J. Am. Chem. zyxwvut SOC. zyxwvu 1991, 113, zyxwvu 967-971 967 was prepared by LiAlH, reduction of the corresponding methyl ester (Aldrich). The remaining substrates were purchased from Aldrich, ex- cept trons-(CH,)CH=CHCH20H (Fluka, contained zyxwvutsrq -5% of the cis isomer). All substrates were distilled from 3-A molecular sieves under either argon or vacuum (0.1 mmHg) and stored under argon. The N,- zyxwvut 0-bis(trimethylsilyl)acetamide, p-toluenesulfonyl isocyanate, and tetra- cyanoethylene (TCNE) were used as supplied (Aldrich), except that T C N E was recrystallized from hot chlorobenzene before use. Preparation of CH2=€HCH20D. A solution of allyl alcohol (1 8.4 mL, 270 mmol) in dry Et20 (50 mL) was added dropwise over 2.25 h to EtMgBr (1.5 M in Et20, 200 mL, 300 mmol) under argon. The slightly cloudy mixture was then quenched by the addition of D 2 0 (16.2 mL), and the resultant mixture was refluxed for 0.5 h. The clear su- pernatant was removed by cannula and the white residue extracted with Et20 (2 X 75 mL, Et20 extract removed by cannula). The Et20 was removed by careful distillation and the residue distilled through a vacu- um-jacketed Vigreux column, yielding allyl alcohol-d (1.98 g, 1 1% yield, bp 98-99 "C). A 'H NMR (acetone-d,; 200-MHz) spectrum showed the residual OH to be -6%. Catalytic Runs. Typically, [Rh(diph~s)]~(ClO~)~ (0.6 mg, 9.987 X IO4 mmol) was dissolved in acetone-d, (0.6 mL) under argon. The substrate (9.987 X mmol, 100 equiv) was then added via syringe. The reactions were monitored by NMR spectroscopy. The [Rh(di- phosphine)(NBD)]+ was hydrogenated to the [Rh(diph~sphine)(sol)~]+ species as follows: the [Rh(diphosphine)(NBD)]+ complex was sus- pended in the solvent. and H2 was bubbled into the solvent for 3 min. The gas flow was stopped, and the NMR tube was shaken for 3 min. This procedure was repeated twice to give clear solutions of the solvent0 complexes. The solutions were then purged with argon, the substrate was added, and the catalysis was monitored by NMR. The results are sum- marized in Tables I and II. The 'H and 2H NMR spectra are presented in Table V. The IH NMR assignments are based on comparison with published spectra (where available), on analogies with the corresponding trimethylsilyl ethers, and on the usual principles of N M R spectroscopy. The relevant literature references are given in Table V. The enantiomeric excess from the [Rh((S)-binap)(S),]+ catalyzed ketonization of the enol 11 was determined by the method outlined previo~sly.~' (26) Schuetz, R. D.; Millard, F. W. J. Org. Cfiem. 1959, 24, 297. (27) Fraser, R. R.; Petit, M. A.; Saunders, J. K. J. Cfiem. SOC., Cfiem. Commun. 1971, 1450. (28) Capon. B.; Siddhanta, A. K. J. Org. Cfiem. 1984, 49, 255 (for enol). Heathcock, C. H.; Buse, C. T.; Kleschick, W. A.; Pirrung, M. C.; Sohn. J. E.; Lampe, J. J. Org. Cfiem. 1980, 45, 1066 (for trimethylsilyl ether). [R~(C~P~OS)(CH~=CHCH~CH~OH)](CIO~)*~. [Rh(cyphos)- (NBD)]CIO, (8.29 mg, 1.157 X mmol) was partially dissolved in acetone-d6 (0.6 mL) under argon. A cycle of bubbling hydrogen gas through the mixture for 3 min followed by shaking for 3 min was re- peated three times to dissolve and hydrogenate the catalyst precursor. Argon gas was then bubbled through the solution for 5 min to remove the excess of hydrogen gas. The solution was cooled to -40 "C, and CH2=CHCH2CH20H (0.77 mg, 1.068 X mmol, 0.93 equiv) was added to the mixture via syringe. Integration of the I'P NMR signals indicated that 85% of the Rh was in the form of the adduct 3; the remainder was an unidentified species, believed to be bridging hydride complexes observed when an excess of hydrogen gas is used to hydro- genate the NBD complex at high [Rh].IS This ratio is reflected in the 'H NMR spectrum as well. The NMR data are presented in Table V. IH NMR assignments are based on decoupling experiments and the usual principles of N M R spectroscopy. The compound was relatively stable, decomposing - 13% after 22 min at 20 "C. Chemical Reactions. The catalysis to generate the enol was performed as described above. Carbon monoxide was bubbled through the solution for 2 min to deactivate the catalyst at the point when all of the allylic alcohol was consumed (see Table I). The reagent (9.987 X mmol, 1 equiv) was then either injected (N,O-bis(trimethylsilyl)acetamide, p- toluenesulfonyl isocyanate) or added as a solid. The yields based on the amount of enol originally present in solution are summarized in Table VI. The NMR data are summarized in Table V. The assignments are based on decoupling experiments, deuterium labeling, the usual principles of N M R spectroscopy, and, where possible, comparison to published data. The relevant references are given in Table V. Acknowledgment. This work was supported by grants from the National Institutes of Health. (29) Hanack, M.; Markl, R.; Martinez, A. G. Cfiem. Ber. 1982, 115,772 (trimethylsilyl ether), (30) Schraml, J.; Sraga, J.; HrnEiar, P. Collect. Czech. Cfiem. Commun. 1983, 48, 3097 (for trimethylsilyl ether). See ref 3 for spectrum of enol. (31) Birkofer, L.; Dickopp, H. Cfiem. Ber. 1969, 102, 14 (trimethylsilyl ether). (32) Stang, P. J.; Mangum, M. G.; Fox, D. P.; Haak, P. J. Am. Cfiem. Soc. 1974, 96, 4562 (trimethylsilyl ether). (33) House, H. 0.; Czuba, L. J.; Gall, M.; Olmstead. H. D. J. Org. Cfiem. 1969, 34, 2324 (trimethylsilyl ether). (34) Rasmussen, J. K.; Hassner, A. J. Org. Cfiem. 1974, 39, 2558 (tri- methvlsilvl ether). (3j) Heathcock. C. H.; Davidsen, S. K.; Hug, K. T.; Flippin, L. A. J. Org. Cfiem. 1986, 51, 3027. Solvent, Counterion, and Secondary Deuterium Kinetic Isotope Effects in the Anionic Oxy-Cope Rearrangement Joseph J. Gajewski* and Kyle R. Geet Contribution from the Department of Chemistry, Indiana University, Bloomington, Indiana 47405. Received July zyxwvut 16, zyxwv I990 Abstract: The potassium and sodium alkoxides of 3-methyl-I ,5-hexadien-3-01 follow first-order kinetics in the process of undergoing the anionic oxy-Cope rearrangement in tetrahydrofuran (THF) and dimethyl sulfoxide (DMSO). The first-order rate constant for the rearrangement of the potassium alkoxide in DMSO is ca. 1000 times faster than that in THF, as is the first-order rate constant in T H F in the presence of 1 equiv or excess 18-crown-6. The rate constants in THF are independent of initial alkoxide concentration; in contrast, the first-order rate constants in DMSO are inversely proportional to the initial alkoxide concentration, and addition of potassium salts to the DMSO solution results in a retardation of rearrangement rate. Addition of and equiv of 18-crown-6 in THF gave first-order behavior only over the first 25% of reaction with an initial rate constant linearly related to that with 1 equiv of crown ether. Secondary deuterium kinetic isotope effects have been determined at the bond-breaking and bond-making sites in the Cope rearrangement of the potassium alkoxide in THF, in T H F in the presence of 18-crown-6, and in DMSO. The isotope effects indicate a highly dissociative transition state with substantial bond breaking of the C3-C4 bond and little bond making between the allylic termini (C1 and C6). The effects of aggregation and ionic dissociation are discussed in the context of mechanistic pathways proposed for the rearrangement in T H F and in DMSO. In 1975, Evans and Golob reported the remarkable rate ac- celeration for the Cope rearrangement of the potassium alkoxide of alcohol 1 of 1010-1017 relative to that for the alcohol itself.' 'Taken from the Ph.D. Thesis of K.R.G., Indiana University, May 1990. In general, hydroxy and alkoxy substitution at C3 of 1,Sdienes affect the rate of thermal sigmatropic 3,3 rearrangements only while a simple change of the hydroxyl substituent (I) Evans, D. A.; Golob, A. M. J. Am. Gem. Soc. 1975, 97,4765. 0002-7863/91/1513-967$02.50/0 0 1991 American Chemical Society