Evaluation of Chelation Effects Operative during Diastereoselective Addition of the Allylindium Reagent to 2- And 3-Hydroxycyclohexanones in Aqueous, Organic, and Mixed Solvent Systems

Article abstract of DOI:10.1021/jo980974y

The unprotected 2- and 3-hydroxycyclohexanones 1-8 were prepared by methods that skirted as much as possible their proclivity for α-ketol rearrangement (where the possibility for such isomerization exists). The diastereofacial selectivity of their reactio

Full text of DOI:10.1021/jo980974y

5604
J . Org. Chem. 1998, 63, 5604-5616
Eva lu a tion of Ch ela tion Effects Op er a tive d u r in g
Dia ster eoselective Ad d ition of th e Allylin d iu m Rea gen t to 2- a n d
3-Hyd r oxycycloh exa n on es in Aqu eou s, Or ga n ic, a n d Mixed Solven t
System s
Leo A. Paquette* and Paul C. Lobben
Evans Chemical Laboratories, The Ohio State University, Columbus, Ohio 43210
Received May 21, 1998
The unprotected 2- and 3-hydroxycyclohexanones 1-8 were prepared by methods that skirted as
much as possible their proclivity for R-ketol rearrangement (where the possibility for such
isomerization exists). The diastereofacial selectivity of their reaction with the allylindium reagent
in water, 50% aqueous THF, and anhydrous THF is described. The neighboring R-hydroxyl
substituent is construed to be capable of engaging in chelation, thereby controlling the stereo-
chemical outcome of the coupling process. When the hydroxyl substituent is oriented in the
equatorial plane, kinetic acceleration accompanies exclusive entry of the allyl group from the
equatorial direction. Steric congestion in the vicinity of the binding hydroxyl and ketonic centers
is well tolerated. Alternative projection of the OH group into the more crowded axial region may
not curtail chelation. For coordination to occur, however, a twist-boat conformation must initially
be adopted. While the evidence suggests that this may indeed occur in water, the necessity of
crossing the added energy barrier precludes the attainment of rates that are competitive with those
exhibited by the equatorial epimers (competition experiments). Placement of the hydroxyl group
at C-3 provides no evident opportunity for chelation control. However, excellent stereoselectivity
is seen upon axial orientation of the 3-OH group. This phenomenon is attributed to steric and/or
electronic effects alone or in combination.
The causative factors underlying the diastereofacial
substitution is involved (e.g., OCH3, OMOM,8a,b,9 or
spirotetrahydrofuranyl9,10). More sterically bulky polar
groups (e.g., OBn and OTBS) direct C-C bond formation
via alternative Felkin-Ahn transition states.8,11 Dimeth-
ylamino12 and carboxyl substituents13 that neighbor the
carboxaldehyde functionality likewise exhibit a useful
capacity for precoordination to In(III) under aqueous
conditions.
selectivity of 1,2-additions to carbonyl compounds con-
tinue to be actively probed because of the central position
this process holds in asymmetric organic synthesis.1 This
is particularly true for allyl organometallics of different
type, which are now recognized to enter into nucleophilic
attack via one or another structurally well-defined cyclic
transition state.2,3 In recent years, allylindium reagents
have been developed for use in aqueous media,4 alongside
corresponding investigations aimed at elucidating the
extent to which π-facial stereocontrol might be achieved
in water.5 Where R- and â-hydroxy aldehydes are
concerned, the coupling reactions take place rapidly,
chelated intermediates are involved, and significantly
heightened levels of syn-1,2-diols and anti-1,3-diols are
produced.6-8 The allyl group continues to be introduced
anti to an adjacent ether oxygen, provided that minimal
Intramolecular chelation can also materialize within
an allylindium reagent when a second polar functional
group such as hydroxyl and carbomethoxy is present.14
Although this class of reagents has shown considerable
promise for accomplishing effective 1,4-asymmetric
induction,15-17 questions persist concerning the precise
manner in which unprotected polar groups resident
within the organometallic species dictate the observed
π-facial discrimination.
(1) Masamune, S.; Choy, W.; Petersen, J . S.; Sita, L. R. Angew
Chem., Int. Ed. Engl. 1985, 24, 1.
(2) Roush, W. R. In Comprehensive Organic Synthesis; Heathcock,
C., Ed.; Pergamon Press: Oxford, 1991; Vol. 2, p 1.
(3) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207.
(4) (a) Li, C. J . Chem. Rev. 1993, 93, 2023. (b) Li, C. J . Tetrahedron
1996, 52, 5643. (b) Li, C.-J .; Chan, T.-H. Organic Reactions in Aqueous
Media; J ohn Wiley and Sons., Inc.: New York, 1997.
(5) Paquette, L. A. In Green Chemistry: Frontiers in Benign Chemical
Synthesis and Processing; Anastas, P., Williamson, T., Eds.; Oxford
University Press: Oxford, in press.
(9) Paquette, L. A.; Lobben, P. C. J . Am. Chem. Soc. 1996, 118, 1917.
(10) Paquette, L. A.; Stepanian, M.; Mallavadhani, U. V.; Cutarelli,
T. D.; Lowinger, T. B.; Klemeyer, H. J . J . Org. Chem. 1996, 61, 7492.
(11) Isaac, M. B.; Paquette, L. A. J . Org. Chem. 1997, 62, 5333.
(12) Paquette, L. A.; Mitzel, T. M.; Isaac, M. B.; Crasto, C. F.;
Schomer, W. W. J . Org. Chem. 1997, 62, 4293.
(13) Bernardelli, P.; Paquette, L. A. J . Org. Chem. 1997, 62, 8284.
(14) (a) Paquette, L. A.; Isaac, M. B. Heterocycles 1998, 47, 107. (b)
Paquette, L. A.; Rothhaar, R. R.; Isaac, M. B.; Rogers, L. M.; Rogers,
R. D. J . Org. Chem. 1998, 63, in press.
(6) Li, C. J .; Chan, T. H. Tetrahedron Lett. 1991, 32, 7017.
(7) (a) Kim, E.; Gordon, D. M.; Schmid, W.; Whitesides, G. M. J .
Org. Chem. 1993, 58, 5500. (b) Gao, J .; Ha¨rter, R.; Gordon, D. M.;
Whitesides, G. M. J . Org. Chem. 1994, 59, 3714.
(8) (a) Paquette, L. A. Mitzel, T. M. Tetrahedron Lett. 1995, 36, 6863.
(b) Paquette, L. A.; Mitzel, T. M. J . Am. Chem. Soc. 1996, 118, 1931.
(c) Paquette, L. A.; Mitzel, T. M. J . Org. Chem. 1996, 61, 8799.
(15) (a) Maguire, R. J .; Mulzer, J .; Bats, J . W. Tetrahedron Lett.
1996, 37, 5487. (b) Maguire, R. J .; Mulzer, J .; Bats, J . W. J . Org. Chem.
1996, 61, 6936.
(16) Paquette, L. A.; Bennett, G. D.; Chhatriwalla, A.; Isaac, M. B.
J . Org. Chem. 1997, 62, 3370.
(17) Paquette, L. A.; Bennett, G. D.; Isaac, M. B.; Chhatriwalla, A.
J . Org. Chem., 1998, 63, 1836.
S0022-3263(98)00974-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 07/21/1998

Diastereoselective Addition of the Allylindium Reagent
J . Org. Chem., Vol. 63, No. 16, 1998 5605
Sch em e 1
Sch em e 2
The insensitivity of allylindium reagents to moisture
obviously dismisses the need for protecting groups for
either reactant. For this reason, it is of special impor-
tance that the mechanistic role of the “naked” polar
functionality be clearly appreciated. Such studies have
not heretofore been possible because of the preexisting
requirement that the great majority of organometallic
reactions be performed on protected substrates in anhy-
drous solvents. The systems selected for the present
allylindation study include 2-hydroxycyclohexanone (1),
the four conformationally restricted 4-tert-butyl homo-
logues 2-5, the transposed ketone 6, and the stereoiso-
meric 3-hydroxy derivatives 7 and 8. This group of
butylcyclohexanone (Scheme 1). When the hydroxy
ketone mixture generated in this manner was found to
decompose/polymerize irreversibly on standing for several
hours at room temperature or on overnight storage at
-10 °C, recourse was made to direct silylation. This
protection maneuver led to the discovery that the spectral
properties of 9 and 10 were found not to be suitably
matched. Consequently, their individual structures were
unambiguously determined by 1H/1H COSY 90 NMR
experiments. Defined by these studies was the existence
of a single methylene spacer between the OTBS and t-Bu
substituents in 9, but two intervening CH2 groups in 10.
The evidence was unmistakable that the initially gener-
ated 2-hydroxy ketones 2 and 3 had undergone equilibra-
tion either during their formation or their silylation. This
observation conforms with earlier claims that “3 could
not be obtained free of 2” 19 and recent studies by Parker
et al. involving 2,4-dihydroxycyclohexanone systems.21
Attention is called to the equatorial preference of this
equilibration process (Scheme 2), which contrasts with
the axial preference exhibited by 2-halocyclohexanones.22
The deprotection of 9 and 10 with fluoride ion was
expectedly met with modest levels of R-ketol rearrange-
ment. The extent to which this occurred (∼10%) pre-
cluded the acquisition of useful amounts of 6 by this
route. Therefore, this isomer was independently pre-
pared from the known acetal 119 by hydrolysis and
stereoselective (axial attack)23 1,2-reduction to give 12
(Scheme 3). Following conversion to 13, the benzylidene
double bond was cleaved ozonolytically to deliver iso-
merically pure 10 in 76% yield. As with 9 and the other
protected hydroxy ketones described here, 10 was stable
to long-term storage at or below room temperature.
Access to the rather elusive trans-4-tert-butyl-2-hy-
droxycyclohexanone (3) proved to be more circuitous. The
observations summarized in Scheme 4 revealed to us that
a nonhydrolytic deprotection strategy had to be imple-
mented. Reduction of keto acetal 149 with Dibal-H in
hexanes at -78 °C proceeded very efficiently to give a
substrates fulfills certain relevant requirements: (a)
positioning the potential anchoring hydroxyl both R and
â to the carbonyl, (b) orienting the OH substituent both
axially and equatorially in fixed chair conformations, and
(c) modulating the level of steric crowding in the immedi-
ate vicinity of the hydroxyl (and carbonyl) as in 4 and 5
and its placement relative to the tert-butyl group (as in
6). Consequently, 1-8 expand upon the acyclic systems
and nicely complement the R-alkoxycyclohexanones probed
so successfully earlier.8,9
(21) (a) Parker, K. A.; Dermatakis, A. J . Org. Chem. 1997, 62, 6692.
(b) Parker, K. A.; Kim, H.-J . J . Org. Chem. 1992, 57, 752.
(22) (a) Lambert, J . B. In The Conformation Analysis of Cyclo-
hexenes, Cyclohexadienes, and Related Hydroaromatic Compounds;
Rabideau, P., Ed.; VCH: Weinheim, 1989; Chapter 2. (b) Denmark, S.
E.; Dappen, M. S.; Sear, N. L.; J acobs, R. T. J . Am. Chem. Soc. 1990,
112, 3466.
(23) The stereochemical course of this reduction was established on
the strength of NOE experiments performed on 13. The key interac-
tions are illustrated here:
Resu lts
P r epar ation of th e Hydr oxycycloh exan on es. Com-
mercially available 1 was obtained as a dimer which
equilibrates with the monomer when dissolved in water
under ambient conditions. Despite existing reports
detailing the synthesis of 2 and 3,18,19 difficulties were
encountered on attempted Rubottom oxidation20 of 4-tert-
(18) Vedejs, E. Dent, W. H., III J . Am. Chem. Soc. 1989, 111, 6861.
(19) Cain, C. M.; Cousins, R. P. C.; Combarides, G.; Simpkins, N.
S. Tetrahedron 1990, 46, 523.
(20) Rubottom, G. M.; Gruber, J . M. J . Org. Chem. 1978, 43, 1599.

5606 J . Org. Chem., Vol. 63, No. 16, 1998
Lobben
Sch em e 3
Sch em e 5
Sch em e 4
chairlike character develop in the transition state. Se-
quential ozonolysis and hydrogenolysis of 19 afforded 3.
Our normal practice was to transform 20 into 3 on an
as-needed basis.
2-Hydroxycyclohexanones 4 and 5 were prepared by a
modification of the procedure originally devised by
Vedejs,31a and by Aube´ and Wang.31b Monomethylation
of the anion resulting from deprotonation of the dimeth-
ylhydrazone of 4-tert-butylcyclohexanone, with subse-
quent periodate cleavage,32 provided the pair of diaster-
eomeric ketones 21 (Scheme 6). Epoxidation of the
thermodynamic silyl enol ether, generated by the Krafft-
Holton protocol,33 and ensuing desilylation afforded 4 and
5 in a 2.8:1 ratio. Pure samples were readily acquired
through combined application of chromatographic and
crystallization methods.
favorable 1.7:1 ratio of 15 and 16, which epimers were
readily amenable to chromatographic separation. How-
ever, all attempts to remove the acetal group in 15
without being disruptive of stereochemistry (including
conditions specifically recommended for this purpose24-26
)
returned only 2 and 6 (in varying proportions) without
any evidence for 3.27
These complications were bypassed by Wittig olefina-
tion28 of 4-tert-butylcyclohexanone and submission of 17
to allylic oxidation with tert-butyl hydroperoxide in the
presence of catalytic levels of selenium dioxide29 (Scheme
5). The high diastereoselectivity accompanying introduc-
tion of the hydroxyl group in 1829 (>97:3 by 1H NMR
integration) is attributed to the strong kinetic preference
of ene product A for [2,3]-sigmatropic rearrangement30
via B rather than C. Only in the first instance does
Last, the 3-hydroxycyclohexanones 7 and 8 were
synthesized according to the directives given by Evans,
Chapman, and Carreira,34,35 except that the final ketal
(31) (a) Vedejs, E.; Dent, W. H., III.; Kendall, J . T.; Oliver, P. A. J .
Am. Chem. Soc. 1996, 118, 3556. (b) Wang, Y. Ph.D. Thesis, University
of Kansas, 1993. We thank Professor J . Aube´ for making part of this
thesis available to us.
(24) Carren˜o, M. C.; Ruano, J . L. G.; Garrido, M.; Ruiz, M. P.;
Solladie´, G. Tetrahedron Lett. 1990, 31, 6653.
(25) Lipshutz, B. H.; Pollart, D.; Monforte, J .; Kotsuki, H. Tetrahe-
dron Lett. 1985, 26, 705.
(32) Corey, E. J .; Enders, D. Tetrahedron Lett. 1976, 3.
(26) (a) Still, I. W.; Shi, Y. Tetrahedron Lett. 1987, 28, 2489. (b) Kim,
K. S.; Song, Y. H.; Lee, B. H.; Hahn, C. S. J . Org. Chem. 1986, 51,
404.
(27) The readiness with which axial R-alkyl groups experience
epimerization has been previously documented: Cross, B.; Whitham,
G. H. J . Chem. Soc. 1961, 1650.
(28) (a) Wittig, G.; Schoellkopf, U. Organic Syntheses; Wiley: New
York, 1973; Collect. Vol. V, p 751. (b) Greenwald, G.; Chaykovsky, M.;
Corey, E. J . J . Org. Chem. 1963, 28, 1128.
(29) The axial disposition of the hydroxyl was clearly evident on the
basis of the appearance of the H-2 methine proton as a doublet of
doublets (J ) 2.6, 2.6 Hz) at δ 3.99. The two small coupling constants
arise from neighboring equatorial-equatorial and equatorial-axial
spin interactions.
(33) Krafft, M. E.; Holton, R. A. Tetrahedron Lett. 1983, 24, 1345.
(34) Evans, D. A.; Chapman, K. T.; Carreira, E. M. J . Am. Chem.
Soc. 1988, 110, 3560.
(35) In view of certain discrepancies noted between our 1H and 13C
NMR data for 7 and 8 and those reported in ref 34 (note the identity
of the carbon shifts); our data for these two compounds are given here
in full. For 7: mp 118-119 °C.; 1H NMR (300 MHz, CDCl3) δ 4.61 (br
s, 1 H), 2.55-2.38 (m, 4 H), 2.37-2.24 (m, 1 H), 2.07 (dddd, J ) 12.8,
12.8, 12.8, 4.6 Hz, 1 H), 1.97-1.87 (m, 1 H), 1.91 (br s, 1 H), 1.53 (ddd,
J ) 12.5, 3.4, 1.5 Hz, 1 H), 1.01 (s, 9 H); 13C NMR (75 MHz, CDCl3)
ppm 211.8, 70.2, 50.9, 50.0, 41.5, 32.6, 28.8, 21.4. For 8: mp 67.5-
68.5 °C.; 1H NMR (300 MHz, CDCl3) δ 4.15 (dd, J ) 8.8, 3.9 Hz, 1 H),
2.54 (dd, J ) 15.4, 4.0 Hz, 1 H), 2.44 (dd, J ) 15.5, 5.0 Hz, 1 H), 2.38-
2.25 (m, 2 H), 1.55-1.40 (m, 2 H), 0.98 (s, 9 H); 13C NMR (75 MHz,
CDCl3) ppm 211.9, 68.6, 52.3, 47.8, 38.5, 32.8, 27.7, 21.6.
(30) Sharpless, K. B.; Lauer, R. F. J . Am. Chem. Soc. 1972, 94, 7154.

Diastereoselective Addition of the Allylindium Reagent
J . Org. Chem., Vol. 63, No. 16, 1998 5607
Sch em e 6
Sch em e 8
Sch em e 7
manner,38 studies involving this activated form of indium
were not pursued further at this time.
Comparable coupling of 2-hydroxycyclohexanone (1) to
allyl bromide in the presence of indium afforded exclu-
sively diastereomer 24 irrespective of the reaction condi-
tions (Scheme 8, entries 4-6). The diastereoselectivity
was determined by 1H NMR integration of unpurified
reaction mixtures.39 The considerably more extended
reaction time required for the allylation performed in
THF constitutes a trend that was observed throughout
this study. The high diastereoselectivity exhibited by 1
warrants direct comparison with 2-methoxycyclohex-
anone for which a maximum 14:1 kinetic preference for
attack anti to the alkoxy substituent has been reported.9
These findings conform to a mechanistic model in which
the indium atom coordinates simultaneously to the
ketonic carbonyl and the adjacent oxygenated center as
depicted in D (R ) R′ ) H). Chelation in this fashion
simultaneously activates the ketone toward nucleophilic
attack, locks the cyclohexanone into a specific chair
conformation (if this feature is not already incorporated
into the substrate), and sets the stage for equatorial
delivery of the allyl unit. The improved chelating ability
of OH relative to OCH3 can be traced to improved steric
accessibility by the metal atom.8
hydrolysis was accomplished with pyridinium p-toluene-
sulfonate in refluxing aqueous acetone.
Com p a r a tive Beh a vior of 4-ter t-Bu tylcycloh ex-
a n on e a n d 1-3. To provide useful calibration, 4-tert-
butylcyclohexanone was initially treated with allyl bro-
mide and powdered indium metal in three different
media (Scheme 7). The strong equatorial bias provided
by this tertiary alkyl substituent has the utilitarian role
of strongly biasing conformational population while serv-
ing as a stereochemical marker. Irrespective of whether
the reaction medium was pure water, dry THF, or a 1:1
mixture thereof, alcohol 22 was found to predominate.
The preferred equatorial approach of the allylindium
reagent conforms to the stereoselectivity previously noted
for other allylmetal reagents.3,36 The preparation of 22
was greatest in THF, a finding suggestive of the pos-
sibility that solvation in water has the effect of facilitat-
ing axial attack to some degree. Another notable feature
is the reaction time in THF. When commercial indium
powder was utilized in entry 3 as it was in the aqueous
experiments, consumption of the ketone was complete
only after 48 h. However, when the indium was produced
by the electrolysis of InCl3 with platinum electrodes,37
the requisite reaction time was dramatically reduced to
4 h. Since comparable rate enhancements have previ-
ously been reported for zinc produced in a similar
Full stereochemical assignment to 24 was made pos-
sible following conversion to acetonide 25. The added
(37) (a) Tokuda, M.; Satoh, S.; Katoh, Y.; Suginome, H. In Electro-
organic Synthesis; Little, R. D., Weinberg, N. L., Eds.; Dekker: New
York, 1991; p 83. (b) Tanaka, H.; Yamashita, S.; Hamatani, T.;
Nakahara, T.; Torii, S. In Recent Advances in Electroorganic Synthesis;
Torii, S., Ed.; Elsevier: New York, 1987; p 307. (c) Habeeb, J . J .; Tuck,
D. G. J . Organomet. Chem. 1977, 134, 363. (d) Habeeb, J . J .; Said, F.
F.; Tuck, D. G. J . Organomet. Chem. 1980, 190, 325.
(38) Tokuda, M.; Kurono, N.; Mimura, N. Chem. Lett. 1996, 1091.
(39) A diastereoselectivity of >97:3 represents the perceived detec-
tion limit of the instrument.
(36) (a) Aluminum, magnesium, lithium, potassium, and sodium:
Gaudemar, M. Tetrahedron, 1976, 32, 1689. (b) Chromium: Hiyama,
T.; Okude, Y.; Kimura, K.; Nozaki, H. Bull. Chem Soc. J pn. 1982, 55,
561. (c) Tin: Naruta, Y.; Ushida, S.; Maruyama, K. Chem. Lett. 1979,
919. (d) Indium: Reetz, M. T.; Haning, H. J . Organomet. Chem. 1997,
541, 117. (e) Indium: Araki, S.; Ito, H.; Butsugan, Y. J . Org. Chem.
1988, 53, 1831. (f) Zinc: Wilson, S. R.; Guazzaroni, M. E. J . Org. Chem.
1989, 54, 3087.

5608 J . Org. Chem., Vol. 63, No. 16, 1998
Lobben
Sch em e 9
Sch em e 10
Eva lu a tion of th e Ster eoisom er ic Ter tia r y r-Ke-
tols 4 a n d 5. An earlier investigation has demonstrated
that incorporation of an ether oxygen into a neighboring
spirotetrahydrofuran ring can be met with a significant
dropoff in chelation to an allylindium reagent.9 This
phenomenon has been attributed to an enhancement in
local steric shielding, a structural feature toward which
this organometallic reagent is quite responsive. As a
logical extension of possible steric variants, we became
particuarly interested in the rather different functional
group assembly resident in 4 and 5. Both cyclohexanones
carry an unprotected R-alcohol, although the tertiary
status of these OH groups is guaranteed to reduce their
accessibility to some degree. Our concern was whether
complete chelation control would continue to be opera-
tional, particularly in water. Clearly, 4 and 5 are not
subject to potential complications stemming from tauto-
meric equilibration.
The experiments undertaken with 4 uniformly gave
rise to 30 as the only detectable product (Scheme 11). In
contrast to the earlier examples, no reaction was observed
in THF when the reaction components were admixed in
the usual way. However, preformation of the allylindium
reagent did lead to rapid and efficient coupling to deliver
30. The dramatic reduction in reaction time seen in entry
16 is noteworthy as it likely merits consideration when
preparative-scale experiments are contemplated.
The equatorial orientation of the allyl group in 30 was
established by way of NOE difference experiments per-
formed on both the free diol and its acetonide 31 (see
Experimental Section). In the latter instance, the en-
hanced rigidification provided by the fused dioxolane ring
was met with considerable enhancement of the NOE
response in every instance.
structural rigidification accompanying this transforma-
tion allowed for the successful implementation of NOE
difference experiments. Particularly revealing in this
instance was the enhancement of the signals due to allyl
protons H-7, H-7′, and H-8 upon irradiation of axial
proton H-2 (see Experimental Section).
The fact that the 2-hydroxy substituent in 2 is equa-
torially disposed suggested that transition state D (R )
t-Bu, R′ ) H) should again be operational. Indeed, the
allylindation of 2 delivered only 26 in each instance
(entries 7-9, Scheme 9). Comparable efficiency was
realized throughout, although the elapsed time necessary
to attain this level of conversion was again much slower
in THF. The identification of 26 was established by
chemical means. Its selective O-methylation with sodium
hydride and methyl iodide at 0 °C gave rise to the known
ether alcohol 27.9
With attention now focused on 3, its remarkable
propensity for facile tautomeric equilibration was mani-
fested again. During the course of reaction with the
allylindium reagent, 23% of the product mixture gener-
ated in water arose from attack on 2 and 6. In 50%
aqueous THF, the extent of these side reactions dropped
only modestly to the 18% level. Contrastingly, allylations
performed in pure THF were free of this complication.
The yields provided in Scheme 10 have been adjusted to
reflect the efficiency with which 3 itself is engaged in
allylindation. The stereochemical course of this process
is such that only 28 is produced (entries 10-12). The
trans disposition of the hydroxyl group in 28 was
confirmed by conversion into the known 29.9 The forma-
tion of a lone diastereomeric product further attests to
the powerful chelating ability of hydroxy substituents in
aqueous media. In the present circumstances, complex-
ation of the indium reagent to 3 is thought to foster
adoption of a twist-boat conformation as in E (R ) H) in
order to maximize the level of the two O- - -In interac-
tions. This structural arrangement allows for ready
delivery of the allyl unit from the pseudoequatorial
direction. For comparison, the methyl ether of 3 exhibits
a chelate/nonchelate ratio of 9.6:1.9
A reversal in the relative configuration of the carbinol
carbon as in 5 provided complementary results demon-
strating the superiority of the aqueous indium protocol
for controlling the stereoselectivity of C-allylation (Scheme
12). In H2O or aqueous THF, the reaction is complete in
2-2.5 h and only diol 32 is formed. The couplings in THF
afforded approximately 2:1 mixtures of 32 and 33, with
the product distribution proving to be relatively insensi-

Diastereoselective Addition of the Allylindium Reagent
J . Org. Chem., Vol. 63, No. 16, 1998 5609
Sch em e 11
Sch em e 13
of trans diol 32 displays significant enhancements to the
protons resident on the equatorial allylic substituent,
interactions that would not be observed if the allyl side
chain was axial as in 33. Conversely, as a consequence
of the 1,3-diaxial relationships involving the allyl group,
H-3a, and H-5a in cis diol 33, enhancements at the 3-5%
level were seen upon appropriate double irradiation of
the associated signals. The relevant data are compiled
in the Experimental Section.
The DEPT technique takes advantage of heteronuclear
three-bond coupling, with the degree of signal enhance-
ment achieved by pulsing a selected hydrogen frequency
and observing a 13C nucleus being maximized when the
dihedral angle relationships are at 0° or 180°. As for 33,
models show the dihedral angle between H-6a and C-8
to be approximately 180° (see F ) such that significant
1,3J interaction should be seen for this isomer but not for
32, which lacks this trans-diaxial relationship. Such was
indeed the case. In addition, acetonide 34 generated
conventionally from 3341 gave similar evidence for strong
heteronuclear coupling between H-6a and C-8, indicating
its ground-state conformation to be that given by G.
Sch em e 12
The excellent control of diastereofacial selectivity
reflected in entries 17 and 18 together with the axial
nature of the tertiary hydroxyl in 32 conforms to the
ready adoption by R-ketol 5 of twist-boat conformation
E (R ) CH3) when water is present. This chelation-
controlled pathway clearly does not enjoy a comparable
overwhelming kinetic advantage when the reaction me-
dium is anhydrous. The data show that the allylindium
reagent exhibits a bonding affinity for tertiary hydroxyl
groups in water that is strong and measurably in excess
over that exhibited toward hindered ethereal oxygen
centers.9
tive to the mode of generation of the allylindium reagent.
Evidently, the anhydrous conditions and not the reaction
time are controlling of competing axial and equatorial
attack.
The stereodisposition of the allyl groups in 32 and 33
was defined by NOE difference and long-range semi-
selective DEPT experiments.40 The axially oriented H-6a
Ster eoch em ica l Cou r se of th e Allyla tion of 6. The
results obtained for the allylation of 6 are summarized
in Scheme 13. To our delight, this transposed system
also underwent nucleophilic attack cleanly and exclu-
(40) (a) Bax, A. J . Magn. Reson. 1983, 52, 330; 1983, 53, 517. (b)
Rutar, V. J . Am. Chem. Soc. 1983, 105, 4095. (c) Rutar, V. J . Magn.
Reson. 1984, 56, 87. (d) Wimperis, S.; Freeman, R. J . Magn. Reson.
1984, 58, 348. (e) Lin, L.-J .; Cordell, G. A. J . Chem. Soc., Chem.
Commun. 1986, 377.
(41) Diol 32 was completely unreactive to these conditions.

5610 J . Org. Chem., Vol. 63, No. 16, 1998
Lobben
Sch em e 14
Sch em e 15
(Scheme 7). The axial orientation of the allyl unit in 40
was clearly apparent from NOE difference data, particu-
larly when direct comparison was made with 39. In the
latter instance, the effect of irradiating H-3 on the
intensity of the H-7,H-7′ signal is particularly diagnostic
of a 1,3-diaxial interaction. Long-range semiselective
DEPT studies also conclusively established the axial
nature of the allyl group in 40. Thus, independent
irradiation of H-2a and H-6a resulted in pronounced
transfer to C-7 (see H).
Com p etition Exp er im en ts. In the absence of kinetic
studies, it is not possible to distinguish between chelation
as a product-determining event or an unproductive
reversible process.42,43 However, if chelated transition
states are responsible for providing a lower energy
pathway to products, it must also be true that this
process must take place more rapidly than others not
involving chelation. Presently, this important parameter
sively from the equatorial direction to give 35. Entries
21-23 reveal that the equatorial status of the differently
positioned R-hydroxyl substituent remains ideally situ-
ated for coupling via a chelation-controlled transition
state related to that described by D. As in preceding
examples, the stereochemistry of 35 was unambiguously
defined by NOE difference experiments.
Cou p lin gs In volvin g t h e 3-H yd r oxycycloh ex-
a n on es 7 a n d 8. The conventional allylindation of 7 led
uniquely to production of diastereomer 36 irrespective
of the reaction medium (entries 24-26, Scheme 14). In
this instance, the experiment performed in THF was
allowed to proceed for 3 days and resulted in only 39%
conversion to 36. The structural definition of 36 was
initially made by means paralleling those utilized earlier.
Thus, when NOE difference experiments were performed
(see Experimental Section), the independent irradiation
of allylic protons H-7 and H-7′ as well as vinylic proton
H-8 was found to induce no transfer to H-5a as would be
expected of a diaxial relationship. On the other hand,
H-2e, H-2a, H-6e, and H-6a proved responsive, indicating
that the allyl group was equatorially disposed. This
conclusion was ultimately confirmed by the conversion
of 36 into 37 and 38 under conditions sufficiently mild
to preclude the possibility that the trans diol would react
with equal efficiency, if at all. The atom connectivity in
38, established by 1H,1H COSY NMR analysis, eliminated
the possibility that a rearrangement may have material-
ized during its formation. Particularly diagnostic were
the W-plan couplings observed between H-2e/H-6e and
H-3e/H-5e.
has been probed by means of a series of competition
experiments. At issue is whether those processes that
exhibit highly controlled diastereoselectivity toward the
allylindium reagent are indeed mediated by more reac-
tive indium chelates. To simplify matters, we have
assumed that chelation/nonchelation phenomena and
conformational factors are the only relevant kinetic
considerations. Although strictly speaking this need not
be so,44 it can be argued that the correlation is reasonable
because of the close structural similarity of the ketones.
Several important facts can be culled from the data
compiled in Table 1. First, an equatorial R-hydroxy
group exerts a significant accelerating effect not available
to its axial counterpart. Inspection of entries 30-32,
particularly the 8.5-fold greater reactivity of 4 relative
to 4-tert-butylcyclohexanone, is most informative. Since
(42) (a) Chen, X.; Hortelano, E. R.; Eliel, E. L.; Frye, S. V. J . Am.
Chem. Soc. 1990, 112, 6130. (b) Frye, S. V.; Eliel, E. L.; Cloux, R. J .
Am. Chem. Soc. 1987, 109, 1862.
(43) Laemmle, J .; Ashby, E. C.; Neumann, H. M. J . Am. Chem. Soc.
1971, 93, 5120.
The allylindation of 8 proceeded smoothly with forma-
tion of two homoallylic alcohols (Scheme 15). The ratios
of 39 to 40 reported in entries 27-29 parallel very closely
those obtained earlier with 4-tert-butylcyclohexanone
(44) Das, G.; Thornton, E. R. J . Am. Chem. Soc. 1993, 115, 1302.

Diastereoselective Addition of the Allylindium Reagent
J . Org. Chem., Vol. 63, No. 16, 1998 5611
Ta ble 1. Com p etitive In d iu m -P r om oted Allyla tion s in
for the axial alcohols involves the crossing of an energy
barrier45 having no parallel in the equatorial diastereo-
mers.
Wa ter a t 25 °C
Caution must be exercised in evaluating the data for
entry 34 because of the requirement for dimer to mono-
mer conversion in 1 prior to its availability for reaction.
Notwithstanding, the results illustrate the greater reac-
tivity of 1.
At the outset of this study, a modest amount of
information was available concerning the ability of OH
groups in â-hydroxy aldehydes to effectively chelate the
allylindium reagent under aqueous conditions.8a,b In fact,
these observations prompted our examination of 7 and 8
in the present context. While it is obvious that the
equatorial orientation of the hydroxyl in 8 precludes the
possible operation of remote chelation of any type, isomer
7 has the option of undergoing allylation via the com-
plexed transition state I. As pointed out previously, the
diastereoselectivity exhibited by 8 expectedly conforms
closely to that shown by the parent 4-tert-butylcyclohex-
anone. In addition, it is marginally less reactive than 7
(entry 36) which, however, gives rise exclusively to 36
(Scheme 14). The equatorial disposition of the allyl group
in 36 unequivocally excludes the involvement of transi-
tion state I, which must culminate in axial allylation. The
data are totally inconsistent with the notion that the
indium reagent experiences dual coordination to both
oxygen atoms in 7. The substantial steric congestion at
play in I may be contributory to its nonoperability. The
origin of the strong preference for equatorial attack
exhibited by 7 is less clear. The possibility exists that
the size and electronegativity of the axial OH substituent
serve singly46 or in combination to disfavor axial entry
of the allylindium reagent.
Con clu sion s
Our investigation into the facial selectivity of the
indium-promoted allylation of 2-hydroxycyclohexanones
provides experimental evidence for possible chelation
control in these carbonyl addition reactions, particularly
in water. Coordination of the unprotected OH to the
nucleophilic allylindium reagent is notably effective when
it is projected into equatorial space. As demonstrated
by competition experiments, rate accelerations are mani-
fested irrespective of the secondary or tertiary nature of
the neighboring hydroxyl. The stereoselectivity demon-
strated by the axial R-ketols remains exceptionally high.
In these examples, kinetic effects are not as dramatic.
This phenomenon is attributed to the possible involve-
ment of twist-boat conformations for realizing effective
chelation by In(III) to both oxygen centers. The need to
cross this added energy barrier does not appear to be
inhibitory of stereocontrol, which is excellent in water.
a
All experiments were conducted minimally in duplicate, and
the reported data represent the average of these experiments.b The
relative rate ratio was ascertained by quantitative analysis of the
unreacted ketones remaining. c The relative rate ratio was ascer-
tained by quantitative analysis of the alcohol products formed.
d
The reaction was performed at an initial pH of 7.0 as in all of
the other cases. e The reaction was performed in water at an initial
pH of 4.0.
2 and 4 exhibit comparable kinetic advantages over 41
(entries 31 and 33), support is gained for the contention
that the effects of steric screening in the vicinity of
equatorial OH groups on the reactivity of the allylindium
reagent in water is virtually inconsequential. The ap-
preciably diminished reactivity of 5 toward 4 (entry 30)
and 42 (entry 35) makes evident in direct competition
the fact that the axial R-hydroxyls are not similarly
advantaged. In this sterically more demanding environ-
ment, the distinction between free OH groups and
ethereal oxygen centers is seen to be leveled (entry 40).
Since excellent allylation diastereoselectivity is exhibited
by 3 and 5 in water, the implication is that complexation
likely operates as well in these circumstances. The
absence of significant kinetic acceleration may reside in
the need to involve a twist-boat conformation of the
reactant as in E. Consequently, the pathway to product
The present work establishes further that 3-hydroxy-
cyclohexanones are not subject to chelation-controlled
diastereoselection. Although they are not activated
toward allylation by precoordination, stereocontrol is not
(45) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of
Carbon Compounds; J ohn Wiley and Sons: New York, 1994; Chapter
11.
(46) (a) Cieplak, A. S. J . Am. Chem. Soc. 1981, 103, 4540. (b)
Cieplak, A. S.; Tait, B. D.; J ohnson, C. R. J . Am. Chem. Soc. 1989,
111, 8447.

5612 J . Org. Chem., Vol. 63, No. 16, 1998
Lobben
0.88 (s, 9 H), 0.13 (s, 3 H), 0.02 (s, 3 H); 13C NMR (75 MHz,
CDCl3) ppm 209.6, 77.0, 49.8, 42.2, 35.8, 32.6, 27.3, 25.8, 25.2,
18.5, -4.6, -5.4; HRMS (EI) m/z (M+) calcd 284.2172, obsd
284.2169.
necessarily precluded. The response of 7 is particularly
instructive in this respect.
It will be noted that equatorial attack operates regard-
less of the equatorial or axial orientation of the hydroxyl
group. This fact leaves open the possibility that stereo-
control in these cyclohexyl systems is dictated in part by
the rather large diameter of indium, such that axial
approach of the allylindium reagent is not sterically
tolerated. To the extent that this phenomenon operates,
the rate acceleration for equatorial hydroxy groups might
be attributable to other factors such as hydrogen bridging
to the neighboring carbonyl group. Notwithstanding the
overall balance of forces including chelation effects, their
combined contribution leads to high stereocontrol from
the equatorial side.
Anal. Calcd for C16H32O2Si: C, 67.55; H, 11.34. Found C,
67.67; H, 11.39.
Desilyla tion of 9 a n d 10. A solution of 9 (1.40 g, 4.95
mmol) in dry THF (5.5 mL) was treated with tetra-n-butyl-
ammonium fluoride (5.44 mL of 1.0 M in THF, 5.44 mmol),
stirred for 2 h at 20 °C, and diluted with brine. The separated
aqueous layer was extracted with ether, and the combined
organic phases were dried and evaporated. The residue was
quickly eluted through a column of silica gel with 10% ethyl
acetate in hexanes to furnish 400 mg (48%) of 2. The semisolid
was used immediately without further characterization. High-
field 1H NMR analysis showed that 11% of 2 had isomerized
to 6.
Comparable treatment of 10 (761 mg, 2.67 mmol) afforded
268 mg (59%) of 6 as a semisolid containing 9% of 2. This
material was utilized without delay.
Exp er im en ta l Section 47
cis-4-ter t-Bu tyl-2-(ter t-bu tyld im eth ylsiloxy)cycloh ex-
a n on e (9) a n d tr a n s-5-ter t-Bu tyl-2-(ter t-bu tyld im eth yl-
siloxy)cycloh exa n on e (10). A cold (-78 °C) solution of
diisopropylamine (3.64 mL, 26.0 mmol) in dry THF (100 mL)
was blanketed with N2, treated via syringe with n-butyllithium
(18.5 mL of 1.3 M in hexanes, 24.0 mmol), allowed to warm to
0 °C during 1 h, and recooled to -78 °C prior to the sequential
addition of 4-tert-butylcyclohexanone (3.08 g, 20.0 mmol)
dissolved in dry THF (100 mL) and of chlorotrimethylsilane
(3.81 mL, 30.0 mmol) via syringe. The reaction mixture was
slowly warmed to 0 °C over 4 h, diluted with triethylamine
(100 mL), and concentrated in vacuo. The residue was
dissolved in pentane, filtered through Celite (pentane wash),
and evaporated. Distillation gave the silyl enol ether (bp 55
°C/0.6 Torr) as a pale yellow viscous oil (3.85 g, 85%).
A 2.26 g (10.0 mmol) sample of the above product was
dissolved in CH2Cl2 (100 mL), treated portionwise with purified
m-chloroperbenzoic acid (2.07 g, 12.0 mmol), stirred for 8 h,
and diluted with saturated NaHCO3 solution. The separated
organic phase was dried, concentrated, dissolved in dry THF
(90 mL), and treated directly with tetra-n-butylammonium
chloride (10 mL of 1.0 M in THF, 10 mmol). The reaction
mixture was stirred for 8 h and partitioned between saturated
NaHCO3 solution and ether. The aqueous phase was extracted
with ether and the combined organic phases were dried and
carefully concentrated on a rotary evaporator.
tr a n s-2-[(E)-Ben zyliden e]-4-ter t-bu tylcycloh exan ol (12).
A solution of 11 (1.0 g, 3.5 mmol) in 75% aqueous acetone (20
mL) containing concentrated hydrochloric acid (0.10 mL) was
refluxed for 2 h, diluted with brine, and extracted with ether.
The combined organic layers were dried and concentrated to
provide the ketone, which was taken up in THF (9 mL) and
added to a cold (0 °C), freshly prepared solution of lithium
trimethoxyaluminum hydride [from 0.53 g (14 mmol) of lithium
aluminum hydride and 1.69 mL (42 mmol) of anhydrous
methanol in 11 mL of dry THF]. After 1 h at this temperature,
the reaction mixture was treated dropwise with H2O (0.53 mL),
15% sodium hydroxide solution (0.53 mL), and H2O (1.59 mL),
and extracted several times with ether. The combined ethereal
layers were washed with brine, dried, and concentrated.
Purification of the residue by chromatography on silica gel
(elution with 10% ethyl acetate in hexanes) afforded 0.54 g
(63% overall) of 12 as a white solid: mp 101-102 °C; IR (KBr,
cm-1) 3284; 1H NMR (300 MHz, CDCl3) δ 7.29-7.12 (m, 5 H),
6.50 (s, 1 H), 4.08-4.04 (m, 1 H), 2.98 (dt, J ) 13.0, 2.5 Hz, 1
H), 2.23-2.17 (m, 1 H), 1.87-1.82 (m, 1 H), 1.70 (br s, 1 H),
1.44-1.14 (series of m, 4 H), 0.82 (s, 9 H); 13C NMR (75 MHz,
CDCl3) ppm 144.9, 137.9, 128.8, 128.0, 126.0, 118.1, 73.3, 49.3,
37.6, 32.8, 29.5, 27.4, 25.9; HRMS (EI) m/z (M+) calcd 244.1827,
obsd 244.1830.
Anal. Calcd for C17H24O: C, 83.55; H, 9.90. Found C, 83.39;
H, 9.84.
To a 340 mg (2.0 mmol) sample of this material in dry DMF
(4 mL) was added imidazole (361 mg, 2.40 mmol). The reaction
mixture was stirred for 2 h and diluted with brine. After
repeated extraction of the aqueous phase with ether, the
combined organic layers were dried and concentrated. Chro-
matography of the residue on silica gel (elution with 6% ethyl
acetate in hexanes) afforded 80 mg (14%) of the less polar 10
and 111 mg (20%) of 9, both as colorless oils.
[[tr a n s-2-[(E)-Ben zyliden e]-4-ter t-bu tylcycloh exyl]oxy]-
ter t-bu tyld im eth ylsila n e (13). A solution of 12 (484 mg,
1.98 mmol), imidazole (230 mg, 3.38 mmol), and tert-butyldi-
methylsilyl chloride (358 mg, 2.37 mmol) in dry DMF (4 mL)
was stirred for 12 h, diluted with brine, and extracted with
ether. After the predescribed workup and chromatography on
silica gel (elution with 5% ethyl acetate in hexanes), there was
1
For 9: IR (neat, cm-1) 1732; H NMR (300 MHz, CDCl3) δ
obtained 506 mg (71%) of 13 as a colorless oil: IR (neat, cm-1
)
1130, 1108; 1H NMR (300 MHz, CDCl3) δ 7.34-7.17 (m, 5 H),
6.61 (s, 1 H), 4.08 (ddd, J ) 11.0, 4.9, 1.6 Hz, 1 H), 3.03 (ddd,
J ) 13.0, 2.5, 2.5 Hz, 1 H), 2.15-2.06 (m, 1 H), 1.90-1.86 (m,
1 H), 1.55-1.19 (series of m, 4 H), 0.98 (s, 9 H), 0.89 (s, 9 H),
0.16 (s, 3 H), 0.14 (s, 3 H); 13C NMR (75 MHz, CDCl3) ppm
144.4, 138.5, 128.8, 127.9, 125.8, 119.0, 74.2, 49.3, 37.9, 32.8,
29.6, 27.5, 26.0, 26.0, 18.5, -4.8, -4.8; HRMS (EI) m/z (M+)
calcd 358.2692, obsd 358.2707.
4.18 (ddd, J ) 11.6, 6.4, 1.0 Hz, 1 H), 2.40 (ddd, J ) 13.8, 4.3,
2.7 Hz, 1 H), 2.27 (ddd, J ) 13.8, 5.9, 1.1 Hz, 1 H), 2.25-2.19
(m, 1 H), 2.02 (dddd, J ) 15.7, 3.0, 3.0, 3.0 Hz, 1 H), 1.60 (dddd,
J ) 12.2, 12.2, 2.6, 2.6 Hz, 1 H), 1.47 (dd, J ) 12.0, 12.0 Hz,
1 H), 1.37 (dddd, J ) 13.4, 13.4, 13.4, 4.3 Hz, 1 H), 0.91 (s, 9
H), 0.89 (s, 9 H), 0.12 (s, 3 H), 0.01 (s, 3 H); 13C NMR (75 MHz,
CDCl3) ppm 209.4, 76.7, 46.0, 39.4, 38.5, 32.3, 28.0, 27.6, 25.8,
18.5, -4.6, -5.4; HRMS (EI) m/z (M+) calcd 284.2172, obsd
284.2145.
Anal. Calcd for C23H38OSi: C, 77.03; H, 10.68. Found C,
77.14; H, 10.71.
Anal. Calcd for C16H32O2Si: C, 67.55; H, 11.34. Found C,
67.68; H, 11.43.
Ozon olytic Clea va ge of 13. A cold (-78 °C) solution of
13 (506 mg, 1.41 mmol) in a 1:1 mixture of CH2Cl2 and
methanol (14 mL) containing 1% triethylamine was ozonolyzed
until a faint blue color appeared. The reaction mixture was
purged with oxygen, treated with dimethyl sulfide (0.65 mL),
warmed to room temperature, diluted with 1% NaHCO3
solution, and extracted with CH2Cl2. The combined organic
layers were dried and concentrated to leave a residue, puri-
For 10: IR (neat, cm-1) 1731; 1H NMR (300 MHz, CDCl3) δ
4.14 (ddd, J ) 12.0, 6.6, 1.0 Hz, 1 H), 2.46 (ddd, J ) 12.9, 2.9,
2.9 Hz, 1 H), 2.14 (dddd, J ) 12.7, 3.3, 3.3, 3.3 Hz, 1 H), 2.04
(ddd, J ) 13.0, 13.0, 1.1 Hz, 1 H) 1.91 (dddd, J ) 12.7, 3.0,
3.0, 3.0 Hz, 1 H), 1.67-1.36 (series of m, 3 H), 0.89 (s, 9 H),
(47) For generic experimental details, see ref 9.

Diastereoselective Addition of the Allylindium Reagent
J . Org. Chem., Vol. 63, No. 16, 1998 5613
fication of which by chromatography on silica gel (elution with
6% ethyl acetate in hexanes) gave 306 mg (76%) of 10.
CDCl3) δ 7.45-7.26 (m, 5 H), 4.96 (dd, J ) 2.0, 2.0 Hz, 1 H),
4.85 (dd, J ) 1.6, 1.6 Hz, 1 H), 4.58 (d, J ) 12.3 Hz, 1 H), 4.35
(d, J ) 12.3 Hz, 1 H), 3.99 (dd, J ) 2.6, 2.6 Hz, 1 H), 2.46-
2.35 (m, 1 H), 2.30-2.16 (series of m, 2 H), 1.99-1.91 (m, 1
H), 1.80 (dddd, J ) 12.4, 12.4, 3.2, 3.2 Hz, 1 H), 1.34 (ddd, J
) 13.0, 13.0, 3.0 Hz, 1 H), 1.14 (ddd, J ) 25.4, 12.4, 4.1 Hz, 1
H), 0.94 (s, 9 H); 13C NMR (75 MHz, CDCl3) ppm 147.9, 139.2,
128.2, 127.4, 127.1, 110.7, 78.8, 68.8, 40.9, 34.3, 32.0, 30.7, 28.6,
27.5; HRMS (EI) m/z (M+) calcd 258.1984, obsd 258.1980.
Hyd r id e Red u ction of 14. Diisobutylaluminum hydride
(6.82 mL of 1.0 M in hexanes, 6.82 mmol) was added via
syringe to a cold (-78 °C), magnetically stirred solution of 14
(1.11 g, 5.25 mmol) in hexanes (14 mL) under N2. The reaction
mixture was warmed to room temperature during 4 h, quenched
with saturated Rochelle salt solution, stirred overnight, and
extracted with ether. The combined organic layers were dried
and concentrated to leave a residue which was subjected to
chromatography on silica gel (gradient elution with 17 f 33%
ethyl acetate in hexanes). There was isolated 662 mg (57%)
of the less polar axial isomer 15 and 370 mg (33%) of the more
polar 16.
tr a n s-2-(Ben zyloxy)-4-ter t-bu tylcycloh exa n on e (20). A
cold solution of 19 (1.29 g, 5.0 mmol) in methanol (50 mL) was
ozonolyzed in the predescribed manner to give 714 mg (55%)
of 20 after silica gel chromatography (elution with 5% ethyl
acetate in hexanes): colorless oil; IR (neat, cm-1) 1719, 1455,
1
1367; H NMR (300 MHz, CDCl3) δ 7.40-7.27 (m, 5 H), 4.51
For 15: colorless crystals, mp 57-58 °C: IR (KBr, cm-1
)
1
(d, J ) 11.9 Hz, 1 H), 4.41 (d, J ) 11.9 Hz, 1 H), 3.72 (dd, J )
2.7, 2.7 Hz, 1 H), 2.82 (ddd, J ) 13.6, 13.6, 6.0 Hz, 1 H), 2.31-
2.23 (m, 2 H), 2.12-2.03 (m, 1 H), 1.96 (dddd, J ) 12.4, 12.4,
3.4, 3.4 Hz, 1 H), 1.52 (ddd, J ) 14.1, 12.6, 3.1 Hz, 1 H), 1.42
(dddd, J ) 12.7, 12.7, 12.7, 4.0 Hz, 1 H), 0.91 (s, 9 H); 13C NMR
(75 MHz, CDCl3) ppm 212.6, 137.5, 128.3, 127.6 (2C), 80.9,
71.0, 40.3, 37.7, 34.2, 31.9, 28.1, 27.4; HRMS (EI) m/z (M+)
calcd 260.1776, obsd 260.1780.
3510; H NMR (300 MHz, CDCl3) δ 3.98-3.90 (m, 4 H), 3.63
(s, 1 H), 2.20 (s, 1 H), 1.96-1.86 (m, 2 H), 1.72-1.64 (m, 1 H),
1.56 (dddd, J ) 13.1, 3.4, 3.4, 1.8 Hz, 1 H), 1.48-1.35 (m, 2
H), 1.22 (dddd, J ) 12.9, 12.9, 12.9, 3.5 Hz, 1 H), 0.85 (s, 9 H);
13C NMR (75 MHz, CDCl3) ppm 109.0, 70.0, 65.3, 64.3, 39.3,
31.8, 30.3, 30.2, 27.5, 24.1; HRMS (EI) m/z (M+) calcd 214.1569,
obsd 214.1566.
Anal. Calcd for C12H22O3: C, 67.26; H, 10.35. Found C,
67.36; H, 10.43.
Anal. Calcd for C17H24O2: C, 78.42; H, 9.29. Found C,
78.57; H, 9.32.
For 16: colorless crystals, mp 55-56 °C; IR (KBr, cm-1
)
Deben zyla tion of 20. A solution of 20 (714 mg, 2.74 mmol)
in ethanol (9 mL) containing 275 mg of 10% palladium on
carbon was stirred under an atmosphere of hydrogen for 1.5
h. The reaction mixture was filtered through Celite, concen-
trated, and subjected to flash chromatography (silica gel,
elution with 20% ethyl acetate in hexanes). There was
obtained 302 mg (64%) of 3 as a semisolid that was used
immediately. 1H NMR analysis indicated that modest levels
of 2 and 6 (ca. 20% combined) had been produced under these
conditions.
1
3493; H NMR (300 MHz, CDCl3) δ 4.12-3.94 (series of m, 4
H), 3.64-3.57 (m, 1 H), 2.03-1.92 (series of m, 2 H), 1.82 (ddd,
J ) 13.1, 13.1, 3.0 Hz, 1 H), 1.67-1.60 (m, 1 H), 1.36 (ddd, J
) 13.1, 13.1, 3.7 Hz, 1 H), 1.29-1.09 (series of m, 3 H), 0.86
(s, 9 H); 13C NMR (75 MHz, CDCl3) ppm 109.5, 73.5, 65.5, 65.3,
45.9, 33.6, 33.5, 32.2, 27.6, 24.1; HRMS (EI) m/z (M+) calcd
214.1569, obsd 214.1552.
Anal. Calcd for C12H22O3: C, 67.26; H, 10.35. Found C,
67.33; H, 10.42.
Hyd r olysis of 15. Major ketal 15 (241 mg, 1.00 mmol) was
refluxed in 75% aqueous acetone (20 mL) containing concen-
trated hydrochloric acid (0.01 mL) for 10 h, cooled to 20 °C,
diluted with brine, and partitioned several times with ether.
The combined organic extracts were dried and concentrated
to afford 120 mg (71%) of a mixture of 2 and 6 but none of 3,
P r ototyp ica l In d iu m -P r om oted Cou p lin gs. A. In Dr y
THF . To a magnetically stirred suspension of indium powder
(168 mg, 1.46 mmol) in dry THF (5 mL) was introduced allyl
bromide (0.126 mL, 1.46 mmol) via syringe and the mixture
was agitated for 5 min prior to introduction of neat ketone
(0.97 mmol). Reaction was allowed to proceed for 3 h at room
temperature, 10% hydrochloric acid was then added, and the
usual workup conditions were applied.
1
as determined by H NMR at 300 MHz.
tr a n s-2-(Ben zyloxy)-4-ter t-b u t ylm et h ylen ecycloh ex-
a n e (19). A mixture of sodium hydride (4.90 g, 0.20 mmol)
and freshly distilled DMSO (100 mL) was heated to 75-80 °C
for 45 min and then cooled to 0 °C. Freshly prepared
methyltriphenylphosphonium iodide (80.85 g, 0.20 mmol) was
introduced as a solution in warm DMSO (200 mL) via a
pressure-equalized addition funnel. The reaction mixture was
warmed to 20 °C during 1 h and the product was separated
from most of the DMSO by fractional distillation under
reduced pressure. Residual DMSO in the higher boiling
fractions was removed by flash column chromatography on
silica gel (elution with hexanes) to afford 26.84 g (80%) of 17
which was directly oxidized.
B. In 50% Aqu eou s THF . To a reaction vessel containing
allyl bromide (26 mg, 0.21 mmol), indium powder (37 mg, 0.32
mmol), and the ketone (0.19 mmol) were added THF (2.0 mL)
and water (2.0 mL). The reaction mixture was stirred at room
temperature until ketone consumption was complete (TLC
analysis), quenched with 10% HCl, and extracted with ether.
The combined organic phases were washed once with brine,
dried, and concentrated to leave an oily residue that was
purified chromatographically.
C. In Wa ter . A mixture of allyl bromide (30 mg, 0.25
mmol), indium powder (29 mg, 0.25 mmol), ketone (0.15 mmol),
and water (2.0 mL) was stirred at room temperature until
ketone consumption was complete (TLC analysis). A workup
identical to that in B was employed.
tert-Butyl hydroperoxide (32 mL) was added to a mixture
of selenium dioxide (441 mg, 3.98 mmol) and salicyclic acid
(1.10 g, 7.95 mmol) in CH2Cl2 (40 mL), to be followed by 17
(12.11 g, 79.5 mmol). The reaction mixture was stirred for 40
h, diluted with brine, and extracted with CH2Cl2. The
combined organic layers were washed with 10% KOH solution,
dried, and concentrated. Chromatographic purification (silica
gel, gradient elution with 5 f 20% ethyl acetate in hexanes)
provided 10.05 g (75%) of 18 as the only observed product. This
material was stirred with benzyl bromide (8.48 mL, 71.7
mmol), sodium hydride (1.94 g, 77.6 mmol), and tetra-n-
butylammonium iodide (2.21 g, 5.98 mmol) in dry THF (300
mL) at 0 °C for 2 h and at room temperature overnight. After
careful dilution with brine, the product was extracted into
ether, dried, and concentrated prior to chromatographic pu-
rification (silica gel, elution with 2.5% ethyl acetate in hex-
anes). There were isolated 7.84 g (51%) of 19 as a faint yellow
oil: IR (neat, cm-1) 1365, 1086, 1065; 1H NMR (300 MHz,
In all cases, product yields have been adjusted to discount
those allylation products resulting from attack on contaminat-
ing tautomeric forms. The product distributions represent the
average derived from duplicate runs.
Cou p lin g In volvin g 4-ter t-Bu tylcycloh exa n on e. Iso-
mer separation accomplished on silica gel (elution with 10%
ethyl acetate in hexanes).
1
For 22:36a colorless oil, IR (neat, cm-1) 3385 (br); H NMR
(300 MHz, C6D6) δ 5.84-5.75 (m, 1 H), 5.07-4.97 (m, 2 H),
2.04 (ddd, J ) 7.4, 1.1, 1.1 Hz, 2 H), 1.56-1.33 (series of m, 6
H), 1.15-1.05 (m, 2 H), 0.87 (s, 9 H), 0.92-0.75 (m, 2 H); 13C
NMR (75 MHz, C6D6) ppm 134.5, 118.0, 69.8, 49.1, 48.2, 37.7,
32.4, 27.8, 22.8.
For 23:36a colorless solid, mp 59.5-60.5 °C; IR (KBr, cm-1
)
3312 (br); 1H NMR (300 MHz, C6D6) δ 5.98-5.84 (m, 1 H),

5614 J . Org. Chem., Vol. 63, No. 16, 1998
Lobben
5.10-5.01 (m, 2 H), 2.18 (d, J ) 7.3 Hz, 2 H), 1.73-1.67 (m, 2
H), 1.52-1.45 (m, 2 H), 1.30 (ddd, J ) 12.1, 12.1, 3.5 Hz, 2 H),
1.13 (s, 1 H), 0.99-0.86 (m, 3 H), 0.78 (s, 9 H); 13C NMR (75
MHz, C6D6) ppm 134.5, 118.0, 71.2, 47.6, 41.4, 38.9, 32.2, 27.7,
24.5.
Cou p lin g In volvin g 1. Single product formed; purified by
silica gel chromatography (elution with 50% ethyl acetate in
hexanes).
For 24: colorless solid, mp 73-74 °C; IR (KBr, cm-1) 3405
1
(br), 3269 (br); H NMR (300 MHz, CDCl3) δ 5.94-5.80 (m, 1
H), 5.13-5.07 (m, 2 H), 3.42 (dd, J ) 9.3, 3.9 Hz, 1 H), 2.37
(dd, J ) 13.7, 7.5 Hz, 1 H), 2.27 (dd, J ) 13.7, 7.4 Hz, 1 H),
2.19 (br s, 1 H), 1.72-1.16 (series of m, 8 H); 13C NMR (75
MHz, CDCl3) ppm 133.8, 118.5, 73.2, 73.1, 43.5, 34.1, 30.2,
23.2, 21.0; HRMS (EI) m/z (M+) calcd 156.1150, obsd 156.1160.
For 33: colorless oil; IR (neat, cm-1) 3430 (br); 1H NMR (300
MHz, CDCl3) δ 5.95-5.81 (m, 1 H), 5.17-5.07 (m, 2 H), 2.33
(dd, J ) 14.0, 7.9 Hz, 1 H), 2.20 (dd, J ) 14.0, 6.6 Hz, 1 H),
2.19 (s, 1 H), 1.94 (s, 1 H), 1.76-1.44 (series of m, 5 H), 1.24
(dd, J ) 13.3, 13.3 Hz, 1 H), 1.18 (s, 3 H), 1.08-0.95 (m, 1 H),
0.85 (s, 9 H); 13C NMR (75 MHz, CDCl3) ppm 133.7, 118.4,
75.2, 74.3, 41.7, 38.1, 36.9, 32.0, 31.9, 27.5, 24.5, 23.5; HRMS
(EI) m/z (M+) calcd 226.1933, obsd 226.1956.
Anal. Calcd for C9H16O2: C, 69.19; H, 10.32. Found C,
69.03; H, 10.25.
Cou p lin g In volvin g 2. Single product formed; purified by
chromatography on silica gel (elution with 20% ethyl acetate
in hexanes).
For 26: colorless solid, mp 99-100 °C; IR (KBr, cm-1) 3288
1
(br); H NMR (300 MHz, CDCl3) δ 5.97-5.83 (m, 1 H), 5.15-
5.10 (m, 2 H), 3.42 (dd, J ) 11.2, 4.5 Hz, 1 H), 2.42 (dd, J )
13.6, 7.7 Hz, 1 H), 2.26 (dd, J ) 13.6, 7.3 Hz, 1 H), 1.93 (br s,
2 H), 1.84-1.72 (m, 2 H), 1.51-1.44 (m, 1 H), 1.37-1.18 (m, 3
H), 1.06-0.91 (m, 1 H), 0.86 (s, 9 H); 13C NMR (75 MHz, CDCl3)
ppm 134.1, 118.6, 74.2, 72.4, 46.4, 44.3, 34.8, 32.2, 31.9, 27.5,
21.5; HRMS (EI) m/z (M+ - C3H5) calcd 171.1385, obsd
171.1376.
Anal. Calcd for C17H30O2: C, 74.29; H, 11.57. Found C,
74.13; H, 11.52.
Anal. Calcd for C13H24O2: C, 73.54; H, 11.39. Found C,
73.57; H, 11.29.
Cou p lin g In volvin g 3. Single product; purified by chro-
matography on silica gel (elution with 20% ethyl acetate in
hexanes).
For 28: colorless semisolid; mp 86-90 °C; IR (KBr, cm-1
)
1
3404 (br); H NMR (300 MHz, CDCl3) δ 5.94-5.80 (m, 1 H),
5.06-4.99 (m, 2 H), 3.43 (br s, 1 H), 2.35 (dd, J ) 13.6, 7.4 Hz,
1 H), 2.13 (dd, J ) 13.6, 7.6 Hz, 1 H), 1.73-1.56 (m, 3 H),
1.50-1.35 (m, 5 H), 1.06 (s, 1 H), 0.86 (s, 9 H); 13C NMR (75
MHz, CDCl3) ppm 134.3, 118.7, 72.3, 71.7, 44.4, 40.5, 33.1,
32.0, 30.5, 27.6, 22.2; HRMS (EI) m/z (M+ - C3H5) calcd
171.1385, obsd 171.1390.
Cou p lin g In volvin g 6. Single product; isolation accom-
plished by chromatography on silica gel (elution with 25% ethyl
acetate in hexanes).
For 35: colorless solid, mp 63-67 °C; IR (neat, cm-1) 3381
1
(br); H NMR (300 MHz, CDCl3) δ 5.99-5.84 (m, 1 H), 5.16-
5.10 (m, 2 H), 3.37 (dd, J ) 11.4, 4.8 Hz, 1 H), 2.47 (dd, J )
13.6, 7.7 Hz, 1 H), 2.28 (dd, J ) 13.6, 7.4 Hz, 1 H), 1.92 (bs, 1
H), 1.84-1.69 (series of m, 4 H), 1.61-1.47 (m, 1 H), 1.38
(dddd, J ) 12.4, 12.4, 3.1, 3.1 Hz, 1 H), 1.09-0.89 (series of
m, 2 H), 0.84 (s, 9 H); 13C NMR (75 MHz, CDCl3) ppm 134.1,
118.7, 73.6, 73.2, 44.9, 41.6, 36.0, 32.0, 30.8, 27.5, 25.1; HRMS
(EI) m/z (M+ - C3H5) calcd 171.1385, obsd 171.1385.
Anal. Calcd for C13H24O2: C, 73.54; H, 11.39. Found C,
73.54; H, 11.44.
Anal. Calcd for C13H24O2: C, 73.54; H, 11.39. Found C,
73.72; H, 11.48.
Cou p lin g In volvin g 4. Single diastereomer; purified by
gradient elution (17-50% ethyl acetate in hexanes) on silica
gel.
For 30: colorless oil; IR (neat, cm-1) 3450 (br); 1H NMR (300
MHz, CDCl3) δ 6.03-5.89 (m, 1 H), 5.19-5.10 (m, 2 H), 2.51
(dd, J ) 13.7, 8.3 Hz, 1 H), 2.24 (s, 1 H), 2.15 (s, 1 H), 2.11
(dd, J ) 13.7, 6.7 Hz, 1 H), 1.77 (ddd, J ) 10.3, 2.8, 2.8 Hz, 1
H), 1.59-1.43 (m, 3 H), 1.40-1.23 (m, 2 H), 1.22 (s, 3 H), 1.08
(dddd, J ) 12.1, 12.1, 3.5, 3.5 Hz, 1 H), 0.84 (s, 9 H); 13C NMR
(75 MHz, CDCl3) ppm 134.5, 119.1, 75.0, 74.1, 45.0, 40.7, 38.9,
33.9, 32.1, 27.4, 23.1, 21.7; HRMS (EI) m/z (M+) calcd 226.1933,
obsd 226.1929.
Anal. Calcd for C14H26O2: C, 74.29; H, 11.57. Found C,
74.05; H, 11.63.
Cou p lin g In volvin g 5. Note that 33 was produced only
when anhydrous THF was the reaction medium. The products
were purified by chromatography on silica gel (elution with
11% ethyl acetate in hexanes).
For 32: colorless oil; IR (neat, cm-1) 3490 (br); 1H NMR (300
MHz, CDCl3) δ 5.98-5.84 (m, 1 H), 5.17-5.10 (m, 2 H), 2.44
(dd, J ) 13.7, 7.5 Hz, 1 H), 2.35 (dd, J ) 13.7, 7.6 Hz, 1 H),
1.76-1.19 (series of m, 7 H), 1.38 (s, 1 H), 1.34 (s, 1 H), 1.24
(s, 3 H), 0.85 (s, 9 H); 13C NMR (75 MHz, CDCl3) ppm 134.5,
119.0, 74.4, 73.6, 42.0, 40.3, 37.4, 32.7, 32.0, 27.5, 24.6, 21.5;
HRMS (EI) m/z (M+) calcd 226.1933, obsd 226.1936.
(1R*,2R*)-1-Allyl-1,2-(isop r op ylid en ed ioxy)cycloh ex-
a n e (25). A solution of diol 24 (156 mg, 1.0 mmol) in DMF (1
mL) was treated with 2-methoxypropene (0.19 mL, 2.0 mmol)
and pyridinium p-toluenesulfonate (25 mg, 0.10 mmol), stirred
at 0 °C for 45 min, diluted with brine, and extracted several
times with ether. The combined organic layers were dried and
concentrated, and the residue was purified chromatographi-
cally (elution with 9% ethyl acetate in hexanes). There was
isolated 172 mg (88%) of 25 as a colorless oil: 1H NMR (300
MHz, CDCl3) δ 5.98-5.84 (m, 1 H), 5.12-5.02 (m, 2 H), 3.90
Anal. Calcd for C17H26O2: C, 74.29; H, 11.57. Found C,
74.17; H, 11.60.

Diastereoselective Addition of the Allylindium Reagent
J . Org. Chem., Vol. 63, No. 16, 1998 5615
(d, J ) 2.6 Hz, 1 H), 2.37 (dddd, J ) 14.4, 6.9, 1.2, 1.2 Hz, 1
H), 2.28 (dd, J ) 14.4, 7.5 Hz, 1 H), 2.12-2.03 (m, 1 H), 1.71-
1.44 (series of m, 6 H), 1.49 (s, 3 H), 1.33 (s, 3 H), 1.18-1.08
(m, 1 H); 13C NMR (75 MHz, CDCl3) ppm 133.8, 117.7, 106.8,
80.1, 76.2, 40.1, 34.3, 28.5, 27.0, 26.2, 22.7, 19.8; HRMS (EI)
m/z (M+) calcd 196.1463, obsd 196.1464.
14.7, 14.7, 4.1 Hz, 1 H), 1.66-1.57 (m, 1 H), 1.51 (s, 3 H), 1.46
(s, 3 H), 1.49-1.38 (m, 1 H), 1.28 (m, 3 H), 1.08 (dd, J ) 14.3,
12.7, Hz, 1 H), 0.93 (dddd, J ) 10.9, 10.9, 4.6, 4.6 Hz, 1 H),
0.84 (s, 9 H); 13C NMR (75 MHz, CDCl3) ppm 134.7, 117.8,
106.7, 83.1, 82.7, 41.3, 39.7, 37.2, 32.8, 32.1, 30.5, 29.8, 27.4,
26.3, 23.4; HRMS (EI) m/z (M+ - CH3) calcd 251.2012, obsd
251.2012.
Anal. Calcd for C12H20O2: C, 73.43; H, 10.27. Found C,
73.51; H, 10.20.
Anal. Calcd for C17H30O2: C, 76.64; H, 11.35. Found C,
76.35; H, 11.33.
Cou p lin g In volvin g 7. Single product formed; purified by
chromatography on silica gel (elution with 20% ethyl acetate
in hexanes).
For 36: colorless oil; IR (neat, cm-1) 3334 (br); 1H NMR (300
MHz, CDCl3) δ 5.92-5.78 (m, 1 H), 5.15-5.05 (m, 2 H), 4.26
(m, 1 H), 2.97 (br s, 2 H) 2.17-2.15 (m, 2 H), 1.88 (ddd, J )
14.4, 3.2, 3.2 Hz, 1 H), 1.81-1.72 (m, 2 H), 1.56-1.49 (m, 1
H), 1.44 (dd, J ) 14.3, 3.0 Hz, 1 H), 1.39 (ddd, J ) 13.4, 13.4,
3.9 Hz, 1 H), 1.00-0.92 (m, 1 H), 0.96 (s, 9 H); 13C NMR (75
MHz, CDCl3) ppm 133.3, 118.9, 72.5, 69.1, 50.8, 48.1, 42.8,
37.8, 32.6, 28.6, 16.4; HRMS (EI) m/z (M+) calcd 212.1776, obsd
212.1777.
(1R*,2R*,4R*)-1-Allyl-4-ter t-bu tyl-2-m eth oxycycloh ex-
a n ol (27). A suspension of sodium hydride (4.0 mg, 0.17
mmol) and diol 26 (30 mg, 0.14 mmol) in dry THF (1.5 mL)
was stirred under N2 for 30 min prior to the addition of methyl
iodide (0.18 mL, 2.8 mmol). The reaction mixture was warmed
to room temperature during 12 h and then partitioned between
brine and ether. The combined organic layers were dried and
evaporated. Chromatography of the residue on silica gel
(elution with 8% ethyl acetate in hexanes) gave 22 mg (69%)
of 27 and returned 5 mg (17%) of unreacted 26. The spectral
data for 27 proved identical to those previously reported.9
(1R*,2S*,4R*)-1-Allyl-4-ter t-bu tyl-2-m eth oxycycloh ex-
a n ol (29). Methylation of 28 (31 mg, 0.15 mmol) in the
predescribed manner gave 29 (15 mg, 48%) while returning
11 mg (36%) of unreacted diol. The spectral data for 29
completely matched those previously reported.9
(1S*,2S*,4S*)-1-Allyl-4-ter t-b u t yl-1,2-(isop r op ylid en e-
d ioxy)-2-m eth ylcycloh exa n e (31). Heating 30 (80 mg, 0.38
mmol) with 2-methoxypropene (0.15 mL, 1.5 mmol) and
pyridinium p-toluenesulfonate (9 mg) in CH2Cl2 (3.8 mL) at
the reflux temperature for 14 h followed by workup in the
predescribed manner (elution with 2.5% ethyl acetate in
hexanes) afforded 83 mg (83%) of 31 as a colorless oil: 1H NMR
(300 MHz, CDCl3) δ 5.99-5.85 (m, 1 H), 5.14-5.03 (m, 2 H),
2.72 (dd, J ) 14.1, 5.4 Hz, 1 H), 2.19 (dd, J ) 10.8, 3.4 Hz, 1
H), 1.91 (dd, J ) 14.1, 8.6 Hz, 1 H), 1.74-1.61 (m, 3 H), 1.53
(s, 3 H), 1.47 (s, 3 H), 1.24 (s, 3 H), 1.21-1.14 (m, 2 H), 0.95-
0.86 (m, 1 H), 0.84 (s, 9 H); 13C NMR (75 MHz, CDCl3) ppm
134.2, 117.9, 106.6, 83.3, 82.5, 45.4, 41.7, 39.4, 32.8, 32.0, 30.3,
30.1, 27.3, 22.2, 21.6; HRMS (EI) m/z (M+) calcd 266.2246, obsd
266.2285.
Anal. Calcd for C13H24O2: C, 73.54; H, 11.39. Found C,
73.44; H, 11.34.
(1R*,5S*,6S*)-1-Allyl-6-ter t-b u t yl-3,3-d im et h yl-2,4-d i-
oxa bicyclo[3.3.1]n on a n e (37). A solution of 36 (0.101 g, 0.48
mmol), 2-methoxypropene (0.18 mL, 1.9 mmol), and PPTS (12
mg, 0.05 mmol) in CH2Cl2 (4.8 mL) was refluxed under N2 for
12 h, cooled to 25 °C, and diluted with brine. The aqueous
layer was separated and extracted with CH2Cl2. The combined
organic layers were dried and concentrated to leave an oil that
was purified by chromatography on silica gel (elution with
2.5% ethyl acetate in hexanes) to afford 77 mg (64%) of 37 as
a colorless oil: IR (neat, cm-1) 1377; 1H NMR (300 MHz,
CDCl3) δ 5.90-5.76 (m, 1 H), 5.09-5.00 (m, 2 H), 4.50 (dd, J
) 4.4, 0.9 Hz, 1 H), 2.51 (ddd, J ) 14.1, 5.3, 3.1 Hz, 1 H), 2.29-
2.12 (m, 2 H), 1.86 (dddd, J ) 12.9, 12.9, 12.9, 4.5 Hz, 1 H),
1.71-1.64 (m, 1 H), 1.59-1.53 (m, 1 H), 1.57 (s, 3 H), 1.36 (s,
3 H), 1.25 (ddd, J ) 13.1, 13.1, 4.7 Hz, 1 H), 1.25-1.20 (m, 1
H), 0.94 (s, 9 H), 0.87 (ddd, J ) 12.6, 3.7, 1.0 Hz, 1 H); 13C
NMR (75 MHz, CDCl3) ppm 134.1, 117.7, 96.8, 72.4, 68.2, 52.7,
47.5, 38.3, 33.3, 32.6, 31.9, 31.5, 28.7, 19.2; HRMS (EI) m/z
(M+) calcd 252.2089, obsd 252.2089.
(1R*,5S*,6S*)-1-Allyl-6-ter t-bu tyl-2,4-dioxabicyclo[3.3.1]-
n on a n -3-on e (38). A solution of 36 (30 mg, 0.14 mmol), 1,1′-
carbonyldiimide (22 mg, 0.014 mmol), and a catalytic amount
of DMAP in benzene (1.4 mL) was heated to reflux under N2
for 12 h. TLC analysis showed the continued presence of
starting material, and therefore a second equivalent of 1,1′-
carbonyldiimide was added and heating was continued under
N2 for an additional 12 h. The reaction mixture was concen-
trated and the residue was purifed by chromatography on silica
gel (elution with 20% ethyl acetate in hexanes) to yield 20 mg
Anal. Calcd for C17H30O2: C, 76.64; H, 11.35. Found C,
76.55; H, 11.31.
(58%) of 38 as a white solid, mp 90-91 °C: IR (KBr, cm-1
)
1718; 1H NMR (300 MHz, CDCl3) δ 5.86-5.72 (m, 1 H), 5.19-
5.09 (m, 2 H), 4.91-4.90 (m, 1 H), 2.46-2.31 (m, 2 H), 2.16-
2.00 (m, 2 H), 1.86-1.76 (m, 1 H), 1.72-1.51 (series of m, 3
H), 1.28-1.23 (m, 1 H), 0.98 (s, 9 H); 13C NMR (75 MHz, CDCl3)
ppm 150.0, 131.1, 119.9, 81.9, 75.6, 50.8, 44.9, 35.4, 34.0, 32.5,
28.2, 17.9; HRMS (EI) m/z (M+ + H) calcd 239.1647, obsd
239.1666.
Anal. Calcd for C14H22O3: C, 70.56; H, 9.30. Found C,
70.40; H, 9.36.
Cou p lin g In volvin g 8. Diols 39 and 40, produced in ratios
which differed with the reaction solvent (Scheme 15), were
separated by gradient elution chromatography on silica gel
(20-33% ethyl acetate in hexanes).
(1R*,2R*,4S*)-1-Allyl-4-ter t-bu tyl-1,2-(isop r op ylid en e-
d ioxy)-2-m eth ylcycloh exa n e (34). A solution of 33 (71 mg,
0.31 mmol), 2-methoxypropene (12 µL, 1.26 mmol), and pyri-
dinium p-toluenesulfonate (8 mg) in CH2Cl2 (3 mL) was
refluxed for 10 h and processed similarly (elution with 2.5%
ethyl acetate in hexanes) to furnish 40 mg (48%) of 34 as a
colorless oil: 1H NMR (300 MHz, CDCl3) δ 5.96-5.82 (m, 1
H), 5.12-5.03 (m, 2 H), 2.44 (ddd, J ) 14.1, 5.6, 1.2 Hz, 1 H),
2.19 (dd, J ) 14.1, 8.7 Hz, 1 H), 2.00 (ddd, J ) 14.3, 3.4, 2.6
Hz, 1 H), 1.83 (ddd, J ) 13.8, 5.0, 5.0 Hz, 1 H), 1.76 (ddd, J )
For 39: colorless solid; mp 118-119 °C; IR (KBr, cm-1) 3381
1
(br); H NMR (300 MHz, C6D6) δ 5.82-5.68 (m, 1 H), 5.07-
4.96 (m, 2 H), 3.85-3.80 (m, 1 H), 1.99-1.97 (m, 2 H), 1.63
(ddd, J ) 12.8, 4.2, 2.9 Hz, 1 H), 1.52-1.41 (m, 1 H), 1.38-
1.28 (m, 2 H), 1.11 (s, 9 H), 1.08-0.83 (series of m, 3 H); 13C
NMR (75 MHz, C6D6) ppm 134.0, 118.5, 72.1, 69.7, 53.4, 48.7,

5616 J . Org. Chem., Vol. 63, No. 16, 1998
Lobben
48.2, 37.0, 33.1, 29.6, 22.2; HRMS (EI) m/z (M+ - C3H5) calcd
171.1385, obsd 171.1373.
21.1; HRMS (EI) m/z (M+ - C3H5) calcd 171.1385, obsd
171.1382.
Anal. Calcd for C13H24O2: C, 73.54; H, 11.39. Found C,
73.47; H, 11.27.
Anal. Calcd for C13H24O2: C, 73.54; H, 11.39. Found C,
73.39; H, 11.26.
For 40: colorless solid, mp 105-106 °C; IR (KBr, cm-1) 3296
Ack n ow led gm en t. Funds in support of this re-
search were generously provided by the U.S. Environ-
mental Protection Agency (Grant R82-4725-010). The
assistance of Mr. Michael Gunzburger in the prepara-
tion of 7 and 8 is acknowledged with appreciation.
1
(br); H NMR (300 MHz, CDCl3) δ 5.95-5.81 (m, 1 H), 5.19-
5.10 (m, 2 H), 3.84-3.78 (m, 1 H), 2.23-2.21 (m, 2 H), 2.16
(br s, 1 H), 1.98 (br s, 1 H), 1.88-1.67 (series of m, 2 H), 1.60
(dd, J ) 13.2, 7.9 Hz, 1 H), 1.50-1.40 (m, 1 H), 1.29-1.11
(series of m, 3 H), 0.98 (s, 9 H); 13C NMR (75 MHz, CDCl3)
ppm 133.1, 119.2, 71.9, 69.7, 52.4, 45.0, 44.6, 36.6, 32.8, 29.1,
J O980974Y