constants (cis) and 20y6 IV (trans). After an additional 30 determined graphically from runs made a t 165 ==I min. IV was converted completely to the cis iso- 2' are: 5.3 X IO+ set.-' for I, 4.2 X loe5 set.-' mer. h separate sample of I11 remained un- for TI, 4.2 X lop5 set.-' for V and S.5 X 10-5 for changed on treatment with potassium t-butoxide 1'1. The rate constants for V and V I are slightly solution. The greater thermodynamic stability of greater than for the cyclohexyl analogs4 I t is inthe cis isomer 111 is expected, since the sulfonyl teresting to note that the sulfone xanthate VI, which is believed to decompose by a stepwise procgroup is known to be large.8 ess, pyrolyzes twice as rapidly as does the corresThe formation of I11 from VI (over-all trans elimination) is not unexpected in view of the fact ponding sulfide xanthate I1 where the reaction is that cis-2-p-tolylsulfonylcyclohexylS-methyl xan- believed to be concerted. Evidently the ease of thate undergoes tyans elimination on pyrolysis in formation of the dipolar intermediate from VI wins preference to cis elimination.2 The present results out over the concerted process, despite the energetic economy of the latter and the fact that oxida.SCH3 tion of the sulfur greatly increases the steric effect of the sulfur grouping. I n pyrolytic cis eliminations the advantages of the concerted mechanism are always offset to some extent by the increased repulsions arising by the requirement of coplanarity for the maximum assistance of bond-making to bond-breaking. These repulsions are probably often large enough to necessitate a compromise betureen these two factors. The stepwise mechanism derives no benefit from coplanarity of the groups, and a staggered conformation may be used for the transition state. This may be an important factor in making the rate of decomposition of VI faster than that of 11. Since the stepwise mechanism provides an advantage for the decomposition of VI relative to 11, it seems probable that V will also utilize this mechanism (as shown) in preference to the concerted one. The 8-fold and 5-fold rate advantages of V and I , respectively, over VI, can be attributed to the increased importance of steric hindrance in the decomposition of VI. VIb
. ..~____-
IV (20%)
I11 (SOYG)
( 8 ) F. G. Bordwell and G. D. Cooper, THISJOURNAL, 73, 5184 (lS51); i b i d . , 79, 916 (1957): E. I,. Eliel and R. S. Ro, ibid.,79, 5995 (1957).
(9) Such carbanions would probably maintain a tetrahedral configuration since there is evidence t h a t there is little or no angular reyuirement for conjugation involving a sulfonyl group (Iid overlap) [see S. Oae and C . C. Price, ibid., 80, 4938 119.58), and referenrvs cited therein].
Dec. 5, 1958
PYROLYSIS O F XANTHaTE AND SULFITE
ESTERSOF BUTA4NOLS
6385
Preparation and Pyrolysis of dZ-threo-3-p-Tolylsulfonyl-2I n the preceding paper in this series4 it was S-Methyl Xanthate.-Using a procedure similar to that shown that methyl sulfites often parallel S-methyl butyl described above, 12.4 g. (9570) of a viscous xanthate was xanthates as to products formed on pyrolysis, and obtained from 8.5 g. (0.037 mole) of alcohol, 1.8 g. of poi t was proposed that the two reactions proceeded tassium, 50 ml. of dry benzene and excess carbon disulfide by similar mechanisms. However, acidic by-prod- and methyl iodide. Decomposition of the xanthate gave g. (5870) of material, b.p. 145-155' (0.5 mm.). Infraucts are formed in methyl sulfite pyrolyses and 3.2 red analysis indicated that the product was primarily cis-?,sensitive olefins may be rearranged during the @-tolylsulfonyl-2-butene containing no more than 5% of the pyrolysis, whereas S-methyl xanthates rarely, if trans isomer. Preparation and Pyrolysis of dl-erythro-3-p-Tolylsulfonylever, give rise to rearrangement products. Pyrolysis of the methyl sulfite of dl-threo-3-p-tolylthio- 2-butyl S-Methyl Xanthate.-Decomposition of 6.0 g. of xanthate, prepared as described above, gave 3.2 g. of 2-butanol and oxidation gave I11 (stereoselective crude an oil b.p. 145-175' (2 mm.). This sample was chromatocis elimination). However, unlike the xanthate, graphed over activated silica gel using hexane and benzene the methyl sulfite of dl-erythro-3-p-tolylthio-2-buta-as successive eluents. Fractions 9 and 10, representing no1 also gave predominantly I11 (75%) (trans 82Y0 of the sample and an over-all 4870 yield, were identiby infrared analysis as mixtures of about 80% cis-2-pelimination product) under these conditions. fied tolylsulfonyl-2-butene and 2070 of the trans isomer. From our previous experience4i t seems highly probPyrolysis of th~eo-3-p-Tolylsulfonyl-2-butylS-Methyl able that rearrangement has occurred in this in- Xanthate in the Presence of Added trans-2-p-Tolylsulfonylof 6.5 g. of the xanthate in the presstance, the unoxidized form corresponding to I11 2-butene.-Pyrolysis ence of 2.0 g. of t~ans-2-p-tolylsulfonyl-2-butenegave 5.3 g. being the more stable product (according to this of a mixture containing about 4070 of trans-olefin. Theoretassumption). ically 307, of trans-olefin should ha\ e been present barring The results of the pyrolysis of the methyl sulfites rearrangement or formation of trans-olefin during the pyof dl-threo- and dl-erythro-3-p-tolylsulfonyl-2-buta-rolysis. Failure of Rearrangement of threo- or erythro-3-p-Tolylnols paralleled those from the corresponding S- sulfonyl-2-butanol on Treatment with Potassium.-The methyl xanthates. Since the sulfone I V has been potassium salts of the alcohols (2.7 g.) were prepared in shown to be very resistant to acid-catalyzed rear- benzene medium, as described above. T h e mixture conof brown granular potassium salt suspended in benrangement (see,above) these sulfites are probably sisting zene was transferred to a separatory funnel and shaken with decomposing by the stepwise mechanism utilized 30 ml. of lOol, hydrochloric acid. The benzene layer was by the sulfone S-methyl xanthates. This is not washed with water, dried over anhydrous magnesium sulfate surprising since cis-2-p-tolylsulfonylcyclohexyl and the solvent removed on the steam-bath. The infrared of the residue (SOY0 recovery) from the erythro methyl sulfite was found to undergo stepwise trans spectrum isomer gave no evidence for the presence of the threo isomer, elimination even more rapidly than did the corre- which has two strong doublets a t 10.18, 10.38 and 10.85, sponding xanthate.4,10 11.05 p . Similarly, the strong band a t 9.98 1.1, which is characteristic of the evythro-alcohol, was absent in the spectrum of the recovered threo-alcohol. Preparation and Pyrolysis of dl-threo-3-p-Tolylthio-2-butyl Rearrangement of trans-2-p-Tolylsulfonyl-2-butene.S-Methyl Xanthate (I).-dl-threo-3-p-Tolylthio-2-butanol Five hundred milligrams of trans-2-p-tolylsulfonyl-2-butene (9.8 g., 0.05 mole) was added t o a stirred suspension of 1.95 was refluxed for 30 minutes with a solution of 20 ml. of tg. (0.05 g. atom) of metallic potassium in 100 ml. of dry butyl alcohol t o which had been added 5 mg. of metallic benzene. T h e mixture was stirred for 2 hr. a t room tempera- potassium. The cooled solution was poured into 50 ml. of ill1 the potassium ap- water and extracted with chloroform. The chloroform exture and then for 2 hr. a t 50-55'. peared t o have reacted a t the end of this period. Carbon tract was washed thoroughly with water, dried over anhydisulfide (15 ml.) was added, the solution was refluxed for drous sodium sulfate, filtered and the solvent removed on a 8 hr. and 20 ml. of methyl iodide was added. The solution steam-bath. The resulting solid, 385 mg., was established by was stirred overnight, filtered and washed thoroughly with infrared analysis to be a mixture of 80% cis- and 207, transwater. After drying over anhydrous sodium sulfate, the 2-p-tolylsulfonyl-2-butenes. Further treatment of this solution was filtered and the solvent removed, first on a product (8070 cis- and 20% trans-olefin) with 20 ml. of tsteam-bath, then under vacuum. T h e residual oil weighed butyl alcohol containing 5 mg. of potassium followed by iso13.5 g. (94Y0). The crude xanthate (7.0 g., 0.022 mole), lation as previously described gave 260 mg. of pure ciswas placed in a 25-ml. vacuum distillation flask fitted with a olefin. Similar treatment of cis-2-p-tolylsulfonyl-2-butene 3-inch Vigreux head. T h e flask was heated a t 200" for 30 with potassium t-butoxide solution gave only unchanged cismin., vacuum was applied slowly and the product was dis- olefin. tilled a t 0.1 mm. keeping the pot temperature a t 200-220'. Kinetic Measurements.-The kinetic runs were made in a In this way 3 g. (77yo)of an oil, b.p. 135-147' a t 0.1 mm., series of tared erlenmeyer flasks heated to 165 zk 2" in an was collected. One gram of the sulfide was oxidized with oil-bath, as described previously.* The method is essen10 ml. of 3070 hydrogen peroxide in 20 ml. of acetic acid a t tially that of O'Connor and Nace." 80'. The resulting sulfone melted a t 36-45" Its infrared Preparation and Pyrolysis of Methyl erythrc- and t b eo-3-pspectrum indicated that the product was cis-2-p-tolylsulTolylthio-2-butyl Sulfites.-The methyl sulfite was prepared fonyl-2-butene containing less than 5Yo of the trans isomer, by a procedure similar to that described in the preceding since the strong 1 4 . i i.1 peak characteristic of the t r a m isomer5 paper.' Pyrolysis of 6.1 g. (0.0223 mole) of crude erythrowas absent. methyl sulfite under vacuum a t 200-220' gave 3.9 g. (9iyo) Preparation and Pyrolysis of dl-erytizro-3-p-Tolylthio-2of distillate. A 3.8-g. sample was oxidized with hydrogen butyl S-Methyl Xanthate (11).--Using quantities and pro- peroxide to give material melting a t 35-43'. Infrared cedure similar to that described for the lhreo isomer, 13.8 g. analysis indicated about 25Yo of trans-2-p-tolylsulfonyl-2( 9 6 7 , ) of crude S-methyl xanthate was obtained. Pyroly- butene and 7570 of cis isomer to be present. The strong sis of 9 g. (0.0314 mole) of this xanthate gave 3.2 g. (58%) 14.7 band of the trans isomer mas absent in a comparable of olefin, b.p. 120-130" (0.5 mm.). Oxidation of a small olefin sample obtained by pyrolysis of the threo-methyl sample gave a colorless solid, m.p. 45-47'. Infrared anal- sulfite. ysis6 indicated that the compound ivas primarily trans-2-pPyrolysis of Methyl erythro- and threo-3-p-Tolylsulfonyl-2tolylsulfonyl-2-butene, although as much as 10% of cis butyl Sulfites.-A 6.1-g. (0.02 mole) sample of crude erythroisomer could escape detection. sulfite was heated under nitrogen for 30 min. a t 170" and then the product was vacuum distilled (with concurrent de(10) As pointed out by a Referee, it is also possible to interpret composition) using a pot temperature of 180-190". In one the results of the sulfide methyl sulfite pyrolyses using ion pair mecha-
Experimental
nisms [see E. L. Eliel, J. W. McCoy and C. C.Price, J . Org. Ckcm., 22, 1533 (1957) and our previous paper41 to account for the lack of stereoselectivity.
( 1 1 ) G. L. O'Connor and H.
(1952).
R. Nace, THISJOURNAL, 74, 5454
HERBERT 0. HOUSE,j.WARRENBIAKERAND DALEA. XIADDEN
6386
hour 3.8 g. (91%) of material was obtained, b.p. 160-164" (0.4 mm.). Since the infrared spectrum showed a weak hydroxyl band, the product was chrornatographed over silica gel usiiig hexane as the solvent and benzene as the eluent. I n this way 2.6 g. (62%) of olefinic product was obtained which consisted of about 15:h of trans-2-p-tolyl-
[ CON~RIRUTIONFROM
THE
\'ol.
so
sulfonyl-2-butene and 85% of the c i s isomer. The pyrolysis of the threo-methyl sulfite anti product iqolxtion werr carried out as described for the erythvn Isoiner. 'l'lit. product obtained contained db
